Altman, M. C. (2017) - The Palgrave Kant Handbook PDF

Palgrave Handbooks in
German Idealism
Series Editor
Matthew C. Altman
Palgrave Handbooks in German Idealism is a series of comprehensive and authoritative
edited volumes on the major German Idealist philosophers and their critics.
Underpinning the series is the successful Palgrave Handbook of German Idealism
(2014), edited by Matthew C. Altman, which provides an overview of the period, its
greatest philosophers, and its historical and philosophical importance.
Individual volumes focus on specific philosophers and major themes, offering a
more detailed treatment of the many facets of their work in metaphysics, epistemol-
ogy, logic, ethics, aesthetics, political philosophy, and several other areas. Each volume
is edited by one or more internationally recognized experts in the subject, and
contributors include both established figures and younger scholars with innovative
readings. The series offers a wide-ranging and authoritative insight into German
Idealism, appropriate for both students and specialists.
More information about this series at

http://www.springer.com/series/14696
The Palgrave Kant Handbook

Edited by Matthew C. Altman
The Palgrave Schopenhauer Handbook (forthcoming)
Edited by Sandra Shapshay
The Palgrave Hegel Handbook (forthcoming)
Edited by Marina Bykova and Kenneth R. Westphal
The Palgrave Fichte Handbook (forthcoming)
Edited by Steven Hoeltzel
The Palgrave Handbook of German Romantic Philosophy (forthcoming)
Edited by Elizabeth Millán
The Palgrave Schelling Handbook (forthcoming)
Edited by Sean J. McGrath and Kyla Bruff
The Palgrave Handbook of Transcendental, Neo-Kantian, and Psychological Idealism
(forthcoming)
The Palgrave Handbook of Critics of Idealism (forthcoming)
Also by Matthew C. Altman

A COMPANION TO KANT’S CRITIQUE OF PURE REASON
THE FRACTURED SELF IN FREUD AND GERMAN PHILOSOPHY
(coauthored)
KANT AND APPLIED ETHICS: The Uses and Limits of Kant’s Practical
Philosophy
THE PALGRAVE HANDBOOK OF GERMAN IDEALISM (edited)
Matthew C. Altman
Editor
The Palgrave
Kant Handbook
Editor
Matthew C. Altman
Central Washington University
Ellensburg, Washington, USA
Palgrave Handbooks in German Idealism

ISBN 978-1-137-54655-5 ISBN 978-1-137-54656-2 (eBook)
DOI 10.1057/978-1-137-54656-2
Library of Congress Control Number: 2017947736
© The Editor(s) (if applicable) and The Author(s) 2017 Corrected publication April 2018
The author(s) has/have asserted their right(s) to be identified as the author(s) of this work in accordance with
the Copyright, Designs and Patents Act 1988.
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the
whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does
not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective
laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Cover illustration: Caravaggio, David with the Head of Goliath, c. 1607. Oil on wood. Kunsthistorisches
Museum, Vienna.
Printed on acid-free paper
This Palgrave Macmillan imprint is published by Springer Nature

The registered company is Macmillan Publishers Ltd.
The registered company address is: The Campus, 4 Crinan Street, London, N1 9XW, United Kingdom
Series Editor’s Preface
The era of German Idealism stands alongside ancient Greece and the French
Enlightenment as one of the most fruitful and influential periods in the
history of philosophy. Beginning with the publication of Kant’s Critique of
Pure Reason in 1781 and ending about ten years after Hegel’s death in 1831,
the period of “classical German philosophy” transformed whole fields of
philosophical endeavor. The intellectual energy of this movement is still very
much alive in contemporary philosophy; the philosophers of that period
continue to inform our thinking and spark debates of interpretation.
After a period of neglect as a result of the early analytic philosophers’
rejection of idealism, interest in the field has grown exponentially in recent
years. Indeed, the study of German Idealism has perhaps never been more
active in the English-speaking world than it is today. Many books appear
every year that offer historical/interpretive approaches to understanding the
work of the German Idealists, and many others adopt and develop their
insights and apply them to contemporary issues in epistemology, metaphy-
sics, ethics, politics, and aesthetics, among other fields. In addition, a number
of international journals are devoted to idealism as a whole and to specific
idealist philosophers, and journals in both the history of philosophy and
contemporary philosophies have regular contributions on the German
Idealists. In numerous countries, there are regular conferences and study
groups run by philosophical associations that focus on this period and its key
figures, especially Kant, Fichte, Schelling, Hegel, and Schopenhauer.
As part of this growing discussion, the volumes in the Palgrave Handbooks
in German Idealism series are designed to provide overviews of the major
figures and movements in German Idealism, with a breadth and depth of
coverage that distinguishes them from other anthologies. Chapters have been
v
vi Series Editor’s Preface
specially commissioned for this series, and they are written by established and
emerging scholars from throughout the world. Contributors not only pro-
vide overviews of their subject matter but also explore the cutting edge of the
field by advancing original theses. Some authors develop or revise positions
that they have taken in their other publications, and some take novel
approaches that challenge existing paradigms. The Palgrave Handbooks in
German Idealism thus give students a natural starting point from which to
begin their study of German Idealism, and they serve as a resource for
advanced scholars to engage in meaningful discussions about the movement’s
philosophical and historical importance.
In short, the Palgrave Handbooks in German Idealism have comprehen-
siveness, accessibility, depth, and philosophical rigor as their overriding goals.
These are challenging aims, to be sure, especially when held simultaneously,
but that is the task that the excellent scholars who are editing and contribut-
ing to these volumes have set for themselves.
Matthew C. Altman
Preface
This is the first volume on a single philosopher in the Palgrave Handbooks in

German Idealism series, and fittingly we begin with Immanuel Kant.
Although there were idealists before him, Kant formulated a version of
idealism that was so compelling and so challenging that it dominated
European philosophy for fifty years. His legacy continues to impact our
intellectual orientations, such that few contemporary philosophers would
seriously entertain the idea that we see things just as they are, without
transforming our experience through the activity of judgment. His work
cannot be ignored: most continental philosophy can be traced to Kant,
directly or indirectly, and he has been alternately criticized and embraced
by analytic philosophers.
Along with Plato and Aristotle, Kant is one of the most important figures
in the history of philosophy. He made original and significant contributions
to all of the major areas of philosophy: epistemology, metaphysics, logic, and
philosophy of mind; ethics and aesthetics; philosophy of science and philo-
sophical anthropology; philosophy of religion; political theory and philoso-
phy of history; and philosophy of education. His work has been analyzed in
many thousands of books and articles, Kant conferences meet regularly
around the world, and several journals are devoted to the study of his
thought.
Although no one resource can cover all of his philosophical achievements,
the Palgrave Kant Handbook provides readers with the most comprehensive
secondary source on Kant in terms of both breadth and depth. This volume
includes thirty-three newly commissioned chapters by some of the most
accomplished scholars in the field, addressing all of the major areas of
Kant’s philosophy. Contributors provide accessible introductions to the
vii
viii Preface
topics and also defend their own (sometimes controversial) interpretations.

Thus the book serves as a touchstone for appreciating Kant’s philosophical
and historical importance, suitable both for students hoping to understand
his positions and for Kant scholars who want to participate in contemporary
debates. In this sense, the book both contributes to an ongoing conversation
and provokes new thinking about Kant.
Contents
1 Introduction: Kant the Revolutionary 1

Matthew C. Altman
Part I Biographical and Historical Background
2 Kant’s Life 21
Steve Naragon
3 Kant and His Philosophical Context: The Reception and

Critical Transformation of the Leibnizian-Wolffian
Philosophy 49
Manuel Sánchez-Rodríguez
Part II Metaphysics and Epistemology
4 Transcendental Idealism: What and Why? 71

Paul Guyer
5 Noumenal Ignorance: Why, for Kant, Can’t We Know

Things in Themselves? 91
Alejandro Naranjo Sandoval and Andrew Chignell
ix
x Contents
6 Kant’s Concept of Cognition and the Key to the Whole

Secret of Metaphysics 117
Chong-Fuk Lau
7 Apperception, Self-Consciousness, and Self-Knowledge

in Kant 139
Dennis Schulting
Part III Logic
8 The Place of Logic within Kant’s Philosophy 165

Clinton Tolley
Part IV Relation between Theoretical and Practical Reason
9 The Primacy of Practical Reason 191

Ralph C. S. Walker
10 A Practical Account of Kantian Freedom 211

Matthew C. Altman
11 Moral Skepticism and the Critique of Practical Reason 243

David Zapero
Part V Ethics
12 How a Kantian Decides What to Do 263

Allen W. Wood
13 Duties to Oneself 285

Oliver Sensen
14 Demandingness, Indebtedness, and Charity: Kant on

Imperfect Duties to Others 307
Kate Moran
15 Kant and Sexuality 331

Helga Varden
Contents xi
16 Kant in Metaethics: The Paradox of Autonomy, Solved

by Publicity 355
Carla Bagnoli
Part VI Aesthetics
17 Feeling the Life of the Mind: Mere Judging, Feeling,

and Judgment 381
Fiona Hughes
18 On Common Sense, Communicability, and Community 407

Eli Friedlander
19 Immediate Judgment and Non-Cognitive Ideas: The

Pervasive and Persistent in the Misreading of Kant’s
Aesthetic Formalism 425
Jennifer A. McMahon
20 Sublimity and Joy: Kant on the Aesthetic Constitution

of Virtue 447
Melissa McBay Merritt
Part VII Philosophy of Science
21 “Proper Science” and Empirical Laws 471

John H. Zammito
22 From General to Special Metaphysics of Nature 493

Michael Bennett McNulty and Marius Stan
Part VIII Philosophy of Religion
23 Kant on Faith: Religious Assent and the Limits to

Knowledge 515
Lawrence Pasternack
24 The Fate of Religion within the Boundaries of Mere Reason 539

Martin Moors
xii Contents
Part IX Political Philosophy
25 The Critical Legal and Political Philosophy of Immanuel

Kant 567
Howard Williams
26 A Cosmopolitan Law Created by Cosmopolitan Citizens:

The Kantian Project Today 593
Soraya Nour Sckell
27 Kant’s Mature Theory of Punishment, and a First Critique

Ideal Abolitionist Alternative 617
Benjamin Vilhauer
Part X Anthropology, History, and Education
28 Denkungsart in Kant’s Anthropology from a Pragmatic

Point of View 643
Patrick R. Frierson
29 Kant on Emotions, Feelings, and Affectivity 665

Alix Cohen
30 The Philosopher as Legislator: Kant on History 683

Katerina Deligiorgi
31 Becoming Human: Kant’s Philosophy of Education and

Human Nature 705
Robert B. Louden
Part XI The Kantian Aftermath, and Kant’s Contemporary Relevance
32 Kant after Kant: The Indispensable Philosopher 731

Michael Vater
33 Kant, the Copernican Devolution, and Real Metaphysics 761

Robert Hanna
Contents xiii
34 Contemporary Kantian Moral Philosophy 791

Michael Rohlf
35 Conclusion: Kant the Philosopher 815

Matthew C. Altman
36 Erratum E1
Index 823
Notes on Contributors
Matthew C. Altman is Professor of Philosophy and Chair of the Philosophy

& Religious Studies Department at Central Washington University. He is
the author of A Companion to Kant’s “Critique of Pure Reason” (2008) and
Kant and Applied Ethics (2011), coauthor of The Fractured Self in Freud and
German Philosophy (2013), editor of The Palgrave Handbook of German
Idealism (2014), and series editor of the Palgrave Handbooks in German
Idealism.
Carla Bagnoli is Professor of Philosophy at the University of Modena, and

Professorial Fellow at the University of Oslo. She is the author of several
essays on Kantian ethics and Kantian constructivism. She is the editor of
Morality and the Emotions (2011) and Constructivism in Ethics (2013).
Andrew Chignell is Professor of Philosophy at the University of

Pennsylvania. His research focuses on Kant and early modern philosophy,
but he has additional interests in contemporary epistemology, aesthetics,
philosophy of religion, and applied ethics. He remains deeply noumenally
ignorant.
Alix Cohen works at the University of Edinburgh. She is the author of Kant
and the Human Sciences: Biology, Anthropology and History (2009), editor of
Kant’s Lectures on Anthropology: A Critical Guide (2014) and Kant on Emotion
and Value (2014), and associate editor of the British Journal for the History of
Philosophy.
xv
xvi Notes on Contributors
Katerina Deligiorgi is Reader in Philosophy at the University of Sussex. She

is the author of The Scope of Autonomy: Kant and the Morality of Freedom
(2012) and Kant and the Culture of Enlightenment (2005), and editor of
Hegel: New Directions (2006).
Eli Friedlander is Professor of Philosophy at Tel Aviv University. Among his

publications are Signs of Sense: Reading Wittgenstein’s “Tractatus” (2001), J. J.
Rousseau: An Afterlife of Words (2005), Walter Benjamin: A Philosophical
Portrait (2011), and Expressions of Judgment: An Essay on Kant’s Aesthetics
(2015). His current research is devoted to Walter Benjamin’s Arcades
Project.
Patrick R. Frierson is Associate Professor of Philosophy and Garrett Fellow

in the Humanities at Whitman College in Walla Walla, Washington. He is
the author of Freedom and Anthropology in Kant’s Moral Philosophy (2003),
What Is the Human Being? (2013), and Kant’s Empirical Psychology (2014);
and coeditor of Kant: Observations on the Feeling of the Beautiful and Sublime
and Other Writings (2011).
Paul Guyer is the Jonathan Nelson Professor of Humanities and Philosophy at

Brown University. He is the author, editor, and translator of two dozen works on
and by Kant, and is General Coeditor of the Cambridge Edition of the Works of
Immanuel Kant. The second edition of his monograph Kant and A History of
Modern Aesthetics (in three volumes) both appeared in 2014.
Robert Hanna is an independent philosopher and co-director of the

Contemporary Kantian Philosophy project. He has held research or teaching
positions in Brazil, Canada, Luxembourg, the UK, and the USA, and has
authored or coauthored six books, including Kant and the Foundations of
Analytic Philosophy (2001), Kant, Science, and Human Nature (2006),
Rationality and Logic (2006), Embodied Minds in Action (coauthored,
2009), In Defense of Intuitions: A New Rationalist Manifesto (coauthored,
2013), and Cognition, Content, and the A Priori: A Study in the Philosophy of
Mind and Knowledge (2015).
Fiona Hughes is Senior Lecturer in Philosophy at the University of Essex.

She has written articles on Kant’s aesthetics and epistemology, on phenom-
enology, and on the relation between philosophy and the arts. She is the
author of Kant’s Aesthetic Epistemology: Form and World (2007) and Kant’s
“Critique of Aesthetic Judgement”: A Reader’s Guide (2010).
Notes on Contributors xvii
Chong-Fuk Lau is Professor of Philosophy at The Chinese University of

Hong Kong. He is the author of Hegels Urteilskritik (2004), A New
Interpretation of Hegel (2014, in Chinese), a number of articles in Kant-
Lexikon (2015), and numerous papers on Kant and Hegel in journals such as
Review of Metaphysics, Kant-Studien, Kantian Review, Kant Yearbook,
Idealistic Studies, Owl of Minerva, Hegel-Jahrbuch, and Perspektiven der
Philosophie.
Robert B. Louden is Distinguished Professor and Professor of Philosophy at

the University of Southern Maine. His publications include Kant’s Human
Being (2011), The World We Want (2007), Kant’s Impure Ethics (2000), and
Morality and Moral Theory (1992). A former president of the North
American Kant Society (NAKS), Louden is also coeditor and translator of
two volumes in the Cambridge Edition of the Works of Immanuel Kant.
Jennifer A. McMahon is Professor of Philosophy, University of Adelaide. She

is the author of Art and Ethics in a Material World: Kant’s Pragmatist Legacy
(2014) and Aesthetics and Material Beauty: Aesthetics Naturalized (2007). She
also wrote the entry on beauty in Oxford Bibliographies Online: Philosophy
(2012) and the Routledge Companion to Aesthetics (2001, 2005).
Michael Bennett McNulty is Assistant Professor of Philosophy at the

University of Minnesota, Twin Cities. He has published a series of articles
on Kant’s conception of chemistry and its historical context. More generally,
his research interests are in Kant’s theoretical philosophy and his views on the
special sciences.
Melissa McBay Merritt is Senior Lecturer in Philosophy at the University of

New South Wales in Sydney, Australia, and Book Reviews Editor for the
Australasian Journal of Philosophy. She has written widely on Kant, and has
recently completed a book manuscript on Kant’s conception of virtue, titled
Kant on Reflection and Virtue (forthcoming).
Martin Moors is Emeritus Professor of Philosophy at the Institute of

Philosophy of the Catholic University Leuven (Belgium), where he taught
courses on metaphysics and philosophy of religion. His research and pub-
lications focus on modernity, German Idealism, Kant, and Schelling. As
visiting professor, he teaches at Peking University and Philosophicum
Thomas Aquinas at Kabgayi (Rwanda).
xviii Notes on Contributors
Kate Moran is Associate Professor of Philosophy at Brandeis University in

Waltham, Massachusetts. She is the author of Community and Progress in
Kant’s Moral Philosophy (2012) and various papers on Kant’s moral philoso-
phy. Her current research includes an examination of indebtedness and
dependence in Kant’s moral and political philosophy.
Steve Naragon is Professor of Philosophy at Manchester University

(Indiana), where he has taught since 1991. He coedited and co-translated
Kant’s Lectures on Metaphysics (1997) and maintains a website on Kant’s
teaching activities (www.manchester.edu/kant). He is currently transcribing
and translating Herder’s notes from Kant’s metaphysics lectures.
Alejandro Naranjo Sandoval is a graduate student in Philosophy at

Princeton University. His main interest is Kant’s theoretical philosophy in
the first and third Critiques and his lectures on logic. He also works on
Leibniz, philosophy of mind, philosophy of language, and logic.
Lawrence Pasternack is Professor of Philosophy at Oklahoma State

University. He is the author of “Religion within the Boundaries of Mere
Reason”: An Interpretation and Defense (2014), the Stanford Encyclopedia
entry on Kant’s philosophy of religion, as well as numerous papers on
Kant appearing in such outlets as Kant-Studien, Journal of the History of
Philosophy, Faith and Philosophy, and Religious Studies.
Michael Rohlf is Associate Professor of Philosophy at The Catholic

University of America and previously taught at Brown University and
Skidmore College. He has written articles on Kant’s moral philosophy,
metaphysics, and epistemology. He is the author of the general entry on
Kant in the Stanford Encyclopedia of Philosophy (2010).
Manuel Sánchez-Rodríguez is Associate Professor at the University of

Granada, Spain. His work focuses primarily on the study of the German
Enlightenment from Leibniz to Kant. He has translated Kant’s Lectures on
Anthropology into Spanish.
Dennis Schulting is a former Assistant Professor of Metaphysics and Its

History at the University of Amsterdam, the Netherlands, and obtained his
PhD in philosophy from the University of Warwick, England. He is the
author of two monographs on Kant’s Transcendental Deduction, Kant’s
Deduction and Apperception: Explaining the Categories (2012) and Kant’s
Notes on Contributors xix
Radical Subjectivism: Perspectives on the Transcendental Deduction (2017), and

the editor of Kantian Nonconceptualism (2016).
Soraya Nour Sckell is Coordinating-Investigator at the Centre of

Philosophy, University of Lisbon. She is also Director of the research
program on cosmopolitanism at the Collège International de Philosophie,
Paris. She is the author of À Paz Perpétua de Kant: Filosofia do Direito
International e das Relações Internacionais (2004); editor of The Minority
Issue: Law and the Crisis of Representation (2009); and coeditor of
Reconnaissance, identité et intégration sociale (2009), War and Peace: The
Role of Science and Art (2010), La fascination de la planète: L’éthique de la
diversité (2012), Interculturalité et transfert (2012), and Le Soi et le Cosmos
d’Alexander von Humboldt à nos jours (2015).
Oliver Sensen is Associate Professor of Philosophy at Tulane University. He

has written articles on Kant’s moral philosophy as well as human dignity. He
is the author of Kant on Human Dignity (2011), editor of Kant on Moral
Autonomy (2012), and coeditor of Kant’s “Tugendlehre” (2013) and Kant’s
“Lectures on Ethics”: A Critical Guide (2015).
Marius Stan is Associate Professor of Philosophy at Boston College. His

published research is on natural philosophy from Newton to Kant. With Eric
Watkins, he coauthored the entry on Kant’s foundations of science for the
Stanford Encyclopedia of Philosophy.
Clinton Tolley is Associate Professor of Philosophy at the University of

California, San Diego. He is the author of a number of articles on Kant and
modern German philosophy, and is currently at work on a book on Kant’s
transcendental idealism. He is also the coeditor and co-translator of The New
Anti-Kant (2014).
Helga Varden is Associate Professor in Philosophy and in Gender &

Women’s Studies at the University of Illinois at Urbana-Champaign. Her
main research interests are in legal-political philosophy, Kant’s practical
philosophy, Locke’s political philosophy, feminist philosophy, applied ethics,
and the philosophy of sex and love.
Michael Vater is a retired Professor of Philosophy at Marquette University

in Milwaukee. He has translated Schelling’s Bruno (1984), collaborated with
David W. Wood on The Philosophical Rupture between Fichte and Schelling
xx Notes on Contributors
(2013), and published numerous book chapters devoted to Fichte, Schelling,

and Hegel.
Benjamin Vilhauer is Associate Professor and Chair of the Philosophy

Department at CUNY City College. He works primarily on Kant’s theory
of freedom and contemporary free will theory. His articles have appeared in
journals including Philosophical Studies, Philosophical Quarterly, British
Journal for the History of Philosophy, and Canadian Journal of Philosophy.
Ralph C. S. Walker studied at McGill University, Montreal, and then at

Oxford, before becoming a Fellow of Magdalen College, Oxford, in 1972.
He has remained there since, though spending periods of academic activity in
overseas universities. Now retired, he continues to teach and write, with a
special interest in Kant.
Howard Williams is Honorary Distinguished Professor in the School of Law

and Politics at Cardiff University and Emeritus Professor of Political Theory
in the Department of International Politics, Aberystwyth. He is author of
Kant’s Political Philosophy (1983); Concepts of Ideology (1988); Hegel,
Heraclitus and Marx’s Dialectic (1989); International Relations in Political
Theory (1992); International Relations and the Limits of Political Theory
(1996); Kant’s Critique of Hobbes (2003); and Kant and the End of War
(2012). He is a founding editor of the journal Kantian Review and editor of
the Philosophy of Immanuel Kant. He is currently writing a study titled The
Legacy of Immanuel Kant in Political Philosophy.
Allen W. Wood is Ruth Norman Halls Professor at Indiana University and

Ward W. and Priscilla B. Woods Professor emeritus at Stanford University.
Wood has been professor at Cornell University and Yale University, and
visited at the University of Michigan, University of California at San Diego,
and Oxford University. He is author of twelve books and editor of eleven
others.
John H. Zammito is John Antony Weir Professor of History at Rice

University. He specializes in the intellectual history of the German
Enlightenment, Idealism, and Romanticism, concentrating especially on
the relationship between Kant and Herder. He has authored some eighty
articles and published the following monographs: The Genesis of Kant’s
“Critique of Judgment” (1992), Kant, Herder, and the Birth of Anthropology
Notes on Contributors xxi
(2002), and A Nice Derangement of Epistemes: Post-Positivism in the Study of

Science from Quine to Latour (2004).
David Zapero is a Fellow of the Alexander von Humboldt Foundation at the

Philosophy Department of the University of Bonn. He works on skepticism,
the philosophy of mind, and German philosophy. He is currently complet-
ing a monograph on skepticism about necessity.
Note on Sources and Key to Abbreviations
Works by Kant are referenced in the text parenthetically, using the abbreviations
listed below. When available, authors have used the standard English translations.
Where there is no mention of an English version, the translation is the author’s own.
Works cited only in footnotes are given with their full publication information.
As is customary in Kant scholarship, each parenthetical reference to Kant’s
writings gives the volume and page number(s) of the Royal Prussian Academy
edition (Kants gesammelte Schriften), which are included in the margins of the
translations. At the end of each of the following entries, I list the volume number
of Kants gesammelte Schriften in which the German version appears.
A/B Critique of Pure Reason (1781, 1787). Trans. and ed. Paul Guyer and Allen
W. Wood. Cambridge: Cambridge University Press, 1998. (Ak 3, 4) The
volume number is not included in references to the Critique of Pure Reason.
Ak Kants gesammelte Schriften. 29 vols. Ed. Preussische Akademie der
Wissenschaften and successors. Berlin: Reimer, later de Gruyter, 1900–.
References to this edition are given in the form Ak 8:5, indicating volume
and page number. Where applicable, the number of the Reflexion (R) is
given in addition to the volume and page number.
An Anthropology from a Pragmatic Point of View (1798). Trans. Robert B.
Louden. In Anthropology, History, and Education, ed. Günter Zöller and
Robert B. Louden, 231–429. Cambridge: Cambridge University Press,
2007. (Ak 7)
ANM Attempt to Introduce the Concept of Negative Magnitudes into Philosophy
(1763). In Theoretical Philosophy, 1755–1770, trans. and ed. David
Walford and Ralf Meerbote, 203–41. Cambridge: Cambridge University
Press, 1992. (Ak 2)
xxiii
xxiv Note on Sources and Key to Abbreviations
APL “M. Immanuel Kant’s Announcement of the Programme of His Lectures

for the Winter Semester 1765–1766” (1765). In Theoretical Philosophy,
1755–1770, trans. and ed. David Walford and Ralf Meerbote, 287–300.
Cambridge: Cambridge University Press, 1992. (Ak 2)
BL The Blomberg Logic (1770s). In Lectures on Logic, trans. and ed. J. Michael
Young, 1–246. Cambridge: Cambridge University Press, 1992. (Ak 24)
C Correspondence. Trans. and ed. Arnulf Zweig. Cambridge: Cambridge
University Press, 1999. (Ak 10–13)
CB “Conjectural Beginning of Human History” (1786). Trans. Allen W.
Wood. In Anthropology, History, and Education, ed. Günter Zöller and
2007. (Ak 8)
CF The Conflict of the Faculties (1798). Trans. Mary J. Gregor and Robert Anchor.
In Religion and Rational Theology, ed. Allen W. Wood and George di Giovanni,
233–327. Cambridge: Cambridge University Press, 1996. (Ak 7)
CJ Critique of the Power of Judgment (1790). Trans. Paul Guyer and Eric
Matthews. Ed. Paul Guyer. Cambridge: Cambridge University Press,
2000. (Ak 5)
CPrR Critique of Practical Reason (1788). In Practical Philosophy, trans. and ed.
Mary J. Gregor, 137–271. Cambridge: Cambridge University Press, 1996.
(Ak 5)
DS “Concerning the Ultimate Ground of the Differentiation of Directions in
Space” (1768). In Theoretical Philosophy, 1755–1770, trans. and ed. David
Press, 1992. (Ak 2)
DWL The Dohna-Wundlacken Logic (1792). In Lectures on Logic, trans. and ed. J.
Michael Young, 425–516. Cambridge: Cambridge University Press, 1992.
(Ak 24)
EAT “The End of All Things” (1794). Trans. Allen W. Wood. In Religion and
Rational Theology, ed. Allen W. Wood and George di Giovanni, 221–31.
EP “Essays regarding the Philanthropinum” (1776–1777). Trans. Robert B.
Louden. In Anthropology, History, and Education, ed. Günter Zöller and
2007. (Ak 2)
FI “First Introduction to the Critique of the Power of Judgment.” Trans. Paul
Guyer and Eric Matthews. In Critique of the Power of Judgment, ed. Paul
Guyer, 1–51. Cambridge: Cambridge University Press, 2000. (Ak 20)
FS “The False Subtlety of the Four Syllogistic Figures” (1762). In Theoretical
Philosophy, 1755-1770, trans. and ed. David Walford and Ralf Meerbote,
85-105. Cambridge: Cambridge University Press, 1992. (Ak 2)
Note on Sources and Key to Abbreviations xxv
G Groundwork of the Metaphysics of Morals (1785). In Practical Philosophy,

trans. and ed. Mary J. Gregor, 41–108. Cambridge: Cambridge University
Press, 1996. (Ak 4)
ID On the Form and Principles of the Sensible and the Intelligible World
[Inaugural Dissertation] (1770). In Theoretical Philosophy, 1755–1770,
trans. and ed. David Walford and Ralf Meerbote, 375–416. Cambridge:
Cambridge University Press, 1992. (Ak 2)
IUH Idea for a Universal History with a Cosmopolitan Aim (1784). Trans. Allen
W. Wood. In Anthropology, History, and Education, ed. Günter Zöller and
2007. (Ak 8)
JL The Jäsche Logic (1800). In Lectures on Logic, trans. and ed. J. Michael
LAn Lectures on Anthropology. Trans. Robert R. Clewis, Robert B. Louden, G.
Felicitas Munzel, and Allen W. Wood. Ed. Allen W. Wood and Robert B.
Louden. Cambridge: Cambridge University Press, 2012. (Ak 25) This
volume includes some or all of the lecture notes of Michael Friedländer
(1775–1776) (25:469–728), Friedrich Christian Starke (Menschenkunde)
(1781–1782?) (25:853–1203), and Christian Coelestin Mrongovius
(1784–1785) (25:1209–1429).
LE Lectures on Ethics. Trans. Peter Heath. Ed. Peter Heath and J. B.
Schneewind. Cambridge: Cambridge University Press, 1997. (Ak 27, 29)
This volume includes some or all of the lecture notes of J. G. Herder
(1762–1764) (27:3–89), Georg Ludwig Collins (1784–1785) (27:243–
471), C. C. Mrongovius (1784–1785) (29:597–642), and Johann
Friedrich Vigilantius (1793–1794) (27:479–732).
LF Thoughts on the True Estimation of Living Forces (1746–1749). Trans.
Jeffrey B. Edwards and Martin Schönfeld. In Natural Science, ed. Eric
Watkins, 1–155. Cambridge: Cambridge University Press, 2012. (Ak 1)
LM Lectures on Metaphysics. Trans. and ed. Karl Ameriks and Steve Naragon.
Cambridge: Cambridge University Press, 1997. (Ak 28, 29)
LP Lectures on Pedagogy (1803). Trans. Robert B. Louden. In Anthropology,
History, and Education, ed. Günter Zöller and Robert B. Louden, 437–85.
LRT Lectures on the Philosophical Doctrine of Religion (1817). Trans. Allen
W. Wood. In Religion and Rational Theology, ed. Allen W. Wood and
George di Giovanni, 339–451. Cambridge: Cambridge University
Press, 1996. (Ak 28)
MFS Metaphysical Foundations of Natural Science (1786). Trans. Michael
Friedman. In Theoretical Philosophy after 1781, ed. Henry Allison and
Peter Heath, 181–270. Cambridge: Cambridge University Press, 2002.
(Ak 4)
xxvi Note on Sources and Key to Abbreviations
MM The Metaphysics of Morals (1797). In Practical Philosophy, trans. and ed.

Mary J. Gregor, 363–602. Cambridge: Cambridge University Press, 1996.
(Ak 6)
NE A New Elucidation of the First Principles of Metaphysical Cognition (1755). In
Theoretical Philosophy, 1755–1770, trans. and ed. David Walford and Ralf
Meerbote, 3–45. Cambridge: Cambridge University Press, 1992. (Ak 1)
NF Notes and Fragments. Trans. Curtis Bowman, Paul Guyer, and Frederick
Rauscher. Ed. Paul Guyer. Cambridge: Cambridge University Press, 2005.
Where applicable, the number of the Reflexion (R) is given in addition to
the volume and page number.
NS Natural Science. Ed. Eric Watkins. Trans. Lewis White Beck, Jeffrey B.
Edwards, Olaf Reinhardt, Martin Schönfeld, and Eric Watkins.
Cambridge: Cambridge University Press, 2012.
OBS Observations on the Feeling of the Beautiful and Sublime (1764). Trans. Paul
Guyer. In Anthropology, History, and Education, ed. Günter Zöller and
2007. (Ak 2)
OBSn “Selections from the Notes on the Observations on the Feeling of the
Beautiful and Sublime” (1764–1765). In Notes and Fragments, trans.
Curtis Bowman, Paul Guyer, and Frederick Rauscher, ed. Paul Guyer,
1–24. Cambridge: Cambridge University Press, 2005. (Ak 20) Because this
translation does not include the Akademie pagination in the margins,
references to OBSn include the page of the translation followed by the
Akademie volume and page number.
OD On a Discovery whereby Any New Critique of Pure Reason Is to Be Made
Superfluous by an Older One (1790). Trans. Henry Allison. In Theoretical
Philosophy after 1781, ed. Henry Allison and Peter Heath, 283–336.
OP Opus postumum (1804). Trans. Eckart Förster and Michael Rosen. Ed.
Eckart Förster. Cambridge: Cambridge University Press, 1993. (Ak 21, 22)
OPA The Only Possible Argument in Support of a Demonstration of the Existence of
God (1763). In Theoretical Philosophy, 1755–1770, trans. and ed. David
Press, 1992. (Ak 2)
OT “What Does It Mean to Orient Oneself in Thinking?” (1786). Trans. Allen
W. Wood. In Religion and Rational Theology, ed. Allen W. Wood and
George di Giovanni, 7–18. Cambridge: Cambridge University Press, 1996.
(Ak 8)
PG Physical Geography (1802). Trans. Olaf Reinhardt. In Natural Science, ed.
Eric Watkins, 441–679. Cambridge: Cambridge University Press, 2012.
(Ak 9)
Note on Sources and Key to Abbreviations xxvii
PM The Employment in Natural Philosophy of Metaphysics Combined with

Geometry, of Which Sample I Contains the Physical Monadology (1756). In
Theoretical Philosophy, 1755–1770, trans. and ed. David Walford and Ralf
Meerbote, 47–66. Cambridge: Cambridge University Press, 1992. (Ak 1)
PP Toward Perpetual Peace (1795). In Practical Philosophy, trans. and ed. Mary
J. Gregor, 315–51. Cambridge: Cambridge University Press, 1996. (Ak 8)
Pro Prolegomena to Any Future Metaphysics That Will Be Able to Come Forward
as a Science (1783). Trans. Gary Hatfield. In Theoretical Philosophy after
1781, ed. Henry Allison and Peter Heath, 49–169. Cambridge: Cambridge
University Press, 2002. (Ak 4)
Rel Religion within the Boundaries of Mere Reason (1793). Trans. George di
Giovanni. In Religion and Rational Theology, ed. Allen W. Wood and
George di Giovanni, 55–215. Cambridge: Cambridge University Press,
1996. (Ak 6)
RH “Review of J. G. Herder’s Ideas for the Philosophy of the History of Humanity.
Parts 1 and 2” (1785). Trans. Allen W. Wood. In Anthropology, History,
and Education, ed. Günter Zöller and Robert B. Louden, 124–42.
RL “On a Supposed Right to Lie from Philanthropy” (1797). In Practical
Philosophy, trans. and ed. Mary J. Gregor, 605–15. Cambridge: Cambridge
RP What Real Progress Has Metaphysics Made in Germany since the Time of
Leibniz and Wolff? (1793, 1804). Trans. Peter Heath. In Theoretical
Philosophy after 1781, ed. Henry Allison and Peter Heath, 349–424.
RPT “On a Recently Prominent Tone of Superiority in Philosophy” (1796).
Trans. Peter Heath. In Theoretical Philosophy after 1781, ed. Henry Allison
and Peter Heath, 429–45. Cambridge: Cambridge University Press, 2002.
(Ak 8)
RS “Review of Schulz’s Attempt at an Introduction to a Doctrine of Morals for all
Human Beings regardless of Different Religions” (1783). In Practical
TelP “On the Use of Teleological Principles in Philosophy” (1788). Trans.
Günter Zöller. In Anthropology, History, and Education, ed. Günter Zöller
and Robert B. Louden, 195–218. Cambridge: Cambridge University Press,
2007. (Ak 8)
TP “On the Common Saying: That May Be Correct in Theory, but It Is of No
Use in Practice” (1793). In Practical Philosophy, trans. and ed. Mary J.
Gregor, 277–309. Cambridge: Cambridge University Press, 1996. (Ak 8)
xxviii Note on Sources and Key to Abbreviations
TPP “Proclamation of the Imminent Conclusion of a Treaty of Perpetual Peace

in Philosophy” (1796). Trans. Peter Heath. In Theoretical Philosophy after
1781, ed. Henry Allison and Peter Heath, 451–60. Cambridge: Cambridge
UNH Universal Natural History and Theory of the Heavens (1755). Trans. Olaf
Reinhardt. In Natural Science, ed. Eric Watkins, 182–308. Cambridge:
Cambridge University Press, 2012. (Ak 1)
VL The Vienna Logic (1780s). In Lectures on Logic, trans. and ed. J. Michael
WE “An Answer to the Question: What Is Enlightenment?” (1784). In Practical
List of Figures
Fig. 17.1 Rembrandt van Rijn, Militia Company of District II under the
Command of Captain Frans Banninck Cocq, commonly known
as The Night Watch, 1642. Oil on canvas. Rijksmuseum,
Amsterdam. 401
Fig. 29.1 The intentionality of the emotions 674
xxix
Chronology of Kant’s Life
This chronology includes all of Kant’s writings, with both a standard English
title (in boldface) and the original German or Latin title.
Kant published many shorter items in either of two local newspapers:
Wochentliche Königsbergische Frag- und Anzeigungs-Nachrichten (KFAN) and
the Königsbergsche Gelehrte und Politische Zeitungen (KGPZ). Beginning in
1784, nearly all of Kant’s longer essays appeared in the Berlinische
Monatsschrift (BM), and beginning in 1785, many of his shorter pieces
appeared in the Jena-based Allgemeine Literatur-Zeitung or its
Intelligenzblatt (both: ALZ).
1724 April 22: Kant is born in Königsberg (East Prussia; later Kaliningrad,
Russia), the first son and fourth child of Johann Georg Kant, a
harness maker, and his wife Anna Regina (Reuter) Kant.
1732 Easter: Kant begins his studies at the Collegium Fridericianum, a
Pietist boarding and day-school.
1737 December 18: Death of Kant’s mother (Anna Regina Kant,
1697–1737).
1740 July 20: Coronation in Königsberg of Friedrich II (1712–1786,
“the Great”).
September 24: Kant matriculates at the university in Königsberg
(Albertus-Universität), studying philosophy, mathematics, the nat-
ural sciences, and some theology.
1746 March 24: Death of Kant’s father (Johann Georg Kant, 1683–
1746) from a debilitating stroke suffered a year and a half earlier.
Kant finishes writing most of his first publication: Thoughts on the
True Estimation of Living Forces.
xxxi
xxxii Chronology of Kant’s Life
1748 August: From now until his promotion to Privatdozent at the

university (1755), Kant supports himself as a house tutor with
two families in the Königsberg area – the Andersch family in
Judtschen (east of Königsberg 100 km) and the von Hülsen family
in Arnsdorf (southwest of Königsberg 125 km).
1749 Kant’s first book is finally published: Thoughts on the True
Estimation of Living Forces (Gedanken von der wahren
Schätzung der lebendigen Kräfte).
1754 June: “Examination of the Question Whether the Axial
Rotation of the Earth . . . Has Changed since Its Origin”
(Untersuchung der Frage, ob die Erde in ihrer Umdrehung um die
Achse . . . einige Veränderung seit den ersten Zeiten ihres Ursprungs
erlitten habe) is published in a local Königsberg newspaper
(KFAN).
Before August 10: Kant returns to Königsberg.
August/September: “The Question, Whether the Earth Is
Ageing, Considered from a Physical Point of View” (Die
Frage, ob die Erde veralte, physikalisch erwogen) is published in
KFAN.
1755 March: Kant’s second book, Universal Natural History and
Theory of the Heavens (Allgemeine Naturgeschichte und Theorie
des Himmels), is published anonymously.
April 17: Kant presents his Magisterarbeit, “Succinct Exposition
of Some Meditations on Fire” (Meditationum quarundam de igne
succincta delineatio), to the dean of the philosophy faculty; first
published in 1839.
May 13: Kant successfully completes the examen rigorosum for the
Magister degree.
June 12: Kant receives the Magister degree in a public ceremony in
the large auditorium of the university, after which he gives a brief
(but now lost) speech in Latin.
September 27: Kant gives his Latin pro receptione disputation on A
New Elucidation of the First Principles of Metaphysical
Cognition (Principiorum primorum cognitionis metaphysicae nova
dilucidatio), after which he becomes a Magister legens with the right
to offer lectures at the university as a Privatdozent.
October 13: Kant gives his first lecture of the winter semester and
of his career.
November 1: An earthquake with an epicenter near Lisbon devas-
tates the city.
Chronology of Kant’s Life xxxiii
1756 January 24/31: Kant publishes the first of three articles that he will
write this year on earthquakes (the first and third in KFAN): “On
the Causes of Earthquakes on the Occasion of the Calamity
that Befell the Western Countries of Europe towards the End of
Last Year” (Von den Ursachen der Erderschütterungen bei
Gelegenheit des Unglücks, welches die westliche Länder von Europa
gegen das Ende des vorigen Jahres betroffen hat).
February: History and Natural Description of the Most
Noteworthy Occurrences of the Earthquake that Struck a
Large Part of the Earth at the End of the Year 1755
(Geschichte und Naturbeschreibung der merkwürdigsten Vorfälle des
Erdbebens, welches an dem Ende des 1755sten Jahres einen großen
Teil der Erde erschüttert hat) is published locally as a pamphlet.
March 23: Kant submits to the philosophy faculty his third Latin
disputation, The Employment in Natural Philosophy of
Metaphysics Combined with Geometry, of Which Sample One
Contains the Physical Monadology (Metaphysicae cum geometria
junctae usus in philosophia naturali, cuius specimen I. continet mon-
adologiam physicam), in support of his (unsuccessful) application
for Martin Knutzen’s old position as associate professor of Logic
and Metaphysics.
April 10: Public defense of his Physical Monadology.
April 10/17: Kant publishes his third article on earthquakes:
“Continued Observations on the Earthquakes That Have
Been Experienced for Some Time” (Fortgesetzte Betrachtung der
seit einiger Zeit wahrgenommenen Erderschütterungen).
April 25: Lecture announcement for the summer semester: New
Notes to Explain the Theory of the Winds (Neue Anmerkungen
zur Erläuterung der Theorie der Winde).
1757 April: Lecture announcement for the summer semester: Plan and
Announcement of a Series of Lectures on Physical Geography
with an Appendix Containing a Brief Consideration of the
Question: Whether the West Winds in Our Regions Are Moist
Because They Travel over a Great Sea (Entwurf und
Ankündigung eines Collegii der physischen Geographie nebst dem
Anhange einer kurzen Betrachtung über die Frage: Ob die
Westwinde in unsern Gegenden darum feucht seien, weil sie über ein
großes Meer streichen).
Kant applies (unsuccessfully) for a teaching position at the
Kneiphof School, one of the three Latin schools in Königsberg.
xxxiv Chronology of Kant’s Life
1758 January 22: The Russian occupation of Königsberg begins, lasting

until 1762.
April: Lecture announcement for the summer semester: New
Doctrine of Motion and Rest and the Conclusions Associated
with It in the Fundamental Principles of Natural Science (Neuer
Lehrbegriff der Bewegung und Ruhe und der damit verknüpften
Folgerungen in den ersten Gründen der Naturwissenschaft).
December: Kant applies (unsuccessfully) for the full professorship
of Logic and Metaphysics.
1759 July: Kant is introduced to his fellow townsperson Johann Georg
Hamann.
October 7: Lecture announcement for the winter semester: An
Attempt at Some Reflections on Optimism (Versuch einiger
Betrachtungen über den Optimismus).
1760 June: Publishes an open letter to the mother of a recently deceased
student: Thoughts on the Premature Death of Mr. Johann
Friedrich von Funk (Gedanken bei dem frühzeitigen Ableben des
Herrn Johann Friedrich von Funk).
1762 August: The Russian occupation of Königsberg ends.
Early August: Johann Gottfried Herder arrives in Königsberg and
attends Kant’s lectures for the next two years.
Early October: Lecture announcement for the winter semester:
The False Subtlety of the Four Syllogistic Figures (Die falsche
Spitzfindigkeit der vier syllogistischen Figuren).
Mid-December: The Only Possible Argument in Support of a
Demonstration of the Existence of God (Der einzig mögliche
Beweisgrund zu einer Demonstration des Daseins Gottes).
December 31: Kant submits his Prize Essay to the Prussian Royal
Academy; it will be published in 1764.
1763 June: Attempt to Introduce the Concept of Negative Magnitudes
into Philosophy (Versuch den Begriff der negativen Größen in die
Weltweisheit einzuführen).
August 10: Kant writes to Charlotte von Knobloch regarding
Emmanuel Swedenborg and related matters. This interest will
culminate in Kant’s Dreams of a Spirit-Seer, published in 1766.
1764 January: Observations on the Feeling of the Beautiful and
Sublime (Beobachtungen über das Gefühl des Schönen und
Erhabenen).
February: “On the Adventurer Komarnicki” (Raisonnement über
einen Abentheurer Jan Pawlikowicz Zdomozyrskich Komarnicki) is
Chronology of Kant’s Life xxxv
anonymously published in a local newspaper (KGPZ), followed by

five installments of the “Essay on the Maladies of the Head”
(Versuch über die Krankheiten des Kopfes).
March 23: “Review of Silberschlag’s Work: Theory of the
Fireball that Appeared on 23 July 1762” (Rezension von
Silberschlags Schrift: Theorie der am 23. Juli 1762 erschienenen
Feuerkugel) is anonymously published in KGPZ.
April: “Inquiry concerning the Distinctness of the Principles of
Natural Theology and Morality” (Untersuchung über die
Deutlichkeit der Grundsätze der natürlichen Theologie und der
Moral). Referred to as Kant’s “Prize Essay,” it is published along-
side Moses Mendelssohn’s first-place essay.
April/May: An extensive and favorable review of Kant’s Only
Possible Argument brings his name and writings to the attention
of a much wider audience, most importantly in Berlin.
August 5: Kant declines the offered Poetry chair.
1765 April: Kant concludes his final course of lectures on mathematics,
having previously given them nearly every semester.
Summer: Kant declines a possible offer of the Mathematics chair at
Halle.
October: Lecture announcement for the winter semester, including
a more detailed discussion of his pedagogical plans: Announcement
of the Programme of his Lectures (Nachricht von der Einrichtung
seiner Vorlesungen in dem Winterhalbenjahre von 1765–1766).
November 13: Correspondence begins with the Swiss-German
polymath Johann Heinrich Lambert (1728–1777).
December 31: In a letter to Lambert, Kant writes: “I have finally
reached the point where I feel secure about the method that has to
be followed.”
1766 January 31: Anonymous publication of Dreams of a Spirit-Seer
Elucidated by Dreams of Metaphysics (Träume eines Geistersehers,
erläutert durch Träume der Metaphysik), which criticizes claims to
knowledge of the supersensible.
February: Kant begins work as assistant librarian at the Castle
Library in Königsberg (his first salaried income), continuing until
April 1772.
February: Correspondence begins with the Berlin scholar Moses
Mendelssohn (1729–1786).
xxxvi Chronology of Kant’s Life
April 21: Markus Herz matriculates at the university (the third

Jewish student to do so), studying here (and with Kant) until
September 1770, when he returns to Berlin to practice medicine.
1768 “Concerning the Ultimate Ground of the Differentiation of
Directions in Space” (Von dem ersten Grunde des Unterschieds der
Gegenden im Raume) is published serially in KFAN.
1769 December 15: Kant declines an offer of the Logic and Metaphysics
chair at Erlangen.
1770 January 12: Kant declines a possible offer to teach at the university
in Jena.
March 31: King Friedrich II formally offers Kant the Logic and
Metaphysics chair at the university in Königsberg (Albertus-
Universität).
May 2: Official installation by the Academic Senate of Kant into
his professorship.
August 21: Kant gives his pro loco Latin disputation, the so-called
“Inaugural Dissertation”: On the Form and Principles of the
Sensible and the Intelligible World (De mundi sensibilis atque
intelligibilis forma et principiis).
1771 August 23: Kant publishes an anonymous review of a work on
comparative anatomy: “Review of Moscati’s Work, Of the
Corporeal Essential Differences between the Structure of
Animals and Humans” (Rezension zu Peter Moscati, Von dem
körperlichen wesentlichen Unterschiede zwischen der Struktur der
Tiere und Menschen), in KGPZ.
November 22–25: Local pastor and mathematician Johann Schultz
(1739–1805) publishes an anonymous review of Kant’s disserta-
tion in KGPZ. Based on this review, Kant calls him “the best head
for philosophy that I know in this region” (Ak 10:133).
1772 Kant considers getting married, but gets no further than that.
October: Kant begins lecturing on anthropology this semester,
continuing to offer this course every winter semester until his
retirement.
1774 October 10: Winter semester begins, with Kant lecturing on
metaphysics, anthropology, and moral philosophy. It is from
this semester that his best known set of moral philosophy notes
stem, published by Paul Menzer (1924) and translated into
numerous languages.
Near the end of the year: Kant is offered the rectorship of the
Mitau academy, which he declines.
Chronology of Kant’s Life xxxvii
1775 April: Announcement for the summer semester of his physical

geography lectures that also includes an essay on race: “Of the
Different Races of Human Beings” (Von den verschiedenen Racen
der Menschen, zur Ankündigung der Vorlesungen der physischen
Geographie im Sommerhalbjahr 1775).
1776 March 28: Kant anonymously publishes the first of two essays in
support of Johann Bernhard Basedow’s Philanthropinum, an
experimental school in Dessau, “Two Essays regarding the
Philanthropinum” (Zwei Aufsätze, betreffend das Basedow’sche
Philanthropinum), in KGPZ; the second essay will appear the
following year in the March 27 issue.
April 22: Kant begins his first semester as dean of the philosophy
faculty; this also involves joining the Academic Senate.
1777 February 28: Kant offers remarks (in Latin) at J. G. Kreutzfeld’s
inaugural dissertation: “Concerning Sensory Illusion and Poetic
Fiction” (titled Entwurf zu einer Opponenten-Rede in Ak; first
published in 1910).
August 18: Mendelssohn visits Kant’s lecture hall.
1778 February 28: Prussian Culture Minister Baron von Zedlitz
encourages Kant to accept a chair at the university in Halle, with
a steep increase in salary. Kant declines the offer.
1781 May 11: Critique of Pure Reason (Kritik der reinen Vernunft)
appears at the Easter book fair.
1782 January 19: The Garve/Feder review of the Critique of Pure Reason
is published anonymously in the Göttingischen Anzeigen von gelehr-
ten Sachen.
February 4: Kant publishes an anonymous “Notice of Lambert’s
Correspondence” (Anzeige des Lambert’schen Briefweschsels) in
KGPZ.
April 18: Kant writes a one-paragraph introduction, “A Note to
Physicians” (Nachricht an Ärzte), to a report on the 1782 influ-
enza epidemic that had been translated from the English into
German and reprinted in KGPZ.
1783 April: Prolegomena to Any Future Metaphysics That Will Be
Able to Come Forward as a Science (Prolegomena zu einer jeden
künftigen Metaphysic, die als Wissenschaft wird auftreten können).
April: “Review of Schulz’s Attempt at an Introduction to a
Doctrine of Morals for All Human Beings Regardless of
Different Religions” (Rezension von Schulz, Versuch einer
xxxviii Chronology of Kant’s Life
Anleitung zur Sittenlehre für alle Menschen, ohne Unterschied der

Religion) is published locally.
1784 At Kant’s request and continued urging, Johann Schultz publishes
a sketch and explanation of Kant’s Critique: Erläuterungen über des
Herrn Professor Kant Critik der reinen Vernunft.
May 22: Kant moves into his own house at 87–88
Prinzessinstraße.
September: Kant finishes writing his Groundwork of the Metaphysics
of Morals.
November: Idea for a Universal History with a Cosmopolitan
Aim (Idee zu einer allgemeinen Geschichte in weltbürgerlicher
Absicht) is the first of sixteen essays that Kant will publish in the
Berlinische Monatsschrift (BM).
December: “An Answer to the Question: What Is
Enlightenment?” (Beantwortung der Frage: Was ist Aufklärung?)
is published in BM.
1785 January 6: “Review of J. G. Herder’s Ideas for a Philosophy of
the History of Humanity, Parts 1 and 2” (Rezension zu Johann
Gottfried Herder, Ideen zur Philosophie der Geschichte der
Menschheit) is published anonymously in three installments in
the Schütz/Hufeland Allgemeine Literatur-Zeitung (ALZ).
March: “On the Volcanoes on the Moon” (Über die Vulkane im
Monde) is published in BM.
April: Groundwork of the Metaphysics of Morals (Grundlegung
zur Metaphysik der Sitten).
May: “On the Wrongfulness of Unauthorized Publication of
Books” (Von der Unrechtmäßigkeit des Büchernachdrucks) is pub-
lished in BM.
November: “Determination of the Concept of a Human Race”
(Bestimmung des Begriffs einer Menschenrasse) is published in BM.
1786 January: “Conjectural Beginning of Human History”
(Mutmaßlicher Anfang der Menschengeschichte) is published in BM.
Metaphysical Foundations of Natural Science (Metaphysische
Anfangsgründe der Naturwissenschaften).
April 18: “Review of Gottlieb Hufeland’s Essay on the Principle
of Natural Right” (Rezension von Gottlieb Hufeland, Versuch über
den Grundsatz des Naturrechts) is published in ALZ.
May 1: Kant begins his first term as rector of the university (for the
summer semester), an office that rotates each semester among the
members of the Academic Senate and involves presiding over
Chronology of Kant’s Life xxxix
meetings, matriculating students, and administering justice within

the university. During this time, Kant must arrange for the cor-
onation of Friedrich Wilhelm II (nephew and successor to
Friedrich the Great).
May: Two anonymous reviews (by Hermann Andreas Pistorius) of
the Critique of Pure Reason are published in the Allgemeine deutsche
Bibliothek.
August 17: Death of Friedrich II.
August: Karl Leonhard Reinhold publishes his Letters on the
Kantian Philosophy (Briefe über die Kantische Philosophie) in Die
Teutsche Merkur (August 1786, January-August 1787). Their
popularity pushes the Critique of Pure Reason to center stage after
years of relative neglect.
August 29: A cabinet order prohibits the philosophy professors at
Marburg from lecturing on Kant’s writings during the coming
winter semester.
September 19: Coronation in Königsberg of Friedrich Wilhelm II
(1744–1797).
October: “What Does It Mean to Orient Oneself in Thinking?”
(Was heißt: sich im Denken orientieren?) is published in BM.
October 12: Correspondence begins with Karl Leonhard Reinhold
(1757–1825).
December 7: Kant becomes a corresponding member of the Berlin
Academy of Sciences.
Kant contributes a preface: “Some Remarks on Ludwig Heinrich
Jakob’s Examination of Mendelssohn’s Morning Hours” (Einige
Bemerkungen on Ludwig Heinrich Jakob, Prüfung der
Mendelssohnschen Morgenstunden).
1787 Kant publishes a highly revised second edition of the Critique of
Pure Reason (the so-called “B-edition”).
1788 January/February: “On the Use of Teleological Principles in
Philosophy” is published in Teutsche Merkur.
Critique of Practical Reason (Kritik der practischen Vernunft).
April: “Kraus’s Review of Ulrich’s Eleutheriology” (Kraus’
Recension von Ulrich’s Eleutheriologie; published in ALZ) is written
by Kant’s close colleague, C. J. Kraus, but based in part on a (still
extant) draft supplied by Kant.
July 9: J. C. Wöllner, the new Minister of Education and Religious
Affairs, issues an “Edict concerning Religion” aimed at suppressing
the display of Enlightenment beliefs among teachers and clerics.
xl Chronology of Kant’s Life
October 4 (or October 10, 1786): “On the Philosophers’

Medicine of the Body” (De medicina corporis, quae philosophorum
est) is given as a Latin address at the completion of Kant’s term as
rector, either in 1786 or 1788. It is published posthumously in
1881.
December 19: Wöllner issues an “Edict on Censorship” to sup-
press irreligious publications in Berlin.
1789 March 3: By special order of the king, “as a sign of our full
satisfaction,” Kant is given a yearly raise of 220 talers, making
him the highest paid professor in all of Prussia.
Kant’s powers of concentration take a turn for the worse.
Beginning with the 1789 summer semester, he will teach only
two courses per term.
July 14: Soldiers and citizens storm the Bastille in Paris, marking
the beginning of the French Revolution.
Kant writes a “First Introduction to the Critique of the Power of
Judgment” (published in part in 1794 and in full in 1914).
Johann Schultz’s Prüfung der Kantischen Critik der reinen Vernunft
is published, with a second volume appearing in 1792. Kant
publicly acknowledges this work as the definitive account of his
philosophy.
1790 March: The two-page essay “On the Propensity to Fanaticism
and the Means to Oppose It” (Über den Hang zur Schwärmerei
und die Mittel dagegen) appears anonymously in a book by Kant’s
acquaintance and biographer L. E. Borowski.
April: Publication of Critique of the Power of Judgment (Kritik
der Urteilskraft) and On a Discovery Whereby Any New Critique
of Pure Reason Is to Be Made Superfluous by an Older One
(Über eine Entdeckung, nach der alle neue Kritik der reinen Vernunft
durch eine ältere entbehrlich gemacht werden soll).
June 12: “On the Announcement of an (Illegitimate) Edition of
I. Kant’s Minor Writings” (Über die Ankündigung einer
[unrechtsmäßigen] Ausgabe von I. Kants kleinen Schriften) is a brief
public notice (in ALZ) complaining of an unauthorized edition of
his essays.
September: “Schultz’s Review of Eberhard’s Magazine” (Zur
Recension von Eberhards Magazin, II. Band), published in ALZ, is
written by Kant’s close colleague, the mathematician Johan
Schultz, but based in part on a (still extant) draft supplied by Kant.
Chronology of Kant’s Life xli
1791 July 4: J. G. Fichte visits Kant’s classroom for the first time,
remaining in Königsberg until October.
September: “On the Miscarriage of All Philosophical Trials in
Theodicy” (Über das Misslingen aller philosophischen Versuche in
der Theodicee) is published in BM.
1792 April: “On Radical Evil in Human Nature” (Über das radikale
Böse in der menschlichen Natur) is published in BM. This is the first
of four parts of his 1793 Religion.
August 22: Kant jump-starts Fichte’s career with a notice in the
ALZ of Fichte’s authorship: “On the Author of the Attempt at a
Critique of All Revelation” (Über den Verfasser des Versuchs einer
Kritik aller Offenbarung).
1793 June 22: “To the Bookdealers” (An die Herren Buchhändler)
appears in the ALZ, Kant’s second public notice on pirated edi-
tions of his writings.
September: “On the Common Saying: That May Be Correct in
Theory, but It Is of No Use in Practice” (Über den Gemeinspruch:
Das mag in der Theorie richtig sein, taugt aber nicht für die Praxis) is
published in BM.
Religion within the Boundaries of Mere Reason (Religion inner-
halb der Grenzen der bloßen Vernunft).
November: Kant begins writing (but does not finish or publish)
What Real Progress Has Metaphysics Made in Germany since
the Time of Leibniz and Wolff? (Welches sind die wirklichen
Fortschritte, die die Metaphysik seit Leibnitzens und Wolf’s Zeiten
in Deutschland gemacht hat?) in response to a prize essay question
posed by the Berlin Academy of Sciences. It is eventually edited
and published by F. T. Rink in April 1804.
1794 May: “Something Concerning the Influence of the Moon on the
Weather” (Etwas über den Einfluß des Mondes auf die Witterung) is
published in BM.
June: “The End of All Things” (Das Ende aller Dinge) is pub-
lished in BM.
August 8: Kant becomes a corresponding member of the
Petersburg Academy of Sciences.
October 1: Kant receives a cabinet order from the king, written by
his minister Wöllner, censoring Kant for his writings on religion,
in particular the recently published and reissued Religion within the
Boundaries of Mere Reason.
xlii Chronology of Kant’s Life
1795 August: Toward Perpetual Peace (Zum ewigen Frieden. Ein phi-
losophischer Entwurf). This short work sells quickly, requiring a
new, expanded edition the following year.
October 14: Wöllner and G. F. Hillmer (the philosophy censor),
in the name of the king, issue an order to the Academic Senate in
Königsberg forbidding all professors from lecturing on Kant’s
Religion within the Boundaries of Mere Reason.
1796 May: “On a Recently Prominent Tone of Superiority in
Philosophy” (Von einem neuerdings erhobenen vornehmen Ton in
der Philosophie) is published in BM.
July 23: Kant stops lecturing during the middle of the summer
semester.
August: Kant contributes an afterword: “Remarks on
Sömmering’s On the Organ of the Soul” (Bemerkungen zu
Sömmering’s Über das Organ der Seele).
October: “Settlement of a Mathematical Dispute Founded on
Misunderstanding” (Ausgleichung eines auf Mißverstand beruhen-
den mathematischen Streits) is a one-page note published in BM.
December: “Proclamation of the Imminent Conclusion of a
Treaty of Perpetual Peace in Philosophy” (Verkündigung des
nahen Abschlusses eines Traktats zum ewigen Frieden in der
Philosophie) is published in BM.
1797 January 5: “Declaration regarding Hippel’s Authorship”
(Erklärung wegen der von Hippel’schen Autorschaft) is a brief public
notice (in ALZ) correcting a belief that Kant had written several of
the anonymously published works by Hippel.
January: The Metaphysics of Morals: Doctrine of Right (Die
Metaphysik der Sitten – Rechtslehre), part one of the Metaphysics of
Morals.
June 14: “Declaration against Schlettwein” (Erklärung gegen
Schlettwein) is Kant’s single contribution to an attempted public
debate by Johann August Schlettwein in the ALZ.
August: The Metaphysics of Morals: Doctrine of Virtue (Die
Metaphysik der Sitten – Tugendlehre), part two of the Metaphysics of
Morals.
September: “On a Supposed Right to Lie from Philanthropy”
(Über ein vermeintes Recht, aus Menschenliebe zu lügen) appears in
Berlinische Blätter, a successor to the Berlinische Monatsschrift.
November 16: Death of Friedrich Wilhelm II.
Chronology of Kant’s Life xliii
1798 April 4: Kant becomes a corresponding member of the Siena

Academy of Sciences.
June 5: Coronation in Königsberg of Friedrich Wilhelm III
(1770–1840, son of Friedrich Wilhelm II).
Kant publishes a brief pamphlet, “Explanatory Notes on the
Metaphysical Foundations of the Doctrine of Right”
(Erläuternde Anmerkungen zu den metaphysischen Anfangsgründen
der Rechtslehre), in response to an anonymous review of the
Rechtslehre.
Anthropology from a Pragmatic Point of View (Anthropologie in
pragmatischer Hinsicht abgefaßt) is prepared and published by
Kant, based on the popular anthropology lectures that he offered
every winter semester beginning in 1772–1773.
“On the Power of the Mind to Master Its Morbid Feelings by
Sheer Resolution” (Von der Macht des Gemüths durch den blossen
Vorsatz seiner krankhaften Gefühle Meister zu seyn) appears in C. W.
F. Hufeland’s medical journal as an open letter; later incorporated
as the third part of The Conflict of the Faculties (Der Streit der
Facultäten), also published that year.
On Turning Out Books: Two Letters to Mr. Friedrich Nicolai
from Immanuel Kant (Über die Buchmacherei. Zwei Briefe an
Herrn Friedrich Nicolai von Immanuel Kant).
1799 August: “Declaration regarding Fichte’s Wissenschaftslehre”
(Erklärung in Beziehung auf Fichte’s Wissenschaftslehre) is published
in the ALZ.
1800 Kant writes a brief “Preface to Reinhold Bernhard Jachmann’s
Examination of the Kantian Philosophy of Religion” (Vorrede zu
Reinhold Bernhard Jachmann: Prüfung der Kantischen
Religionsphilosophie), a book that Jachmann had undertaken in
part at Kant’s behest.
“Postscript to Christian Gottlieb Mielcke’s Lithuanian-
German and German-Lithuanian Dictionary” (Nachschrift
eines Freundes zu Heilsbergs Vorrede zu Christian Gottlieb Mielckes
Littauisch-deutschem und deutsch-littauischem Wörterbuch).
Logic (Immanuel Kants Logik, ein Handbuch zu Vorlesungen), edi-
ted by G. B. Jäsche, is prepared and published at Kant’s request,
but with no input or oversight by Kant.
1801 June 24: “Public Notice regarding the Illegitimate Edition,
Published by Vollmer, of Imm. Kant’s Physical Geography”
(Nachricht an das Publicum, die bey Vollmer erschienene
xliv Chronology of Kant’s Life
unrechtmäßige Ausgabe der physischen Geographie von Imm. Kant

betreffend) is published in ALZ.
November 14: Kant retires from his professorship and the
Academic Senate.
1802 Physical Geography (Immanuel Kants physische Geographie), edited
by F. T. Rink, is published under similar circumstances to the
Jäsche Logic of 1800.
1803 On Pedagogy (Über Pädagogik), edited by F. T. Rink, and pub-
lished under similar circumstances to the above, although the
provenance of the text is much less clear.
July: “His memory has almost completely left him” (an anon-
ymous report).
1804 February 12: Kant dies in Königsberg after a lengthy illness.
February 28: Kant is interred in the “Professor Crypt”
(Professorgewölbe) in the north wall of the Königsberg Cathedral.
Steve Naragon
1
Introduction: Kant the Revolutionary
Matthew C. Altman
Beginning with the publication of the Critique of Pure Reason in 1781, Kant
changed the trajectory of Western philosophy. His “Copernican revolution
in philosophy” challenged two thousand years of accepted doctrine.
Although philosophers disagreed about how to achieve knowledge – through
reasoning, the senses, divine revelation, and so on – there was general
consensus that truth is mind-independent and that our knowledge claims
are attempts to track what is the case regarding things as they are in
themselves, apart from consciousness. By contrast, Kant argued that the
world we represent is dependent on our subjective conditions for the possi-
bility of experience, with the implication that our knowledge is limited to
appearances. Responding to Kant’s discovery of the bounds of human
knowledge, the writer Heinrich von Kleist experienced what is called his
Kant-Krise (Kant crisis), stating: “The thought that we here on earth may
know nothing, nothing at all of Truth . . . has shattered me in the innermost
sanctum of my soul. – My single, my highest goal is sunk from sight and I
have no other.”1 Philosophy as whole can be said to have experienced such a
crisis on a large scale.
1
Heinrich von Kleist to Ulrike von Kleist, 23 March 1801, in An Abyss Deep Enough: Letters of Heinrich
von Kleist, trans. and ed. Philip B. Miller (New York: Dutton, 1982), 97–98.
M. C. Altman (*)
Department of Philosophy and Religious Studies, Central Washington University,
Ellensburg, USA
e-mail: matthew.altman@cwu.edu
© The Author(s) 2017 1

M. C. Altman (ed.), The Palgrave Kant Handbook, Palgrave Handbooks
in German Idealism, DOI 10.1057/978-1-137-54656-2_1
2 M. C. Altman
Kant destroyed a world of metaphysical speculation, and he created a

different philosophical terrain in which epistemology, the philosophy of
language, and the philosophy of mind set about understanding our place
in the world. Using the method of critique, “we consider [a concept] only in
relation to our cognitive faculties, hence in relation to the subjective condi-
tions for thinking it” (CJ 5:395). Kant would apply this approach not only to
the theory of knowledge, but to other areas of philosophy, including ethics
and aesthetics, most famously in his two other critiques, the Critique of
Practical Reason (1788) and the Critique of the Power of Judgment (1790),
respectively.
Philosophers responded in different ways to the critical philosophy.
Kantian idealism was extended and transformed by Reinhold, Fichte,
Schelling, and Hegel, and that movement gave rise to British Idealism and
the work of Green, Bradley, and Bosanquet. G. E. Moore and Bertrand
Russell took their refusal of idealism as the starting point for analytic
philosophy. However, other Anglo-American philosophers have drawn heav-
ily on Kant’s epistemology and metaphysics, including Peter Strawson,
Gareth Evans, and Wilfrid Sellars; and more recently the Pittsburgh (Neo-
Hegelian) School of John McDowell and Robert Brandom. In addition,
nearly every continental philosopher has been directly or indirectly influ-
enced by Kant. There is a line of descent from Kant through Schopenhauer
to Nietzsche, Derrida, Foucault, and Judith Butler; through Hegel to Marx,
the Frankfurt School, and Slavoj Žižek; and through Neo-Kantianism to
Husserl, Heidegger, Sartre, and Beauvoir. Kant’s thinking has been adopted
and applied in contemporary ethical theory by Onora O’Neill, Christine
Korsgaard, and Allen Wood; in the philosophy of mind by Hilary Putnam,
John Campbell, and Robert Hanna; in epistemology by Richard Rorty; in
aesthetics by Clement Greenberg; and in social/political philosophy by John
Rawls, Jürgen Habermas, and Charles Taylor. Given the breadth and depth
of his impact across the analytic-continental divide and in numerous fields of
philosophical inquiry, Kant’s importance cannot be overestimated.
The transcendentals
To introduce Kant’s philosophy and its importance, I will focus on his
revolutionary approach to three ideas: the True, the Good, and the
Beautiful. On some classical accounts, these are known as the “transcen-
dentals,” or properties of being that do not depend on how things affect
1 Introduction 3
the senses of finite knowers.2 For example, Plato identifies the True, the
Good, and the Beautiful as forms that can be intellectually apprehended,
or objects of knowledge apart from mere appearances. For Kant, how-
ever, the transcendentals are not qualities of being itself; rather, they
depend in essential ways on the activity of the subject. His discussions of
these ideas correspond roughly to the three Critiques – truth in the
Critique of Pure Reason, goodness in the Critique of Practical Reason,
and beauty in the Critique of the Power of Judgment – and in each case
we see how Kant transforms traditional approaches to them through his
Copernican turn. Kant shows how truth, goodness, and beauty depend
on or are the result of judgment – because, crucially, thinking is judging
– but he also explains how better and worse claims regarding knowledge,
ethics, and aesthetics are nonetheless possible. In other words, Kant
demonstrates how subjectively grounded norms can get us to objective
claims about truth and value.
Among other reasons, the threat of relativism was a particular concern for
Kant because of the work of David Hume. Hume defines causal claims and
inductive inferences as merely our habitual, subjective associations of con-
stantly conjoined occurrences. Similarly, moral distinctions are the result of
feelings of approval and disapproval, and standards of taste are established by
consensus among “true judges” who are especially attuned to the sentiment
of beauty. Truth, goodness, and beauty thus are subjective in the sense that
they are about our responses to things rather than the things themselves;
feelings need not be indicative of objective matters of fact.
Once we challenge the idea that the True, the Good, and the Beautiful
are properties of being, we run the risk of relativism. Kant, however, shows
how we could both validate such judgments by appealing to subjective
concepts and resist the relativism that would seem to follow. This is one of
Kant’s most important philosophical innovations, and it is one of the ways
that he can help to correct a false dichotomy under which many people
still labor: that all things are relative if they are mind-dependent, and that
the way to objectivity is to exclude the transformative power of thought.
Kant is trying to change what the terms ‘objective’ and ‘subjective’ mean,
2
In the B-edition of the Critique of Pure Reason, Kant adds a short section that reinterprets the ancient
theory of the transcendentals, which he identifies as the claim that “quodlibet ens est unum, verum, bonum
[every being is one, true, and good]” (B113). Kant says that, while the ancients identified these as
“predicates of things [Dinge],” they are really “logical requisites and criteria of all cognition of things in
general [Erkenntniß der Dinge überhaupt]” (B113–14). In this introduction, I am not addressing the
different interpretations of the transcendentals or Kant’s use of the term in the first Critique. Rather, I
am using the transcendentals as a device to focus on three revolutionary ideas in Kant.
4 M. C. Altman
or trying to clarify their different meanings. Judgments about the True,

the Good, and the Beautiful can be subjective in the sense that they
depend on the activity of the subject, but they are not subjective in the
sense of being true only relative to the contingent opinions of particular
individuals, be it one person or a group of true judges. There are shared
objects and shared moral and aesthetic standards, even though objectivity
depends on judgments according to concepts, the moral law is self-legis-
lated, and beauty is a function of how something affects our cognitive
faculties.
The True: Why do our a priori concepts track

the world?
Most philosophers prior to Kant tried to explain how we gain knowledge
of a mind-independent world. Kant confronted two such positions in the
form of empiricism and rationalism. According to Locke, the senses
receive information directly from things, and the resulting sensations
give us representations of the world that are either accurate themselves
or correctable by other, more reliable sensations. According to Descartes,
reasoning about the world corrects our unclear and indistinct percep-
tions, and we can thus discover the world as it must be – corresponding
to our a priori mathematical concepts, and so on. In his pre-critical
period, Kant largely followed Leibnizian-Wolffian rationalism in claiming
that an investigation of concepts amounts to an investigation of the
world itself.
Kant says that he was awakened from his “dogmatic slumber” by con-
fronting Hume’s skepticism, and specifically the problem of causality (Pro
4:260). According to Hume, causal relations could be established neither a
posteriori through the senses nor a priori through reason. The empirical
observation of two events that are constantly conjoined, one after the
other, cannot establish a necessary causal relation; and, because the effect is
distinct from the cause, the effect cannot be discovered merely by concep-
tually analyzing the cause. Hume concludes that we cannot justify our
commitment to causality as an objective phenomenon, and he explains that
what we identify as causal necessity really has a contingent, psychological
basis: when we see two events constantly conjoined, we subjectively associate
them, and that feeling of expectation – when we see one event and expect the
other – is all that causality is.
1 Introduction 5
Kant rejects Hume’s “skeptical solution” because causal claims are meant
to describe the world rather than our feelings and perceptions. When I say
that too much weight caused the bridge to collapse, for example, I am not
merely describing my subjective association of the two, or else any association
would be equally valid. There would be no such thing as a right or wrong
causal claim. Instead, a causal claim is a judgment about how two objects or
events are in fact related and an insistence that, because we experience a
shared world, you should also assent to that claim.
On Kant’s reading, Hume’s account of causality does not have skepticism
as its aim, but rather poses the problem of how to validate the use of concepts
with regard to things (Pro 4:258–59). Hume simply did not understand
where an objective concept of causality would come from if not from reason
(analytic a priori) or from the senses (synthetic a posteriori) – the two prongs
of Hume’s fork. So, he left it to others to explain how such a thing is possible:
“The discussion was only about the origin of this concept, not about its
indispensability in use” (Pro 4:259). Kant begins his explanation by noting
that “every event has a cause” is, unlike “every effect has a cause,” synthetic
rather than analytic – that is, we cannot just unpack the meaning of ‘event’ to
find out that it must be caused. In addition, causal laws are supposed to hold
universally and the resulting event is supposed to follow necessarily from the
cause, so it must be a priori rather than an a posteriori generalization derived
from experience. Hume could not conceive of such a thing, and yet Kant says
that there are several other claims of this kind that we make about the world,
claims that, like causality, are synthetic and a priori, such as the truths of
geometry and arithmetic and the principle of the conservation of mass.
Kant then confronts the problem of explaining how concepts that are not
based on experience can tell us anything about the world. As Kant phrases it,
“How are synthetic propositions a priori possible?” This is “the exact pro-
blem on which everything hinges” (Pro 4:276; see also B19, B73). Synthetic
a priori judgments are not possible if the empiricists or the rationalists are
right. Both of them assume that there is a world out there, independently of
our thinking, which we are trying to get at either through our perception of
the world or through our rational conception of it. And they assume that our
a priori concepts necessarily show up in the world that we experience,
without ever explaining why there is this correspondence. But, Kant says, if
the world is not separate from my thinking – that is, if my representation of
the world is in part created by me because of how I conceive of it – then it
makes sense that the world would conform to my understanding, because the
world that I experience is structured by that very understanding, or is in part
the result of my thinking. In the preface to the first Critique, Kant compares
6 M. C. Altman
this shift in our worldview to Copernicus’s challenge to the Ptolemaic model

of the universe:
This would be just like the first thoughts of Copernicus, who, when he did not
make good progress in the explanation of the celestial motions if he assumed
that the entire celestial host revolves around the observer, tried to see if he
might not have greater success if he made the observer revolve and left the stars
at rest. Now in metaphysics we can try in a similar way regarding the intuition
of objects. If intuition has to conform to the constitution of the objects, then I
do not see how we can know anything of them a priori; but if the object (as an
object of the senses) conforms to the constitution of our faculty of intuition,
then I can very well represent this possibility to myself. (Bxvi-xvii)
Just as the seemingly retrograde motions of the planets cannot be explained

given an immobile earth at the center of the universe, the fact that a priori
concepts apply to objects is inexplicable if the world that we experience is
taken to be a fixed point, absolutely distinct from the understanding. Kant’s
Copernican revolution in philosophy reconceives what we know by claiming
that it depends on the activity of the observer. That is why we can discover
what the world is like by looking at how we understand the world.
Kant goes on to validate this “experiment” by showing that space and time
are pure forms of sensible intuition and that a Table of Categories can be
derived from the forms of logical judgment, and that together these are
conditions for the possibility of experience. He concludes that, because our
experience of the world is in part due to these subjective conditions, then
objects exist for me as “mere appearances” (A45–49/B62–66; Pro 4:287–88).
I can never know things as they are apart from these conditions, or things as
they are in themselves. This is transcendental idealism.
Kant’s position does not devolve into a kind of Berkeleyan idealism according
to which the world is merely an idea because, he says, in addition to being a
transcendental idealist, he is also an empirical realist. Although our subjective
conditions of knowing (the forms of sensible intuition [space and time] and the
categories) determine the form of our experience, the matter of our experience
(raw sense data) is actually given to us by things in themselves – and both are
necessary for experience (A50–51/B74–75; Pro 4:318). In fact, in his Refutation
of Idealism, Kant argues that our experiences can only have a temporal order if
they are correlated with a mind-independent reality (B274–79). Thus our
experience is only possible if there is a world distinct from consciousness.
Transcendental idealism is an epistemological doctrine about what we know –
we can only know appearances and never things in themselves – not an
1 Introduction 7
ontological doctrine that says the external world is entirely a product of thinking
or (worse yet) merely a figment of my imagination.
Kant makes an important move here: one can recognize the mind-
dependence of the world we inhabit without capitulating to the relativist
idea that there is no shared standard of truth. There are better and worse
ways to apply the categories. We misidentify causes, get numbers of things
wrong, think that nonexistent things are real, and so on. We apply the
categories by representing what is common to objects, as indicated by
certain “marks [Merkmale]” by which we relate them to one another
(A320/B377; JL 9:91, 94). Thus we are receptive to what the world gives
us and have to track the world with our concepts, even though the resulting
experience is deeply informed by our thinking, such that we can only know
the world as an appearance.
Kant shows that objective claims about causality (among other things) are
possible by showing that we apply a priori concepts to our sensible intuitions,
and that our knowledge extends only to appearances. But he has also avoided
the subjectivist challenge to the existence of a mind-independent world and the
related worry that knowledge is relative to the individual perceiver: we are all
using the same categories and applying them to a shared world. Our cognition
is limited to appearances because it is dependent on the rules of thinking, but
we can still make knowledge claims that are more or less justified.
The Good: Why be moral?

Many moral theories struggle with the issue of why we should do what is right.
Establishing that something is the right thing to do is, for many theorists,
supposed to give us reason to do it, yet the question “Why be moral?” is not
nonsensical, and it poses a special problem for theories that have us pursue ends
such as virtue or happiness, follow God’s commands, or adhere to cultural
expectations or the natural law. Kant presents his ethics as an alternative to these
views, and in doing so he also addresses the question of moral motivation:
ethical obligation is a form of self-constraint, so the question of why I ought to
be moral does not arise for Kant in the same way that it does for these other
positions. I ought to be moral because I am the one holding myself to account;
in acting at all, I am committing myself to act rightly.
Kant begins the Groundwork of the Metaphysics of Morals (1785) by
claiming that such seemingly good things as intellectual talents, gifts of
fortune, virtue, and happiness are only conditionally good. It is bad for a
8 M. C. Altman
criminal to have the keen intelligence, courage, and wealth that would help
him to more effectively carry out his evil plans. Even happiness is only good
if someone deserves to be happy (G 4:393–94). The goodness of all of these
things depends on the person having a good will. Unlike those other things, a
good will is the only unconditional good, in the sense that its goodness does
not depend on what it “effects or accomplishes” (G 4:394; see also CPrR
5:24–27). Thus no principle is suited actually to tell us what is right and
wrong except for a deontological moral principle. To have a good will means
that you act from the motive of duty: you do something not because it
satisfies some desire, but because it is right; that is, you do what is right even
if you do not want to do it or you have nothing to gain by it.
Since the good will is unconditionally good, it is good in every case. It does
not depend on your circumstances or the particular things that you want,
and so moral requirements must hold universally. If it is moral, then every-
one ought to do it; if they ought to do it, then they have to be able to do it
(“ought implies can”); and if everyone cannot do it, then it must not be right.
In other words, because we cannot base morality on the particular ends that
we are pursuing – that is, the matter of the action – any right action must be
in accordance with the form of lawfulness itself (“everyone ought to . . .”):
Since I have deprived the will of every impulse that could arise for it from
obeying some law, nothing is left but the conformity of actions as such with
universal law, which alone is to serve the will as its principle, that is, I ought
never to act except in such a way that I could also will that my maxim should
become a universal law. (G 4:402; see also CPrR 5:27)
This is the categorical imperative, and specifically the Formula of Universal

Law: your subjective principle of acting (maxim) must be something that all
rational agents in relevantly similar circumstances could adopt (G 4:401n,
420n; MM 6:389). If what you are doing depends on everyone else acting
differently – for example, you are able to lie effectively only if most other
people do not lie and there is a general sense of trust that you take advantage
of – then it must not be the right thing to do (G 4:402–3). You would be
making an exception of yourself, excusing yourself from what is rationally
required in favor of what you happen to want (G 4:424).
Kant strictly distinguishes motivation by the moral law from motivation by
inclinations. In both cases, I am choosing freely what to do; I am responsible
for adopting my maxim. We are the sort of beings who weigh options, decide
what is best, and, as Kant puts it, we set the ends that we pursue. This is what
it means to have a will: not only to act according to laws, but to act according
1 Introduction 9
to our representation of laws (G 4:412). It is this reflective distance from our

inclinations that makes us free and makes us capable of morality. Kant
distinguishes the capacity to choose among different possible incentives
(Willkür) from the ability to act on the basis of a principle that I give to
myself through pure practical reason (Wille). When I act for the sake of duty, I
am choosing to do what reason demands rather than giving myself over to my
natural inclinations – acting for reasons all the way down. Thus only moral
action is fully rational and autonomous (CPrR 5:33).
So where does moral obligation come from? We act rationally only if we
adhere to the conditions of rational activity. In the theoretical sphere, our
reasoning is constrained by the forms of logical judgment and the under-
standing is constrained by the categories. In thinking, we commit ourselves
to the constraints of reason, such that we cannot, for example, deny that the
conclusion follows from the premises in a valid argument, or assert that an
event is uncaused. Similarly, in acting, we commit ourselves to the con-
straints that make practical reasoning possible. Specifically, asserting that I
have a good reason to do something means that I am committed to its
generalization: someone in my position, under my circumstances, should act
as I do. What makes it a good reason instead of a bad reason is its
universalizability. By reasoning at all, then, we are committing ourselves to
standards that determine whether our reasons for acting are good or bad,
much as we are committed to certain logical and conceptual ways of thinking
with regard to truth claims. Since none of us is perfectly rational, we are
necessitated by these rules to act as we ought to act. This is why the moral law
constrains us as an imperative.
There are two key points to note about this. First, these constraints on our
reasoning are not a limitation on our freedom, but rather an expression of
our freedom. If we could somehow relinquish our capacity to reason and be
driven entirely by instinct, like animals, we would not be bound by the moral
law, but we would not be freer than we are now. In that case, our actions
would be constrained by our representations – specifically, by what end
seems best to fulfill our strongest desire. Although we would be driven by
psychological causes, insofar as we (unlike mere machines) would act on the
basis of representations of possible ends, this would nonetheless be what Kant
calls “the freedom of a turnspit, which, when once it is wound up, also
accomplishes its movements of itself” (CPrR 5:97). And of course, the
complete absence of any constraints, even if such a thing were possible,
would leave us with arbitrariness rather than freedom. Considering shareable
reasons to determine the right course of action is not somehow less free than
making your decisions by the flip of a coin.
10 M. C. Altman
Second, Kant claims that we place these practical constraints on ourselves

through an act of self-legislation: autonomy is properly defined as “the
property of the will by which it is a law to itself” (G 4:440; see also CPrR
5:33). Whether we are justifying our actions to others or merely considering
what we ought to do, we are implicitly committing ourselves to standards of
justification – basically, to the idea that there are better and worse reasons, a
distinction that is made possible by constraints on our reasoning. By enga-
ging in the process of reason-giving, we are committed to the formal condi-
tions of rational willing itself. As we have seen, what makes rational willing
rational, rather than merely a capitulation to inclination, is that our maxims
are shareable by other like-situated agents. We thus impose the categorical
imperative on ourselves. For this reason, the question “Why be moral?” poses
no threat to Kant’s ethics. In acting at all, we bind ourselves to the moral law.
The question then shifts to the value of our capacity to reason and whether
we should act rationally. Kant has at least two different responses to this.
First, he says that, when we act, we must act under the idea of freedom (G
4:448). From the first-person perspective, we take ourselves to weigh reasons
and decide freely what to do. Thus the question of whether we should be
moral does not arise because we cannot help but take ourselves to be acting
freely and, as such, we are bound by the categorical imperative. Kant’s second
response is more metaphysically substantive. Once we shift from asking
about the value of morality to asking about the value of reasoning, Kant
has a clear answer: things become good for us only when we set them as ends,
so the capacity to set ends – to determine that things are good – has intrinsic
value, because it is the condition of other things being valued. From this, he
concludes that we are obligated to respect the humanity in ourselves and
others as an end in itself – the so-called Formula of Humanity of the
categorical imperative (G 4:428–29). On this account, we ought to do
what is right because, in doing so, we respect the value that we have as
self-determining beings.
The Beautiful: How can aesthetic judgments be

justified?
In his early work on aesthetics, Observations on the Feeling of the Beautiful and
Sublime (1764), Kant claims that our response to beauty in art and nature is
subjective, and that the feeling varies depending on one’s innate sensitivity,
sex, and nationality, among other things. Even in the Critique of Pure Reason
1 Introduction 11
(1781), Kant insists that a critique of taste, which would subject our aesthetic
judgments to principles of reason, is impossible. Like Hume, Kant concludes
that aesthetic rules or criteria can be nothing but empirical generalizations
based on what people do in fact feel (A21n/B35n).
In the Critique of the Power of Judgment (1790), however, Kant claims that
there are better and worse aesthetic judgments, and that some judgments of
taste can lay claim to universal validity. As we saw earlier, theoretical judg-
ments about objects are right or wrong depending on whether we properly
apply the categories. If two people present contradictory explanations of what
caused an event, they cannot both be true. However, if we have different
tastes – for example, I like chocolate ice cream best and you prefer strawberry
– then neither of us is right or wrong. We simply like different things. Kant
claims that judgments of taste are distinct in that they are about feeling and
they can be evaluated against a shared standard of taste. On the one hand,
judgments of beauty have to do with how someone responds to nature or art.
The fact that something is beautiful (or interesting, graceful, garish, etc.) is
not some fact that can be found in the work itself. It has to do with how the
viewer responds to it. So, in this sense, it is like the experience of enjoying a
particular flavor of ice cream. But it also seems that there is a right and a
wrong answer when it comes to aesthetic judgments. If you say that a velvet
Elvis is a better piece of art or is more beautiful than Michelangelo’s David,
you are wrong. The question then becomes how judgments of beauty, which
seem subjective and depend essentially on feelings, could lay claim to uni-
versal validity. If you do not appreciate a magnificent natural scene or a great
piece of art, you should. How does Kant make sense of this kind of constraint
with regard to our aesthetic judgments?
What makes the aesthetic feeling of beauty different from the mere
gratification we get from pleasant things is that only in the first case do we
experience “a disinterested and free satisfaction” (CJ 5:210). One can
appreciate the beauty of nature or art even apart from having some desire
that is satisfied by it. For example, you can take pleasure in a great sculpture
even if you are not a gallery owner who is trying to sell it. In addition, when
we make a judgment of taste, Kant says, we are not bringing particular
experiences under some general rule as we are when we make a theoretical
judgment. With determinate (theoretical) judgments, the concept is suffi-
cient to determine whether a given particular is an instance of the concept.
Aesthetic judgments, however, are a kind of reflective judgment, by which
Kant means that we apprehend particular representations and try to bring
them under an “indeterminate principle of a purposive arrangement of
nature in a system, as it were for the benefit of our power of judgment”
12 M. C. Altman
(FI 20:214). That is, the various elements of nature or a work of art seem to
be arranged in order to affect our sensibility in a way that pleases us, even
though we cannot explain why they have that effect.
Kant says that art has the quality of being purposive without purpose
(Zweckmäßigkeit ohne Zweck), meaning that it looks like nature or the artist is
following a rule or that there is some reason for the different parts being there,
but we cannot figure out what it is (CJ 5:236). If the purpose of art were clear,
or if we simply applied concepts to what we are doing, then there would be rules
to follow to make good art or to make appropriate judgments about art. It
seems like there are rules out there – there are right and wrong ways to do things
– but we cannot discover them. Aesthetic judgments have to do with feelings
rather than facts. Such judgments have the form of a determinate judgment:
“Michelangelo’s David is beautiful” seems to be a case of bringing the object
(Michelangelo’s David) under a general concept (beautiful) by means of the ‘is.’
However, the “determining ground” of an aesthetic judgment is subjective
feeling rather than an objective concept (CJ 5:203). Specifically, Kant says that
we take pleasure in the “harmony” or “free play” of the imagination and the
understanding (CJ 5:217–18). When we experience something beautiful, the
imagination is not bound by any particular concept or purpose, so it acts freely
and without constraint, while our rational interest in unity is satisfied insofar as
it seems like there is some principle, albeit an indeterminate principle, under
which it can be conceptualized by the understanding.
Although there is no objective rule of taste and aesthetic judgments are
essentially subjective, this does not entail that there is no standard of
taste by which we can evaluate people’s aesthetic judgments. Because
they are based on the feeling of disinterested pleasure, aesthetic judg-
ments depend on how art or nature affects my sensibility apart from the
particular desires that distinguish me from others – my wanting to sell it,
for example. Therefore, insofar as we all have the same basic faculties –
we share a “common sense [sensus communis]” (CJ 5:238–40, 293–96) –
aesthetic judgments have “subjectively universal validity” (CJ 5:215). If
we do not let our particular desires and other idiosyncrasies obstruct our
appreciation of nature or art, and if we attend properly to its form, then
we will judge the same things to be beautiful. This is why when someone
makes an aesthetic judgment, he makes a demand on others that, if it is
a correct appraisal, constrains them in a way that they are not con-
strained by his preference for chocolate ice cream:
Many things may have charm and agreeableness for him, no one will be
bothered about that; but if he pronounces that something is beautiful, then
1 Introduction 13
he expects the very same satisfaction of others: he judges not merely for himself,
but for everyone, and speaks of beauty as if it were a property of things. Hence
he says that the thing is beautiful, and does not count on the agreement of
others with his judgment of satisfaction because he has frequently found them
to be agreeable with his own, but rather demands it from them. (CJ 5:212–13)
The fact that we have common feelings of disinterested pleasure in the same
objects, consistently over time and across cultures, lends empirical support to
the claim that these objects are beautiful (CJ 5:231–32). Art theory and art
history identify what tends to make a given piece of art good or not good,
and make us aware of the features that tend to produce the free play of our
cognitive faculties, without ever being able to discover rules for making or
identifying good art.
One commonly hears that art is subjective. One of Kant’s lasting con-
tributions to aesthetics is to concede the truth of that claim and yet also to
explain why some pieces of art are valued by vast majorities of the human
population, despite divergent desires and interests; why there is value in art
instruction and art appreciation, even though none of us can discover rules of
taste; and why aesthetic judgments are not merely expressions of personal
approval, but claims that place a demand on others and their feelings.
Summary of chapters
The preceding notes on Kant’s epistemology, ethics, and aesthetics give only
a sense of Kant’s philosophy and its significance. The chapters in this book
bring together many of the most important Kant scholars working today, and
they give a fuller picture of why Kant’s work continues to be the focus of so
much excellent scholarship, in so many different areas of philosophy.
The book begins with Part I, which sets Kant’s life and work in their
historical context. The first chapter gives a brief biography of Kant, describ-
ing his social and religious circumstances, his early education, his work as a
professor, and the goals of the critical philosophy (Naragon). Chapter 3
examines how Kant relates to the Leibnizian-Wolffian philosophy, specifi-
cally how Leibniz’s system of pre-established harmony theory is adopted and
transformed in Kant’s critical theory of reflective judgment and teleology
(Sánchez-Rodríguez).
Part II focuses on Kant’s metaphysics and epistemology. It begins with the
overarching question of how to define transcendental idealism, and specifically
14 M. C. Altman
how to interpret Kant’s distinction between appearances and things in them-

selves, either as two aspects or two worlds, and whether Kant denies that things
in themselves are spatiotemporally located (Guyer). Other chapters examine
particular elements of Kant’s theoretical philosophy, including his claim that we
cannot know anything beyond appearances (“noumenal ignorance”), and
different arguments for this claim (Naranjo Sandoval and Chignell); Kant’s
approach to cognition, and how Kant establishes the objective validity of
representations (Lau); and the transcendental unity of apperception, and spe-
cifically how self-consciousness and self-identity depend on or are made possible
by apperception (Schulting). Part III includes one chapter on Kant’s logic. Kant
distinguishes transcendental logic from merely formal approaches to logic, and
because of that, logic becomes crucial in the first Critique by serving as a possible
source of a priori cognition (Tolley).
Part IV provides a transition between Kant’s theoretical philosophy and
his ethical theory by examining how theoretical and practical reason are
related. Chapter 9 defends Kant’s claim that practical reason has primacy
over theoretical reason. The claim is validated because only practical reason
can provide a kind of knowledge that theoretical reason cannot – namely,
knowledge that the will is free (Walker). The second chapter in Part IV
defends a practical interpretation of Kant’s theory of freedom, according to
which we are committed to moral responsibility only from the practical
standpoint, as opposed to the determinism of the theoretical standpoint
(Altman). Finally, chapter 11 explains how Kant’s transcendental deduction
of freedom attempts to answer the challenge of moral skepticism (Zapero).
Part V covers topics in Kant’s ethics, beginning with how the categorical
imperative functions in moral decision making. The chapter argues against
the idea that Kant’s theory is best understood in terms of a decision
procedure; rather, in recognizing the dignity of rational nature, we must
use judgment to find out what is morally required of us in particular
circumstances, and given the sort of people we want to become (Wood).
Other authors apply Kant’s ethical theory to more specific issues: explaining
that duties to oneself have priority because of a fundamental commitment to
the law of one’s own reason (Sensen), resolving the seeming tension between
duties of assistance and the obligation to avoid dependence on others
(Moran), and, in the context of a particular account of human nature,
confronting issues regarding human sexuality, including marriage and pros-
titution (Varden). Chapter 16 addresses Kant’s importance for contemporary
metaethics and defends Kantian constructivism as a plausible account of
moral obligation, claiming that Kant is committed to a “dialogical” inter-
pretation of autonomy (Bagnoli).
1 Introduction 15
Contributors turn their attention to Kant’s aesthetics in Part VI. The

section begins with an investigation of how Kant’s aesthetic theory fits with
the critical philosophy as a whole. Chapter 17 makes the case that Kant gives
an account of feeling that operates as a non-cognitive and yet reflective form
of awareness, and that this exhibits a cooperation of the faculties that is
necessary for the possibility of any cognition (Hughes). Authors then con-
sider how the idea of common sense functions in the articulation of aesthetic
judgments (Friedlander), and how Kant’s aesthetic formalism should be
understood in the larger context of the critical philosophy, and contrary to
recent misinterpretations (McMahon). The final piece of this section exam-
ines the Kantian sublime and how it furthers our moral development, and
specifically the cultivation of virtue (Merritt).
Two chapters on Kant’s philosophy of science comprise Part VII.
Chapter 21 investigates the aims of science in Kant’s philosophy, situates
his theory in its historical context, and shows how his notion of a “proper
science,” which is to discover universal and necessary truths, differs from our
modern conception of empirical natural science (Zammito). The second
chapter explains Kant’s framework for natural science by focusing on the
general metaphysics of the first Critique and its relationship to the special
metaphysics of nature in the Metaphysical Foundations of Natural Science
(McNulty and Stan).
Kant’s philosophy of religion is related both to his epistemology, since
there he shows why speculative metaphysics is impossible, and to his ethical
theory, since he claims that religion ultimately must serve moral purposes.
Part VIII begins with a chapter on faith and knowledge, specifically pure
rational faith as it relates to the highest good and the practical postulates, and
Kant’s approach to historical/ecclesiastical faith. It concludes that pure
rational faith can be understood as a kind of holding-to-be-true, even within
the constraints of transcendental idealism (Pasternak). Chapter 24 considers
what it means for religion to be “within” the boundaries of reason, and how
Kant defines grace, Christ, the church, and religious rituals and practices in
ways that further morality (Moors).
Part IX includes three chapters on Kant’s political theory, beginning with
a chapter on how politics and law are related to metaphysics and morality;
Kant’s account of the social contract, property relations, civil society, and the
state; and the relationship among the different levels of law – national,
international, and cosmopolitan (Williams). Contributors also explore
Kant’s cosmopolitanism and its contemporary relevance, especially for
cross-border associations and struggles for identity (Sckell); and critically
reappraise Kant’s retributivist theory of punishment in light of the first
16 M. C. Altman
Critique, concluding with a non-retributive, “ideal abolitionist” alternative,

which appeals to a version of Rawls’s original position (Vilhauer).
Broadly speaking, Part X considers Kant’s theory of human nature and the
ways that it informs his philosophies of anthropology, history, and educa-
tion. Chapter 28 illuminates Kant’s conception of “pragmatic” anthropology
by considering his treatment of humans’ “mode of thought [Denkungsart],”
and what it means to study human beings “as free” (Frierson). Chapter 29
examines Kant’s general theory of sensibility, giving a Kantian account of the
nature of emotions, feelings, and desires and making a case for a definition of
emotions that recognizes they are somewhat controllable while also denying
that they are essentially motivational forces (Cohen). Chapter 30 identifies
Kant’s conception of history as a unified whole and defends his practical
justification of philosophical history (Deligiorgi). And Chapter 31 reevalu-
ates Kant’s philosophy of education as a central doctrine of the critical
philosophy, since, according to Kant, we only realize our humanity through
education (Louden).
The final section of the book, Part XI, provides a philosophical-historical
evaluation of Kant’s reception among his contemporaries and his continuing
importance. The first chapter traces Kant’s legacy along three different paths:
how he defines the history of philosophy, how he defines the social nature of
intelligence by his characterization of the cognizing/acting subject, and how
he supports an Enlightenment agenda and a concern for greater human
freedom (Vater). Chapters 33 and 34 focus on Kant’s relevance for con-
temporary philosophers in particular: first, for metaphysics, and specifically
how Kant’s philosophy can be used to correct widespread confusions in
contemporary analytic philosophy (Hanna); and second, for ethics, and
specifically how certain limitations of Rawls’s Kantian constructivism can
be corrected with Wood’s metaethical realism, and vice versa (Rohlf).
Conclusion
Kant has had an enormous impact on intellectual history, and indeed on the
very ways that we understand ourselves and our experience. We struggle with
the implications of the Copernican revolution in philosophy whenever we
consider, for example, how the activity of thinking shapes our experience of
the world, the nature of moral constraint, or the possibility of shared
aesthetic judgments. Kant’s work is the focus of extensive scholarship, it is
invoked in a number of contemporary debates (in politics, ethics, art,
1 Introduction 17
religion, business, psychology, and more), and it remains a source of inspira-

tion for contemporary philosophers to develop their own views. If nothing
else, this volume testifies to the continuing vibrancy of Kant’s philosophy. It
engages previous work on Kant, provides a snapshot of the current state of
Kant studies, and establishes fertile ground for future research.
Part I
Biographical and Historical Background
2
Kant’s Life
Steve Naragon
However brief the sketch of Kant’s life, one generally learns at least three
things: he never married, he never traveled, and he ordered his life so rigidly
that the housewives of Königsberg could set their clocks to his daily walks. Like
the spare lines of a caricature, these provide some sense of the man, but one
could also point out in reply that Kant was by no means the first bachelor in
the history of philosophy nor the only bachelor in Königsberg; that Kant
actually managed to get out of town quite often, if not very far; and that only
in his later years, when it appeared that his life-ambitions were outstripping his
life, did he buckle-down and fashion himself into something more like a
machine. The following will add some shading to those spare lines.1
1
I have drawn primarily from the following early sources: Johann Christoph Mortzfeld, Fragmente aus
Kants Leben: Ein biographischer Versuch (Königsberg: Hering und Haberland, 1802); Ludwig Ernst
Borowski, Darstellung des Lebens und Charakters Immanuel Kants, Von Kant selbst genau revidirt und
berichtigt (Königsberg: Nicolovius, 1804); Reinhold Bernard Jachmann, Immanuel Kant geschildert in
Briefen an einen Freund (Königsberg: Nicolovius, 1804); Ehregott Andreas Christoph Wasianski,
Immanuel Kant in seinen letzten Lebensjahren: Ein Beitrag zur Kenntnis seines Charakters und häuslichen
Lebens aus dem täglichen Umgange mit ihm (Königsberg: Nicolovius, 1804); Johann Gottfried Hasse,
Letzte Äusserungen Kants von einem seiner Tischgenossen (Königsberg: Nicolovius, 1804); Friedrich
Theodor Rink, Ansichten aus Immanuel Kants Leben (Königsberg: Goebbels und Unzer, 1805); and
material gathered in 1804 for Samuel Gottlieb Wald’s memorial address for Kant, but first published in
Rudolf Reicke, Kantiana: Beiträge zu Immanuel Kants Leben und Schriften (Königsberg: Theile, 1860).
Emil Arnoldt assessed the above and other material in his “Kants Jugend und die fünf ersten Jahre seiner
S. Naragon (*)
Department of Philosophy and Religious Studies, Manchester University,
North Manchester, USA
e-mail: ssnaragon@manchester.edu

22 S. Naragon
Kant’s early education

Immanuel Kant (1724–1804), arguably the most important Western philoso-
pher since the Middle Ages, was born into a home of very modest means, at the
far eastern end of the Baltic Sea in the bustling port city of Königsberg (now the
Russian city of Kaliningrad). He was the fourth child and first surviving son in his
family, with one older and three younger sisters and a younger brother who
survived into adulthood.2 He was born the same year as Königsberg itself, which
had been formed by consolidating the three medieval towns of Kneiphof (a small
island located where the New and Old Pregel rivers met), Altstadt (to the north),
and Löbenicht (to the east). The Kants lived in that part of Königsberg given over
to members of the harness and saddle guilds, in the Vorderste Vorstadt, just south
of the Kneiphof island. At the east end of this island sat the fourteenth-century
red brick cathedral, or Domkirche, where Kant’s parents had been married and
where Kant and his siblings were later baptized. On the north side of this
cathedral stood a few low buildings that made up the Academia Albertina, a
Lutheran university founded in 1544 in the wake of the Protestant Reformation.
Kant received his first taste of student life at a German school in his
neighborhood, where a single teacher gave instruction on reading, writing,
arithmetic, and Christianity. There were a dozen of these so-called “church
schools” – all Lutheran – scattered around the city and always affiliated with
a church overseeing the instruction. A few of them were bilingual, offering
instruction in German as well as in Polish or Lithuanian; one school was
entirely in French, to serve the small community of French Huguenots; and
another served the children from Catholic homes.3
Kant would not have attended this German school for many years, since
he enrolled at the Collegium Fridericianum – a Pietist Latin school – when he
was eight. This was at the urging of Franz Albrecht Schultz, a forty-year-old
Privatdocentur,” Altpreussische Monatsschrift 18 (1881): 606–86. Still definitive is Karl Vorländer,

Immanuel Kant, der Mann und das Werk, 2 vols. (Leipzig: Meiner, 1924); and the more recent (and
in English) Manfred Kuehn, Kant: A Biography (Cambridge: Cambridge University Press, 2001), both
of which I have made constant use. Finally, an excellent brief developmental summary of Kant’s writings
can be found in Paul Guyer, Kant (London: Routledge, 2006), ch. 1. An earlier version of the present
essay appeared as “Kant’s Career in German Idealism,” in The Palgrave Handbook of German Idealism,
ed. Matthew C. Altman (London: Palgrave Macmillan, 2014), 15–33.
2
Vorländer, Immanuel Kant, 2:385. The Kant family papers are printed in Arnoldt, “Kants Jugend,” 608.
3
Johann Friedrich Goldbeck, Nachrichten von der Königlichen Universität zu Königsberg in Preußen, und
den daselbst befindlichen Lehr-, Schul- und Erzeihungsanstalten (Leipzig: Buchhandlung der Gelehrten,
1782), 254–55; and Fritz Gause, Die Geschichte der Stadt Königsberg in Preussen, 3 vols., 2nd ed. (Köln:
Böhlau, 1996), 2:126.
2 Kant’s Life 23
Pietist theologian and pastor who had arrived in Königsberg just the year
before, and who must have noticed an intelligence in the young boy that
deserved better cultivation. Schultz was the consistory advisor and pastor at
the Altstadt Church, which managed the German school Kant was attending,
and it was perhaps this connection that led Kant’s mother to attend Schultz’s
Bible study classes, to which she brought her older children, and that also led
to Schultz’s occasional visits to the Kant home.4
Local children hoping to attend the university and too poor for a private tutor
needed to study at one of the three Latin city schools or else at the Collegium
Fridericianum, which had beds for about seventy-five boarders, and which also
accepted day-students, of which Kant was one. From Easter 1732 (having just
turned eight) until Michaelmas 1740 (when he was sixteen), Kant walked across
town each day to attend this school. Classes began every morning at 7 a.m. and
ended at 4 p.m., with time set aside for play and for worship. He attended as a
charity student, and without Schultz’s intervention he likely would not have gone
at all. Kant was grateful to Schultz for this,5 however much he disliked his years
there – Kant’s friend Hippel later wrote that “terror and fear would overcome
him as soon as he thought back to the slavery of his youth”6 – and his extreme
distaste for institutional religion likely began at this time, as well.
Kant’s studies at the Collegium included Latin and theology for all seven-
teen of his semesters there, as well as Greek for at least ten semesters and
Hebrew for eight, French for six, handwriting for eleven (at one point he fell
back a level), singing for six, geography for at least four, history for three,
antiquities for five, poetry for four, arithmetic for nine, mathematics for two,
and philosophy beginning in his next to last year.7
The curriculum included nothing from the natural sciences, nor was there
any study of modern literature. Instruction in Hebrew and Greek focused on
Bible translation, with no classical Greek works. Of more relevance to Kant’s
later philosophical career was the study of Cicero in the context of the Latin
class. Kant excelled in Latin, becoming a fine stylist, and “even as an old man
4
Vorländer, Immanuel Kant, 1:20–21.
5
Borowski claims that Kant hoped to erect a monument in Schultz’s honor; see his biography (Darstellung,
152) and his notes for Wald (quoted in Reicke, Kantiana, 31). In the latter he refers to Schultz as one of Kant’s
valued teachers at the Collegium Fridericianum, alongside Kant’s Latin teacher, Johann Friedrich Heydenreich.
6
Reported by Theodor Gottlieb von Hippel in his autobiography and reprinted in Friedrich
Schlichtegroll, Biographie des Königl. Preuß. Geheimenkriegsraths zu Königsberg, Theodor Gottlieb von
Hippel, zum Theil von ihm selbst verfaßt (Gotha: Perthes, 1801), 78–79.
7
On Kant’s experiences here, see Heiner F. Klemme, Die Schule Immanuel Kants: Mit dem Text von Friedrich
Schiffert über das Königsberger Collegium Fridericianum, 1741 (Hamburg: Meiner, 1994), esp. 32–60.
24 S. Naragon
recited the most beautiful passages of Latin poets, orators, and historians.”8
But in general, as Kant once mentioned to a former classmate, “any sparks in
us for philosophy or math could not be blown into a flame by those men,” to
which the classmate replied: “But they were good at blowing them out.”9
Pietism and rationalism in Königsberg

Kant came of age in the wake of two strong challenges to orthodox Lutheran
theology: Pietism and rationalism. Pietism was a revivalist, anti-intellectual
movement within the context of the German Lutheran church that was
inspired by Philipp Jakob Spener (1635–1705) and his Pia Desideria
(1675), and it was similar to the Methodism that was sweeping England.
In the words of Isaiah Berlin:
[Pietism] laid stress on the depth and sincerity of personal faith and direct
union with God, achieved by scrupulous self-examination, passionate, intensely
introspective religious feeling, and concentrated self-absorption and prayer,
whereby the sinful, corrupt self was humbled and the soul left open to the
blessing of divine, unmerited grace.10
Kant’s family belonged to this Pietist movement, as did all his instructors at
school, so this religion deeply informed the first sixteen years of Kant’s life.
The spiritual center for Pietism at the time was the university at Halle, the
largest of the four Prussian universities (the others were Königsberg,
Frankfurt/Oder, and Duisburg). August Hermann Francke (1663–1727)
had studied under Spener at Dresden and then brought the movement to
Halle, and he was instrumental in helping Friedrich Wilhelm I (who reigned
from 1713 to 1740) install Pietists at the university in Königsberg.
Rationalism was championed by Christian Wolff (1679–1754), also at Halle,
who developed a scholasticism consistent with the scientific advances of his day.
He viewed the special revelation of scripture as consistent with, but separable
from, the natural revelation of rational theology. Just a year before Kant was born,
the Pietists had convinced the king to expel Wolff from Halle, having been
particularly scandalized by the rectoral address Wolff gave on July 12, 1721 –
8
Wald, quoted in Reicke, Kantiana, 6.
9
Borowski, Darstellung, 161–62.
10
Isaiah Berlin, Three Critics of the Enlightenment: Vico, Hamann, Herder, ed. Henry Hardy (Princeton:
Princeton University Press, 2000), 258.
2 Kant’s Life 25
“On the Moral Philosophy of the Chinese” – in which he argued that Chinese (i.
e., Confucian) and Christian ethics were fundamentally the same, and thus that
ethics as such was not in need of a special Christian revelation.11
This struggle between the Pietists and the rationalists played itself out in
Königsberg as well, but a peculiar blend of these two forces also emerged.
Schultz had studied under both Francke and Wolff in Halle, and he managed
to reconcile these seemingly antagonistic positions. He was sent to
Königsberg precisely because of his Pietism, assumed the directorship of
the Collegium Fridericianum shortly after arriving in town, and eventually
helped oversee all the churches and schools. Yet in his inaugural dissertation
he argued that faith and reason can be harmonized, and that Wolff’s
philosophy is acceptable and even useful for the faith.
Wolff himself had maintained that “if anyone has ever understood him, it
is Schultz in Königsberg.”12 Kant’s student and later close friend, T. G. von
Hippel (1741–1796), studied theology under Schultz and wrote that he
“taught me theology from a different perspective, bringing in so much
philosophy that one was led to believe that Christ and his Apostles had all
studied in Halle under Wolff.”13
Martin Knutzen (1713–1751), under whom Kant would later study, had
nearly completed his own studies when Schultz arrived in Königsberg, and
under his influence soon developed much the same blend of rationalism and
Pietism – what Erdmann described as a Pietist content of divine revelation
trussed up in the Wolffian form of definitions, theorems, and lemmas.14
Both of these men were of considerable importance for Kant, shaping the
intellectual backdrop of his early years as a student at the university.
Kant’s university studies

Kant’s transition into university life must have been exciting. Near the end of
his last term at the Collegium Fridericianum, in the summer of 1740, the old
king died and on July 20 his son arrived in town to be installed as the new
11
Lewis White Beck, Early German Philosophy: Kant and His Predecessors (Cambridge: Harvard
University Press, 1969), 258–59.
12
Wald, quoted in Reicke, Kantiana, 6; a nearly verbatim quote is given by Hippel in Schlichtegroll,
Hippel, 160.
13
Hippel, quoted in Schlichtegroll, Hippel, 162.
14
Benno Erdmann, Martin Knutzen und seine Zeit: Ein Beitrag zur Geschichte der Wolfischen Schule und
insbesondere zur Entwicklungsgeschichte Kants (Leipzig: Voss, 1876), 116.
26 S. Naragon
king, Friedrich II (later dubbed “the Great”), beginning what was to become
a forty-six-year reign promoting Enlightenment ideals throughout the land,
and particularly in the universities. Wolff was coaxed back to Halle from his
chair in Marburg, and Pietists everywhere were put on notice. Two months
after the coronation, Kant matriculated at the Academia Albertina. He was
sixteen years old, a standard age for such beginnings.
Given Kant’s meager financial resources (he received some help from an
uncle), it is striking that he never took a teaching position at his old Latin
school to support himself, as many other poor students did (including
Herder, thirty years later). Were his experiences there such that he could
not bring himself to re-enter that world? Nor is there any evidence that Kant
ever applied for one of the many stipendiums available to poorer students.15
If he lived at home with his father and younger siblings (his mother had died
three years earlier), it was not for long, as we read of him living with friends
and making small amounts on the side; for instance, he was in demand as a
tutor. Christoph Friedrich Heilsberg, a student friend, wrote that Kant
promised me his assistance, gave me books on the newer philosophy, and

helped me with the lectures by Professors Ammon, Knutzen, and Teske, at
least the hardest parts; all of this was done out of friendship. In the meantime
he tutored several students for a small fee, which each gave to him readily.16
Kant also made money playing billiards and cards:
His only recreation was playing billiards, a game in which Wlömer and I were his
constant companions. We had nearly perfected our game, and rarely returned
home without some winnings. I paid my French teacher altogether from this
income. As a consequence, persons refused to play with us, and we abandoned this
source of income, and chose instead L’Hombre, which Kant played well.17
The little we know of Kant’s studies is that he attended lectures by C. F. Ammon

(mathematics), J. G. Teske (experimental physics), Knutzen (mathematics, logic,
metaphysics, and moral philosophy), and Schultz (theology).18 He may also have
15
Arthur Warda, “Ergänzungen zu E. Fromms zweitem und drittem Beitrage zur Lebensgeschichte
Kants,” Altpreussische Monatsschrift 38 (1901): 402.
16
Heilsberg, quoted in Reicke, Kantiana, 48.
17
Ibid., 49.
18
Ibid., 48; and Borowski, Darstellung, 28. See also Manfred Kuehn, “Kant’s Teachers in the Exact
Sciences,” in Kant and the Sciences, ed. Eric Watkins (Oxford: Oxford University Press, 2001), 11–30.
2 Kant’s Life 27
attended J. D. Kypke’s lectures on logic and metaphysics, since they were free.
Rink reports that he took classes for “about three years.”19
Kant’s relationship with Knutzen is a puzzle. Kant’s early biographers
describe a close mentoring relationship between Knutzen and Kant, and
most accounts since have repeated and embellished this. According to
Ludwig Ernst Borowski (1740–1831), one of Kant’s early students and ear-
liest biographers, Knutzen “was the teacher with whom Kant felt most
connected. He attended all his courses on philosophy and mathematics
without interruption. . . . Knutzen . . . found in Kant splendid talents . . . even-
tually loaning him works by Newton.”20 Yet when Knutzen mentions his
better students, he does not mention Kant; nor does Kant mention
Knutzen.21 Kant’s first work (Thoughts on the True Estimation of Living
Forces [1746–1749; LF 1:3–181]), which Borowski describes as Knutzen-
inspired, was dedicated not to Knutzen but to J. C. Bohl, a professor of
medicine at the university. Insofar as Knutzen had a favorite student, it was
Friedrich Johann Buck (1722–1786), the same Buck who in 1759 would be
given the Professorship in Logic and Metaphysics, instead of Kant.22 So the
relationship between Knutzen and Kant could not have been very close; what
Kant did receive from him appears to have been this: an introduction to
Wolffian metaphysics and to Newtonian science.
Becoming professor Kant

“I do not intend to teach philosophy, but rather how to philosophize.”23
Kant repeated this sentiment throughout his teaching career. He sought to
19
Rink, Ansichten, 27.
20
Borowski, Darstellung, 28, 29, 163–64. See also Jachmann, Immanuel Kant, 10; and C. J. Kraus,
quoted in Reicke, Kantiana, 7.
21
See Hans-Joachim Waschkies, Physik und Physikotheologie des jungen Kant: Die Vorgeschichte seiner
Allgemeinen Naturgeschichte und Theorie des Himmels (Amsterdam: Grüner, 1987), 20n4.
22
Both Erdmann (Martin Knutzen) and Waschkies (Physik und Physikotheologie) promote the view of a
close relationship between Knutzen and Kant. A more skeptical position is found in Kuehn, Kant, 78–
84; and Kuehn, “Kant’s Teachers,” 22–23.
23
From a recently discovered reflection of Kant’s, reproduced in Steve Naragon and Werner Stark, “Ein
Geschenk für Rose Burger. Notizen und Hinweise zu einem neu aufgefundenen Kant-Blatt,” Kant-
Studien 104, no. 1 (March 2013): 5. In the context of discussing Kant’s logic lectures, Jachmann wrote
that “it was never his intention merely to recite a logic to his listeners, but rather to teach them to think”
(Immanuel Kant, 28–29). See also the Dohna logic lectures of 1792: “Not to learn philosophy – but
rather to learn to philosophize, otherwise it remains only imitation” (DWL 24:698). See also Kant’s
discussion of the “zetetic method” of the Announcement (1765; APL 2:307).
28 S. Naragon
help his students master an activity, rather than a set of dogmas – how to
think, rather than what to believe. Kant “compelled his hearers to think for
themselves,” according to Herder, who studied with Kant in the early 1760s.
Even during Kant’s first semesters as an instructor, Kant would “always
remind us that he would not teach philosophy, but rather how to philoso-
phize, etc. . . . To think for oneself – to investigate for oneself – to stand on
one’s own feet – were expressions he uttered constantly.”24
Kant was the first major modern philosopher to spend his life teaching at a
university, and most of his immediate followers sought to make that their
home as well. When Kant decided on this academic path is unclear, but it
appears to have come rather late. Most students in the eighteenth century
took classes for two to three years without seeking a degree; only those
wishing to teach at the university needed one, either a doctorate to teach
in theology, medicine, or law, or a master’s degree to teach in the philosophy
faculty.25
Kant stayed at the university, or at least with his friends in Königsberg, for
eight years, until poverty forced him to leave in the summer or fall of 1748 to
serve as a private tutor in the countryside, which he did for about five years26 –
three years in one home, and two in a second – working with young boys
ranging in age from seven to fourteen. These tutoring positions were usually
taken by young theology students waiting on their first church appointment,
although this clearly was not Kant’s situation, who listed himself at the time as
a “student of philosophy.”27
24
Borowski, Darstellung, 84, 188. The same sentiment is found in Kant’s essay “An Answer to the
Question: What Is Enlightenment?” (1784; WE 8:35), and at the end of “What Does It Mean to Orient
Oneself in Thinking?” (1786): “Thinking for oneself means seeking the supreme touchstone of truth in
oneself (i.e. in one’s own reason); and the maxim of always thinking for oneself is enlightenment” (OT
8:146n).
25
Most universities were arranged with four faculties: philosophy, plus the three higher faculties of
theology, law, and medicine. Graduation was possible from any of them. A graduate of the philosophy
faculty became a Doctoris Philosophiae seu Magistri – a “doctor of philosophy or magister,” although he
was commonly called a “Magister” (designated by an “M.” printed before one’s name) and a graduate
from any of the higher faculties was called a “Doctor” (designated by a “D.” printed before one’s name).
The words themselves are nearly synonymous: a magister is one who teaches, and a doctor is one who
has been taught.
26
Waschkies, Physik und Physikotheologie, 25–27, offers the best evidence for this disputed date
of Kant’s departure. This early chronology is contested. For an account, see “The Hofmeister” at
www.manchester.edu/kant/Students/studentHofmeister.htm.
27
Bernard Haagen, “Auf den Spuren Kants in Judtschen. Ein Beitrag zur Lebensgeschichte des
Philosophen,” Altpreußische Monatsschrift 48 (1911): 382–411, 528–56; at 390.
2 Kant’s Life 29
Of the eight years between entering the university and leaving

Königsberg, likely no more than three years were spent attending lec-
tures, so how did Kant spend those remaining five years? Had he been
pursuing an academic career, he would have written a Latin dissertation
to present to the philosophy faculty, a prerequisite for graduation, but
Kant did not do this. Instead he wrote a book on what he took to be an
important physics problem of his day, and he wrote it in German for
publication outside the university. This first publication – written
between 1744 and 1747 and with printing concluded in 1749 – was a
240-page work that attempted to reconcile the Cartesian and Leibnizian
accounts of force (Living Forces [1746–1749; LF 1:3–181]),28 and this
was followed by a book nearly as long that offered a Newtonian account
of the formation of the universe, also in German and published in 1755
(Universal Natural History and Theory of the Heavens [UNH 1:217–
368]). Apart from that, Kant wrote a few shorter pieces for the local
paper in the summer of 1754, perhaps after his return to Königsberg, on
the rotation and age of the earth.
This scholarly activity suggests that Kant was looking to create a
name for himself outside of academia, a path certainly in keeping with
the careers of many of the individuals with whose ideas he was engaged:
Descartes, Newton, Leibniz, and many of their followers.29 And yet he
returned to Königsberg in the summer of 1754, perhaps by then with
the intention of teaching. He submitted his master’s thesis (“Succinct
Exposition of Some Meditations on Fire” [NS 1:371–384]) the follow-
ing spring on April 17, 1755, sat for the oral exam on May 13, and
received his degree on June 12. He was then required to submit a
second Latin thesis to receive the privilege of teaching at the university.
This work – New Elucidation of the First Principles of Metaphysical
Cognition (NE 1:387–416) – was publicly defended on September 27.
And thus did Kant’s teaching career at the university begin. It lasted
almost forty-one years.
28
This matter had already been resolved by Jean le Rond d’Alembert in 1743, although Kant had not
heard the news, which presumably had not yet reached Königsberg.
29
Kant was also studying the work of academics such as Wolff, who taught at Halle and Marburg, and
Crusius, who taught at Leipzig. But Lambert and Maupertuis were outside the university, as was Euler
(other than for a brief stint at St. Petersburg). Hume and Rousseau were non-academics, but Kant did
not read them until later.
30 S. Naragon
Kant’s lectures and his students

Kant’s life was shaped by the rhythms of the academic year, which in
Prussian universities was divided by Michaelmas (September 29) and
Easter (fluctuating between March 22 and April 25), with a new semester
beginning about two weeks after each of these dates: winter semester the
second week of October, and summer semester sometime in April or May.
Most classes – and all “public lecture” classes – met four times a week.
The philosophy faculty in eighteenth-century Königsberg included eight
full professors, the occasional associate professor, and a fluctuating number of
unsalaried lecturers (Privatdozenten) whose only remuneration came at the
end of the semester, collected directly from the students and normally at the
rate of four thaler per head.30 Only two of the full professorships were what
today would be thought of as philosophy, namely, logic/metaphysics and
practical philosophy; the others were mathematics, physics, Oriental lan-
guages, rhetoric/history, Greek, and poetry. These salaried professors were
required to offer “public lectures” (normally one each semester, and these
students attended for free), but they also would offer various “private
lectures” (alongside the Privatdozenten), which were paid for by the students
directly.
Kant taught as a lecturer for twenty-nine semesters (1755 to 1770) before
finally receiving the professorship in mathematics, which he quickly
exchanged for the professorship in logic and metaphysics,31 teaching in
that capacity for fifty-three semesters (1770 to 1796).32
The records are incomplete, but Kant appears to have taught logic most often
(56 times), followed by metaphysics (53) and physical geography (49). He
taught these three courses nearly every semester until he became a full professor,
after which he taught each of them once a year. His first course on anthropology
30
Goldbeck, Nachrichten, 102, estimates student living expenses (room, board, and firewood) at sixty
thaler per year, so these tuition fees for private lectures were not trivial. For comparison, Kant’s starting
annual professor’s salary in 1770 was 166 2/3 thaler; see the Cabinet order from King Friedrich II of
March 31, 1770 (reprinted at Ak 10:94).
31
F. J. Buck held the logic/metaphysics position – he had been preferred over Kant in 1759 when J. D.
Kypke’s death made available the position – and Kant petitioned the government to give the vacant
mathematics chair to either K. A. Christiani (1707–1780), the current professor of moral philosophy, or
to Buck, with the resulting vacancy to go to Kant. The king chose the latter course, but it appears from
his letter that Kant would have been equally comfortable assuming either chair (moral philosophy or
logic/metaphysics).
32
Most of the data regarding Kant’s teaching comes from Emil Arnoldt, Gesammelte Schriften, ed. Otto
Schöndörffer, 10 vols. (Berlin: Cassirer, 1906–1911), vols. 4–5.
2 Kant’s Life 31
was given in 1772–1773, and every winter semester thereafter (for a total of 24
semesters). These four courses formed the core of his teaching as a full professor,
with metaphysics and anthropology offered in the winter, and logic and
physical geography in the summer. Kant also offered private lectures in mathe-
matics nearly every semester at the beginning of his career, but abruptly stopped
after 1763–1764 (15 semesters total). Theoretical physics (21) and moral
philosophy (28) were alternated during much of his career, along with natural
law (12), which he first taught in 1767, and philosophical encyclopedia33 (10),
which he first taught in 1767–1768. Occasional courses were given on natural
theology (4) and pedagogy (4).
Kant and his early biographers claim that his classrooms were always well-
attended, and the records tend to bear this out, but not always. Other well-
regarded instructors were teaching the same courses as Kant – for instance,
there might be four or five private courses on metaphysics, apart from the
public course offered by the full professor – and yet there were fewer than
400 students enrolled at the university.34 The competition for students must
have been intense, especially among the unsalaried lecturers whose income
was entirely tuition-based. Of the courses for which we have records, almost
10% of Kant’s private offerings were cancelled due to low enrollment.
Kant was nonetheless an engaging lecturer. Christian Friedrich Jensch
heard Kant during the winter semester of 1763–1764:
How interesting Kant was in his lectures. He would enter the room in a sort of
enthusiasm, saying: we left off here or there. He had memorized the main ideas
so deeply and vividly that the entire hour was lived in these alone; often he took
little notice of the textbook over which he was lecturing.35
Kant would bring his copy of the required textbook with him to class36 and
sometimes notes on loose sheets of paper, although many of his textbooks were
33
This was an introductory course that surveyed the philosophical disciplines (logic, metaphysics,
practical philosophy) and their history.
34
Franz Eulenburg, Die Frequenz der deutschen Universitäten von ihrer Gründung bis zur Gegenwart
(Leipzig: Teubner, 1904), 296.
35
Jensch, quoted in Johann Friedrich Abegg, Reisetagebuch von 1798 (Frankfurt: Insel, 1976), 251.
36
By government decree, professors had to use an approved textbook in each of their courses. Purgstall
describes the tattered condition of Kant’s copy of Meier’s logic text (this was near the very end of Kant’s
teaching career): “He always brings the book along. It looks so old and soiled, I believe that he has
brought it daily to class with him for forty years. All the blank leaves are covered with writing in a small
hand, and besides, many of the printed pages have leaves pasted on them, and lines are frequently
crossed out, so that, as you might imagine, scarcely anything of Meyer’s Logic is left.” Even by the early
32 S. Naragon
bound with interleaved blank pages, giving ample room for writing down his
notes. Over the years these pages were entirely filled, yet he rarely read from
these notes or the textbook, instead engaging the author in a conversation and
using the text as an organizing principle and as a springboard for his own ideas.
We have more reports of the lectures in his later years, and these were
mixed. C. F. Reusch, a son of the physics professor, attended Kant’s lectures
in the mid-1790s:
To a young man of 15–16 years under those circumstances, not much of his
philosophical lectures could be put into a context that made them under-
standable; what I grasped was an occasional illuminating point or spark in the
soul. I don’t believe that it went any better at that time with the older students.
In contrast, his physical geography lectures were quite understandable, even
highly intellectually stimulating and entertaining.37
Another student from the same period, an Austrian nobleman by the name of
W. J. G. von Purgstall (1773–1812), wrote that Kant’s “presentation is
entirely in the tone of ordinary speech and, if you will, not very beau-
tiful. . . . Yet even though he does not look all that great, even though his
voice is unclear, everything that his delivery lacks in form is richly replaced by
the excellence of the content,” and then adds how helpful Kant’s lectures
were for understanding his published works:
One never leaves his auditorium without bringing home some elucidating hint
into his writings, and it is as though one arrived at the easiest and shortest way
to understanding many difficult sentences in the Critiques of Pure and
Practical Reason. . . . He simply enters directly into the subject and talks
about it as though he would never dream the materials could be very difficult,
and wholly convinced that anyone could understand it.38
Kant often chose a student from the audience to look in the eye while lecturing,
using this as a gauge of how well he was being understood. He must have found
many of those eyes discouraging. Borowski studied with him during his earliest
years and noted that “a lively attentiveness was always required. Without this his
1760s, Jensch reported that Kant’s copy of Baumgarten’s Metaphysica was “covered with notes all over”
(Karl Hugelmann, “Ein Brief über Kant,” Altpreussische Monatsschrift 16 [1879]: 608–9).
37
Christian Friedrich Reusch, Kant und seine Tischgenossen: Aus dem Nachlaß des jüngsten derselben, des
Geh. Ob.-Reg.-Rats Dr. Chr. F. Reusch (Königsberg: Tag & Koch, 1848), 291–92.
38
From a letter written by Purgstall and quoted in Hugelmann, “Ein Brief über Kant,” 608–9.
2 Kant’s Life 33
lectures couldn’t be understood, and one would get lost.”39 Kant was generally
hard to understand – in content, and sometimes in delivery – and students were
advised to take his easier classes first (physical geography, anthropology, moral
philosophy) or else begin with an easier professor.40
Kant’s interests and where they led

Two things fill the mind with ever new and increasing admiration and
reverence, the more often and more steadily one reflects on them: the starry
heavens above me and the moral law within me. (CPrR 5:161)
Kant’s philosophical project, broadly understood, was to reconcile the phy-

sical and moral worlds – the world of Newtonian mechanics with the world
of persons – and doing this required some hard and innovative work in
metaphysics. The lines quoted above, and found on a plaque once adorning
Kant’s tomb in Königsberg, come from the end of his Critique of Practical
Reason (1788). The passage continues:
I do not need to search for them and merely conjecture them as though they
were veiled in obscurity or in the transcendent region beyond my horizon; I see
them before me and connect them immediately with the consciousness of my
existence. (CPrR 5:161–62)
These two worlds were basic facts for Kant and were captured by the names
of Newton and Rousseau. Kant encountered Newton while still a teenager in
Knutzen’s lecture hall; he read Rousseau twenty years later in the early 1760s,
and this second encounter was just as transformative as the first. Rousseau
was for Kant a second Newton, as suggested in a remark written into his copy
of Observations on the Feeling of the Beautiful and Sublime (1764):
Newton saw for the first time order and regularity combined with great
simplicity, where before him was found disorder and barely paired multiplicity;
39
Borowski, Darstellung, 85, 185–86.
40
Wald, quoted in Reicke, Kantiana, 18. Hippel, who matriculated at the university in 1756, wrote that
he took the less challenging courses from Buck before attending Kant’s lectures (Hippels sämmtliche
Werke, vol. 12: Hippels Leben, ed. Gottlieb Hippel [Berlin: Reimer, 1835], 91). Kant was aware of these
difficulties and encouraged students to attend K. L. Pörschke’s lectures first in preparation (Jachmann,
Immanuel Kant, 30).
34 S. Naragon
and since then comets run in geometrical courses. Rousseau discovered for the
first time beneath the multiplicity of forms human beings have taken on their
deeply buried nature and the hidden law by the observation of which provi-
dence is justified. . . . After Newton and Rousseau, God is justified. (OBSn 9
[Ak 20:58–59])
Another remark suggests how Rousseau transformed Kant’s moral landscape:
I feel a complete thirst for knowledge and an eager unrest to go further. . . .

There was a time when I believed that this alone could constitute the honor of
mankind, and I had contempt for the rabble who know nothing. Rousseau
brought me around. This blinding superiority disappeared, I learned to honor
human beings. (OBSn 7 [Ak 20:43–44])
Unfortunately, these two worlds – Newton’s physical world of material

bodies understood with and governed by causal laws, and Rousseau’s
moral world in which each human, as a free and rational being, is of
inestimable worth living in a community of equals – are not easily held
together. For how is freedom possible in Newton’s universe? And with-
out freedom, what becomes of the moral universe? If every event in the
physical universe is the direct result of one or more previous events, then
every event happens necessarily and human freedom is an illusion. The
laws of nature appeared to leave no room for the concerns of morality
and the freedom it assumes.41
That was one puzzle confronting Kant. A second puzzle arrived at about
the same time in the form of David Hume, who famously argued that
physical causation is nothing more than a subjective sense of connection
between two events based on regularities encountered in the past, and that
there is no objectively necessary connection between these events, which are
themselves entirely “loose and separate.”42 Kant viewed this as a serious
41
Ten years earlier, Kant considered a similar challenge – this time from the side of Wolffian rationalism.
In A New Elucidation of the First Principles of Metaphysical Cognition (1755; NE 1:387–416), Kant
addressed this conflict between rationalism and human freedom, but at the time sided with Wolff over
Crusius’s “liberty of indifference” (NE 1:398–405). Kant eventually abandoned the rationalist account of
freedom – calling it “the freedom of a turnspit” in his Critique of Practical Reason (CPrR 5:97).
42
David Hume, Philosophical Essays Concerning Human Understanding [later editions: An Enquiry
Concerning Human Understanding] (London: Millar, 1748), §7 (“Of the Idea of Power or Necessary
Connexion”), pt. 2 (pp. 119–27). Kant first raised this worry about causal connection in his essay
Attempt to Introduce the Concept of Negative Magnitudes into Philosophy (1763; ANM 2:167–204): “I
fully understand how a consequence is posited by a ground in accordance with the rule of identity:
2 Kant’s Life 35
problem for the natural sciences, as these are meant to be systems of objective
and necessary causal laws. But if Hume is correct, such laws are merely
contingent empirical generalizations.
Kant’s intellectual life has traditionally been understood as falling into two
periods – the pre-critical and the critical – with the publication of his Critique of
Pure Reason (1781) marking the divide. These two puzzles fell on the pre-
critical side, and their solution marked the divide. The metaphysical doctrine
providing the solution, and which defined his new “critical philosophy,” was
what Kant called transcendental idealism, which holds that knowledge is
possible only when the mind partly constitutes the thing being known.
Kant arrived at transcendental idealism by asking a deceptively simple
question: “How is experience of an objective, public world possible?” He
concluded that this requires the mind to structure the experienced world:
first, by the sensibility passively receiving and shaping (as spatiotemporal) an
unknowable given, and second, by the understanding actively structuring
this spatiotemporal array into the world of physical objects. The mind is no
longer a passive recipient of sensations, but instead actively structures those
sensations into an objective world, and each mind does this, and does this in
the same way, resulting in a public, shared world.
Transcendental idealism redraws the boundary between the knowing subject
and the known object and, like any boundary, it has two sides, one humbling
and one affirming. The humbling side limits our knowledge claims to the world
of appearances (the phenomenal world), denying that we are capable of spec-
ulative insight into reality, thus humbling traditional metaphysics into silence.
The affirming side reminds us that this phenomenal world that we can know
just is, after all, the spatiotemporal world of material objects in which we live
and play and pursue science. What is more, transcendental idealism shows us
that at least some propositions about this world are a priori knowable, namely,
the formal part contributed by the knowing self. We cannot have a priori
knowledge of any particular causal laws, but we can know a priori that such laws
exist to be discovered empirically. Thus Kant’s two puzzles are solved, for
although the phenomenal world is entirely law-governed by causal relations,
it is at least possible that there exists a noumenal (real) self that is free and thus
that morality is possible.
analysis of the concepts shows that the consequence is contained in the ground. . . . But what I should
dearly like to have distinctly explained to me, however, is how one thing issues from another thing,
though not by means of the law of identity” (ANM 2:202).
36 S. Naragon
In planning out his new system, Kant had imagined writing a methodo-
logical propaedeutic (which turned into the Critique of Pure Reason) followed
by a metaphysics of nature and a metaphysics of morality.43 The former
appeared in 1786 as the Metaphysical Foundations of Natural Science (MFS
4:467–565), while the latter did not appear until 1797 as the two-part
Metaphysics of Morals (MM 6:205–355, 373–493): the Doctrine of Right
(concerning the nature of law and the state) and the Doctrine of Virtue
(concerning the system of moral duties that bind individuals). Kant’s best
known and most closely studied work on moral philosophy was also the first
that he published: the relatively short Groundwork of the Metaphysics of
Morals (1785; G 4:387–463), in which he introduced the concept of the
categorical imperative and made autonomy a central feature of how we
understand morality.
Responses to the Critique

Kant’s new critical philosophy did not enjoy a promising start.44 The
Critique of Pure Reason is a hard read today and it was perhaps just as
hard for those living in Kant’s day and speaking his language. Some of
the brightest minds, such as Moses Mendelssohn and Johann Wolfgang
von Goethe, complained in all sincerity that they could not make sense
of it.
This poor reception helped motivate Kant to write a summary intro-
duction, the Prolegomena to Any Future Metaphysics (1783; Pro 4:255–
383), which was further shaped by an early anonymous review written by
the popular philosopher Christian Garve (and heavily edited by J. G. H.
Feder), which viewed the Critique as belonging to the tradition of
Humean skepticism and Berkeleyan idealism.45 While Kant had only
43
An early version of this plan can be found in Kant’s letter to J. H. Lambert (December 31, 1765) (C
10:56).
44
Apart from Kuehn’s biography (Kant), see also Manfred Kuehn, “Kant’s Critical Philosophy and Its
Reception – the First Five Years (1781–1786),” in The Cambridge Companion to Kant and Modern
Philosophy, ed. Paul Guyer (Cambridge: Cambridge University Press, 2006), 630–63; Karl Ameriks,
Kant and the Fate of Autonomy: Problems in the Appropriation of the Critical Philosophy (Cambridge:
Cambridge University Press, 2000); Brigitte Sassen, Kant’s Early Critics: The Empiricist Critique of the
Theoretical Philosophy (Cambridge: Cambridge University Press, 2000); and Frederick C. Beiser, The
Fate of Reason: German Philosophy from Kant to Fichte (Cambridge: Harvard University Press, 1987).
45
Published in the January 19, 1782 issue of the Göttingen Gelehrten Anzeigen.
2 Kant’s Life 37
respectful words for Hume’s work, he wished to sharply distinguish his

own brand of idealism from what he called the “dogmatic idealism” of
George Berkeley, and several additions in the 1787 second edition of the
Critique, such as the Refutation of Idealism, were responding to the
Garve/Feder review.
Not until K. L. Reinhold’s Letters on the Kantian Philosophy (1786–
1787)46 did the Critique begin to receive its proper audience, and
camps soon formed of Kantians and anti-Kantians. C. G. Schütz and
Gottlieb Hufeland’s Allgemeine Literatur-Zeitung (1785–1804), a daily
newspaper from Jena featuring book reviews, provided an early sympa-
thetic forum for the new Kantian philosophy, while the Wolffian J. A.
Eberhard published the Philosophisches Magazin (1788–1792), featuring
articles critical of Kant and in which Eberhard himself argued that
Kant’s “new” philosophy was, at best, a rehashing of Leibniz and
Wolff.47 Kant spent the 1780s and 1790s filling out his critical philo-
sophy and responding to critics, for which he often enlisted the aid of
colleagues. Of these, most notable was Johann Schultz (1739–1805), a
mathematics professor at Königsberg, whose review of Kant’s 1770
Dissertation was found by Kant to be so insightful that he later asked
for Schultz’s help in promoting the Critique of Pure Reason. In 1797
Kant publicly declared Schultz to be his most reliable expositor.48 Not
even Schultz was a blind follower, however, and problems he raised in
1785 with what is commonly viewed as a key section of the Critique of
Pure Reason, namely, the Transcendental Deduction, nearly led to a
falling out with Kant, but also encouraged him to revise that section
heavily for the second edition.49
46
Reinhold published his letters in installments in C. M. Wieland’s Teutsche Merkur (August 1786 to
October 1787), and Kant publicly thanked Reinhold in “On the Use of Teleological Principles in
Philosophy” (1788; TelP 8:160, 184).
47
Kant replied to Eberhard with On a Discovery whereby Any New Critique of Pure Reason Is to Be Made
Superfluous by an Older One (1790; NCR 8:187–251), and also enlisted his colleague Johann Schultz to
critically review Eberhard’s magazine.
48
Kant, “Against Schlettwein” (1797; Ak 12:367–68). See Johann Schultz, Erläuterungen über des Herrn
Professor Kant Critik der reinen Vernunft (Königsberg: Dengel, 1784); and Johann Schultz, Prüfung der
Kantischen Critik der reinen Vernunft, 2 vols. (Königsberg: Hartung, 1789).
49
Schultz’s criticisms are found in his anonymous review of J. A. H. Ulrich, Institutiones logicae et
metaphysicae scholae suae scripsit (Jena: Cröker, 1785), in Allgemeine Literatur-Zeitung (December 13,
1785), 247–49, translated into English in Sassen, Kant’s Early Critics, 210–14. See also Kant’s
Metaphysical Foundations of Natural Science (1786; MFS 4:467–565), where he publicly answers
Schultz, in part by demoting the Transcendental Deduction’s role (MFS 4:474–76).
38 S. Naragon
Kant’s watch, Mr. Green, and the art

of conversation
The best known Kant anecdote comes from Heinrich Heine, who began by
noting that “the life story of Immanuel Kant is difficult to describe, because
he had neither a life nor a story”:
He lived a mechanically ordered, almost abstract bachelor existence in a quiet

isolated lane in Königsberg, an old city on the north-eastern border of
Germany. In my opinion, not even the great cathedral clock there went
about its daily labor with less passion and more regularity than its compatriot
Immanuel Kant. Getting up, drinking coffee, writing, giving lectures, eating,
walking: everything had its set time, and the neighbors knew that it was exactly
half past three when Immanuel Kant, in his grey frock-coat, rattan cane in
hand, emerged from his front door and strolled in the direction of small
Lindenallee, which is still called “Philosopher’s Way” on his account.50
Another anecdote – likely from the late 1780s or early 90s – also involved Kant
and a clock, this time his own watch to which he had attached an amber pendant
containing an ancient fly. During a discussion with friends about past geological
epochs, Kant lifted up his watch and said to the fly: “How great our knowledge of
[those epochs] would be, if only you could tell us how things were in your time.”51
The amber pendant may have been nothing more than an innocent
decoration, but the symbolism is significant enough, as is the watch itself,
given Heine’s description of Kant’s clock-like existence. But Kant had not
50
Heinrich Heine, On the History of Religion and Philosophy in Germany and Other Writings, trans.
Howard Pollack-Milgate, ed. Terry Pinkard (Cambridge: Cambridge University Press, 2007), 79. The
claim that housewives set their clocks to Kant’s passing was likely inspired by Jachmann’s account of the
evenings Kant would spend at Green’s house, which always ended at the same time: “The group always
disbanded so punctually at 7 o’clock that I often heard the neighbors on the street say: It can’t yet be
seven, since Professor Kant has yet to pass by” (Immanuel Kant, 81–82). Heine’s claim regarding the
name of the walk sounds likely, but is false, as it bore that name long before Kant ever walked there; see
Christopher Hartknoch, Alt- und Neues Preussen, oder Preüssischer Historien (Frankfurt & Leipzig:
Hallervorden, 1684), 395.
51
Karl Gottfried Hagen, “Bemerkungen, die Entstehung des Bernsteins betreffend,” in Beiträge zur
Kunde Preußens, 8 vols. (Königsberg, 1821), 4:207–27; the Kant anecdote is in a note on p. 209.
Jachmann (Immanuel Kant, 147) mentions Hagen as one of Kant’s regular dinner guests. A photograph
of presumably the same watch was published in a German newspaper sometime after 1933, although the
amber pendant appears to have gone missing. The source of the clipping is unknown, but it was
reproduced in Lorenz Grimoni and Martina Will, eds., Immanuel Kant: Erkenntnis – Freiheit – Frieden
(Husum: Husum, 2004), 183.
2 Kant’s Life 39
always been like that, and any truth to Heine’s caricature can be attributed to
Kant’s association with his good friend Mr. Green, who was, in a sense,
Kant’s second watch.
Joseph Green (1727–1786) was an English merchant who had been living
in Königsberg for a number of years before he met Kant, quite by chance, in
what was probably the summer of 1765. They were in a group discussing the
conflict between Britain and its American colonies, with Kant vigorously
defending the American cause and roundly condemning the British, all of
which became too much for Green, who challenged Kant to a duel. Things
might have gone badly had not Kant explained his position so dispassionately
and convincingly that Green was not only pacified but also charmed, and the
two strolled off together as friends.52
The two bachelors remained close until Green’s death twenty years later, with
Kant regularly taking his meals at Green’s house. It was from Green that Kant
developed his later penchant for strict routine and for living one’s life according to
maxims, and their common acquaintance, T. G. von Hippel, the Königsberg
mayor, featured Green in his anonymous play, The Man Who Lived by the
Clock.53 Whatever truth there is in Heine’s famous image of Kant’s “mechani-
cally-ordered life” appears to have stemmed from this association with Green.
At least by the 1780s54 Kant was spending nearly every evening at Green’s,
returning home at seven, except on Saturdays, when he left at nine. They
were not dining alone, however, at least not usually. Jachmann described the
rhythm of these evening gatherings:
Kant went to Green’s every afternoon to find him asleep in his armchair. Kant
would sit down, meditate for a while and also fall asleep; the bank director
Ruffmann would usually come along and do the same, until finally Motherby
would arrive in the room at a certain hour and wake up the party, which would
then converse over the most interesting topics until seven o’clock.55
52
August Hagen, “Kantiana,” Neue Preußische Provinzial-Blätter 6 (1848): 8; and Jachmann, Immanuel
Kant, 77–79. The proposed dating of the event is argued for in Kuehn, Kant, 154–55.
53
Der Mann nach der Uhr, oder der ordentliche Mann, published in 1765, first performed in Hamburg
the following year (Hagen, “Kantiana,” 9–10). Jachmann, as well as Hagen, draws this connection to
Green (Immanuel Kant, 80).
54
As suggested by Hamann’s correspondence (quoted in Vorländer, Immanuel Kant, 2:28). C. J. Kraus
also notes that the daily gatherings occurred in later years, when Green’s gout prevented him from
leaving his house easily (Reicke, Kantiana, 60).
55
Jachmann, Immanuel Kant, 81.
40 S. Naragon
Kant made many new acquaintances there, including Green’s younger busi-
ness partner, Robert Motherby (1736–1801), and Kant eventually found
himself spending every Sunday (along with Green) in the Motherby home,
where Kant would play with the children.56
Green was companionable, well-regulated, and clearly possessed of a good
mind, as we are told that Kant had him read every line that went into the Critique
of Pure Reason (1781).57 Green’s death in 1786 was hard for Kant and appears to
have marked his life in various ways. He gave up all evening commitments after
that and confined himself to eating only a midday meal, which he now took in his
own home purchased a few years earlier, and perhaps it was Green’s death that led
Kant to have a kitchen installed in 1787 and to hire a cook, although he
continued his pattern of dining at Motherby’s each Sunday.58 For the first few
years Kant’s younger colleague, C. J. Kraus, the full professor of practical
philosophy, was a daily dinner guest, having agreed to split the costs with
Kant.59 Jachmann notes that there were usually one or two dinner companions,
and at most five, since the table could seat no more than six.60
Two memoirs – by Hasse (1804) and by Reusch (1848) – refer to Kant’s
dinner parties in their titles, underscoring the centrality of this aspect of
Kant’s life.61 Kant was above all else a social creature and a famous con-
versationalist, and several hours of each day were devoted to this practice,
normally around the table. Abegg wrote that “Kant could sit at the table until
seven or eight at night, so long as someone stayed to talk with him,”62 and all
topics of conversation were open at dinner, except philosophy. An Austrian
56
Borowski, Darstellung, 102; and Jachmann, Immanuel Kant, 51–52, 145, 148. The Motherbys had
eleven children in all. Jachmann later served as a tutor to one of the sons, William Motherby, who
attended Kant’s lectures in the early 1790s and later distinguished himself as a physician.
57
Jachmann, Immanuel Kant, 80.
58
Earlier, in the 1760s, Kant seems to have spent many of his Sundays just north of the city in Moditten
with the head forester Michael Wobser. This is where Kant stayed for a few weeks in 1763 while
composing Observations on the Feeling of the Beautiful and Sublime; see Dr. Michelis, “Kant – Hauslehrer
in Judtschen?” Kant-Studien 38, nos. 1–2 (Jan. 1933): 492–93.
59
This arrangement came to an end around 1789 after a falling out between the two men, on which see
Kuehn, Kant, 331–34. Kant continued with his dinners, of course, only without Kraus, and when they
were both invited to someone else’s home, “they never sat right next to each other, but also not very far
from each other” (Abegg, Reisetagebuch, 255–56).
60
Jachmann, Immanuel Kant, 146–47.
61
The title of Hasse, Letze Äusserungen, could be translated as “Kant’s last remarks, by one of his dinner
companions,” and Reusch, Kant und seine Tischgenossen, as “Kant and his dinner companions.”
62
Abegg, Reisetagebuch, 255. This topic might strike some as too trivial to discuss, but these dinners
were a central part of Kant’s life. Kuehn draws a plausible connection between this conversational form
of life and Kant’s critical philosophy (Kant, 273–74).
2 Kant’s Life 41
nobleman who studied Kant’s philosophy under Reinhold at Halle and then
traveled to Königsberg in 1795 to attend Kant’s lectures on logic and
physical geography, wrote that Kant
invited me to lunch every fourth day, and I’ve accepted a few times. He has not
yet spoken a word of theoretical philosophy, and I never try to bring it up, but
always let him guide the conversation, and it seems that he does not enjoy
discussing abstractions.63
In Anthropology from a Pragmatic Point of View (1798), Kant discusses the

natural tension between pursuing the moral and the physical good. Balancing
this tension is at the core of humanity, Kant argues, and nothing strikes the
balance quite so well as a good dinner party: “The good living that still seems to
harmonize best with true humanity is a good meal in good company” (An 7:278),
and earlier in that text Kant notes that “there is no situation in which sensibility
and understanding unite in one enjoyment that can be continued as long and
repeated with satisfaction as often as a good meal in good company” (An 7:242).
Kant loved to eat, but not alone – a practice “unhealthy for a scholar who
philosophizes,” as one then foregoes the differing ideas that dinner companions
provide (An 7:279–80). These dinners could be seen as examples of the public
sphere where one follows the maxims of the sensus communis, namely to think for
oneself (so that one might make a contribution to the conversation), to think
from the perspective of everyone else, and to think consistently (CJ 5:294). Kant’s
concern with the rules governing these dinners fell well beyond mere table
etiquette. For instance, the number of guests was to be such as could support a
single conversation – Kant recommended between three and nine – thus allowing
each to talk to all (An 7:278–79); dogmatic assertions were discouraged (An
7:281); and the discussion was to be held in a spirit of trust and discretion (An
7:279) – all with an eye toward a free and open exchange of ideas.64
In the early years, Kant dined in public inns when he lacked an invitation
into someone’s home, but these invitations came frequently and from an
impressive assortment of aristocrats, military officers, and government officials.
63
Quoted in Hugelmann, “Ein Brief,” 610. See also Hasse, Letze Äusserungen, 6–7. Nor did Jachmann
ever hear Kant mention any of his writings during these dinnertime conversations (Immanuel Kant,
137). Vorländer offers a close description of the rituals surrounding Kant’s dinner parties in his home
(Immanuel Kant, 2:297–300).
64
This idea is usefully explored in Alix A. Cohen, “The Ultimate Kantian Experience: Kant on Dinner
Parties,” History of Philosophy Quarterly 25, no. 4 (Oct. 2008): 315–36.
42 S. Naragon
Kant enjoyed a close relationship with the Keyserlings, especially with Countess
Charlotte, who sat him on her right side so long as no visiting dignitary required
that place of honor instead. Kant had a standing invitation to dinner at their
palace every Tuesday, a tradition continuing until near her death in 1791.65
When one reflects on Kant’s humble origins, it is remarkable how far he
had come in polite society. His years tutoring in the von Hülsen family (c.
1751–1753) were perhaps his first introduction to that society, after which
his visits with the Keyserlings, beginning after his return to Königsberg in
1754, completed the education. The Russian occupation of Königsberg that
soon followed (lasting between 1758 and 1762)66 opened up society and its
dinners and dances to a wider range of citizens, with professors now mingling
with the aristocratic and military classes. Kant thrived in this environment.
Completing the system: Kant’s third Critique

The Critique of the Power of Judgment (1790) was Kant’s most influential
work for the generation of philosophers that followed, bringing to center
stage the concepts of purpose and systematicity, and marking the emergence
of aesthetics as a serious philosophical discipline. Kant himself gave the book
a preeminence in his system, writing in the preface that “with this I bring my
entire critical enterprise to an end” (CJ 5:170).
When Kant published the Critique of Pure Reason in 1781, he did not have in
mind to write any more critiques: one was to be quite enough. But while working
up the second edition of this Critique (published in 1787), the material that he
was developing on practical reason – primarily his defense of God, freedom of the
will, and the immortality of the soul as “postulates of practical reason” – expanded
to the point that an entirely separate treatment was in order,67 which he then
published as the Critique of Practical Reason (1788; CPrR 5:1–164).
The origins of the third Critique are less straightforward, with the idea to
write it falling hard on the heels of working out the second. A letter to C. G.
65
C. J. Kraus, quoted in Reicke, Kantiana, 60. Johann Ludwig Schwarz was staying with the Keyserlings
in February or March of 1787 and mentions dining with Kant four of the five days he was there, so
clearly Kant was often more than just a Tuesday guest. See Schwarz, Denkwürdigkeiten aus dem Leben
eines Geschäftsmannes, Dichters und Humoristen (Leipzig: Kollmann, 1828), 180.
66
This was a peacefully conducted arrangement that occurred during the Seven Years War.
67
A note in the November 21, 1786, issue of the Allgemeinen Literaturzeitung claims that the new
edition of the Critique of Pure Reason would also include a “Critique of Pure Practical Reason” (Ak
3:556).
2 Kant’s Life 43
Schütz in June 1787 (C 10:490) indicated Kant’s intentions, and six months
later we find him writing in a letter to Reinhold:
I am now at work on the critique of taste, and I have discovered a new sort of a priori
principles. . . . For there are three faculties of the mind: the faculty of cognition, the
faculty of feeling pleasure and displeasure, and the faculty of desire. . . . This
systematicity put me on the path to recognizing the three parts of philosophy . . . :
theoretical philosophy, teleology, and practical philosophy. (C 10:514–15)
Teleological explanation, where one understands nature as a system of ends

or purposes rather than as a large clockwork following Newtonian laws, had
yet to be properly addressed by Kant.
The structure of the third Critique strikes the casual reader as simply odd, for
Kant appears to have published two books under a single cover: the first concerns
aesthetics (the nature of the beautiful and the sublime, of genius, and of the moral
dimensions of aesthetic judgment), while the second concerns primarily biology
(the use of purpose or final causation in our explanation of living organisms, as
well as the purpose of nature as a whole). What unites these two halves is the
concept of purposiveness guiding judgment, whether that judgment is aesthetic
or teleological. With aesthetic judgments, beautiful works of art or natural objects
fill us with a disinterested pleasure suggesting a harmonious fit between the mind
and the object contemplated; with teleological judgments of nature, organisms
appear purposive in their growth and development.
With this third Critique, Kant intended to complete his critical project by
bridging the “incalculable gulf fixed between the domain of the concept of
nature, as the sensible, and the domain of the concept of freedom, as the
supersensible” (CJ 5:175–76). He hoped to demonstrate the possibility of
these two realms forming a coherent whole, and so to bring within a single
focus the starry heavens above (the domain of nature as legislated a priori by the
understanding) and the moral law within (the domain of freedom as legislated a
priori by reason). Unifying theoretical and practical philosophy under a single
principle or system was a preoccupation passed on to those following Kant.
Kant and religion

Prussia’s first patron of the Enlightenment, Friedrich the Great, died on
August 17, 1786, and was succeeded by his religiously conservative nephew,
Friedrich Wilhelm II. Kant’s long-time admirer and ally at the Berlin court,
44 S. Naragon
K. A. von Zedlitz, was eventually replaced as Minister of Education and

Religious Affairs by J. C. Wöllner – the man whom Friedrich the Great had
described as “a deceitful and intriguing parson” – and after just one week in
office, on July 9, 1788, Wöllner issued a religious edict aimed at suppressing
the display of Enlightenment beliefs among teachers and clerics, followed in
December by a censorship edict to exert more control over religious pub-
lications. Kant himself was reprimanded on October 1, 1794, for his pub-
lication the previous year of Religion within the Boundaries of Mere Reason
(1793; Rel 6:1–202), and perhaps also for his more recent essay “The End of
All Things” (1794; EAT 8:327–39) that satirized the government censors.
Kant was forbidden to teach or write on matters of religion because he had,
in the words of the cabinet order, misused his philosophy “to distort and
disparage many of the cardinal and foundational teachings of the Holy
Scriptures and of Christianity” (Ak 11:525). Kant responded to the Berlin
government on Sunday, October 12, the day before the beginning of the
winter semester, writing that he would forgo any further public discussion of
religion, both in lecture and in publications (C 11:525–26, 530), a pledge
that he is said to have kept until the death of Friedrich Wilhelm II in 1797.68
Kant’s early life had been deeply informed by religion, and he remained
steadily engaged with religious questions until the very end. In his publica-
tions, beginning with his New Elucidation (1755) and The Only Possible
Argument in Support of a Demonstration of the Existence of God (1763), he
criticized the ontological proofs found in Descartes and Wolff, but then
developed a proof based on the necessary conditions for the possibility of
existence in general. By the time of his critical writings, however, Kant was
quite certain that all such proofs for God’s existence fail and that the scope of
natural theology was rather narrow. In general, religious doctrine is beyond
the domain of human knowledge, incapable of either proof or disproof, but
since certain supersensible ideas – including God, the afterlife, and freedom
of the will – have important practical implications for us, Kant found it
important “to deny knowledge in order to make room for faith” (Bxxx).
In sum, the purpose of religion is to bolster our moral lives in community
by helping to make actual the moral kingdom of ends (see, for instance, Rel
6:97–99). At the same time, any religion demanding assent to a creed is an
68
And yet we know from the Vigilantius metaphysics notes from that very semester that Kant did
indeed lecture on natural theology, some 36 pages’ worth, although unfortunately nearly all of this
section of the notes has gone missing, leaving us with only the very end of the discussion (LM 29:1040).
2 Kant’s Life 45
affront to our humanity and a breeding ground for hypocrites. In a letter of April
28, 1775, to the Swiss theologian J. C. Lavater (1741–1801), Kant wrote:
You ask for my opinion of your discussion of faith and prayer. Do you realize
whom you are asking? A man who believes that, in the final moment, only the
purest candor concerning our most hidden inner convictions can stand the test
and who, like Job, takes it to be a crime to flatter God and make inner
confessions, perhaps forced out by fear, that fail to agree with what we freely
believe. I distinguish the teachings of Christ from the report we have of those
teachings. In order that the former may be seen in their purity, I seek above all
to separate out the moral teachings from all the dogmas of the New Testament.
(C 10:175–76)
As for special revelation, Kant rejected from the very start any use of it to
explain physical phenomena, and eventually of morality as well. His
Universal Natural History and Theory of the Heavens (1755) offered an
account of the universe’s design based strictly on Newtonian mechanics. In
addition, his three earthquake essays of 175669 sought to dissuade his readers
from viewing the Lisbon earthquake, and earthquakes in general, as anything
more than physical events. They are neither punishments meted out by an
angry god nor do they offer any clues about God’s nature or existence. Kant
had little patience for claims of special revelation, which he found deeply
problematic – for by what criterion could we ever be certain that some event
had a divine origin, much less what it might mean? One special revelation
considered central to Judaism, Christianity, and Islam – namely, God’s
command that Abraham sacrifice his son Isaac – was singled out for special
scorn in his Conflict of the Faculties (1798):
For if God should really speak to a human being, the latter could still never
know that it was God speaking. It is quite impossible for a human being to
apprehend the infinite by his senses, distinguish it from sensible beings, and be
acquainted with it as such. – But in some cases the human being can be sure
that the voice he hears is not God’s; for if the voice commands him to do
something contrary to the moral law, then no matter how majestic the
69
“On the Causes of Earthquakes on the Occasion of the Calamity that Befell the Western Countries of
Europe towards the End of Last Year” (NS 1:419–27), “History and Natural Description of the Most
Noteworthy Occurrences of the Earthquake that Struck a Large Part of the Earth at the End of the Year
1755” (NS 1:431–61), and “Continued Observations on the Earthquakes That Have Been Experienced
for Some Time” (NS 1:465–72).
46 S. Naragon
apparition may be, and no matter how it may seem to surpass the whole of
nature, he must consider it an illusion. (CF 7:63)
[Here Kant adds a note:] We can use, as an example, the myth of the sacrifice
that Abraham was going to make by butchering and burning his only son at
God’s command (the poor child, without knowing it, even brought the wood
for the fire). Abraham should have replied to this supposedly divine voice:
“That I ought not to kill my good son is quite certain. But that you, this
apparition, are God – of that I am not certain, and never can be, not even if this
voice rings down to me from (visible) heaven.” (CF 7:63n)
Personally, Kant appears to have had little use for organized religion, and
perhaps just as little for a personal god. As an adult he rarely passed through a
church door,70 and when a friend asked Kant near the end of his life what he
thought about an afterlife, Kant replied: “Nothing certain.”71 Johann Brahl,
a frequent dinner guest and longtime editor of the Hartung newspaper, noted
in 1798 that, “while Kant postulates God, he does not himself believe in it,”
nor does he fear death.72
Kant’s end
Two stories have come down to us of how Kant might have died, but did
not; and in both cases someone had thought to murder him during one of his
regular afternoon walks. The first involved a deranged butcher, whom Kant
skillfully talked down. The second involved an escaped prisoner who resolved
to shoot dead the first person he met, which turned out to be Kant on his
walk, but the sight of the elderly professor so moved the convict that he
instead shot a young boy who happened by.73 As it turns out, Kant died
peacefully in his own bed, just a month shy of his eightieth birthday, on
February 12, 1804 – although for all practical purposes he had disappeared
sometime the previous year. In both body and mind, by the end Kant was an
entirely wasted man.
70
Reusch, Kant und seine Tischgenossen, 5.
71
Hasse, Letzte Äusserungen, 28–29.
72
Abegg, Reisetagebuch, 147. Professor Pörschke, a former student and then colleague of Kant’s, told
Abegg that Kant had assured him that “he had been teaching for a long time without ever doubting any
of the Christian dogma, [but] gradually one piece after another fell away” (Abegg, Reisetagebuch, 184).
73
Hagen, “Kantiana,” 16.
2 Kant’s Life 47
A number of publications appeared under Kant’s name during his last

years. The occasional pieces were no more than a few paragraphs in length: a
short preface to R. B. Jachmann’s book on religion (1800; Ak 8:441), an
afterword for a German-Lithuanian dictionary (1800; Ak 8:445), a public
notice denouncing Fichte’s philosophical innovations (1799; C 12:370–71),
and another denouncing Gottfried Vollmer’s unauthorized publication of
Kant’s physical geography lectures (1801; Ak 12:372). Younger colleagues
edited three volumes from manuscripts – G. B. Jäsche’s Logic (1800; JL 9:1–
150), and F. T. Rink’s Physical Geography (1802; PG 9:151–436) and
Pedagogy (1803; LP 9:439–99) – but Kant had no hand in any of these.
Kant was buried on February 28, 1804, just north of the church where he
had been baptized eighty years earlier. This was in an arcade given over for
the remains of professors; Kant was interred at the far eastern end. Wasianski
had been present at Kant’s death and filed this last report: “the mechanism
faltered and the machine stopped moving.”74
74
Wasianski, Immanuel Kant, 217.
3
Kant and His Philosophical Context:
The Reception and Critical Transformation
of the Leibnizian-Wolffian Philosophy
Manuel Sánchez-Rodríguez
Introduction
Kant’s relation to the modern philosophers that preceded him can be read
from different perspectives. First, we should take into account that Kant
explicated the theses of other philosophers explicitly in the context of his
open confrontation with them in order to defend transcendental philosophy
as an alternative within the intellectual context of the period. At the same
time, he considered criticism to be a historical consequence of the state of
indifference into which the scholarly world had plunged as a result of the
endless disputes in the field of metaphysics (Ax-xi). Hence, to Kant himself
we owe the historical understanding of criticism as the synthesis between
rationalist dogmatism and empiricist skepticism. This was Kant’s historical-
philosophical reconstruction of authors such as Locke, Hume, Leibniz, and
Wolff, for example, which cannot be evaluated without also cautiously
examining Kant’s own philosophical interest in them and the historical
context in which these currents of thought were received.
This chapter has been translated into English by David Nesbitt.
M. Sánchez-Rodríguez (*)
Department of Philosophy II, University of Granada, Granada, Spain
e-mail: msr@go.ugr.es

50 M. Sánchez-Rodríguez
This brings us to a second focus with respect to the historical relation

between Kant and modern philosophers. He did not always explicitly or
clearly declare the paths that had led him to his own theoretical positions, the
antecedents that occasioned him to reflect, the problems that prodded his
thought, or, in each of these contexts, the relation of dependence or the
distance with respect to the philosophers that had stirred his reflection.
Certainly, the historical self-understanding of criticism made by Kant is
important for the historian of philosophy, primarily because it has shaped
our understanding of the philosophical meaning of modern thinkers up to
the present day. However, a critical study of such a construction of Kant with
respect to the philosophy of his times is equally necessary as a means to
comprehend more thoroughly both the historical meaning of transcendental
philosophy and the way that Kant understood his own philosophical project.
Here we cannot focus on the multitude of classical and modern philosophical
approaches that Kant confronted. Nor would it be possible even to expound on
all the lines of influence that might converge in each of these approaches. But
we can examine one of these particular aspects to exemplify the manner in
which Kant publicly presents his relation to his predecessors, as well as to assess
in a critical way the intellectual development that led him to that relation.
In particular, the aim of this chapter is to defend the idea that, in the
Critique of the Power of Judgment, Kant preserves and redraws some of the
main theses of Leibniz’s philosophy, although from a radical critical trans-
formation of them in the general context of his critique of metaphysics. With
this critical transformation of Leibniz, Kant offers both a response to British
empiricism, which undoubtedly exerted a fundamental influence on the
formation of Kantian philosophy, and a critique of metaphysics. But I will
also argue that this critical transformation is proposed by Kant from a
theoretical position whose origin and development is only comprehensible
precisely in the context of the strong influence exerted on Kant’s thought by
Leibnizian-Wolffian psychology, logic, and aesthetics, even though this
historical backdrop may not seem evident in the text of the third Critique.
In particular, I will focus on Kant’s declaration in On a Discovery whereby
Any New Critique of Pure Reason Is to Be Made Superfluous by an Older One
(1790), according to which his theory of reflective judgment and the con-
ception of the faculties that accompany that theory reformulate, in a critical
sense, Leibniz’s system of pre-established harmony. On the one hand, in this
way Kant inscribes in his system a critical teleology that rests on a critique of
the metaphysical aspirations of teleology. On the other hand, he also refuses
to strip away all rational ground of the idea of the purposiveness of nature, as
might be expected given the influence of Hume’s skepticism.
3 Kant and His Philosophical Context 51
Multiple references to Leibniz appear in the writings of Kant,1 and in

these we find both critical questioning as well as clear recognition of the
brilliance of this thinker. Although the Critique of Pure Reason is clearly
aimed at Leibniz and at Leibnizian-Wolffian philosophy as one of the main
targets of his attacks, Kant’s assessment of them vacillates and fails to reach
equilibrium when you consider the body of his writings as a whole.2 Of
course, this assessment presupposes a particular interpretation of Leibniz’s
philosophy that is biased by the very philosophical position of Kant, so we
must note some of his presuppositions. First, Kant’s reading of Leibniz
appears to be conditioned by the fact that, even at the end of the eighteenth
century, many of Leibniz’s writings were unknown, especially those of his
earliest phase, even though works such as Nouveaux essais sur l’entendement
humain (New Essays on Human Understanding, completed by 1704 but
published in 1765), the Essais de théodicée (Essays on Theodicy, 1710), the
Monadologie (Monadology, 1714), and the Meditationes de cognitione, veritate
et ideis (Meditations on Knowledge, Truth, and Ideas, 1684) had been
widely disseminated.3 Second, linked to the former point, the comparative
studies of sources indicate that Kant generally had access to the ideas of
Leibniz through the interpretations of authors in the Leibnizian-Wolffian
tradition, such as Wolff and Baumgarten. Though of Leibnizian origin, these
interpretations do not explain with exactitude the theses of Leibniz himself or
submit them in a systematic way to clear transformations. Despite this, on
occasion Kant simply ascribes such interpretations to Leibniz himself or
presents the latter’s philosophy in an excessively rigid or simplified manner.
This chapter cannot critically review this biased distortion of Leibniz by
recovering the original spirit of his philosophy through an analysis of texts to
which Kant had no access; even current scholars find it difficult to formulate
the essential meaning of a philosophy so complex and alive as that of
Leibniz.4 Furthermore, it is of special interest here to examine this bias of
1
Anja Jauernig, “Kant’s Critique of the Leibnizian Philosophy: Contra the Leibnizians, but Pro Leibniz,”
in Kant and the Early Moderns, ed. Daniel Garber and Béatrice Longuenesse (Princeton: Princeton
University Press, 2008), 41.
2
Catherine Wilson, “Leibniz’s Influence on Kant,” in Stanford Encyclopedia of Philosophy, ed. Edward N.
Zalta (Winter 2012), http://plato.stanford.edu/entries/kant-leibniz.
3
On the dissemination and reception of Leibniz in the eighteenth century, see Catherine Wilson, “The
Reception of Leibniz in the Eighteenth Century,” in The Cambridge Companion to Leibniz, ed. Nicholas
Jolley (Cambridge: Cambridge University Press, 1995), 442–74.
4
Daniel Garber holds, not without reason, that the Kantian interpretation was based on “a gross
oversimplification of the real Leibniz” or on “a caricature of his real thought.” At the same time, he
admits that, “in a sense, it was impossible for Kant to get Leibniz right simply because there was no
single Leibniz to get right” (“What Leibniz Really Said?” in Kant and the Early Moderns, 67, 78).
the Kantian interpretation to the extent that this enables us to understand

more fully the reception of Leibniz’s thought and of the Leibnizian-Wolffian
philosophy in transcendental philosophy, because this reception forms part
of how Kant understands his own philosophy in the context of the intellec-
tual milieu of his period.
The opening of a critical recovery of Leibniz

in the confrontation with Eberhard
In What Real Progress Has Metaphysics Made in Germany since the Time of
Leibniz and Wolff? (1793), Kant declares that the efforts of Gottfried Wilhelm
Leibniz (1646–1716) and Christian Wolff (1679–1754) are unfruitful in
establishing a metaphysics that includes knowledge of the supersensible. With
his monadology and his system of pre-established harmony, Leibniz had sought
to offer an alternative solution to the theory of substance that could explain, in a
coherent way founded on metaphysical principles (as opposed to the
Cartesians), the nature of substances and the possibility of their reciprocal
and orderly communication in reality as a whole. With this theory, Leibniz
thought that he was not only resolving a metaphysical problem, since he
believed that the principles supporting his theory enabled him to explain and
found the observable order of phenomena, but also that he was giving sense to
the rationality of the natural philosophy of his time by complementing the
mechanism of the modern thinkers with teleological principles. Although
individual substances, or monads, are as such radically isolated from each
other and neither communication nor causal relations are possible between
them, we can explain the order between them and their belonging to the totality
of reality if we assume that their activity and development occur in a pre-
established harmony due to their common origin based on a sufficient and
purposive ground. Each individual monad represents, from a certain perspec-
tive and through its own states of perception, the totality of the universe and
therefore its relation with the rest of the monads. In the case of the spiritual
monad, the different perceptions of the soul are governed by an active principle
of unity and continuity independently of their clarity and distinctness. As long
as this active principle determines the individual identity of the spiritual monad
as well as the development and unfolding of its perceptions, and at the same
time it is in a pre-established harmony with its body, which expresses a relation
with the rest of the monads, Leibniz presupposes by this a correspondence of
the perceptions of the mind with respect to the whole of reality.
In Real Progress, Kant presents the main objections to this metaphysical

conception that are raised from the position reached in the Critique of Pure
Reason. Metaphysics cannot hope to achieve its aim of advancing from
sensible to supersensible cognition without having previously undertaken a
critique of pure reason that determines the objective validity and therefore
the scope and legitimacy of its aims. For this, it does not suffice to conduct a
critique of understanding in general and of the possibility of a priori knowl-
edge by means of mere concepts; rather, it is necessary to scrutinize, by means
of principles, human reason in its specificity – that is, “our mode of cognition
of objects” a priori (B25; see also Bxvii; Pro 4:350–51; RP 20:324–25),
which is essentially discursive because the understanding depends on a
heterogeneous faculty of cognition and therefore is not reducible to sensi-
bility. A metaphysics such as that of Leibniz, which attempts to account for
the conditions for the possibility of reality merely from the analysis of the
concepts and pure principles of reason, will not be able to fulfill its mission of
offering objectively valid knowledge of the supersensible. Rather, this dog-
matic metaphysics fails to rise above mere logical analysis, by which we can
certainly think reality, but in no way can know it, given that our concepts
can have objective validity and be used in a priori synthetic judgments only
to the extent that they refer in general to those objects that are given to
human reason, by means of the a priori forms of sensible intuition.
Therefore, metaphysics should be supported not only in an ontological
system “of all the concepts of understanding and principles” that serve as a
condition for the possibility of cognition in general (RP 20:260); we also
should take these principles to determine a priori the objectivity of the
objects only insofar as they are given to sensibility.
The metaphysics of Leibniz and Wolff was not intended merely to be
presenting the possibilities that the principle of contradiction offers to a
logical development of concepts, but such intentions are judged by Kant to
be unrewarding. First, by the principle of sufficient reason, these thinkers
sought to explain the existence of objects to which their concepts refer, but
according to the way that Kant presents this principle – “Everything has its
ground” – this affirmation refers to things in general as thought of by mere
concepts, and not to objects as given to the understanding by means of
sensibility (RP 20:277). That the principle refers to the ground of all that we
can think does not signify that it also refers to the cause of the objects of the
senses, since it is not possible through a logical step to deduce reality from
mere concepts. Rather, this principle lacks objective validity if it does not
satisfy the condition established by the Transcendental Deduction of the
Pure Concepts of the Understanding (A84–130, B116–69). In On a
Discovery, Kant reminds Johann Augustus Eberhard (1739–1809) of this

after the latter had followed Wolff and Alexander Gottlieb Baumgarten
(1714–1762) in the logical derivation of the synthetic principle of sufficient
reason from the analytical principle of contradiction: “that no category can
contain or bring forth the least cognition, if it cannot be given a correspond-
ing intuition, which for us human beings is always sensory” (OD 8:198).5
While the Leibnizians found the principle to be a mere logical principle and
do not demonstrate its objective validity, any attempt to take a step through
it in the field of metaphysics will not be legitimate, if in fact metaphysics
seeks to broaden our cognition concerning reality. Kant recognizes that
Hume was the first to place metaphysics in a predicament by posing the
question of the possibility of the a priori synthesis in relation to the principle
of causality (Pro 4:310; RP 20:266). Like Hume, Kant thought that one
should not fall into confusion between the cause by which an event takes
place and the epistemological ground by which we think about such an event
according to mere concepts.6 Kant also judged that the skeptical critique of
Hume, through which it is impossible to grasp by mere reason the possibility
of causality, could be extended to the rest of the categories that are generally
presupposed in natural philosophy. However, Kant did not go along with the
Scottish philosopher in his conviction that such concepts come from experi-
ence, nor that their intended necessity is a mere illusion due to custom, since
for Kant such concepts apply universally and necessarily to sensible objects
because they are an a priori condition for the objectivity of experience.
According to Kant, the second reason that Leibniz and Wolff did not
manage to make their intended leap from merely logical analysis to meta-
physics was because they had not recognized the specificity and heterogeneity of
sensibility as opposed to the understanding. For both Leibniz and Wolff,
sensory cognition is ultimately interpreted as being one degree lower within
the same intellectual cognition, so that intuition would in reality offer
confused access to the concept of objects as they are in themselves (A44/
B61–62; OD 8:216–20; RP 20:278–79). While Locke had committed the
error of sensibilizing the concepts, Leibniz and Wolff had intellectualized the
5
Johann Heinrich Lambert (1728–1777) had already criticized Wolff’s ambition to achieve cognition of
supersensible objects of metaphysics that cannot be demonstrated in experience. For Christian August
Crusius (1715–1775), also, philosophy should not be taken as the science of possible things, aimed at
discovering the reasons underlying the natural order, as Wolff believed, but rather as the analysis of
human cognition starting from the recognition of the finiteness of that cognition. In this context, see
Wilson, “Reception of Leibniz,” 458–59.
6
Predrag Cicovacki, “Kant’s Debt to Leibniz,” in A Companion to Kant, ed. Graham Bird (West Sussex:
Wiley-Blackwell, 2006), 86.
phenomena (A268–70/B324–26).7 By this subterfuge, rationalism implied

that it is possible and legitimate to cognize by the senses the same intellectual
basis of things that can also be thought of in a different way by the under-
standing, after a logical and reflexive analysis of the confused perceptions of
the mind. The senses and understanding know the same things, only with
different degrees of distinctness. For Kant, however, sensibility and the
understanding are two different sources of knowledge, and they presuppose
equally different principles. In Kant’s view, the claim of the rationalists
would imply not leaving the sphere of logic at all and claiming knowledge
through completely empty concepts, whether confused or different, as long
as the function of the understanding consists in applying its concepts to
intuitions that he cannot give to himself, but that must previously be given to
formal susceptibility of sensibility, which presupposes in turn that the subject
must be affected by sensations (OD 8:222).
Therefore, according to Kant’s objection to Leibniz and Wolff, “so far as
the speculative power of reason is concerned, no knowledge of the super-
sensible is possible (Noumenorum non datur scientia)” (RP 20:277).8 Any
attempt to construct a metaphysical project of the speculative-dogmatic
types, aimed at the knowledge of supersensible objects, is declared to be
fruitless in principle. However, transcendental philosophy, on the assump-
tion that reason is not merely theoretical, can in fact reflect on the concept of
the supersensible, and in this case
we would have to investigate the supersensible thing, not in respect of what it is in

itself, but only with regard to how we have to think it, and assume its nature to be,
in order for it to be apposite, for ourselves, to the practico-dogmatic object of the
pure moral principle, namely the final end, which is the highest good. (RP 20:296)
Therefore, philosophy cannot pursue knowledge of any supersensible object,

but only of the universal and necessary mode by which human reason thinks
of them, as a subjective condition of the practical realization of the moral
object.
7
Most comparative studies of sources show that the positions of Leibniz and Wolff on sensibility do not
coincide and that Kant is ascribing ideas to Leibniz that we in fact find only in Wolff. See, for example,
Catherine Wilson, “Confused Perceptions, Darkened Concepts: Some Features of Kant’s Leibniz-
Critique,” in Kant and His Influence, ed. George MacDonald Ross and Tony McWalter (Bristol:
Thoemmes, 1990), 73–103; and Michel Fichant, “Leibniz a-t-il ‘intellectualise les phenomenes’?
Elements pour l’histoire d’une meprise,” in De la sensibilité: Les esthétiques de Kant, ed. François Calori,
Michael Fessel, and Dominique Pradelle (Rennes: Presses Universitaires de Rennes, 2014), 37–70.
8
On the problem of the uncognizability of the thing in itself, see Chapters 5 and 6 of this volume.
Also, in the theoretical-speculative sphere, reason should assume the

concept of the supersensible in order to ensure the idea of systematicity
and order in nature that presupposes the understanding in his empirical use.
But here we find ourselves faced with a subjective demand that is not based
on constitutive principles of objective validity. As we will see, with this, Kant
holds that it is necessary to continue to uphold the Leibnizian idea that
nature underlies an order which agrees with the cognitive aims of the under-
standing, though under the assumption of its critical reversal. The oscillation
in the Kantian assessment of Leibniz’s thesis depends on this tension. That is,
Kant did not so much refute either the utility or suitability of the principles
criticized, but rather their interpretation as referring to things in themselves
once dogmatic philosophy fails to appreciate the relation of such principles to
the human mode of knowing, essentially dependent on sensibility. Having
confused phenomena with concepts and having assumed in his monadology
that bodies are ultimately composed of monads, Leibniz conceived of “a sort
of enchanted world” (RP 20:285). Because of the aim of explaining and
founding reality on mere concepts, the system of pre-established harmony is
considered by Kant to be “the strangest figment ever to be excogitated by
philosophy” (RP 20:284).9 In both cases, the error lies in having not appre-
ciated the transcendental distinction between the understanding and sensibility.
Kant’s ambiguous position is what explains why, in the confrontation with
Eberhard in On a Discovery, Kant could even speak (not without certain irony)
of an apology for Leibniz, inasmuch as he thought that the same principles that
he submitted to a critique could be recovered if interpreted correctly – that is, in
the critical sense. Had Eberhard, in his frontal attack on transcendental philo-
sophy, attempted to demonstrate that all that could be considered of value in
the Critique of Pure Reason had already been said by Leibniz, Kant would have
responded that, to the contrary, it is possible to detect and recover a criticism in
Leibniz under the condition that we do not take the intention of this thinker
verbatim and, primarily, that we abandon his theory of sensibility.
In this respect, the principle of sufficient reason should be interpreted,
according to Kant, as merely subjective and belonging to a critique of pure
reason.10 With this, Leibniz recognized that a merely logical principle, the
9
In this respect, it has been argued that Kant’s reception of monadology and the theory of pre-established
harmony is profoundly determined by the modifications of both ideas by Wolff and Baumgarten. See
Gaston Robert, “¿Qué tan leibnizianos eran los ‘leibnizianos’ Ch. Wolff y A. G. Baumgarten? Reflexiones
acerca de la teoría de la armonía preestablecida,” Ideas y Valores 63, no. 154 (April 2014): 107–35.
10
As Juan Antonio Nicolás claims, if it is possible to find a justification of the universal application of
the principle of sufficient reason in Leibniz, it is of the transcendental type, whereas Leibniz recognized it
as a condition for the possibility of rationality at the ontological, epistemological, and logical levels. See
principle of contradiction, would not suffice to reveal something more than

what resides in mere concepts. At that point, he proffered the idea that
metaphysics cannot be based merely on the analytical principle of contra-
diction, but that it is necessary also to provide a principle that offers the
reason or the ground of synthetic judgments (OD 8:247–48).
In relation to his monadology, Kant judges that Leibniz in reality, when
read in the critical sense, could not have been referring to more than the
supersensible substrate through which we think reality by means of an idea
of reason. And for this “we should not therefore let ourselves be disturbed
by his account of sensibility as a confused mode of representation, but must
rather replace it by another, more suited to his purpose,” with the aim of
lending consistency to the Leibnizian system itself (OD 8:248; see also
MFS 4:507–8).11 The philosophical knowledge provided by the monadol-
ogy could therefore be fundamentally preserved by criticism if it is held that
such knowledge does not refer to phenomena but to the mode by which our
reason can and should think the supersensible substrate of phenomena.
The system of pre-established harmony

and purposiveness without purpose
In relation to the hypothesis of pre-established harmony, Kant undertakes a
critical transformation of Leibniz. If Leibniz’s intention was to found an order
of reality and the possibility of a correspondence of reason with reality, Kant
judges this rational demand to be fulfilled by his theory of faculties, as founded
on the Transcendental Deduction in the first Critique and, especially, on the
theory of reflective judgment. For this reason, the idea of a harmony between
two substances should be replaced by the idea of a harmony between faculties of
knowledge involved in judgment: sensibility and the understanding. Reason
obligates us to assume the concordance between sensibility and the under-
standing by virtue of the possibility of experience, though it is not possible to
adduce any reason from it, especially, according to Kant, concerning the
problem posed in the grounding for reflective judgment in the Critique of the
Power of Judgment. Specifically, we cannot know
Juan Antonio Nicolás, “Universalität des Prinzips vom zureichenden Grund,” Studia Leibnitiana 22,
no. 1 (1990): 90–105.
11
Jauernig holds that Kant’s statements in On a Discovery should lead us to conclude that his criticisms
of Leibniz and the Leibnizians are in reality directed only at the latter (“Kant’s Critique of the Leibnizian
Philosophy,” 45). However, the assimilation of Leibniz by Kant is based on a clear critical correction of
the former, which is reflected also in On a Discovery.
why, as otherwise fully heterogeneous sources of cognition, they always con-

form so well to the possibility of empirical cognition in general, but especially
(as the Critique of Judgment will intimate) for the possibility of an experience of
nature under its manifold particular and merely empirical laws, of which the
understanding teaches us nothing a priori, as if nature were deliberately ordered
for our comprehension; this we could not further explain (and neither can
anyone else). (OD 8:249–50)
With his theory of pre-established harmony, Leibniz did not aim to demonstrate
this agreement between the apparent purposiveness of nature, intended by God,
and the possibility of a rational system of empirical knowledge. However, he did
presuppose that by the idea of purposiveness we should think of the super-
sensible as an underlying causality “of all things outside us” and not as a
supersensible causality that underlies “the mental powers in us” (OD 8:250).
Kant thus replaces the idea of a purposiveness in a divine will that precedes the
creation of the individual substances by a purposiveness in the supersensible
causality and of the subjective origin that – we should think – underlies the
manner in which the faculties of knowledge interact. Leibniz ultimately
explained the agreement between reason and reality by the assumption of a
common ground for both of them: supersensible causality of an archetypal
understanding that thinks and creates reality as the best of all possible worlds,
and that is therefore in agreement with the epistemological aim of reason.
Although both Leibniz and Kant interpret this idea as a principle of reason
that we must assume in order to investigate and unravel the structure of nature,
Leibniz did in fact draw metaphysical consequences from this idea: we should
accept, at least with moral certainty, a supersensible causality, common to
reality and reason, because, if we do not, we could not explain the possibility
of knowledge. For Kant, however, we cannot assume the existence of the
supersensible on the basis of our subjective manner of thinking the super-
sensible. The critical philosophy recognizes that this idea defines only the
manner by which human reason can reflect on nature, not how nature is in
itself, since from a subjective need of reason we cannot deduce an objective need
relative to the constitution of reality and to its relation to human reason. Thus
the principle of the formal purposiveness of nature has no objective validity,
based on a founded probability or moral certainty, but rather only heautono-
mous validity (CJ 5:185–86; FI 20:225).12 This is a subjective principle of
12
Concerning the grounding of the theory of reflective judgment within the general framework of
Kant’s theory of rationality, see Manuel Sánchez Rodríguez, Sentimiento y reflexión en la filosofía de Kant:
Estudio histórico sobre el problema estético (Hildesheim: Olms, 2010), 191–242.
reason, which legislates the manner in which we should a priori reflect on nature if
we aim to judge phenomena, in order to construct a system of empirical
knowledge. In the sensible experience of an indeterminate representation,
therefore, the inscription of such a particular in the ordered and interconnected
system of reality is not known or presupposed in a confused but objective way, as
Kant assumes that Leibniz and his followers thought. Rather, in reflection on
this particular representation, the subject must ideally refer this particular
representation to the regulative and heuristic idea of a system of empirical
knowledge that agrees with a supersensible order of nature. However, this
principle is reduced by criticism to a transcendental condition that merely
governs the relation between the faculties involved in judgment. That is, fitting
the order of reality to the demands of reason cannot be guaranteed at the
expense of the lack of distinction between sensibility and the understanding;
rather, as opposed to Leibnizianism, we must admit that the representation is
given to sensibility in its irreducible lack of determination with respect to the
epistemological demands of empirical understanding, insofar as what is given to the
senses in a certain way does not depend at all on the spontaneity of the
understanding, whether human or divine. The agreement between sensibility
and the understanding for the determination of a representation as singular
cannot be grounded or ensured a priori, but it is possible to determine a priori
that this subjective relation between faculties must have an underlying super-
sensible causality (the supersensible in us), by which we can and must ideally
refer our act of judging to the idea of cognition in general or to the possible
approval of any rational subject.
The treatment of the concept of purposiveness without purpose expresses in an
exemplary way the position of Kantian criticism as opposed to rationalism and
empiricism. With Leibniz and against Hume, Kant accepts that the assumption
of an agreement between the order of nature and the cognitive aims of reason is
an a priori principle of rationality, and that this is not explained merely by habit or
custom. For criticism, however, this idea does not describe how reality is but
rather the way that we think reality. Therefore, with Hume and against Leibniz,
the principle of uniformity of nature has a merely subjective origin, but the claim
for the necessity of this principle is not explained simply by its pragmatic utility in
the investigation of nature. On the contrary, the principle normatively defines
what is involved in rationally assessing or judging phenomena if one intends to
claim a meaningful understanding of them.13
13
On the confrontation with Hume in this regard, see Henry E. Allison, “Reflective Judgement and the
Application of Logic to Nature: Kant’s Deduction of the Principle of Purposiveness as an Answer to
Hume,” in Strawson and Kant, ed. Hans-Johann Glock (Oxford: Clarendon, 2003), 181–83.
The Leibnizian-Wolffian tradition

in the development of the theory of reflective
judgment
Up to this point, we have reviewed the reception of the Leibnizian-
Wolffian philosophy in the criticism of Kant based on his own declaration
with respect to this reception and how Kant himself understands it.
However, as indicated at the beginning of this chapter, it is also necessary
to look at Kant’s relation to the philosophy of his time from an additional
standpoint, which seeks to detect the outside lines of thought that converge
in transcendental philosophy and that are not always recognized by Kant.
The historical context of his intellectual development that led to his theory
of reflective judgment, and his conception of the relation between the
faculties of knowledge that accompanies it, offer an occasion for such an
examination.
Certainly, with his concept of purposiveness without purpose and the
subjective-rational principle of the supersensible, Kant reaches a position of
his own from which he confronts both dogmatic rationalism and empiricist
skepticism. In On a Discovery, he argues that we can read his Critique of
Pure Reason as the true apology for Leibniz. At first this may appear to mean
that, in Kant’s judgment, this work preserves and recovers the fundamental
aspects of Leibnizian thought after a critical transformation. However, at
least with respect to the hypothesis of pre-established harmony, his own
explanation in On a Discovery reveals that it is primarily the theory of
reflective judgment, explicated in the third Critique, that manages to
confront the philosophical problem that Leibniz had tried to solve, namely
the justification of the idea that nature presents an inner order that enables
the construction of a system of empirical laws, by virtue of the speculative
aims of reason.14 From the very position reached by Kant in the Critique of
the Power of Judgment, this is not a problem that the first Critique resolves
satisfactorily – not only because the concept of the purposiveness of nature
cannot be derived from the a priori legislation of pure understanding, but
also because the theory of reflective judgment approaches, from a different
14
Allison also argues that we should detect in the Critique of the Power of Judgment the “true apology”
for Leibniz, and that this work can be read as an attempt to reconcile Leibnizian realism of the universals
with the Lockean theory of concepts as products of the “workmanship of the understanding.” See Henry
E. Allison, “The Critique of Judgment as a ‘True Apology’ for Leibniz,” in Kant und die Berliner
Aufklärung: Akten des IX. Internationalen Kant-Kongresses, ed. Volker Gerhardt, Rolf-Peter
Horstmann, and Ralph Schumacher, 5 vols. (Berlin: de Gruyter, 2001), 1:286–99.
perspective, the treatment of this problem that we find in the Appendix to

the Transcendental Dialectic of the Critique of Pure Reason.15
As we have seen, with the distinction between sensibility and the under-
standing in the Inaugural Dissertation and the solution to the Transcendental
Deduction of the Pure Concepts of the Understanding in the Critique of Pure
Reason, Kant had succeeded in proscribing by principle any claim to cogni-
tion of the supersensible. The criticism thus offered the grounding for the
validity and scope of an a priori cognition at the same time as it established a
negative critique with respect to the aims of rationalist metaphysics.
However, we should remember that Leibniz considered metaphysical prin-
ciples to be simultaneously indispensable to ground the natural philosophy of
his time, because they complemented the mechanism of modern thinkers
with teleological ideas. Notwithstanding his critical confrontation with
metaphysics, Kant did not renounce this interest of Leibniz, so that his
objectives included not only establishing limits to the aims of theoretical
cognition in order to leave room for a practical development of pure reason,
but also justifying the pertinence of teleological maxims in theoretical reason,
with the goal of ensuring the claims of rationality that were presupposed in
natural philosophy itself. Now, by recognizing the distinction between
sensibility and the understanding and the discursivity of the latter given its
dependence on the former, Kant not only could prevent errors and contra-
dictions of metaphysics but also felt forced to look for an alternative path by
which to account for this teleological dimension of human rationality.
Leibniz, from a conception dependent on the general framework of his
hypothesis of pre-established harmony, presupposed by principle a corre-
spondence of reality, thought, and language,16 constitutive of the rationality
of the human being. Kant, however, after establishing the line between the
sensible and the intellectual, between the phenomenal and the noumenal,
had to think of how the principle of purposiveness functions in the logic of
the faculties of knowledge involved in judgment without breaking the critical
limits established in the Inaugural Dissertation and the Critique of Pure
Reason. We have verified that he considers his theory of reflective judgment
to offer an alternative in the criticism to Leibniz’s solution. This is not the
place to undertake a detailed analysis of the Critique of the Power of Judgment,
but it is possible to show that Kant’s intellectual development, by which he
15
Sánchez Rodríguez, Sentimiento y reflexión, 110–27.
16
Benson Mates, The Philosophy of Leibniz: Metaphysics and Language (New York: Oxford University
Press, 1986), 4–5.
reaches this alternative solution, at the same time is comprehensible only in

the historical context of the reception, discussion, and confrontation of
criticism with respect to the Leibnizian-Wolffian philosophy of the mid-
eighteenth century.
The explication of the problem of the final grounding of the reason in the
Critique of the Power of Judgment shows these limits of transcendental logic in
relation to this problem. As Kant clarifies in On a Discovery, as long as
sensibility and the understanding are two completely heterogeneous faculties,
it is impossible to provide any common ground that ensures a priori the
agreement between what is given to sensibility in its singularity and the
empirical law or the concept by which we can determine it to be in agree-
ment with a possible system of empirical knowledge. Certainly, the singular
ones are already determined in general by the categories of transcendental-
determinate judgment, for they are given to us as phenomena and thus are
possible experiences. However, from this a priori legislation of the under-
standing, it does not follow that the phenomena, beyond their sensible
formality and objective validity, present an order and reciprocal agreement
in such a way that it would be possible to form a system of interconnected
empirical laws according to a principle of teleological order (FI 20:203, 214;
CJ 5:179–80, 183).17
For this reason, the grounding of reflective judgment cannot be based on a
principle or concept of objective validity that enables us to decide a priori the
correctness of this faculty in judging phenomena. In relation to the empirical
content of our judgments, Kant already contended in the Critique of Pure
Reason that it is not possible to offer criteria that a priori determine the material
truth of those judgments (A57–59/B82–83), nor is it possible to use rules that
help judgment to subsume a particular case under a universal (A132–34/B171–
74; CJ 5:169). Hence, the theory of transcendental-determinate judgment
offers no rules to judge objects in concreto, but rather “the general condition
for rules” (A135/B174) – that is, the transcendental and determinate condition
that constitutes the objective validity of judgments of knowledge in general.
Transcendental principles do not establish a priori the particular objective
correctness, or the particular truth of these judgments; rather, they speak
about their general objectivity, inasmuch as they determine a priori phenom-
ena’s subsumption under the activity of the understanding, but they leave
totally undetermined their subsumption under the understanding’s empirical
concepts and particular laws (A128, B165, A480–81/B508–9).
17
Sánchez Rodríguez, Sentimiento y reflexión, 218–29.
This limitation on the scope of transcendental logic is a critical restriction

derived from the fundamental principle that sensibility and the understand-
ing are two heterogeneous faculties, whose agreement, which should be
presupposed in the judgment, cannot be completely determined a priori. It
may be thought that the recognition of this lack of determination of judg-
ment with respect to the concepts of the understanding marks a rigid
distinction between the critical philosophy, on one hand, and the theory of
cognition of the Leibnizian-Wolffian philosophy on the other. This would
begin by a principle of continuity between the sensible and the intellectual
and could thereby presuppose that knowledge of the sensible is completely
determined, though in a confused way, by principles of intellectual origin.
Such a reading could result from the historical self-understanding pro-
pounded by Kant himself concerning this problem, which we have examined
up to now. However, in such a case, we would be using the concept of
“Leibnizian-Wolffian philosophy” in an excessively lax way and ignoring the
complexity of Kant’s historical relation with this tradition. Certainly, this
recognition of the limits of logic, together with a progressive recognition of
the autonomy of sensibility as opposed to the understanding, has its main
precedents in the authors of this tradition, such as Baumgarten or Georg
Friedrich Meier (1718–1777), who had already undertaken a critique of this
conception of Leibnizian philosophy.
In Kant’s unpublished writings of around 1770, we find reflections on
judgment in concreto, or “‘sound understanding [gesunden Verstand]’”
(CJ 5:169),18 and the issue related to its justification or grounding. Although
Kant completely lacks the theoretical framework that he subsequently develops
in his transcendental logic and later in the theory of reflective judgment, it
becomes evident that in these early writings he already recognizes that any
attempt to determine or guide judgment in concreto by rules or precepts should
be declared futile (Ak 15:325 [R737], 16:23–24 [R1580]; NF 16:22 [R1579];
LAn 25:158, 359–60). Both the arguments presented as well as the terminology
and the existence of clear parallels should lead us to consider these texts as pre-
critical antecedents to the subsequent recognition that it is not possible to offer
rules or precepts that could justify the objective correction of judgment, as we
find in the first and third Critiques.19
18
“The correct use of [the power of judgment] is so necessary and generally required that nothing other
than this very faculty is meant by the name ‘sound understanding’” (CJ 5:169).
19
See Manuel Sánchez Rodríguez, “Logica naturalis, Healthy Understanding and the Reflecting Power
of Judgment in Kant’s Philosophy: The Source of the Problem of Judgment in the Leibnizian-Wolffian
Logic and Aesthetics,” Kant-Studien 103, no. 2 (Jan. 2012): 188–206.
However, this appraisal that the correctness of judgment cannot be

completely determined by universal rules or precepts of the understanding
does not originate with Kant. Rather, he inherits it directly from those
authors who had progressively begun to recognize the autonomy of sensi-
bility as opposed to the intellectual, and had claimed, mainly against the
corrective ideal of philosophical knowledge present in Wolff, that the scope
of logic is inadequate in this regard.20 While Wolff claimed that all of the
rules presupposed in a confused way by the natural skill of cognition – logica
naturalis – could be completely formalized in a distinct way and made
explicit by the logica artificialis, some later philosophers began to defend
the irreducibility of logica naturalis presupposed in the judgment of sound
understanding, within the general context of an incipient recognition of the
specificity of sensibility.
In the posthumously published logic and anthropology lectures from 1771
to 1773, Kant expounds for the first time on the problem of the grounding of
judgment in concreto, and his thoughts are directly related to this debate, just
as it can be found in Baumgarten and later in Meier, his main disciple and
apologist. From this moment on, until 1789, a continuity can be traced in
Kant’s reflections on the issue of the grounding of judgment.21 Baumgarten
and Meier provide no advance whatever in the fundamentals of the theory of
reflective judgment or the subjective principle of the formal purposiveness of
nature. However, we have seen that these ideas presuppose a fundamental
critique of Leibniz – that is, the defense of a discontinuity between the
sensible and the intellectual, giving rise to the impossibility of a grounding of
judgment by objectively valid principles of the understanding. Although only
Kant establishes and grounds this delineation between sensibility and the
understanding, and even draws the consequences of this delineation for the
critique of Leibnizian metaphysics, it is also true that he already finds in the
work of Baumgarten and Meier the first crack that starts the breakup of
dogmatic-speculative metaphysics.
The historical developments of Leibniz’s ideas within the framework of
aesthetics, empirical psychology, and logic in German philosophy laid the
first foundations for the progressive recognition of the autonomy and
irreducibility of sensibility as well as for a treatment of the concepts of
metaphysics in subjectivist and immanent terms. The reception and disse-
mination of the Meditationes (1684) of Leibniz, where he admits the
20
Ibid., 193–97.
21
Ibid., 199–202.
possibility, in disagreement with Descartes, that knowledge can be clear

despite being characterized by confusion, plays a fundamental role in this
historical development.22 In his Deutsche Logik of 1712, Wolff draws
links to ideas explicated in this essay. In the fundamentals, he does not
distance himself from the position of Leibniz. His main novelty, how-
ever, stems from having placed this logical system of metaphysical knowl-
edge in a psychological framework, which would serve as the basis for the
subsequent treatment of confused cognition from a theory of perfect-
ibility of subjects. In this respect, the progressive delineation of the
sensible is given by Wolff for the distinction between an upper part
and a lower part of the faculty of knowledge, in such a way that
Leibniz’s division between confused and distinct cognition, though it
continues to be interpreted in a gradual sense, gives rise in Wolff to a
doctrinal division between different levels in the faculty of knowledge
itself.23 It will be empirical psychology that will study cognitive faculties
and processes of the lower part of knowledge. The division between
logica naturalis and logica artificialis referred to above presupposes this
doctrinal distinction of empirical psychology.
Baumgarten continues this development of Leibnizianism launched by
Wolff. While the latter philosopher discussed a lower part of the faculty of
knowledge, Baumgarten recognizes in its specificity a lower faculty of knowl-
edge, whose explication receives autonomous treatment within the system of
his Metaphysics, particularly in the part dedicated to his empirical psychology,
in which confused or indistinct cognition receives the designation of “sensi-
tive representation.”24 With this, the possibility emerges of a positive treat-
ment of the sphere of the sensible, not merely a negative consideration of its
lower degree of intellectual knowledge. Thus, the representation that is
known in a clear and confused way by the lower faculty of knowledge
provides a different kind of knowledge, not so much deficient knowledge.
In this case, it is extensive clarity, which is differentiated specifically from
intensive clarity provided by analysis.25
22
Manuel Sánchez Rodríguez, “La teoría de los tipos de representación en Leibniz y sus principales
influencias en la estética y la lógica de la Ilustración alemana,” Cultura: Revista de História e Teoria das
Ideias 32 (2013): 271–94.
23
Christian Wolff, Psychologia empirica (1732), in Gesammelte Werke, ed. Jean École (Hildesheim:
Olms, 1972), pt. 2, vol. 6, §§54–55.
24
Alexander Gottlieb Baumgarten, Metaphysica (17391, 17574), §521 (Ak 15:9).
25
Ibid., §§531–32 (Ak 15:13, 20–22).
The importance of this influence has been pointed out in the formation
of Kant’s concept of sensible cognition, as presented in the Inaugural
Dissertation.26 Nevertheless, the grounding of the formal conditions of
sensible cognition in this period, though nurtured in this historical tradi-
tion, is not carried out by Kant without a clear confrontation with and
correction to Baumgarten, who, Kant claims, continues to consider the
sensible as a lower degree of intellectual cognition.27 For Kant, sensibility
and the understanding are two heterogeneous sources of cognition, whose
possibility should be founded on principles that are specific to each. From
there, Kant sharply rejects the lack of distinction of the two levels in the
aesthetic conception of the German rationalists, for which beauty is con-
sidered the sensible or confused knowledge of perfection in the phenomenon
(LAn 25:386; BL 24:77). The rationalists continue to preserve the classical
ideal that it is possible to create an aesthetic as a doctrinal body by which
the correctness of the judgment of beautiful objects can be discerned (A21n/
B35n). For Kant, “[Henry] Home, more correctly, called aesthetics critique,
since it yields no rules a priori that determine judgment sufficiently, as logic
does” (JL 9:15).
In this respect, Kant’s position on the issue of the aesthetic judgment of
beauty coincides with his recognition that it is impossible to ground the
correctness of a judgment of sound understanding or, according to later
terminology, of reflective judgment: how can judgment of the sensible
singular make the claim of validity where, in principle, understanding cannot
offer concepts or rules of objective validity that guide this activity
(NF 15:432 [R988])? The solution to this problem, which Kant will address
in the third Critique as a single issue, cannot be given by his philosophical
predecessors. On one hand, even though their aesthetics and rationalist
psychology had opened the way for a theory of faculties that took into
consideration the specificity of sensibility, they remained committed to the
idea that a sensible – confused – representation of perfection is possible. On
the other hand, in English criticism it is not possible to find a historical
26
Giorgio Tonelli, Kant, dall’estetica metafisica all’estetica psicoempirica: Studi sulla genesi dell crticismo
(1754‒1771) e sulle sue fonti (Torino: Memorie dell’Academia delle Scienze di Torino, 1955); and
Takeshi Nakazawa, Kants Begriff der Sinnlichkeit: Seine Unterscheidung zwischen apriorischen und
aposteriorischen Elementen der sinnlichen Erkenntnis und deren lateinischen Vorlagen (Stuttgart-Bad
Cannstatt: Fromman-Holzboog, 2010), 197–256.
27
Sánchez Rodríguez, Sentimiento y reflexión, 24–54; and Manuel Sánchez Rodríguez, “Philosophische
Kritik des Geschmacks und objektive Gültigkeit des Schönen bei Kant in der Zeit der Dissertatio,”
Historia philosophica 10 (2012): 11–24.
development through which such a claim expressed in judgment can be

based on a priori criteria. In addition, the transcendental deduction and
the theory of transcendental-determinate judgment that is derived from it
only allows thinking the relation between sensibility and the understanding
with respect to the objective validity of the judgment.
The debate over the concept of genius in Germany at the end of the
1770s offered Kant the occasion to reflect on the subjective aspect of the
relation between the faculties.28 This debate presents a clear monadological
stamp inasmuch as the spirit of genius was explained as active productivity
grounded on a basic force (Grundkraft) that underlies the relation between
the faculties of knowledge and in this way makes their harmony possible.
In Kant’s unpublished writings and the Lectures on Anthropology, we find
the reception of this conception of genius of Leibnizian origin, still
incompatible with the limits imposed by the Transcendental Dialectic of
the first Critique. Through such reflections, Kant tentatively explores the
idea of an indeterminate ground of the interrelation and play between the
faculties of knowledge. In debates on the concepts of genius, spirit, and
basic force in authors such as Meier, Johann Georg Sulzer (1720–1779),
and Johannes Nikolaus Tetens (1736–1807), and after abandoning the
metaphysical presuppositions still present in them, Kant finds a path
to start a critical, immanent, and subjectivist transformation of the
Leibnizian hypothesis of pre-established harmony between substances
through the idea of a free and harmonic play between faculties of knowl-
edge. The incorporation of this line of thinking into the critical system
became possible by grounding the rational concept of the supersensible as
merely subjective and by considering it to be the ultimate principle
underlying this harmonic relation between faculties and ensuring subjec-
tive validity of the a priori concept of purposiveness of nature as a
principle that the power of judgment gives to itself for reflection
(NF 15:436 [R992]).
The exemplary case of the critical transformation of Leibnizianism by
Kant shows that, to study the historical-philosophical context of his work,
we should not only examine his own understanding of transcendental
philosophy. In fact, if Kant can assert against Leibnizians such as
Eberhard that criticism represents the true apology for Leibniz with respect
28
On the origin of the concept of genius in Kant in the context of Leibnizian-Wolffian philosophy, see
Manuel Sánchez Rodríguez, “La influencia del leibnicianismo en la génesis de la teoría del genio de
Kant: El Geist como fuerza básica,” in Leibniz en la filosofía y la ciencia modernas, ed. Manuel Sánchez
Rodríguez (Granada: Comares, 2010), 535–52.
to the pre-established system of harmony, insofar as he recovers the essen-

tial philosophical sense of this idea in his theory of reflective judgment,
then certainly this transformation is not historically comprehensible with-
out taking into account the influence that other Leibnizians exercised over
Kant’s intellectual development.29
29
Research for this chapter was supported by a Ramón y Cajal Senior Grant and the Research Project
“Leibniz en español,” Ministry of Economy, Spain.
Part II
Metaphysics and Epistemology
4
Transcendental Idealism: What and Why?
Paul Guyer
What?
In the first edition of the Critique of Pure Reason (1781), Kant defines the
doctrine that he there calls “transcendental idealism” thus:
I understand by the transcendental idealism of all appearances the doctrine

that they are all together to be regarded as mere representations and not as
things in themselves, and accordingly that space and time are only sensible
forms of our intuitions, but not determinations given for themselves or con-
ditions of objects as things in themselves. (A369)
This is from the Fourth Paralogism of Pure Reason in the Transcendental

Dialectic. In summing up the argument of Transcendental Aesthetic, Kant
had used similar language, although not in the form of an explicit definition:
Time and space are accordingly two sources of cognition, from which different
synthetic cognitions can be drawn a priori, of which especially pure mathe-
matics in regard to the cognitions of space and its relations provides a splendid
example. Both taken together are, namely, the pure forms of all sensible
intuition, and thereby make possible synthetic a priori propositions. But
these a priori sources of cognition determine their own boundaries by that
P. Guyer (*)
Department of Philosophy, Brown University, Providence, USA
e-mail: paul_guyer@brown.edu

72 P. Guyer
very fact (that they are merely conditions of sensibility), namely that they apply
to objects only so far as they are considered as appearances, but do not present
[darstellen] things in themselves. Those [appearances] alone are the field of their
validity, beyond which no further objective use of them takes place. (A38–39/
B55–56)
In the Prolegomena to Any Future Metaphysics, published two years later and
intended as a popularization of the Critique, though in the event also a
defense of it, he suggests that the doctrine might better have been named
“critical idealism.” Here he says:
What I called idealism did not concern the existence of things (the doubting of
which, however, properly constitutes idealism according to the received mean-
ing), for it never came into my mind to doubt that, but only the sensory
representation of things, to which space and time above all belong; and about
these last [space and time], hence in general about all appearances, I have only
shown: that they are not things (but mere modes of representation), nor are
they determinations that belong to things in themselves. (Pro 4:293)
Kant also states that, on his “system,” “the existence of the thing that appears
is not thereby nullified, as with real idealism, but it is only shown that
through the senses we cannot cognize it at all as it is in itself” (Pro 4:289).
Putting these several statements together, it would seem that transcendental
idealism is the view that space and time are the forms of our sensible
intuitions, defined by Kant as that through which cognition “relates imme-
diately” to its objects and as given through our senses (A19/B33); that
spatiality and temporality are thus necessary features of all our sensible
representations of objects but only of those representations, as well as of
the appearances of objects identical to or dependent upon those representa-
tions; and that while there are objects distinct from our representations of
them that give us sensible representations and appear to us through those
sensible representations, those objects themselves are not spatial and/or
temporal or in space and/or time. (I will henceforth usually use the expres-
sion “spatiotemporal” in place of this more cumbersome locution.)
It might seem that the content of this philosophical doctrine is clear
enough, and that the chief question about it would be why on earth
anyone would hold it, as opposed to thinking that such a fundamental
feature of our sensible representation of the world as its spatiotemporality
is not more than sufficient evidence that the world really is spatiotem-
poral. Kant points to his chief reason for holding transcendental idealism
4 Transcendental Idealism 73
in our first quotation when he connects it to the possibility of synthetic a

priori cognition of space and time and their properties, of which math-
ematical knowledge of space (that is, geometry) is a “splendid example.”
But before we can consider Kant’s argument (or arguments) for trans-
cendental idealism, in particular his argument from synthetic a priori
knowledge, we must consider what he means by the doctrine. For in
spite of the straightforward appearance (pardon the pun) of the state-
ments we have quoted, what Kant means by “appearances” and “things
in themselves,” as well as what he means by his statements that space and
time are not things in themselves or properties of them, have proven
controversial. But we had better not stray too far from the question of
Kant’s argument (or arguments) for transcendental idealism, for it is my
view that we can only determine what Kant means by transcendental
idealism by seeing what follows from his argument (or arguments) for it
– although of course understanding his arguments will presuppose at
least a provisional understanding of some of his key terms.
Since a 1982 article by Karl Ameriks1 and the 1983 first edition of Henry
Allison’s book Kant’s Transcendental Idealism: An Interpretation and Defense,2
many have distinguished between “one-object” or “one-world” and “two-
object” or “two-world” interpretations of transcendental idealism. On one-
object approaches, appearances and things in themselves are held to be, in
some way, numerically identical objects, hence the term “one-object”; while
on two-object approaches, appearances and things in themselves are held to
be numerically distinct objects, hence the name “two-object” approach.
Allison has been a chief proponent of a “one-world” interpretation, although
he was anticipated by two decades by Graham Bird,3 and presented it in
1983 as an “epistemological” or “methodological” rather than “metaphysi-
cal” or “ontological” position: his claim was that appearances and things in
themselves are not two kinds of things but two concepts or ways of con-
sidering things, one of which, that is, the concept of things as appearances,
includes in it the essential feature and condition of the possibility of our
1
Karl Ameriks, “Recent Work on Kant’s Theoretical Philosophy,” American Philosophical Quarterly 19,
no. 1 (Jan. 1982): 1–24; reprinted in his Interpreting Kant’s “Critiques” (Oxford: Clarendon Press,
2003), 67–97.
2
Henry E. Allison, Kant’s Transcendental Idealism: An Interpretation and Defense (New Haven: Yale
University Press, 1983), rev. ed. (New Haven: Yale University Press, 2004).
3
See Graham Bird, Kant’s Theory of Knowledge: An Outline of One Central Argument in the “Critique of
Pure Reason” (London: Routledge & Kegan Paul, 1962). Bird has more recently defended his approach
at length in The Revolutionary Kant: A Commentary on the “Critique of Pure Reason” (Chicago: Open
Court, 2006).
74 P. Guyer
cognition of things, namely their spatiotemporality (what Allison called

“epistemic conditions”), while the other, that is, the concept of things in
themselves, considers things, the very same things, in abstraction from these
conditions.4 This one-object interpretation has also been called a “two-
aspect” rather than two-object interpretation, on the ground that it interprets
Kant as considering one and the same (set of) objects under two different
aspects. As Tobias Rosefeldt has pointed out, however, there can also be an
ontological rather than epistemological two-aspect theory, on which the
distinction between appearances and things in themselves is taken as a
contrast between two aspects or kinds of properties of things rather than
just two ways of considering them.5 And this is not an empty possibility,
since at least one author has claimed that Kant’s expression “thing in itself”
refers to the intrinsic properties of things while “appearance” refers to the
relational properties of those same things – the same things, but not the same
properties.6 This view seems particularly unattractive, and the one passage in
which Kant might suggest it seems particularly confused. But let us not
worry about that just yet. The point for now is that both of these kinds of
“one-object” approach are contrasted to a “two-objects” or “two-world”
approach, on which appearances and things in themselves are held to be
two ontologically, numerically distinct sets of objects, although sets of objects
that are of course related to each other, in that things in themselves are in
some way the grounds of appearances, although not in a straightforward
causal way, since causality is held to be a spatiotemporal relationship and
things in themselves, on a two-world approach, are not supposed to be
spatiotemporal.
In an important paper, Ralph Walker has argued that the “one-world” and
“two-world” terminology is metaphorical and misleading in a number of
ways.7 One way in which I find it misleading is that no one except a Reidian,
direct-realist interpreter of Kant, should there be such a thing, would deny
that on any interpretation of things in themselves Kant, like every other
early-modern philosopher but Reid, admits at least two domains of objects,
4
Allison in turn credited two books by Gerold Prauss, Erscheinung bei Kant: Ein Problem der “Kritik der
reinen Vernunft” (Berlin: de Gruyter, 1971); and Kant und das Problem der Dinge an sich (Bonn: Bouvier,
1974) for help in formulating his position.
5
Tobias Rosefeldt, “Dinge an sich und sekondäre Qualitäten,” in Kant in der Gegenwart, ed. Jürgen
Stolzenberg (Berlin: de Gruyter, 2007), 167–209.
6
Rae Langton, Kantian Humility: Our Ignorance of Things in Themselves (Oxford: Clarendon Press,
1998).
7
Ralph C. S. Walker, “Kant on the Number of Worlds,” British Journal for the History of Philosophy 18,
no. 5 (Dec. 2010): 821–43.
namely our own representations and, at least in the case of outer sense, the
objects of those representations. (In the case of inner sense, where we have
representations of our own representations, the object of one representation
may be another particular representation, but not a different kind of object.
This may cause problems of its own.8) The question is thus not whether
Kant recognizes one set of objects or two, but rather whether he recognizes
two sets of objects, namely representations and things in themselves, or three,
namely representations, things in themselves, and appearances, ontologically
distinct from both representations and things in themselves. But a view like
that seems particularly unattractive, and it seems that the real question at
issue ought to be: how does Kant understand the relationship between the
two sets of objects, namely our representations and their external objects, that
he, like everyone else of this time, takes for granted? Does he suppose that
our representations are spatiotemporal and that their ontologically indepen-
dent objects are also spatiotemporal, although we can and for some reason
should have a concept of them that omits their spatiotemporality; or does he
hold that, while our representations of things are spatiotemporal and there-
fore the way things appear to us or their appearances are also spatiotemporal,
for some reason those things as they are in themselves are not spatiotemporal?
Because the one-world/two-world contrast is so misleading, I think it
would be better to contrast “conceptual” to “ontological” approaches to
transcendental idealism: on conceptual approaches, spatiotemporality is
merely omitted from the concept of things in themselves, while on an
ontological approach, spatiotemporalty is denied of things in themselves,
or they are asserted to be non-spatiotemporal.9 My claim, in contrast to
interpretations like those of Allison and Bird, is then that Kant’s position is
the latter, and that given his actual arguments for transcendental idealism it
can only be the latter – although in fact his arguments for this position, dear
as especially the foremost among them is to him (and vital as the whole
doctrine is to his moral philosophy), are not very good.10 In other words, it is
8
This is the very problem that the earliest critics of Kant’s not-yet-named transcendental idealism,
namely Johann Heinrich Lambert, Johann Georg Sulzer, and Moses Mendelssohn, raised in their 1770
letters to Kant about his inaugural dissertation (see C 10:103–10, 111–13, 113–16, esp. 116), the
question, namely: how could states of inner sense merely appear temporal when consciousness of them is
also a state of inner sense, that is, when they appear to other states of inner sense? There will not be room
here for further discussion of this special although important issue.
9
I will use the term “ontological” rather than “metaphysical” because Kant uses the latter term in so
many different ways that it might be best to avoid it when trying to talk about his views in one’s own
voice.
10
Langton’s approach, on Rosefeldt’s classification an ontological rather than methodological one-world
approach, would then turn out to be simply an ontological approach, although one on which
76 P. Guyer
better to correctly understand Kant’s doctrine and reject it than to defend it

by misunderstanding it.
We must turn to Kant’s actual arguments because his terminology alone
will certainly not settle the question of which approach to the interpretation
of transcendental idealism is correct. At least if we keep using the one-object/
two-object language, it is easy to find passages that speak for each approach.
One-object, two-aspect interpreters appeal to passages such as this: “the same
objects can be considered from two different sides, on the one side as objects
of the senses and the understanding for experience, and on the other side as
objects that are merely thought at most for isolated reason striving beyond
the bounds of experience” (Bxviii-xix note), and “appearance . . . always has
two sides, one where the object is considered in itself (without regard to the
way in which it is to be intuited . . . ), the other where the form of the
intuition of this object is considered” (A38/B55). But two-object theorists
can equally appeal to passages such as these: “what we call outer objects are
nothing other than mere representations of our sensibility, whose form is
space, but whose true correlate, i.e., the thing in itself, is not and cannot be
cognized through them” (A30/B45); “external objects (bodies) are merely
appearances, hence also nothing other than a species of my representations”
(A370); and “we have to do only with our representations; how things in
themselves may be (without regard to the representations through which they
affect us) is entirely beyond our cognitive sphere” (A190/B235). Lucy Allais
has recently provided an extensive catalogue of passages speaking for each
side of the debate, which puts it beyond doubt that the debate can never be
settled just by producing texts that seem to speak for one approach or the
other.11 In fact, the passages show that Kant uses the term “appearance” itself
in a systematically ambiguous way: sometimes it means the way in which
objects appear to us, that is, the representations through which an object
appears to us; but sometimes it means the objects that appear to us in a
certain way through those representations, as they appear to us. When Kant
states that “appearances are only representations of things that exist” (B164),
he is using the term in the former way; but when he says, as already quoted,
that appearance “always has two sides, one where the object is considered in
itself (without regard to the way in which it is to be intuited . . . ), the other
where the form of the intuition of this object is considered” (A38/B55), he is
spatiotemporality is denied of only one part of things, their thing-in-itself part. To say this is already to
suggest how problematic this approach is.
11
Lucy Allais, Manifest Reality: Kant’s Idealism and His Realism (Oxford: Oxford University Press,
2015), 28–33.
using it in the latter way. If “appearance” is used in the former way, thus as a
generic term for more specific experiences such as a glimpse or a whiff, then
of course appearances are ontologically distinct from the things that appear,
events in a perceiver rather than objects perceived or states of them; but if
“appearance” is used in the latter way, then it can be taken to refer to the
object that appears rather than the experience of it, and thus appearance and
thing in itself can be ontologically or numerically identical.12 But this fact
about the usage of “appearance” leaves entirely unsettled the question of what
Kant means when he says that the object that appears may be considered
with or without “regard to the way in which it is to be intuited”: is he saying
just that the way in which we intuit objects, that is, spatiotemporality, may
be omitted from the concept of the object that appears, which of course leaves
open that spatiotemporality nevertheless is a property of that object; or is he
denying that spatiotemporality is a property of the thing that appears, being
instead only a property of the way the thing appears that is due to the way we
represent it?
My position is that Kant’s denial of the spatiotemporality is not in the first
instance a claim about the concepts of things in themselves but about those
things themselves, the things that ground our representations of them and
appear to us in a certain way because of the nature of our representations of
them, and that spatiotemporality is to be excluded from the concept of things
in themselves only because it is denied of the things themselves. Kant
certainly seems to speak in the material rather than formal mode, that is,
about objects rather than concepts, when he first starts his argument for
transcendental idealism, in drawing “Conclusions from the above concepts”
following (what he labeled in the second edition of the Critique) the “meta-
physical” and “transcendental” expositions of the concept of space, namely
that “Space represents no property at all of any things in themselves nor any
relation of them to each other, i.e., no determination of them that attaches to
objects themselves and that would remain even if one were to abstract from
all subjective conditions of intuition” (A26/B42). As I have argued before, if
Kant were just speaking about concepts, he could not infer from the fact that
12
I say this for the moment, speaking in the voice of the one-object theorist. Of course, once Kant
denies that things in themselves are spatiotemporal, then there arises a question about how non-
spatiotemporal objects can be numerically identical with spatiotemporally individuated objects, and
indeed there are arguments, such as the Third Paralogism, in which Kant clearly argues that they need
not be and cannot be determined to be. On this issue, see, among others, Dennis Schulting, “Kant’s
Idealism: The Current Debate,” in Kant’s Idealism: New Interpretations of a Controversial Doctrine, ed.
Dennis Schulting and Jacco Verburgt (Dordrecht: Springer, 2011), 13–16; and Nicholas F. Stang, “The
Non-Identity of Appearances and Things in Themselves,” Noûs 48, no. 1 (March 2014): 106–36.
78 P. Guyer
a concept of an object abstracts from some property of it that the object itself
lacks that property, any more than we could infer from the fact that a job
application omits any mention of the candidate’s gender that the candidate
has no gender.13 But Kant does not seem to be talking about what is omitted
from a concept, and committing the fallacy of inferring what is true (or
rather false) of an object from what is omitted from a concept of it; he seems
to be talking of things themselves, and straightforwardly denying that they
have a property or stand in a relation, that of spatiality.
But as I said, what Kant means by transcendental idealism cannot be
settled verbally, since statements supporting both main interpretations can so
readily be found. Rather, the only way to settle what he means by a statement
even as apparently unequivocal as that at A26/B42 is to examine the argu-
ment from which it is supposed to follow and to determine what that
argument actually entails. So that is what I will now do – and the upshot
of so doing will be the conclusion that what follows from Kant’s argument
for the claim at A26/B42, indeed what follows from all his arguments on
behalf of transcendental idealism, is the denial that things in themselves are
spatiotemporal, thus the assertion that the things that necessarily appear to us
spatiotemporally are not themselves spatiotemporal.
Why?
Kant offers several arguments for transcendental idealism. But there is one
that he offers first and most frequently. The chief premise for this argument
is stated immediately following the “Conclusions” I have already quoted
from A26/B42. Kant follows that statement thus: “For neither absolute nor
relative determinations can be intuited prior to the existence of the things to
which they pertain, thus be intuited a priori” (A26/B42). Kant’s claim is that
the spatiality of things in themselves would be incompatible with our a priori
knowledge of spatiality, and that since we do have such a priori knowledge,
then things in themselves cannot be spatial. (Kant will subsequently make the
analogous argument concerning temporality.) In turn, it is from this denial
that things in themselves are spatial that Kant infers that “space is nothing
other than merely the form of all appearances of outer sense, i.e., the
subjective condition of sensibility, under which alone outer intuition is
13
See my review of Henry Allison’s 1990 book Kant’s Theory of Freedom in Journal of Philosophy 89,
no. 2 (Feb. 1992): 99–110.
possible for us” (A26/B42) (and will likewise later draw the analogous
inference that temporality is only the form of inner sense, although as already
noted this claim may have problems of its own). That is, he does not infer
that spatiotemporality is not a property of things in themselves from a prior
premise that spatiotemporality is the form of our intuition (or, as Allison
called it, the epistemic conditions of our knowledge), but infers that spatio-
temporality is merely the form of our sensibility from the preceding argument
that it is not a property of things in themselves. Our challenge is thus to
understand why he thinks that it would be incompatible with our a priori
knowledge of spatiotemporality if things in themselves were actually
spatiotemporal.
But first, a quick review of the arguments about space (and mutatis
mutandis, time) leading up to Kant’s “Conclusions.” In what Kant labels
the Metaphysical Exposition in the second edition, he argues, first, that
spatiality is the a priori form of our intuition of outer objects, because we
cannot represent objects as distinct from each other or as outside of ourselves
except by representing them as separated from each other or our own bodies
in space (A23/B38), and because we can represent space without representing
objects but not vice versa (A24/B38–39); for both of these reasons, he claims,
we need an antecedent, a priori representation of space. Then he argues that
our a priori representation of space itself is singular and thus an intuition
because we can represent any space, no matter its size, as a region of a larger
space, and because we therefore represent space as one, infinite manifold
rather than as any number, finite or infinite, of instances of a general concept.
(Similar arguments are made for the case of time.) In the second edition of
the Critique, Kant separates out from these arguments, under the rubric of
the Transcendental Exposition of the Concept of Space, an argument he had
previously sandwiched between the first two and last two of the arguments of
the Metaphysical Exposition, namely that we must have an a priori intuition
of space and its structure in order to explain the synthetic a priori cognition
that we have in geometry (B40–41). (In the case of time, he says that the
representation of temporal succession is necessary to represent the possibility
of change between states that would be contradictory without succession,
and thus the possibility of motion [B48–49].)
For our purposes, two points about these arguments should be noted.
First, while the Transcendental Exposition may assume that we have syn-
thetic a priori cognition of mathematics, specifically geometry, and presum-
ably of a specific geometry, namely Euclidean (although Kant does not
explicitly assert the latter in the present passage), the arguments of the
Metaphysical Exposition make much more general claims, what we might
80 P. Guyer
call topological claims, and do not depend upon the assumption that we have
synthetic a priori cognition of any particular geometry or metric of space.
Second, and perhaps even more important, the arguments as presented thus
far establish, or would if sound, only that we must have an a priori repre-
sentation of space and its form (as unitary and infinite), not that space is
merely a representation or, conversely, that space is not a property of things
in themselves. Yet that is what Kant’s conclusion at A26/B42 seems to claim.
The arguments for that conclusion begin only after the Metaphysical and
Transcendental Expositions of space have been completed, although as I have
said only Kant’s actual arguments for the conclusion can establish how it
should be interpreted.
The first of these arguments is the one that turns on the alleged incompat-
ibility of our synthetic a priori cognition of the properties of space with
spatiality being a property of things in themselves or space itself being such a
thing (again, mutatis mutandis for time). It is not immediately obvious why
the possibility of a priori cognition of properties of an object should imply
that such properties are not properties of that thing in itself, but only of our
representation of the object and thus merely of the way it appears to us. In
everyday cases, after all, we may know that there are certain conditions for
our perception, and thus know, at least relatively a priori (see B2), that an
object that we succeed in perceiving under those conditions really does, or
will, satisfy those conditions. For example, if we know that the unaided
human ear can only detect sounds between approximately 20 Hz and
20 MHz, then it would seem that we can also know, in advance or a priori,
that any particular sounds that we do succeed in hearing must, really, in
themselves, fall between 20 Hz and 20 MHz. Why should we not similarly
infer from the fact that we must represent an outer object spatially in order to
represent it at all, that any particular outer object we do succeed in perceiving
really is spatial, in itself, and that this is why we do succeed in perceiving it?
This objection, or the idea of a “missing alternative,” namely that spatio-
temporality could be a property of both our representations and things in
themselves, has traditionally been associated with the name of Friedrich
August Trendelenburg, although it was already raised by Hermann
Andreas Pistorius in his 1786 review of Johann Schultz’s 1784 Elucidations
of Professor Kant’s “Critique of Pure Reason.”14 But Kant had clearly already
14
Pistorius’s review was originally published in Allgemeine deutsche Bibliothek 66 (1786): 92–123; and is
reprinted in Albert Landau, ed., Rezensionen zur Kantischen Philosophie 1781–1782 (Bebra: Landau,
1991), 326–52; and Bernward Gesang, ed., Kants vergessener Rezensent: Die Kritik der theoretischen und
praktischen Philosophie Kants in fünf Rezensionen von Hermann Andreas Pistorius, Kant-Forschungen
meant to block this objection in the first edition of the Critique, before
Pistorius ever raised it. Kant’s argument is essentially that we know the
synthetic a priori propositions about space and time in general demonstrated
in the metaphysical expositions as well as the specific mathematical proposi-
tions at issue in the transcendental expositions to be necessarily true, that
they are necessarily true of our representations of outer and inner objects,
since space and time are the necessary forms of such representations, but that
if true of things in themselves, these propositions would only be contingently
true of them, which is contrary to the initial premise that they are necessarily
true; hence they cannot be true of things in themselves at all, that is,
spatiotemporality cannot be a property of things in themselves nor space
and time things in themselves of any kind. This is not a conclusion about or
from the concept of things in themselves; it is a conclusion about things in
themselves, themselves. The omission of spatiotemporality from the concept
of things in themselves follows from the denial of spatiality to those things
themselves.
Kant was so committed to this argument that he put it first in his “General
remarks on the transcendental aesthetic” and repeated it in the Prolegomena
and elsewhere. In the “General remarks,” Kant starts from the example of a
geometrical proposition supposedly known a priori and thus as necessarily
true, in this case the Euclidean proposition that an enclosed figure can be
constructed only with three straight lines, but the same argument could
apply more generally to the not specifically geometrical propositions of the
Metaphysical Exposition, such as that two simultaneously existing objects
can be intuited as distinct only by being intuited as located in different
regions of space, or two events not located in different regions of space can be
intuited as distinct only by being intuited in successive moments of time. In
any case, here is the argument:
If there did not lie in you a faculty for intuiting a priori; if this subjective
condition regarding form were not at the same time the universal a priori
condition under which alone the object of this (outer) intuition is itself
Band 18 (Hamburg: Meiner, 2007), 3–25. Trendelenburg made the objection in 1840 in the first
edition of his Logische Untersuchungen (Berlin: Bethge); in the third edition of 1870 (Leipzig: Hirzl), the
objection is found at vol. 1, 158–70. In Frederick Beiser’s words, the objection is that “all Kant’s proofs
for the subjectivity of space and time in the ‘Transcendental Aesthetic’ do not exclude the possibility that
they [space and time] are also valid of things-in-themselves” (Late German Idealism: Trendelenburg and
Lotze [Oxford: Oxford University Press, 2014], 112). Beiser’s discussion of Trendelenburg’s debate over
this objection with Kuno Fischer extends from pp. 108–21 of this book. Another recent discussion is
Andrew Specht, “F. A. Trendelenburg and the Neglected Alternative,” British Journal for the History of
Philosophy 22, no. 3 (May 2014): 514–34.
82 P. Guyer
possible; if the object (the triangle) were something in itself without relation to
your subject: then how could you say that what necessarily lies in your
subjective conditions for constructing a triangle must also necessarily pertain
to the triangle in itself ? for you could not add to your concept (of three lines)
something new (the figure) that must thereby necessarily be encountered in the
object, since this is given prior to your cognition and not through it. (A48/
B65–66, emphasis added)
Kant’s argument is that while spatiotemporality in general or the specific

geometrical property that he describes might be necessarily true of our
representations, they would not be necessarily true of objects independent
of our representations; but they are necessarily true throughout their
domains, so they must not be true at all of objects that are independent of
our representations, for of them they could only be contingently true. Thus
spatiotemporality and all that it involves must be only the subjective condi-
tion of representation, thus “merely the form of all appearances” in the sense
of our representations, but also of those objects as they appear to us but not
as they are in themselves.
It is no fluke that Kant offers us this argument in the Critique of Pure
Reason. We find exactly the same argument in the Prolegomena, although this
time made specifically with reference to geometry, without parentheses:
it absolutely would not follow from the representation of space, a representa-

tion that serves a priori, with all the various properties of space, as foundation
for the geometer, that all of this, together with what is deduced from it, must
be exactly so in nature. . . . because it is simply not to be seen how things would
have to agree necessarily with the image that we form of them by ourselves and
in advance. If, however, this image – or better, this formal intuition – is the
essential property of our sensibility by means of which alone objects are given
to us, and if this sensibility represents not things in themselves but only their
appearances, then it is very easy to comprehend, and at the same time to prove
incontrovertibly: that all outer objects of our sensible world must necessarily
agree, in complete exactitude, with the propositions of geometry, because
sensibility itself, through its form of outer intuition (space), with which the
geometer deals, first makes those objects as possible, as mere appearances. (Pro
4:287, emphasis added)
This time Kant can confine his argument to the propositions of geometry
rather than just using a geometrical proposition as an example because, in
accordance with the “analytical” method of the Prolegomena, his entire
argument for transcendental idealism starts from the assumption that we
have synthetic a priori cognition in mathematics, which is not the case on the
“synthetic” method of the Critique (see Pro 4:274–75).15 That detail aside,
Kant’s argument is otherwise the same as in the Critique: if the propositions
at issue were true of objects other than our representations, they could at best
be known to be contingently, not necessarily true of those objects; but they
are known to be necessarily true throughout their domain, so they cannot be
true of mind-independent things at all, but only of our representations of
them, thus of appearances as representations and of mind-independent
objects only insofar as they are appearances, that is, only insofar as they
appear to us through our representations of them.
A further passage from Kant’s unpublished notes shows how committed
he was to this argument:
If no space were given a priori in our subject as the form of its sensible intuition
and objects outside us were merely given in this form, then no synthetic
propositions that at the same time hold of actual outer objects would be
possible a priori. For if we were to derive the representations from the objects as
they are given in themselves, then everything would depend merely on experience
and no synthesis would hold a priori together with the necessity of the judgments,
at least not objectively. (NF 18:271 [R 5637], emphasis added)
Once again, the claim is that if the synthetic propositions about space that we
know a priori and therefore know to be necessarily true were to hold of
mind-independent things as well as of our own representations of them, we
could only know them to be contingently true of the former, and therefore
could not know them to be necessarily true throughout their domain after
all. But we do, so they must not be true of things in themselves at all.
Erich Adickes’s preferred dating for the note I have just quoted is 1780–
1783, so it might have been made in preparation for the composition of the
Prolegomena and not demonstrate Kant’s commitment to this argument for
transcendental idealism beyond the second edition of the Critique, where the
passage from the “General remarks” is preserved unchanged. But Kant
continues to suggest the same argument a decade later. Here is a passage
from his drafts for a submission to the Berlin Academy competition on the
question “What real progress has metaphysics made in Germany since the
15
On the contrast between the methods of the Prolegomena and Critique, see my “The Prolegomena and
the Critique of Pure Reason,” in Kants “Prolegomena”: Ein kooperativer Kommentar, ed. Holger Lyre and
Oliver Schliemann (Frankfurt am Main: Klostermann, 2012), 277–98.
84 P. Guyer
time of Leibniz and Wolff?” This draft was written in the period 1792–1795,
although Kant never submitted it to the Academy. He writes:
it is not the form of the object, as it is in itself, which makes intuition a priori
possible, but rather that of the subject, namely the form of sense, of that kind
of representation which he, the subject, is capable of. For if this form were to be
taken from the object itself, we would first have to perceive this, and could
become aware of its nature only in this perception. But that would then be an
empirical intuition a priori. But whether it be the latter or not is something of
which we can persuade ourselves, as soon as we attend to whether the judgment
which attributes this form to the object carries necessity with it, or not; for in
the latter case it is merely empirical. (RP 20:266–67)
Kant’s claim is that “the judgment which attributes” the form of spatiotem-
porality “to the object” does “carry necessity with it.” Therefore, it cannot be
a judgment supposed to be true of “the object itself,” for in that case it would
only be contingent and known merely empirically, so it must be true only of
the representation of the subject, of which it can be necessarily true and
known a priori. Once again, this argument is about what properties things in
themselves can, or more precisely, cannot have, and not in the first instance
an argument about the concept of things in themselves. It introduces a
concept of such things that lacks reference to spatiotemporality only because
such things themselves must lack spatiotemporality.
Kant thus remained committed to this argument throughout his mature
career. It seems clear to me that this is Kant’s preferred argument for
transcendental idealism, precisely because it blocks the missing alternative
objection that Pistorius and Trendelenburg would raise. This is not to say
that it is a very good argument. It is open to obvious questions: How do we
know that the propositions about spatiotemporality in general and mathe-
matics in particular are necessarily true throughout their domain? Why could
they not be necessarily true of our representations of things but only con-
tingently true of things themselves? Or, to put the objection another way, if
the premise is that synthetic a priori knowledge is possible only of represen-
tations or appearances and not of things in themselves, why can we not
construct the modus tollens argument that, since spatiotemporality can be
only contingently and not necessarily true of things in themselves, then we
cannot have synthetic a priori cognition of the relevant propositions through-
out their domain after all, just as well as the modus ponens argument that
Kant actually offers? I do not see a non-question-begging answer to this
question. But Kant himself clearly did not doubt his own argument, and the
main point is, again, that it is clearly leads to the ontological conclusion that
things in themselves are not spatiotemporal and not merely to an abstraction
from spatiotemporality in the concept of things in themselves.
Although this was Kant’s preferred argument for transcendental idealism,
he did have other arguments. We will have to consider these more briefly,
but the important point is that all of these too lead to the ontological denial
that things in themselves are spatiotemporal rather than the merely concep-
tual point that we can form a concept of things that abstracts from their
spatiotemporality. Kant adds two further arguments to the one just
expounded in the second edition of the Critique. The first is that all our
intuition “contains nothing but mere relations, of places in one intuition
(extension), alteration of places (motion), and laws in accordance with which
this alteration is determined (moving forces)”; but “through mere relations
no thing in itself is cognized”; therefore outer and inner sense contain in their
representations “only the relation of an object to the subject, and not that
which is internal to the object in itself” (B66–67).16 This argument, that
because spatiotemporal properties are relations but things in themselves can
have no relations (which is the obvious inspiration for Rae Langton’s inter-
pretation of “Kantian humility”),17 relies on a definition of things in them-
selves as completely non-relational that Kant here pulls out of thin air. It is
also incompatible with several of his most fundamental claims about things
in themselves, namely that they “affect” us in our empirical intuitions and, in
the practical sphere, that they are the ultimate locus of our choice, which
consists in nothing less than the relation of our power of choice (Willkür) to
maxims. But in spite of these problems, it is clear that this argument leads to
an ontological conclusion, the denial of spatiotemporal relations to things in
themselves, not to a merely conceptual abstraction.
The same is true of the second argument that Kant added to the second
edition of the Transcendental Aesthetic. This is the argument that since
16
Kant suggests that there are two exceptions to the major premise of this argument, namely “the feeling
of the pleasure and displeasure and the will [den Willen]” (B66). This may not have much bearing on
Kant’s present argument, but his exception of the feeling of pleasure and pain is inconsistent with his
“transcendental definition” of pleasure as a state in which we wish to remain and pain a state that we
wish to leave (e.g., FI 20:230–31; CJ 5:222). Both pleasure and pain then involve a relation between
among the present state of a subject and its wishes concerning its future state. The case of the will may be
more complicated: if Kant means by this Willkür, that is clearly relational, involving the subject’s choice
of (relation to) maxims, fundamental or otherwise; if by Wille Kant means, as he later does, pure
practical reason, then perhaps what he is alluding to is his position that the pure practical reason, unlike
merely prudential practical reason, is not determined by any relation to an external end but by its own
form, and in this sense is non-relational.
17
See note 6 above.
86 P. Guyer
God’s intuition of objects is not spatiotemporal, then objects as they are in

themselves cannot be spatiotemporal, so if for this reason “one will not make
[the conditions of space and time] into objective forms of all things, then no
alternative remains but to make them into subjective forms of our kind of
outer as well as inner intuition.” Actually, Kant goes even further, and claims
that if space and time were properties of things in themselves then God
himself would have to be spatiotemporal – “as conditions of all existence in
general they would also have to be conditions for the existence of God”
(B71–72)! This is an obvious fallacy, assuming as it does that if spatiotem-
porality were a genuine property of any things in themselves then it would
have to be a property of all. Even apart from this last step, the argument still
rests on a theological premise that has no place in Kant’s theoretical philo-
sophy, and seems a desperate shot at the final buzzer. But whatever its force,
it is clearly an argument that things as they really are cannot be spatiotem-
poral, not that we can form a concept of things that abstracts from their
spatiotemporality. Thus all of Kant’s arguments in the Transcendental
Aesthetic point to the ontological rather than merely conceptual interpreta-
tion of transcendental idealism.
Finally, in the Transcendental Dialectic, specifically the Antinomy of Pure
Reason, Kant offers what he calls an indirect rather than direct proof of
transcendental idealism. There is hardly space here to deal with this in detail,
but my conclusion is that this argument also requires the ontological rather
than merely conceptual interpretation. As is well known, in the first two,
“mathematical” antinomies, Kant argues that if space and time were real
properties of things in themselves, then they would have to be both finite and
infinite in extent and divisibility, which is of course impossible; but if space
and time are regarded merely as indefinitely extendable and divisible forms of
intuition (not infinite given magnitudes, by the way!), not applying to things
in themselves at all, then neither of the contradictory propositions in each
pair has to be asserted. In the third and fourth, “dynamical” antinomies,
Kant argues that there is a contradiction between the endless causal deter-
minism required by spatiotemporality and the spontaneity required by
reason, as well as between the endless chain of contingency required by the
former and the necessary being posited by the latter, but that antinomial
contradiction can be avoided if we confine spatiotemporality to the mere
appearances of things and allow the possibility that things as they are in
themselves are not spatiotemporal. Then there is room for spontaneity and
necessary existence in the domain of things in themselves. For example, in
the case of spontaneity, specifically the spontaneity of an “acting subject,” the
solution to the Third Antinomy is that “this acting subject, in its intelligible
character, would not stand under any conditions of time, for time is only the
condition of appearances but not of things in themselves” (A539/B567).
This argument would not work if temporality were excluded merely from the
concept of things as they are in themselves, for then such things could still be
temporal and still be subject to ordinary causality even though that is not
contained in or implied by such a concept of things in themselves. The
argument has to be that things in themselves may really be non-spatiotem-
poral, for only then may real room for spontaneity and necessary being be
preserved. This is not the place to evaluate Kant’s indirect proof of trans-
cendental idealism in the Antinomy, but the point is just that the Antinomy
clearly requires an ontological rather than merely conceptual interpretation
of transcendental idealism.18
Both Kant’s direct and indirect proofs of transcendental idealism, then,
can only be understood as arguments for the ontological rather than merely
conceptual version of the doctrine.
What? again
Kant’s arguments for transcendental idealism thus lead to what writers such
as Lucy Allais and Karl Ameriks19 have recently called a “moderate” meta-
physical interpretation. The moderate metaphysical position holds that there
are things that exist independently of our representations of them, that these
things necessarily appear to us in certain ways, namely spatiotemporally, but
that these things do not themselves have the properties they appear to us to
have, namely spatiality and temporality. Yet, the interpretation continues,
they must have some properties, or there must be a way they are, and the
latter must in some way ground the former, that is, the properties they have
must ground the way they appear to us to be, even though we cannot know
what these properties are and thus cannot know how they ground the way
they appear to us or the properties they appear to us to have. For example,
18
I have tried to show how the antinomies can be resolved without transcendental idealism in Kant and
the Claims of Knowledge (Cambridge: Cambridge University Press, 1987), ch. 18 (pp. 385–415). For
other arguments that Kant’s resolution of the antinomies requires what I am calling the ontological
rather than merely conceptual interpretation of transcendental idealism, see Robert Merrihew Adams,
“Things in Themselves,” Philosophy and Phenomenological Research 57, no. 4 (Dec. 1997): 801–25; and
Karl Ameriks, “Kant’s Idealism on a Moderate Interpretation,” in Kant’s Idealism, 38–40.
19
In addition to “Kant’s Idealism on a Moderate Interpretation,” which is also reprinted in Ameriks’s
recent collection Kant’s Elliptical Path (Oxford: Clarendon Press, 2012), 75–99, see in the same
collection “On Reconciling the Transcendental Turn and Kant’s Idealism,” 100–119.
88 P. Guyer
Allais says that “transcendental idealism must be understood as containing

substantial metaphysical commitments: to the mind-dependence of things as
they appear to us, and to the existence of an aspect of reality that grounds the
appearances of things, and which we cannot cognize.”20 This position is
metaphysical insofar as it asserts the existence of mind-independent things in
addition to our mind-dependent representations of things. It is doubly
moderate, first insofar as it does not, what would be radical, deny the
existence of such things, and second insofar as it does not, what would
likewise be radical, hold that we can have knowledge of the properties of
such things by means of pure (theoretical) reason as contrasted to intuition.
(Of course, Kant argues that we can specify the properties of certain things in
themselves, namely God and our own free wills, on practical grounds, and
assert the existence of these things on practical grounds as well. That is
another story.21) The moderate metaphysical interpretation of transcenden-
tal idealism is the position that we can, indeed must, assert the existence of
things that do not depend on our representations even though we can only
determinately attribute to them properties that do depend on our representa-
tions, since all the properties that we attribute to things are spatial and/or
temporal in form and, Kant holds, spatiality and temporality themselves are
representation-dependent properties. We have no reason not to suppose that
things in themselves do have properties and that these properties ground the
properties that they appear to us to have, but we cannot say what the
properties of things in themselves are apart from our representation of
them or how they ground the properties that things appear to have in virtue
of the way we represent them.
I have no doubt that this is general form of Kant’s own understanding of
transcendental idealism. His clearest statement of it came in the Prolegomena
to Any Future Metaphysics, in a set of remarks added to its version of the
Transcendental Aesthetic in which Kant strove to distinguish his position
from the “subjective idealism” of Berkeley, with which his position had been
identified in the Göttingen review of the first edition of the Critique of Pure
Reason (redacted by Johann Georg Heinrich Feder from a draft by Christian
Garve).22 Here Kant writes:
20
Allais, Manifest Reality, 7.
21
Robert Adams in particular has emphasized the practical dimension of Kant’s transcendental idealism.
See Adams, “Things in Themselves,” 813–16.
22
The review drafted by Garve but redacted by Feder appeared in the Zugabe zu den Göttingischen
Anzeigen von gelehrten Sachen, 19 January 1782, 40–48; and is reprinted in Landau, Rezensionen, 10–17.
Translations of both the published version and Garve’s original text, which was subsequently published
There are things given to us as objects of our senses existing outside us, yet we
know nothing of them as they may be in themselves, but are acquainted only
with their appearances, i.e., with the representations that they produce in us
because they affect our senses. Accordingly, I by all means avow that there are
bodies outside us, i.e., things which, though completely unknown to us as to
what they may be in themselves, we know through the representations which
their influence on our sensibility provides for us, and to which we give the
name of a body – which word therefore merely signifies the appearance of this
object that is unknown to us but is nonetheless real. Can this be called
idealism? It is the very opposite of it. (Pro 4:289)
Why is this position the very opposite of idealism? Because, Kant explains,
what I called [transcendental] idealism did not concern the existence of things
(the doubting of which, however, properly constitutes idealism according to
the received meaning), for it never came into my mind to doubt that, but only
the sensory representation of things, to which space and time above all belong;
and about these last, hence in general about all appearances, I have only shown:
that they are not things (but mere modes of representations), nor are they
determinations that belong to things in themselves. (Pro 4:293)
Here Kant claims that there are things that not only appear to exist outside of us
(outside of our empirical selves) in virtue of appearing spatiotemporal, but also
that these things do exist independently of us and our representations, that these
things affect or influence us, specifically our sensibility, yet that while spatiality
and temporality are the necessary features (“above all”) of all our representations,
in virtue of which things do appear to exist outside us, these properties are not
things in themselves nor determinations of them. Thus though the existence of
things in themselves is known to us, their properties – “what they may be in
themselves” – are not, except as the indeterminate grounds of the influence or
affection of those things on us. This is the moderate metaphysical position that
Allais, for example, ascribes to Kant: it asserts rather than denies the existence of
things in addition to and independent from our representations of them, but also
denies that these things can really be spatiotemporal and thus, since all the
properties we can know are spatiotemporal, denies that we can know what the
properties by means of which these things influence or affect us are.
in Allgemeine deutsche Bibliothek, supplement to vols. 37–52 (1783): 838–62, can be found in Brigitte
Sassen, ed., Kant’s Early Critics: The Empiricist Critique of the Theoretical Philosophy (Cambridge:
Cambridge University Press, 2000), 53–77. Feder subsequently defended his own, empiricist approach
to both sensibility and understanding in Über Raum und Caussalität (Göttingen: Dieterich, 1787).
90 P. Guyer
This is Kant’s position, and whatever might be said about the strength of
the arguments in its behalf, it is coherent. I have been critical of those
arguments, not just here but in work extending back to Kant and the Claims
of Knowledge (1987). But in closing, let me make one point in Kant’s behalf.23
Many have worried that the influence or affection of things in themselves
upon our representations that Kant assumes must be “noumenal causality,”
but that Kant’s position is then incoherent precisely because causality is
necessarily a spatiotemporal relation – succession in accordance with a rule –
and thus a relation in which non-spatiotemporal objects cannot stand. My
view is that this is an objection that Kant needlessly brought upon himself by
including causality rather than the relation of ground and consequence on the
initial list of unschematized categories. If he had included only the abstract
relation of ground and consequence, the objective correlate of the logical form
of hypothetical judgment, on the initial list of categories and reserved actual
causality for the subsequent list of schemata, as he did when he included
“inherence and subsistence” on the list of categories (A80/B106) but reserved
the temporally-inflected concept of enduring substance for the list of sche-
mata (A144/B183), he could have spoken, as indeed he often did, of things in
themselves as the ground although not the spatiotemporal cause of our
representations without risk of self-contradiction. When he says that we can
conceive of things in themselves without intuition although we cannot cognize
them without intuitions, he is assuming precisely that we can conceive of
them by means of unschematized categories, since the schemata arise by
adding intuition to the unschematized categories. Thus I think that the
moderate metaphysical interpretation of transcendental idealism is coherent.
Still, that is the most that I think can be said in behalf of transcendental
idealism. Dear as it was to Kant’s heart, and in spite of the opening it gave him
to revive a commitment to a libertarian conception of free will that he himself
had originally rejected,24 the arguments for it remain profoundly problematic.
23
See also Adams, “Things in Themselves,” 819–20.
24
See Kant’s first publication in pure philosophy, the NE of 1755, Proposition IX, 1:400–405.
5
Noumenal Ignorance:
Why, for Kant, Can’t We Know
Things in Themselves?
Alejandro Naranjo Sandoval and Andrew Chignell
Nothing is so firmly believed, as that which we least know.

– Michel de Montaigne, Essays
Introduction
Kant is famous for distinguishing between things as they appear to us
(“appearances [Erscheinungen],” phenomena), and things as they are “in
themselves” or as we think of them (Dinge an sich, noumena).
Transcendental Idealism is the thesis (roughly) that some of the most impor-
tant features of appearances –including their spatiotemporal-causal features –
originate not in the world but in the perceiving or cognizing mind itself.
Noumenal Ignorance is the following related but distinct thesis:
Noumenal Ignorance: We cannot cognize or have any substantive positive

synthetic knowledge about particular things in themselves.
A. Naranjo Sandoval
Department of Philosophy, Princeton University, Princeton, USA
e-mail: ans@princeton.edu
A. Chignell
Department of Philosophy, University of Pennsylvannia, Philadelphia, USA
e-mail: chignell@sas.upenn.edu

92 A. Naranjo Sandoval and A. Chignell
It is crucial to note that the truth of Noumenal Ignorance is consistent with

the claim that we have knowledge of analytic truths regarding things in
themselves (such as: things in themselves are things). It is also consistent
with the claim that we have very general synthetic knowledge about them
(such as: some things in themselves affect us and thereby ground appear-
ances) and/or negative knowledge about them (such as: things in themselves
are not in space and time in the way that appearances are). Furthermore,
Noumenal Ignorance is consistent with the claim that we can think coherently
about things in themselves (hence: “nou-mena”) and that some of those
thoughts are preferable to others for theoretical and/or practical reasons.
Finally, Noumenal Ignorance is consistent with the claim that we can have
rational belief or “faith” (Vernunftglaube) concerning the substantive proper-
ties of particular things in themselves. We do not propose to defend the truth
of any of these claims here. We just want to point out that they are consistent
with Noumenal Ignorance.
It is also crucial to note that Noumenal Ignorance is not entailed
by Transcendental Idealism. One might think that the spatiotemporal-
categorial structure of the world is not fundamentally real, and thus
count as a transcendental idealist, and yet still hold that we can know
some informative truths about whatever is fundamental. Kant himself
held this combination of positions in the 1770 text typically referred to
as “The Inaugural Dissertation” (officially titled Dissertation on the Form
and Principles of the Sensible and the Intelligible World). For the 1770
Kant, space and time are a function of the transaction between mind and
world, but ultimate or “intelligible” reality is still at least partially
knowable.
By 1781, Kant has changed his mind about the latter point and argues, in
the first Critique, that fundamental reality “in itself” is not knowable (or,
more precisely, that we cannot know anything substantive and positive about
particular things in themselves). In the first part of this chapter, we look at a
few leading Anglophone interpretations of Kant’s argument for Noumenal
Ignorance. We group these into two different camps: the first camp contains
commentators who presuppose a “one-world” view of Transcendental
Idealism; the second contains commentators who work with a “two-world”
view. Although this grouping has been useful to us in discussing the state of
play regarding Noumenal Ignorance, it is not our goal here to discuss the
debate between one-world and two-world theorists in any detail.1
1
For more on Transcendental Idealism, see Paul Guyer’s chapter in this volume.
5 Noumenal Ignorance 93
In the second part of the chapter, each of us sketches his own account of
Noumenal Ignorance. The first (Naranjo Sandoval) emphasizes some of the
logical and semantic themes related to this doctrine and claims that the
reason we cannot have knowledge of things in themselves is that to know
objects we must be able to refer to them – in a sense explained later on – and
that we cannot, in this same sense, refer to things in themselves. The second
(Chignell) suggests that Kant’s evolving views about the metaphysics and
epistemology of modality in the critical period are what ultimately motivate
his commitment to Noumenal Ignorance.
As may already be apparent, there are many different roads leading to
Noumenal Ignorance. It is possible – even likely – that Kant himself
recognized this, and that this explains why he seems to defend the
doctrine differently in different passages. But, again, the main goals of
the present chapter are to present some important recent accounts of the
doctrine and then articulate what we each take to be Kant’s most
significant argument for it.
One-world ignorance
Allison: Discursivity + Epistemic conditions = Ignorance
Henry Allison’s seminal book, Kant’s Transcendental Idealism: An

Interpretation and Defense, appeared in 1983 and initiated a vigorous reap-
praisal of the nature, scope, and cogency of Kant’s arguments for
Transcendental Idealism and Noumenal Ignorance. Although the view
Allison defended was original in many important respects, it also drew on
(and made accessible in English) Gerold Prauss’s influential picture articu-
lated earlier in Erscheinung bei Kant (1971) and other writings. Because the
view ascribes a one-world metaphysics to Kant and thus avoids what many
analytic philosophers regard as the most pernicious forms of idealism,
Allison’s interpretation also drew new readers to the first Critique. After
twenty years of discussion and criticism, Allison released a heavily revised
version of his book; this 2004 edition is the one from which we will quote
here.
Allison’s central thesis is that Noumenal Ignorance follows from
(i) Kant’s commitment to what Allison calls “the discursivity thesis” and
(ii) Kant’s way of cashing out the distinction between things in themselves
and appearances.
Regarding (i): The “discursivity thesis” says that human cognition arises
from the combined operation of the receptive faculty of sensibility and the
spontaneous faculty of understanding via concepts.2 Allison does not think
that Kant explicitly argues for the first part of the thesis – namely that objects
have to be “given” to us in what he calls “sensible intuition.” But he does
believe Kant endorses that broadly empiricist point in passages like this: “It
comes along with our nature that intuition can never be other than sensible,
i.e., that it contains only the way in which we are affected by objects. . . .
Without sensibility no object would be given to us” (A51/B75).
According to Allison, the arguments of the Transcendental Aesthetic
constitute a case for this first part of the thesis, since they conclude with a
commitment to what Allison calls “sensible epistemic conditions.” For
Allison, an epistemic condition is a condition under which a subject with
faculties like ours is able to represent, cognize, and know about a domain of
objects.3 The epistemic conditions imposed by the structure of sensibility,
according to Kant, are twofold: all of the objects or states we can cognize
must exist along a single temporal continuum, and all of the outer objects or
states we can cognize must be related to one another in a unitary space.
But, Allison emphasizes, “sensible intuition alone is not sufficient to yield
cognition of objects . . . it provides the data for such cognition but does not
itself amount to cognition.”4 Hence the second part of the discursivity thesis:
there are conceptual as well as sensible epistemic conditions on human
cognition. Thus the arguments of the Transcendental Analytic are designed
to show that the “material” or “data” given to us in intuition must be such
that it is able to be brought under general rules or concepts. The universality
of this feature, which Allison calls the “original orderability” of the sensible
material, can obtain only on the assumption that such orderability is “a
contribution of the cognitive subject.”5
Allison takes the following quote from Kant to reveal as much: “The
capacity (receptivity) to acquire representations through the way [durch die
Art] in which we are affected by objects is called sensibility” (A19/B33,
emphasis added). This is pretty slim evidence, but Allison takes “way” here
2
“To claim that human cognition is discursive is to claim that is requires both concepts and sensible
intuition” (Henry E. Allison, Kant’s Transcendental Idealism: An Interpretation and Defense, rev. ed.
[New Haven: Yale University Press, 2004], 13).
3
“By an epistemic condition is here understood a necessary condition for the representation of objects,
that is, a condition without which our representations would not relate to objects or, equivalently,
possess objective reality” (ibid., 11).
4
Ibid., 14.
5
Ibid.
to refer to the manner in which sensible intuition is given – it is and must be

given in a “way” that makes it susceptible to conceptualization, and at least
part of this “way” is determined by the formal structures of the mind.
Regarding (ii): Allison says that the distinction between appearances and
things in themselves has to be understood in relation to the epistemic condi-
tions discussed under (i). That is, to talk about “appearances [Erscheinungen]”
or empirical “objects [Objecte]” is to consider entities as satisfying the sensible
and conceptual epistemic conditions under which alone we are able to cognize
them. To talk about “things in themselves” is to consider the same entities
independently of those epistemic conditions.6 It follows from this, almost by
definition, that while there is only one world of entities, these things cannot be
cognized when considered independently of these conditions (that is, they
cannot be cognized as things in themselves).7 Allison does not say much about
the difference between cognition and knowledge, but since he calls these
epistemic conditions, he presumably thinks that without cognition of things
in themselves, we also cannot have substantive synthetic knowledge of them.
Langton: Receptivity + Irreducibility = Ignorance
Other commentators develop accounts that, like Prauss’s and Allison’s, take
Kant to be positing one “world” or set of entities, but depart from them in
offering fundamentally metaphysical rather than epistemological motivations for
the doctrine of Noumenal Ignorance. Rae Langton (and more recently Lucy
Allais, Tobias Rosefeldt, Kristopher McDaniel, and one of the present authors
[Chignell]) argue that the appearance/thing in itself distinction is ultimately
grounded in a distinction between types of properties. Although the differences
between these various “metaphysical one-world” pictures are illuminating and
worth discussing, here we will focus on Langton’s account as articulated in
Kantian Humility: Our Ignorance of Things in Themselves (1998).8
6
Allison follows Prauss in providing a series of key quotations in which Kant himself speaks of the
distinction as one about how one and the same set of entities is “considered [betrachtet]” – either under
epistemic conditions or not (ibid., 52–57).
7
This does not mean that, for Allison, Noumenal Ignorance is a trivial or purely conceptual truth. For, as
he correctly points out, “this would follow only if the distinction [between appearances and things in
themselves] were itself obvious or trivial. But this is far from the case with the transcendental distinction,
which . . . rests upon a radical reconceptualization of human knowledge as based on . . . epistemic
conditions” (ibid., 19).
8
See Lucy Allais, “Kant’s One World: Interpreting ‘Transcendental Idealism,’” British Journal for the
History of Philosophy 12, no. 4 (Nov. 2004): 655–84; Lucy Allais, Manifest Reality: Kant’s Idealism and
According to Langton, Kant’s claim that we cannot know things in

themselves is best understood as the claim that we cannot know the
“intrinsic natures” of the (one world of) substances that exist. That is, we
cannot know the intrinsic properties of things (properties that are instan-
tiated even in worlds where the substance is completely alone).9 Langton’s
argument for this appeals to texts that suggest that the distinction between
things in themselves and appearances seems to just be the distinction
between substances in respect of their intrinsic properties and substances in
respect of their key extrinsic properties.10 She then argues that our minds are
affected by another substance in respect of the latter’s relational (and thus
extrinsic) properties, but not by its intrinsic properties. (Affection is a
relation, after all.) In Kant’s words, appearances “consist entirely of rela-
tions” (A285/B341), and “we have no insight into the inner in things”
(A277/B333). More expansively:
Now through mere relations no thing in itself is cognized; it is therefore right

to judge that since nothing is given to us through outer sense except mere
representations of relation, outer sense can also contain in its representation
only the relation of an object to the subject, and not that which is internal to
the object in itself. (B67)
Clearly if all of our perceptual knowledge of the world comes by way of

receptivity to what affects us, then the only way we could know about
things in themselves is by inferring from knowledge of relational proper-
ties to knowledge of intrinsic properties. In other words, the fact that
perception involves an affection relation between world and mind does
not just entail that the only properties that can be known (especially by
way of inference) are relational. Rather, Langton’s Kant also believes, contra
His Realism (Oxford: Oxford University Press, 2015); Tobias Rosefeldt, “Dinge an sich und sekundäre
Qualitäten,” in Kant in der Gegenwart, ed. Jürgen Stolzenberg (Berlin: de Gruyter, 2007), 167–209; Kris
McDaniel, “A Philosophical Model of the Relation between Things in Themselves and Appearances,”
Noûs 49, no. 4 (Dec. 2015): 643–64; and Andrew Chignell, “Skepticism, Common Sense, and Mind-
Dependence,” in The Sensible and Intelligible Worlds, ed. Karl Schafer and Nicholas Stang (Oxford:
Oxford University Press, forthcoming).
9
See Rae Langton, Kantian Humility: Our Ignorance of Things in Themselves (Oxford: Clarendon, 1998),
18. Langton quotes from the Kemp Smith edition, which sometimes translates Kant’s das Innere und
Äussere as “the intrinsic and the extrinsic.” The Guyer/Wood translation uses more literal (and possibly
less technical) terminology: “the inner and the outer” (A265/B321).
10
See, for example: “The inner determinations of a substantia phaenomenon in space . . . are nothing but
relations, and it is itself entirely a sum total of mere relations. . . . As object of the pure understanding, on
the contrary, every substance must have inner determinations and forces that pertain to its inner reality”
(A265/B321).
Leibniz, that relational (extrinsic) properties cannot be ontologically

reduced to (that is, do not even supervene on) categorial (intrinsic)
ones. This means that we have no inferential route from facts about the
one to facts about the other, and thus cannot know things in themselves.
In Langton’s words:
Receptivity implies that we can have knowledge of what can affect us.
Irreducibility implies, in Kant’s view, that intrinsic properties cannot affect
us. If Kant maintains this view throughout his philosophical career – as I shall
argue he does – then one can trace a path to humility [i.e., Noumenal
Ignorance]. If substances affect us, but it is not through their intrinsic properties
that we are affected, then their intrinsic properties remain unknown.11
One of the biggest concerns about Langton’s view, raised most prominently
by Karl Ameriks, is that this “short argument” to humility (from receptivity
and irreducibility) seems to leave no role for Kant’s idealism about space and
time to play, not to mention the other key arguments of the Aesthetic and
Analytic. Kant himself, though, seems to indicate that idealism is part of
what ultimately underwrites Noumenal Ignorance (which Langton calls
“humility”). This may not be a concern for Langton in the end, since she
openly admits that her metaphysical one-world picture is not meant to be a
form of genuine idealism after all.12
Two-world ignorance
Van Cleve: Receptivity + Reducibility = Ignorance
In his influential study of the first Critique, Problems from Kant (1999), James
Van Cleve allows that, for Kant, we can know about some of the “relational
or structural features of things in themselves” – such as that they ground
appearances.13 Hence he agrees with Langton that what Kant means by
11
Langton, Kantian Humility, 5.
12
For Ameriks’s criticism, see Karl Ameriks, “Kant and Short Arguments to Humilty,” in Interpreting
Kant’s “Critiques” (Oxford: Clarendon, 2003), 135–57. For Langton’s assessment of her own view,
consider: “[My interpretation] does ascribe to a notorious idealist a position that is not idealism, not
anti-realism of any kind, but rather epistemic humility” (Langton, Kantian Humility, 6).
13
James Van Cleve, Problems from Kant (Oxford: Oxford University Press, 1999), 136. Van Cleve also
suggests that it might be possible to know that there is some structural similarity (if not outright
isomorphism) between facts about things in themselves and facts about appearances (ibid., 155–62).
Noumenal Ignorance is only that we cannot have knowledge of the intrinsic

properties of fundamentalia.
Van Cleve’s argument for this reading of Noumenal Ignorance makes use of
the following components:
(i) A “two-world” account of the distinction between things in themselves

and appearances.
(ii) A specific interpretation of Kant’s theory of relational and intrinsic
properties.
(iii) Kant’s claim that we can only know objects through their powers to
affect us.
With respect to (ii) and (iii), Van Cleve’s view is similar to Langton’s, as
we will see in a moment. But with respect to (i) there is a big difference: Van
Cleve follows other prominent Anglophone commentators (such as Paul
Guyer, Jonathan Bennett, and P. F. Strawson) in ascribing full-blown
metaphysical idealism to Kant (although Van Cleve seems to find this part
of Kant’s philosophy more palatable than these earlier commentators do).
According to this kind of idealism, the existence and key features of appear-
ances ontologically depend on minds that can cognize them, while the
existence and intrinsic features of things in themselves do not.14 As evidence,
Van Cleve points to passages in which Kant gives a deflationary account of
appearances, such as: “appearances do not exist in themselves, but only
relative to the same [cognizing] being, insofar as it has senses” (B164). He
also cites those in which Kant expresses the counterpart claim regarding
things in themselves: “that [something] should exist in itself without relation
to our senses and possible experience, could of course be said if we were
talking about a thing in itself” (A493/B521–22). It follows, according to Van
Cleve, that appearances are phenomenal – all truths about them are reducible
to or “derivable from truths about states of perceivers.”15 This also means
that appearances and things in themselves are of different ontological kinds:
there are “two worlds” that are metaphysically distinct – the phenomenal
world and the noumenal world.
14
“In dozens of passages, Kant tells us that appearances have no being apart from being represented. . . .
Things in themselves, by contrast, are things that exist independently of human representation or
cognition. They exist whether perceived or not and have whatever properties they do independently of
us” (ibid., 6–7).
15
Ibid., 11.
One of the main differences between these two worlds has to do with
whether the objects in them consist of mere relations or not. For Van Cleve,
as for Langton, appearances are constituted by all and only relational properties.
So when we cognize them we are cognizing relations that they bear to each
other, to themselves and their parts, as well as to space and time and to our
minds. As Kant says: “everything in our cognition that belongs to intui-
tion . . . contains nothing but mere relations, of places in one intuition (exten-
sion), alteration of places (motion), and laws in accordance with which this
alteration is determined (moving forces)” (B66–67).
Van Cleve and Langton also agree that this is not the case with things
in themselves: they do have intrinsic or inner properties. Van Cleve’s
argument to this conclusion relies on the view that Kant endorses a
broadly Leibnizian view of relations, according to which the relational
properties of things in themselves supervene on their intrinsic proper-
ties.16 He reads Kant as saying as much when he states: “if I think of
mere things in general, then the difference in the outer relations certainly
does not constitute a difference in the things themselves, but rather
presupposes this” (A280/B336). If this is correct, then it follows that
knowing the relational properties of things in themselves (as they affect
us) is not to know everything about them: further and more fundamental
facts about things in themselves always remain.
At this point, the crucial claim is (iii): we only know objects in virtue of
the dispositional properties they have to affect us or other objects.17 Van
Cleve does not explicitly cite his evidence for attributing this view to Kant.
Arguably, however, he has in mind passages where Kant claims that intuition
is needed for the cognition of objects – as in the well-known dictum:
“Thoughts without content are empty, intuitions without concepts are
blind.” Intuitions, moreover, are “the way in which we are affected by
objects” (A51/B75).
The key is that, for Van Cleve, such dispositional properties are not
intrinsic to objects – they are either relational properties themselves or
“powers . . . that are manifestable in [these relational properties].”18 This
16
Van Cleve claims that this thesis “underlines two of his assumptions in the ‘Amphiboly’ section of the
Critique of Pure Reason: things in themselves cannot have their nature exhausted by relations, and they
cannot differ numerically without differing qualitatively” (ibid., 47). Even though they both agree on
the conclusion – that is, that things in themselves have intrinsic properties – this is in sharp contrast to
Langton’s views, according to which Kant originally held the reducibility of relations in the pre-critical
period but later on came to believe that relational properties are irreducible to intrinsic ones.
17
“The only properties we know of in external things are their powers to affect us” (Van Cleve, Problems
from Kant, 153).
is because dispositional properties do not satisfy the definition of intrin-

sicality that Van Cleve attributes to Kant, according to which an intrinsic
property is one that the object has in every world in which it exists.19
The sugar cube’s disposition to dissolve in water, for example, is not
intrinsic, since it is a power that sugar has only in worlds with laws
similar to ours. In other words, there are worlds in which sugar cubes
exist and yet the laws of nature differ such that the sugar cubes do not
dissolve in water.20
So Van Cleve’s Kant thinks, in sum, that we can only know objects in
virtue of knowing the dispositional properties they have to affect us, but
those properties (or powers in which those properties are “manifestable”)
are not intrinsic – they are relations between objects and minds. If our
knowledge is always of non-intrinsic manifestations of powers, then we
cannot know anything about the intrinsic properties of things. This does
not threaten our knowledge of phenomenal objects, however: after all,
their very existence consists in such relational properties – that is all
there is to know about them. The opposite is the case, however, with
things in themselves: they are more than mere relations – they have
intrinsic, non-relational features. Given Kant’s account of cognition it
looks like substantive knowledge of those features, at the very least,
always escapes us.
But a puzzle remains: Why could we not “recover” some information
about the intrinsic properties of things by knowing them through their
dispositions to affect us? That is, given that Van Cleve holds that the
relational properties of things in themselves are reducible to their intrinsic
properties, it is not clear why we cannot infer from our knowledge of the
former to at least some knowledge of the latter. Here Langton’s view may
have the advantage: her claim that relational properties of things in them-
selves are irreducible blocks off any inferential path toward knowledge of the
intrinsic properties of things.
19
More precisely, Van Cleve references Moore’s test: P is intrinsic if and only if “when anything
possesses it, that same thing or anything exactly like it would necessarily or must always, under all
circumstances, possess it in exactly the same degree” (ibid., 152).
20
Some metaphysicians would disagree with Van Cleve here and claim that dispositions are categorial,
intrinsic properties that are only “triggered” in the right kinds of circumstances.
Hogan: Freedom + Practical knowledge = (Theoretical)

ignorance
Desmond Hogan is not primarily concerned to offer a reconstruction of

Kant’s arguments for Noumenal Ignorance; rather, he seeks to explain how
Kant could both describe things in themselves as unknowable and yet also
claim certainty (and thus knowledge) of some substantive features of them.
Indeed, the kind of ignorance claim that he focuses on is unusual because it
results from a substantive metaphysical claim. This metaphysical claim also
plays a key role in Kant’s argument for the subjectivity of space and time and
in this way indirectly helps to motivate his more general restriction of our
knowledge to appearances.21
According to Hogan, when Kant says that we cannot have knowledge of
things in themselves, he at times simply means that this knowledge exceeds
our cognitive capacities and at others that there is no ground through which
certain key features of things in themselves could be known. Hogan’s focus is
on the latter claim – although it is important to note that he does not think
that Noumenal Ignorance is exhausted by it.
Hogan’s reconstruction relies on three key ideas:
(i) At times, “a priori knowledge” means, for Kant, “knowledge through the
ground.”
(ii) There are no grounds through which free actions can be known.
(iii) For the critical Kant, we have practical knowledge that we are free.
Regarding (i): Hogan recognizes that in some passages “a priori knowl-

edge” means “non-empirical knowledge.” He plausibly contends, however,
that in other passages the phrase refers to a concept that Kant inherited from
his rationalist predecessors – namely, “knowledge through the ground.” On
this understanding, we know something a priori “when we understand it
‘through’ the thing or things that ground it metaphysically.”22 So someone
knows a priori that the sun rises every morning only if she knows a priori the
complex facts about planetary motion that ground our observation of
sunrises.
21
All citations are from Desmond Hogan, “How to Know Unknowable Things in Themselves,” Noûs
43, no. 1 (March 2009): 49–63. Hogan has expanded on some of the details of his view in Desmond
Hogan, “Noumenal Affection,” Philosophical Review 118, no. 4 (Oct. 2009): 501–32.
22
Hogan, “How to Know Unknowable Things in Themselves,” 52.
Hogan offers both textual and philosophical reasons to think that Kant
was, at least sometimes, trading on this secondary meaning of “a priori
knowledge.” One of the key passages is found in the Transcendental
Aesthetic: “time could not precede the objects as a determination or order
attaching to the things themselves as their condition and be cognized and
intuited a priori through synthetic propositions” (A33/B49). Setting aside
some complexities, Hogan’s claim is that Kant is not only making the
epistemological point that, if the fundamental structure of time depended
on things in themselves, then any a priori knowledge of this structure would
exceed our cognitive capacities. Rather, he is also making the metaphysical
point that there would be no grounds through which synthetic propositions
about the structure of time could be non-empirically known. This suggests
(according to Hogan) that the sense of a priori at play here is “knowledge
through the ground.”23
Hogan’s innovative suggestion, then, is that Kant’s claim that we cannot
have knowledge of things in themselves means (at least sometimes) that there
are important features of reality that simply lack a ground through which they
could be non-empirically known. Most significantly, Hogan argues, Kant
follows the Pietist theologian and philosopher Christian August Crusius in
holding that the free acts of human beings necessarily lack a ground through
which they can be known: “Kant does not merely think that a priori knowledge
of free acts exceeds our cognitive powers. He rather means to point out that in
the case of a feature of reality lacking a determining ground there is nothing
through which it could be cognized in principle.”24
As evidence for this reading of the argument for Noumenal Ignorance, Hogan
points to passages in which Kant expresses the incompatibilist sentiment that,
on the assumption that we are free, there are no antecedent causes determining
our actions and hence no grounds through which they could be known25:
23
In fact, these two senses of “a priori knowledge” are related. For Hogan, Kant held that knowledge
through the ground depends on non-empirical knowledge precisely because we cannot know via
empirical observation alone the causes or grounds of things – non-empirical principles are ultimately
required for this purpose. In our example regarding sunrises, some causes could be known empirically,
but in the last analysis one would have to invoke either the Principle of Contradiction, if the truths
involved are necessary, or the Principle of Sufficient Reason. This allows Hogan to read some of Kant’s
epistemic claims that we cannot non-empirically know some things – which is by far the most frequent
formulation of Noumenal Ignorance – as metaphysical claims that there are no grounds through which
these things could be known.
24
Hogan, “How to Know Unknowable Things in Themselves,” 56.
25
Note that this interpretation of Noumenal Ignorance is not limited to the unknowability of facts
regarding agents. In conversation Hogan has indicated that, since free acts are unknowable, and since
[In the case of free actions] the conditions are lacking under which reason can
comprehend [einsehen] something; these are the determining grounds. But our
free actions have no determining grounds, thus we also cannot comprehend
them. This is a ground for comprehending the limits of the understanding
[Dieses ist ein Grund, die Schranken des Verstandes einzusehen], but not for
denying the matter. (LM 28:271)26
Note that Hogan’s reading also depends on (iii): the claim that Kant held
that we can have (practical) knowledge (Wissen), and not merely belief
(Glaube), that we are free. He presents evidence for that conclusion from
Kant’s mature writings; for example27:
[There is] one idea of reason (which is . . . incapable of theoretical proof of its
possibility) among the facts, and that is the idea of freedom, the reality of
which, as a particular kind of causality . . . can be established through practical
laws of pure reason. . . . It is the only one among all the ideas of pure reason
whose object is a fact [deren Gegenstand Tatsache ist] and which must be
counted among the scibilia. (CJ 5:468)28
So Hogan’s overall picture is that Kant thinks we can know (practically,

at the very least) that we are free, and this prevents us from having a
priori knowledge (that is, knowledge through the ground) of any of our
free acts. Since the phenomenal world has been shown, in other parts of
the Critique, to be fully deterministic, these unknowable free acts must
occur at the level of things in themselves, and the knowable world of
appearances must be ontologically distinct from the world of things in
themselves. Thus the doctrine that things in themselves are not in space
agents can freely act on (some) features of reality, it follows that those features of reality are unknowable
too, in the sense that they lack grounds through which they could be known.
26
Another clear passage presented by Hogan is: “[contingent changes such as free actions], in so far as
they appear to have about them an indeterminacy in respect of determining grounds and necessary laws,
harbour within themselves a possibility of deviating from the general tendency of natural things towards
perfection” (OPA 2:110–11).
27
Hogan is in good company in this regard: Allison and Ameriks have also defended the claim that, at
the time Kant wrote the first edition of the first Critique, he believed that there is theoretical proof that
we are absolutely free. See Karl Ameriks, Kant’s Theory of Mind: An Analysis of the Paralogisms of Pure
Reason (Oxford: Clarendon, 1982), 189–98; and Henry E. Allison, Kant’s Theory of Freedom
(Cambridge: Cambridge University Press, 1990), chs. 1–3.
28
Hogan recognizes that in passages like this it seems that absolute freedom could only be known to be
real if we know that the moral law is binding. Hence it seems that absolute freedom has only practical
(and not theoretical) grounding. But, he alleges, Kant holds that facts that have practical grounding have
the same epistemic standing as facts that have theoretical grounding (CPrR 5:121; RP 20:310).
and time is “something against which not the least ground for uncer-
tainty can be raised” (Pro 4:289); it is “not just a mere hypothesis, . . .
but a demonstrated truth” (RP 20:268). Likewise, “it is . . . indubitably
certain, and not merely possible or even probable, that space and time, as
the necessary conditions of all (outer and inner) experience, are merely
subjective conditions of all our intuition” (A48–49).
Hogan’s emphasis on freedom as a motivation for one of Kant’s signature
doctrines raises an exegetical concern. If Kant really thought that our
inability to know anything a priori about free action is part of the motive
for Noumenal Ignorance (not to mention Transcendental Idealism), then why
did he not just come out and say it in the early parts of the Critique? Why did
he focus almost entirely on the arguments from geometry, chronometry, and
the principles of pure understanding instead?
A second concern is that the view ends up giving us a surprising amount of
knowledge of things in themselves. Yes, we cannot know a priori precisely
what an incompatibilistically free agent did or will choose, but we do know
(practically) that there is such freedom (as Hogan-cum-Kant claims), that it
must exist at the non-deterministic level of things in themselves, and that it
must be exemplified by minds or mind-like substances (the bearers of free
will). This makes it seem as though the spirit (if not the letter) of Noumenal
Ignorance has been left behind.
The logical and semantic side of noumenal

ignorance (Naranjo Sandoval)
Let us take a step back. Recall that Kant’s official rationale for why we cannot
know things in themselves is that in order to cognize an object we need to
have both intuitions of it and concepts that apply to it:
In the analytical part of the critique it is proved . . . that we have no concepts of

the understanding and hence no elements for the cognition of things except
insofar as an intuition can be given corresponding to these concepts, conse-
quently that we can have cognition of no object as a thing in itself, but only
insofar as it is an object of sensible intuition, i.e. as an appearance. (Bxxv-xxvi;
see also B165, B146)
Kant clearly says that we can think about things in themselves – that is,
subsume them under concepts in accordance with purely logical laws –
and so what we lack in order to cognize such things are intuitions of
them.29 This immediately raises two questions: First, what does Kant mean
when he says that we do not have intuitions of things in themselves? Second,
why are intuitions required for us to cognize – and ultimately know – objects?
A full answer to the first question is one of the crown jewels of any
satisfactory interpretation of Transcendental Idealism. For such an answer
would involve, among other things, an account of the distinction between
things in themselves and appearances, as well as a picture of the kind of
knowledge that can result from our interaction – through either affection or
causation – with them. Given the intense interest in Kant’s idealism,
discussions of Noumenal Ignorance have largely focused on answering the
first question. The second question, however, is frequently neglected. In my
view, this is a serious omission: to understand why our lack of intuitions of
things in themselves precludes our knowing them it is not enough to show
that and why there is this lack. Rather, we also need to understand why
intuitions are necessary for knowledge of things in the first place. The way
to do this, I think, is to study and contrast the role that intuitions and
concepts play in Kant’s theories of cognition, truth, reference, and
knowledge.30
The central goal of my interpretation, then, is to remedy this omission.31
Fortunately, this does not require wholesale revisions of most previous
accounts of the motivations behind Noumenal Ignorance, since the aim in
those discussions has been to answer the first question, whereas my aim is to
answer the second. Van Cleve and Langton, for example, agree that
Noumenal Ignorance is the result of Kant’s claim that we have knowledge
of objects insofar as they affect us – and thus only if we have intuitions of
them, since intuition “contains . . . the way in which we are affected by
objects” (A51/B75). They do not specify, however, how or in what sense
representations with intuitional content encode this information, or why
this intuitionally-encoded information is required to have knowledge of
objects.
29
“Even if we cannot cognize [erkennen] . . . objects as things in themselves, we at least must be able to
think [denken] them as things in themselves” (Bxxvi–xxvii).
30
The view sketched here is a simplified version of an interpretation that I (Naranjo Sandoval) am
currently preparing for presentation elsewhere. Here, I focus mostly on Kant’s theory of truth and
cognition. A fuller study, I think, would also consider the role that intuition plays in the justification of
knowledge as a propositional attitude.
31
For an important exception, see Eric Watkins and Marcus Willaschek, “Kant’s Account of
Cognition,” Journal of the History of Philosophy 55, no. 1 (Jan. 2017): 83–112. See also the replies to
that article (in the same volume) by Stefanie Grüne and Andrew Chignell.
In my view, the answer to this question can be located in the following two
ideas. First, Kant holds that in order for epistemic agents to know an object they
must be able to refer to it. Second, he believes that only representations that
have intuitional content can genuinely refer to objects. Naturally, this assumes a
technical sense of “reference” that will be explained below, one which is more
stringent than the contemporary sense – for according to the latter certain
definite descriptions, which Kant would consider purely conceptual representa-
tions, succeed in referring (for example, consider the description “the first man
to walk on the moon”). Understanding these two claims, as well as Kant’s
reasons for holding them, will occupy us for the rest of this section.
A final prefatory remark: Kant is not always clear about which kinds of
mental representation he takes to be candidates for substantive knowledge – at
times he uses the genus “cognition [Erkenntnis],” at times the species “judgment
[Urteil].” Even if one takes Kant’s more specific usage at face value, as I will
here, it is not clear that all judgments could in principle count as substantive
knowledge – this is not the case with analytic judgments, for example. For this
reason, I use the neutral term “t-judgments” to refer to those judgments which
Kant thinks are capable of counting as substantive, non-trivial knowledge.32
In order to understand the first claim outlined above – that is, that
reference is required for knowledge – it is necessary to note that, for Kant,
t-judgments are essentially representations about individual objects that are
also grounded on individual objects.33 That they are about objects means that
t-judgments contain representations that are related by means of a copula (‘is,’
for example) and that bear an intentional relation of description or expression
to actual or possible objects. That they are grounded on objects means that the
representations contained in t-judgments bear, via intuition, a non-represen-
tational relation of grounding or determination to actual individual objects.34
For example, the judgment “my copy of the first Critique is on the table”
contains a representation of my copy of the first Critique. In turn, this
representation is related through intuition to my copy of the book – and to
32
The “t” in “t-judgments” stands for “theoretical,” although I do not claim here that all theoretical
judgments are t-judgments and vice versa. Some general features of t-judgments include that they have
content and are truth-evaluable attempts at describing the world. However, commentators will disagree
on further details.
33
Here I believe that Kant uses “object” to refer to any state of affairs or feature of reality which could be
cognized – these might include propositions. For a similar take on what Kant means by “object” in the
context of his theory of cognition, see Houston Smit, “Kant on Marks and the Immediacy of Intuition,”
Philosophical Review 109, no. 2 (April 2000): 239–43.
34
Here I remain neutral on the nature of this relation, since explaining it would require addressing
complicated issues concerning intuition and intuitional content.
no other object. In Kant’s words: “Judgment is therefore the mediate cogni-

tion of an object, hence the representation of a representation of it. In every
judgment there is a concept that holds of many, and that among this many
also comprehends a given representation, which is then related immediately to
the object” (A68/B93, emphasis added).35
The hope, then, if one aims for true t-judgments, is that the representa-
tions contained in them accurately describe the objects which ground them. I
believe that this natural thought is the guiding principle of Kant’s theory of
truth. Accordingly, in the introduction to the Transcendental Logic, he
unambiguously states his commitment to a correspondence theory, accord-
ing to which t-judgments are true only insofar as their informational features
correspond to the features of the objects which ground them: “What is
truth? The nominal definition of truth, namely that it is the agreement of
cognition with its object, is here granted and presupposed” (A58/B82).
For our purposes, it is most relevant that Kant draws the following
conclusion from his commitment to a correspondence theory: for t-judg-
ments to be true they must contain representations that not only accurately
describe their objects but also distinguish them from all others. For short, we
will say that these representations truly refer to their objects, where we can
think of a representation which truly refers as one which is “tailor-made” to
its object – it “fits” that very object and no other. Kant’s reasoning for this
claim is best explained through an illustration.
Suppose I form the true t-judgment “the painting acquired by the
Metropolitan Museum is an original Vermeer.” Since my t-judgment is truth-
evaluable, it must contain a representation of the painting in front of me that
distinguishes it from other similar objects as a necessary condition of its truth.36
For suppose it did not – suppose my representation of the original is informa-
tionally identical to my representation of the counterfeit. Then Kant’s correspon-
dence theory would not only explain the truth of my t-judgment but also – per
35
For more on the relation between judgments and their objects in the context of Kant’s views on logic,
see John MacFarlane, “Frege, Kant, and the Logic in Logicism,” Philosophical Review 111, no. 1 (Jan.
2002): 50–51.
36
This is not to say that we can consciously distinguish between similar objects – for example, between
the original and the counterfeit. It means only that our representations possess sufficient informational
complexity to individuate objects from all others, provided that we can form true t-judgments about
them. Note, too, that the sense of “containment” involved here is not linguistic. That is, the sentence
expressing the content of my t-judgment contains the expression “the painting acquired by the
Metropolitan Museum”; but my t-judgment also contains a representation of the painting (whether
the content of this representation can be expressed by that definite description depends on one’s views
on the semantics of descriptions). I hope to clarify this relation of containment between the judgment
and its component representations elsewhere.
impossibile – its falsity! For the t-judgment could be considered a representation of

both the original and the counterfeit, even though it is only grounded on the
former.37 As a representation about the original, the correspondence between
representation and object would render the t-judgment true; as a representation
of the counterfeit, it would render the t-judgment false.38 Kant summarizes this
argument soon after subscribing to a correspondence theory:
[Since] truth consists in the agreement of a cognition with its object, then this
object must thereby be distinguished from others; for a cognition is false if it
does not agree with the object to which it is related even if it contains some-
thing that could well be valid of other objects. (A58/B83; see also JL 9:51)
It seems, then, that in order for t-judgments to be true they must contain
representations that truly refer to their objects. It follows, under the uncon-
troversial assumption that knowledge entails truth, that to have knowledge of
an object one must have representations that truly refer to it.39 The key to
my interpretation, then, is the additional claim that it is only possible for
representations to truly refer if they have intuitional content.40
The argument for this key claim proceeds by elimination, as do many of
Kant’s other arguments to the conclusion that sensibility has a certain feature.
We start with the observation that we have representations that truly refer to
37
Note that this means that Kant does not hold a causal (or in this case “grounding”) theory of
reference. The object of the t-judgment is the ground of the t-judgment’s truth or falsity, but whether
the t-judgment fails or succeeds at referring to that object depends only on the informational features of
its representations.
38
I am left with the difficult task of explaining how certain representations describe one object and no
other. I believe that Kant solved this problem by postulating a thoroughgoing correspondence between a
true t-judgment and its object; that is, if a t-judgment about an object is true then it must contain a
representation that expresses accurately and exhaustively the features of the object. This would be a
natural choice given Kant’s doctrine of thoroughgoing determination, according to which, given a
predicate P, any object falls under the extension of P or its negation. See, for example, A571/B579. I
think it is possible to avoid worries about the finiteness of our cognitive capacities by claiming that this
thoroughgoing correspondence obtains between the object and the intuitional content of the representa-
tion, which might not be introspectible in its entirety. This would agree with our claim here that only
representations with intuitional content can truly refer to objects. The defense of this claim, however,
will have to await another occasion.
39
For example, in his analysis of Kant’s notion of knowledge (Wissen), Andrew Chignell includes the
requirement of truth. See Andrew Chignell, “Kant’s Concepts of Justification,” Noûs 41, no. 1 (March
2007): 57.
40
It should be noted that, for Kant, the ability to distinguish objects from each other is not only
necessary for knowledge but also epistemically beneficial for other lower, less demanding propositional
attitudes, such as hypothesis and, more broadly, opinion, the epistemic merits of which come in degrees.
That is, insofar as agents have representations that better distinguish objects from others, their cognition
of these objects is epistemically improved. Elaborating on this would be part of the fuller project hinted
at in note 30.
objects – this much is required, as we have seen, for the truth of substantive
judgments. We know, too, that the mind contains two cognitive faculties:
sensibility – the receptivity to the world that results in sensations and ultimately
intuitions – and understanding – the ability to spontaneously think about the
objects of these intuitions by way of concepts (A50/B74). Furthermore, these
faculties are independent of one another: neither can perform the functions of
the other (A52/B76). Through either the understanding or sensibility, then, we
must be able to form representations that truly refer to objects. So if it can be
shown that the understanding is incapable of providing these representations –
as I believe it can – then our result follows.
In my view, the best way to show this is to highlight the hierarchical
structure of concepts that Kant introduces in his lectures on logic and in
a section of the Appendix to the Dialectic titled “On the regulative use
of the ideas of pure reason.”41 This structure is based on the extensional
notion of a “mark contained under a concept,” which means a repre-
sentation that instantiates the concept but is not identical to it. For
example, the concept “Siamese cat” is contained under the concept “cat”
(see VL 24:911; JL 9:95). Kant proceeds to define the following hier-
archy on these marks: “concepts are called higher (conceptus superiores)
insofar as they have other concepts under themselves, which, in relation
to them, are called lower concepts” (JL 9:96).
For present purposes, the relevant point is that each of our representations
of objects has a determinate, well-defined location in this hierarchical struc-
ture. That is, it is a fact about logic that for any representation r of a given
individual object, and given any concept F, either r falls under F or it does
not.42 In the Vienna Logic Kant presents the following examples: “All
[humans] are virtuous or vicious as to character, men or women as to sex,
young or old as to age, learned or unlearned as to cognition” (VL 24:927; see
also JL 9:147). If this is right, then for any representation of an individual
object there corresponds a unique list of concepts under which it falls.43
41
For a discussion of Kant’s hierarchical structure of concepts, see Eric Watkins, “Kant on Infima
Species,” in Kant und die Philosophie in weltbürgerlicher Absicht: Akten des XI. Internationalen Kant-
Kongresses, ed. Stefano Bacin, Alfredo Ferrarin, Claudio La Rocca, and Margit Ruffing, 5 vols. (Berlin: de
Gruyter, 2013), 5:283–94.
42
For Kant’s doctrine of thoroughgoing determination, see A571/B579.
43
In fact, uniqueness is only guaranteed if we consider ordered lists of concepts, that is, lists such that, for
every two contiguous concepts in them, the second is a species of the first. For example, one’s
representation of an individual cat is correlated to the following ordered list of concepts: thing, living
thing, animal, mammal, four-legged, meowing, etc. We also require lists to be finite, given concerns
about the finiteness of our conceptual cognitive capacities. For the sake of simplicity, this is left implicit
here.
But does the converse also hold? That is, can some such list correspond to
a unique individual object? If so, then this list would be a conceptual
representation that truly refers to its object, since it would accurately repre-
sent it and no other. In turn, this would mean that the understanding has the
capacity to form representations which truly refer, since through it we can
represent objects under the hierarchy of concepts.44
It will come as no surprise that I think that this is not the case. According to
Kant, the human understanding “needs manifoldness and variety in things
despite their agreement under the same genus” (A654/B682); that is, it requires
that every concept, regardless of its specificity, contain under it a multitude of
numerically distinct marks. We could always envisage by means of concepts
distinct possible individuals falling under conceptual descriptions – even those
as specific as “a Roman general who crossed the Rubicon and defeated
Pompey,” for example. This has the important consequence that “every
genus requires different species, and these subspecies, and since none of the
latter once again is without a sphere [eine Sphäre], (a domain as conceptus
communis), reason demands in its entire extension [Erweiterung] that no species
be regarded as in itself the lowest” (A655/B683; see also VL 24:911; JL 9:97).45
Since every genus can be divided into subspecies, and since these sub-
species necessarily have a non-empty extension, there can be no lowest
species, that is, no species that is not also a genus.46 This is enough to
guarantee that any given list of concepts in the hierarchical structure cannot
truly refer to a unique object – the last concept on the list will always contain
further marks under it, each of which has a non-empty extension, and so the
list will accurately describe a multitude of objects.
44
It is necessary to note that the understanding orders concepts in this hierarchical structure because it
abides by the logical principles of homogeneity, specification, and continuity (A658/B686). Given that
these logical principles are regulative, they are also necessary rules about the features of representations
and the relations between them. From this it follows that the understanding must order concepts in this
way, and so there is no way, independently of the hierarchical structure, in which the understanding
could form representations that truly refer.
45
The idea here seems to be that it is always possible to conceive of numerically distinct marks under a
given concept, regardless of its content. In accordance with this reading, Kant states that the reason why
there is no lowest species is that “as soon as I have a concept that I apply to individua, it would still be
possible for there to be still smaller differences among the individua” (VL 24:911). If this interpretation
is correct, this lends credence to Chignell’s views, as presented in the next section, that Kant’s Noumenal
Ignorance is at least partly informed by his views on modality.
46
Note that this marks a departure from Leibnizian doctrine, according to which each individual object
has a complete concept such that, if one understood it perfectly, one would be able to derive from it all
concepts under which it falls. For example, see G. W. Leibniz, “Discourse on Metaphysics,” in
Philosophical Essays, trans. Roger Ariew and Daniel Garber (Indianapolis: Hackett, 1989), 51.
As promised, then, we have established that through the understand-

ing alone we cannot form representations that truly refer to objects. As
I have argued above, this entails that only representations with intui-
tional content can truly refer. There is also clear textual evidence that
Kant was thinking along these lines. Kant’s denial of a lowest species is
frequently accompanied by remarks suggesting that intuitions play the
logical role that lowest species would perform if they existed and, in
particular, that they succeed in referring to actual individual objects.
Kant explicitly condones this conclusion in his argument against a
lowest species in the Vienna Logic, where he places intuition at the
root of the hierarchical structure because sensible representations are
related to individual objects:
The conceptus infimus cannot be determined. For as soon as I have a concept

that I apply to individua, it would still be possible for there to be still smaller
differences among the individua, although I make no further distinction. Now
in this gradation, one concept is always higher than another, until I come to the
highest. The lowest cognition is intuition, because it is always concerned with
something unique. (VL 24:911, final emphasis added; see also A656/B684)47
At the beginning of this section, we considered the following question: why

do we need intuitions of objects in order to know them? The answer that I
have defended here is, in short, that for truth – and hence for knowledge –
epistemic agents require representations that truly refer to objects and that
only representations with intuitional content can be said to truly refer (where
“to truly refer” has the technical sense of accurately describing an object and
no other).48 This type of reference is part of what allows t-judgments to be
candidates for knowledge. Because our representations of things in them-
selves do not contain intuitional content, then, they cannot truly refer to
actual individual objects and thus cannot constitute substantive knowledge
about them.
47
Note that I am not only claiming that intuitions are required for representations to refer to a single
object. This is what commentators have called the singularity of intuitions. See, for example, Michael
Friedman, “Kant’s Theory of Geometry,” Philosophical Review 94, no. 4 (Oct. 1985): 455–506.
Rather, I am making the stronger claim that (some) representations with intuitional content contain all
the information required for distinguishing the object from all others – this is the sense in which they are
at the bottom of the hierarchical structure of concepts.
48
For a similar interpretation of Noumenal Ignorance, see Watkins and Willaschek, “Kant’s Account of
Cognition.” However, our accounts differ in important respects, including, for example, the way we
understand reference and our arguments to the conclusion that we can only refer through intuitions.
Modal motives for ignorance (Chignell)

My view of these issues is grounded in Kant’s evolving picture of modality
and our knowledge of it.49 In the Nova Dilucidatio (1755), The Only
Possible Argument (1763), and other pre-critical texts, Kant simply assumes
that things can be “given” to the mind as possible and compossible – given
to us in thought, so to speak, just by way of conceiving them (see OPA
2:77). This is a classic rationalist assumption: think of Descartes clearly and
distinctly conceiving the possibility of a distinct immaterial mind or a
supreme being, or of contemporary modal rationalists taking something
like “ideal positive conceivability” as a guide to possibility. The early Kant
goes on to generate a theistic proof according to which all of these
possibilities we can conceive must be grounded in an actual supreme
being.50
By 1781, however, Kant had given up the assumption that real possibi-
lities are “given” to thought in a non-problematic way such that we can go on
to demand a ground or explanation for them in actuality. We can think up
(denken) various things, and we can individuate at least some of them by
appeal to the different predicates in our concepts of them. So the concepts
are not genuinely empty or nonsensical, despite Kant’s rhetorical flourishes to
that effect. The problem in the critical period is that Kant comes to regard
such thoughts as unable to “give” objects about which we can make knowl-
edge-claims, even on the basis of otherwise “irrefutable” arguments. So what
changed?
What changed, I think, is that during the 1760s Kant became convinced
that there is a metaphysical difference between “logical” and “real possibi-
lity,” and he also came to think that what he there calls “real opposition [reale
Entgegensetzung]” or “real repugnance [Realrepugnanz]” sometimes obtains
between logically consistent positive properties. In the Negative Magnitudes
essay of 1762, for instance, Kant cites numerous examples of a kind of real
opposition between two or more properties that “cancels out” (the verb here
is aufheben) their respective effects: opposed winds on a sail, opposed physical
forces, opposed emotions, and so forth.
49
For the full description of this account, from which some of the material in this section is drawn, see
Andrew Chignell, “Modal Motivations for Noumenal Ignorance: Knowledge, Cognition, Coherence,”
Kant-Studien 105, no. 4 (Dec. 2014): 573–97.
50
For the whole argument, see Andrew Chignell, “Kant, Modality, and the Most Real Being,” Archiv für
Geschichte der Philosophie 91, no. 2 (Jan. 2009): 157–92.
Having noticed that there can be non-logical opposition in empirical

contexts like this, Kant starts to worry (in my view) that something similar
might obtain between supersensibles, too, and that non-logical opposition
in that context might make the thing absolutely impossible. Thus in
Negative Magnitudes and more clearly in the Only Possible Argument of
1763, he cites a few cases that involve what I have elsewhere called subject-
rather than merely predicate-canceling real repugnance. This is a metaphy-
sical opposition between properties that makes any being that would jointly
instantiate them really impossible (think of your favorite Kripke-style cases
here: being red all over and being green all over). Kant’s own examples
include the putative fact that “the impenetrability of bodies, extension
and such like, cannot be attributes of that which has understanding and
will [die Undurchdringlichkeit der Körper, die Ausdehnung u.d.g. können
nicht Eigenschaften von demjenigen sein, der da Verstand und Willen hat].”
It is not that being extended and having a mind are logically inconsistent:
there is no way to generate a contradiction from their conjunction using
standard rules and definitions. Rather, it is that “these predicates can by no
means co-exist together as determinations in a single subject [nimmermehr in
einem einzigen Subject als Bestimmungen neben einander können statt fin-
den]” (OPA 2:85, emphasis added). The “cannot” and “can” in these
sentences refer to real modalities: a thing that is both extended and has a
mind cannot really be.51
By the time of the Critique, Kant saw the rationalist’s neglect of non-
logical constraints on possibility as one of their most serious sins. In the
Phenomena/Noumena chapter, for example, he points out that the real
possibility of something simply cannot be established by mere thinking:
That the not-being of a thing does not contradict itself is a lame appeal to a
logical condition, which is certainly necessary for the concept but far from
sufficient for real possibility. (A244/B302, emphasis added)
In the Amphiboly of the Concepts of Reflection he likewise complains that

with respect to the concept of God they
find it not merely possible but also natural to unite all reality in one being
without any worry about opposition, since they do not recognize any
51
For further defense of these claims about subject-canceling real repugnance in the pre-critical period,
see Andrew Chignell, “Kant and the ‘Monstrous’ Ground of Possibility,” Kantian Review 19, no. 1
(March 2014): 53–69.
opposition except that of contradiction (through which the concept of a thing

would itself be canceled out), and do not recognize the opposition of reciprocal
destruction. (A273–74/B329–30)
The error here is not just metaphysical but also epistemological: Leibniz and
Wolff do not recognize the distinction between logical and real modality and
so do not see that our grasp of the former may not be sufficient for “insight
[Einsicht] into whether all realities could be united together in one object
[Objekt], and hence into how God is possible” (LRT 28:1026).52
This criticism of Leibniz-Wolff applies equally to Kant’s pre-critical self:
the items with which Kant started his pre-critical proof (finite possibilia) and
the item with which he ended it (the most perfect being [ens perfectissimum])
are presupposed a priori to be really as well as logically possible, simply
because we can think consistently about them. But if there are non-logical
constraints on real possibility ‒ constraints that we do not reliably track in
some other way – then that presupposition looks unmotivated.
In my view, it is this recognition of the logical/real modality distinction
that moves Kant to endorse a new condition on cognition: “I can think
whatever I like,” he says, “as long as I do not contradict myself, i.e., as long as
my concept is a possible thought, even if I cannot give any assurance as to
whether or not there is a corresponding object [Objekt] somewhere within
the sum total of all possibilities.” But “to cognize an object, it is required that
I be able to prove its [real] possibility (whether by the testimony of experi-
ence from its actuality or a priori through reason)” (Bxxvi note).
Things in themselves are not given in intuition as actual such that their
real possibility is trivially entailed. They are also not given in mere thought
such that their real possibility is established. Hence we cannot – “whether by
the testimony of experience from its actuality or a priori through reason” –
prove that individual noumena are really possible. For all we know, our
52
In the Metaphysical Foundations of Natural Science, Kant asks us to think of a case of “two motions”
that are “combined in precisely opposite directions in one and the same point” (MFS 4:491). In such a case
the two predicates do not cancel one another out and leave the point at rest (as they would do if we were
merely thinking of opposed forces – see, for example, A265/B321 and OP 22:283). Rather, the
opposition cancels the entire subject to which they are ascribed: “representing two such motions at
the same time in exactly the same point within one and the same space would be impossible, and thus so
would the case of such a composition of motions itself” (MFS 4:491). A few pages later, in a reflection
on this case, Kant explains that “the representation of the impossibility of these two motions in one body
is not the concept of its rest, but rather of the impossibility of constructing this composition of opposite
motions” (MFS 4:494). Similarly, later in the Metaphysical Foundations Kant indicates that a material
being “is impossible through mere attractive forces without repulsive forces,” and that this impossibility
has its basis in “the essence of matter” rather than in a logical contradiction (Demnach ist Materie durch
bloße Anziehungskräfte ohne zurückstoßende unmöglich) (MFS 4:511).
speculative thoughts about such entities are mere flights of fancy – will o’ the
wisp fantasies about things that are really impossible. Furthermore, if sub-
stantive knowledge about particular things in themselves (that is, knowledge
that goes beyond saying that they exist, are not in space and time, are the
ground of appearances, etc.) would have to be based in cognition, and
substantive judgments about such things cannot meet the modal condition,
then Noumenal Ignorance turns out to be true.53
In my view, this concern about our ability to track real modality in mere
thought – a concern that arose out of Kant’s career-long wrestling with
rationalist metaphysics – is crucial to his development of Noumenal
Ignorance. But it is worth noting that this account coheres nicely with the
one put forward by Naranjo Sandoval in the previous section.54 Perhaps the
best thing for us to say, together, is that Kant requires both that we be able to
truly refer to particular entities and that we be able to prove that they are
really possible, in order to have substantive knowledge of them.
Conclusion
In this chapter, we have traced various interpretations of Kant’s doctrine of
Noumenal Ignorance. In some cases, we have discussed its relation to other
key theses of the critical philosophy, such as the existence of epistemic
conditions, or the distinction between appearances and things in themselves;
in others, we have examined theses that previous commentators have not
associated with Noumenal Ignorance, for example, Kant’s doctrine regarding
the difference between real and logical possibility, or his claim that we know
that we are free.
So instead of a unified picture of Kant’s commitment to Noumenal
Ignorance, according to which this doctrine is clearly the result of some single
53
A more precise articulation of the modal condition:
Necessarily, S cognizes that p only if, for any object referred to in p, if it is really possible then S is in
a position to prove its real possibility, and if it is really impossible then S is in a position to prove its
real impossibility.
Obviously more needs to be said about what “proving” real possibility amounts to in this context. For
my latest efforts in that regard, see Andrew Chignell, “Knowledge, Discipline, System, Hope: The Fate
of Metaphysics in the Doctrine of Method,” in Kant’s “Critique of Pure Reason”: A Critical Guide, ed.
James R. O’Shea (Cambridge: Cambridge University Press, 2017), 259–79.
54
See note 45 for a further way in which these two views might complement each other.
argument in the Analytic or the Aesthetic, a much more complicated picture

has emerged. Although commentators can reasonably differ about which of
them is most significant, it seems clear that, for Kant, there are many
theoretical and practical tributaries that lead to the same ocean of
Noumenal Ignorance.55
55
For helpful comments on earlier drafts of this material, we would like to thank Derk Pereboom and
Desmond Hogan.
6
Kant’s Concept of Cognition and the Key
to the Whole Secret of Metaphysics
Chong-Fuk Lau
Kant’s Critique of Pure Reason had an enormous impact on the development of

epistemology. However, certain core concepts in contemporary epistemology
may have rather misguided interpretations of the Critique. A particularly tricky
problem is Kant’s concept of cognition (Erkenntnis). The German word
Erkenntnis was often rendered by early Kant translators as “knowledge” (and
erkennen as “to know”), without being distinguished from another German word,
Wissen.1 This would be a correct translation in ordinary cases, as Erkenntnis and
Wissen have similar connotations in everyday usage and Erkenntnistheorie is also
the German word for “epistemology” or “theory of knowledge.” However,
Erkenntnis and Wissen are two significantly different concepts in Kant’s philoso-
phy. While the former is a more important concept in the Critique, it is the latter
that stands in the center of contemporary epistemology, corresponding to the
concept of propositional knowledge or knowing-that. Rendering Erkenntnis as
“knowledge” would not only blur Kant’s distinction between Erkenntnis and
Wissen, but might also lead to a misunderstanding of Kant’s project. More recent
Kant translators have made a better choice in adopting “cognition” and “to
1
Norman Kemp Smith renders both Erkenntnis and Wissen as “knowledge” in his translation of Kant’s
Critique, which was regarded as the standard for a few decades. See Immanuel Kant, Critique of Pure
Reason, trans. Norman Kemp Smith (London: Macmillan, 1929).
C.-F. Lau (*)

Department of Philosophy, The Chinese University of Hong Kong, Hong Kong
e-mail: cflau@cuhk.edu.hk

118 C.-F. Lau
cognize” as the translations of Erkenntnis and erkennen, thereby reserving “knowl-

edge” and “to know” for Wissen.2
To a certain extent, Kant himself has contributed to the confusion with his
sometimes ambiguous usage of the term Erkenntnis: the term can refer to
objectively valid judgment (B3–4; RP 20:266), but more basically it means
objective representation with consciousness (A320/B377; JL 9:91). I character-
ize the former, more restrictive sense as the judgmental sense of cognition to
contrast it with the latter, more generic one as the representational sense. All
judgmental cognitions are representational cognitions, but not vice versa.
Clarifying the difference and relation between them will help us better under-
stand the central concern of the Critique. Kant does not see the epistemological
issues of truth and justification, which are otherwise central to the concept of
knowledge, as his basic problem. What is essential across various kinds of
cognition is rather an appropriate relation to the object that connects us to the
world and confers objective validity on intuitions, concepts, and judgments.
Kant’s concept of cognition directly addresses the question that he iden-
tifies as “the key to the whole secret of metaphysics” in a letter to his former
student, Marcus Herz, on February 21, 1772:
As I thought through the theoretical part, considering its whole scope and the
reciprocal relations of all its parts, I noticed that I still lacked something
essential, something that in my long metaphysical studies I, as well as others,
had failed to consider and which in fact constitutes the key to the whole secret
of metaphysics, hitherto still hidden from itself. I asked myself this question:
What is the ground of the relation of that in us which we call “representation”
to the object? (C 10:129–30)
Behind the standard epistemological questions of what the nature of knowl-

edge is and how beliefs can be true and justified, there is, for Kant, a more
fundamental question of how representations or cognitions can point beyond
themselves to be about something. “Kant sees,” as Robert Brandom suggests,
“that the epistemological question has semantic presuppositions. The issue for
him is not knowledge, but intentionality.”3 Robert Hanna even considers “the
2
See Immanuel Kant, Critique of Pure Reason, trans. Werner Pluhar (Indianapolis: Hackett, 1996); and
Immanuel Kant, Critique of Pure Reason, trans. and ed. Paul Guyer and Allen W. Wood (Cambridge:
Cambridge University Press, 1998). Kant himself would have agreed to render Erkenntnis in English as
“cognition,” as he often identifies Erkenntnis with the Latin word cognitio (A320/B376–77, A835–36/
B863–64; JL 9:55, 63).
3
Robert Brandom, “Kantian Lessons about Mind, Meaning, and Rationality,” Southern Journal of
Philosophy 44, no. S1 (Spring 2006): 51.
6 Kant’s Concept of Cognition 119
most general transcendental question of all” to be “the Semantic Problem:

how are objective mental representations – that is, cognitions – possible?”4
The semantic turn gives Kant a key to unlock the secret of metaphysics, for
determining how cognition can be related to objects also reveals what sorts of
things can be objects of cognition and what will remain uncognizable. The
intrinsic connection with the concept of object gives Kant’s concept of
cognition a metaphysical dimension that is absent in the contemporary
concept of knowledge.
The following discussion begins with Kant’s concept of knowledge as a
mode of judging alongside with opinion and belief (or faith). It then analyzes
the truth-conditions in Kant’s concepts of knowledge and cognition, respec-
tively. I discuss different senses of cognition in comparison with related
notions such as representation, sensation, and intuition, and I expound on
the concepts of objective validity and real possibility. What is at stake is the
relation of cognition to the object or the condition of representational
purport, which reveals itself as a normative problem. I argue that the
possibility of representational relation depends on a set of necessary rules,
by virtue of which correct or objectively valid representations can be distin-
guished from incorrect or invalid ones. The chapter concludes with remarks
on Kant’s thesis of the uncognizability of things in themselves and the
possibility of extending the realm of cognizable objects beyond possible
experience by the practical use of reason. I show that Kant’s concept of
cognition offers a foundation for a novel metaphysics that has both a
theoretical as well as a practical dimension.
Knowledge, truth, and cognition

In ordinary language, cognition tends to signify the mental processes of
information processing, while knowledge refers to the right product of
such cognitive processes. Taking Kant’s distinction of cognition and knowl-
edge as a distinction between process and product has prima facie plausi-
bility,5 but they differ more fundamentally. For Kant, knowledge is an
4
Robert Hanna, Kant and the Foundations of Analytic Philosophy (Oxford: Clarendon, 2001), 66. See also
Hanna’s analysis of Kant’s concept of cognition in Kant and the Foundations of Analytic Philosophy, 18.
5
See Rudolf A. Makkreel, “The Cognition-Knowledge Distinction in Kant and Dilthey and the
Implications for Psychology and Self-Understanding,” Studies in History and Philosophy of Science 34,
no. 1 (March 2003): 149.
120 C.-F. Lau
epistemic concept applicable only within theoretical philosophy, whereas the

concept of cognition has a much wider scope. In the Doctrine of Method,
knowing is defined as one of the three modes of taking something to be true:
Taking something to be true [Fürwahrhalten], or the subjective validity of

judgment, has the following three stages in relation to conviction (which at the
same time is valid objectively): having an opinion [Meinen], believing
[Glauben], and knowing [Wissen]. Having an opinion is taking something
to be true with the consciousness that it is subjectively as well as objectively
insufficient. If taking something to be true is only subjectively sufficient and is
at the same time held to be objectively insufficient, then it is called believing.
Finally, when taking something to be true is both subjectively and objectively
sufficient it is called knowing. (A822/B850)
These three different modalities of judging can be characterized as proble-

matic, assertoric, and apodictic, respectively, in accordance with Kant’s table
of judgment (JL 9:66). If one makes a judgment without certainty or
conviction, one is merely holding an opinion, which is provisional or
problematic. If one is convinced of the truth of something without objective
certainty, the belief is an assertoric judgment that can better be called faith.6
Only if one’s judgment is based on objectively sufficient grounds can it be
apodictic and called knowledge.
Kant maintains that, in the theoretical domain, the condition of subjective
sufficiency depends on the condition of objective sufficiency. If one has
objectively sufficient grounds, then one knows what one takes to be true; if
one does not, then it is a mere opinion, but never a belief in the sense of faith.
Faith is confined to domains in which it is not just the case that one fails to
acquire objective evidence, but that there is no objective evidence for the
matter at all. More precisely, faith belongs to the practical domain. Whereas
Kant restricts the concept of knowledge within the theoretical domain in a
way that excludes the possibility of practical knowledge,7 he accepts the
existence of practical cognition. Cognition is a much broader concept than
knowledge in a number of ways. I will return to the practical dimension of
6
Guyer and Wood render the German word Glauben into English as “believing” or “belief.” Since the
concept of belief in contemporary epistemology means taking something to be true regardless of whether
it is justified or not, it is less restrictive than Kant’s concept of Glauben, which is a kind of belief that is
sufficient only subjectively but not objectively. Pluhar makes a better choice in rendering Kant’s Glauben
as “faith,” which is usually understood as believing on insufficient evidence.
7
Although Kant does not accept practical knowledge (praktisches Wissen), he allows the possibility of
practical science (praktische Wissenschaft) (CPrR 5:8), which consists of practical cognitions.
cognition at the end of this chapter and focus on its theoretical dimension
first.
Cognition is not included as one of the three modes of judging
something to be true because it is not necessarily a judgment.
Cognition can mean something similar to propositional knowledge,
when it is conceived as “a judgment from which proceeds a concept
that has objective reality, i.e., to which a corresponding object can be
given in experience” (RP 20:266). I call this the judgmental sense of
cognition, which refers to a judgment with objective reality or validity.8
When Kant talks about theoretical or empirical cognition, he is mostly
using the concept of cognition in the judgmental sense.
However, even in the judgmental sense, cognition is still different from
knowledge. Above all, they differ in their relation to the condition of truth. In
contemporary epistemology, truth is recognized as a necessary condition of
propositional knowledge.9 One knows that p only if it is true that p. While
knowledge is defined in terms of truth, Kant does it the other way around and
defines truth through cognition: “The nominal definition of truth . . . is the
agreement of cognition with its object” (A58/B82). Cognition is thereby under-
stood as a judgment with objective validity, which is not necessarily a true
judgment. Thus cognition can be either true or false, depending on whether or
not the judgment agrees with what it is about. More precisely, “a cognition is false
if it does not agree with the object to which it is related even if it contains
something that could well be valid of other objects” (A58/B83). Whereas “false
knowledge” is a contradiction in terms, “false cognition” is not.10
Kant certainly maintains that what is false is not knowledge, but he does
not define knowledge explicitly in terms of truth. Instead, he opts for a
definition of knowledge by means of epistemic concepts alone, replacing
truth with the condition of objective sufficiency. Conceptually, if knowledge
is defined as a kind of belief that is objectively and subjectively sufficient,
8
Although Kant sometimes uses the notion of objective reality in a stronger sense than that of objective
validity, the two notions are often used synonymously. I will treat them as equivalent in this chapter.
9
Knowledge is classically defined as justified true belief, which, however, has been shown to be
inadequate by Edmund Gettier’s famous counterexamples. It is debatable whether it is possible to
offer an adequate definition of knowledge by modifying the condition of justification or replacing it with
other conditions, or even whether the concept of knowledge is analyzable, but epistemologists rarely
dispute that knowledge is a kind of true belief.
10
The difference is in line with the ordinary usage of the two terms, as noted by G. E. Moore: “The
word ‘cognition’ itself is sometimes confined, as its etymology suggests, to awareness or consciousness of
what is true, in which case it is equivalent to ‘knowledge.’ But a ‘false cognition’ would not be so
generally recognised as a contradiction in terms, as ‘a false experience’ or ‘false knowledge’” (“Experience
and Empiricism,” Proceedings of the Aristotelian Society 3 [1902–1903]: 83).
122 C.-F. Lau
knowledge can still, in principle, fail to be true, as objectively and subjec-

tively sufficient grounds do not necessarily guarantee truth. For example, the
standard theory of elementary particles may well be based on sufficient
grounds, but like any scientific theory, there is always a possibility that it
may one day be proven false. This possibility indicates that truth cannot be
guaranteed by epistemic conditions. Nevertheless, Kant chooses to define
knowledge by epistemic concepts instead of truth, and the reason for it may
lie in Kant’s critical attitude toward the traditional conception of truth.
Kant’s reservation is that the correspondence conception is a correct but
merely “nominal definition of truth,” which fails to provide a “general and
certain criterion of the truth of any cognition” (A58/B82).
More concretely, Kant complains that the traditional correspondence
conception unavoidably falls into “a miserable circle” (A58/B82). If truth
consists in the agreement of cognition with an object, there must be a way to
compare them, determining whether the cognition in me agrees with the
object outside me. But Kant believes that the comparison would be trapped
in a circle because “I can compare the object with my cognition . . . only by
cognizing it. Hence my cognition is supposed to confirm itself” (JL 9:50).11
In response, Kant suggests that the agreement-relation has to be supplemen-
ted by epistemic criteria that make the distinction of truth from falsehood or
illusion possible. In the case of empirical cognition, Kant identifies “the
connection of appearances necessarily determining one another in accor-
dance with universal laws, which one calls nature” as “the mark of empirical
truth” (A451/B479). The truth of a judgment is thus determined by whether
it coheres with other judgments according to the rules that determine the
structure of possible experience (Pro 4:290). This kind of epistemic criterion
provides judgments with objectively sufficient grounds that distinguish them
as knowledge from mere opinion and faith.
For Kant, truth is implicitly a necessary condition of knowledge, but the
same does not apply to cognition. Truth is “an objective property” that a
cognition can have or fail to have (JL 9:65). Truth belongs to a set of four
properties that constitute what Kant calls a perfect cognition. The four
properties are classified according to the general structure of the table of
categories: “A cognition is perfect (1.) as to quantity if it is universal; (2.) as to
quality if it is distinct; (3.) as to relation if it is true; and finally (4.) as to
11
Kant ridicules the situation with the following analogy: “it is just as when someone makes a statement
before a court and in doing so appeals to a witness with whom no one is acquainted, but who wants to
establish his credibility by maintaining that the one who called him as witness is an honest man” (JL 9:50).
modality if it is certain” (JL 9:38).12 Since a perfect cognition has to satisfy

the above fourfold condition, it proves to be a more restrictive concept than
knowledge. While the conditions of truth and certainty qualify a perfect
cognition as knowledge, the conditions of universality and distinctness
narrow down the scope of perfect cognition to a specific subset of knowledge.
The four categories of perfection are about different aspects of a cognition,
but Kant does grant priority to truth over the other three properties, affirm-
ing that truth is “a principal perfection of cognition, indeed, the essential and
inseparable condition of all its perfection” (JL 9:49).
Even though falsehood would not disqualify a cognition as such, this does
not mean that any arbitrary judgment can be a cognition. A cognition can
turn out to be false, but in a sense it cannot be entirely false, according to
Kant:
A cognition is either wholly true, or either partly true or partly false. A

cognition can never be utterly false, however, although it can frequently be
wholly true. . . . Thus a cognition that is presented as thoroughly false will
nevertheless always be only in part false and will always contain a certain
degree of truth. (BL 24:93–94)
Every cognition has some truth in it because even if a cognition does not in
fact agree with the object it is about, it has, at least, to satisfy the formal
conditions that allow it to be about or related to an object. Only judgments
with objective validity are cognitions. Objective validity is a specifically
Kantian notion that can be ascribed to various types of mental representa-
tions, including intuitions, concepts, and judgments. Objectively valid judg-
ments may be false, but they must nevertheless have a real possibility of being
true, which is more restrictive than merely logical possibility. I will come
back to explain the notions of objective validity and real possibility later.
Cognition, perception, and intuition

In Kant’s terminology, cognition is often taken in a sense that is closer to
perception than knowledge. Kant’s cognition is correlated with objects or
things rather than with facts or propositions, as in the case of knowledge. As
Wilfrid Sellars notes, “For Kant there is a kind of awareness of objects in the
12
Each of these four categories of perfection can be further divided into aesthetic and logical ones, based
on the distinction between intuitive and discursive cognition (JL 9:36).
124 C.-F. Lau
broad sense, an awareness of them as being of a certain sort which is prior to

propositional knowledge, prior to judgments, prior to anything having the
explicit form of a judgment.”13 In fact, Kant defines cognition in its broader
sense as objective perception in the Stufenleiter passage, in which a hierarchical
definition of a series of key concepts of Kant’s epistemology is given:
The genus is representation in general (repraesentatio). Under it stands the

representation with consciousness (perceptio). A perception that refers to the
subject as a modification of its state is a sensation (sensatio); an objective
perception is a cognition (cognitio). The latter is either an intuition or a
concept (intuitus vel conceptus). The former is immediately related to the object
and is singular; the latter is mediate, by means of a mark, which can be
common to several things. (A320/B376–77)
The definition continues with distinctions between pure and empirical

concepts as well as between concepts of the understanding and reason
(ideas), without including the concept of knowledge. Cognition is objective
perception, which means “all representations related with consciousness to an
object” (JL 9:91). I call this broader, more inclusive sense of cognition the
representational sense.
Kant’s epistemology often gives the impression that cognition requires
both intuitions and concepts (A68/B93, A78–79/B104). As Kant’s
famous dictum goes, “Thoughts without content are empty, intuitions
without concepts are blind” (A51/B75). However, this double condition
applies only to judgmental cognitions, not necessarily to representational
ones, in which case either an intuition or a concept can count as a
cognition, each being a specific kind of objective perception. While
intuitions are singular representations that are related to objects imme-
diately, concepts are universal representations whose relation to objects
has to rely on mediation through intuitions. Intuitions and concepts
form two types of cognition, which Kant refers to as intuitive and
discursive cognition, respectively (JL 9:36). There is no question that
the paradigmatic kind of objective cognition discussed in the Critique is
13
Wilfrid Sellars, Kant and Pre-Kantian Themes: Lectures by Wilfrid Sellars, ed. Pedro Amaral
(Atascadero, Cal.: Ridgeview, 2002), 44. This is in line with the tradition of epistemology of Kant’s
time: “In the eighteenth century its core sense [of Erkenntnis] was intimately connected with the notion
of reference. To have Erkenntnis of a thing was to have in one’s mind a presentation, an idea, an image, a
token referring to that thing. . . . This makes understandable the close connection in eighteenth-century
philosophy between good reference and knowledge: To know is to have a good picture, the right
concept, the correct name, of a thing” (Rolf George, “Vorstellung and Erkenntnis in Kant,” in Interpreting
Kant, ed. Moltke S. Gram [Iowa City: University of Iowa Press, 1982], 35).
a conjunction of intuition and concept, which is discursive and judg-

mental, but Kant does also grant the possibility of intuitive cognition.
There are three different senses of intuitive cognition that can be more
specifically characterized as original, mathematical, and nonconceptual cogni-
tion, respectively. The first kind of intuitive cognition is something beyond
human reach and possible only for an infinite mind such as God with an
intellectual or original intuition. The original intuition is “one through
which the existence of the object of intuition is itself given” (B72). Finite
rational beings like humans are equipped only with a derived intuition,
whose cognitive capacity is dependent on the passive affection by objects.
The divine form of intuitive cognition is original in the sense that its objects
are originated from the very act of cognition. Being capable of cognizing
things as they are in themselves without recourse to concepts, the original
cognition of God is nonconceptual. If thinking consists in the use of con-
cepts, then God does not think at all, but rather grasps objects intuitively,
“for all of its cognition must be intuition and not thinking, which is always
proof of limitations” (B71).
In contrast to the original cognition of God, “the cognition of every, at
least human, understanding is a cognition through concepts, not intuitive
but discursive” (A68/B93). All judgmental cognitions have to rely on the
application of concepts, but there are still two types of human cognition
that can be called intuitive, namely, mathematical and nonconceptual
cognition. Traditionally, the difference between mathematics and philoso-
phy was considered to be a difference in the objects of the two disciplines,
with the former dealing with quantity and the latter with quality. Kant sees
a more fundamental difference in the nature of their cognitions, identifying
mathematical and philosophical cognition as intuitive and discursive,
respectively (JL 9:23). Mathematical cognition is said to be intuitive, not
because it does not involve concepts. It is intuitive because mathematical
concepts can be exhibited or constructed a priori in intuition and their
objective validity is guaranteed by the pure forms of intuition.
Mathematical cognition is, as Kant describes it, rational cognition “from
the construction of concepts,” which differs from philosophical cognition,
which is “rational cognition from concepts” (A713/B741). The legitimate
possibility of constructing concepts a priori in intuition provides mathe-
matics with a kind of immediate objective validity that qualifies it as
intuitive. In its foundation, mathematical cognition is intuitively certain
and even akin to the original cognition of God in that it can, in a sense,
“create” its objects of cognition, for in mathematics “one makes something
the object of intuition for himself a priori” (JL 9:23).
126 C.-F. Lau
When Kant speaks of intuitive cognition, he is usually referring to math-

ematical cognition, but there is a third kind of intuitive cognition, which
seems to be devoid of concepts, as presented in the following example:
If a savage sees a house from a distance, for example, with whose use he is not
acquainted, he admittedly has before him in his representation the very same
object as someone else who is acquainted with it determinately as a dwelling
established for men. But as to form, this cognition of one and the same object is
different in the two. With the one it is mere intuition, with the other it is
intuition and concept at the same time. (JL 9:33)
Kant counts a visual perception of an object as an intuitive cognition when

the perceiver does not have appropriate concepts at his disposal. The same
perceptual situation would allow another person with appropriate concepts
to have a discursive cognition of the house. Nonetheless, the savage should at
least possess the concept of a thing, which allows him to arrive at a discursive
cognition of the object as a something in front of him. As a finite rational
being, the savage should be capable of using categories such as substance and
causality to make sense of the world around him. If all the categories are
taken away, there is no cognition possible, not even a purely intuitive one
(A253/B309). In this case, the savage could still have sensation that is caused
by the house, but he would not have any cognition of an object, just as
animals “do not cognize” any object either (JL 9:65). Now, if the savage can
cognize an object, he must have the basic categories at his disposal. In the
case that he lacks the concept of house, he will have a “mere intuition” of the
house that is perceptually richer in content than the concept of a mere
something. Being richer than the concepts he possesses, his intuitive cogni-
tion has what is now called nonconceptual content, in which sense I call this
kind of cognition nonconceptual cognition. If intuitions without concepts
are blind, nonconceptual cognition can be called blind cognition.
It should be noted, however, that this third kind of intuitive cognition
counts as cognition only in the representational sense. Nonconceptual cogni-
tion is non-discursive and non-judgmental, which comes with the price that
nonconceptual cognition is not truth-apt: “For truth and illusion are not in
the object, insofar as it is intuited, but in the judgment about it insofar as it is
thought” (A293/B350). Mathematical and nonconceptual cognition are two
different kinds of intuitive cognition; yet, they differ not primarily in the way
in which they are intuitive, but more fundamentally in the way in which they
are cognition. Nonconceptual cognition is about a more primitive relation of
representation to object that first makes truth and falsehood possible. It is the
representational relation that constitutes the core of the concept of cognition.
Sensation, representation, and objective validity

For Kant, everything that is mental – from intuitions, concepts, and ideas to
even moral consciousness and aesthetic judgments – belongs to the genus of
representation. Kant considers representation to be such a general concept
that it is neither definable nor explainable (JL 9:33; VL 24:805).
Representation is not yet cognition; to qualify as a cognition in its broadest
sense, a representation still has to satisfy two additional conditions, i.e., being
conscious and objective. The two conditions give rise to four combinations: 1)
cognition as objective conscious representation, 2) sensation as subjective
conscious representation, 3) acquaintance (Kennen) as objective unconscious
representation, and 4) obscure representation as subjective unconscious
representation.14
Although it may appear contradictory to have representations without
being conscious of them (An 7:135), Kant accepts the existence of
unconscious representations, even admitting that “the field of obscure
representations is the largest in the human being” (An 7:136). Kant
rejects the Cartesian-Lockean presupposition that the mind is transparent
to itself. There can be representations that belong to the mind but are
nevertheless inaccessible to it. For Kant, consciousness has an intensive
magnitude that can vary and differ in infinitesimally small degrees
(A176/B218), and a totally unconscious representation is simply an
extreme on a continuous spectrum of consciousness. Even though we
may not be directly aware of obscure representations, we may be indir-
ectly conscious of having them by means of their effects (An 7:135). In
any case, obscure representations do not belong to the subject matter of
14
In the Jäsche Logic, Kant discusses seven degrees or grades of cognition, starting from the most
primitive one (JL 9:64–65): 1) representing, 2) perceiving in the sense of representing something
consciously, 3) being acquainted with something, 4) cognizing in the sense of being consciously
acquainted with something, 5) understanding in the sense of cognizing something conceptually, 6)
having insight into something in the sense of cognizing it through reason, and finally 7) comprehending
something in the sense of cognizing it “through reason or a priori to the degree that is sufficient for our
purpose” (JL 9:65). Since Kant defines cognition (as the fourth degree) as the state of being “acquainted
with something with consciousness” (JL 9:65), it implies that the state of being acquainted with some-
thing (as the third degree) is understood as a kind of objective representation without consciousness. In
this sense, objective unconscious representation can be called acquaintance, although it also belongs to
obscure representation.
128 C.-F. Lau
epistemology or metaphysics (JL 9:33), not even to pragmatic anthropol-

ogy; instead, “the theory of obscure representations belongs only to
physiological anthropology” (An 7:136).
Kant calls conscious representation perception, but perception can be
considered under two different aspects: Considered as a modification of the
state of a subject, it is sensation; only insofar as it is related to an object, is it
cognition. The difference between sensation and cognition can be under-
stood by means of what is known to contemporary philosophy of mind as the
difference between the phenomenal and the representational aspect of con-
scious experience. The phenomenal aspect refers to the conscious qualia or
the what-it-is-likeness of experience, but it is the representational aspect that
features in information processing and contains the intentional content of a
mental state. Kant defines sensation as “the effect of an object on the capacity
for representation, insofar as we are affected by it” (A19/B34). The effect is
private and internal to the cognitive subject and includes a variety of con-
scious qualia such as the sensations of colors, sounds, and warmth, which
form “the proper quality of empirical representations” (Pro 4:307). However,
Kant maintains that “the sensations of colors, sounds, and warmth, . . . since
they are merely sensations and not intuitions, do not in themselves allow any
object to be cognized, least of all a priori” (B44).
Sensation is something real in the mind and can play a part in empirical
cognition, but its cognitively relevant role has to be mediated by intuition
(A20/B34). Sensation is only the subjective, qualitative, material aspect of
conscious perception, a modification of the internal state of the mind that
does not belong properly to the objective world because it lacks a relation to
objects. Sensation is a representation without the function of pointing
beyond itself to be about something. If the term “representational” is under-
stood as carrying the property of aboutness or intentionality, then not every
representation is representational. Only object-directed representations are
representational and thus objective, whereas sensation is subjective in the
sense of having no representational purport. Cognition turns out to be the
most general concept that encompasses all objective or representational
representations.
The key question that comes with Kant’s concept of cognition is not how
a cognition can be true or justified, but rather how a cognition can be
appropriately related to an object or how a representation can be representa-
tional-of something. As Brandom puts it,
Kant’s most basic transcendental question does not, as his own characterization
of his project suggests, concern the condition of the possibility of synthetic
knowledge a priori, but the conditions of the intelligibility of representational

objectivity: of states or episodes that answer for their correctness to how it is
with the objects they represent.15
The traditional epistemological problems presuppose a certain kind of rela-

tion between our thought and the world that can be evaluated in terms of
truth and justification. But the very possibility of such a relation has yet to be
explained. Along with digging deeper into the semantic foundation of the
epistemological problems, Kant has to develop a new set of normative
concepts to address the new problem. The success or failure of representa-
tional purport is evaluated primarily in terms of objective validity, which
means literally validity for objects. For any cognition to have objective
validity, it must “be related to an object, and . . . have significance and
sense in that object” (A155/B194). The notion of objective validity is
applicable to all forms of objective representations – from intuitions, con-
cepts, and judgments to even cognitions in the practical respect – while being
inapplicable to subjective representations such as sensation. For different
sorts of cognition, the object-relatedness also plays out differently, corre-
sponding to their specific way of being related to objects, including truth as a
specific sort of objective validity.16
In the case of intuition, objective validity means a reference to an actual or
possible object that can be either given in inner or outer senses (in empirical
cognition) or constructed a priori (in mathematical cognition). For a concept
to have objective validity, it has either to be applicable to some actual or
possible objects or to belong to the necessary conditions of the possibility of
cognition. As the object-relatedness for all sorts of cognition is established in
the end through intuition (A19/B33), the objective validity of concepts has
to be mediated by it. Since empirical concepts are formed a posteriori through
empirical intuition, their formation already guarantees a connection to actual
or possible objects. What is more problematic is the objective validity of pure
concepts, which are not abstracted from experience but arise from the
understanding (JL 9:92). Kant’s worry is that without being related to objects
through intuition, such concepts would be “entirely empty of content” or
“rather a mere play” (A239/B298). A central task of the Critique consists in
15
Brandom, “Kantian Lessons,” 51.
16
I analyze the concept of objective validity in greater detail and distinguish between two orders of it in
Chong-Fuk Lau, “Transcendental Concepts, Transcendental Truths and Objective Validity,” Kantian
Review 20, no. 3 (Nov. 2015): 445–66. See also Hanna, Kant and the Foundations of Analytic Philosophy,
83–95.
130 C.-F. Lau
proving that the twelve categories are universally applicable to empirical

objects, although they are pure concepts originating from the understanding.
The Transcendental Deduction attempts to show that these pure concepts
belong to the necessary conditions that first make empirical cognition
possible. They acquire objective validity indirectly through the temporal
schematization that makes their content homogenous with what can be
given in intuition (A137–38/B176–78).
Since there is no other use of concepts than in judgment (A68/B93), the
objective validity of concepts can be defined in terms of properties that can
be ascribed to judgments. If a judgment consists of concepts that have no
objective validity, it means that the concepts in the judgment are not
applicable to any actual or possible objects and therefore that the judgment
cannot be possibly true, unless it is an analytic judgment which expresses
merely logical or conceptual relations. The judgment may be still mean-
ingful, if the concepts involved are internally consistent and have intensional
meaning. However, since such concepts are not applicable to actual or
possible objects, they are empty extensionally. It is to be noted that concepts
without objective validity can still count as cognition in the representational
sense, as they do have their representational purport, even though it fails to
reach any possible object. Although they do not form any cognition in the
judgmental sense, they allow to us to think by virtue of their representational
content, as long as the concepts are internally consistent: “To think of an
object and to cognize an object are thus,” as Kant emphasizes, “not the
same” (B146). Both kinds of cognitive activities make use of concepts, but
only the latter kind is subject to the conditions of intuition and capable of
guaranteeing a proper relation to objects.
If cognition in the judgmental sense is a judgment in which all the
concepts involved have objective validity, then it must be truth-apt, even
though it may turn out to be false. Truth-aptness in this context is not
equivalent to logical consistency or freedom from contradictions; instead, the
scope of truth-aptness is defined by what Kant calls the real possibility of
things, in contrast to the logical possibility of concepts (Bxxvi, B308). Kant’s
theory of categories defines its own modal categories, which determine what
kinds of things have real possibility, actuality (existence), and necessity. In
the Postulates of Empirical Thought, the modal categories are analyzed in
reference to the conditions of experience, among which real possibility is
defined as such: “Whatever agrees with the formal conditions of experience
(in accordance with intuition and concepts) is possible” (A218/B265). Real
possibility is a much more restrictive concept than its logical counterpart. A
(real-)possible object has to satisfy the formal conditions of experience that
are determined by the spatiotemporal forms of sensibility and the categories

of the understanding. There is, so to speak, a basic skeleton for every possible
object, and every single one has to be fitted into the spatiotemporally and
causally organized unity of experience. This scope of possible objects sets the
boundary for all judgmental cognitions.
Objectivity, normativity, and cognizability

According to the concept of real possibility, what can be counted as a
possible object of cognition is something to which spatiotemporal properties
and the categories of substance and causality can be properly ascribed, instead
of merely mental representations in our consciousness. Mental representa-
tions can also be regarded as objects of cognition, but only in a thin and non-
substantial sense. Kant clarifies the difference as follows:
Now one can, to be sure, call everything, and even every representation, insofar
as one is conscious of it, an object; only what this word is to mean in the case of
appearances, not insofar as they are (as representations) objects, but rather only
insofar as they designate an object, requires a deeper investigation. Insofar as
they are, merely as representations, at the same time objects of consciousness,
they do not differ from their apprehension. (A189–90/B234–35)
The thinner, logical sense of object can be ascribed to any conscious repre-
sentation, including any sensations and thoughts, but this kind of object
exists only privately in the subject, without belonging as a constituent to the
objective world. Since our internal perception is in continual flux, no
persistent object can be found in the stream of inner consciousness (A107).
An object in a thicker, substantial sense, however, requires an identifiable
unity to which properties can be ascribed. In Kant’s words, “An object . . . is
that in the concept of which the manifold of a given intuition is united”
(B137). It is this proper sense of object that “requires a deeper investigation.”
The difference can be illustrated with the examples of the perceptions of a
house and of a ship driven downstream, discussed in the Second Analogy
(A190–92/B235–37). In both cases, every conscious sensation of color or
shape is an object in the thin sense, which can occupy the role of a subject in
a judgment and be described using appropriate concepts. However, the
sensations succeed each other and vanish quickly. There are no objective
rules as to the order in which different parts of the house or the ship have to
be perceived. Yet, if we are to cognize the house as a standing object in
132 C.-F. Lau
contrast to the ship as a moving one, we have to apply the categories of

substance and causality appropriately and organize various conscious repre-
sentations that could be perceived in different orders in accordance with
specific rules. More precisely, the representations of the house have to be
perceived as belonging to an object simultaneously, although they may
succeed one another in the mind. By contrast, those representations of the
ship driven downstream have to be perceived as a series of events that happen
to a moving object. In the latter case, the subjective succession of perceptions
in my mind is cognized as an objective succession of events outside me.
It turns out that what makes a representation capable of being
representational-of-something is the fact that it is subject to necessary
rules, in accordance with which a representation is combined with
others into a unified object. As Kant puts it, “appearance, in contra-
distinction to the representations of apprehension, can thereby only be
represented as the object that is distinct from them if it stands under a rule
that distinguishes it from every other apprehension, and makes one way
of combining the manifold necessary” (A191/B236, emphasis added).
The condition of rule-governedness constitutes the core of Kant’s
definition of object in its substantial sense: “That in the appearance
which contains the condition of this necessary rule of apprehension is
the object” (A191/B236). If a representation can somehow point
beyond itself to be about something, this relation is only possible
under a set of necessary rules that govern the relation between the
representation and the object represented. This is similar to the relation
between cause and effect in causality. Causality would be inconceivable
if events were not governed by laws. Only by virtue of a causal law is it
possible to pair up specific events as cause and effect, thereby distin-
guishing them from other events that occur at the same time. More
precisely, events that succeed each other can be related as cause and
effect only if there is a strict law which those successive events instanti-
ate. Donald Davidson calls it the Principle of the Nomological
Character of Causality.17 This principle can be traced back to Kant,
who defines the schema of causality as “the succession of the manifold
insofar as it is subject to a rule” (A144/B183).
Now just as the relation between cause and effect is necessarily mediated
by a causal law, a cognition can be appropriately brought into relation with
17
Donald Davidson, “Mental Events,” in Essays on Actions and Events, 2nd ed. (Oxford: Clarendon,
2001), 208.
an object only by means of a necessary rule, by virtue of which the cognition

counts as a representation of what is represented. For Kant, cognitions
“consist in the determinate relation of given representations to an object”
(B137), and it is the rule that makes the object-relation of a cognition
determinate by mapping a cognition to a specific set of objects. However,
it should be noted that a rule that governs representational relations differs
from a causal law in an important respect: while causality describes factual
relations between events, the rule that governs the relation of a representation
to its object is of a normative nature. We can certainly misidentify causal
relations, but causal laws just necessitate what happens and causal determina-
tion itself makes no mistake. In the case of representational relations, the
rules are norms that can possibly be violated. There are correct and incorrect
representations, objectively valid and invalid concepts, true and false judg-
ments, as well as perfect and imperfect cognitions, depending on whether the
rules that govern the respective kind of representations are sufficiently
followed. Nevertheless, although it is certainly possible to have misrepresen-
tations in individual cases – for example, to misapply the category of sub-
stance to cases in which the category of causality would be appropriate – it
does not mean that the rules are not objective. If it were not the case that
most concepts have been, by and large, correctly applied, there would be no
basis at all to claim that we have wrongly applied a concept. The possibility
of violating a rule presupposes that it has been consolidated as a rule by
sufficient instances of correct application.18
The normative approach offers a plausible way to understand what is
revolutionary in Kant’s Copernican revolution. Instead of assuming that
our cognition has to conform to objects, Kant takes it as his revolutionary
idea to argue that “the objects must conform to our cognition” (Bxvi). Kant
maintains that the Copernican revolution is only possible on the basis of his
transcendental idealism, which takes objects of cognition not as things in
themselves but as appearances. Since Kant sometimes equates appearances
with “mere representations” (A490–91/B518–19), he is often accused of a
form of Berkeleyan idealism or phenomenalism.19 According to this inter-
pretation, objects have to conform to our cognition because they are nothing
18
The idea is similar to what Davidson calls the principle of charity, which implies the impossibility of
universal mistake: “Global confusion, like universal mistake, is unthinkable, not because imagination
boggles, but because too much confusion leaves nothing to be confused about and massive error erodes
the background of true belief against which alone failure can be construed” (Davidson, “Mental Events,”
221).
19
See Peter Strawson, The Bounds of Sense: An Essay on Kant’s “Critique of Pure Reason” (London:
Routledge, 1975), 21–22.
134 C.-F. Lau
but mental representations that are organized or constructed according to the

forms of our cognitive faculty. However, the normative approach can make
sense of Kant’s revolution even if objects are recognized as material things
existing outside the mind. If Kant is successful in showing that the categories
of understanding are the necessary rules that determine the basic structure of
what can be counted as an object, the forms can be a priori valid for all
possible objects, even if they do not exist in our mind. Everything that does
not conform to the conditions just would not be identifiable as an object of
cognition for us.
Admittedly, every kind of rule comes with specific implications and con-
straints. If all possible objects have to satisfy the rules developed in the Critique,
then things that do not fall within the spatiotemporally and causally structured
world of appearance can never be cognized. If we agree, say, with Paul Guyer
that “transcendental idealism is nothing other than the thesis that things in
themselves, whatever else they may be, are not spatial and temporal,”20 then
things in themselves or noumena would never be counted among possible
objects of cognition.21 In fact, they cannot even be said to exist because they do
not meet the formal conditions that define real possibility, let alone the more
restrictive material conditions that define existence or actuality (A217/B266).
In this sense, things in themselves or noumena cannot and do not exist.22 They
are what Kant calls an empty concept without object, which is a specific type of
nothingness: “the object of a concept to which no intuition that can be given
corresponds is = nothing, i.e., a concept without an object, like the noumena,
which cannot be counted among the possibilities although they must not on
that ground be asserted to be impossible (ens rationis)” (A290/B347). A
noumenon turns out to be a thought-entity, which can be thought of, but
never cognized, “because it is a mere invention” (A292/B348).
As we know, Kant repeatedly claims that noumena or things in themselves
can never be cognized. If the above analysis of Kant’s concept of cognition is
right, then the uncognizability thesis means that there are no such things as
20
Paul Guyer, Kant and the Claims of Knowledge (Cambridge: Cambridge University Press, 1987), 333.
See also Chapter 4 of this volume.
21
Things in themselves and noumena will be taken as roughly referring to the same thing in this
chapter, although there are subtle differences between the two concepts. See Henry E. Allison, “Things
in Themselves, Noumena, and the Transcendental Object,” Dialectica 32, no. 1 (March 1978): 41–76.
22
I explain the nonexistence of things in themselves in greater detail in Chong-Fuk Lau, “Kant’s
Epistemological Reorientation of Ontology,” Kant Yearbook 2 (2010): 123–33. J. G. Fichte picks up
on this idea in Kant and develops a version of transcendental idealism in which he explicitly denies the
existence of a thing in itself. For a defense of this claim in Fichte (as true to the spirit of Kant), see
Matthew C. Altman, “Fichte’s Transcendental Idealism: An Interpretation and Defense,” in The
Palgrave Handbook of German Idealism, ed. Matthew C. Altman (London: Palgrave, 2014), 323–25.
things in themselves that can be objects of cognition. This, however, should

not be confused with the claim that there can never be any knowledge about
them, since the uncognizability thesis does not straightforwardly imply that
there can be no true or justified beliefs about things in themselves.23 For
example, the claim that things in themselves are not spatiotemporal can be a
true and justified judgment, without violating Kant’s uncognizability thesis.
In fact, the uncognizability of things in themselves is explained by the fact
that they are not spatiotemporal and thus do not fall within the cognizable
realm of possible objects. A true judgment in which the term “things in
themselves” serves as a grammatical subject does not imply that it is a
judgment with objective validity, expressing a judgmental cognition of things
in themselves. The claim that things in themselves are not spatiotemporal can
be true analytically as a logical implication derived from Kant’s theory of
cognition if the theory itself is largely true. The truth of the claim is
consistent with the fact that the concept of things in themselves is a mere
invention without objective validity and remains unrelated to any possible
object. It is possible to think of an object that has no real possibility and to
arrive at true judgments about the concepts that are used to describe it.
However, it is just a thought or a judgment, possibly a true one, but never
a cognition. If the true judgment is justified, it can be counted as a piece of
knowledge about the concept, say, of things in themselves without constitut-
ing a cognition of them.
Cognition, faith, and metaphysics

Determining the boundaries and conditions of cognizability is a major task
of Kant’s Critique, but there can be different types of cognition with their
respective boundaries and conditions. Although Kant rules out the possibility
of theoretical cognition beyond possible experience, he leaves room for
another kind of cognition that may reach out further to the intelligible or
supersensible world. As Kant famously proclaims, “I had to deny knowledge
in order to make room for faith” (Bxxx). Faith is a kind of belief that is based
on subjectively, but not objectively, sufficient grounds. A main concern in
Kant’s distinction between opinion, faith, and knowledge in the Doctrine of
23
See Karl Schafer, “Kant’s Conception of Cognition and Our Knowledge of Things in Themselves,” in
The Sensible and Intelligible Worlds, ed. Karl Schafer and Nicholas Stang (Oxford: Oxford University
Press, forthcoming).
136 C.-F. Lau
Method is to establish a legitimate realm of cognition that is not confined to

the theoretical use of reason. There can be beliefs that are justified non-
epistemically. Kant maintains, for example, that beliefs in the freedom of the
will, the immorality of the soul, and the existence of God can be justified in
the practical respect as a condition of morality or the idea of the highest good
(A825–26/B853–54).
The distinction between knowledge and cognition has another impor-
tant consequence here. Even though there can never be any knowledge or
theoretical cognition affirming, say, the immorality of the soul, it can
nevertheless be an object of practical cognition. Practical cognition has a
different kind of relation to a different kind of object, as compared to the
theoretical one. While theoretical cognition is related to its object by
“merely determining the object and its concept,” practical cognition is
object-related by “making the object actual” (Bx). The difference in the
object-relation can be traced back to a more fundamental difference in
the nature of the object (CPrR 5:57). In a narrower sense, practical
cognition is cognition of what ought to exist, i.e., imperatives, in contrast
to theoretical cognition as cognition of what does exist (A633/B661). In a
broader sense, the object of practical cognition includes not only impera-
tives, but also grounds for possible imperatives such as the belief in the
immorality of the soul (JL 9:86).
No matter whether it is directly about imperatives or indirectly about
grounds for imperatives, the object of practical cognition involves something
that can never be met in the whole of nature:
The ought expresses a species of necessity and a connection with grounds

which does not occur anywhere else in the whole of nature. In nature the
understanding can cognize only what exists, or has been, or will be. It is
impossible that something in it ought to be other than what, in all these time-
relations, it in fact is; indeed, the ought, if one has merely the course of nature
before one’s eyes, has no significance whatever. (A547/B575)
Since imperatives or judgments about oughtness do not describe facts, there

is no objective ground that could possibly qualify them as knowledge.
However, this does not imply that there is no way to distinguish normative
judgments that are based on sufficient reason from those that are not. A
major task for Kant’s practical philosophy is to develop and justify another
kind of objective validity that can be ascribed to practical cognition with
respect to its relation to objects.
Yet, Kant makes a more controversial move with the idea that the rules
that confer objective validity on rationally based imperatives can confer
objective validity on a factual judgment that constitutes a necessary ground
for the imperatives. Even if the ground is about something in the super-
sensible world, Kant considers it legitimate to ascribe to it a kind of objective
validity in the practical respect. In this way, Kant allows practical cognition
of a metaphysical object, about which there can never be any theoretical
cognition or knowledge24:
From a moral point of view, . . . we do have sufficient reason to assume a life for
man after death (the end of his earthly life) and even for eternity, and hence the
immortality of the soul; and this doctrine is a practico-dogmatic transition to
the super-sensible, i.e., to that which is a mere Idea, and can be no object of
experience, yet possesses objective reality, albeit valid only in a practical sense.
(RP 20:309)
Here I do not mean to defend Kant’s commitment to this kind of meta-

physical belief, nor am I in a position to give a detailed account of Kant’s
theory of practical cognition. The brief discussion in this final section aims
to indicate how Kant’s concept of cognition leaves room for a kind of
practical relation of representations to objects, thus providing a foundation
for a unified theory of theoretical and practical reason. Kant’s concept of
cognition offers the key to the whole secret of metaphysics which is not just
a theoretical enterprise, but also a practical one, forming a system of
rational cognition consisting of the metaphysics of nature and the meta-
physics of morals.
24
See Karl Schafer, “Practical Cognition and Knowledge of Things-in-Themselves,” in The Idea of
Freedom: New Essays on the Interpretation and Significance of Kant’s Theory of Freedom, ed. Dai Heide and
Evan Tiffany (Oxford: Oxford University Press, forthcoming).
7
Apperception, Self-Consciousness,
and Self-Knowledge in Kant
Dennis Schulting
Introduction
Kant’s theory of transcendental apperception, which concerns the rules for
unitary self-consciousness, is a central part of the critical project for establish-
ing the necessary (and formally sufficient) conditions of cognition or knowl-
edge.1 The principle of transcendental apperception, in Kant’s words, is “the
supreme principle of all use of the understanding,” and, since the under-
standing is the “faculty of knowledge,” transcendental apperception is the
principle of all knowledge (B136–37, translation modified). This is a rela-
tively uncontroversial aspect of Kant’s theory of apperception, although it
should be said that there is hardly even general agreement about the details of
its central tenets.2 But there does seem to be a general agreement about the
The original version of this chapter was revised. An erratum to this chapter can be found at
https://doi.org/10.1057/978-1-137-54656-2_36
1
I understand Kant’s Erkenntnis to be more than what is understood by the English “cognition” and less
than the English term “knowledge” (for which the German equivalent is Kenntnis or Wissen). For all its
potentially misleading connotations to contemporary readers, I shall in general be using the term
“knowledge” for Kant’s Erkenntnis, and make no specific distinction with the term “cognition.” For
further discussion on this issue, see Dennis Schulting, Kant’s Radical Subjectivism: Perspectives on the
Transcendental Deduction (London: Palgrave Macmillan, 2017), ch. 2.
2
For an account, see Dennis Schulting, Kant’s Deduction and Apperception: Explaining the Categories
(Basingstoke: Palgrave Macmillan, 2012); and Schulting, Kant’s Radical Subjectivism.
D. Schulting (*)
Independent Scholar, Carbonera, Italy
e-mail: ds196901@gmail.com

140 D. Schulting
view that Kant’s theory of transcendental apperception is first and foremost

constitutive of objective knowledge, and not so much a theory about the
possibility of self-knowledge or a sui generis theory of self-consciousness. It is
generally held that the Kantian principle of unitary self-consciousness may be
about the rules for uniting one’s representations, which enables the knowl-
edge of objects, but is not about self-consciousness per se.3 That is to say,
transcendental apperception is more about what enables us to think about
something other than thought than it is about how we are aware of ourselves,
whether as thinkers or just as having some kind of self-awareness (self-feeling,
proprioception). In the past, Dieter Henrich has spoken of the “grounding
function [Begründungsfunktion],” for Kant, of self-consciousness for some-
thing else, namely objective knowledge.4
The view that Kant’s theory of apperception is less concerned with self-
consciousness per se than it is with self-consciousness as the ground of
objective knowledge is often motivated by Kant’s remark, in the introductory
section of the Paralogisms chapter of the Critique of Pure Reason, that we can
think of the self only in terms of the predicates that we think, never an sich, in
abstraction from these predicates of thinking. We can never really think the
self itself, as it is apart from any necessary constitutive conditions, under which
alone, like any other entity or object, it can be thought. We only think our
thoughts, and do not have thereby access to some putative underlying identical
self, to whom these thoughts belong. That is to say, if I were to try to
determine the identical underlying self, whose thoughts are thought, I would
move in a circle, “since we must always already avail ourselves of the repre-
sentation of it [i.e., the ‘I’] at all times in order to judge anything about it.”
Kant says that “we cannot separate ourselves from this inconvenience” (A346/
B404). Especially, this latter remark would appear to mean that the transcen-
dental perspective of the “I think,” by means of which we think our thoughts,
prevents us from having an immediate access to ourselves qua thinking selves,
and so by implication prohibits a sui generis theory of self-consciousness. The
only way to gain cognition of one’s self is by treating it in the same way as
objects are cognized, which might create the impression that on Kant’s
account self-consciousness strictly speaking is only a derivative kind of self-
consciousness. So it would seem that Kant’s treatment of the “I think” – which
3
Cf. Klaus Düsing, “Gibt es einen Zirkel des Selbstbewußtseins? Ein Aufriß von paradigmatischen
Positionen und Selbstbewußtseinsmodellen von Kant bis Heidegger,” in Subjektivität und Freiheit:
Untersuchungen zum Idealismus von Kant bis Hegel (Stuttgart-Bad Canstatt: Frommann-Holzboog,
2002), 111–40.
4
See Dieter Henrich, Fichtes ursprüngliche Einsicht (Frankfurt am Main: Klostermann, 1967), 10.
7 Apperception, Self-Consciousness, and Self-Knowledge 141
is what is being discussed in this particular section in the Paralogisms, but is in

fact the principle of apperception (B132; see further below the account in the
section “Transcendental Apperception and Self-Consciousness”) – does not
provide us a theory of self-consciousness an sich. And indeed, as Kant says, the
“I” of the “I think” is but a “transcendental subject of thoughts = x.” The
“consciousness in itself” does not distinguish “a particular object,” but, “inso-
far as it is to be called a cognition,” is rather “a form of representation in
general” through which “anything” can be thought (A346/B404). The “I
think,” as the “vehicle” of my thoughts, “of all concepts whatever” (A341/
B399), thus seems a neutral placeholder for thoughts that are being thought,
by whoever thinks them. That is, “the proposition I think (taken problema-
tically) contains the form of every judgment of understanding whatever and
accompanies all categories as their vehicle.” Moreover, the “proposition ‘I
think’ is,” as Kant says, “taken here only problematically; not insofar as it may
contain a perception of an existence (the Cartesian cogito, ergo sum)” (A347–
48/B405–6). It is clear, from these passages at least, that on Kant’s account the
function of the “I” is squarely aimed at thought or thinking as such, in
particular the understanding as the form of any possible knowledge, not at
self-consciousness (or at least self-consciousness in a narrower sense). The view
expressed in the Paralogisms is not arbitrary, but is grounded in the discursive
logic that underpins Kant’s theory of knowledge, which is shown in the logic
of transcendental apperception, which we shall discuss further below.
Relatedly, Kant also makes a crucial distinction between inner sense and
apperception, unlike the rationalists, for whom inner sense is self-conscious-
ness, and implies immediate cognition of the self an sich through empirical
experience (i.e., inner sense, empirical self-consciousness).
In this chapter, I wish to concentrate on two connected elements of
Kant’s theory of self-consciousness: the transcendental conditions for
establishing the identity of self-consciousness, which first enable the
awareness thereof, namely self-consciousness strictly speaking, and the
relation between self-consciousness and self-knowledge. I contend that
two mistaken assumptions underlie the critique of Kant’s “derivative” or
so-called “reflection-theoretical” view of self-consciousness, namely the
belief that it does not accommodate a sui generis theory of self-conscious-
ness: (1) that the identity of self is somehow a priori given, and presum-
ably any act of transcendental apperception, which is interpreted as an act
of reflection, always already presupposes this a priori self-identity, and (2)
that the awareness of the identity of self-consciousness ipso facto amounts
to self-knowledge. Concerning assumption (1), often it is thought that
Kant’s so-called reflective “I think,” which accompanies my
142 D. Schulting
representations, is secondary to, or derivative of, the transcendental unity of

self-consciousness, or indeed, secondary to the identity of self-consciousness. In
the section “Transcendental Apperception and Self-Consciousness,” I address
this assumption from an interpretive point of view, by looking more closely at
Kant’s argument in §16 of the B-Deduction (B131–36). This will show that
Kant’s view of self-consciousness is in fact not derivative, and that instead any
account of self-consciousness and the identity of self is first made possible by
transcendental consciousness or transcendental apperception, which is nothing
but the act itself of accompanying, through the “I think,” one’s representations
as one’s own. Transcendental consciousness is an original consciousness, which
a priori grounds any form of self-consciousness or self-knowledge, and is “the
consciousness of myself, as original apperception” (A117n). In the last section
of the chapter (the section “Self-Knowledge”), I consider (ad 2) why, for Kant,
awareness of the identity of self-consciousness does not ipso facto amount to
self-knowledge, and explain that, in addition to transcendental self-conscious-
ness, what Kant calls the “affection” of inner sense is needed for self-knowledge
to be possible. But first, in the next section, I address some more general,
systematic issues, which directly bear on the aforementioned topics.
Self-consciousness and identity

Apart from interpretive grounds (which are addressed in the sections
“Transcendental Apperception and Self-Consciousness” and “Self-Knowledge”),
there are also philosophical reasons for why Kant’s account of self-consciousness
does not account for self-consciousness in the sense of strict self-knowledge, which
explains the misguidedness of the criticism that Kant’s theory of apperception
lacks the means to account for a sui generis self-consciousness. The general
criticism is that if the identity of self is first established in the reflection on oneself
(a turning back into oneself), then the self-identity and the knowledge thereof
is not immediate, but secondary to the reflection. But at the same time, it is
argued, the reflection presupposes the identity of the self in order to be able to carry
out the reflection, for the reflection is of course done by the same person or
self whose identity is reflected upon. This model ipso facto cannot attain deter-
mination of self-identity per se, because it fundamentally misconstrues the nature
of self-consciousness or the “I” as a function of thought or cognition. The
difference between the “I” as a function of thought and the “I” as referring to
self-consciousness strictly speaking can be indicated by seeing the former in
terms of a proposition of the form (a) “I know that I ϕ” or “(I know:) I ϕ”
(what Ernst Tugendhat calls “epistemic self-consciousness”5), where ϕ stands

for a predicate describing a generic conscious state, and the latter in terms of a
self-relation that is expressed by the Fichtean identity proposition (b) “Iy=Ix,”6
where y stands for the agent or subject of the act of self-consciousness and x for
the object of the act of self-consciousness. In (b) there is an identity between the
positions on both sides of the copula, between the knowing “I” (“Iy”) and the
known “I” (“Ix”); (b) expresses self-consciousness, namely, an intimate and
immediate, non-propositional knowledge of one’s self-relation or one’s being
related to oneself.7 In (a) there is prima facie no expression of a self-relation, nor
of an identity between “I” and the proposition “I ϕ,” that is, an identity
between knowing (the “I”) and known (putatively the proposition “I ϕ”); the
“I ϕ” expresses a cognitive act performed by an “I,” rather than an object of
knowledge for an “I.” Kant’s notion of self-consciousness, according to his
theory of apperception (which I will delineate in the section “Transcendental
Apperception and Self-Consciousness”), would be describable as (a) but not as
(b). It is significant that Tugendhat calls (a) a form of “epistemic self-conscious-
ness,” as this is precisely the reason why Kant’s view of self-consciousness is
criticized for lacking the means to describe a sui generis self-consciousness.
Though the opposite is often claimed,8 it should be clear that Kant’s afore-
mentioned circle is not the circularity of presupposing what one is supposed to be
reflectively aware of, namely the identity of the self that is both known entity and
knower of this entity (shown by the identity proposition “Iy=Ix”).9 This inter-
pretation of the circularity to which Kant alludes in the Paralogisms’ introductory
section sees self-consciousness in terms of the representationalist subject-object
model “s represents o,” where, in the case of self-representation, s is supposed to be
identical to o in conformity with the Fichtean proposition “Iy=Ix.” But in Kant’s
case the problem does not concern the putative identity between the subject as the
thinker (“Iy” or henceforth “Is”) and the thinker as her own object (“Ix” or
henceforth “Io”), but rather the identity between my thoughts as accidents and
their substratum, putatively the substance in which they inhere (A349–50). The
circularity, for Kant, concerns the unavoidability of having to employ the “Is” as
the “vehicle” of my thoughts, in whatever attempt to determine the substance
5
Ernst Tugendhat, Selbstbewußtsein und Selbstbestimmung: Sprachanalytische Interpretationen, 6th ed.
(Frankfurt am Main: Suhrkamp, 1997), 50.
6
See Johann Gottlieb Fichte, Grundlage der gesamten Wissenschaftslehre (1794), eds. Fritz Medicus and
Wilhelm G. Jacobs (Hamburg: Meiner, 1997), 14.
7
See Henrich, Fichtes ursprüngliche Einsicht, 13.
8
Paradigmatically, Henrich, Fichtes ursprüngliche Einsicht.
9
See Düsing, “Gibt es einen Zirkel des Selbstbewußtseins?” 117–18.
144 D. Schulting
underlying it (putatively “Io”). That is to say, to determine the substratum to

which my thoughts collectively belong, I need to employ the “vehicle” with
which I think these thoughts in the first place, which points to a circularity that is
non-vicious, however unavoidable. This may seem to be the same circularity as
the one that appeared so problematic to Fichte (and his contemporary followers),
namely, that the subject doing the thinking is also the object of my thought whose
identity I am trying to establish, which creates the vicious circularity of trying to
determine the identity of that which one has already made an appeal to in the
very attempt to determine it.10
But for Kant the identity of the logical self of discursive thinking (the “Is”) is
unproblematic, and does not presuppose knowledge of the identity of “Io”: I
know that my thoughts are mine to the extent and to the extent only that I
accompany them, as Kant says in the well-known proposition at the start of his
argument for the conditions for self-consciousness in §16 of the B-Deduction;
in fact, it is an “analytical proposition,” namely defining the rule that “it is only
because I can combine a manifold of given representations in one conscious-
ness that it is possible for me to represent the identity of the consciousness in
these representations itself, i.e., the analytical unity of apperception is only
possible under the presupposition of some synthetic one,” which thus first
“ground[s] . . . the identity of apperception itself” (B133–35). (I will return to
this passage in the next section.) So, in one sense at least, I know about the
identity between myself as thinker of my thoughts and the thoughts I thereby
accompany or think, just because the identity of my self is established in the
very act of accompanying my representations, which, according to Kant,
requires an a priori act of synthesis (more on this further below). I am aware
of the identity of the synthesis that first establishes the identity of the repre-
sentations I accompany, just because I am the one doing the accompanying
(that is, performing the synthesis). Kant speaks of an identity in terms of an
analytical unity of consciousness or apperception that obtains between all of the set
of representations that are accompanied by an “I think,” for any such occurrent
accompanying of representations (B133–34).
The circularity of which Kant speaks in the Paralogisms does not concern
the alleged problem with establishing, along the s=o model, the identity
between thinker and what is thought, whose identity is, as I argued, unpro-
blematic for Kant, for any set of representations r accompanied by any
instantiated “I think” (“Is”) is identical to “Is” to the extent and to the extent
10
Kant describes this as follows: “I cannot cognize as an object itself that which I must presuppose in
order to cognize an object at all” (A402).
only that r is accompanied by “Is”; there is thus an identity between the “I” as
subject (“Is”) and its predicates as its “object” (cf. A197/B242: any representa-
tion can be called the “object” of another representation). Rather, the
circularity problem, for Kant, concerns the mistaken attempt to establish
the identity between the thinker “Is” and any substance (“Io”) that may
underlie the subject of thought (“Is”). So what is at issue here is establishing
the identity between the subject and its putative substratum in terms of an
unconditional noumenal substance, and the possibility of this Kant categori-
cally rejects. But knowing the identity of an underlying substance, hence
knowledge about the identity between “Is” and “Io,” is not presupposed in
whatever self-conscious act of accompanying my representations which
determines the identity of those representations as mine. To suppose it
were, would in Kant’s view precisely reverse the order of what grounds and
what is grounded; hence his circularity objection: putatively being able to
determine the subject as grounding substance presupposes the transcendental
act of apperception, and not the other way around, as Henrich, following
Fichte, believes it does.
I should point out again, though, that the identity between the thinking “Is”
and its predicates – the identity at issue in Kant’s argument about apperception –
is not one strictly between two propositions “I know” and “I ϕ” which relate to
each other as main and subordinate clauses in the proposition “I know that I ϕ,”
but the identity is one between the “I” in the main clause and the “I” in the
subordinate clause, based on the idea that, when I ϕ, I know that I am the one ϕ-
ing. The knowledge – if that is the proper term – that I have of the identity
concerned here is non-propositional, because it is prior to, and presupposed by,
knowing the propositional content of “I ϕ.” This kind of non-propositional
knowledge of one’s identity comes with ϕ-ing, for any act of ϕ-ing which I am
aware of as doing – this is perhaps expressed better in the formulation “(I know:)
I ϕ.” It is the consciousness of oneself in terms of the identity of an action or
function that unites a manifold of representations. This is illustrated by Kant’s
claim in A108, also quoted by Henrich in this context,11 where Kant writes:
Thus the original and necessary consciousness of the identity of oneself is at the
same time a consciousness of an equally necessary unity of the synthesis of all
appearances in accordance with concepts, i.e., in accordance with the rules that
not only make them necessarily reproducible, but also thereby determine an
object for their intuition, i.e., the concept of something in which they are
11
Dieter Henrich, Identität und Objektivität: Eine Untersuchung über Kants transzendentale Deduktion
(Heidelberg: Winter, 1976), 64 and throughout.
146 D. Schulting
necessarily connected; for the mind could not possibly think of the identity of itself
in the manifoldness of its representations, and indeed think this a priori, if it did
not have before its eyes the identity of its action, which subjects all synthesis of
apprehension (which is empirical) to a transcendental unity, and first makes
possible their connection in accordance with a priori rules. (A108, emphasis
added)
Kant may seem to say here that the apperceptive mind has an a priori awareness
of the identity of the self, suggesting that the awareness of its identity is somehow
prior to the action of synthesis, which is central to the argument in the context of
this passage. But what Kant in fact says is that the identity of the mind is
simultaneous with “its action” of subjecting the synthesis of apprehension to a
transcendental unity. The awareness of the self’s identity is dependent on an
“action” of synthesis or unification of representations (I return to this argument
in the section “Transcendental Apperception and Self-Consciousness”), and is
not a priori given and as such immediately intuitable.
By contrast, the Fichtean position (“Iy=Ix”) assumes an intuitive evidence
of the identity of the “I” in any act of self-awareness – hence, Fichte employs
the term Selbstanschauung to indicate this, whereas Kant specifically and
consistently says that our mode of cognizing is not one based on self-
intuition in terms of an intellectual intuition, thus presupposing a priori
knowledge of an identical “I” across various instances of consciousness,
that is, an identical substance underlying self-consciousness, to which I
have an intuitive and thus immediate access (see, for example, B153,
B157–59; compare B135, B138–39, B148–49). Kant acknowledges a kind
of immediate intuition underlying the intellectual “I,” in apperception, but
this is only an undefined or indeterminate empirical intuition, a “feeling”
(Pro 4:334n), of the existence of myself as the one thinking at any one
particular instance of the “I think,” not across a manifold of such instances
(see B422–23n). Importantly, the Fichtean self-intuition assumes an aware-
ness of the identity of the self qua self, whereas, for Kant, any intuition that
were involved as implied by or underlying the act of apperception, would as
such be merely a fleeting, instantaneous sensation. Kant’s own term
Selbstanschauung, which he uses in a footnote in §25 of the B-Deduction,
is precisely the opposite of Fichte’s, since it is not an intellectual intuition.12
12
In the Wissenschaftslehre nova methodo, however, Fichte appears to deny that his notion of intellectual
intuition is different from what Kant in fact means when speaking of the way we think of ourselves.
According to Fichte, the immediate intuition of myself as a thinker, whereby the thinking “I’” and the
thought “I” are one, is the inner intuition of the active “I,” and is intellectual; it forms the immediate
ground or basis on which I am able to think of myself. Fichte says that “this does not contradict the
For Kant, self-intuition is a sensible intuition under the a priori form of time,
which, if I were to attain knowledge of my self’s phenomenal identity (as an
observable person), is required in addition to the intellectual awareness that I
have of myself in the apperceptive act that accompanies my ϕ-ing, which is an
awareness of myself purely as an intelligence, as a logical subject (which, as
such, is empirically or psychologically speaking, empty, for the representation
of oneself in the act of apperception, namely the consciousness “that I am” in
the very act, is “a thinking, not an intuiting” [B157]). As Kant says, “the
determination of my existence can only occur in correspondence with the
form of inner sense,” which is time. As a result, in Kant’s view, I do not have
“cognition of myself as I am, but only as I appear to myself” (B158). Apart
from this sensible self-intuition,
I do not have yet another self-intuition, which would give the determining in
me, of the spontaneity of which alone I am conscious, even before the act of
determination, in the same way as time gives that which is to be determined,
thus I cannot determine my existence as that of a self-active being, rather I
merely represent the spontaneity of my thought, i.e., of the determining, and
my existence always remains only sensibly determinable, i.e., determinable as
the existence of an appearance. (B157–58n)
This departs significantly from the Fichtean notion of immediate evidence of

self-identity in an intellectual intuition. In the proposition “I know that I ϕ” or
“(I know:) I ϕ,” as an expression of what Tugendhat calls epistemic self-
consciousness, one does not just assume the identity between the “I” in the
main clause and the “I” in the that-clause, based on putative intuitional
evidence. There are a priori rules that govern knowledge of the identity between
Kantian system [and that] Kant only rejects a sensible intellectual intuition, and rightly so” (J. G. Fichte,
Wissenschaftslehre nach den Vorlesungen von Hr. Pr. Fichte, in Gesamtausgabe der Bayerischen Akademie der
Wissenschaften, pt. IV, vol. 2 [hereafter GA IV/2], ed. Reinhard Lauth et al. [Stuttgart Bad-Canstatt:
Frommann-Holzboog, 1978], 31). Importantly, Fichte says here – and this might be seen as contrasting
with my suggestion earlier, in the main text – that the “I” at issue here is only the “I” “for me,” “insofar
as I conceive of its concept through an immediate consciousness”; “another being of the ‘I’, as substance,
soul etc. is not at all at issue here” (GA IV/2:29, emphasis added). Therefore, the intellectual intuition
only provides one awareness of the activity of thinking of oneself, not of one’s putative substantial self or
soul (what Fichte refers to as a “fixed” or “resting” “I,” in contrast to an “acting” “I” [GA IV/2:31]). It
would thus appear that neither Kant nor Fichte have in mind a conception of the substantial “I,” when
they refer to the thinking “I.” But despite Fichte’s criticism of Kant as beholden to a reflexive s=o model
of self-consciousness, I should like to stress that Kant, too, believes that the “I” is first produced in the
activity of thinking (see B132). It remains to be seen to what extent, however, Fichte’s and Kant’s views
in fact concur, since Kant explicitly sees the “I” of thinking as a mere logical “I,” which is empty, whereas
Fichte portrays the “I” as self-positing, not as “beforehand already a substance,” but as such still positing
its own “essence” (GA IV/2:31). This latter element goes beyond what Kant would endorse.
148 D. Schulting
them, by means of which this knowledge is established – and not somehow

assumed on the basis of putative immediate intuitional evidence. These are the
transcendental rules, namely the categories, of which Kant speaks.13
For Kant, in the mere employment of the deictic expression or indexical “I”
no person is identified. Hence, there is no risk of misidentification. To identify
myself as a person, I need to have an empirical intuition of myself beyond the
mere representation “I,” in the same way as this would be required for any
other object in space and time. In employing the indexical “I,” which is not a
universal representation in terms of a concept, under which other representa-
tions can be subsumed (in the way that the concept “red” is), I am not
specifically referring to “Dennis Schulting,” but merely reflexively referring
to myself as the one who has any particular occurrent representations by self-
ascribing my representations to myself. The representation “I” does not in and
of itself single out me in particular, namely myself as this person with the name
“Dennis Schulting,” precisely because it is or can be employed by anyone who
self-reflexively ascribes his or her representations to him or herself, for any
actual occurrence of such self-ascription. “I” is merely a formal representation,
by means of which one self-ascribes one’s own representations to oneself, for
any representer who so self-ascribes (cf. A122). The rules that govern this use
of the formal representation “I,” or “I think,” is what Kant calls the conditions
that make up transcendental apperception. Kant argues for this in §16
(B131–36). This will be delineated in the next section.
The identity at issue in “epistemic” self-consciousness (and which I shall
argue is the right model for self-consciousness simpliciter) is the identity of
which I have knowledge in the following way: I know that I ϕ but this does
not mean that there is a separate act of knowing in addition to the act of ϕ-
ing, or that I have a second-order reflection upon myself as having certain
knowledge, let alone that I need to have knowledge of my substantial self-
identity in order to be able to ϕ. Tugendhat is right that there is strictly
speaking no identity between the “I” in the main clause and the proposition
“I ϕ,” in the sense that “I” as knowing is identical to the content of whichever
act of ϕ-ing she does (unlike in the Fichtean proposition “Iy=Ix”). But there is
an identity of sorts between the “I” that knows she is ϕ-ing and the very act of
her ϕ-ing. In other words, the identity lies between the act of knowing and
the act of doing: in ϕ-ing I am simultaneously aware of myself as ϕ-ing, or, I
know of myself as ϕ-ing, while ϕ-ing. I am not aware of myself as my own
13
Tugendhat dismisses Kant’s transcendental theory as “obscure” (Selbstbewußtsein und
Selbstbestimmung, 52). But it seems to me that here lies precisely the key to solving the riddle concerning
the alleged circularity. See the section “Transcendental Apperception and Self-Consciousness.”
object (as in the quasi-tautological proposition “Is=Io”), but I am aware of

myself as doing something, namely thinking this or that. The relation
between knowing that I ϕ and the fact that I ϕ, between the knowledge of
ϕ-ing and the act of ϕ-ing, is not dyadic, as the reflection-theoretical model
suggests, but the expression of an identity lies in the adverbial nature of the
relation between knowledge and act.14 To put it in Kantian terminology, I
apperceptively know that I ϕ while ϕ-ing. The relation between the apper-
ceptive knowledge and the act of ϕ-ing is one of identity nonetheless, as
Robert Pippin says.15
Transcendental apperception and self-consciousness

In §16 of the B-Deduction, as a first step in the argument toward the
formulation of the conditions of possible knowledge, Kant elucidates his
theory of self-consciousness. I am here not interested in the grounding
function of self-consciousness for knowledge, as Kant sees it, though it is
directly implied by the theory of self-consciousness.16 Kant argues that the
identity of self-consciousness is grounded in an act of original apperception,
which he calls pure or also transcendental apperception, to contrast it with
empirical apperception, which I shall not consider further here (though I do
discuss some aspects of inner sense, an alternate term for empirical appercep-
tion, in the section “Self-Knowledge”).17 Kant starts famously with the
proposition which says that the “I think must be able to accompany all
my representations.” In this proposition, Kant states that there is a necessary
unitary relation between the representation “I think” and the manifold of all
those representations that are accompanied by it. The unitary aspect of the
relation is made clear by Kant in stating that representations can be “my
representations” only if they “all together belong to one self-consciousness
14
See Robert B. Pippin, “Kant on the Spontaneity of Mind,” in Idealism as Modernism: Hegelian
Variations (Cambridge: Cambridge University Press, 1997), 43.
15
Robert B. Pippin, “The Significance of Self-Consciousness in Idealist Theories of Logic,” Proceedings
of the Aristotelian Society 114, no. 2, pt. 2 (July 2014): 155.
16
I have dealt with the grounding function of apperception for knowledge in Schulting, Kant’s
Deduction and Apperception; and Schulting, Kant’s Radical Subjectivism.
17
For a brief discussion of the distinction between transcendental and empirical apperception and their
relation, see Schulting, Kant’s Radical Subjectivism, ch. 4. For a different and more detailed view, see
Christian Onof, “Kant’s Conception of Self as Subject and Its Embodiment,” Kant Yearbook 2 (May
2010): 147–74.
150 D. Schulting
[zu einem Selbstbewußtsein]” (B132, translation modified18). The state of

together belonging to me signifies a certain kind of necessary unity among
the representations that are accompanied by the “I think,” which, as we shall
see, first constitutes the identity of self-consciousness. This unity of their
belonging together is what Kant calls “the transcendental unity of self-
consciousness” (B132). All too commonly, one is committed to reading
this transcendental-unitary relation between the “I think” and its representa-
tions in too strong a modal sense. The strong modal reading of the principle
of transcendental apperception says, roughly (whereby PTA stands for
Principle of Transcendental Apperception):
PTA = Necessarily, all representations of which I am conscious are subject to the

unifying act of apperception.
This definition of transcendental apperception is problematic. The prin-

ciple of apperception, that is, the “I think” proposition, specifically does not
state that, necessarily, for any and all representations r there is an actual or
potential instance of self-consciousness or a self-aware agent S, so that r is eo
ipso unified with all other representations represented, either potentially or
actually, by S. That would be a rather intemperate claim, and it would also
conflict with several of Kant’s statements regarding the possibility of repre-
sentations or even instances of consciousness which an arbitrary representer R
is not apperceptively aware of having (for example, the representations had
by infants, who are not yet able to employ the very concept of “I” [see An
7:127–28]).19 It would be odd to claim that on the grounds of occurrent
representations not being accompanied by the “I think” (that is, represented
by S), by implication R does not represent, or, is not in some non-appercep-
tive sense minimally aware of the representations it has. Stressing the neces-
sary potentiality of apperception does not help here, for infants represent
without ever apperceiving their representing, so that for them there is not
even a potentiality of apperception.20
18
Like Pluhar, Kemp Smith, and Meiklejohn, I read zu einem Selbstbewußtsein as meaning “to one self-
consciousness,” for if the indefinite article were meant here, the preposition and the pronoun would
normally have been contracted to zum.
19
For further references, see Dennis Schulting, “Transcendental Apperception and Consciousness in
Kant’s Lectures on Metaphysics,” in Reading Kant’s Lectures, ed. Robert R. Clewis (Berlin: de Gruyter,
2015), 89–113.
20
For more discussion, see Schulting, Kant’s Deduction and Apperception, ch. 8.
On a more moderate reading, the principle of apperception states that for any
and all representations r to be part of the set of all my representations, r must be
part of the set of representations that are conjointly accompanied by my identical
self as the agent of representing, by means of an analytic unity of consciousness,
which is common to all those representations accompanied by the same self.
This reading is in line with the criterial principle expressed at B138:
All my representations in any given intuition must stand under the condition
under which alone I can ascribe them to the identical self as my representa-
tions, and thus can grasp them together, as synthetically combined in an
apperception, through the general expression I think. (B138)
Notice that Kant puts emphasis on the indexical “my.” The criterion for
the unity among the representations in order to “belong to me,” to be related
to my identical subject, is the “analytical unity of apperception” (B132–33,
italics added). There lies a strict reciprocity, or analytic unity, between the
indexical “I” of the act of apperception or self-ascription and the “my” as the
indexical contained in the accompanied manifold of representations, of all
the representations that I ascribe to myself. This makes sense, as the only
representations that “I” ever accompany, or am able to accompany, will be
“my” representations, not yours, hers, or x’s, not even those that happen to
be occurrent in my head, but to which I do not currently direct my
attention.21 Therefore, the relation between the “I think” and its (my own)
representations is a biconditional one. The moderate reading of apperception
can be defined thus:
PTA* = A representation r is accompanied by subject S if and only if r is

analytically united with all representations that have the same relation to S and
S accompanies these representations conjointly, for which a certain condition
of combining must be fulfilled.
The condition of combining mentioned in the above definition, as well

as in the earlier quoted passage at B138, is the condition of a priori
synthesis, which is the explaining ground of the analytic principle PTA*.
The “thoroughgoing identity of the apperception of a manifold given in
intuition” (B133), which is expressed by PTA*, is first made possible by a
priori synthesis, or the original-synthetic unity of apperception;
21
For a detailed account of the modality of apperception, see Schulting, Kant’s Deduction and
Apperception, ch. 6.
152 D. Schulting
“thoroughgoing identity of self-consciousness could not be thought” with-

out the presupposition of a necessary synthesis of the representations given
in intuition (B135). This means that for Kant the identity of my repre-
sentations as belonging to me through their analytic unity, that is, my self-
identity as thinker, is not a priori given, of which I would have a priori
knowledge (through immediate intuitional evidence, say), but must be
established by a spontaneous act of unification of my representations –
Kant calls the “I think” “an act of spontaneity,” which “cannot be
regarded as belonging to sensibility,” and so is not given in or with it,
and is in fact the self-consciousness that accompanies all other representa-
tions, and “cannot be accompanied by any further representations”
(B132), which constitutes it as spontaneity proper. The “I think” or self-
consciousness is spontaneous, original, properly basic, yet it rests on an a
priori synthesis, which is the aforementioned transcendental unity of self-
consciousness, which, as an “original combination” (B133), holds the
representations accompanied by the “I think” together as self-same repre-
sentations. Does the originality and spontaneity of self-consciousness or
the “I think” and the fact that it rests on an a priori synthesis not amount
to a contradiction? No. For the “I think” and the a priori synthesis or the
transcendental unity of self-consciousness, or also called the “original-
synthetic unity of apperception” (B131, heading §16; B135), are equipri-
mordial. Kant explains:
The thought that these representations given in intuition all together belong to
me means . . . the same as that I unite them in one self-consciousness, or at least
can unite them therein, and although it is itself not yet the consciousness of the
synthesis of the representations, it still presupposes the possibility of the latter,
i.e., only because I can comprehend their manifold in one consciousness do I
call them all together my representations. . . . Synthetic unity of the manifold of
intuitions, as given a priori, is thus the ground of the identity of apperception
itself, which precedes a priori all my determinate thinking. (B134, translation
modified and emphasis added)
The a priori synthetic unity as the “ground of the identity of appercep-

tion itself” is a logical ground. It is not as if it preceded the act of
apperception; rather, it enables it in the very act of apperception.
Identity of self-consciousness consists in the act of unifying or accom-
panying (in the above-described sense), by means of a priori synthesis,
one’s self-same representations, which is the unification of one’s own
representations in a transcendental unity of self-consciousness (B132).
This act of unifying is in fact the act of a priori synthesis, for it is an act that puts
representations together in a synthetic unity as originally belonging together
(originally, to the extent that they are all equally mine). Hence, Kant calls it (in
the heading to §16 [B131]) the “original-synthetic unity of apperception,”
which combines the unity and act aspects of a priori synthesis. The sharing of
the analytic unity (the indexical “I”) among the representations accompanied by
the “I” is rigorously coextensive with the a priori putting them together in a
synthetic unity, a representational whole of representations as together belong-
ing to the “I,” whose same (analytic) feature they share, namely by being my
representations – of course, these representations belong together synthetically,
in an a priori way, only to the extent that they share this same analytic feature.
The element of togetherness is emphasized by the synthesis, while the sharing of
the same indexical “my” is made manifest by the element of analytic unity. This
identity among my representations is first made possible through the fact that I
put them together by accompanying them all as my representations, “by my
adding one representation to the other and being conscious of their synthesis”
(B133).22 The reciprocity between the analytic and synthetic unities of apper-
ception is specifically asserted by Kant in the following passage:
It is only because I can combine a manifold of given representations in one

consciousness that it is possible for me to represent the identity of the
consciousness in these representations itself, i.e., the analytical unity of
apperception is only possible under the presupposition of some synthetic one.
(B133, italics added)23
All this comports with my earlier statements in the section “Self-

Consciousness and Identity” that there is a relation of identity between the
“I” of “I know” and the “I” of “I ϕ” in the proposition “(I know:) I ϕ.” The
so-called “epistemic” self-consciousness expressed here (as Tugendhat terms
it) is the transcendental unity of self-consciousness, which first constitutes the
thoroughgoing identity of self-consciousness, that is, self-consciousness sim-
pliciter, which is not based on, or grounded in, an intuitional evidence of an a
priori self-sameness, but in an original, spontaneous act that first establishes
“identity of the consciousness in these representations” (B133), namely,
22
This addition should not be read as if the synthesis were an a posteriori one, which might be suggested
by Kant’s words here. The act of addition, that is, synthesis, is what happens in the background of any
act of apperceiving representations as one’s own, and which constitutes self-consciousness. The act of
addition or synthesis is not something that I need to do consciously.
23
I have extensively argued for what I call the rigorous coextensivity of the analytic and synthetic unities
of apperception in Schulting, Kant’s Deduction and Apperception, esp. 110–13.
154 D. Schulting
the representations in any manifold in intuition, which I accompany and

take as together belonging to me. For Kant, through the “I” neither a
manifold nor its combination is given, which I could intuitively represent,
as in an intellectual intuition (B135), so I am not intuitively aware of my
own a priori identity (which is what Fichte does suggest is the case in the
background of his proposition “Iy=Ix”). I have to combine the representations
in any given empirical intuition in order to first attain awareness of my
identity, namely of myself as the self accompanying these representations.24
Hence, Kant says that
I am therefore conscious of the identical self in regard to the manifold of the

representations that are given to me in an intuition because I call them all
together my representations, which constitute one. But that is as much as to
say that I am conscious a priori of their necessary synthesis, which is called the
original synthetic unity of apperception, under which all representations given
to me stand, but under which they must also be brought by means of a
synthesis. (B135–36)
The necessary synthesis which a priori grounds the identity of my self-

consciousness is obviously not manifestly present in the proposition “(I know:)
I ϕ,” and neither am I of course literally (psychologically) conscious of the a
priori synthesis as such, yet in any apperceptively accompanying my ϕ-ing, I am
implicitly aware of the necessary combining of the various representations as
mine in whatever cognition I have, that is, regardless of the specific content of
my ϕ-ing. It is this apperceptive implicit awareness of a priori synthesis that
paradigmatically constitutes self-consciousness, that is, an awareness of my iden-
tity of the action of combining my representations as my own for any case of
cognition or ϕ-ing – it is the mind’s consciousness of “the identity of the function
by means of which [a] manifold is synthetically combined into one cognition”
(A108, emphasis added). But, importantly, the consciousness of the identity of
this combinatory function of the cognitive mind does not constitute knowledge,
strictly speaking, of my self’s identity, that is, knowledge of self in a more
traditional sense of self-consciousness. For knowledge of self something more is
required, on Kant’s account, in a way that is parallel to what establishes possible
knowledge of spatiotemporal objects. This I address in the next section.
24
I must set aside here issues of modality that are alluded to in the above-quoted passage at B133. For
more discussion, see Schulting, Kant’s Deduction and Apperception, ch. 6.
Self-knowledge
One of the central aspects of Kant’s transcendental theory of self-conscious-
ness, which crucially separates his from previous theories, is the distinction
between transcendental apperception and inner sense, the latter also called
empirical apperception (A107). Philosophers in the rationalist tradition of
rational psychology (Leibniz, Wolff, Baumgarten) did not distinguish
between inner sense and transcendental apperception. For them, inner
sense was self-consciousness simpliciter. They did not differentiate pure
from empirical forms of self-consciousness.25 For the rational psychologists,
this meant that the experience in inner sense amounted to a self-conscious-
ness that is identical to the experience of self as a substance, and as a person.
This identity is conveyed by the idea of immediate evidence of one’s self
through inner sense, which provides knowledge of one’s immaterial soul. For
the rational psychologist, the empirical experience of oneself in inner sense or
self-consciousness amounted to an intimate, empirically manifest self-
acquaintance that, on this very basis, grounds any further rational knowledge
claims about the soul (i.e., its immateriality, substantiality, personality).
At the end of the previous section, I pointed out that by contrast, for Kant,
self-consciousness does not ipso facto entail self-knowledge. As Kant says, “the
consciousness of oneself is . . . far from being a cognition of oneself [ein
Erkenntnis seiner selbst]” (B158). This is precisely because, for Kant, trans-
cendental apperception is not to be identified with inner sense, which, if it
were, would yield an intuitional knowledge of the self’s innermost nature, by
means of an intellectual intuition (B159), namely, knowledge of how I
would be in myself (B157), just as his rationalist predecessors believed.
However, Kant acknowledges, his differentiation of self-consciousness as
transcendental apperception from inner sense does create a paradox, for it
seems to divide the self into an active and a passive part, which raises the
question, as Kant says (B153), of how it is at all possible that we “relate to
ourselves passively” (as is the case in inner sense, which Kant calls the
“passive subject” at B153). It also invites the question as to why it is that
we are not able to attain cognition of ourselves as we, qua thinking beings, are
in ourselves if it is the case that we do have a direct awareness of what makes
us thinkers, namely of the original synthetic unity of apperception in any act
of cognition. This paradox regarding the active and passive subjects is
For discussion, see Schulting, “Transcendental Apperception and Consciousness in Kant’s Lectures on
25
Metaphysics.”
156 D. Schulting
addressed in an excursus in the B-Deduction (§24), but it is described by

Kant in more detail in a long, crossed-out passage from his manuscript for
the 1798 Anthropology from a Pragmatic Point of View.26 It is worth quoting
that passage in full here (the crossed-out passage is appended to §7 in the
Anthropology):
How then is the great difficulty to be removed, in which consciousness of

oneself still presents only the appearance of oneself, and not the human being
in himself? And why does it not present a double I, but nevertheless a doubled
consciousness of this I, first that of mere thinking but then also that of inner
perception (rational and empirical); that is, discursive and intuitive appercep-
tion, of which the first belongs to logic and [the] other to anthropology (as
physiology)? The former is without content (matter of cognition), while the
latter is provided with a content by inner sense. . . .
And no one doubts that we could not equally make inner observations of
ourselves and make experiences in this way, but if we dare now to speak of
objects of inner sense (which as sense always provides appearances only) it is
because we are able to reach only cognition of ourselves, not as we are, but as
we appear (internally) to ourselves. There is something shocking in this
proposition, which we must consider more carefully. – We allow a judgment
of this kind regarding objects outside us, but it looks quite absurd to apply it to
what we perceive within ourselves. – That some word-twisters take appearance
and semblance (Erscheinung und Schein) for one and the same thing and say that
their statements mean as much as: “it seems (scheint) to me that I exist and have
this or that representation” is a falsification unworthy of any refutation.
This difficulty rests entirely on a confusing of inner sense (and of empirical self-
consciousness) with apperception (intellectual self-consciousness), which are
usually taken to be one and the same. The I in every judgment is neither an
intuition nor a concept, and not at all a determination of an object, but an act
of understanding by the determining subject as such, and the consciousness of
oneself; pure apperception itself therefore belongs merely to logic (without any
matter and content). On the other hand, the I of inner sense, that is, of the
perception and observation of oneself, is not the subject of judgment, but an
object. Consciousness of the one who observes himself is an entirely simple
representation of the subject in judgment as such, of which one knows
everything if one merely thinks it. But the I which has been observed by
itself is a sum total of so many objects of inner perception that psychology
has plenty to do in tracing everything that lies hidden in it. And psychology
26
See also the account at B68–69 in the Critique.
may not ever hope to complete this task and answer satisfactorily the question:
“What is the human being?”
One must therefore distinguish pure apperception (of the understanding) from
empirical apperception (of sensibility). The latter, when the subject attends to
himself, is also at the same time affected and so calls out sensations in him, that
is, brings representations to consciousness. These representations are in con-
formity with each other according to the form of their relation, the subjective
and formal condition of sensibility; namely, intuition in time (simultaneously
or in succession), and not merely according to rules of the understanding.27
As is made clear in this passage, Kant makes a distinction between the logical
conditions for pure self-consciousness, and the conditions for a concrete self-
consciousness, for “observation of oneself” or self-perception. The former con-
ditions are dealt with in §16 of the B-Deduction, which I addressed in the
previous section: it concerns the consciousness or a priori cognition of the pure
apperceptive “I,” who is always implicitly aware of herself as ϕ-ing. In pure self-
consciousness, we do not perceive ourselves as we are in ourselves, but merely as
intelligences (B158n), as subjects aware of their ϕ-ing. The latter conditions, by
contrast, concern the perception of the self as an object, which is to be treated in
the same way that objects outside of oneself are perceived – this concerns Kant’s
view of the strict parallelism between the knowledge of the self as an object and
the knowledge of outer objects, a topic that resurfaces in the Refutation of
Idealism (B274–79). In fact, there is no difference in principle between the way I
can know, strictly speaking,28 about myself and how I cognize objects outside me:
in both cases, I know only an appearance, not the intrinsic nature of how I am in
myself or things in themselves, based on the notion that this knowledge, in both
cases, is constrained by empirical intuition (B156). This solves the paradox that
Kant noted at the beginning of the excursus at B152–56. Kant writes:
But how the I that I think is to differ from the I that intuits itself (for I can
represent other kinds of intuition as at least possible) and yet be identical with
the latter as the same subject, how therefore I can say that I as intelligence and
thinking subject cognize my self as an object that is thought, insofar as I am
also given to myself in intuition, only, like other phenomena, not as I am for
the understanding but rather as I appear to myself, this is no more and no less
27
Immanuel Kant, Anthropology, History, and Education, ed. Günter Zöller and Robert B. Louden
(Cambridge: Cambridge University Press, 2007), 253, boldface added.
28
This knowledge is more than just the “knowledge” of merely thinking the subject of judgment, as the
highlighted Anthropology passage suggests.
158 D. Schulting
difficult than how I can be an object for myself in general and indeed one of
intuition and inner perceptions. (B155)
In order to perceive oneself in terms of self-observation or self-knowledge, that

is, when I am “an object for myself,” “a determinate sort of intuition,” more in
particular, a “self-intuition” (B155, B157, B157n) is required, “in addition to
that which I think myself” (B158), that is, in addition to the purely logical
original-synthetic unity of apperception or the faculty of combination, of
which I am aware as the “I” of pure self-consciousness. This self-intuition “is
grounded in an a priori given form, i.e., time, which is sensible and belongs to
the receptivity of the determinable” (B157n). Through the synthesis of the
understanding in sensibility, which is a “transcendental synthesis of the
imagination,” it is then possible to determine “sensibility internally,” by
ordering the time determinations in inner sense. This determinative synthesis
is exercised “on the passive subject,” that is, on inner sense (B153). Kant calls
this the affection of the inner sense or internal affection (in the literature
sometimes referred to as “self-affection,” though Kant himself does not use this
term as such; he speaks of being “internally affected by our selves” at B156; see
also B68).29 Inner sense is “mere form of intuition,” which “does not yet
contain any determinate intuition at all” (B154),30 while pure apperception is
“the source of all combination” (B154; cf. B129–30) and is what provides the
determination to the mere manifold in inner sense (the passive subject). What
Kant means by the fact that the understanding, through the determining act of
pure apperception, affects the inner sense by “bringing it under an appercep-
tion” (B153), is that the combination that first yields knowledge of an object is
produced by the understanding (hence, the name “productive imagination,”
while the imagination is the “effect of the understanding on sensibility and its
first application” [B152]). This affection of inner sense, as the way manifolds
of representations in an empirical intuition are determined, holds true, mutatis
mutandis, for internal as well as outer objects, so for knowledge or determinate
perception of oneself as well as of objects outside me – again confirming Kant’s
29
The term self-affection suggests that only self-perception is concerned in the affective determination of
inner sense, but internal affection is involved in all determination of sensible manifolds, for any sort of
empirical object, as is made clear in the preceding passages of §24. The difference is that in the cognition
of outer objects, we are also externally affected.
30
For detailed discussion on the differentiation between mere form of intuition and determinate
intuition in relation to space as one of the two pure forms of intuition, see Christian Onof and
Dennis Schulting, “Space as Form of Intuition and as Formal Intuition: On the Note to B160 in
Kant’s Critique of Pure Reason,” Philosophical Review 124, no. 1 (Jan. 2015): 1–58; and Schulting, Kant’s
Radical Subjectivism, ch. 7.
strict parallelism between subject and object as far as the epistemology of

knowledge is concerned. Insofar as the conditions of its possibility are con-
cerned, self-observation or self-perception is thus no different, and no more
problematic, than the perception of outer objects.
In and of itself, the a priori synthesis, which determines, and thus affects,
the mere manifold in inner sense, “is nothing other than the unity of the
action of which [the self of understanding] is conscious as such even without
sensibility” (B153). The self-intuition is merely empirical, but the internal
affection mentioned earlier is nothing but the determination by the under-
standing of this intuition in inner sense – the affection itself is thus nothing
psychological, as the term might suggest. I do not have “yet another self-
intuition, which would give the determining in me, of the spontaneity of
which alone I am conscious, even before the act of determination” (B158n,
italics added). Therefore, “I cannot determine my existence as that of a self-
active being, rather I merely represent the spontaneity of my thought”
(B158n).31 This is crucially different from any kind of self-knowledge that
supposes that we have an immediately, intuitionally evidential cognition of
ourselves as intelligences, along the lines of Fichte’s Selbstanschauung,
whereby I a priori intuit my “determining” self, even before I exercise it in
whatever act of determinative cognition. Kant categorically rejects the idea of
an intellectual intuition (B159; see also B68, B135, B138–39, B148–49,
B153, B157–59). For Kant, self-knowledge does not consist in, nor does it
presuppose, intuitive knowledge of a strict identity of subject and object in
the sense of Fichte’s quasi-tautology “Iy=Ix.” Rather, it concerns merely an
identity relation between the apperception as the determining act and a
sensible intuition insofar as the latter is determined by it, which consequently
yields a cognition of myself as an appearance only, not as a thing in itself, that
is, not essentially. This identity rests upon, and is first constituted by, a
function of synthesis of a given empirical manifold in inner sense, and
consists strictly speaking in the analytic unity of apperception (B133). I
have no prior intuitive acquaintance with my determining self; I am
acquainted only with my self as the subject of apperception, through affect-
ing manifolds of representations in inner sense. Only the representations that
I combine as together belonging to the unity of my apperception, which
defines an objective unity of consciousness (and hence an object for thought)
31
Compare this with the following passage in the B-Paralogisms: “Thinking, taken in itself, is merely
the logical function and hence the sheer spontaneity of combining the manifold of a merely possible
intuition; . . . in the consciousness of myself in mere thinking I am the being itself, about which,
however, nothing yet is thereby given to me for thinking” (B428–29).
160 D. Schulting
(B137), constitute an object of my thought, regardless of the question

whether this concerns an outer or inner object (myself as an object of self-
perception). I do not have privileged knowledge of my own determining self
as an inner object.
Of course, the “I think” does imply the existence of the “I,” for every
instantiation of the “I,” in every act of apperception, but it does not yet
determine or give thereby the “I” as an object of immediate perception
(cf. B157n, B422n). As Kant says, “in the synthetic original unity of
apperception, I am conscious of myself not as I appear to myself [as an
object of empirical perception], nor as I am in myself [as in a Fichtean
Selbstanschauung], but only that I am” (B157). This comports with the
emphatic distinction between the self-consciousness of pure apperception
and the self-intuition in inner sense, which enables self-perception or self-
knowledge. The existence of the thinking self as “the determining in me”
(B158n), that is, the act of transcendental apperception, is logically implied
in the act: the apperceptive “I” that is ϕ-ing is instantiated in an existing
person doing the ϕ-ing, namely my occurrent thinking self. But to determine
the existence of this person, that is, if I were to determine myself as this
particular person doing my occurrent ϕ-ing, namely,“Dennis Schulting,” requires
more than self-consciousness. In transcendental self-consciousness, I am not
aware of myself as this particular person, “Dennis Schulting”; I am just aware of
myself as ϕ-ing, where “myself” rigidly refers to me, that is, to the person that
ϕ-s, but at the same time does not rigidly designate me as this particular
person, “Dennis Schulting.” For the latter, an additional act of self-affection is
required, which determines a certain manifold in my sensibility that identi-
fies me, objectively, as this particular person, “Dennis Schulting.” This identi-
fying information is objective, in that it concerns information that is
objectively valid, and thus, at least potentially, intersubjectively available
(in the same way that any empirical evidence for outer objects would be
intersubjectively available). This information is not private. I myself was told
or came to know, when I was much younger, that that was my identifying
name. But the identifying information does not play the central role in
identifying myself as myself (though it may play a role). I could identify
myself as myself even if I did not know who I was strictly speaking, or what
name I had (bar severe cognitive malfunction, whereby I had lost, not just
my memory, but also my apperceptive capacities); or indeed if I were
thoroughly mistaken about who I was. Indexicals such as “myself,” “mine,”
“my,” and “I” are instantiated by any and all subjects of acts of ϕ-ing, who
are apperceptively aware of themselves as ϕ-ing, regardless of their particular
names, histories, personalities, and so on.
The internal affection by the understanding of one’s inner sense is the

“concretization,” as it were, of the combinatory capacity of pure appercep-
tion, of my pure intelligence or the a priori act of synthesis in any act of ϕ-
ing. This holds for inner as well as outer objects of perception. It is clear,
both from the accounts of pure, transcendental self-consciousness, or pure
apperception, and the inner affection of one’s inner sense by oneself, which
enables self-observation, that Kant is not concerned with the psychological or
anthropological aspects of subjectivity, the aspects that most people associate
with subjectivity, with our innermost, purely subjective or private experi-
ences or mental states. Indeed, as Kant says in the above-quoted crossed-out
passage from the Anthropology, “the I which has been observed by itself is a
sum total of so many objects of inner perception that psychology has plenty
to do in tracing everything that lies hidden in it.” Transcendental philoso-
phy’s view of self-consciousness, and even of self-knowledge, is more limited
in its scope and need not concern itself with all these “many objects of inner
perception,” but is merely concerned with the a priori constraints under
which both self-consciousness and self-knowledge can take place, the condi-
tions namely that first define what it means to be able, in the strictest sense,
to have consciousness or indeed knowledge of oneself at all.32
32
I would like to thank Christian Onof and Marcel Quarfood for their comments on an earlier draft of
this paper.
Part III
Logic
8
The Place of Logic within Kant’s Philosophy
Clinton Tolley
Logic and the Copernican turn

At first glance, it might seem that logic does not play a central role in Kant’s
critical philosophy. Kant himself authored no books or essays on logic during
the critical period1; indeed, in his whole career, he wrote only one essay
specifically on logic, his early 1762 essay “False Subtlety,” on the figures of
the syllogisms – hence, well before his so-called “Copernican” turn. The
most well-known remarks Kant makes about logic during the critical period
itself can surely suggest he does not take this discipline to be of much interest
for his own revolutionary program. At the outset of the B-edition preface,
Kant famously claims that, since the time of Aristotle, logic has been “unable
to take a single step forward, and therefore seems in every respect to be
finished and complete” (Bviii, translation modified). Indeed, immediately
thereafter Kant contrasts the already “finished and complete” standing of
logic with the “much more difficult” task that the Critique itself will aim to
1
The book that appeared in 1800 under the title of Immanuel Kants Logik was not authored by Kant
himself, but was written up by one of his students, G. B. Jäsche, on the basis of Kant’s lecture notes, and
there is no evidence that Kant himself ever reviewed Jäsche’s manuscript at any stage of its composition.
See Terry Boswell, “On the Textual Authenticity of Kant’s Logic,” History and Philosophy of Logic 9,
no. 2 (1988): 193–203; and J. Michael Young, “Translator’s Introduction,” in Lectures on Logic by
Immanuel Kant (Cambridge: Cambridge University Press, 1992), xv–xxxii.
C. Tolley (*)
Department of Philosophy, University of California, San Diego, La Jolla, USA
e-mail: ctolley@ucsd.edu

166 C. Tolley
accomplish: that of getting “reason [Vernunft]” on “the secure path of a

science” (Bix).
This impression can seem to be further confirmed when we look into the
content of Kant’s critical philosophy itself. The signature doctrine that Kant
takes to resolve the various conflicts that reason gets itself into, as it tries to
find its way to science – that is, the transcendental idealism underlying Kant’s
Copernican revolution – might seem to be a doctrine primarily concerned
with correcting a misunderstanding of the nature of our sensibility rather than
one concerning thought, inference, or reasoning per se. For one thing, the
core of Kant’s idealism is presented and developed within the Transcendental
Aesthetic, which is the science of sensibility (A52/B76), rather than in the
section of the Critique entitled “Logic.” Transcendental idealism consists in
the claim that what is immediately given in our sensible intuitions – what
Kant calls “appearances [Erscheinungen],” and the space and time that they
fill – are objects that “cannot exist in themselves, but only in us,” by being
contained “in” our “representations [Vorstellungen]” (A42/B59; see also
A490–94/B518–22). And when Kant does turn, finally, to the task of
using transcendental idealism to diagnose what goes wrong with our reason
itself, in the Transcendental Dialectic, the problems that reason falls into are
explicitly stated not to be due to reason’s failure to operate in accordance
with any “logical principle,” but rather due to reason’s attempt to go beyond
acting in accord with logical principles to asserting the objective validity of
certain “transcendental principle[s]” (A648/B676).
A closer look at Kant’s critical writings, however, shows these sorts of
initial impressions to be deeply misleading. Recent advances in scholar-
ship have helped to make it increasingly clear that Kant’s thoughts about
logic stand at the center of his philosophical development, throughout
his career.2 For several reasons this should come as no surprise. For one
2
For the significance of Kant’s views on logic for his early writings, see Peter Yong, “God, Totality and
Possibility in Kant’s Only Possible Argument,” Kantian Review 19, no. 1 (March 2014): 27–51; and
Nicholas F. Stang, Kant’s Modal Metaphysics (Oxford: Oxford University Press, 2016). For the sig-
nificance of Kant’s changing views on logic for the emergence of the critical philosophy, see R. Lanier
Anderson, The Poverty of Conceptual Truth: Kant’s Analytic/Synthetic Distinction and the Limits of
Metaphysics (Oxford: Oxford University Press, 2015). For the centrality of Kant’s conception of logic
within the critical philosophy itself compare Michael Friedman, Kant and the Exact Sciences (Cambridge:
Harvard University Press, 1992); Béatrice Longuenesse, Kant and the Capacity to Judge: Sensibility and
Discursivity in the Transcendental Analytic of the “Critique of Pure Reason,” trans. Charles T. Wolfe
(Princeton: Princeton University Press, 1998); John MacFarlane, “Frege, Kant, and the Logic in
Logicism,” Philosophical Review 111, no. 1 (Jan. 2002): 25–65; Clinton Tolley, “Kant’s Conception
of Logic” (PhD diss., University of Chicago, 2007); Huaping Lu-Adler, “Kant on Proving Aristotle’s
Logic as Complete,” Kantian Review 21, no. 1 (March 2016): 1–26; and Anderson, Poverty of Conceptual
Truth.
8 Place of Logic within Kant’s Philosophy 167
thing, Kant gave lectures on logic continuously, every year except one,
and more frequently than on any other topic.3 Indeed, his own appoint-
ment was as a professor of logic (and metaphysics). With respect to the
critical period in particular, Kant makes clear (in the very same B-preface
passage noted above) that his critique of reason itself actually “presupposes
[voraussetzt] a logic for the assessment [Beurteilung]” of the alleged bits of
“information [Kenntnis]” that are taken to make up the science of reason
(Bix, translation modified). What is more, by far the largest part of the
first Critique itself is actually classified as a kind of logic – namely, what
Kant calls a “transcendental logic” (A50–704/B74–732). Finally, as we
will see below in more detail, at the outset of each main part of the
Critique’s Transcendental Logic (the Analytic and the Dialectic), Kant
explicitly points to the findings of the traditional logic – more specifi-
cally, its account of the forms of judging and inferring – as providing the
key starting point for the relevant stage in the investigation of the
possibility of the science of reason itself (see A299/B356).4
In what follows I will limit my task primarily to spelling out in more
detail how Kant’s thinking about logic during the critical period shapes
the account of philosophy that he gives in the Critiques. I will focus
especially on the role that Kant accords to logic within theoretical philo-
sophy. I will proceed as follows. First, I will provide an account of what
Kant means by claiming that logic is the science of “understanding in
general” and the activity of thinking. I will then turn to Kant’s motiva-
tions behind his formation of the idea of a new “transcendental” logic,
drawing out in particular how he means to differentiate it from the
traditional “merely formal” approaches to logic, insofar as transcendental
logic investigates not just the basic forms of the activity of thinking but
also its basic contents. I will then show how Kant’s understanding of both
of these logics directly factor into the first Critique’s more general project
of the critique of reason, now considered not just as a capacity for a
certain kind of thinking (inferring), but as a possible source of a priori
cognition. I will end by taking up an even broader perspective, to show
how Kant takes the findings of logic to provide architectonic structure
even to parts of philosophy outside of the doctrine of specifically theore-
tical cognition.
3
Compare Tolley, “Kant’s Conception of Logic,” 30. See also Chapter 2 of this volume.
4
It is worth noting that “the doctrine of reason [Vernunftlehre]” was a common title for logic at the time.
It was, in fact, the title of the textbook by Georg Meier that Kant used for his own lectures on logic.
168 C. Tolley
Logic as the science of understanding (thinking)

Kant takes the subject matter of logic to be what he calls “the understanding
[Verstand],” which he takes to be a “capacity [Vermögen]” of our “mind
[Gemüt]” for a certain kind of representational activity. More specifically,
logic is the “science [Wissenschaft]” which specifies the “rules [Regeln]” or
“laws [Gesetze]” according to which this capacity acts or is “used.”5 The most
general name for the representational activity which is distinctive of the
understanding is “thinking [Denken]” (A69/B94; see also A51/B75; Pro
4:304). Thinking itself consists in“unifying representations in one conscious-
ness [Bewußtsein]” (Pro 4:304, translation modified). Thinking thus con-
trasts with merely having a manifold of representations in mind, since it
involves a unifying of them. Thinking also contrasts with merely having
representations in mind unconsciously (see An 7:135); it involves bringing
representations to consciousness. Thinking is, however, dependent upon
having representations already present in mind, since our understanding
“only reflects” on what has already been given to our mind, rather than
being able to “intuit” (receive representations) on its own (Pro 4:288,
emphasis added).
The resulting “one consciousness that unifies the manifold” of representa-
tions (what Kant also calls a “consciousness of this unity of the synthesis”) is
what he calls a “concept [Begriff],” as he thinks that the very word suggests just
this idea of consciously grasping together (A103). For this reason, “to think”
can be understood as essentially: “to represent something to oneself in a
concept” (DWL 24:695; see also A69/B94) – where what is represented “in”
(or through) the concept is a unity of some other representations.
A concept itself “rests on” what Kant calls a “function,” which is “the unity
of the action [Handlung] of ordering different representations under a com-
mon [gemeinschaftliche] one” (A68/B93). Kant holds that our understanding
possesses a variety of distinct “functions of thinking” (A70/B95), each of which
leads to a different kind of consciousness of a unity of a manifold of repre-
sentations. This consciousness comes in four basic kinds: mere conceiving (JL
9:91–92), judging (JL 9:101–2), inferring (JL 9:114–15), and systematic
5
As Kant puts it at the outset of the Transcendental Logic and elsewhere, logic is “the science of the rules
of understanding in general” (A52/B76). Very similar definitions can be found in Kant’s lectures and
notes (Reflexionen) on logic. Compare, for instance, the Latin rendering given in the 1790s Vienna Logic:
“Definition. Logica est scientia regularum universalium usus intellectus” (VL 24:792; see also Ak 16:46
[R1628]); see also the earlier (1773–1775) Reflexion 1603: “Logic is an a priori science of the [universal]
pure laws of the understanding and reason in general” (NF 16:33).
ordering, as is exemplified in a science (JL 9:139–40). This differentiation in

basic kinds of thinking also correlates with a differentiation in which aspect of
“understanding in general” is responsible for each type of thinking. Kant
thinks that it is understanding in a more “specific” sense that is responsible
for concepts, whereas it is the “power of judgment [Urteilskraft]” that is
responsible for judging (A130/B169), and “reason [Vernunft]” that is the
capacity for inferring (A299/B355) and ordering (A832/B860), respectively.
Here is how he exemplifies the first three kinds of thinking in his logic lectures:
The understanding is the faculty of representation of the universal [Allgemeine]

as such. {E.g., the definition of man in general.}
The power of judgment is the faculty of representing the particular as contained
under the universal {Caius is a man}[,] or the faculty of subsumption.
Reason is the faculty of the derivation [Ableitung] of the particular from the
universal . . . {All men are mortal. Sempronius is a man, too. Sempronius is
mortal.} (DWL 24:703–4)6
Although each of these kinds of acts of thinking are distinct from one
another, what they all have in common is that they are acts of unifying
representations together in one consciousness, that is, grasping them in a
unity.
Now, by taking logic to be first and foremost about acts of thinking and
the exercise or use of our “powers of mind” (to represent, subsume, derive,
and so on) to “unify” things in “consciousness” in various ways, Kant follows
the early modern tradition in the philosophy of logic by taking its subject
matter to be something essentially mental and hence psychological.7 This,
however, does not mean that logic coincides with the empirical study of the
mind. This is because Kant does not think that the manner in which logic
investigates thinking is restricted to how individual acts of thinking are given
to the mind through inner “sensation [Empfindung]” or empirical “intuition
[Anschauung],” in inner appearances – let alone is logic thought to be some-
how restricted to these inner appearances themselves. Rather, Kant thinks
6
This is so, even if Kant often uses the term “understanding in general” in what he calls its “broad
designation,” which encompasses all three of these “powers of the mind [Gemütkräfte]” (A131/B169; see
also An 7:196–97). This broad designation also carries over for the use of the term “thinking” (see LM
29:888–89).
7
This subordination of logic to psychology is made especially vivid in the classification that Alexander
Baumgarten gives in his Acroasis logica, §37. See Clinton Tolley, “The Relation between Ontology and
Logic in Kant,” in International Yearbook for German Idealism, ed. Dina Emundts and Sally Sedgwick,
vol. 12 (Berlin: de Gruyter, 2017), 95–98.
170 C. Tolley
that there can be a “pure [reine]” logic, which “has no empirical principles”
and so “draws nothing” from the empirical science of the mind. This
contrasts with what Kant calls “applied [angewandte]” logic, which would
provide “a representation of the understanding and the rules of its necessary
use in concreto,” which by contrast “can all be given only empirically,” and so
which “requires empirical and psychological principles” (A54–55/B78–79,
emphasis added).
Even so, in both its pure and applied form, logic is a science whose subject
matter is a specific sort of mental or psychological activity – namely, think-
ing. In this it contrasts, first, with other sub-branches of psychology, which
are distinguished from logic by the specific mental capacity they have in view.
The most prominent contrasting sub-branch in the first Critique is what
Kant calls “aesthetic,” understood to be “the science of the rules of sensibility
in general,” where “sensibility” itself is understood to be “the receptivity of
our mind to receive representations” (A51/B75) – in particular, to receive
sensations and intuitions (A50–51/B74–75). The subject matters of aesthetic
and logic are therefore importantly disjoint, insofar as “these two faculties or
capacities [Fähigkeiten] cannot exchange their functions,” since “the under-
standing is not capable of intuiting anything” and “the senses are not capable
of thinking anything” (A51/B75).
Now, by having as their subject matter something specifically mental or
psychological (namely, a specific capacity for acts of representing), both
logic and aesthetic contrast with two other types of sciences: on the one
hand, they contrast with sciences whose subject matter is something speci-
fically not psychological, for example, physics, understood as the science of
corporeal substance; on the other hand, they contrast with sciences whose
subject matter is not specifically psychological, for example, ontology, under-
stood as the science of the most universal predicates of being “in general”
(see A845–46/B873–74). The latter contrast is especially worth emphasiz-
ing, insofar as Kant’s conception of logic therefore stands at some remove
from more recent conceptions of logic which, following Bertrand Russell,
take logic itself to be the science with the most universal domain.8 For his
part, Kant takes the subject matter of logic to have a very specific domain,
since not everything is an act of thinking; indeed, not even everything
mental or psychological is such an act (namely intuiting). In other words,
8
Compare Warren Goldfarb, “Frege’s Conception of Logic,” in Future Pasts: The Analytic Tradition in
Twentieth-Century Philosophy, ed. Juliet Floyd and Sanford Shieh (Oxford: Oxford University Press,
2001), 25–41; for discussion, see MacFarlane, “Frege, Kant, and the Logic in Logicism.”
for Kant, as for his predecessors, the domain of logic is subordinate to both
psychology and (a fortiori) ontology.9
Yet while Kant is fairly traditional in his understanding of the subject
matter of logic, Kant departs sharply from his early modern predecessors, and
looks more distinctively modern, in his understanding of the manner in
which logic treats this subject matter. As noted above, Kant is quite explicit
that he takes logic to constitute a “science [Wissenschaft]” of the understand-
ing and the laws of thinking, whereas earlier authors (for example, the
authors of the Logique of Port Royal, as well as Georg Meier, the author of
Kant’s logic textbook) had taken logic to present “the art of thinking.”10 In
§43 of the third Critique, Kant himself sharply distinguishes “art [Kunst]”
from science: “Art as a skill of human beings is also distinguished from
science (to be able [Können] from to know [Wissen]), as a practical faculty is
distinguished from a theoretical one, as technique is distinguished from
theory (as the art of surveying is distinguished from geometry)” (CJ 5:303;
see also DWL 24:747). By classifying logic as a science rather than an art,
Kant is thereby claiming that logic conveys knowledge (a theory) of thinking,
rather than teaching the practical skill (technique) of how to be able to think.
One can have the art (skill) of thinking (and so be able to think) without
“knowing” thinking in a scientific manner. Logic provides this theoretical
knowledge of thinking itself.
From the science of thinking and to the science

of its contents (concepts)
So far we have been considering the subject matter of logic at a fairly abstract
level, as the understanding or thinking “in general.” And though we have
touched upon the various forms that thinking can take (conceiving, judging,
inferring, systematizing), and have also noted that Kant thinks we can
investigate thinking through two routes – a priori and empirically (in
“pure” and “applied” logic, respectively) – all of the foregoing specifications
9
Again compare Tolley, “Relation between Ontology and Logic in Kant.” See also Hilary Putnam,
“Rethinking Mathematical Necessity,” in Words and Life, ed. James Conant (Cambridge: Harvard
University Press, 1994), 245–63. Note as well that, with respect to the traditional logic, Kant also is at
some distance from those, such as Bolzano and Frege, who take logic to be concerned first and foremost
with the contents of thinking (for Bolzano: “propositions [Sätze] an sich”; for Frege, “thoughts
[Gedanken]”) rather than the acts of thinking or their ultimate objects.
10
For further references to pre-Kantian specifications of logic as an “art,” see Tolley, “Kant’s
Conception of Logic,” 52–70.
172 C. Tolley
of thinking are limited in the following respect: they specify differences only
on what might be thought of as the subject-related side of thinking, or
thinking qua activity of a subject. The difference, for example, between
judging and inferring is a difference in the form of the act a thinking subject
engages in; similarly, the difference between considering thinking “purely”
and considering thinking as it is actually realized in an individual, concrete,
existent subject, and given “empirically” through intuition, is a difference in
the kind of relation that the investigating subject bears to the activity of
thinking.
While Kant accepts that this traditional approach to thinking is valid as far
as it goes, he also argues that we can and must go beyond the tradition by
taking up a new approach to thinking within logic. Kant’s proposal is that
logic should equally consider the object-related side of thinking – that is, the
fact that in each act of thinking our mind becomes representationally
“related” to (“directed” at) some object or other. As Kant sees it, by remain-
ing with a more subject-directed characterization of thinking, the traditional
logic has been treating the understanding “without regard to the difference of
the objects to which it may be directed”; it has done this because it means to
be concerned especially with what is “universal [allgemein]” for thinking as
such – “the absolutely necessary rules of thinking” – what pertains to any “use”
of the understanding, regardless of what kinds of objects the thinking is
about (A52/B76, emphasis added). This is so, even if it was recognized that
we could undertake a study of some “particular [besondere] use” we make of
our understanding and thinking, in which case we would be concerned with
“the rules for correctly thinking about a certain kind of objects” (A52/B76,
emphasis added). As Kant sees it, this latter kind of study would also yield a
“logic,” but one that is associated with specifically “this or that science,”
depending on the species or sub-domain of objects in question; more
specifically, Kant takes this sort of logic to function as an “organon” for
some specific science (A52/B76). (Strikingly, in his lectures, Kant calls
mathematics just this sort of organon [see DWL 24:696; JL 9:13].) It is,
however, only the investigation of what pertains “universally” to all acts of
thinking, regardless of their objects, that has primarily occupied what Kant
calls “elementary logic” (A52/B76).11
At this point, however, Kant raises the possibility of an entirely new kind
of investigation of thinking and the understanding. Whereas the traditional
11
By introducing such divisions with logic, Kant is picking up (and partially reorganizing) various
threads from his predecessors in early modern philosophy of logic; compare Tolley, “Kant’s Conception
of Logic,” 25–29.
logic had either considered thinking as to its standing as an activity, in

abstraction from all of the differences among the possible objects of thought,
or considered the rules for thinking about this or that specific kind of object,
Kant proposes an approach to thinking that somehow lies between these two.
It will be like the traditional “universal” (or “general”) logic, in that it will
not focus on the thinking of some particular kind of object, and so will not
be restricted to the thinking involved in this or that science. Yet unlike the
traditional logic, it will not limit itself to the consideration of thinking as
mental activity that takes certain forms; rather, it will be more object-directed
than this, insofar as it will instead investigate whether there are certain
equally “elementary” representational relations to objects that are themselves
universal across all “uses” of the understanding.
The representational “relation [Beziehung]” that thinking bears to its
object is what Kant calls its “content [Inhalt]” (A58/B83). One way to put
what has been distinctive of the traditional “universal” logic, therefore, is that
it “abstracts . . . from all content” of thinking, “i.e. from any relation of it to
the object, and considers only . . . the form of thinking in general” (A55/B79,
emphasis added). The new science of understanding that Kant proposes is “a
logic in which one did not abstract from all content,” but instead investigated
whatever content might pertain to thinking considered per se – that is, what
content would pertain to the “pure thinking of an object” (A55/B80), by
means of which our understanding is “related to objects a priori” (A57/B82).
Because this content would, in effect, come simply from thinking itself, it
would be content that “cannot be ascribed to the objects” thought about
(A55–56/B80), in the sense that the content does not come to mind due to
the objects themselves being given through our sensibility; the content would
have “neither empirical nor aesthetic origin” (A57/B81), but would instead
be “originally [uranfänglich] given a priori in ourselves” (A56/B80).
This new science of the a priori elementary contents that make possible the
pure thinking of objects is what Kant here calls “transcendental logic”
(A57/B81), in contrast to the approaches of the previous “pure general”
logic, which he now characterizes as having provided a “merely formal logic”
since it “abstracts from all content” and “concerns itself merely with the form
of thinking . . . in general” (A131/B170). Transcendental logic will still count
as a pure logic, though, because it is still a science of the understanding per se:
in it “we isolate the understanding” (from, for example, sensibility), in order
to “elevate from our cognition merely the part of our thought” – namely,
certain contents – “that has its origin solely in the understanding”
(A62/B87).
174 C. Tolley
The first task of this new transcendental logic is thus to demonstrate that
there is such “pure” content present a priori in all acts of thinking whatso-
ever, simply in virtue of their being acts of thinking at all. Kant’s thesis is that
there is, in fact, a set of concepts that have their “origin” in the understanding
itself, and that these concepts correspond (more or less) to those which
Aristotle (and subsequent metaphysicians) had identified as representing
the most fundamental “categories” of objects (see B105). In order to show
that and how such “elementary concepts” (A83/B109)12 could have their
origin in the understanding itself, Kant undertakes the ingenious strategy of
showing how such (transcendental) elementary concepts can be seen as
necessarily coordinate with the most elementary forms of thinking discovered
by the traditional (formal) logic. This is what Kant calls the “metaphysical
deduction” of the categories from the “universal logical functions of think-
ing” (B159).
A key step in Kant’s metaphysical deduction of the pure concepts from the
logical forms of thinking is his argument that we first need to identify a single
form of thinking (act of understanding) as that which in some sense “con-
tains all the rest” (as he puts it in the Prolegomena [4:323]), in order to
provide a “principle” that will explain why all of the forms of acts that logic
had classified as cases of thinking should after all be brought under the single
heading of acts of understanding in general (in its “broad designation”). This
leads to one of Kant’s most influential theses in the philosophy of logic –
namely, that judgment is what plays this unifying role, with the forms of
judging in particular being what can serve as the most elementary delimita-
tion of the activity of understanding: “we can . . . trace [zurückführen] all
actions of the understanding back to judgments, so that the understanding
in general can be represented as a faculty for judging” (A69/B94; see also FS
2:59).13 Kant thinks that concepts themselves, for example, can be under-
stood as essentially “predicates of possible judgments” (A69/B94); in fact,
Kant goes so far as to claim that the understanding “can make no other use of
these concepts than that of judging by means of them” (A68/B93, emphasis
added). Similarly, inferring itself is analyzed by Kant as an act of “judging
mediately” (A330/B386), such that an inference can be understood to be
“nothing but a judgment mediated by [a] subsumption” – that is, a further
judgment (A307/B364). Later Kant is even more emphatic: “the
12
Kant also calls them “basic [Grund-],” “root [Stamm-],” “original [ursprüngliche],” “primitive” con-
cepts (B107–8).
13
For helpful discussion on the points in this section, see Longuenesse, Kant and the Capacity to Judge.
understanding shows its power [Vermögen] solely in judgments” (RP 20:271,

emphasis added).
By taking judging as the basic principle for the classification of the various
forms of acts of understanding in the traditional logic, Kant thinks he is also
in possession of the basic principle for the derivation of fundamental contents
of understanding within his new logic. This is because acts of judging
themselves are acts of representing objects as being a certain way. As we saw
above, a concept itself is the representation (consciousness) “common” to
several representations. And as “predicates of possible judgments,” concepts
“are related to some representation of a still undetermined object” (A69/
B94). More specifically, they are related (by way of the logical form of
judging) to the representations that would function as the subject-term in a
judgment, but then are also thereby related (representationally) to the object
represented by the subject. The very act of unifying or combining represen-
tations in the way that is distinctive of a form of judging is something that at
the same time adds a further kind of representational relation (content) to
the combination of the representations in question. Hence, not only is a
judgment “the representation of a relation between two concepts” (or more
generally: between representations), as “the logicians” of Kant’s day say it is,
but it is more specifically a representation of an “objective unity of given
representations” (B140–42) – or, as Kant puts it elsewhere, “the representa-
tion of a representation of an object” (emphasis added), whether these
representations themselves that are unified in one consciousness are already
concepts or are other sorts of representations, such as intuitions (A68/B93).
More generally, Kant takes a distinctive form of objective representation –
a distinctive “relation to an object” – to arise in each “function of thinking”
qua form of judging. It is here that Kant finds the systematic origin of the
pure concepts or categories, as elementary “contents” that arise “in” the acts
of understanding itself:
The same understanding, therefore, and indeed by means of the very same actions
[Handlungen] through which it brings the logical form of a judgment . . . also
brings a transcendental content into its representations . . . on account of which
they are called pure concepts of the understanding that pertain to [gehen auf]
objects a priori. (A79/B105, emphasis added)
In fact, Kant thinks there will “arise exactly as many pure concepts of the
understanding, which pertain to [gehen auf] objects . . . as there were logical
functions of all possible judgments” (A79/B105, translation modified) – as is
manifest by the parallels between the table of forms of judging and the table
176 C. Tolley
of categories that Kant gives in §§9–10 of the first Critique. As he describes

this process in the Prolegomena §39, to uncover the elementary contents of
the new transcendental logic, Kant thereby needed only to reconsider the
elementary forms (“functions”), which had long been uncovered by the
traditional logic, as forms responsible not just for unifying representations
into a relation, but as themselves “related to objects in general” in virtue of
the representationality of the relevant form of thinking itself:
Here lay before me now, already finished though not yet free of defects, the
work of the logicians, through which I was put in the position to present a
complete table of pure functions of the understanding, which were however
undetermined with respect to every object. . . . I related these functions of
judging to objects in general . . . and there arose pure concepts of the under-
standing. (Pro 4:323–24)
For example, through unifying representations via the categorical form of

judgment, which relates two representations formally as subject and predicate,
there arises a representation of some object (represented by the subject-
representation) as bearing some property (as represented by the predicate-
representation). Kant takes this to show that thinking itself, by means of the
“same act” that unifies the representations into this form of judgment,
thereby represents its object according to the pure concept (category) of
substance, as that in which the relevant property inheres.
Finally, Kant thinks transcendental logic can also show that this sort of
elementary content arises not just in the forms of judging, to yield the pure
concepts of understanding (categories), but also in the forms of inferring that
distinguish the activity of reason, to yield what Kant calls “ideas”: “As in the case
of the understanding, there is in the case of reason a merely formal, i.e., logical
use, where reason abstracts from all content”; but then “a division of reason into
a logical and a transcendental faculty occurs here,” too, as with the under-
standing; hence, “from the analogy with concepts of the understanding, we can
expect both that the logical concept will put in our hands the key to the
transcendental one and that the table of functions of the former will give us the
family tree of the concepts of reason” (A299/B355–56).14 Because the “logical
faculty” of reason is that of “drawing inferences mediately” (A299/B355), Kant
14
Interestingly, there do not seem to be parallel pure contents (concepts) that arise out of forms of acts of
the power of judgment, though it is of course this power that is responsible for generating pure judgments
(“principles [Grundsätze]”) concerning the application of concepts to objects. See section (“From the
Science of Thinking to the Critique of Cognition from Reason”); see also A159/B198.
concludes that, just as “the forms of judgments . . . brought forth categories,” so

too “we can expect that the form of inferences of reason [Vernunftschlüsse] . . .
will contain the origin of special concepts a priori that we may call pure concepts
of reason or transcendental ideas” (A321/B378, translation modified).15
From the science of thinking to the critique

of cognition from reason
With his discovery of the possibility of transcendental logic, Kant
thereby uncovers a distinctive angle of approach within logic to thinking
in general – namely, an approach that looks at thinking neither in
abstraction from all of its content, or all relation that it bears to objects,
nor by focusing only on its relation to some objects, in this or that
particular scientific domain. Rather, transcendental logic looks at the
object-relatedness of thinking “in general,” the distinctive representa-
tional relation to objects that thinking itself “brings into” representa-
tions, thanks to the forms of its own activity.
Kant’s successors were quick to pick up on the novelty of both Kant’s
thesis of the possibility of a universal material or contentful transcendental
logic, and were also heavily influenced by his concomitant reconception of
traditional logic as “merely formal” by comparison; both remained central
features of the specifically “Kantian” tradition within the philosophy of logic
in the nineteenth century.16 Even so, for Kant himself, this recarving of the
aspects of thinking (understanding “in general”), in order to better articulate
the subdivisions within logic, was of a more immediate, instrumental use in
his larger project of the first Critique and the critical philosophy more
15
In fact, the ideas arise not directly from the relevant logical forms of unifying representations
(concepts) in individual inferences, but only from the further acts of synthesizing all inferences of a
specific form in relation to whatever would function as the “unconditioned” that contains the “totality”
of the grounds or conditions for whatever is represented as being conditioned in any given individual
inference (see A322–23/B379–80). In this respect, these contents are perhaps more closely related to the
fourth kind of thinking noted above – namely, that of systematically ordering into a scientific unity.
(Compare the discussion below in the next section.) For our purposes, however, we can bracket the
complications introduced into the parallel metaphysical deduction of the ideas by this further inclusion
of a reference to the whole or “totality” of conditions and the unconditioned. For more on this, see Eric
Watkins, “Kant on the Unconditioned,” unpublished manuscript.
16
For a recounting of some of this history, see Jeremy Heis, “Attempts to Rethink Logic,” in The
Cambridge History of Philosophy in the Nineteenth Century, ed. Allen W. Wood and Songsuk Susan Hahn
(Cambridge: Cambridge University Press, 2012), 95–132.
178 C. Tolley
generally – namely, the project of the critique of reason as a possible source of

a priori “cognition [Erkenntnis]” of objects.
In Kant’s critical philosophy, for cognition of an object, something
more than the mere thinking of an object is required – namely, the
object must be “given” in a separate kind of representation. In the case
of theoretical cognition (as opposed to practical cognition [see Bx]),
objects can only be “given” in representations that Kant calls “intuitions
[Anschauungen].” Intuitions cannot come about through the understand-
ing itself, but come to mind instead from our “sensibility [Sinnlichkeit]”:
as we saw above, “without sensibility no object would be given to us,”
because “the understanding is not capable of intuiting anything” (A51/
B75). Cognition of an object, therefore, cannot arise from mere thinking
alone, but only when an object is also “given” to us in an intuition and
then “thought in relation to that representation” – that is, in relation to
the intuition of the object; “neither concepts without intuition corre-
sponding to them in some way nor intuition without concepts can yield
a cognition” (A50/B74). As Kant puts matters elsewhere:
To think of an object and to cognize an object are thus not the same. For two
components belong to cognition: first, the concept, through which an object is
thought at all (the category), and second, the intuition, through which it is
given; for if an intuition corresponding to the concept could not be given at all,
then it would be a thought as far as its form is concerned, but without any
object, and by its means no cognition of anything at all would be possible,
since, as far as I would know, nothing would be given nor could be given to
which my thought could be applied. (B146)
As Kant notes in the B-edition preface, the domain of possible thoughts

therefore ranges much wider than the domain of possible cognitions:
To cognize an object, it is required that I be able to prove its possibility (whether

by the testimony of experience from its actuality or a priori through reason). But
I can think whatever I like, as long as I do not contradict myself, i.e., as long as
my concept is a possible thought, even if I cannot give any assurance whether or
not there is a corresponding object somewhere within the sum total of all
possibilities. (Bxxvi note)
As Kant goes on to say in this footnote, cognition can only be of objects of

which there is a possible concept and which are themselves “really possible”;
thought, by contrast, can be of any objects whatsoever, whether really
possible or not, just so long as the concept of such an object is “logically

possible” (Bxxvi note).17
For Kant’s overarching purposes of the “critique” of pure reason, it is
crucial that Kant means for reason to be considered, not as to its (merely
formal-logical) standing as the capacity for a certain kind of thinking (namely
inferring), nor even as to its standing as the source of certain concepts (the
pure transcendental-logical concepts [ideas] of reason), but rather as “the
faculty that provides the principles [Prinzipien] of cognition a priori,” with
“pure reason” as “that which contains the principles for cognizing something
absolutely a priori” (A11/B24, emphasis added). Crucially, then, for this sort
of investigation, what we have seen described above as the first task of
transcendental logic – namely, the systematic identification of certain pure
contents of thinking that have their origin entirely in the understanding in
general (including reason) and arise wholly out of acts of thinking, or what
Kant calls the “metaphysical deduction” of pure concepts – can function only
as a necessary but insufficient step for the critique of the possibility that
reason is a source of a priori cognition.
This sort of analysis of purely intellectual content is necessary because, as
we have just seen, all acts of cognizing include acts of thinking, in addition to
intuiting. Consequently, all content of cognition includes transcendental-
logical content (“the category,” in addition to the content supplied from
intuition). For this reason, Kant takes the presentation of the system of the
pure concepts constitutive of pure a priori thinking to provide the systematic
framework for the analysis of the possibility of pure a priori cognition.
Yet as long as the pure content in question remains purely intellectual
(that is, having its source purely in acts of the understanding), this framework
of pure concepts can only yield an analysis of pure thinking of objects and
cannot construct (on its own) any pure cognition of objects. For the latter,
we would need to demonstrate that there is or can be intuitions “correspond-
ing” to these pure concepts, so as to be able to “give” the objects of these
concepts to mind. But then, just as thinking by itself (whether conceiving,
judging, inferring, or systematically ordering) is not sufficient for cognizing,
the estimation of the possibility of pure cognition from the understanding in
general, and reason in particular, cannot come from merely logical analysis,
understood as either formal-logical analysis of the forms of thinking or the
17
For more on the distinction between thinking and cognizing, and the conditions for cognition, see
Clinton Tolley, “The Generality of Kant’s Transcendental Logic,” Journal of the History of Philosophy 50,
no. 3 (July 2012): 417–46; and Eric Watkins and Marcus Willaschek, “Kant’s Account of Cognition,”
Journal of the History of Philosophy 55, no. 1 (Jan. 2017): 83–112.
180 C. Tolley
transcendental-logical analysis of pure concepts (contents) of thinking con-

sidered per se, as they “arise” in acts of understanding alone. Something
more, therefore, is required for the critique of reason as a source of pure
cognition – namely, information about our sensibility and its intuitions, and
an estimation of the possibility of establishing a priori a relation between the
contents (ideas) of reason and those of sensibility.
As a preliminary to this estimation concerning reason, the Critique first
synthesizes the findings of the Transcendental Logic’s metaphysical deduction
of the pure concepts of understanding with the findings of the Transcendental
Aesthetic. The resulting “transcendental” deduction provides Kant with a basic
model for showing “how subjective conditions of thinking should have objective
validity, i.e., yield conditions of the possibility of all cognition of objects”
(A89–90/B122, emphasis added).18 While we cannot hope to go into the
details of this deduction here, what is worth noting is that, in the B-edition
especially, we can see Kant beginning with an analysis of the conditions for
thinking – more specifically: the conditions for “accompanying” certain repre-
sentations with the “I think,” or becoming “conscious” of these representations
in a “unity” (see B-deduction §§15–20 [B129–43]) – and then moving to the
conditions for cognizing objects through the unified consciousness of such
representations (see B-deduction §§21–26 [B144–65]). Kant himself draws
attention to this shift in focus, from the pure concepts as conditions of thinking
to their function as conditions of cognizing, both at the key transitional sections
(§§21–22) and then again at the outset of the concluding summary of the
deduction (§27).19
18
Here we can see, in Kant’s reconception of logic, a twofold response to Hume’s worries concerning
what Kant is identifying as the pure concepts. On the one hand, with the metaphysical deduction of
such concepts out of the traditional-logical forms of thinking, Kant means to demonstrate, against
Hume, that concepts like that of substance-inherence and cause-effect in fact have a “purely logical” or
intellectual origin, rather than an empirical or aesthetic one, or an origin as “a bastard of the imagina-
tion”; that is, Kant means to demonstrate, to the contrary, that our understanding (and reason) on its
own – independently of experience, imagination, or sensibility – does have “the capacity to think such
connections in general” (Pro 4:257–58, translation modified). On the other hand, Kant nevertheless
agrees with Hume’s related worry that the mere fact of our possession of such concepts does not on its
own demonstrate either the existence of any actual objects that correspond to such concepts or that we
have the capacity to cognize these objects. That is, Kant accepts that, beyond the first response to Hume’s
challenge concerning the pure concepts (the metaphysical deduction), a second response is necessary,
concerning the question of the role of such concepts in our claims to cognition of objects: with what
right (quid juris) do we take there to be objects corresponding to these concepts, and with what right do
we claim to be able to cognize these objects? And while the first response to Hume can be given within
logic alone, the second requires appeal to the Aesthetic.
19
This arguably provides the proper template for understanding the difference between the two “steps”
of the B-deduction that Dieter Henrich brought into focus, though Henrich himself does not char-
acterize the significance of the transition in the way I am doing here (i.e., according to the distinction
After demonstrating, in general, that the pure contents of thinking sup-

plied by our understanding are also conditions that make cognizing really
possible, Kant then turns to the task of specifying how thinking can have a
universal and necessary relation to all possible objects of intuition. This takes
two stages: first, in the Schematism, Kant identifies “schemata” or “determi-
nations” of sensible patterns that can be found in every possible sensible
intuition and that are thereby fit to stand as “mediating” correlates between
the pure concept’s purely intellectual content and the indeterminate, infinite
manifold that is given in intuition itself (see A138–41/B177–80). Because
Kant thinks that time is both an a priori sensible content and also what
provides the form to the “one totality in which all of our representations are
contained” (namely, our own “inner sense” [A155/B194]), the requisite
schemata can be given in terms of temporal patterns (“time-determinations”)
that would correlate with the pure concepts. For example, the schema or
determination in sensible intuition for the pure concept of substance is “the
representation of the real as a substratum of empirical time-determination in
general, which therefore endures while everything else changes”
(A144/B183). Second, Kant provides judgments or “basic propositions
[Grundsätze, principles in this sense]” which “contain in themselves the
grounds [Gründe] of other judgments” concerning objects – namely, “all
cognition of its object,” for example, all cognition of substance
(A148–49/B188). In the case of substance, this basic proposition is: “In all
change of appearances substance persists, and its quantum is neither
increased nor diminished in nature” (B224).
While this synthesis in relation to the pure concepts (categories) of our
understanding gives an important model for how pure thinking could be
transformed into a priori cognition, it is not yet sufficient for a critique of
reason in particular, as to how its own pure thinking might serve as a possible
source for a priori cognition. For this, Kant needs to determine if and how
the pure concepts (ideas) of reason (concepts of the immortal soul, the world-
whole, and God) can also be shown a priori to have the requisite relation to
the objects of intuition. Kant’s main conclusion here is famously negative:
“no objective deduction of these transcendental ideas is really possible, such as
we could provide for the categories” (A336/B393; see also A663/B691).
Nevertheless, Kant thinks that the “principles [Grundsätze]” that reason
arrives at, on the basis of attempting to relate these ideas to intuition, can
between thinking and cognizing). See Dieter Henrich, “The Proof-Structure of Kant’s Transcendental
Deduction,” Review of Metaphysics 22, no. 4 (June 1969): 640–59.
182 C. Tolley
in fact be shown to have “objective but indeterminate validity,” insofar as

they “serve as a rule of possible experience,” as a “heuristic principle” for the
“elaborating [Bearbeitung]” of experience (A663/B691), so as “to preserve the
greatest systematic unity in the empirical use of our reason” (A670/B698).
Reason is therefore shown to be a source of “necessary maxims” that serve
“not as constitutive principles for the extension of our cognition to more
objects than experience can give, but as regulative principles for the systema-
tic unity of the manifold of empirical cognition in general” (A671/B699).
That is, this “systematic unity or the unity of reason” serves as a “logical
principle” that is “subjectively and logically necessary, as method” for the
application of reason to the objects of intuition (as cognized in experience),
rather than a “transcendental principle of reason” which would somehow
demonstrate that things “are in themselves determined to systematic unity”
(A647–48/B675–76).
***
Now, for the broader critique of the possibility of a priori cognition, such
an inclusion of further material beyond what can “arise” in the understand-
ing or reason alone is surely necessary, given Kant’s understanding of the
conditions of cognition itself. What is less clear, however, is whether these
steps beyond the metaphysical deduction are themselves ultimately best
thought of as investigations that lie within logic strictly speaking, rather
than in some other kind of discipline, such as critique. For it would seem
that, in each of these further steps (the Transcendental Deduction, the
Schematism, the Principles, the Dialectic), Kant is clearly drawing upon
material from the Aesthetic, concerning sensibility, and so is going beyond
the findings of the science of understanding per se, studied in isolation from
all other capacities. (This is so, even though all of these sections are of course
officially contained under the heading of Transcendental Logic in the first
Critique.)
In any case, this line of questioning also leads us quite close to another
topic associated with Kant’s philosophy of logic, with respect to which the
significance of the thinking/cognizing distinction promises to help clarify
matters. This is the question of how best to understand the significance of
logic (formal and transcendental) for Kant’s distinction between analytic and
synthetic judgments. The first thing to note here is that the difference between
analytic and synthetic judgments is a distinction based on the content of
judgments and not their form (in Kant’s sense):
Judgments may have any origin whatsoever, or be constituted in whatever

manner according to their logical form, and yet there is nonetheless a distinction
between them according to their content, by dint of which they are either merely
explicative and add nothing to the content of the cognition, or ampliative and
augment the given cognition; the first may be called analytic judgments, the
second synthetic. (Pro 4:266, initial emphases added)
This, however, implies that traditional (pure general, “merely formal”) logic
does not know of this difference, because it abstracts from the content of
thinking (even that of pure concepts) “in general” (see A79/B105; compare
A154/B193). Hence, the distinction between analytic and synthetic judg-
ments is actually not one that can be made within formal logic, but only in
transcendental logic (see OD 8:242).
This should be kept in mind when considering Kant’s discussion of the
relation between analytic judgments and the “principle [Satz]” of contra-
diction. As Kant sees it, this principle governs all judgments “in general,”
whether mere thoughts or cognitions, whether synthetic or analytic, and it “is
valid irrespective of their content [unangesehen ihres Inhalt gilt],” and “says
that contradiction entirely annihilates and cancels them” (A151/B190, trans-
lation modified; see also OD 8:195). When understood in this way, the
principle “belongs merely to logic,” by which he means the traditional
(“merely formal”) logic; yet as Kant goes on to note, this principle can also
be put to a more specific use outside of (formal) logic – namely, a “positive
use,” to “cognize sufficiently” the truth of specifically analytic judgments
(A151/B190). Indeed, Kant calls this principle the sufficient “principle
[Prinzipium] of all analytic cognition” (A151/B190). With this, however,
the focus has moved beyond the merely necessary conditions for thinking in
general, and on to the conditions for a specific sort of cognition in particular
(A151/B190–91).
Finally, though it is not uncommon to find claims to the effect that, for
Kant, logic itself (presumably formal logic) “is analytic,” whereas for exam-
ple, mathematics and metaphysics “are synthetic,”20 it is not exactly clear
what this could mean. As we have already seen, if it states truths about
anything, formal logic states truths about thinking itself, its forms, and the
laws that govern the activity of thinking. (As Kant’s lectures have it, logic is
the “self-cognition of the understanding” [JL 9:14, emphasis added].) Yet
there does not seem to be any reason to think that these judgments (about
the understanding, about thinking) will (let alone must) have contents that
take the form of an analytic judgment in particular, such that with the
20
Anderson, Poverty of Conceptual Truth, 103; also 31.
184 C. Tolley
content of their predicate concepts is “already thought in” the content of

their subject concepts (B11). Nor (at least to my knowledge) does Kant
himself ever state explicitly that the truths contained within logic are analytic
judgments.
The role of logic across Kant’s philosophical

architectonic
What has come to light in the foregoing is the following basic threefold
progression in how logic functions within the broader critical philosophy:
1) first, there is traditional logic, which provides the specification of the basic
forms of thinking, in abstraction from all of the content of thinking (its
relation to objects);
2) second, there is transcendental logic, which provides the specification of
those basic pure contents (concepts, categories, ideas) of thinking which
arise from acts of the understanding (and reason) itself, in abstraction
from its relation to sensibility, that is, purely intellectual content (so: the
pure concepts as “unschematized” [see OD 8:223–24; RP 20:272]); and
3) third, there is the critical investigation of the understanding in general, and
reason in particular, as a capacity not just for thinking but for cognizing
objects a priori, which (given Kant’s account of cognition) necessarily brings
into consideration information that lies outside of the understanding itself,
information pertaining to sensibility and its representations (intuitions, their
forms), as well as the possibility of representations (like schemata) that
mediate between thinking and intuiting.
Concerning 1): We have already touched upon the fact that, so far as the
traditional “merely formal” logic is concerned, the acts of understanding
under investigation range over much more than acts of cognizing. As
Kant describes it in §12 of the B-deduction, what is required to count as
an act of understanding is simply what he there calls a kind of “quali-
tative unity,” or “that under which the unity of the grasping-together
[Zusammenfassung] of the manifold . . . is thought,” a unity which is
present not just in a cognition but is also manifest in “the unity of the
theme in a play, a speech, or a fable” (B114, translation modified). Kant
here also calls the unity in question simply the “unity of the concept,”
which recalls our earlier discussion of thinking itself (and conceiving) as
occurring wherever there is a unifying of representations together “in one

consciousness.”
What we should now also note, in relation to 2), is that something similar
can be said even of Kant’s new transcendental logic, at least in its strict sense,
since Kant also allows for our understanding “in general” to be used in ways
distinct from theoretical cognition altogether. Perhaps most importantly, the
(“unschematized”) pure concepts, and in particular, the pure concepts (ideas)
of reason, can be used to form thoughts (judgments) about objects of which
we can have no cognition, but about which Kant thinks our reason gives us
grounds to hold certain judgments to be true. Perhaps the primary instances
of this use of the pure concepts is found in the formation of the theoretical
judgments that God exists and that our own soul is immortal. For both of
these judgments, Kant thinks that we have rational (if practical) grounds to
hold them to be true, even while both of the relevant objects are such as to lie
beyond the sphere of objects of possible (theoretical) cognition (see CPrR
5:120–21; CJ 5:467–68).
The same, it seems, must be said about certain more speculative judgments
articulated in the course of the first Critique itself, concerning the existence of
“things in themselves,” “noumena,” the “grounds” of appearances, and so on.
It has been common (since the time of Kant’s first readers) to criticize Kant for
a kind of inconsistency here, insofar as he at once rejects the idea that we can
have cognition of any substances or causes outside of the possibility of an
intuition of them, while also seeming to insist on (or at least assume) the truth
of judgments involving pure concepts like that of substance and or cause but
that are about just such non-sensible objects – for example, judgments con-
cerning some kind of causal interaction between the things that serve as the
grounds of appearances and our own sensibility.21 Even so, Kant himself is
quite explicit that he is only assuming that we can think of such objects (and
can also “assume” that they exist), not that we can cognize them. Compare
what Kant writes in the B-preface about the objects that are responsible for
appearances: “even if we cannot cognize these same objects as things in
themselves, we at least must be able to think them as things in themselves.
For otherwise there would follow the absurd proposition that there is an
appearance without anything that appears” (Bxxvi; see also B312).22
21
For a discussion of this sort of criticism (with references to various historical instances of it), along
with a defense of Kant against this sort of charge, see especially Karl Ameriks, Interpreting Kant’s
Critiques (Oxford: Clarendon, 2003).
22
Compare as well Kant’s remarks at the end of the Prolegomena about reason’s need to “assume” and
“think” of the existence of certain intelligible beings, in order to make sense of appearances (Pro 4:355).
186 C. Tolley
A further, final, use of the (again “unschematized”) pure concepts

worth noting is one that occurs outside of theoretical cognizing in
particular, one that makes possible specifically practical cognition. This
sort of cognition, too, includes a specific kind of “relat[ion] to its object”
(that is, content) – namely, that of “making [machen] the object actual”
(Bix–x). The question thus arises as to whether practical cognition, like
theoretical cognition, involves certain purely intellectual contents. In the
second Critique, Kant attempts to show not just that there is such pure
content (practical “categories”) but that the “categories” involved in
practical cognition are in fact “without exception, modi of a single
category [of understanding], namely that of causality”; in other words,
whatever further sort of content practical cognition will include, at the
very least it will include thinkable content: “the determinations of a
practical reason can take place . . . conformably with the categories of
the understanding” (CPrR 5:65). To be sure, as in the theoretical case,
it is only with some further “determination” of the purely intellectual
content of the pure concepts (here: an application to desires) that our
understanding in general (as reason) is finally able to practically cognize
its objects – so that these pure concepts “become [practical] cognitions”
of objects – rather than just “think” of them (CPrR 5:66).23 Yet it is
only because the original pure concepts of understanding themselves do
not contain “in themselves” any specifically sensible-intuitive (spatial,
temporal) determinations that they can also find application to sensi-
ble-inclinational determinations as well, and thereby figure in the con-
tents of both theoretical and practical cognitions. What is more, as the
Table in the second Critique makes clear (CPrR 5:66), here again, Kant
takes the transcendental-logical categories to provide a key elementary
framework for the whole system of practical reason.24
Now, the purely intellectual standing of the subject matter of logic is also
what enables Kant to use the framework of the traditional-logical forms
(“functions”) and the transcendental-logical categories even in his investiga-
tion in the third Critique of the “reflective” use of understanding in the
23
For more on the categories of freedom, compare Susanne Bobzien, “Die Kategorien der Freiheit bei
Kant,” in Kant: Analysen-Probleme-Kritik, ed. Hariolf Oberer and Gerhard Seel (Würzburg:
Könighausen & Neumann, 1988), 193–220; and Ralf M. Bader, “Kant and the Categories of
Freedom,” British Journal for the History of Philosophy 17, no. 4 (Sept. 2009): 799–820.
24
Compare the remarks from Kant’s lectures on logic, where it is explicitly allowed that formal logic
“can have to do with practical cognition” as well as “speculative cognition,” since “nothing belongs to
logic except the logical form of all cognitions, i.e., the form of thought, without regard to the content”
and “practical cognition is distinct from speculative cognition as to content” (VL 24:903).
aesthetic exercise of the power of judgment.25 In fact, given his conception of

the generality of logic, Kant is committed to saying that, in any domain
where our activity of thinking and understanding can get a grip (wherever
there is intelligibility), this activity will take the forms disclosed by traditional
logic, and the domain will be represented (in part) through the pure intel-
lectual content disclosed by transcendental logic (as the investigation of our
understanding “in isolation”). As we noted above, the former are universally
and necessarily constitutive of what it is to be an exercise of the under-
standing in general in the first place, if any “use” of the understanding is to
“take place” at all (A52/B76).26 The latter articulate what it is to be “an
object in general,” regardless of what specific kind, regardless even of whether
(really) possible or not, “whether it is something or nothing” (A290/B346),
whether it is already given or to be “made actual.” This sort of generality
allows logic itself – formal and transcendental, taken together – to make a
crucial and quite significant contribution to the underlying unity and sys-
tematicity of Kant’s critical philosophy as a whole.27
25
Concerning the judgments of the beautiful, compare §1: “In seeking the moments to which this
power of judgment attends in its reflection, I have been guided by the logical functions for judging (for a
relation to the understanding is always contained even in the judgment of taste)” (CJ 5:203n).
Concerning judgments of the sublime, compare §24 (CJ 5:247), in which Kant deploys the distinction
between mathematical and dynamical categories from the first Critique (see §11 [B109–13]). (The
connections between either the logical forms or categories and the dimensions of teleological judgment
are much less explicit.)
26
As Kant anticipates in §11 of the B-edition, the table of categories not only “completely contains all
the elementary concepts of the understanding,” but it also contains “even the form of a system of them
in the human understanding” (B109–10).
27
I would like to thank Eric Watkins and Samantha Matherne for helpful discussion of earlier versions
of this material.
Part IV
Relation between
Theoretical and Practical Reason
9
The Primacy of Practical Reason
Ralph C. S. Walker
In the Critique of Practical Reason, Kant has a short section on the primacy of
practical reason over speculative reason (or theoretical reason, as he also calls
it: I shall use these terms interchangeably). Its primacy consists in its ability
to establish firm conclusions – or at least one firm conclusion – about how
things really are. Theoretical reason cannot do that. It can yield analytic
truths, which, however, tell us nothing about reality; it can yield truths about
the spatiotemporal world of our possible experience; it can also demand of us
that we should think in ways that go beyond the realm of possible experience,
guided by regulative principles. That is as far as theoretical reason can take
us. Beyond that it can only speculate idly.
In this same section he says that practical reason and speculative
reason are ultimately the same: “it is . . . only one and the same reason
which, whether from a theoretical or a practical perspective, judges
according to a priori principles” (CPrR 5:121). I shall return to this
briefly, but I am more concerned here with the consequence he draws:
that because they are ultimately the same, they cannot contradict one
another. They have different perspectives and different “interests.” The
interest of reason in its speculative use is in “cognition of the object up to
the highest a priori principles”; the interest of reason in its practical use
is in the determination of the will. Practical reason has primacy because
R. C. S. Walker (*)
Department of Philosophy, Magdalen College, Oxford, UK
e-mail: ralph.walker@magd.ox.ac.uk

192 R. C. S. Walker
it “has of itself original a priori principles with which certain theoretical

positions are inseparably connected, while these are withdrawn from any
possible insight of speculative reason” – freedom of the will, immortality
of the soul, and the existence of God being what he has in mind.
Speculative reason “must accept these propositions . . . although they are
transcendent for it,” so that there is a “union of pure speculative with
pure practical reason in one cognition [zu einem Erkenntnisse]” (CPrR
5:120–21).
The reality of freedom

People have often doubted that anything of this kind can be established
by practical reason, and doubted too whether Kant can seriously have
meant that it can be. He is saying that practical reason can deliver
“cognitions” which we must “accept.” In the first Critique, and in
other places as well, he limits “cognition [Erkenntnis, Wissenschaft]” to
the field of possible experience, apparently implying that other cognitive
states are substandard. He must be using “cognition” in a different way
here, and indeed he says as much: it is “cognition from a practical point
of view” (Erkenntnis in praktischer Absicht [CPrR 5:133], in praktischer
Beziehung [CPrR 5:137], in praktischer Rücksicht [TPP 8:418]). The
question is what this amounts to.
Actually there are two questions, only one of which I shall discuss. For
the three postulates – of God, freedom, and immortality – are not all on
the same footing. The one on which the others depend, and on which
Kant places greatest weight, is that of freedom: absolute freedom of the
will (Willkür), wholly independent of the determinism of “alien causes”
(G 4:446). Exactly what he thinks about the other two is a fascinating
question, and he clearly found it so himself, for both in his published
writings and in his notes his view seems to vacillate between holding that
to take morality seriously “I should act as if there are a God and a future
life” (NF 18:678 [R6351]) and holding that one is entitled to a secure
confidence about these things. However, the case for freedom is different.
The case for the primacy of practical reason is established if there is one
thesis that it can establish decisively, and which theoretical reason
cannot. The thesis is itself of course theoretical, as Kant says (CPrR
5:120), in the sense that it makes a factual claim, though it is grounded
in practical reason and no speculative proof could ever establish that
9 Primacy of Practical Reason 193
claim.1 It is the thesis that the will is free.2 Here, and here only, we
achieve knowledge about the ultimate reality that underlies the world of
our possible experience, the spatiotemporal world of appearances that is
the object of our everyday cognition and our science. Such cognition as
we may have of God and immortality – if “cognition” is the right word
here – is secondary to our knowledge of freedom, and derivative from it.3
It is easy to feel that this cannot be right, and that despite this passage we
should interpret him as saying that theoretical reason, and only theoretical
reason, gives us truth; and that practical reason is concerned with action, so
that it can at most ask us to believe in certain things, including the reality of
freedom, while secretly knowing that we have no grounds for that. For it seems
pretty clear how – according to Kant – theoretical reason works to give us truths
about the world. The Critique of Pure Reason shows how the rational, active
faculty of the mind4 orders the data of sense through synthesis and through the
deployment of the categories to provide us with knowledge of the spatiotemporal
world, a kind of cognition limited only by the bounds of possible experience.
Given that we can find the truth about the world by these means – the truth
about the only world we can know, thus setting aside the unknowable things in
themselves – it would seem clear that we can never be epistemologically justified
in making truth-claims that go beyond the limits of possible experience. Morality
may make it desirable for us to believe in freedom, but since freedom is not an
object of possible experience it can do no more than that.
This is how it would be if theoretical, not practical, reason were primary
(as he says, CPrR 5:20). But it is not. In holding that practical reason is
primary instead, he is indeed claiming that it provides us with a kind of
1
Marcus Willaschek points out quite rightly that we often do think it entirely reasonable to accept
factual claims on practical grounds, without being able to justify them theoretically: e.g., that I am not a
brain in a vat (“The Primacy of Practical Reason and the Idea of a Practical Postulate,” in Kant’s
“Critique of Practical Reason”: A Critical Guide, ed. Andrews Reath and Jens Timmermann [Cambridge:
Cambridge University Press, 2010], 168–96). This is a helpful suggestion so far as the beliefs in God and
immortality are concerned, but in Kant’s view the Fact of Pure Reason directly proves the reality of
freedom (CPrR 5:47).
2
Sebastian Gardiner argues that this cannot be what Kant has in mind in the section on the Primacy of
Practical Reason, because the reality of freedom had been established earlier in the book (“The Primacy
of Practical Reason,” in A Companion to Kant, ed. Graham Bird [Oxford: Blackwell, 2006], 262–64). I
cannot see that this follows; Kant is pointing out its special status, and its importance as the basis for the
inferences to God and immortality.
3
Kant is much less confident about the claims for God and immortality than about the claim for
freedom. Often he seems to think that what morality requires is just that we must believe that there is a
God, and that the soul is immortal, if we are to take morality seriously. That would not do for freedom.
4
This includes the understanding and the productive imagination, but when Kant is using the term
“theoretical reason” broadly he includes these under it.
194 R. C. S. Walker
knowledge that is of a higher order than anything that theoretical reason can
give us: because it is knowledge of ultimate reality, not just of the world of
appearances.
The fundamental principle of practical reason is the categorical imperative:
the moral law that binds all rational beings. He considers this to be too
evident to need any justification beyond itself, and too basic to permit any
such justification. It is “given, as it were, as a fact of pure reason of which we
are a priori conscious and which is apodictically certain” (CPrR 5:47). He
would say the same about the fundamental principles of theoretical reason,
like the principle of contradiction; he says, “We can become aware of pure
practical laws just as we are aware of pure theoretical principles, by attending
to the necessity with which reason prescribes them to us and to the setting
aside of all empirical conditions to which reason directs us” (CPrR 5:30).
There is also a further parallel between practical and theoretical reason.
Among other things, theoretical reason gives us conclusions about the con-
ditions required for possible experience; among other things, practical reason
is concerned to give us conclusions about the conditions required for the
moral law to be binding upon us. So practical reason does not just tell us
what to do. It can deliver factual conclusions based on a premise that Kant
considers undeniable. To question whether experience is possible would be
silly; in Kant’s eyes it would be equally silly to question the bindingness of
the moral law. These starting-points have to be taken as given.
Kant holds that our awareness of the moral law enables us to establish the
reality of freedom, which is something that theoretical reason could never do.
In our awareness of the moral imperative we can see that we ought to do
things we are not inclined to do, and that we ought to do them whatever
background influences may be acting on us. But we could not be morally
bound to do something unless we were free to do it. Freedom of the will is
the ratio essendi of the moral law: there could not be a moral law binding us
unless we were free to choose whether to follow it or not follow it. So
practical reason forces upon us the conclusion that we are free, whereas
theoretical reason can do no more than leave the question of freedom
open. He takes himself to have established in the Critique of Pure Reason
that theoretical reason can only recognize it as a possibility, because it is
incapable of deciding the matter. So practical reason can at least take us this
much further than theoretical reason can. And if Kant is right that freedom
requires the immortality of the soul and the existence of God, then it can
take us a remarkably long way.
He does not regard our assurance of freedom as being in any way second-rate.
He says it constitutes cognition, Erkenntnis, although the source of this
cognition lies in the moral law (TPP 8:419); in a note he calls it morally certain,
and adds, “Moral certainty is the greatest certainty” (NF 16:482 [R2716]). In
the Critique of the Power of Judgment he says that freedom “is the only one
among all the ideas of pure reason whose object is a fact and which must be
counted among the scibilia [knowables]” (CJ 5:468). The contrary impression
may be given by his saying that it is cognition “from a practical point of view,”
but this only means that it has its source in the moral law. Sometimes he says that
it is “subjective” in a way in which our cognition of the world of appearances is
“objective,” but this again is only intended to reflect its source in the subject’s
awareness of the moral law, in contrast with the cognition of objects in the world
around us that the principles of theoretical reason allow us to have.5 Again, when
he says that his moral proof of God’s existence is kat’anthrôpon rather than
kat’alêtheian – in relation to man, rather than in relation to truth – his point
must be the same: it is grounded in our awareness of the moral law. It is the only
satisfactory proof, for no theoretical proof – no proof “in relation to truth” – can
get further than encouraging us to think of nature as having some intelligent
cause. Theoretical reason can deal with that “truth” that belongs within the
world of appearances, but beyond that it can only provide regulative ideas, ways
in which we should think but which cannot tell us how things are. Thus,
theoretical reason can tell us to use the conception of an intelligent cause as a
regulative idea for the purposes of science, and it can go no further than that.
Practical reason can (CJ 5:463; cf. A739/B767; RP 20:306).
So, in Kant’s eyes at least, our cognition of freedom is secure. He takes the
moral law to be given as an objective imperative, very much in the way that
the laws of logic are given. Terminology here can be confusing: our awareness
of it is subjective, being ours, but the law itself is objective in the sense of
being independent of what anyone thinks or feels about it. It must be,
because – like the laws of logic – it is binding on all rational beings. It is
therefore independently real, and real in its own right. Given Kant’s trans-
cendental idealism, this means that it has a reality that the spatiotemporal
world lacks, since what is true in the world of appearances depends on
human cognitive capacities. In saying that it is binding on all rational beings,
Kant means that it motivates all rational beings. It does not motivate them so
5
It is true that in the first Critique he says that its grounds are “only subjectively sufficient,” and
therefore calls it a kind of belief (Glaube) rather than knowledge (Wissenschaft) (A822/B850), though
one that is unshakable (A828/B856). But what he must mean here is again that it derives from one’s own
awareness of the moral law, so that he could not communicate it to others who were “entirely indifferent
in regard to moral questions” (A829/B857). Elsewhere he calls it “rational faith [Vernunftglauben],” and
says it is “not inferior in degree to knowing [Wissen], even though it is completely different from it in
kind” because it is grounded in the moral law (OT 8:141).
196 R. C. S. Walker
strongly as to compel them to act, for that would leave no room for freedom,
but its motivating force is always there, leaving them the choice whether to
act from the moral motive or from some alternative inclination (Rel 6:41).
This is often thought to be an untenable position. But the idea that
moral judgments have motivational force is one that many contemporary
philosophers share, even though few would go so far as to claim that they
motivate all rational beings.6 It is true that many of those who accept that
they have motivational force would go on to deny that they are objective, in
the sense of being independent of what anyone thinks or feels about them.
John Mackie dismisses the idea that they could be objective in that sense by
saying that they would have to be “entities or qualities or relations of a very
strange sort, utterly different from anything else in the universe.”7 But that
is not an argument. And they would not be unique in the universe. Logical
laws must have a very similar status: independent of anything we think or
feel about them, and yet motivating us, not indeed to act in a particular
way, but to draw particular conclusions from given premises. I shall return
to logical laws in due course, but the view that the moral law is an objective
imperative is by no means untenable; indeed, many moral objectivists
consider it true.
As to its entailing freedom, Kant must be right that if there is a law of that
kind, which tells us what we ought to do and thereby motivates us to do it,
then it must at least be possible for us to act on that motivation.8 This will be
true whether or not the moral law turns out to place the specific requirements
on us that Kant thinks it does; and it is of course important that for Kant
himself, and for many moralists who take a similar approach, “acting on” a
motivation does not imply success in achieving its goals; what matters is the
serious attempt (G 4:394).
We must therefore be able to make the choice, in appropriate circum-
stances, to act on that motivation rather than on whatever alternative
motivation we also have. So we must be free, at least in some reasonably
strong sense. Whether it has to be quite in the sense Kant has in mind,
6
Though Kant never makes it quite clear what makes someone a “rational being.” Young children and
the insane would seem to be excluded. Perhaps he might also exclude someone wholly lacking “the
moral predisposition” (An 7:324).
7
J. L. Mackie, Ethics: Inventing Right and Wrong (Harmondsworth: Penguin, 1977), 38.
8
Lawrence Pasternak takes this to rest on the principle that “ought implies can” (“The Development
and Scope of Kantian Belief: The Highest Good, the Practical Postulates and the Fact of Reason,” Kant-
Studien 102, no. 3 [Sept. 2011]: 306). That is not clear. The fact that the law motivates me to do
something seems a powerful reason for thinking that I must be able at least to try to do it (which is all
that Kant needs); that does not have to depend on the general principle that “ought implies can.”
whereby this freedom is absolute and unconditioned by “alien causes”

(G 4:446), is as yet less clear; this is another matter to which we shall return.
Imperatives of theoretical reason

Kant sees theoretical reason as the active faculty by which we synthesize the
given manifold and form judgments about the world. It has three parts:
general logic, transcendental logic, and what he calls “reason” in a narrower
sense – that extension of reason which requires us to think in ways that go
beyond the limits of possible experience, in accordance with regulative
principles that guide us in framing a more and more unitary conception of
nature (though they do not yield us any truths). Kant sees it as a set of rules
for how we ought to think about the world.
We often think of theoretical reason as a means to truth, and understand
Kant as seeking to provide us, in the first Critique, with the principles that
we need to find out what is true: principles that confine truth to the
spatiotemporal world of our experience. This cannot be altogether wrong,
but it may also be misleading. Contemporary philosophy is centrally
concerned with truth. We look for truth-makers, we explain things in
terms of truth-conditions, and we put forward theories about the nature
of truth and other theories about how to attain it. We even explain logic in
terms of truth. Kant is much less interested in it. He even says – admittedly
in an early note – “Truth has no value in itself” (OBSn 23 [Ak 20:175]).
Certainly he has things to say about truth in the first Critique, but truth
does not have the centrality there that we might expect in a modern work.
He does offer a sort of definition of truth, as “the agreement of a cognition
with its object”; but he calls this a nominal definition, meaning that it fails
to reveal the nature of truth, and he says it is absurd to ask for a general
criterion of truth (A58–59/B82–83). His focus is elsewhere: on synthesis,
on judgment, on the conditions of self-conscious experience. Truth in its
strictest sense – truth about the world as it is in itself – is in general
unreachable, and there is no point speculating about what lies beyond
our cognitive capacities.
Where we see theoretical reason as concerned with truth, Kant sees it
as concerned with rules. General logic has to do with “the absolutely
necessary rules of thinking” (A52/B76), transcendental logic with “laws of
the understanding and reason” (A57/B81). Of course, we also see theo-
retical reason as having to do with rules, but we see these rules as good
198 R. C. S. Walker
rules because they yield us the truth. For Kant it is the other way round:
within the spatiotemporal world, the truth is what the rules yield. This is
essential to his understanding of ordinary, everyday cognition, and of
scientific cognition as well. Practical reason gives us rules that tell us what
we ought to do; theoretical reason gives us rules that tell us what we
ought to believe.
In saying that the rules of “general logic” – which we would just call logic –
are “necessary,” he is using that word in the same way as when he calls moral
requirements “necessary”: meaning not that we always do conform to them,
but that we ought to. They determine which forms of inference we may validly
use, and which forms we should reject as invalid. Nothing in this conception
of validity depends on the idea that valid forms of inference are those which
always yield true conclusions if their premises are true. No doubt there is
heuristic value in explaining validity to students by saying that a valid form of
inference is one that always leads from true premises to true conclusions, but
from Kant’s point of view to put it that way is to get things backwards; and on
this at least he must be right. A valid form of inference is one that we are
entitled to employ, and that we ought to accept. If someone argues that all
men are mortal, and A is a man, so A is mortal, no independent knowledge of
the truth of the conclusion is needed to confirm the validity of the inference. It
is because we see the inference to be valid and the premises true that we see we
ought to accept the conclusion as true.
The rules of “general logic” cannot by themselves give us cognition about
the way the world is. For that we need “transcendental logic,” which is again
a system of rules that he claims to derive from the logical structure of
judgments as applied in space and time. These rules are the Principles of
Pure Understanding. He argues that the possibility of experience depends on
these principles (A156/B196). He cannot mean by this that we must believe
them to be true; that is obviously false, since many people have not believed
that every alteration is governed by causal law, and one might well doubt
whether sensations must have an infinitely variable degree of intensity. He
means that they are the rules that we should use for making sense of our
experience, though we need not formulate them to ourselves so long as we
deploy them. His argument about causality is that it is only by using the
causal principle that one can establish that one event happens before another
in the objective time-order – where to establish something is to show that it is
something we ought to accept.
This is not to deny that the Principles of Pure Understanding do give us
truths about the world of appearances. Kant certainly thinks he has shown
that every alteration must have a cause. But they function as rules, without
which “appearances could never amount to cognition of an object corre-

sponding to them” (A159/B198). They are rules that we have to use to gain
knowledge of the world of appearances, but this is not a straightforward
matter of finding things out: the world of appearances is in part the product
of the application of these rules. Because they are rules, we may misuse them,
as when we ascribe a cause too hastily. In that case the judgment we make is
wrong, and we can find out that it is wrong by paying more attention to the
data given in sensibility and applying the rule more carefully. We may also
fail to make use of these principles where we ought to, and thereby miss
something. We can fail to see something as an object because we have given it
no attention, or because we are drugged and see things only as color patches.
In the section on the Postulates of Empirical Thought, what emerges is
that reality within the world of appearances is determined by consonance
with the formal and material conditions of experience (the categories and
forms of intuition on the one hand, what is given in sensation on the other
[A218–22/B265–69]). As he puts it in Reflexion 5642, truth consists merely
in “the thoroughgoing interconnection of representations in accordance with
laws of the understanding” (Ak 18:281). This provides, in effect, a coherence
theory of what constitutes truth within the world of appearances: so we can
ascribe to Kant a coherence theory of truth, if “truth” means that everyday
conception of truth that fits the world as we experience it. Truth is what
coheres with the formal and material conditions of experience.9 From his
point of view, however, this is only a very limited kind of truth. It is what we
can and should regard as the truth for the purposes of everyday life and of
science; but in reality it is not really truth at all, for ultimately truth must be a
matter of correspondence with things as they are in themselves, not with
things as we are led to construe them through the operation of human
cognitive capacities.
Judgment is a form of synthesis, and both judgment and the more
elementary synthesis involved in concept formation are essential to what
makes self-conscious experience possible (A79/B105). Kant holds that we
can only make judgments, or synthesize intuitions together under concepts,
through using the categories. The essential thing about both of them is their
normative character, and he must certainly be right about that, even if he is
wrong about the categories. Judging is inevitably a rule-governed activity;
that the same applies to concept formation is a major insight of Kant’s,
9
I go into this in more detail in Ralph C. S. Walker, The Coherence Theory of Truth: Realism, Anti-
Realism, Idealism (London: Routledge, 1989), 61–82.
200 R. C. S. Walker
further to be developed by Wittgenstein and others. It cannot be just a

matter of associating ideas, as some empiricists held, or there would be no
room for the dimension of right and wrong in the way concepts are formed
or applied; there would be no room to recognize that someone is making a
mistake when she starts applying the concept “green” to blue things, instead
of just associating ideas in an unusual way (B141–42). And all the judging
that we do, and all the synthesis we carry out, is spontaneous, “self-active,” in
the sense that it is we who do it, following rules or perhaps deviating from
them on our own initiative.
The nature of these rules, according to Kant, reflects the requirements of
general and transcendental logic. This applies whether they are rules for
judging or rules for that elementary synthesis which is needed to bring
intuitions under a concept. They are binding on all rational beings,10 in
the sense that they ought to follow them, and because they bind us in this
way we might call them imperatives. Kant does not, because he reserves the
word for the commands of practical reason. But the rules of theoretical
reason are equally rules that bind us, and they bind us categorically. If we
thought they applied to us only when we are seeking the truth, they would be
hypothetical rather than categorical imperatives, but as we saw, that would be
to get things backwards: the truth (about the world of appearances) just is
what we arrive at by following them.
Thus there is at least a similarity between theoretical and practical reason,
as Kant understands them. They are both sources of rules, and rules that are a
priori. That may be all he meant by saying that they were ultimately the
same, though people often think otherwise because he did say in the
Groundwork that he needed “to present the unity of practical with speculative
reason in a common principle,” which he does not seem to have done (G
4:391). On the other hand, his use of words like “principle” can be quite
vague; the principle of contradiction is said to be the “the supreme principle
of all analytic judgments,” though it is clearly impossible to derive them all
from it (A150/B189).11
These are the rules that govern – or should govern – our thinking and our
acting. To say that the factual judgments that we make about the world are
10
Though rational beings with different forms of intuition from ours would of course have to apply
them differently in consequence; and none of this would apply to a being with intellectual intuition.
11
Admittedly, “principle” translates Grundsatz here and Prinzip at G 4:391, but Kant tends to use them
more or less interchangeably. Some of the lines on which he might have sought to achieve a more
substantive unity are developed by Onora O’Neill, Constructions of Reason: Explorations of Kant’s
Practical Philosophy (Cambridge: Cambridge University Press, 1989).
true, or justified, or constitute knowledge of how things are, is to say that we

ought to believe them. In our judgments about the spatiotemporal world, we
do not reach a truth that is objective independently of what anyone thinks or
feels about it, because space and time themselves belong to our human
cognitive capacities. So do the rules for applying the schematized categories
to the data given in intuition. We reach a truth about the world of appear-
ances. This is simply a truth about how we must understand the world, given
human cognitive capacities.
Many physicalists would reject all of this fundamentally, arguing that
the appearance of normativity is nothing more than an illusion, and that
all the processing that goes on is due to the physical activity of the brain.
If we are effectively elaborate computers, producing outputs determined
by an internal program, the account that Kant offers must just be wrong.
Some aspects of it could be rescued, but synthesis could not be self-active
in the way Kant thinks; the rules governing it would be the sort of thing
that could be captured in machine code, and “reason” would amount to
a specific kind of brain activity. If, however, one rejects that account of
things, much of what Kant has to say is very plausible. Perhaps we may
have reservations about the categories, and about his treatment of space
and time, but that our minds have to operate in certain ways to make
sense of the world seems evidently true, and that we have to make use of
the notions of object and cause, in ways that go well beyond the actual
evidence, seems true as well. That we can be aware of the world around
us only by processing mentally the data given to our senses seems equally
undeniable, and this processing must affect not only our grasp of sec-
ondary qualities but that of primary qualities as well and our under-
standing of spatial and temporal relationships.12 Indeed, it is plain
enough that all we can know about the world is its effects on our
sense-organs, and that this is bound to raise a question about what
underlying reality there might be to produce these effects. Whether or
not this be transcendental idealism is perhaps a verbal question; if it is,
the case for transcendental idealism is a very powerful one, though I
cannot defend it here.13
12
Cf. Prolegomena, §13, note 2 (4:289); and Lucy Allais, Manifest Reality: Kant’s Idealism and His
Realism (Oxford: Oxford University Press, 2015), esp. pt. 2.
13
See Rae Langton, Kantian Humility: Our Ignorance of Things in Themselves (Oxford: Oxford
University Press, 1998); and (in a non-Kantian context) Thomas Nagel, Mind and Cosmos: Why the
Materialist Neo-Darwinian Conception of Nature Is Almost Certainly False (Oxford: Oxford University
Press, 2012).
202 R. C. S. Walker
What Kant is claiming for our knowledge of freedom is that this, and
perhaps this alone, is knowledge about reality as it is in itself. The world of
our everyday experience is the world as it appears to us: the world of
appearances. It is the world as we construe it, following the rules and
synthesizing the given manifold in accordance with the categories. In our
awareness of the moral law we go beyond that, and discover something about
the underlying reality. Only here do we touch on genuine truth, as opposed
to the factitious truth generated by the Principles of Pure Understanding:
factitious, though of course quite good enough for everyday Erkenntnis.
Reason and freedom

People often say that Kant had no good ground for claiming that we must be
free in his strong sense. It would be enough if we were free in the sense of
“free” used by those often called compatibilists, for whom freedom consists
in doing what you want to do.14 On their view, that would still allow him to
say that we have a choice between doing what morality requires and follow-
ing our own selfish inclinations. It is an important aspect of Kant’s moral law
that it motivates us; our inclinations also motivate us; we act from duty, then,
when we act from the moral motivation, and when we act from inclination
we do not. The choice (they would say) is as much our choice as anything
can be, because it stems from whichever motivation in us is the strongest.
And that, in turn, is entirely compatible with a determinism that holds every
event, mental or physical, to have a sufficient cause determining it in time.
It is impossible for Kant to hold that.15 He calls it the freedom of the
turnspit or the marionette (CPrR 5:97, 101) and dismisses it as “a wretched
subterfuge” (CPrR 5:96). He has to, because the world of appearances is
governed by a thoroughgoing causal determinism, according to the argument
of the first Critique. But the moral law that motivates us is the law of pure
practical reason, and pure practical reason is neither an object nor an event in
time. It does not belong in the deterministic world of appearances. To be
moved by reason is to make an autonomous choice, which is not subject to
the empirical law of causality. It cannot be subject to any such law, because
14
Adding whatever qualifications to this as may be necessary – for example, to exclude the addict whose
first-order wants conflict with his second-order wants.
15
Kant had once held such a position, but before he developed his transcendental idealism (NE 1:398–
406).
the objective imperative that motivates us is not the sort of thing that can
play a part in a system of causal laws.
This may seem an extreme claim to make. But it is hard to see how it can
be avoided. It would not help to suggest that there are only certain conditions
under which practical reason can motivate us to act, because we should still
have to agree with Kant that under those conditions we do have the kind of
freedom that he insists on: the absolute freedom to choose between acting on
the rational motive and not acting on it. Nor would it help to say that a lively
awareness of what reason demands is a mental state that we have, one which
has its own cause, and that when awareness is sufficiently lively we act on it.
For the point is that, if we are to be moved by practical reason, practical
reason itself must be the source of our awareness of it. Practical reason is no
sort of event or state of affairs; practical reason does not belong in the world
of appearances.
That leaves him in the position of apparently having to say that our acts
both are and are not determined by causal law. They must be, because they
are events in time; they cannot be, because they are free. He evidently thinks
there is no contradiction, since he says that the insights of practical and
speculative reason cannot contradict one another (CPrR 5:120). It is tempt-
ing to take up the idea he himself mentions once or twice, that the require-
ment of the Second Analogy is only that every alteration must have a cause,
not that it have a determining cause:
It is easy to see that if all causality in the world of sense were mere nature, then
every occurrence would be determined in time by another in accord with
necessary laws, and hence . . . the abolition of transcendental freedom would
also simultaneously eliminate all practical freedom. For the latter presupposes
that although something has not happened, it nevertheless ought to have
happened, and its cause in appearance was thus not so determining that
there is not a causality in our power of choice such that, independently of
those natural causes and even opposed to their power and influence, it might
produce something determined in the temporal order in accord with empirical
laws, and hence begin a series of occurrences entirely from itself. (A534/B562;
cf. Ak 18:254–55 [R5613, R5616])16
16
Nevertheless, it is hard to reconcile this with the passages in which Kant is at his most deterministic.
A549–50/B577–78 firmly takes the view that the agent’s choices are determined by her empirical
character, to such an extent as to make all her actions to be “determined in the order of nature”; CPrR
5:99 is equally uncompromising. The agent’s intelligible character is supposed to act freely in timelessly
determining her empirical character, but how this could work remains mysterious.
204 R. C. S. Walker
This is a question that would need extensive treatment. What matters here is
not whether Kant can reconcile the determinism of the world of appearances
with the freedom morality requires, but why he has to reject a compatibilism
of the contemporary kind. It is because he is committed to holding that we
can be moved to act by reason, and he believes that we are all committed to
that. Reason is not a cause. In responding to reason we are responding to a
rational imperative: an imperative that is independent of what anyone thinks
or feels about it, and independent of the desires and inclinations that anyone
may have.
Those who think that morality is not a matter of objective imperatives
need not be forced to any such conclusion. If one does think it is a matter of
objective imperatives, the same considerations will apply, whether one
ascribes them to “practical reason” or not, for an objective imperative cannot
be an event in time. The danger of contradiction could be avoided, of course,
by abandoning the thesis that the world of appearances is governed by
thoroughgoing causal laws. Alternatively, one might seek to argue that
these considerations provide good grounds for rejecting the idea that objec-
tive imperatives, which are not themselves events in time, can motivate us.
Neither of these courses is available to Kant.
But now a doubt may arise as to whether practical reason is entitled to
the primacy that Kant claims for it. For just as the moral law is a
principle that binds us, motivating us to act in a certain way, so the
theoretical principles are also principles that bind us, motivating us to
draw the conclusions that are supported by logic and by the evidence as
interpreted in light of the Principles of Pure Understanding. These
theoretical principles also set norms, though they tell us what we ought
to believe rather than what we ought to do. And it is not only our
judgments that they govern, or ought to govern, but also synthesis,
including the most elementary synthesis that brings conceptual order to
the manifold of intuition. Kant repeatedly says that in synthesis we are
spontaneous, “self-active.” This at least means that synthesis does not
operate in a mechanical way – something that is already implied in
saying that it is a matter of following rules. Moreover, the activity of
synthesis, and the similarly structured activity of judgment-forming,
cannot carry themselves out. There must be a subject of synthesis, and
this subject of synthesis cannot itself be an object of cognition
(Erkenntnis), for (as Kant says) “the subject of the categories cannot, by
thinking them, obtain a concept of itself as an object of the categories;
for in order to think them, it must take its pure self-consciousness . . . as
its ground” (B422). The “‘I’ that thinks” cannot have Erkenntnis about
itself because that would require it to synthesize itself. This is a con-

sequence of Kant’s transcendental idealism; it is also a consequence of
the rather commonsensical considerations put forward above to which
the name of “transcendental idealism” might perhaps equally be applied.
If the world as we can know it is in part the product of the processing
of data by the human mind, as it inevitably must be, then the
mind itself, which is the subject of this processing, cannot at the same
time be merely an object of it. And if all we can know about the world
is through the effects that it produces on us, the same must be true of
ourselves, and true generally of the persons we meet as well as of
anything else in the world as we can know it.
Particularly in the first Critique, Kant is very cautious in what he says
about this, being anxious to distinguish his own position from the para-
logistic position of Descartes, for whom the “I” is a substance that persists
through time. He does, though, agree (in a rather offhanded way) that it
can be called a substance, provided one recognizes that one is here using
“substance” in its unschematized sense, and thus not carrying the implica-
tion of permanence in time or, indeed, of mind/body dualism (A350). But
this self that stands outside space and time – though of course it can only be
aware of itself or its own synthetic activity in temporal terms – is a necessary
presupposition of theoretical reasoning; just as much, indeed, as it is a
necessary presupposition of practical reasoning and practical action. And
though judging and synthesizing may not involve will and choice, their
“spontaneous” rule-governed activity cannot be controlled by the causal
determinism of the world as we can know it any more than the free choice
of the will can be. Indeed, just as the moral law presupposes freedom of
choice (Willkür), because otherwise it would be requiring of us things we
are unable to perform, so the rules governing synthesis require a similar
freedom to make or not to make an appropriate judgment, to bring or fail
to bring an intuition under the appropriate concept. So, after all, does not
theoretical reasoning provide as sound a basis for presupposing freedom as
practical reasoning does? He seems to be suggesting this in the Groundwork,
when he says:
one cannot possibly think of a reason that would consciously receive direction
from any other quarter with respect to its judgments, since the subject would
then attribute the determination of his judgment not to his reason but to an
impulse. Reason must regard itself as the author of its principles independently
of alien influences; consequently, as practical reason or as the will of a rational
being it must be regarded of itself as free. (G 4:448)
206 R. C. S. Walker
This makes it seem as though the same case can be made for autonomy in
judging as in acting, though he is here appealing only to how one must think
of oneself. One might wonder why he never develops the idea.
The primacy of the practical

Kant does insist that theoretical reason can never give us information about
what lies beyond the limits of possible experience. On the other hand, this
apparently rules out many of the arguments of the first Critique, arguments
that explore the conditions of possible experience in terms of spontaneous
synthesis, the “I” that thinks, and so on. He explicitly allows that “the
possibility of experience” can be the third thing that unites subject and
predicate in synthetic a priori judgments (A196/B195, A737/B765), so one
might think that would have given him a way to legitimize such claims as
these. Instead, he says that it only allows cognition a priori of “the mere rule
of the synthesis of that which perception may give a posteriori” (A720/B748),
in other words the Principles of Pure Understanding.
He does have a case for that. The first Critique argues that we can give a
kind of sense to using concepts beyond the limits of possible experience, but
that there is no way in which theoretical reason can actually justify any of the
assertions we might make with them: they are at best “transcendental
hypotheses,” which cannot be proved. Now Kant’s accounts of synthesis
and the “I” that thinks might be plausible exceptions if it could genuinely be
shown that these things must be so if experience is to be possible. But Kant
does not show that – he does not present a proof. He presents a highly
persuasive explanation of how experience can be possible, but he does not
establish that it must be correct – that no other explanation could cover the
ground. Philosophers rarely do manage to achieve proofs, and when they try,
they usually fail, as Kant is very effective in showing.
At the time of the first edition of the Critique of Pure Reason (1781) and of
the Groundwork (1785), Kant may have been rather unclear about how to
regard claims such as these. He can say that we must think of ourselves as
making up our own minds in judgment, and that we must think of ourselves
as subjects spontaneously active in synthesis, but he cannot prove that we are
right to think in these ways. He does not describe any of his own claims as
transcendental hypotheses, though, and his rather disparaging treatment of
transcendental hypotheses in the Critique suggests that he did not recognize
that that was what they were. His lack of clarity about the matter may help to
explain why he is so unwilling to make clear and unambiguous assertions

about the need for the self as subject, and for things in themselves that affect
it. In the second edition he is still annoyingly obscure about these matters,
but at least he does commit himself further; his thought has developed, but
he is still not completely clear about these things.
In the first edition as well as the second he does have a lot to say about
synthesis. He calls it spontaneous, and shows the essential role it has to play in
developing our awareness and our knowledge of the world around us. He also
discusses freedom, and maintains that (despite appearances) the determinism
of the spatiotemporal world still leaves room for freedom in the sense of
absolutely free choice. However, he draws an interesting distinction between
practical freedom and transcendental freedom. Transcendental freedom is the
kind of absolute freedom the reality of which theoretical reason cannot
establish. Practical freedom, on the other hand, can be “proved through
experience” (A802/B830). He says that although we are also aware of “laws
that are imperatives, i.e., objective laws of freedom, . . . that say what ought to
happen,” nevertheless, “in the practical sphere” we can leave it open whether
in prescribing these laws reason itself may be “determined by further influen-
ces . . . higher and more remote efficient causes” (A802–3/B830–31).
Kant’s terminology here is confusing, but then it often is. Here he is
making a sharp distinction between “practical concepts” on the one hand,
which belong firmly to the everyday world of appearances and have to do
with feelings (A801n/B829n), and on the other hand the demands of pure
practical reason, which do not. As he says, his concern with the former is
“pragmatic” (A800/B828). Our everyday consciousness of the power to
choose must be what gives us the cognition of practical freedom, but that
leaves open the possibility, as far as theoretical reason is concerned, that our
choice may be determined by “higher and more remote efficient causes” of
which we are not conscious.
If this applies to freedom, it must apply to spontaneity. Indeed, though
Kant repeatedly says that synthesis is spontaneous, carried out by the subject
and not externally determined (e.g., B130), one could accept most of his
arguments in the Deduction and in the Analytic of Principles while at the
same time holding that there may or must be external causes that determine
the syntheses we make and the judgments we form. This is why so much of
what Kant says can be accepted by philosophers such as P. F. Strawson,17
17
P. F. Strawson, The Bounds of Sense: An Essay on Kant’s “Critique of PureReason” (London: Methuen,
1966).
208 R. C. S. Walker
who have no commitment at all to freedom in Kant’s sense – or to the

distinction between appearances and things as they are in themselves. For
them, of course, the subject of synthesis is just the ordinary spatiotemporal
human being, but one who nevertheless has to think and order experience in
roughly the way Kant says.
These philosophers are not transcendental idealists. Kant’s transcen-
dental idealism does demand genuine spontaneity and genuine freedom,
with no higher and more remote causes. But it is only when he comes
to write the Critique of Practical Reason that he can find solid ground
for asserting the reality of these things. The “fact of pure reason,”
carrying with it “apodictic certainty” of obligation and therefore of
our freedom to respond, gives him what he needs. It assures us that
in the choices we make we are free from determining factors of any
kind – even determining factors that might operate at the noumenal
level. We can only be so if we ourselves, as subjects, belong to the
underlying reality of things in themselves, since otherwise we should be
subject to the causal determinism of the Second Analogy. And given
that we, as subjects, do have absolute, unconditioned freedom, the
activity of our minds in synthesis must also be free in that same
unconditioned sense. For if causes of any kind, noumenal or otherwise,
were to constrain the activity of the self in thinking, they would be able
to constrain it in acting too.
It is therefore in a tone of surprise and perhaps relief that, in the preface to
the Critique of Practical Reason, he says that this new book provides “a very
satisfactory confirmation” of what the first Critique had claimed, when it
insisted on letting objects of experience as such, including even our own

subject, hold only as appearances but at the same time on putting things in
themselves at their basis and hence on not taking everything supersensible as a
fiction and its concept as empty of content;
whereas
now practical reason of itself . . . furnishes reality to a supersensible

object . . . namely to freedom (although, as a practical concept, only for practical
use), and hence establishes by means of a fact what could there only be thought.
By this, the strange though incontestable assertion of the speculative Critique,
that even the thinking subject is in inner intuition a mere appearance to itself, gets
its full confirmation in the Critique of Practical Reason. (CPrR 5:6)
It is true that in the parenthetical clause he says that the reality thus
established for freedom is “only for practical use,” but as argued earlier we
should not construe this as implying that it is merely something we should try
to believe for practical purposes. The contrast is with establishing freedom “for
speculative use,” which would mean providing a license for theoretical reason
to spin stories about the nature of the underlying self and its free agency. In
such matters, theoretical or “speculative” reason can make no headway at all:
it can only accept the things practical reason provides for it, and “unite
them . . . with its own concepts . . . in one cognition” (CPrR 5:120–21).
This means that there is something importantly missing from the Critique of
Pure Reason, and from the Groundwork too. For it is the Fact of Pure Reason,
and the Fact of Pure Reason alone, that entitles Kant to assert – to know – that
we are free, free in the transcendental and not just the “practical” or empirical
sense. The free self can be recognized “not as receptivity but as pure spontane-
ity,” though “beyond that [it] is also incapable of knowing anything of its
nature” (RP 20:271). By relying on theoretical reason alone, the transcendental
idealism of the Critique of Pure Reason can be put forward as a conjecture, but its
key claims are given their proper foundation by the Fact of Pure Reason.
Practical reason alone can establish reality of transcendental freedom, and
it does so through the Fact of Pure Reason. This is the one place at which it is
possible to gain a firm footing in making judgments about ultimate reality. It
does not take us very far. But it takes us far enough to vindicate the reality of
transcendental freedom, and hence of spontaneous synthesis; to establish also
that there must be a subject of synthesis that is ultimately real. These are
things that have to be true, if freedom is real; and the Fact of Pure Reason
allows us to know that it is.
So practical reason is indeed primary – for Kant; and by no means
implausibly, though much depends on whether we can accept the Fact of
Pure Reason. Practical reason alone gives us access to ultimate reality,
through our recognition of the moral law as binding on us. Theoretical
reason can enable us to gain what we call the truth about the world, but the
truth about the world is nothing but the product of theoretical reason itself,
working through the active faculties of the mind in synthesizing the given
manifold, judging and finding the appropriate coherence among judgments
in accordance with the categorical imperative that governs it.18
18
When I wrote the above I was not aware of a paper by Jens Timmermann with which I am in
considerable agreement: “The Unity of Reason – Kantian Perspectives,” in Spheres of Reason: New
Essays in the Philosophy of Normativity, ed. Simon Robertson (Oxford: Oxford University Press, 2009),
183–98.
10
A Practical Account of Kantian Freedom
Matthew C. Altman
If in the whole series of all occurrences one recognizes purely natural necessity,
is it nevertheless possible to regard the same occurrence, which on one hand is
a mere effect of nature, as on the other hand an effect of freedom?
– Immanuel Kant, Critique of Pure Reason (A543/B571)
One of the most difficult and important questions in all of philosophy is how
to make sense of free will, and specifically how to reconcile the deep
commitment that we have to freedom in our practical lives with the recogni-
tion that events in the world, including human actions, are causally deter-
mined. Kant claims to have solved this problem, but there is significant
critical disagreement about how to interpret his position. In this chapter,
I examine and evaluate competing interpretations of Kant’s theory of free-
dom and conclude that the only way for us to be free as noumena and
causally determined as phenomena is to conceive of our actions in two
different ways, from either the practical standpoint or the theoretical stand-
point. That is, we can commit ourselves to causal determinism with regard to
objective claims about the world, but in acting we must conceive of ourselves
as free agents. This “practical account” of Kantian freedom has several
M. C. Altman (*)
Ellensburg, USA

212 M. C. Altman
explanatory advantages over both the compatibilist and the libertarian inter-
pretations of Kant’s theory, including being more consistent with the epis-
temic limitations that Kant establishes in the Critique of Pure Reason.
Making room for faith in freedom

To understand how the theory of freedom fits into Kant’s philosophy as a
whole, one must begin with the first Critique. There he demonstrates that,
because of our epistemic constraints, things as they are apart from our way of
knowing them, abstracting from the forms of sensible intuition (space and
time) and the pure concepts of the understanding (the categories), are in
principle unknowable (A27–28/B43–44, A146–47/B186–87; Pro 4:322;
OP 22:30–31). This is transcendental idealism: we can know only appear-
ances and never things in themselves.
When we talk about things in themselves as correlates of sensible repre-
sentations – as “transcendental objects” that we represent as appearances – we
begin to talk about them positively (A108–9, A494–95/B522–23). We
consider such things as they would be apprehended directly, and we conceive
of them as noumena:
the concept of a noumenon, taken merely problematically, remains not only

admissible, but even unavoidable, as a concept setting limits to sensibility. But
in that case it is not a special intelligible object for our understanding; rather
an understanding to which it would belong is itself a problem, namely, that of
cognizing its object not discursively through categories but intuitively in a non-
sensible intuition, the possibility of which we cannot in the least represent.
Now in this way our understanding acquires a negative expansion, i.e., it is not
limited by sensibility, but rather limits it by calling things in themselves (not
considered as appearances) noumena. But it also immediately sets boundaries
for itself, not cognizing these things through categories, hence merely thinking
[denken] them under the name of an unknown something. (A256/B311–12;
see also Bxxvi, A286–89/B342–45; Pro 4:312–13)
In the negative sense, noumena are equated with things in themselves,

standing for whatever things are apart from our way of knowing them
(“an unknown something”). In a positive sense, a noumenon is “an object
of a non-sensible intuition” (i.e., intellectual intuition) rather than our
kind of sensible intuition (B307). Although Kant says that we can form no
cognition (Erkenntnis) of things in themselves, he does say that we
10 Practical Account of Kantian Freedom 213
nonetheless think (denken) of such things through the categories. In doing

so, however, he is not extending our knowledge in any way – quite the
contrary. He is saying that whenever we try to make sense of things in
themselves, we must do so by means of the categories, but that in the
absence of schematization (relating sensible intuitions, through the activity
of judgment, in space and time), we have no criteria for knowing if we are
applying the categories correctly. The thing in itself is thus a boundary
concept that marks off the limit of our understanding (Pro 4:316–17; OP
22:31). If we take claims about noumena to be actual claims about things as
they are, we use those claims to explain phenomena, or we treat our
thinking through the categories as cognition rather than mere thinking,
then we are talking nonsense. As Kant puts it, “the concept would
remain . . . without sense, i.e., without significance [ohne Sinn, d.i. ohne
Bedeutung],” because we would be making conceptual claims about some-
thing that cannot be schematized (A240/B299; see also Pro 4:315).
In the Transcendental Dialectic, Kant says that, despite the limits on what
we can know, human reason is naturally led to speculate about the ultimate
ground of the conditioned phenomena that we experience. In our attempt to
transcend the bounds of sense and apply the categories beyond their proper
use, we make logical mistakes (paralogisms) or arrive at seemingly conflicting
propositions (antinomies). With regard to nature in particular, we try to
make sense of the unconditioned condition of all contingent events. That is,
we try to give a complete explanation that accounts for the causes and effects
of the world taken as a whole. Reason then arrives at two equally plausible
propositions, which form the Third Antinomy: either there is some free
causality that begins the series of causes and effects (the thesis), or the totality
of natural laws is a self-sufficient whole that needs no further explanation (the
antithesis) (A444–51/B472–479).
The argument against the thesis and for the antithesis is easy to under-
stand: Every event in nature must be caused by some prior event. Things do
not just happen. In the Second Analogy, Kant shows that we can only
experience the objective order of succession in time if those events are related
by the law of cause and effect; that gets us beyond the mere play of subjective
perceptions (A189–211/B232–56). “Every event has a cause” is necessarily
and universally true, which means that we cannot appeal to uncaused causes
to explain representations. What makes all of the causes and effects possible is
the infinite series of causes and effects considered together, as the self-
subsistent natural order.
The argument against the antithesis and for the thesis is more difficult, but
Kant believes that it is just as convincing. Imagine that every event must have
214 M. C. Altman
some prior cause, so that the causal chain goes back infinitely. Does that
make sense? On this view, a series of prior events has given rise to your
reading this chapter right now. If someone asks why you are reading it, you
may offer a causal explanation. You could say that you are a conglomeration
of matter, and that preceding physical events were such that your being here
now followed as a result. Alternatively, in psychological terms, you are
reading it because of your desire to learn, which was instilled in you by
your parents, who were influenced by their parents, and so on, all the way
back to the Big Bang and everything that led up to that event. Here is the key
move: If someone asks why that series of causes occurred rather than some
other, no explanation can be given. Things could just as easily have occurred
differently, such that you were not reading this chapter or you were not even
born. But this would mean that no explanation can be given for why you are
reading this chapter, because there is no explanation for the series of events
that gave rise to this moment. The only way to get around this is for there to
be an original, uncaused cause that could explain why this particular causal
series exists. Freedom is necessary at some point to give an account of why
things are as they are.
Although the arguments for the thesis and antithesis are equally strong,
Kant says that the resulting contradiction is only apparent. By distinguish-
ing appearances from things in themselves, transcendental idealism allows
us to affirm both positions. With regard to phenomena, every event has a
cause; the antithesis is true. Yet the category of causality applies only there.
Viewed from the intelligible perspective, freedom as an uncaused cause is
possible; the thesis may also be true, although true of noumena and not
phenomena. Freedom and determinism do not contradict one another
(A531–58/B559–86).
Arguments for freedom

Having made “room for faith” in freedom in the first Critique (Bxxx), Kant
takes two different approaches to demonstrate that we are in fact free, or at
least that we must act as if we are free: in the Groundwork of the Metaphysics of
Morals (1785) he appeals to the idea of freedom, and in the Critique of
Practical Reason (1788) he appeals to the fact of reason. In the earlier work,
Kant is content to claim that, because our practical commitment to freedom
is unavoidable, we are constrained by the categorical imperative just as we
would be if we knew that we were free:
every being that cannot act otherwise than under the idea of freedom is just
because of that really free in a practical respect, that is, all laws that are
inseparably bound up with freedom hold for him just as if his will had been
validly pronounced free also in itself and in theoretical philosophy. (G 4:448)
When we deliberate about what to do, we act on the assumption that there is
more than one option, that we freely choose which action to take, that
whatever we decide will determine how we act, and that once we have
acted we could have done otherwise, even with the same preceding states
of affairs – none of which is epistemically justified. Although theoretically we
understand that only one action is possible given the previous events leading
up to it, we cannot take this cognition into the practical, first-person
perspective: we “take a different standpoint when by means of freedom we
think ourselves as causes efficient a priori than when we represent ourselves in
terms of our actions as effects that we see before our eyes” (G 4:450).
Kant’s argument in the Critique of Practical Reason is that the very
exposition of the moral law, common to both the Groundwork and the
second Critique, establishes that it can bind us even without any correspond-
ing inclinations. Through that analysis we come to see that, with every
action, reason judges its maxim against the demands of morality. When we
deliberate about possible actions, we have an immediate consciousness of
moral constraint that confronts us as a “fact of reason [Faktum der
Vernunft]”:
One need only analyze the judgment that people pass on the lawfulness of their
actions in order to find that, whatever inclination may say to the contrary, their
reason, incorruptible and self-constrained, always holds the maxim of the will
in an action up to the pure will, that is, to itself inasmuch as it regards itself as a
priori practical. (CPrR 5:32)
From the fact that we hold ourselves to account, it follows that we must
really be free, because we can only be constrained to do the right thing
if we are capable of doing the right thing (“ought implies can”).1 We
1
There are many passages in which Kant invokes the proposition that “ought implies can,” including
A548/B576, A807/B835; CPrR 5:30, 142, 143n; Rel 6:45, 47, 50, 62, 64; TP 8:276–77; PP 8:370; and
MM 6:380. Although I am using the proposition for a very limited purpose in this chapter, it plays a
crucial role in some of Kant’s arguments in the practical philosophy and his philosophy of religion. For a
discussion and assessment of Kant’s use of this claim, see Jens Timmermann, “Sollen und Können: ‘Du
kannst, denn du sollst’ und ‘Sollen impliziert Können’ im Vergleich,” Philosophiegeschichte und logische
Analyse 6 (2003): 113–22; and Robert Stern, “Does ‘Ought’ Imply ‘Can’? And Did Kant Think It
Does?” Utilitas 16, no. 1 (March 2004): 42–61.
216 M. C. Altman
must be not only free from determination by material desires, but free to
act on the basis of what pure reason demands: by means of the fact of
reason, we discover that “reason is by means of ideas itself an efficient
cause in the field of experience” (CPrR 5:48; see also CJ 5:468). Kant
qualifies this claim, however: freedom is shown to have “objective though
only practical reality [Realität],” or has “only a practical use [Gebrauch]”
(CPrR 5:48, 55, 56). The very exposition of the moral law demonstrates
that our subjection to the moral is “unavoidable [unvermeidlich] even
though it does not rest upon empirical principles” (CPrR 5:55), that our
claim to be bound by it is intuitively warranted, and that our use of the
concept of freedom in explaining our actions is self-justifying – but only
from the practical standpoint.
Moral motivation: Respect as an incentive

Having established that, from the practical perspective, we must think
that we are free or that freedom is real, Kant insists that it is important,
as part of a critique of practical reason, to show not only that we can
recognize the principle of moral obligation (the principium diiudicationis),
but also that we can act on that recognition, or that reason can deter-
mine the will on the basis of such a principle (serving as a principium
executionis). Motivational skepticism poses a threat here: the worry that
all of our actions are ultimately motivated by our passions or instincts
rather than reason, a position that was held by Hume.2 According to
Kant’s ethical theory, respect for the moral law must be able to serve as a
sufficient motive for action – that is, we must be able to do what is right
because it is right, even if we have conflicting sensible inclinations or do
not have so-called moral feelings, such as sympathy for others or delight
in acting morally, that incline us toward the right actions (CPrR 5:30).
So Kant tries to explain how it is possible for “pathologically affected
[affiziert]” beings not to be “pathologically necessitated [necessitirt],”
and to be capable of motivation purely by the demands of reason (A534/
B562; see also LM 28:254–55).
2
David Hume, A Treatise of Human Nature, ed. L. A. Selby-Bigge, rev. ed. P. H. Nidditch, 2nd ed.
(Oxford: Clarendon, 1978), bk. II, pt. iii, §3 (pp. 413–18).
Kant begins by investigating how the rational and sensible natures of

imperfectly rational beings are affected when the moral law serves as a
motive:
nothing further remains than to determine carefully in what way the moral law
becomes the incentive [Triebfeder] and, inasmuch as it is, what happens to the
human faculty of desire [Begehrungsvermögen] as an effect of that determining
ground upon it. For, how a law can be of itself and immediately a determining
ground [Bestimmungsgrund] of the will (though this is what is essential in all
morality) is for human reason an insoluble problem and identical with that of
how a free will is possible. What we shall have to show a priori is, therefore, not
the ground from which the moral law in itself supplies an incentive but rather
what it effects (or, to put it better, must effect) in the mind insofar as it is an
incentive. (CPrR 5:72)
According to Kant, subjection to the moral law motivates action by virtue of

what he calls respect or reverence (Achtung) (G 4:400–401; CPrR 5:74–88;
Rel 6:23n). Kant acknowledges that he cannot explain how respect is caused
by subjection to the moral law. Explaining how a pure, undetermined moral
interest can exert a motivational/incentivizing effect on or in the sensible
world would transgress the bounds of sense and apply the category of
causality beyond its limited use. Kant therefore concerns himself not with
how such an incentive is possible, but with how imperfectly rational beings
experience constraint by the categorical imperative when it contradicts
material principles of self-love.
For Kant, an imperfectly rational being has both lower and higher
faculties of desire (unteres and oberes Begehrungsvermögen). The former
produces pleasure through the attainment of an end desired subjectively
by one’s sensible nature, and the latter expresses the demand of pure
reason to further humanity as an end in itself. Moral constraint affects
the agent by subordinating the tendency to egoism that would have her
exempt herself from the moral law. Because an imperfectly rational being
has a propensity to evil or self-conceit (Rel 6:29–36), she undergoes
“humiliation [Demütigung]” (CPrR 5:75, 78–79; see also MM 6:436).
That is, she regards her inclinations, in contrast to her moral obligations,
as unworthy of satisfaction. Correspondingly, she experiences a sense of
“exaltation [Erhebung]” in realizing that she can freely legislate her
principles of acting (MM 6:436). In short, this moral feeling of “self-
approbation [Selbstbilligung]” and subsequent “elevation [Erhebung]” of
the rational being, both of which follow from a pure moral interest,
218 M. C. Altman
oppose the lower faculty’s pursuit of happiness as a subjective end,

thereby thwarting sensible pleasure in favor of positive “moral feeling
[moralische Gefühl]” (CPrR 5:78–81). This is reason’s “own kind of
satisfaction [Zufriedenheit],” which results from “the establishment of a
good will . . . even if this should be combined with many infringements
upon the ends of inclination” (G 4:396).3
Noumenal causality
Kant’s identification of the will with pure practical reason means that one can
choose differently regardless of the events that precede one’s choice, since the
choice itself is undetermined by material causes such as inclinations. Yet the
uncaused cause affects the higher and lower faculties of desire, thus introdu-
cing changes into causally determined phenomena. In other words, as a free
act, choosing rightly occurs outside of causally determined events in time;
but it affects events in time by means of moral feelings such as humiliation,
exaltation/elevation, self-approbation, and satisfaction:
Now this acting subject, in its intelligible character, would not stand under any
conditions of time, for time is only the condition of appearances but not of
things in themselves. In that subject no action would arise or perish, hence it
would not be subject to the law of everything alterable in its time-determina-
tion that everything that happens must find its cause in the appearances (of
the previous state). (A539–40/B567–68; see also LE 27:505)
A rational choice must occur nontemporally, because it is not causally

determined (as an event in time). However, a person’s actions must be
motivated by the feeling of respect in order for him to be a responsible
moral agent, whose behavior is grounded in pure practical reason. This is
perhaps the paradigm case of thinking through the categories: we con-
ceive of a free, uncaused choice (noumenal freedom) as the cause of an
action (a phenomenal event in the world), even though the category of
causality cannot be applied to things in themselves (CPrR 5:55–56).
3
For a more detailed account of how pure reason moves us to act, see Stephen Engstrom, “The Triebfeder
of Pure Practical Reason,” in Kant’s “Critique of Practical Reason”: A Critical Guide, ed. Andrews Reath
and Jens Timmermann (Cambridge: Cambridge University Press, 2010), 90–118; and Patrick R.
Frierson, Kant’s Empirical Psychology (Cambridge: Cambridge University Press, 2014), 116–66.
From the sensible perspective, any action that we perform is deter-

mined by prior events. Under transcendental idealism, however, one and
the same event can be understood as having both intelligible and sensible
causes (A536/B564). From the intelligible perspective, rational decisions
change the way the world is in conformity with “its own order according
to ideas” (A548/B576). Phenomena have an intelligible “ground” in the
thing in itself, so actions, as events, can be explained in terms of
noumenal acts:
If . . . appearances do not count for any more than they are in fact, namely, not
for things in themselves but only for mere representations connected in
accordance with empirical laws, then they themselves must have grounds
[Gründe] that are not appearances. Such an intelligible cause, however, will
not be determined in its causality by appearances, even though its effects appear
and so can be determined through other appearances. (A537/B565)
According to this picture of free self-determination, nontemporal choices can

cause actions in time because the content of appearances, although not their
“connect[ions] according to empirical laws,” is determined by (or, more
properly, grounded in) the thing in itself.
As we have seen, Kant’s conception of what it means to be free in a moral
sense involves several different elements:
1) Since we choose freely as noumena, we have different options that are all
possible choices. When someone acts wrongly, Kant says, it does not
matter what the rest of his life was like. We regard “reason as a cause
that, regardless of all the empirical conditions . . . , could have and ought
to have determined the conduct of the person to be other than it is”
(A555/B583; see also MM 6:213).4 In order for a given end to become an
incentive to act, I must “incorporate” it into my maxim (Rel 6:23–24), by
which Kant means that I must take the potential incentive to be some-
thing that I have reason to pursue. This is practical freedom in the negative
sense (G 4:446; CPrR 5:33; MM 6:213–14; LM 28:256–57).
2) We must have the freedom to choose not only among different
possible incentives (Willkür), but also to act on the basis of a principle
4
This is not to say that only agents who can act wrongly are free. Kant contends that God, as a perfect
being with a holy will, can only choose rightly and that God exhibits “true freedom” (LDR 28:1068; see
also 1066, 1097). However, practical freedom in imperfectly rational beings involves the ability to
choose among different maxims, some of which are contrary to the moral law.
220 M. C. Altman
that we give to ourselves through pure practical reason (Wille).5 If we

were free only in the former sense, then we would be free from
determination by material causes (A533–34/B561–62), but we would
be choosing among options that are given to us from without, and
would thus be acting heteronomously. Having freedom of choice and
the ability to act for the sake of duty alone means that we are
responsible moral agents who are capable of autonomous self-determi-
nation. This is practical freedom in the positive sense (G 4:446–47;
CPrR 5:33; MM 6:213–14).
3) I am also free in the sense that my willing is an uncaused cause. It has
no prior event that determines it; my action is determined sponta-
neously. Yet it effects change in the world of appearances. As Kant
phrases it, “a causality must be assumed through which something
happens without its cause being further determined by another pre-
vious cause, i.e., an absolute causal spontaneity beginning from itself
a series of appearances that runs according to natural laws” (A446/
B474; see also G 4:446, 453; CPrR 5:132). This is transcendental or
cosmological freedom (A533/B561, A803/B831; CPrR 5:3, 29, 96–97).
Because being bound to act for the sake of duty requires that we be
free to determine our actions apart from pathological causes – “ought
implies can,” or “ought only if can” – practical freedom depends on or
is made possible by transcendental freedom (A533–34/B561–62,
A802–3/B830–31; CPrR 5:97).
All of these things together – choice, autonomy, and uncaused causation –
make us moral agents.
In addition to these three kinds or characteristics of freedom, Kant also
claims that every event has a cause, and thus that everything that happens,
including human actions, is in principle predictable based on empirical
laws:
4) My actions are determined by prior events in time, since all representa-
tions are subject to the category of causality. This is determinism.
5
Kant’s distinction between Wille and Willkür is given a lot of explanatory weight by Lewis White Beck,
A Commentary on Kant’s “Critique of Practical Reason” (Chicago: University of Chicago Press, 1960),
176–203. The distinction has been challenged by, among others, Nelson Potter, Jr., “Does Kant Have
Two Concepts of Freedom?” in Akten des IV. Internationalen Kant-Kongresses, ed. Gerhard Funke and
Joachim Kopper, 2 vols. (Berlin: de Gruyter, 1978), 2:590–96; and Hud Hudson, “Wille, Willkür, and
the Imputability of Immoral Actions,” Kant-Studien 82, no. 2 (Jan. 1991): 179–96. I employ the
distinction between Wille and Willkür only to clarify different kinds of free activity. My exposition does
not depend on whether Kant consistently maintains it.
In his solution to the Third Antinomy, Kant claims to resolve the seeming
contradiction between determinism and freedom. But it is not clear how we
can be practically and transcendentally free if causal determinism is true. In
other words, how can our actions be attributable to us as agents if they are
also fully explained by prior events? A number of contemporary Kant
scholars have attempted to answer this by interpreting Kant’s philosophy as
a form of either compatibilism or libertarianism, but in the sections that
follow I reveal the shortcomings of these interpretations. Both of them
emphasize one element of Kant’s theory at the expense of others – either
ruling out transcendental freedom in favor of determinism, or qualifying
determinism so that the moral agent can change the course of nature. I will
argue that only a practical account of Kant’s theory of freedom accommo-
dates both determinism and (transcendental and practical) freedom. To do
so, I distinguish theoretical cognition, as it is described in the first Critique,
from the practical point of view that is so crucial in the arguments of the
Groundwork and the second Critique.
Compatibilism and causality

According to Kant, one of his major accomplishments is distinguishing
appearances from things in themselves. Transcendental idealism makes
room for faith in freedom by showing that the category of causality applies
to all phenomena, but only phenomena, so we may be free when considered
apart from the categories. Kant also purports to establish the practical reality
of freedom in the second Critique. Therefore, a successful interpretation of
Kant’s theory of freedom would affirm that our actions are determined and
also validate the conditions of moral agency: having live options when one
deliberates, such that one could have done otherwise; being able to be
motivated by duty alone; and having one’s free choice be causally effective
in the world of appearances.
Because Kant affirms that we are both free and determined, he is often
considered a compatibilist. According to compatibilism, all events, including
human actions, are determined by prior events. Actions are free and attribu-
table to the agent as long as they are caused in the right way, by certain kinds
of psychological causes that are internal to the agent. If I purposely decide to
push my friend into a ditch, that action is caused by desires that I have. To
use Harry Frankfurt’s language, if I want to push her and I approve of that
desire (I want to want to push her), and if those desires bring about my
222 M. C. Altman
action, then I am responsible for or have willed the action – even though my
desires, whether lower- or higher-order, are all caused by prior events.6
Although the correspondence is not perfect, Kant gives us a similar picture
of mental life, with the lower faculty of desire concerned with what I want
and the higher faculty of desire concerned with morality, a particular kind of
reflective desire or interest.
I have all sorts of things going on in inner sense, including desires and
volitions, and these can cause me to act. My action is attributable to me if it
is caused by desires and volitions that I identify as mine rather than by some
external force. There is no inconsistency in saying that my character is the
result of my temperament and my social circumstances, and that my actions
are the result of my character, since on this account I am part of the
phenomenal world. All of these causes have prior causes themselves, so
they could be said to cause an action without introducing any conceptual
problems.
This assumes that I identify myself with my desires, my deliberation, my
act of reason-giving, and so on – which of course I do. But for Kant,
compatibilism is merely the “comparative concept of freedom” (CPrR
5:96), in the sense that having internal mental events that cause me to
push someone into a ditch is in some sense more “mine” than if I trip and
accidentally bump into her. However, this account of voluntary action does
not get us the kind of freedom necessary for moral responsibility, since the
action is still attributable to prior causes that are themselves determined.
Kant claims that morality is possible because, although determinism is true of
appearances, it is not true of things in themselves. If Kant successfully
validates practical and transcendental freedom in his practical philosophy,
my actions cannot be wholly attributable to events, even psychological
events. Instead, my character is the result of an absolutely free choice of
maxim as the governing principle of my will, and that choice is not in time
(A553–54/B581–82).
Ralf Meerbote and Hud Hudson draw on Donald Davidson’s anomalous
monism to defend the compatibilist interpretation of Kant’s theory of free-
dom.7 Meerbote and Hudson claim that mental states are token-identical to
6
Harry G. Frankfurt, The Importance of What We Care About: Philosophical Essays (Cambridge:
Cambridge University Press, 1998).
7
Ralf Meerbote, “Kant on the Nondeterminate Character of Human Actions,” in Kant on Causality,
Freedom, and Objectivity, ed. William L. Harper and Ralf Meerbote (Minneapolis: University of
Minnesota Press, 1984), 138–63; and Hud Hudson, Kant’s Compatibilism (Ithaca: Cornell University
Press, 1994). For Davidson’s theory of anomalous monism, see especially Donald Davidson, “Mental
Events,” in Essays on Actions and Events, 2nd ed. (Oxford: Clarendon, 2001), 214–25.
physical states, but they are type-distinct. Token identity of mental states and
physical states just means that each and every token (i.e., particular) mental
state is identical with some token physical state or other. So, with regard to
freedom, a volition is identical with some physical event in the brain.
However, willing and whatever brain state that we are in when we are willing
are not type-identical, meaning that properties of the will and properties of the
brain state are different, and one set of properties cannot be reduced to the
other. If willing and brain state X are type-distinct, then the “willing” predicate
cannot be dispensed with and replaced by “having brain state X,” even if X
happens to be the correlate of willing in beings like us. Thus, mental states and
physical states are correlated, but we still must talk about our inner lives in
mental terms, with mental properties that are distinct from physical properties.
On the face of it, this seems like a plausible interpretation of Kant’s theory.
Token identity provides for the correlation between mental events and physical
events that is necessary if two different perspectives on the same action are
possible, one with reference to freedom and one with reference to prior causes.
In addition, type distinctness means that mental states cannot simply be
eliminated with a full account of physical states. This is consistent with
Kant’s claim that an explanation in terms of freedom is separate from an
explanation in terms of prior events. Hudson’s and Meerbote’s compatibilism
allows us to give two different descriptions of the same action, one in terms of
physical properties and one in terms of mental properties, which seems to get at
Kant’s distinction between the self as phenomenon and the self as noumenon.
This appeal to tokens and types, however, fails to capture many facets of
Kant’s theory of freedom. First, it does not provide for the possibility of
different actions given the preceding series of events. When compatibilists
say that an agent could have done otherwise, what they mean is that other
states of affairs are logically possible: another action could have taken place, but
if it had, the events and/or laws leading up to that action would have been
different. However, when I blame someone for something, I am not merely
wishing that preceding events had been different, such that she ended up with
a different character. When Kant says that an agent could have done otherwise,
he means that she has more than one option even with her unique causal
history (A554–55/B582–83). If we accept causal determinism in the way that
the compatibilist does, then moral responsibility in Kant’s sense is impossible:
If I say of a human being who commits a theft that this deed is, in accordance
with the natural law of causality, a necessary result of determining grounds in
preceding time, then it was impossible that it could have been left undone;
how, then, can appraisal [Beurteilung] in accordance with the moral law make
224 M. C. Altman
any change in it and suppose that it could have been omitted because the law
says that it ought to have been omitted? (CPrR 5:95)
Compatibilism only says that different prior physical events could have
necessitated different actions, but Kant’s objection to compatibilism is that
some prior physical event or other necessitated her action at all. That cannot
warrant moral “appraisal.” The action must be up to her as a rational agent,
and for it to be up to her, she must be able to perform or not perform the
action under the same empirical conditions.
The second problem with the token-type distinction is that it, and compa-
tibilism in general, does not involve noumenal causality. There are two reasons
for this. For one thing, Hudson and Meerbote can say that mental states and
physical states are correlated, but this is not the same thing as saying that the
physical state is the result of a mental state. According to Davidson, physical
events can only be causally related to other physical events; mental events cause
physical events only in the sense that they are correlated with token-identical
physical events that are causally effective: “every mental event that is causally
related to a physical event is a physical event [i.e., it is the event to which it is
token-identical].”8 It does not make sense for mental states to cause physical
states if what distinguishes them is not their token, which is identical, but their
properties. A mental property causes a physical event only insofar as the former
is (i.e., is token-identical with) a physical event.
Another argument against the compatibilist interpretation is that the activity
of choosing freely and setting our ends cannot simply be mental predicates that
are token-identical to physical states. Kant would agree with Davidson’s dis-
tinction as long as the mental events he describes are psychological events in
inner sense. Kant says that the moral feeling of respect causes me to act, but he
also says that my transcendentally free choice to act for the sake of duty causes
the feeling (or causes the feeling to be effective). In replacing the phenomena-
noumena distinction with the distinction between tokens and types, Meerbote
and Hudson give an account that need not appeal to anything but phenomena.
Mental states are nothing more than events in time that are as determined as
physical tokens, even if their properties are different. Thus their account is
consistent with scientific naturalism, because we only need empirical objects to
explain both physical and mental events. Yet Kant claims that transcendental
idealism “make[s] room for faith” in freedom by distinguishing phenomena
from noumena, and showing that the concepts of the understanding apply only
8
Davidson, “Mental Events,” 224.
to the former (Bxxx). He also claims that “the idea of freedom makes me a
member of an intelligible world,” not that it gives me temporally located mental
predicates (G 4:454). The distinction between mental and physical types is not
equivalent and does not function in the same way as the distinction between
noumena and phenomena.
A similar problem faces Robert Hanna’s so-called “post-compatibilism.”9
According to Hanna, the laws that govern events in the world are determined
insofar as they are explicable in terms of Newtonian physics, but that mechanistic
understanding does not give us the whole picture of reality. As living organisms,
our activity is non-conceptual – our self-organizing activity is explicable in terms
of final causes in addition to efficient causes, the latter of which cannot explain
dependence of the parts on the whole, and vice versa (CJ 5:372–73) – so we can
understand our actions as the effects of “causal-dynamical laws of biological,
conscious, and rational activity, which enrich and supplement the repertoire of
general deterministic, mechanistic natural causal-dynamic laws.”10 According to
Hanna, there is room even within a causally determined world to explain human
actions in terms of purposes rather than causes.
Hanna’s position rests on some debatable assumptions, such as the claim
that Kant is committed to non-conceptualism, the theory that we can have
representational content that is not subject to concepts.11 Even if we grant
that Hanna correctly interprets Kant’s views regarding biological organisms
and their activity, however, it still is not a satisfactory theory of transcen-
dental freedom. The appeal to biological events, even self-organizing events,
could only get us comparative freedom – that is, when compared to the
determinism of events that are conceptually organized along Newtonian
lines. Biological events, as phenomena, are subject to natural laws rather
than the moral law: as Kant says, “the correctness of the principle of the
thoroughgoing connection of all occurrences in the world of sense according
to invariable natural laws is already confirmed as a principle of the transcen-
dental analytic and will suffer no violation” (A536/B564). In addition,
Hanna’s “self-organizing thermodynamic systems” are situated in time,12
9
Robert Hanna, “Freedom, Teleology, and Rational Causation,” in Kant Yearbook, vol. 1: Teleology, ed.
Dietmar H. Heidemann (Berlin: de Gruyter, 2009), 99–142.
10
Ibid., 119.
11
See the special issue of the International Journal of Philosophical Studies devoted to this question
(vol. 19, no. 3 [July 2011]), which includes Robert Hanna, “Beyond the Myth of the Myth of the
Given: A Kantian Theory of Non-Conceptual Content,” 323–98; and Robert Hanna, “Kant’s Non-
Conceptualism, Rogue Objects, and the Gap in the B Deduction,” 399–415. See also Chapter 33 of this
volume.
12
Hanna, “Freedom, Teleology, and Rational Causation,” 119.
226 M. C. Altman
and Kant is clear that the will must be outside of time in order to be free from
determinism:
The causality of reason in the intelligible character does not arise or start
working at a certain time in producing an effect. For then it would itself be
subject to the natural law of appearances, to the extent that this law determines
causal series in time, and its causality would then be nature and not freedom.
(A551–52/B579–80)
Depending on whether we can make sense of Kant as a non-conceptualist,

Hanna could explain why different choices are possible, insofar as they
operate apart from mechanistic laws. However, biological organisms are
not autonomously self-determining agents who have practical freedom in a
positive sense, and their freedom from one kind of deterministic laws cannot
get us to uncaused causation.
According to Kant’s transcendental idealism, freedom and determinism are
compatible because an action can be explained both in causal terms as a result
of either physical or psychological causes (the latter of which is the compatibi-
list picture of freedom) and as determined by an absolutely free, noumenal
choice (the libertarian picture of freedom). This has led to some confusion in
the secondary literature. Whether Kant is a compatibilist depends on how we
define compatibilism. Certainly he is no compatibilist in the sense that Hume,
Ayer, or Frankfurt is a compatibilist.13 They all believe that people can be held
responsible for their actions, but they are also complete causal determinists, or
at least believe that we can be free even if determinism is true,14 and without
the noumena-phenomena distinction, that means that there is no uncaused
cause. For them, I am morally responsible as long as my choices are caused in
the right way – namely, by internal, psychological causes with which I identify.
By contrast, Kant believes that freedom from determination is necessary for
freedom in the moral sense, which makes him an incompatibilist. We only are
fully free when we do what is right because it is right, apart from material
causes such as inclinations. Contrary to what Hanna claims,15 determination
13
Hume, Treatise of Human Nature, bk. II, pt. iii, §§1–2 (pp. 399–412); David Hume, An Enquiry
concerning Human Understanding, in Enquiries concerning Human Understanding and concerning the
Principles of Morals, ed. L. A. Selby-Bigge, rev. ed. P. H. Nidditch, 3rd ed. (Oxford: Clarendon, 1975),
§8 (pp. 80–103); A. J. Ayer, “Freedom and Necessity,” in Philosophical Essays (New York: St. Martin’s,
1963), 271–84; and Frankfurt, Importance of What We Care About, esp. 1–103.
14
See Frankfurt, “Alternate Possibilities and Moral Responsibility,” in Importance of What We Care
About, 1–10.
15
Hanna, “Freedom, Teleology, and Rational Causation,” 125–26, 129.
of one’s actions by higher-order desires, as in the compatibilism of Frankfurt or

Hume, would not be enough, on Kant’s view, to attribute moral responsibility
to the person. To be free, those higher-order desires would have to be ultimately
grounded in pure practical reason. Therefore, Kant believes that freedom and
determinism are compatible only in the sense that we can both appeal to mental
events to explain actions (compatibilism) and that we can explain voluntary
actions with reference to noumenal choices, which must be distinct from
deterministic events (incompatibilism). As Allen Wood puts it, Kant commits
himself to “the compatibility of compatibilism and incompatibilism.”16
This leaves us with a threefold distinction when accounting for human
action, a distinction that parallels Kant’s analysis of the Lockean “thing in
itself.” Empiricists such as Locke distinguish secondary qualities of objects as
they appear to different observers from the object as it really is,17 and Kant
agrees with this distinction in the sense that objective claims can be made about
spatiotemporally located perceptions, with primary qualities such as extension
in space, as distinct from mere sensations. But this empirical distinction
should not be confused with the transcendental distinction between things in
themselves and spatiotemporal objects, because the bearer of Lockean primary
qualities is also a representation for consciousness (A45–46/B62–63,
A372–73). Correlatively, the compatibilist can distinguish between actions
that have external causes (tripping and accidentally bumping into my friend)
and actions that have internal causes (pushing my friend because I want to do
it), but Kant distinguishes both of these empirical causes from what stands apart
from the entire nexus of causes and effects. Only the noumenal agent is capable
of transcendental freedom, so only an appeal to noumena can account for the
moral attribution of blame, properly speaking. Kant sets out this threefold
distinction in a lengthy passage in the second Critique:
in the question about that freedom which must be put at the basis of all moral
laws and the imputation appropriate to them, it does not matter whether the
causality determined in accordance with a natural law is necessary through
determining grounds lying within the subject or outside him, or in the first case
whether these determining grounds are instinctive or thought by reason,
if . . . these determining representations have the ground of their existence in
time and indeed in the antecedent state, and this in turn in a preceding state,
16
Allen W. Wood, “Kant’s Compatibilism,” in Self and Nature in Kant’s Philosophy, ed. Allen W. Wood
(Ithaca: Cornell University Press, 1984), 74.
17
See John Locke, An Essay Concerning Human Understanding, ed. Peter H. Nidditch (Oxford:
Clarendon, 1975), 132–43.
228 M. C. Altman
and so forth, these determinations may be internal and they may have psycholo-
gical instead of mechanical causality, that is, produce actions by means of
representations and not by bodily movements; they are always determining
grounds of the causality of a being insofar as its existence is determinable in time
and therefore under the necessitating conditions of past time, which are thus,
when the subject is to act, no longer within his control and which may therefore
bring with them psychological freedom (if one wants to use this term for a merely
internal chain of representations in the soul) but nevertheless natural necessity;
and they therefore leave no transcendental freedom, which must be thought as
independence from everything empirical and so from nature generally, whether it
is regarded as an object of inner sense in time only or also of outer sense in both
space and time; without this freedom (in the latter and proper sense), which alone
is practical a priori, no moral law is possible and no imputation in accordance with
it. (CPrR 5:96–97; see also LM 28:256–57, 267–68)
Like the empiricist’s claims about the “thing in itself,” which is really an
objective representation and thus merely an appearance, the compatibilist’s
claims about freedom are really about psychological freedom rather than
transcendental freedom. If we were left only with that – with token-identical
phenomena of different types, or self-organizing biological systems – then
moral agency as Kant understands it would be impossible.
If Kant is trying to demonstrate “the compatibility of compatibilism and
incompatibilism,” then a compatibilist reading is not false, since it does
capture a kind of personal responsibility – namely, psychological freedom.
We must make this distinction, even in the world of appearances, between
voluntary actions, which issue from my desires, character, and intentions;
and involuntary actions, which can be explained with reference only to
physical states and not mental states. Since this view ignores noumenal
causality, however, it is bound to be an incomplete interpretation of Kant’s
theory of freedom.
Libertarian agency and personal identity

Libertarianism seems like a more promising approach. First, libertarians hold
that free will is incompatible with determinism. Kant seems to have held this
view to the extent that he thought it necessary to limit determinism to the
realm of appearances. This allows us to choose among different options even
given the same causal history leading up to that point, since our choice, as
noumenon, is not determined by that causal history. In addition, libertarians
assert that a free action is caused not by prior events but by the decision of
the agent: “agent causation” as opposed to “event causation.”18 Actions that
are willed freely are the result of reasons that the agent has, and we often
appeal to those reasons instead of physical causes to explain what a person
does. For example, I can say, “I pushed my friend because I thought it would
be funny,” and that seems sufficient without any reference to my neuro-
chemical state.
With this appeal to the noumenal agent, however, problems begin to
emerge. In order for me to have objective experience, my self-consciousness
must hold together different cognitions so that they can be related to one
another as persisting objects over time (A106–10, B131–39). In addition,
without an experience that is held together, Kant claims that there would
be no unified subject, but a series of unrelated perceptions: “only because I
can comprehend their manifold [i.e., representations] in a consciousness do
I call them all together my representations; for otherwise I would have as
multicolored, diverse a self as I have representations of which I am con-
scious” (B134). Kant establishes a link between temporal unity among
representations and the possibility of being an identical subject of experi-
ence. I can have a coherent experience only if I am a singular consciousness
that has successive experiences, and – what is more important for our
purposes – I can have a particular consciousness only if I synthesize
perceptions into a spatiotemporal manifold. I have to distinguish the
external world of objects from subjective activity if I am to persist as a
unified subject over time; I must be unified, despite the fact that my
cognitions constantly change. The problem is: how could there be a
noumenal agent if my existence as a self-conscious subject depends on my
relation to objective representations?
Moral agency also depends on a corresponding unity of one’s actions as one’s
own from moment to moment. An agent who chooses with no sense of the past
or future and who could not situate herself with regard to her other actions,
prior commitments, and character as it has developed over time would pre-
sumably not be a rational agent but would exist in a kind of dissociative fugue
state. Because time is a pure form of sensible intuition, however, the idea of a
18
See Roderick M. Chisholm, “Human Freedom and the Self,” in Free Will, ed. Gary Watson, 2nd ed.
(New York: Oxford University Press, 2003), 26–37; Randolph Clarke, “Toward a Credible Agent-
Causal Account of Free Will,” Noûs 27, no. 2 (June 1993): 191–203; Randolph Clarke, “Agent
Causation and Event Causation in the Production of Free Action,” Philosophical Topics 24, no. 2
(Fall 1996): 19–48; and Timothy O’Connor, Persons and Causes: The Metaphysics of Free Will (Oxford:
Oxford University Press, 2000).
230 M. C. Altman
temporally persisting intelligible agent is a contradiction. Kant says that free

causality must occur outside of time in order not to be a determined event
(CPrR 5:95–97). A noumenal subject could have no character because it has no
past or present, and no unified subjectivity (in relation to a manifold). Yet our
conception of an agent includes a memory, character traits, and the persistence
of core values, all of which only make sense with regard to the self of inner sense
rather than a noumenal “self.”
Without having a subject who acts, any noumenal choice is an isolated
action that has no relation to past or present. This is a classic problem for
indeterminism: Imagine choosing in a way that is completely uninfluenced
by your circumstances. If we exclude your upbringing, your natural tem-
perament, your cultural mores, your desires and interests, and everything
else that you have been taught, and you are supposed to choose indepen-
dently of those things, how do you decide what to do? For Kant, “the
empirical character is . . . determined in the intelligible character,” meaning
that the ordering of one’s desires and values must be the result of an
uncaused cause, or a choice that is undetermined (A551/B579; see also
Rel 6:25). The choice in that case seems not to be free, but arbitrary,
because it has no basis in who the person is. This is why Hume thinks that
only compatibilism, which traces voluntary actions to a person’s character,
can make sense of moral praise and blame.19
A second problem with noumenal agency is that the consciousness of
myself as an appearance must correspond to the thing in itself that grounds
that appearance. In other words, I have to think that the choices of my
noumenal self are reflected in the choices of my phenomenal self, that my
choices are distinguishable from your choices, that those noumenal choices
are attributable to me, and so on. Indeed, Kant says that the intelligible
character is “indicated through [angegeben durch]” the empirical character as
its “sensible sign [sinnliche Zeichen]” (A546/B574; see also A551/B579). The
problem is that such a correspondence cannot make sense given the claims of
the first Critique. As Ralph Walker notes, a thing is distinguished as an
individual (distinct from other things and identical to itself) by being set in
space and time, and the thing in itself is not in space and time (A48–49/
B66).20 This is why Schopenhauer calls space and time the principium
19
Hume, Treatise of Human Nature, bk. II, pt. iii, §2 (pp. 409–12); and Hume, Enquiry concerning
Human Understanding, §8, pt. iii (pp. 98–99).
20
Ralph C. S. Walker, “Kant on the Number of Worlds,” British Journal for the History of Philosophy 18,
no. 5 (Dec. 2010): 824–25. Although this is not the time to adjudicate this dispute, I should note that
this is in contrast to Allen Wood’s Identity Interpretation, according to which there is an identity
individuationis (principle of individuation), because for there to be a plurality

of distinct objects they must occupy different spatiotemporal positions. Apart
from the principle of individuation, the thing in itself is not a plurality, so it
cannot be composed of distinct individuals.21
Schopenhauer extends this to agency as well – there are no noumenal
agents, only an undifferentiated will – yet, according to Kant, the practical
perspective on our actions requires a kind of personal identity, or the ability
to make moral attributions to individual agents. Thus we are left with
nothing but a kind of faith that my noumenal self is related to a class of
phenomenal actions, even though the idea of a unified noumenal self is not
only theoretically unjustified but theoretically nonsensical – because,
respectively, it is a paralogism to infer noumenal personhood on the basis
of apperception (A361–66, B408–10), and space and time do not apply to
noumena (A26/B42, A32–33/B49). Kant has made room for faith, but it is
unclear what could be the propositional content of that faith. The libertar-
ian alternative to the compatibilist’s comparative freedom leaves us at an
explanatory impasse.
Causality and temporality

We also have to consider the idea of an “uncaused cause.” The first part of
the phrase is unproblematic when applied to noumena, since the category of
causality is restricted in its use to things as appearances. But then libertarians
invoke the concept in claiming that the self as noumenon can be a cause of
actions (“agent causation”). This seems impossible since actions qua events
are absolutely predictable given the history of events leading up to them.
Prior events are the cause of events in the present, not timeless choices that
are supposed to insert themselves somehow into the (otherwise unbroken)
causal chain of determined phenomena.
between things in themselves and objects as they are when thought through the categories, apart from
the objects’ sensible properties. See Allen W. Wood, Kant (Oxford: Blackwell, 2005), 64–76. I do not
see how this interpretation can be squared with some of Kant’s claims in the first Critique, such as: “If by
merely intelligible objects we understand those things that are thought through pure categories, without
any schema of sensibility, then things of this sort are impossible” (A286/B342; see also A256/B311–12).
For an argument against the Identity Interpretation, see Dennis Schulting, “Limitation and Idealism:
Kant’s ‘Long’ Argument from the Categories,” in Kant’s Idealism: New Interpretations of a Controversial
Doctrine, ed. Dennis Schulting and Jacco Verburgt (Dordrecht: Springer, 2011), 173–78.
21
Arthur Schopenhauer, The World as Will and Representation, vol. 1, trans. and ed. Judith Norman,
Alistair Welchman, and Christopher Janaway (Cambridge: Cambridge University Press, 2010), 137–38,
152–53, 358.
232 M. C. Altman
Benjamin Vilhauer confronts this problem in defending his own libertar-

ian interpretation. According to Vilhauer, we must distinguish the form of
experience that is contributed by the category of causality from the matter of
experience to which we apply the categories. Although deterministic causa-
tion in general characterizes phenomena as a result of how we organize things
conceptually, the particular causal laws that we instantiate in our actions and
that relate our actions to prior and future events are the result of what,
noumenally, we decide.22 That is, even though we organize those actions
through the activity of the understanding, thus conceiving of them as
determined events, free activity on the part of the noumenal subject supplies
the content, or the particular events/actions, about which we form causal
judgments. Thus agents are responsible for “limited instantiation scope”
causal laws, or particular law-governed events that give rise to other law-
governed events.23 For example, choosing to get up and open the door has
consequences: the door opens, allowing cold air to come into the house,
which triggers the thermostat and the furnace, and so on – none of which
would have happened had I not chosen to open the door. Eric Watkins
presents a similar, ontologically robust interpretation: the subject in itself
determines the intelligible character, which grounds psychological laws of
behavior, which cause actions.24
Although Vilhauer and Watkins attempt to address perhaps the most
difficult question for libertarians – “how there can be any room in the
deterministic causal series for free will to play a role in shaping its
course”25 – their interpretation grates against Kant’s own descriptions of
both natural events and transcendental freedom. First, it is true that a
representation has formal conditions that are contributed by the knowing
22
Benjamin Vilhauer, “Incompatibilism and Ontological Priority in Kant’s Theory of Free Will,” in
Rethinking Kant, vol. 1, ed. Pablo Muchnik (Cambridge: Cambridge Scholars, 2009), 22–47; Benjamin
Vilhauer, “The Scope of Responsibility in Kant’s Theory of Free Will,” British Journal for the History of
Philosophy 18, no. 1 (Jan. 2010): 45–71; and Benjamin Vilhauer, “Kant and the Possibility of
Transcendental Freedom,” in The Palgrave Handbook of German Idealism, ed. Matthew C. Altman
(London: Palgrave Macmillan, 2014), 105–25, esp. 115–16. Vilhauer echoes Wood’s conception of
intelligible causality in “Kant’s Compatibilism,” 86–89.
23
Vilhauer, “Scope of Responsibility in Kant’s Theory of Free Will.”
24
Eric Watkins, “The Metaphysics of Freedom,” in Kant and the Metaphysics of Causality (Cambridge:
Cambridge University Press, 2005), 301–61. Hanna’s view of the relationship between free choosing
and the form and matter of experience also overlaps in some ways with Vilhauer’s and Watkins’s views.
For example, Hanna writes: “the complete set of general deterministic mechanistic natural causal-
dynamic laws provides a skeletal causal-dynamic architecture for nature, which is then gradually fleshed in
by the one-off laws of self-organizing thermodynamic systems” (“Freedom, Teleology, and Rational
Causation,” 119).
25
Vilhauer, “Kant and the Possibility of Transcendental Freedom,” 113.
subject and sensations that are somehow given by the thing in itself. But
these cannot be separated into a free matter and a determined form. Kant
says that every representation, governed as it is by the category of causality,
could not have been otherwise given the conditions that preceded it. Every
action, composed of form and matter, is in principle predictable. Yet
Watkins says that, since the psychological laws result from free choices,
the actions that they produce do not follow from determined events that
precede them. And Vilhauer says that noumenal activity somehow popu-
lates the form, undetermined, such that we know that whatever happens
must have happened, but we do not know what will happen until the
noumenal choice has been made to introduce “particular causal laws” into
a general explanatory framework: “our phenomenal actions could have
been different, because they would have been determined by different
causal laws if we had chosen differently.”26 For Vilhauer, the noumenal
agent causes some laws outside of the order of time, and those laws give
rise to actions in time. It follows that the empirical laws governing an
agent’s actions are unknowable because they become laws only when she
decides to adopt them. Therefore, on Watkins’s and Vilhauer’s views, our
actions are in principle unpredictable. For Kant, however, future actions,
qua events or representations, are predictable in principle:
if it were possible for us to have such deep insight into a human being’s cast of
mind, as shown by inner as well as outer actions, that we would know every
incentive to action, even the smallest, as well as all the external occasions
affecting them, we could calculate a human being’s conduct for the future
with as much certainty as a lunar or solar eclipse. (CPrR 5:99; see also A550/
B578)
Although Kant doubts whether we could attain such “deep insight,” the
insight that he refers to has to do with a person’s empirical character,
incentives, and circumstances, not insight into the free decisions of a nou-
menal agent – something that is impossible to achieve. If we are free in the
Kantian sense, it is not because prior events could give rise to different
outcomes or human action is unpredictable. In both cases, determinism
would be false. Kant says that determinism is true, even though it is true
only of appearances.
26
Ibid., 116.
234 M. C. Altman
Kant describes transcendental freedom as “the causality of reason in the

intelligible character” (A551/B579; see also A539/B567). The point of the
Third Antinomy is to show that there is no logical contradiction between
determinism and causation by an uncaused, intelligible cause. The libertarian
goes beyond this in an attempt to explain how uncaused causation is
theoretically possible. Watkins says that persons “in themselves” determine
their empirical characters through an intelligible choice of character. Contra
Watkins, however, Kant says that we cannot offer such an explanation given
our epistemic limits. We can only investigate the effects of whatever I
presuppose, from the practical standpoint, that I am choosing as a noumenal
agent. Earlier I quoted this passage from the second Critique: “how a law can
be of itself and immediately a determining ground of the will (though this is
what is essential in all morality) is for human reason an insoluble problem
and identical with that of how a free will is possible” (CPrR 5:72). Noumenal
causation is theoretically nonsensical. To be morally responsible for an
action, it must be the result of a choice on my part. For the compatibilist,
this is easily explained: my desires and beliefs cause my choice, and my choice
causes the action. The libertarian claims that the action can (must?) be
explained as the result of a choice for which I alone am responsible: the
agent alone causes the action ex nihilo. But in what sense is this an explana-
tion? We must think the thing in itself through the categories, but that gives
us no cognition (Bxxvi). Thus when we say that I, as a noumenal agent, cause
the action, this does not really explain anything, since the category of
causality does not apply to the thing in itself in any theoretical sense –
even if I must think through the category from the practical standpoint.
To take seriously Kant’s commitment to both transcendental freedom
and determinism, the libertarian must assume that the uncaused cause (the
choice) and the caused cause (the preceding physical and mental events)
lead to an identical outcome, that the same action was sufficiently deter-
mined by prior determined events and that it was sufficiently determined
by a free, uncaused choice. However, there is no basis on which to assume
such a correspondence. The libertarian emphasizes that my empirical
character is “determined in the intelligible character” (A551/B579), but
my empirical character is also determined by prior phenomenal causes,
including my upbringing, my other social circumstances, and my physical
constitution (my neurobiology, etc.), all of which themselves have prior
causes – my parents’ upbringing and their parents’ upbringing, the physical
circumstances that led to my brain states, and so on ad infinitum.
Therefore, if I am responsible for my character, I must be responsible for
every event that led to my character, or, as Kant puts it, I must have chosen
“all the past which determined it” (CPrR 5:98). To ensure this correspon-
dence between the action as caused by a set of previous phenomenal events
and the action as caused by an uncaused choice, the noumenal agent
becomes a kind of god who determines the entire course of the universe.
As Walker notes, this leads to “the complete collapse of our ordinary system
of moral evaluations,” since it makes us responsible for everything: not only
the lie I told yesterday, but World War I and the Lisbon earthquake.27 I am
Aristotle’s unmoved mover.
The libertarian very quickly gets into confused territory. The schema of
causality by which the category is applied depends on a “succession of the
manifold” in which one event follows another (A144/B183–84). Since time
does not apply to the thing in itself, what would it mean for the noumenal
choice as an uncaused cause to occur before the action as a phenomenal event?
That would amount to placing both phenomena and noumena along the
same timeline. In addition, thinking of the thing in itself as causing anything,
thus conceiving of it as a noumenon, has only practical use; theoretically, it
misapplies the categories. Although Kant claims to have made room for faith
in freedom by distinguishing phenomena from the thing in itself, what that
leaves us with is a kind of thing for which causation is inapplicable, at least
from a theoretical standpoint. This is the negative sense of noumenon
(A256/B311–12). The libertarian has us making a noumenal decision, but
what would it mean for something, such as deliberation, to happen outside of
time? It could not happen at a particular moment, so in what sense would it
even be a “happening”? It seems that we would have to mark a before and
after to indicate that deliberation has begun and concluded, or that I have
changed my mind, even though there is no noumenal temporality. As
Jonathan Bennett writes, “When Kant says of a noumenon that ‘nothing
happens in it’ and yet that it ‘of itself begins its effects in the sensible world’
(B569), he implies that there is a making-to-begin which is not a happening;
and I cannot understand that as anything but a contradiction.”28 So, there
are deep conceptual problems we face when trying to show how, within a
Kantian framework, uncaused causation is possible. Yet that is what the
libertarian is trying to do.
27
Ralph C. S. Walker, Kant: The Arguments of the Philosophers (London: Routledge, 1978), 149. For a
libertarian response to this objection, see Vilhauer, “Scope of Responsibility,” 45–71. See also Wood,
“Kant’s Compatibilism,” 92–93.
28
Jonathan Bennett, “Kant’s Theory of Freedom,” in Self and Nature in Kant’s Philosophy, ed. Allen W.
Wood (Ithaca: Cornell University Press, 1984), 102.
236 M. C. Altman
Any attempt on our part to explain how freedom works – that is, how an
absolutely free choice outside of time causes an action to occur in time – is
bound to fail. The idea of the thing in itself being uncaused is easy; that is
merely a negative claim. But if free action involves an uncaused cause, then
we are thinking the thing in itself through the category of causality, and
insofar as this is a theoretical explanation about how an apparent action
came to be – which is what Vilhauer and Watkins are both attempting –
then we are misapplying the category. The idea of an uncaused cause is
literally nonsense: taking causality away and reintroducing it in the next
breath.
The practical account

I contend that uncaused causation is a way of thinking about what we do
(through the categories) when we consider our actions from the practical
standpoint. Kant distinguishes between two different kinds of stand-
points, based on whether we are using theoretical reason for the purpose
of cognition or practical reason for the purpose of willing. These two
activities involve very different kinds of justification: the former with
reference to intuitions and concepts, and arriving at truth (A651/B679);
and the latter with reference to the hypothetical and categorical impera-
tives, and arriving at rightness (G 4:417–21). In this section, I will argue
that Kant properly addresses the problem of free will only by maintaining
the separate roles for theoretical and practical reason, and by appealing to
them to validate different things: determinism is warranted from the
standpoint of the rational knower, and freedom is warranted from the
standpoint of the rational agent.
As noted earlier, Kant claims that morally responsible agents must have
the following characteristics:
1) We must be free from determination by material causes, so that we can

choose among different possible options and we could have done other-
wise (choice or Willkür).
2) We must be able to act on a pure concern for the moral law itself (which is
made possible by Wille, or the capacity to act on principles).
3) Our free decision must be effective in determining how we act, so that we
are responsible for our actions (agency, being an uncaused cause, or having
noumenal causality).
However, Kant also asserts that events in the world are subject to strict causal
determinism:
4) All phenomena are determined by prior events, including human actions,
which are explicable in terms of psychological causes that are in time.
Thus our actions are in principle absolutely predictable.
As I have shown, many interpretations of Kant’s theory of freedom ignore or fail to

satisfy some of his philosophical commitments. Compatibilists embrace deter-
minism and define freedom without noumenal causality, and libertarians embrace
uncaused causation and deny that there is complete causal determinism.
These approaches share a common problem, and by diagnosing it we can
proceed to a more fruitful and accurate interpretation of Kant’s theory of
freedom. In the first Critique, Kant is clear about what he is and is not doing
in resolving the Third Antinomy:
It should be noted that here we have not been trying to establish the reality of
freedom, as a faculty that contains the causes of appearance in our world of
sense. . . . Further, we have not even tried to prove the possibility of freedom;
for this would not have succeeded either, because from mere concepts a priori
we cannot cognize anything about the possibility of any real ground or any
causality. . . . [To show] that this antinomy rests on a mere illusion, and that
nature at least does not conflict with causality through freedom – that was the
one single thing we could accomplish, and it alone was our sole concern.
(A557–58/B586–87; see also Bxxix; CPrR 5:72, 94)
Establishing the reality of transcendental freedom would mean inferring its

existence from an apparent action, or discovering a noumenal cause on the basis
of phenomena, by means of the concept of causality. This misapplies an a priori
concept beyond its limited use with regard to appearances. Establishing even
the real possibility of freedom would be improperly making an objective claim
about a possible thing. We would be using the category of causality to make a
claim about the cognitive coherence of uncaused causality, which amounts to a
kind of dogmatic metaphysics (the libertarian view); or we would be represent-
ing free activity as possible properties in the world (the compatibilist view).
Kant is explicit about the limits of what we can do when we think through the
categories: beyond the application of pure concepts to the sensible world, “there
are arbitrary connections without objective reality whose possibility cannot be
cognized a priori and whose relation to objects cannot . . . be confirmed or even
made intelligible” (Pro 4:313). Rather than real possibility or intelligibility,
Kant only tries to establish a more modest, negative claim: if we distinguish
238 M. C. Altman
appearances from things in themselves, freedom is logically possible – that is, the
concept of freedom is not self-contradictory. This allows us to think of our
noumenal choice as causing events in the world because, Kant says, “I can think
whatever I like, as long as I do not contradict myself” (Bxxvi note).
Once Kant establishes the logical possibility of freedom in the Critique of
Pure Reason, he then establishes the reality of freedom as something to which we
are committed in our practical lives. Freedom must be presupposed either
because we act under the idea of freedom (Groundwork) or because we must
think that we are free in order to make sense of moral constraint (Critique of
Practical Reason). And this is where thinking through the categories is warranted
(B166n). We cannot explain how noumenal causality is really possible, because
it does not make sense theoretically to apply the category of causality beyond
the limits of cognition. We conceive of ourselves as noumena only in a practical
sense when we think of ourselves as uncaused causes of our actions. We cannot
have knowledge that we are free and we cannot even understand how we could
be free, but when we act – that is, when we consider our actions from the first-
person, practical perspective – we have to conceive of ourselves as free.
This practical account of Kantian freedom is also called the “regulative
idea theory,” the “deflationary view,” or a kind of “commitment theory,”29
and it is defended by Henry Allison, Christine Korsgaard, Graham Bird, and
Andrews Reath.30 It successfully accommodates both determinism and
uncaused causation in a way that the other views do not. Universal natural
determinism is true, and all of our actions are predictable when we take on the
theoretical standpoint – that is, when we are aiming for knowledge (Wissen) of
our actions as events, making “is” claims, or using constitutive principles. In
addition, we can choose among different options, can determine our actions
on the basis of reasons, and can be motivated by respect for the moral law
when we take on the practical standpoint – that is, when we are deciding what
to do and evaluating our maxims, making “ought” claims, or appealing to
regulative principles. As Bird phrases it, “the argument and acceptance of
transcendental freedom do not establish that our practically free choices are
exempt from causal influences but only that those causal influences can be
29
It is called the regulative idea theory by Hanna, “Freedom, Teleology, and Rational Causation”; the
deflationary view by Vilhauer, “Kant and the Possibility of Transcendental Freedom”; and a commit-
ment theory by Galen Strawson, Freedom and Belief (Oxford: Clarendon, 1986), 61–73.
30
Henry E. Allison, Kant’s Theory of Freedom (New York: Cambridge University Press, 1990); Christine
M. Korsgaard, Creating the Kingdom of Ends (Cambridge: Cambridge University Press, 2000), 159–221;
Graham Bird, The Revolutionary Kant: A Commentary on the “Critique of Pure Reason” (Chicago: Open
Court, 2006), 689–718; and Andrews Reath, “Kant’s Critical Account of Freedom,” in A Companion to
Kant, ed. Graham Bird (Oxford: Blackwell, 2006), 275–90.
practically disregarded.”31 Depending on whether we occupy the theoretical

or the practical standpoint, the same action is regarded either as an event that
is determined by prior events, including prior physical, psychological, and
social causes; or as an action that is the result of freely adopted reasons. On
this view, my choice is timeless not in the sense that something “happens”
outside of time and inserts itself into the world of appearances. Rather,
intelligible agency is timeless because time is applicable only to representa-
tions and not to reasons.
Events in space and time can be explained with reference to causes, but
they cannot be justified by an appeal to reasons: as Kant succinctly puts it,
“Matter causes [wirkt]. Will acts [Willkür handelt]” (OP 21:226). Free
activity is constrained by normative principles, such that we can act on
the basis of reasons that are better or worse. If our actions were nothing but
causally determined events, then such evaluative judgments would be
nonsensical, like asking a rock why it rolled downhill. In justifying my
actions by appealing to normative principles, however, I exempt myself
from empirical laws: “ought expresses a species of necessity and a connec-
tion with grounds which does not occur anywhere else in the whole of
nature” (A547/B575). My choices are not explicable in terms of prior
causes. I act under the idea of freedom because I take myself to be acting
on the basis of reasons, and thus Kant claims that I am bound by normative
principles, including the categorical imperative.
The fact that I think of myself as free from the practical perspective is not
the same thing as believing that I am free. A hard determinist believes that
she is not free, and a compatibilist rejects the idea of an uncaused cause, and
both would assert as much even as they deliberate about what to do. The
hard determinist simply believes that freedom is an illusion, and the compat-
ibilist claims that the feeling of freedom is just the feeling that she is able to
do what she wants, even though she also believes that her goals, desires, and
temperament are aspects of her causally determined character. In fact, Kant
himself believes that our actions are determined, since the category of
causality is a synthetic a priori proposition, strictly necessary and universal.32
His point is that the belief that my actions are determined is irrelevant from
the practical point of view:
31
Bird, Revolutionary Kant, 710.
32
Patrick Kain considers the relation between theoretical and practical belief in “Practical Cognition,
Intuition, and the Fact of Reason,” in Kant’s Moral Metaphysics: God, Freedom, and Immortality, ed.
Benjamin J. Bruxvoort Lipscomb and James Krueger (Berlin: de Gruyter, 2010), 211–30.
240 M. C. Altman
Freedom is . . . practically necessary – thus a human being must act according

to an idea of freedom, and he cannot otherwise. But that does not yet prove
freedom in the theoretical sense. All the difficulties and contradictions that the
concept of freedom has caused thereby fall away. One may prove or also refute
freedom in the theoretical sense, as one wants, nevertheless one will still always
act according to ideas of freedom. There are many people who do not concede
certain propositions in speculation, but still act according to them. (LM
29:898; see also RS 8:13)
The idea of freedom is not a constitutive claim about objects but a regulative
idea that I employ in my thought and action. Thus, Allison says, the idea of
freedom has “normative force” in that, when we act, we must think of
ourselves as rational agents, regardless of whether we believe that we are
free or whether we are free in fact.33 Indeed, as Kant says, even if we could
demonstrate theoretically that there is no such thing as freedom, it would not
undermine the practical point of view.
I must conceive of myself as a moral agent whose actions are determined
by reasons rather than causes – that is, as a result of practical deliberation,
constrained by the categorical imperative. Davidson claims that reasons are a
species of cause: a reason is a belief state that is coupled with a desire (or “pro-
attitude”), and together they cause an action, which is a kind of event.34 One
problem with the causal theory of action is that it applies the concept of
causality in a constitutive way, treating both practical reasoning and the
resulting actions as objects of cognition. By contrast, Kant talks about
noumenal causality by thinking through the categories: from the practical
standpoint, we see rational activity as a cause, even though it cannot be
schematized. Kant would grant that the causal theory of action is true, but
only as a description of comparative freedom. From the theoretical stand-
point, actions are events that are caused by psychological states (beliefs and
desires) in time. From the practical standpoint, actions are the result of free
deliberation that is distinct from causally determined appearances. This is the
compatibility of compatibilism and incompatibilism.
Compatibilists and libertarian incompatibilists attempt either to explain or
explain away this sense of our own indeterminism, but they are both bound
to fail as unified accounts of experience. Theoretically, libertarianism does
33
Henry E. Allison, “Kant’s Practical Justification of Freedom,” in Kant on Practical Justification:
Interpretive Essays, ed. Mark Timmons and Sorin Baiasu (Oxford: Oxford University Press, 2013),
287–88.
34
Donald Davidson, “Actions, Reasons, and Causes,” Journal of Philosophy 60, no. 23 (7 Nov. 1963):
685–700.
not make sense for the very reasons that Kant spells out in the antithesis of
the Third Antinomy. An uncaused cause is contrary to one of the defining
features by which we identify and make sense of the external world. When
pressed to explain an action that is determined by no prior events – apart
from one’s neurobiology, one’s upbringing, one’s social circumstances, one’s
inclinations, and so on – the limits of human understanding leave us with
nothing to say. But when a compatibilist says that our actions are as
predictable as any other event, it is a belief that cannot have any purchase
on us as soon as we make even the simplest decision. As Korsgaard puts it,
“In order to do anything, you must simply ignore the fact that you are
programmed, and decide what to do – just as if you were free.”35 Even
though Kant grants that determinism is true, it is irrelevant and cannot be
taken to be true from the practical standpoint. Or, to speak more precisely,
we can believe it without being able to act on that belief. The idea that one
could have done otherwise even if the action is necessitated by prior events
only makes sense if we distinguish the practical standpoint as a presupposi-
tion that we make in acting from the theoretical standpoint from which we
make conceptually governed, objective claims about the world.
When choosing freely, I identify the resulting event as my action. Because
the action is part of a causal chain of events leading up to it, I relate myself
differently to the past. Rather than a series of events to be discovered and
recorded as objects of knowledge, from the practical standpoint the past
becomes something that I affirm, take on as my own, or take responsibility
for. This is what Kant means when he says that the agent “imputes to
himself . . . the causality of those appearances” through which the past “deter-
mines” the present action (CPrR 5:98). This does not mean that I caused the
Lisbon earthquake. Rather, by participating in the world left to me by the
Lisbon earthquake, I transform the world from the sum total of representa-
tions that make me who I am into a place where I make something of myself
through my choices.36
35
Korsgaard, Creating the Kingdom of Ends, 163.
36
This view has also been voiced by the existentialists. For example, Sartre’s enigmatic claims that I am
“as profoundly responsible for the war as if I had myself declared it” and that “I choose being born” are
not claims that I literally cause these things to happen (Jean-Paul Sartre, Being and Nothingness: A
Phenomenological Essay on Ontology, trans. Hazel E. Barnes [New York: Washington Square, 1956],
709–10). Rather, he means that by acting freely I am choosing to make my way about in a world in
which the war is taking place and I have been born. When I take on certain projects in my life, I am
affirming the past that makes them possible. See Dagfinn Føllesdal, “Sartre on Freedom,” in The
Philosophy of Jean-Paul Sartre, ed. Paul Arthur Schilpp (La Salle, Ill.: Open Court, 1981), 392–407.
242 M. C. Altman
Conclusion: Trafficking in the space of reasons

According to Kant, it is impossible to understand theoretically how freedom
is a real possibility. Yet this is what compatibilist and libertarian interpreters
of Kant’s theory of freedom are trying to do. Like many metaphysicians prior
to Kant, Meerbote, Hanna, and Vilhauer attempt to explain how free activity
could cause motivating desires and the actions that follow. Theoretically, that
amounts to metaphysical speculation. Practically, it is irrelevant. The idea of
freedom is a matter of holding ourselves to account, and this is an unavoid-
able assumption that we make when we act.
In his work on free will, Kant focuses on the practical standpoint and what
it is like to deliberate and make moral judgments. The Groundwork is an
investigation into the kinds of justification that are rational: doing what is
right because it is right, acting on maxims that are shareable by all rational
end-setters, and so on. It would be left to Hegel to show how freedom in this
sense is a historically specific and social phenomenon, but Kant is at least
moving toward what Robert Pippin calls a “relational state” theory rather
than a “causal power” theory of freedom, according to which moral respon-
sibility is a matter of holding one another to account rather than primarily a
metaphysical claim about the source of one’s behavior.37 What is practically
significant is not how noumenal causality can be effective, but how we traffic
in the space of reasons in taking on responsibility for and justifying our
actions.38
37
Robert B. Pippin, Hegel’s Practical Philosophy: Rational Agency as Ethical Life (Cambridge: Cambridge
University Press, 2008).
38
I am indebted to Cynthia D. Coe, Michael Fletcher, Wayne P. Pomerleau, and Benjamin Vilhauer for
reading early drafts and suggesting promising directions for this chapter.
11
Moral Skepticism and the Critique of
Practical Reason
David Zapero
It was not until after Kant wrote the Critique of Pure Reason that he
considered the possibility of extending his transcendental inquiry to moral
philosophy. In fact, the first important review of the Critique, written
anonymously by Christian Garve for the Göttinger Gelehrten Anzeigen, played
a decisive role in this respect. In the course of responding to Garve, Kant
came to realize that moral philosophy could be dealt with in the framework
that he had until then only applied to theoretical philosophy.1 Since the
moral law – Kant argues – must be a synthetic a priori principle, the task of
moral philosophy is analogous to the task of theoretical philosophy: its
central question is also how a synthetic a priori principle is possible.
Indeed, when Kant went on to write the Groundwork shortly thereafter, he
emphasized that parallel by dealing with the “how possible” question in a
way that was clearly supposed to be analogous to the way he deals with it in
the first Critique (G 4:453–55).
Yet, it is far from evident what that parallel amounts to. The principles of
pure theoretical reason and the moral law may both be synthetic and a priori,
but they are inevitably quite different kinds of principles. Put in the broadest
1
See Henry E. Allison, Kant’s “Groundwork for the Metaphysics of Morals”: A Commentary (Oxford:
Oxford University Press, 2011), 53–69; and Eckhart Förster, The Twenty-Five Years of Philosophy: A
Systematic Reconstruction, trans. Brady Bowman (Cambridge: Harvard University Press, 2012), 41–56.
D. Zapero (*)
Department of Philosophy, University of Bonn, Bonn, Germany
e-mail: davidzaperomaier@googlemail.com

244 D. Zapero
and most neutral way possible: whereas the former concern how things are,
the latter concerns how things ought to be. So it is not at all obvious how an
inquiry seeking to show that objects of experience are indeed bound by
certain laws could be of use when it comes to dealing with how we ought
to act. One may of course say that in both cases the crucial issue is the
bindingness of a priori principles – but the kind of bindingness that is at
stake is quite different in each case. Or so it would seem. The objects of
experience abide by laws that they cannot infringe upon, whereas we are
bound by a law that we can (and do) breach. So it is not at all obvious what
the “how possible” question amounts to in the context of practical philoso-
phy. If the moral law is indeed a synthetic a priori principle, what does it
mean to show “how it is possible”?
Of course, there is also much dispute about what that question
amounts to in the context of theoretical philosophy. There is, clearly,
no consensus as to what Kant thinks he has established in showing that
the categories of the understanding must be “objectively valid for experi-
ence to be possible” (leaving aside the issue of whether he manages to
establish that, or how he manages to do it). Despite these controversies,
we have a fairly clear sense of the general purpose of his enterprise. That
is mainly due to the fact that Kant gives us quite explicit indications
about who his main interlocutor is and what kind of challenge that
interlocutor raises. In the context of practical philosophy, however,
things are somewhat different. In that context, we do not have one
single interlocutor or one group of interlocutors that plays the role that
Hume does in the case of theoretical philosophy. Kant at least does not
present us explicitly with any such interlocutors. And when it then
comes to the central treatment of the “how possible” question, it is not
clear just what kind of challenge Kant is responding to.
Most commentators believe that, at least by the time he wrote the second
Critique, Kant had quite modest aims. That is, they believe that he does not
seek to address any traditional skeptical challenges about morality. The
skeptic’s questions about why we should be moral and act as the moral law
commands are not, according to them, important questions for Kant. Kant
seeks instead to flesh out his system of morality and to show its persuasive-
ness by doing so.2
2
See, most notably, Henry E. Allison, Kant’s Theory of Freedom (Cambridge: Cambridge University
Press, 1990), 214–49; Karl Ameriks, “Kant’s Deduction of Freedom and Morality,” Journal of the
History of Philosophy 19, no. 1 (Jan. 1981): 53–79; Dieter Henrich, “Der Begriff der sittlichen Einsicht
und Kants Lehre vom Faktum der Vernunft,” in Die Gegenwart der Griechen im neueren Denken:
In the present chapter, I argue against these views and show that – and
how – Kant engages with skeptical worries about morality in the second
Critique. The fact that he does not deal with those issues separately – at least
not in detail – does not imply that he does not deal with them at all. In fact,
his treatment of those issues, I will argue, is central to the second Critique,
and recognizing the role it plays in the overall argument of the Critique is
indispensable even for understanding Kant’s normative ethics.
To show this, I will begin by looking at the textual evidence that is usually
appealed to when making the case that Kant does not engage with skeptical
questions about morality. I show that while he does deny that it is possible to
give a proof of the authority of the moral law, he at the same time provides
clear indications that he has resolved the issue as if he had been able to
provide a proof. In the second section, I sketch an argumentative strategy
that accounts for this peculiar stance: I show how the absence of a separate
treatment of the legitimacy problem can be explained by the fact that that
problem is already dealt with in the context of the presentation of the moral
law. It is in the presentation of the mere idea of that law, I argue, that Kant
already addresses the skeptic’s question of what reason we have to obey the
moral law. It is to the exposition of the moral law that we must turn to find
Kant’s treatment of the skeptic’s challenge. In the third section, I provide
some detailed textual evidence for my interpretation. Finally, I conclude with
some reflections about the most particular feature of Kant’s argumentation,
namely the idea of a practical standpoint.
“Pure practical reason proves its reality by what it

does”: Some preliminary textual evidence
One of the most puzzling claims of the Critique of Practical Reason is made
right at the outset of that work in the context of an attempt to explain its very
subject matter. Since the purpose of the first Critique had been to examine
the legitimacy of the claims made by pure reason in its theoretical use, one
would have expected the second Critique to do the same thing for the
Festschrift für Hans-Georg Gadamer zum 60. Geburtstag, ed. Dieter Henrich, Walter Schulz, and Karl-
Heinz Volkmann-Schluck (Tübingen: Mohr, 1960), 77–115; John Rawls, Lectures on the History of
Moral Philosophy, ed. Barbara Herman (Cambridge: Harvard University Press, 2000), 235–72; and Jens
Timmermann, “Reversal or Retreat? Kant’s Deductions of Freedom and Morality,” in Kant’s “Critique of
Practical Reason”: A Critical Guide, ed. Andrews Reath and Jens Timmermann (Cambridge: Cambridge
246 D. Zapero
practical use of pure reason. One would thus have expected for the second
Critique, which extends the project of critical philosophy to practical reason,
to constitute a Critique of Pure Practical Reason, and not merely a Critique of
Practical Reason. (The first Critique was after all a critique of pure theoretical
reason – “theoretical” was not included in the title simply because at that
time no sequel was planned.)
To account for this fact, Kant introduces the following claim at the
beginning of the preface. The particular form of the second Critique, we
are told, can be explained by a peculiar feature of practical reason itself: there
is no need to criticize the pure part of practical reason in order to determine
whether it “presumptuously oversteps itself (as does happen with speculative
reason),” because if one succeeds in showing “that there is pure practical
reason,” this already is sufficient; pure practical reason “proves its reality
and that of its concepts by what it does, and all subtle reasoning against
the possibility of its being practical is futile” (CPrR 5:3).
This particularity of pure practical reason – that it somehow “proves its
reality . . . by what it does [durch die Tat]” – explains, we are told, why the
focus of the second Critique is different from that of the first. Since pure
theoretical reason makes illegitimate claims and gives rise to aporia, it is
necessary to examine separately the legitimacy of all its claims, so as to
determine which ones are legitimate and which are not. In the case of
practical reason, however, this will not be necessary. Since the legitimacy of
its claims is somehow assured by its practical nature, the critique will focus
on the task of delimiting pure practical reason – and, therefore, it will deal
with practical reason in general.3
At this point, we are given no details about how exactly pure practical
reason achieves the feat of “proving its reality”; we are merely told that it does
this durch die Tat, “by what it does.” But we are given a clear sense of the
magnitude of that feat. Kant suggests that, in the study of practical reason, it
is not necessary to deal separately with the question of the legitimacy of its
principles because that question somehow takes care of itself. In other words,
there is no need to deal with the problem that a transcendental deduction
would have dealt with, namely the problem of the objective reality of its pure
3
Throughout this chapter, I will use “objective reality” and “objective legitimacy” (or “objective
validity”) synonymously. In doing so, I take myself to be faithful to the way that Kant uses those
terms in both the first and second Critiques. (Compare for instance his use of “objective validity
[objektive Gültigkeit]” at CPrR 5:46 with his use of “objective reality [objective Realität]” at CPrR
5:47.) For a different view on this matter, see Henry E. Allison, Kant’s Transcendental Idealism: An
Interpretation and Defense (New Haven: Yale University Press, 1983), 133–36.
principles. Now since it turns out in the course of the Critique that the
supreme principle of pure practical reason is the moral law, Kant’s remark is
even more significant. It suggests that the problem of establishing the validity
of that law will not be necessary. It will not be necessary because its mere
presentation – the mere presentation of pure practical reason – will already
take care of the matter.
Now when it actually comes to the legitimacy problem in the second
Critique, matters take another quite unexpected turn, since Kant there
confronts us with the claim that a transcendental deduction of the moral
law is not possible. That is, it turns out that the kind of proof that was
provided for the principles of pure theoretical reason cannot be provided in
the case of pure practical reason:
With the deduction, that is, the justification of [the moral law’s] objective and
universal validity and the discernment of the possibility of such a synthetic
proposition a priori, one cannot hope to get on so well as was the case with the
principles of the pure theoretical understanding. . . . The objective reality of the
moral law cannot be proved by any deduction, by any efforts of theoretical
reason, speculative or empirically supported, so that, even if one were willing to
renounce its apodictic certainty, it could not be confirmed by experience and
thus proved a posteriori. (CPrR 5:46–47)
These claims about the impossibility of providing a transcendental deduction

of the moral law have led many commentators – indeed, most contemporary
commentators – to conclude that Kant renounces the attempt of justifying
the moral law. And they have good reasons for drawing this conclusion.
Given that Kant takes the moral law to be a synthetic a priori principle, and
given that the task of proving the legitimacy of such a principle is dealt with
in a transcendental deduction, the impossibility of providing a deduction is
tantamount to the impossibility of proving the legitimacy of the moral law.
However, it is also important to note that, while Kant rules out the
possibility of providing a transcendental deduction – and thus a proof of
the legitimacy – of the moral law, there are numerous indications that he
considers that that problem has somehow nevertheless been successfully
resolved. In fact, the infamous passage that I have just cited is a case in
point. For the last sentence ends with the remark: “and [the moral law] is
nevertheless firmly established of itself” (CPrR 5:47). So, while he insists on
the impossibility of providing a proof in the first part of the last sentence, he
goes on to maintain that the law is “firmly established” – and that suggests
that, despite the lack of proof, its legitimacy has been proven. Indeed, if one
248 D. Zapero
reads the whole passage, the tone of those oft-quoted sentences appears in a
somewhat different light than when one reads them out of context. It seems
as though Kant is claiming that, although we cannot provide a transcendental
deduction, the problem of legitimacy has been dealt with in a fully satisfac-
tory way.
There are several other places in which Kant makes assertions that go in
the same direction. Commenting on the doctrine of the fact of reason – to
which we will come in the next section – he tells us: “The objective reality of
a pure will or, what is the same thing, of a pure practical reason is given a
priori in the moral law, as it were by a fact” (CPrR 5:55). Here Kant is much
more specific than in the previous passage. The statement suggests quite
clearly that the goal of a transcendental deduction – establishing the objective
reality of the moral law – has been achieved in some other way, namely by
means of some kind of fact. Somewhat less specific but nonetheless compel-
ling is a statement made even before the deduction issue is raised, when Kant
presents what has been achieved in the Analytic. He tells us that the Analytic
has been able to show that “pure reason can be practical” – and he then adds
that this has been shown “by a fact in which pure reason in us proves itself
actually practical” (CPrR 5:42). While less specific, the statement is also
striking since Kant does not hesitate in referring to a proof that shows that
pure reason is indeed practical.
We are thereby left with a quite remarkable dilemma. On the one hand,
we are told explicitly that a deduction of the moral law is impossible – and
since a deduction is the only kind of proof that seems appropriate to establish
the legitimacy of the moral law, this amounts to saying that we cannot
provide a proof of the legitimacy of that law. On the other hand, we are
told in no uncertain terms that precisely that task has been resolved – indeed,
that it has been resolved in an entirely satisfactory way, as if one had managed
to provide a deduction. In other words, we are told both that a proof of the
legitimacy of the moral law is impossible and that that legitimacy has been
established. Yet, how could it be established except by a proof (that is, a
transcendental deduction)?
The most intuitive way of resolving this dilemma is by rejecting one of its
two horns – that is, either by questioning that a proof is impossible or by
questioning that the legitimacy of the moral law has been proven. Now since
rejecting the first horn is highly problematic, given Kant’s explicit statements
on the impossibility of providing a deduction, one can ultimately only reject
the second horn of the dilemma. That is indeed what most commentators
have sought to do. They downplay Kant’s claim that he has established the
legitimacy of the law without a deduction, and they do so mainly in one of
two ways. They either seek to show that Kant renounces the effort of proving
what a deduction would prove and contents himself with a more modest
solution; or they seek to show that he is not particularly concerned about the
challenges that a deduction would raise: by accepting the impossibility of
providing such a proof, he thus does not make what he considers to be a
significant sacrifice.
There is, however, another way of coming to terms with this difficulty.
One can confront it head on by reflecting on what seems to be the flagrant
contradiction: does the impossibility of a proof necessarily exclude the
possibility of establishing the legitimacy of the moral law? Of course, if the
“establishing” at stake in the second part of the sentence is the same as the
“proving” in the first part, there is clearly a contradiction. But it is far from
clear that that is indeed the case. That is what I have tried to convey by
focusing on the opening remarks of the Critique. What Kant clearly denies is
that the legitimacy of the moral law can be established in a separate treatment
of the matter. If “proof” means an independent argument showing that the
moral law is legitimate, then providing a proof is indeed excluded. But that
does not mean that the issue is not dealt with earlier than expected, that is,
before Kant raises the question of whether a proof is possible. In other words,
one could resolve the difficulty by showing that no full-blown proof is
necessary because the legitimacy is dealt with by a prior step in the argument.
Indeed, that is the claim that I want to defend in what follows. I will show
that the issue of the legitimacy of the moral law is dealt with – and, in Kant’s
eyes, resolved – in the process of presenting the mere idea of that law. To do
so, I will take a stance on a number of controversial interpretative issues
without being able to give any detailed defense of my positions. But this
somewhat dogmatic presentation will allow me to bring into view a possibility
that is usually neglected or overlooked – namely, the possibility that Kant
should be dealing with the legitimacy issue before he even raises the problem
of the deduction. The task of convincing the reader of my interpretation will
be left to the third section. There I will focus on how the argumentative
strategy outlined is presented by Kant in the second Critique.
The bindingness of reason’s form

Since one of the main tasks of Kant’s moral philosophy is to flesh out the idea
of an unconditionally binding imperative, Kant often emphasizes the differ-
ences between such an imperative – a categorical imperative – and a
250 D. Zapero
hypothetical one. That is, he often emphasizes that the former is a “practical
law” binding the actions of all rational beings, whereas the latter is a
conditional principle whose authority depends on certain conditions being
fulfilled. This insistence on what separates those two kinds of imperatives
can, however, easily lead one to lose sight of what they share, that is, what
makes them both imperatives. Indeed, their being imperatives does not
simply reduce to their being prescriptions. The notion of an imperative is a
more specific one for Kant; it is tied to the idea of a distinctly normative kind
of necessity or constraint.
At the beginning of the second Critique, imperatives are characterized in
the following way:
for a being in whom reason quite alone is not the determining ground of the
will, [a practical] rule is an imperative, that is, a rule indicated by an “ought,”
which expresses objective necessitation [Nötigung] to the action and signifies
that if reason completely determined the will the action would without fail take
place in accordance with this rule. (CPrR 5:20)
Practical principles are principles of reason, we are told; and with respect to
finite beings who are not entirely rational, those principles possess a certain
“objective necessitation,” since they express something that such beings ought
to do, but that those beings will not always end up doing. Kant thus defines
the “ought” of practical principles by appealing to reason: he maintains that
the kind of constraint that those principles exert on us is dependent on us
being rational beings.
He is hereby introducing one of the central claims of the second
Critique, namely the claim that the bindingness of practical principles
is always grounded in a certain rational stance, one that he designates by
the term “willing.” The significance of that claim can be grasped by
focusing on just what it excludes – as Kant does in the first few
paragraphs of his book. By insisting on the idea that practical principles
must be grounded in a certain rational stance, Kant seeks primarily to
exclude the possibility that they be grounded in attitudes such as desiring
or wishing. Of course, the fact that I desire some end can also involve a
certain constraint. Indeed, it can lead me to actually pursue that end.
But Kant is attempting to distinguish that kind of efficacious constraint
from another type of constraint, namely a normative one. The constraint
or necessitation that he is interested in is one where I ought to pursue a
certain end – and where my being so obligated leaves open the question
of whether I will actually pursue that end.
For one to be subject to that kind of constraint, Kant claims, it is

insufficient simply to desire an end. My desiring or wishing an end is a
mere fact devoid of any normative significance. Of course, for me, my
desire or wish may be pressing, and it may lead me to pursue the relevant
end. But that kind of compulsion is of a fundamentally different nature
from the one that is at stake in a commitment. Kant thus calls the first
kind of necessity a “subjective necessity,” and he contrasts it with the
“objective necessity” that a commitment involves. In committing to
something or willing it, one submits to the demand that one pursue
the relevant end. Indeed, committing just is committing oneself to
whatever is necessary to achieve that end. It may of course turn out, if
I do not pursue the end, that I was not really committed to it; but that
simply reveals something about my lack of self-knowledge, not about the
nature of committing. The important point is that if I committed to
something, my failure to do what I committed to will be precisely that –
a failure. And Kant is claiming that one cannot say the same in the case
of desiring or wishing. My desiring or wishing something does not entail
the kind of “ought” that can allow for such a failure. Naturally, I may
“want” to pursue the desired end – “want” in the sense of “willing” or
“committing” – and in that case it would indeed be a failure if I do not
pursue the end. But the desiring or wishing has, in and of itself, no
normative force. If I do things differently, I have not failed to live up to
a norm.4
Kant’s treatment of hypothetical imperatives is instructive because it
brings out this issue particularly well. Even when he seeks to oppose
hypothetical to categorical imperatives, Kant points out en passant that the
hypothetical imperative too must be a rational principle: “Reason, from
which alone can arise any rule that is to contain necessity, does indeed put
necessity even into this precept (for otherwise it would not be an imperative),
though it is only a subjectively conditioned necessity” (CPrR 5:20). That is,
since hypothetical imperatives are also imperatives and also possess the dis-
tinctive kind of normative necessity that I have talked about, they are also
grounded in reason. Even though their necessity is “subjectively condi-
tioned,” because it depends on some prior condition being fulfilled (namely,
on the agent being committed to some particular end), they have the kind of
4
On this point, see Christine M. Korsgaard, “The Normativity of Instrumental Reason,” in The
Constitution of Agency: Essays on Practical Reason and Moral Psychology (Oxford: Oxford University
Press, 2008), 27–68; and Robert B. Pippin, Hegel’s Practical Philosophy: Rational Agency as Ethical Life
252 D. Zapero
“objective necessity” that is characteristic of practical principles. Since Kant

considers that that kind of necessity is inextricably associated with a certain
rational stance – willing or committing – he makes sure to point out, even in
the case of this “inferior” sort of imperative, that it too is grounded in that
kind of a stance.
Now these considerations on the specificity and grounding of practical
principles in general have direct consequences for the question of what
shape an unconditionally binding principle can have. Indeed, if one
accepts the considerations just sketched, it turns out that an uncondi-
tionally binding practical principle will have to satisfy at least two
conditions. First, simply in virtue of being a practical principle, that is,
in virtue of possessing the kind of necessity characteristic of such prin-
ciples, it will have to be grounded in willing. Given that – according to
the take on the opening paragraphs of the second Critique that I have
presented – any practical constraint is tied to the rational stance that
Kant denotes by “willing,” even an unconditionally binding practical
principle will have to be grounded in such a stance. Yet, since it is an
unconditional principle that binds all maxims, its bindingness cannot
depend on the features of any particular act of willing. So the principle
– if indeed there is such a principle – will also have to satisfy a second
condition. If it is to hold for all maxims, it must be grounded in a
feature that is also characteristic of all acts of willing. Since willing is the
only possible source of the bindingness of a practical principle, a prin-
ciple that binds all acts of willing will also have to be one that is
grounded in a feature that is common to all such acts. It will have to
be a condition that one submits to simply by willing anything at all.
That, I am suggesting, is what Kant means when claiming that the moral
law must be the form of practical reason (CPrR 5:27). The claim is that any
practical principle that aspires to be unconditionally binding must abstract
from all the features of any particular act of willing. It must do so because, as
long as its authority depends on such particular features (the “content” of
willing), that authority will necessarily be restricted to the acts of willing that
have those features. Since the only possible source for the distinctive kind of
constraint characteristic of practical principles is the act of willing, a principle
that binds all acts of willing must also only depend on something that is
common to all such acts. It is only something common to all and any such
act (that is, the “form” of practical reason) that could possibly constitute the
source for an unconditional authority.
In making this claim, Kant does not yet take a stance on whether there is
indeed such a principle or not. Such a claim is at least not meant to involve a
stance on that question. It is simply meant to flesh out the mere idea of an
unconditional principle by specifying some conditions that any possible
candidate must fulfill. The considerations that we just presented thus belong,
to put it in Kant’s terms, to the mere exposition of the moral law. However,
once one has spelled out the relevant conditions, one is forced to admit that
they are not in fact entirely neutral with respect to the legitimacy issue. That
is, these considerations about the mere idea of an unconditionally binding
practical principle already lead us to questions concerning the reality of such
a principle. For it turns out that the authority of such a principle has a very
particular relation to the act of willing. If there is indeed some condition that
one must satisfy to will anything at all, then that condition will in fact be the
unconditional practical principle that we are looking for.
Here is not the place to discuss Kant’s claim that the form of willing is the
universal law formula which he presents in §7 of the second Critique. What is
of interest here is the role that claim plays in the overall argumentative strategy
that I have outlined. Given the claims that – according to my interpretation –
Kant has introduced before that point in the second Critique, Kant can
conclude, merely by presenting it, that the form of practical reason is in fact
unconditionally binding. For he has already shown that if there is indeed such
a form, that form will constitute an unconditionally binding principle. When
Kant then goes on to present that form, he can – without taking any further
steps – claim that that form is in fact unconditionally binding. The mere
“exposition” of that principle suffices to show that all acts of willing are bound
by that very principle.
One may of course object that, on the account that we have given, the
authority of such a principle is inevitably conditional. One may say: even if
one accepts the argumentative steps that I have outlined, and even if one
accepts Kant’s claim that the universal law formula is the form of practical
reason, one is left with a principle that is only binding if we will something.
The conclusion of the argument thus falls short of what would be needed to
show that the principle in question is unconditionally binding.
Yet, the fact that its authority is tied to the activity of willing does not
make that authority a conditional one. Since it is a practical principle
that we are concerned with, that is, a principle that binds our maxims,
its authority is in no way restricted by the fact that it is “only” binding
when we adopt a maxim. Indeed, it is not meant to be binding in any
other respect. In fact, this dependence on the activity of willing is
precisely what ensures its unconditional bindingness. It ensures that
whatever one wills, one’s maxim will be bound by the principle in
question.
254 D. Zapero
Of course, one may have preferred an argument that allows us to ground

the authority of the moral law on some fact of the matter. And this is indeed
what Kant unequivocally excludes. The kind of fact of the matter that could
serve as a grounding for the authority of the moral law – our capacity for
spontaneity – is inaccessible to us, we are told. Trying to investigate that
capacity would entail overstepping the limits to knowing that have been fixed
by the first Critique. Yet, it is far from clear that this limitation bars Kant
from providing an argument that establishes the authority of the moral law as
if that authority had been established by the proof we may have hoped for.
The argumentative strategy that I have sketched does make the authority of
the law dependent on the actual activity of willing. It is only “insofar” as we
will something that we are bound by the principle in question: on the
explanation that I have provided, its authority comes into play by being a
condition that we have to submit to in order to will anything at all. But given
that it is a practical principle that could not possibly extend to anything else
but the activity of willing, this does not entail any limitation.
Indeed, the difference between the argument that I have sketched and the
argument that one would have perhaps expected is a difference in standpoint.
The argument that I have sketched adopts a particular standpoint: it does not
seek to establish the authority of the moral law from a third-person perspec-
tive, by appealing to some fact of the matter, but it tries to establish that
authority by adopting the point of the view of an agent. Rather than
appealing to some standard that is independent of the activity of willing, it
seeks to show that there is a demand “internal” to willing – internal to any
willing. In other words, the argument that I have sketched does not adopt a
theoretical standpoint, but it adopts the standpoint of an agent, that is, a
practical standpoint.
By taking up such a standpoint, Kant’s argumentation ends up having a
quite peculiar shape. What it would take two steps to do in a theoretical
proof is done in just one. There is no need, from a practical standpoint, to
first present the mere idea of a moral law and then establish its authority, that
is, to show why we should obey it. Since that law is presented as a demand
that is constitutive of willing, that is, as a normative constraint on agency, the
mere exposition of that idea already allows us to establish what Kant calls its
reality. Of course, if one was expecting a theoretical proof, one may refuse to
talk about a proof here, since the authority of the moral law has not been
dealt with separately. Yet, the fact that it has not been dealt with separately
does not mean that it has not been dealt with at all – or even that it has been
dealt with in a less satisfactory manner. The argument that I have sketched
does not deal separately with the question of why we should obey the moral
law – but it does not do so because it does not need to. And it would be a
mistake to take the absence of a separate treatment of that question as a
reason to think that that question has not been given treatment at all.
Indeed, it has – and the way that it is dealt with shows that Kant is
addressing a skeptical worry about morality. Kant is addressing the question
of why one should abide by the moral law, that is, what reason one has to do
as it commands. His strategy, I have suggested, consists in claiming that the
reasons to be moral are built into the standpoint that we adopt qua agents.
He seeks to show that the moral law is a requirement that we have to
conceive ourselves as submitted to merely in order to adopt the point of
view that is characteristic of agency. Willing anything at all, he maintains,
already requires that one have accepted the authority of the demand that is
given expression in the formula of universal law. To defend that claim,
however, he need not provide a separate proof of the authority of the
moral law. Instead, he only needs to unfold his notion of willing and agency.
It is in the presentation of those notions that one finds his treatment of the
skeptical challenge.
It is to this “unfolding” that I will turn now, in order to give some textual
evidence for the interpretation that I have outlined.
“Merely insofar as they have a will”: The doctrine

of the fact of reason
One of the most obscure passages of the second Critique follows Kant’s first
explicit presentation of the moral law. After having arrived at the formula of
universal law in §7, he goes on, in the first remark, to comment on the
unconditional authority of that principle. The remarks are meant to help
clarify the mere idea of that principle – and yet, after a few sentences, the
tone of the explanation changes and one gets the impression that Kant is not
only talking about the notion of the moral law but about the actual authority
of that law, as if the issue of its authority had somehow already been dealt
with. In case there are any doubts about the transition that he is making at
this point, Kant goes on to address the issue explicitly:
Consciousness of this fundamental law may be called a fact [Faktum] of reason

because one cannot reason it out from antecedent data of reason, for example,
from consciousness of freedom (since this is not antecedently given to us) and
because it instead forces itself upon us of itself as a synthetic a priori proposi-
tion that is not based on any intuition, either pure or empirical, although it
256 D. Zapero
would be analytic if the freedom of the will were presupposed; but for this, as a
positive concept, an intellectual intuition would be required, which certainly
cannot be assumed here. However, in order to avoid misinterpretation in
regarding this law as given, it must be noted carefully that it is not an empirical
fact but the sole fact of pure reason which, by it, announces itself as originally
lawgiving (sic volo, sic jubeo). (CPrR 5:31)
What I have been suggesting in the previous section is that Kant, by means of
this fact of reason, seeks to make explicit the full import of the presentation
of the moral law that has preceded this passage. That is, it seeks to make
explicit how the mere presentation of the moral law already dealt with the
question of the validity of that law.
In this section, I seek to show that this is indeed what Kant does, both in
the passage that I have just quoted and the ones that follow it. Once Kant has
succeeded in “extracting” the form of practical reason, he goes on – I will
argue – to make explicit the authority that the relevant principle has merely
in virtue of being the formal principle that it is. Doing so primarily involves
pointing to the way in which that authority is related to the activity of
willing, that is, to the fact that the principle’s authority is a requirement that
one must submit to merely to be able to take up the standpoint of an agent.
There are of course various elements in the passage above that can seem to
point in a different direction. The most significant one is probably Kant’s
gloss on the fact of reason as a consciousness of the practical law. The phrase
suggests that Kant’s appeal to such a fact is an appeal to some sort of explicit
or implicit awareness of the moral law. It seems, in other words, that Kant is
appealing to our common moral consciousness. However, if we look at
another use he makes of that expression a couple lines earlier, in the remark
to §6, it turns out that it is meant in quite a different way. At that point, we
are told that we are “immediately conscious” of the moral law “as soon as we
draw up maxims of the will for ourselves” (CPrR 5:29). We could see this too
as an appeal to ordinary moral consciousness – but it is at any rate remark-
able, to say the least, that Kant should insist on the close relationship
between the consciousness of the moral law, on the one hand, and the
adoption of a maxim, that is, the activity of willing, on the other.
This is, of course, not the place to discuss the details of one of Kant’s most
controversial notions, namely his notion of consciousness. Since it is often
taken as decisive evidence for a certain widespread interpretation of the fact
of reason, I simply seek to point out that the appeal to a “consciousness of the
law” need not be an appeal to some kind of ordinary moral consciousness at
all. In fact, it is quite compatible with an “adverbial” conception of
consciousness according to which “consciousness” refers to the way in which

we perform a certain activity.5 That is, the consciousness of the law can refer
to the fact that adopting a maxim must be done in a certain way, namely that
willing requires that one conceive oneself as subject to the moral law. If that
were indeed so, then it would point to the relation – between the activity of
willing and the authority of the moral law – that I have claimed allows Kant
already to proclaim the legitimacy of the moral law at this place in the second
Critique.
Indeed, the remarks following the ones that I have quoted point in exactly
the same direction:
The fact mentioned above is undeniable. One need only analyze the judg-
ment that people pass on the lawfulness of their actions in order to find that,
whatever inclination may say to the contrary, their reason, incorruptible and
self-constrained, always holds the maxim of the will in an action up to the
pure will, that is, to itself inasmuch as it regards itself as a priori practical.
(CPrR 5:32)
Since Kant appeals here as well to ordinary moral judgment, the passage has
mostly been taken as further evidence for the claim that he seeks to ground
the authority of the moral law on our ordinary moral consciousness. It is
important, however, to begin by looking at the real focus of the passage. Kant
indeed appeals to an ordinary moral judgment, but he does so to bring out
the content of that judgment, namely the fact that reason always holds up our
maxims to the requirements of the moral law. So the focus of the passage is
still on the close relation between the act of adopting a maxim and that
maxim’s being subject to the moral law.
Of course, one may also explain Kant’s appeal to that relation as a way of
insisting on the unconditional authority that is constitutive of the mere notion
of a moral law. There is nothing unusual about Kant pointing to that relation,
one may object: the moral law is supposed to be one that all maxims are subject
to. But the context of these remarks clearly suggests that in appealing to the
unconditional character of the moral law, Kant seeks to explain the moral law’s
legitimacy. As I have tried to show, pointing out that maxims are inexorably
subject to the moral law can do significant argumentative work here because it
brings into view the fact that that law is a condition that is “intrinsic” to
5
For such a view, see Robert B. Pippin, “Kant on the Spontaneity of Mind,” Canadian Journal of
Philosophy 17, no. 2 (1987): 449–75.
258 D. Zapero
willing; that is, the law is a condition that – due to its formal character – one
must submit to simply in order to will anything at all.
It is thus no surprise that Kant should appeal to ordinary moral judgment.
The important point for him is not whether we in fact, empirically, always
(implicitly or explicitly) judge in a certain way or “hold up” our maxims to
the requirements of the moral law. The point is rather that the tendency to
do so should come as no surprise. If we usually hold up our maxim to the
demands of the moral law, “whatever inclination may say to the contrary”
(CPrR 5:32), that is because the very act of adopting a maxim implicitly
already requires us to do so. The fact that we should have the tendency to
judge in that way reflects the fact that, ultimately, we must judge in that way
– since such a comparison is already presupposed by the very adoption of a
maxim.
This point also remains center-stage in the rest of the remarks on the fact
of reason:
Now this principle of morality, just on account of the universality of the

lawgiving that makes it the formal supreme determining ground of the will
regardless of all subjective differences, is declared by reason to be at the same
time a law for all rational beings insofar as they have a will, that is, the ability to
determine their causality by the representation of rules, hence insofar as they
are capable of actions in accordance with principles and consequently also in
accordance with a priori practical principles (for these alone have that necessity
which reason requires for a principle). (CPrR 5:32)
Since Kant here ties the authority of the moral law to the idea of the will
(“this principle of morality . . . is declared by reason . . . a law for all
rational beings insofar as they have a will”), many interpreters have
taken this to be evidence that the demonstration of the legitimacy of
the moral law depends on an ulterior treatment of the freedom of the
will.6 But if one looks more closely at Kant’s formulation, one sees that
he is rather making the opposite claim. For the reference to the will is
qualified by überhaupt (so fern sie überhaupt einen Willen, d.i. ein
Vermögen haben) and thus should rather be translated as: “merely insofar
as they have a will.” That is, the possession of the will is not so much
presented as a separate condition that must be satisfied for the authority
of the moral law to hold. Instead, the will is being referred to as a
capacity that a rational being would have anyway. Indeed, it should
6
See, for instance, Allison, Kant’s Theory of Freedom, 238–41.
come to no surprise that Kant is not worried here about the question of
whether we actually possess a will or not. From a practical standpoint
that issue is resolved from the outset: the argumentation is being con-
ducted from the standpoint of an agent, or a being that wills. And Kant
is insisting on the fact that merely in virtue of adopting that standpoint,
merely insofar as we have a will, we are subject to the moral law. Our
subjection to that law is not something that must be figured out inde-
pendently of the mere presentation of the moral law because that law
constitutes a requirement that we have to conceive ourselves as submitted
to in order to adopt the perspective of an agent, that is, in order to will
anything at all.
It is of course highly significant for Kant that we cannot provide a proof of
the moral law that is analogous to the proof provided in the first Critique.
When he then reaches the point at which he would have to provide such a
deduction, he does not seek to play down the significance of the fact that a
deduction, in the case of practical reason, is unfeasible. Yet, given the results
of the previous section of the book (the section leading up to §7), one would
expect for Kant to claim that the moral law’s authority has been firmly
established, although no proof of that authority can be provided. That is,
one would expect for him say two seemingly contradictory things: that we
cannot provide a proof of the moral law’s authority and that that authority
has been firmly established. This is, as I tried to show in the first section,
precisely what he ends up doing. He insists on the fact that a proof – in fact,
any separate treatment of the issue – is not possible, and at the same time he
maintains that the issue has been resolved in an entirely satisfactory way.
Conclusion
Kant’s treatment of the question of why we should abide by the moral law is
peculiar, I have tried to show, because it relies on a distinction between a
theoretical and a practical standpoint. Indeed, it is because one expects his
treatment of that question to be conducted from a theoretical standpoint that
it is easy to overlook that he deals with the question at all. A proof from a
theoretical standpoint establishes the authority of the moral law from a
perspective that is entirely external to that authority. And it is quite natural
to consider that such a standpoint is the only standpoint from which a proof
can be conducted. If the proof in some way presupposes the legitimacy that it
is meant to establish, one may insist, it would be circular and would hardly
260 D. Zapero
be a proof at all. Any form of argumentation that in some way presupposes

the legitimacy it seeks to establish – one may claim – will be something less
than a proof, a second-best option.
That is why it is easy to overlook Kant’s treatment of moral skepticism: his
treatment goes counter to the expectation that every satisfactory proof must
be conducted from a theoretical standpoint. Indeed, his treatment relies on
the idea that it is possible not to adopt the vantage point that a theoretical
proof would require (a viewpoint from which the authority of morality has
not been established) and yet to provide a demonstration of the authority of
the moral law that is just as satisfying as if one had provided it from such a
vantage point. One can therefore only bring into view his response to
skepticism if one calls into question the idea that every proper proof must
be a “theoretical” one. Kant’s argument is conducted from a practical point
of view. By adopting such a standpoint, he seeks to show that we cannot get
“outside” morality in the sense in which the skeptic claims we can. And since
we cannot get outside (the “not” being a quite peculiar, practical “not”), we
need not provide any kind of separate reason for getting back in.
As I have interpreted it, the spirit of Kant’s argumentation is captured
perspicuously by the Groundwork’s remarks that by proving the reality of
freedom in a practical respect one proves that reality as if it had also been
proven in theoretical philosophy, but one is able to escape the “burden which
weighs upon theory” (G 4:448). I cannot here deal with the question of the
relation between that proof and the one that I have discussed. However, I
hope to have shown that it is just that kind of a strategy that Kant adopts
with respect to the authority of morality in the second Critique. That is, he
seeks to show that there is a possibility of responding to the question of why
we should abide by the moral law without adopting the viewpoint character-
istic of a theoretical proof. He seeks to show that one can respond to that
question – indeed, that we can provide an entirely satisfactory reply (escape
the “burden which weights upon theory”) – by appealing to requirements
that are intrinsic to the practical standpoint.
Part V
Ethics
12
How a Kantian Decides What to Do
Allen W. Wood
It is usually understood to be the principal task of philosophical ethics to

provide moral agents with well-grounded instruction concerning what to do.
It is the aim of this essay to examine Kant’s response to this task. But Kantian
ethics is not only about what to do, so I will also look a bit at the larger
context in which Kantian ethics considers how to decide what to do.
Deciding what you ought to do

We might think that it is an important task of ethics to help us decide what
to do. A more precise account, however, is that it is the task of ethics to help
us decide what we should do, or ought to do. This, at any rate, is the way Kant
would understand the task of ethics, and, more generally, the task of practical
reason. Practical reason is about “ought.”
Not every decision to do something need be preceded by a judgment that
we should or ought to do it. Some decisions are between alternatives that are
indifferent from the standpoint of reason. You can and perhaps must decide
one way or the other, but neither option, as opposed to the other, is something
that you ought to do. Practical reason, however, is about decisions based on
what we ought to do. What we ought to do is what we have good reasons to do.
A. W. Wood (*)
Department of Philosophy, Indiana University, Bloomington, USA
e-mail: awwood@indiana.edu

264 A. W. Wood
This feature of practical reason is neither trivial nor uncontroversial. There

are some philosophers who would apparently like to bypass this function of
practical reason altogether. They want to go directly to deciding what to do
without considering reasons. An uncharitable interpretation would be that
they think we should act for no reason at all. But as they might prefer to put
it, they want to reduce talk about reasons, or about what we should do, to talk
about deciding what to do. The title of one of Allan Gibbard’s books, for
example, is Thinking How to Live.1 About this title, and related phrases used
by Gibbard (such as “Why to care?”), Derek Parfit remarks: “To some of us,
these phrases seem to have a normative word missing. Rather than asking why
to care about something, we would ask why we should care about this thing,
or what reasons we have to care. Rather than concluding what to do, we
would reach conclusions about what we should do.”2
Gibbard thinks that deciding what I ought to do is really no different from
deciding what to do – with the sole addition, however, that we take a certain
attitude (which he calls “endorsement”) toward our decisions when we think of
them as done for reasons or as what we ought to decide. He equates this with
another attitude (“acceptance”) toward the norms associated with the decisions.
Talk about reasons, on Gibbard’s view, is really nothing but expression of
attitudes. Gibbard’s “expressivism” thus offers us a sophisticated deflationary
paraphrase of all our talk about “reasons.” It understands such talk as expressing
our attitude toward what we are deciding to do. Gibbard wants to avoid citing
any reasons why we ought to do it, because, to put it somewhat indelicately,
Gibbard thinks there are no such things as reasons. To judge that you ought, or
that there is a reason, is nothing more than to express a certain attitude: “To call
something rational is to express one’s acceptance of norms that permit it.”3 If
Gibbard gets his way, reasons and oughts would disappear and be replaced by
“normative attitudes.”
I think this gets things exactly backwards. Normative attitudes have to
make sense, which they cannot do if there are no reasons why you ought to
take them. Accepting a norm cannot create or replace a reason, because it
depends on a reason – one’s reason for accepting the norm. If there were no
reasons for accepting norms or taking normative attitudes, then norms and
normative attitudes would all be groundless. You cannot solve this problem
by saying that your reasons for taking these attitudes are nothing but the
1
Allan Gibbard, Thinking How to Live (Cambridge: Harvard University Press, 2003).
2
Derek Parfit, On What Matters, 2 vols. (Oxford: Oxford University Press, 2011), 2:386.
3
Allan Gibbard, Wise Choices, Apt Feelings: A Theory of Normative Judgment (Cambridge: Harvard
expression of further (perhaps “higher order”) attitudes, because you would need a
reason for those attitudes too, or else you would be unjustified in taking them as
well. For instance: You would have to be wrong in thinking that you ought (or that
there could be any reason) to agree with Gibbard’s account of oughts and reasons.
Exactly the same would hold of any of the empirical (for example evolutionary)
theories to which Gibbard wants to appeal to try to convince you of his position.
You would need reasons to accept those theories beyond your own attitudes, since
the superstitious fundamentalists who reject science also have their own opposed
attitudes. Gibbard’s theory about ought and reasons not only places science and
superstition on a par, but – still worse – it leaves nothing with any justification
beyond the (groundless) attitudes (of “acceptance” or “endorsement”) that we
happen to take toward it. In this way, it is self-undermining. Parfit puts it this way:
If we became convinced that there are no truths about what is rational, or about
reasons, or about what we ought to do, we would cease to believe that
normative questions could have answers. Our normative thinking would
then be easier, since we would cease to worry that we might be getting things
wrong. But that would not make our thinking more effective, since it would not
help us to get things right. There would be nothing to get right.4
We could put it another way. Positions like Gibbard’s, even if you give
them some “meta-” name, also make direct claims about what has value, about
what reasons there are. We can see this from the fact that they are sometimes
used to support some ethical theories (such as Hume’s) and reject others (such
as Kant’s). Specifically, their claim about value and reasons is that nothing has
value, there are no reasons, no good or bad, right or wrong, that nothing
matters. When this is pointed out to them, expressivists respond by insisting
that things can still matter to people, claiming that is all we should care about
(or all there ever could be). They emphasize that they go on caring about
things, things matter to them. And they do express these attitudes of caring
and letting things matter. So, they ask, what’s the problem?
But this reply entirely misses the point. For some people do care about things
they have no reason to care about. Some things matter to some people that
should not matter to them. If expressivist theories were correct, then that pitiful
and deplorable state would hold for everyone, all the time. Following their own
theories, these philosophers themselves could have no reason to care about
anything or let anything matter to them. Expressivism, in other words, deprives
4
Parfit, On What Matters, 2:408.
266 A. W. Wood
the expressivists themselves of any genuine reasons for doing what they say they
do. It makes no difference that they accept or endorse what they do, and give
expression to these attitudes. For that might be equally true even of people who
care about what they should not, or who let things matter to them that do not
really matter at all. They express their groundless attitudes too.
Ends and reasons

Deciding what to do often requires deciding, for a good reason, what we
ought to do. The Kantian view is that both must be based on rational
principles. Kant thinks every ought is grounded on objective reasons and
objective principles. For a Kantian, then, the very first step in deciding
what to do – at least where the decision is to be based on reasons – is always
deciding, on rational grounds, what you ought to do, and this involves basing
what you do (whether explicitly or implicitly) on rational principles.
Kantian ethics has a general concept of action. An action is something that is
in the agent’s power, and is chosen as a means to the agent’s end (G 4:427).
Looked at from this standpoint, it is the end that supplies the reason for every
action. But Kantian ethics holds that in many cases – all those involving moral
duties, and many others besides – there must be a reason for setting the end. The
end is good or valuable in some way; for instance, it contributes to your
happiness, and therefore rests on principles of pragmatic or prudential reason;
or it is an end that morality requires you to have, and therefore rests on the moral
law. So instead of saying only that the end is the reason for the action, it is more
appropriate to say that the reason for setting the end is the reason for the action.
Some actions, of course, are better thought of as omissions. The reason for them
is the reason why that action should not be done. I should not do such-and-such
because doing it would render impossible some other end I have set, or because it
would be contrary to the end of my happiness, or because its end is contrary to
some end I ought to set; or simply because it is morally prohibited to do it.
As the cases just cited illustrate, Kant recognizes three kinds of practical reason:
instrumental, prudential, and moral (G 4:414–20).5 Instrumental reason
grounds actions chosen as means to an end set arbitrarily or at your discretion
(because you find it valuable in some way). An omission is required if it would
preclude the attainment of an end. Whether the end is really valuable, as long as
5
For more on this, see Allen W. Wood, The Free Development of Each: Studies in Freedom, Right and
Ethics in Classical German Philosophy (Oxford: Oxford University Press, 2014), ch. 2.
you actually set it, you have an instrumental reason for those actions or omissions
(G 4:417). It belongs to the essence of every finite rational being to value its own
happiness – the idea of the greatest attainable whole of its satisfactions. An action
or omission is prudentially rational if it would contribute to your happiness.
Finally, some ends are morally required: there is a categorical imperative, a
command of reason not based on any presupposed end, for setting those ends.
Kant considers your own perfection (both technical and moral) and the happiness
of others to be such morally required ends (MM 6:385–88). There is a lexical
priority among the three kinds of practical reason: prudential reason trumps
merely instrumental reason, and moral reason trumps prudential reason. Moral
reasons are categorical – they are based on no independently presupposed end,
though as we have just seen, categorical imperatives require us to set certain ends,
so actions that comply with them also fit the Kantian concept of action as
something chosen as a means to an end set by the agent. Moral reasons are
overriding whenever they conflict with instrumental or prudential reasons.
Kantian ethics is an ethics of autonomy or rational self-government. But
we do not literally make the moral law. (That would be incredibly arrogant
of us.) Objective value does not depend on the will of any being – not on
God’s necessarily perfect will, nor on our miserable wills either (G 4:439).
Moral laws have, literally speaking, no author or legislator: they are valid in
themselves (LE 27:261–62, 282–83).6 Autonomy, as Kant uses the word, is
a way of regarding the moral reasons that are binding on us. Because they
are recognized and actions from them are motivated by our own reason,
when we obey moral laws we can regard them as proceeding from our own
wills. Those who do not obey the moral law, however, cannot regard them
in this way. Such people are not autonomous – they cannot even regard
themselves as self-legislating. They must do what they ought to do because
there are objective reasons to do it, whether or not their wills agree with
these reasons.
Kantian ethics is an ethics of freedom. It is binding on us because, on pain
of incoherence, we must affirm ourselves as free both in our theoretical and
practical lives (G 4:448). Kant regards the metaphysical question of freedom
as insoluble, and freedom itself as incomprehensible (G 4:459). The most we
can show is that that there is no logical contradiction in thinking of ourselves
as free and also as determined by natural causality by regarding our actions
from two standpoints (G 4:450–53; cf. A558/B586).
6
See Allen W. Wood, Kantian Ethics (New York: Cambridge University Press, 2008), ch. 6.
268 A. W. Wood
The problem of freedom, however, is not peculiar to Kantian ethics. Any

ethics presupposes that we are free to follow it. Kantian ethics makes no
special demands of this kind that depend on supernatural doctrines.7 Kantian
epistemology, moreover, restricts us to knowledge of the sensible world;
according to it, we could have no grounds for affirming any metaphysical
theories about the supersensible. Kant holds that our practical freedom is
knowable empirically (A802/B830). Kant’s position on the metaphysical
question of freedom is that freedom must be presupposed, but its possibility
is forever incomprehensible to us and we can never gain insight into it (G
4:459). This may be an unwelcome message, but it is not a positive meta-
physical theory. Look at it this way: If it turns out that accepting supernatural
doctrines is necessary to admit freedom, this affects Kantian ethics no more
and no less than any other theory. On the other hand, if Kant is raising a
pseudo-problem about freedom, then his ethical theory is no better off and
no worse off than the theories that see that it is a pseudo-problem.
Duties and virtues

We might think that there is a one-word Kantian answer to what we ought to
do. The word is duty. You ought to do your duty. But of course matters are
much more complicated. In Kantian ethics, “duty” does not refer, as it often
does in everyday life, to what you are pressured into doing by your parents,
your church, or the state, or what you are expected to do as part of some
social practice. “Duty” refers most basically to what you yourself will to do as
a rational and a moral being. Kant’s theory of practical reason allows for ends
that are set for non-moral reasons. Morality overrides these ends if they
conflict with it, but if there is no conflict, then we can have good and decisive
instrumental and prudential reasons to do some actions that are not related
to any duty.
For Kant the setting of an end is not something you do by some arbitrary
act that is isolated from the rest of your life. If an end is set for reasons, it is
based on your judgments about what there are good objective reasons to do.
But the setting of ends is also related to the kind of person you are. Different
people set different discretionary ends, and they form different conceptions of
their happiness. Even the morally required ends – your own perfection and
the happiness of others – allow for latitude and variation, because the duty to
7
See Wood, Kantian Ethics, ch. 7.
pursue these ends is wide, imperfect, and meritorious (MM 6:390). Kant says
that your path through life is not strewn with duties, as with mantraps (MM
6:409). As long as you do not on principle exclude anyone’s happiness from
among your ends, you may promote the happiness of some people and not
others. As long as you include others in your beneficence, you are permitted
also to benefit yourself, and how much is also up to you; you have no direct
duty to promote your happiness either (MM 6:451). You may choose to
develop one talent and not another. Which ones you develop, and how much,
is up to you. Even the kind and degree of moral virtue you cultivate in yourself
is up to you. There is no strict duty to be as virtuous as you can possibly be.
Kantian ethics is conceived as a doctrine of virtue. The focus of Kant’s moral
philosophy is not on employing a general criterion or devising a procedure that
tells us what to do. The question of what you ought to do – the question with
which this essay is concerned – does not occupy the central place in Kantian
ethics that it does in many ethical theories. This is in part because Kant thinks
the job of moral philosophy is not to tell us what to do, but mainly to protect
us against self-deceptive attitudes of excuse through which we try to accom-
modate these demands to our convenience or our self-interest (G 4:405–7).
Virtue is the strength of character needed to persist in doing our duty and in
pursuing the ends of morality when we might be tempted not to pursue them
(MM 6:380–82, 405–9). The focus of Kantian ethics is on what kind of
person you should be and should try to become. For a Kantian, questions
about what to do, and what you ought to do, must always be seen in that
context. Kant’s approach is far closer to that of virtue ethics than most people
realize – including most representatives of virtue ethics.
Kantian ethics has one important feature, not usually appreciated, in
common with Aristotle’s virtue ethics. Kantian duties are conceived in
what I would call “moralized” concepts: Concepts which designate what it
is right or wrong to do. It is a separate question (posed for moral judgment)
to which actions or omissions these concepts apply. For Kant, a lie is always
wrong. But not every false speaking need count as a lie. Analogously, for
Aristotle every intemperate act is wrong, but not every act of drinking wine is
intemperate. In this way, Kantian ethics does involve unexceptionable moral
rules, but (contrary to its undeserved reputation) Kantian ethics is not
necessarily inflexible or excessively strict. Whether it is depends on how the
rules are applied by judgment.
The first part of Kant’s Doctrine of Virtue provides an account of virtue – what
it is (MM 6:379–82, 403–9), the kinds of ends a virtuous person sets (MM
6:382–95), how these ends constrain actions (MM 6:396–98), what feelings
belonging to rational nature help us to constrain them (MM 6:399–403). Only
270 A. W. Wood
then does it address the question of how virtuous agents decide what they ought
to do (MM 6:410–14). Then the second part gives us a doctrine of duties. It
distinguishes duties to yourself (MM 6:417–47) – which are not duties to benefit
yourself, but duties to be worthy of your own humanity as an end in itself, which
is the basic value and motive of all ethics – from duties to others (MM 6:448–74).
Within the latter, it distinguishes duties of love (MM 6:448–61) from duties of
respect (MM 6:462–68).
What Kant calls a theory of duties, however, often reads more like an account
of virtues and vices. The strict duties to oneself involve avoiding certain vices:
vicious habits or ways of being – drunkenness, gluttony, unchastity, mendacity,
avarice, servility (MM 6:424–37). There may be characteristic acts that exemplify
these vices – e.g., drinking too much, greedy acquiring and hoarding, bowing and
scraping – but these acts, even in their conception, usually also involve adopting
certain attitudes and having certain desires or feelings – taking too much pleasure
in food and drink that is bad for you, enjoying too much the possession of
money, adopting a fearful or obsequious attitude toward the powerful. The same
is true of vices that oppose duties of love: envy, ingratitude, malice, gloating
(Schadenfreude); or those that oppose duties of respect: arrogance, defamation,
ridicule (MM 6:458–61, 465–68). The envious or gloating person takes pleasure
in the misfortunes of others, while defamation and ridicule involve not only acts
of defaming or ridiculing but also enjoying the disgrace or the derision of others.
The same is true of positive duties. Self-perfection involves caring about develop-
ing your talents and improving your character (MM 6:444–47). Beneficence
involves cultivating feelings of love for others and sympathizing with them by
actively participating in their situation; gratitude involves honoring another for
the favors they have done you (MM 6:452–58).
Why then does Kant provide us with a taxonomy of duties, rather than of
virtues? Part of the reason is historical: Kant lectured for years on
Baumgarten’s ethics, which contains a lengthy taxonomy of duties.8 But
there is also a philosophical reason. It is a common complaint against virtue
ethics that it does not offer us a clear or specific account of what we ought to
do. But from a Kantian standpoint, this is a misleading way of putting the
objection. For it invites the response that virtue ethics can supply a set of so-
called “v-rules,” telling us to behave generously, courageously, temperately,
or as the generous, courageous, or temperate person would behave.9
8
Alexander Gottlieb Baumgarten, Ethica philosophica (Magdeburg: Hemmerde, 1751).
9
See Rosalind Hursthouse, On Virtue Ethics (Oxford: Oxford University Press, 1999), esp. 13–15,
23–37.
These give us instruction about what to do that is comparable in clarity and

specificity to what rival theories can offer.
But the real problem is not specificity. It is that even if the v-rule
identifies the right actions, it does not provide the right reasons for doing
them. Mill puts it well: “No ethical standard decides an action to be good
or bad because it is done by a good or bad man, still less because it is done
by an amiable, brave or benevolent man, or the contrary.”10 Aristotle, the
hero of virtue theories, agrees with this: He holds that that the virtuous
person is the one who decides on the right thing from correct reason (orthos
logos). The truly virtuous person acts for the right reasons, the reason the
action should be done, and not because the action follows the mean or
exemplifies this or that virtue. This problem is analogous to the one we
found with Gibbard’s expressivism about reasons. If there are no reasons,
then not only does nothing matter, but there can also be no virtues. And if
you do not act for the right reasons, you do not display virtue. Reasons
always have to come first.
Duties as reasons
For Kant, the form of the right moral reason consists in citing the duty that
the action fulfills. Some duties for Kant are strict, perfect, or owed duties,
requiring a specific action. Others are wide, imperfect, or meritorious duties,
where the action itself is not required, but we are required to have an end
which this action, among others, might promote, and this would give us the
right kind of reason for doing it. Of course when an action is a duty (whether
strict or wide), there is a further reason why it is one. This gives us a deeper
reason for doing it than its being a duty. For Kantian ethics, the deepest
reason in all cases is that the action or omission treats rational nature in
persons as an end in itself. For example, beneficent actions harmonize with
the end of rational nature in a person by furthering some part of the person’s
happiness; arrogant actions are forbidden because they dishonor humanity in
the person of those to whom the arrogant person adopts an attitude of self-
conceited superiority.
Kant’s system of duties also specifies certain kinds of actions that you
ought to do and not do. There are prohibitions, for example, on suicide (G
4:422, MM 6:422–24) and lying (MM 6:429–31). These are not as strict
10
John Stuart Mill, Utilitarianism, 2nd ed., ed. George Sher (Indianapolis: Hackett, 2001), 20.
272 A. W. Wood
and unambiguous as they are represented as being in the common carica-

tures and facile criticisms of Kantian ethics. Even the strictest duties in the
Kantian taxonomy represent general rules, which need to be applied to
particular cases through a distinct act of judgment, which is never reducible
to any set of rules, deductions, or any discursive criterion (TP 8:275;
A132–34/B171–74; An 7:199).
Kantian strict duties are specified through a concept that contains in
itself the thought that the actions which fall under it are required or
forbidden. Suicide, for instance, is an act of self-murder, where “murder”
is a homocidium dolosum: wrongful homicide (MM 6:422). There could be
acts that cause the death of someone that are not wrongful, and these
would not fall under the prohibition. There are acts that might result in
your death that would not be suicide. A lie (mendacium) is a wrongful false
speaking (falsiloquium dolosum) (MM 6:238). There could be cases of false
speaking that are not lies.
Stated in such terms, the rules may be without exceptions, but it is always a
question of judgment which particular acts fall under them. Any rule not stated
in such terms, such as a rule against false speaking, always admits in principle of
exceptions. Kant’s Doctrine of Virtue therefore contains a set of “causuistical
questions” whose purpose is to sharpen our judgment through the consideration
of difficult or problematic cases – usually, proposed exceptions to the duty (MM
6:411; cf. 423–24, 428, 431, 433–34, 437, 454, 458). Some of these we know
Kant rejects, others he accepts, and some remain unclear. Kant’s casuistical
questions about suicide, for example, include a half dozen proposed exceptions
– cases where it is suggested that killing oneself would not be a violation of duty
(MM 6:423–24). We know from his lectures that Kant rejects most of them (LE
27:369–75, 627–29). But Kant’s moral theory leaves it up to the reader’s
judgment. Different people might come to different conclusions about difficult
cases. Likewise, despite some quite strict assertions of the prohibition on lying
(MM 6:429), Kant offers some examples of untruthful speaking whose permis-
sibility he entertains (MM 6:431). In his lectures he seems to accept the concept
of a “necessary lie” in which what would normally be wrongful is permissible on
grounds of duress (LE 27:447–48). The views about right that underpin Kant’s
treatment of the notorious example of the “murderer at the door” (RL 8:425–30)
are commonly misunderstood, and are much more reasonable than is usually
appreciated.11
11
See Wood, Kantian Ethics, ch. 14.
How far do duties, even wide duties, constrain our actions? Some readers of
Kant’s Doctrine of Virtue fasten on this remark: “But a wide duty is not to be
taken as permission to make exceptions to the maxim of actions but only as
permission to limit one maxim of duty by another (e.g., love of one’s neighbor
in general by love of one’s parents), by which in fact the field for the practice of
virtue is widened” (MM 6:390). This could be read as saying in absolutely every
situation, some duty – if not a perfect duty, then an imperfect duty – should
always constrain your actions. This is a strict and harsh interpretation of
Kantian ethics. If we accept it, then it would be impermissible to engage in
any conduct that is not the fulfillment of some duty or other. There could be no
room in your life for you to promote your own happiness (except insofar as you
might have an indirect duty to do so [G 4:398]).12
The preponderance of the evidence, however, is against this strict and
harsh interpretation. For Kant claims there are some indifferent actions,
and that a “fantastically virtuous” person who “strews all his steps with
duties, as with mantraps . . . would turn the government of virtue into
tyranny” (MM 6:409). Kant claims that you are permitted to be bene-
volent to yourself, without having a duty to do so (MM 6:451). I do not
think the strict interpretation can allow for that. Further, if our entire
lives were occupied only with the practice of required or meritorious
actions, then there would be no room in them for the exercise of
instrumental or prudential reason, except as ways of fulfilling the strict
command of morality. It seems more reasonable to understand Kant as
allowing you sometimes to decide you ought to do something because it
is instrumentally rational or prudentially rational in cases where no
command of morality constrains what you do.
Thus I do not read the passage quoted above (MM 6:390) as saying that
we must spend every last second of our day fulfilling some duty or other. It
says only that we should not make exceptions to the maxims of virtue. Even
when not doing meritorious actions, we should still regard our virtuous
maxims as guides to our conduct. We must not behave in a way that
abrogates our virtuous maxims. Just as you have no strict duty to be as
virtuous as you can possibly be, you have no strict duty to spend every
moment of the day performing either morally required or meritorious
actions.
12
For interpretations of Kant that take this position, see Faviola Rivera-Castro, “Kantian Ethical
Duties,” Kantian Review 11 (March 2006): 78–101; and Jens Timmermann, Kant’s “Groundwork for
the Metaphysics of Morals”: A Commentary (Cambridge: Cambridge University Press, 2010).
274 A. W. Wood
In Kantian ethics, morality guides your decisions, but not very directly. It
is meritorious to develop your talents and to promote the happiness of
others. You are permitted to seek your own happiness as well. But which
talents you develop, which others you benefit, what ends you set, what
projects you undertake, are up to you. Morality gives you reasons – good,
objective reasons, not reasons deflatable into (groundless) attitudes of endor-
sement – for setting some ends, undertaking some projects, but it does not
tell you which ones. Your decisions about these matters, along with your
situation, will then offer you good and objective reasons for doing some
things and for not doing others. Strict duties will require you to do some
things and to refrain from doing others. But Kantian morality leaves you with
the responsibility for living a decent, meritorious, and meaningful life.
Morality places constraints on how you live, and also gives you reasons for
living some ways rather than others, and moral duties (rather than non-
moral, so-called ‘ground projects’) are what give meaning and content to our
lives, but morality does not dictate your life. In a nutshell, that is the correct
answer to the question: “How does a Kantian decide what to do?”
The moral law: Universalizability, humanity

as end, autonomy
Most people who have heard of Kant would probably have expected quite a
different answer. It is common for readers of Kant’s ethical writings, espe-
cially the Groundwork, as offering more definite procedural guidance in
deciding what we ought to do. In particular, it is often thought that
Kantian ethics, in every situation, would tell you to formulate some maxim
and ask yourself whether it is “universalizable.” This picture of Kantian ethics
is so commonly accepted, both by self-described Kantians and by non-
Kantians, that it feels quixotic when I set out to dispute it. But I have no
choice. It is a gross distortion. It badly misunderstands even those parts of
Kantian doctrine which it caricatures.
In the second section of the Groundwork, Kant formulates the moral law
in three main ways, two of which involve variants:
First formula:
FUL Formula of Universal Law: “act only in accordance with that maxim
through which you can at the same time will that it become a universal law”
(G 4:421; cf. 402);
FLN Formula of the Law of Nature: “act as if the maxim of your action were to
become by your will a universal law of nature” (G 4:421).
Second formula:
FH Formula of Humanity as End in Itself: “So act that you use humanity,
whether in your own person or in the person of any other, always at the same time
as an end, never merely as a means” (G 4:429).
Third formula:
FA Formula of Autonomy: “the idea of the will of every rational being as a will
giving universal law” (G 4:431; cf. 432); or “to choose only in such a way that
the maxims of your choice are also included as universal law in the same
volition” (G 4:440); or “act in accordance with maxims that can at the same
time have as their object themselves as universal laws of nature” (G 4:437;
cf. 432, 434, 438).
FRE Formula of the Realm of Ends: “act in accordance with the maxims of a
member giving universal laws for a merely possible realm of ends” (G 4:439,
translation modified; cf. 433, 437–39).
None of these formulas is to be used directly as a decision procedure in
deciding what to do, or what I ought to do. The second formula (FH) is used
in the Groundwork to ground four general duties that belong to the system of
ethical duties presented in the Doctrine of Virtue. They are selected examples
from the much broader set of duties whose application in Kantian ethics I
have been trying to describe in the previous section.
Two of the duties Kant selects for discussion are narrow duties: the
prohibition on suicide, and the prohibition on making promises you do
not intend to keep – or more generally, on behaving toward others in ways
that involve your having ends they cannot share (G 4:429–30). The wide
duties are to develop your talents and to further the ends belonging to the
happiness of others. A close look at the Doctrine of Virtue shows that Kant
appeals to FH regularly, though often obliquely only by implication, in
arguing for all the ethical duties. He seldom or never grounds duties on
either of the other formulas.
The only possible exception to this last assertion is the way he appeals
to something like FUL in grounding the duty to include the happiness of
others among your ends. “[The moral law] permits you to be benevolent
to yourself on the condition of your being benevolent to every other as
well; for it is only in this way that your maxim (of benevolence) qualifies
for a giving of universal law, the principle on which every law of duty is
276 A. W. Wood
based” (MM 6:451). This case is necessarily unique, because happiness is

the sole nonmoral end that belongs to every finite rational being.
Therefore, the maxim of prudence, setting your own happiness as an
end, is the sole one case of a maxim everyone will necessarily adopt. And
because of the empirical fact that we are interdependent, everyone’s
happiness depending on the voluntary beneficence of others, you can
adopt the maxim of prudence consistent with morality only by including
the happiness of others among your ends. But since this gives rise only to
a wide or imperfect duty, it does not say when, or how, or to what extent
we should promote the happiness of others.
Testing maxims
It is deplorably common for readers of the Groundwork to read the first
formula (FUL, or its variant, FLN, which is the formula given the most
systematic application in the second section) as some kind of general proce-
dure to be applied to actions, at least as a sine qua non test for permissibility.
In Kant’s use of them, these formulas provide a way of ruling out as
impermissible certain maxims that cannot, without contradiction or conflict-
ing volitions, be willed to be universal laws (or laws of nature).
FUL and FLN are unable, however, to justify any of the positive duties,
whether narrow or wide, even those that are assumed in Kant’s examples. For
instance, the impermissibility of the maxim of shortening my life from self-love,
when its continuance promises more misery than pleasure, cannot ground a
general prohibition on suicide, because for all that argument shows, there might
be many other maxims (some of them permissible according to FLN) on which
one might end one’s life. In order to use FUL or FLN to establish a general
prohibition on suicide, one would have to find a way of testing for universaliz-
ability the totality of such maxims. But this seems impossible to do. The same is
true for the general requirement to keep promises, or the prohibition on making
promises you do not intend to keep. Here too there might be many maxims on
which one could make such a promise, and FLN has no way of showing all at
once that all of them are impermissible. Regarding the examples involving
imperfect duties, FLN is used to show that it would be impermissible to adopt
the maxim of refusing to develop one’s talents, or to give aid and sympathy to
others in need. For all that shows, there might be no general duty to develop
your talents or give aid to others, but only a prohibition on maxims that rule out
such conduct on principle.
Kant uses FUL, and especially FLN, to show certain actions to be

impermissible by showing their maxims to be impermissible. These formulae
have therefore often been treated as a general impermissibility test for any
and every maxim on which an agent might act or consider acting. Some have
considered these formulas as a “CI-Procedure” for the “construction” of a
moral doctrine. It is common for self-described “Kantians” to say that we
bring to morality a set of maxims; morality tells us which ones are all right
and which ones are not. But this cannot be how Kant intends the principle of
morality to work. For he realizes that the universalizability test allows for a
degree of arbitrariness that the moral law cannot allow (MM 6:389). The role
Kant intends FUL and FLN to play in decision making is much more limited
and specialized. It is not a general procedure even for distinguishing the
permissible from the impermissible.
The assumption that Kant intends these formulas to be used as general
criteria of permissibility for maxims has also invited criticisms, claiming that
some counterexample maxims either seem intuitively permissible even
though they fail the test, or intuitively impermissible even though they
pass it. Hegel charges Kant’s formula with being empty of content; Mill
declares that Kant “fails, almost grotesquely, to show that there would be any
contradiction, any logical (not to say physical) impossibility in the adoption
by human beings of the most outrageous rules of conduct.”13
Many of the supposed counterexample maxims are bogus or frivolous – they
exhibit either a misunderstanding of how the universalizability criterion is
supposed to work, or they involve obviously sophistical reasoning, suggesting
that the critic never took Kant seriously in the first place. An obvious case is
Hegel’s claim that “Help the poor” cannot be universalized because if it were,
then either there would be no poor left to help or no one with wealth left to
help them.14 Another is Brentano’s charge that Kant’s test rules out the maxim
of public officials to refuse bribes, because if it were universalized, this would
destroy the practice of bribery and leave them with no bribes to refuse.15
13
Mill, Utilitarianism, 4.
14
G. W. F. Hegel, Über die wissenschaftlichen Behandlungsarten des Naturrechts, in Werke in zwanzig
Bänden, ed. Eva Moldenhauer and Karl Markus Michel (Frankfurt am Main: Suhrkamp, 1970),
2:465–66.
15
Franz Brentano, The Foundation and Construction of Ethics: Compiled from His Lectures on Practical
Philosophy by Franziska Mayer-Hillebrand, trans. Elizabeth Hughes Schneewind (Abingdon: Routledge,
1973), 26. For a discussion of Hegel’s bogus counterexample, see Allen W. Wood, Hegel’s Ethical
Thought (New York: Cambridge University Press, 1990), 159–60; and Christine M. Korsgaard, Creating
the Kingdom of Ends (Cambridge: Cambridge University Press, 1996), 87. For a discussion of
Brentano’s, see Timmermann, Kant’s “Groundwork for the Metaphysics of Morals,” 159; and Günther
Patzig, “Der Gedanke eines kategorischen Imperativs,” Archiv für Philosophie 6 (1956): 85–87.
278 A. W. Wood
However, if FUL and FLN are treated as they usually are – as general tests for
the permissibility of any and all maxims – then these tests would certainly fail.
They would clearly be subject to both false negatives (innocent maxims that fail
the test) and false positives (morally objectionable maxims that pass the test).
Some quite permissible, even meritorious, maxims that are specific ways of
fulfilling duties could not themselves be willed to be universal laws, because they
involve conduct whose universal adoption would be impossible. It is meritor-
ious, for instance, to give to charity, and one perfectly innocent way of doing
this would be to adopt the maxim that you will give a larger percentage of your
income to charity than the average person does. But obviously that maxim
would contradict itself if adopted by everyone. False positive maxims can often
(perhaps always) be constructed for any immoral conduct by so restricting the
conditions of a wrongful act in such a way that it is foreseeable that the maxim
would be acted upon only on this one occasion. For example, my maxim might
be to make a deceitful promise to repay money but include in the maxim such
details as the day of the week, the height or clothing of the person, and so on.
A maxim of this kind could not be self-defeating if made a universal law unless
the particular wrong or bad action were self-defeating all by itself, which it
could not be if it were a successful case of immorality.
The standard reply (supposedly on Kant’s behalf) to such false positives is
to say that a maxim with such specific conditions would surely not be the
agent’s “real maxim.” The agent, it is alleged, would still want to make the
lying promise even if other details obtained instead. This reply is unconvin-
cing. If the universalizability test is supposed to provide an impermissibility
criterion for any and all maxims that anyone might propose, then the reply is
irrelevant, since if the test has that aim, it could not possibly matter what any
particular agent’s “real maxim” is. It can also be pointed out that for any
given immoral act, the agent’s maxim would always in fact include quite
specific conditions: any lying promise, for instance, would be made on a
certain day of the week, to a person of such-and-such description, using
such-and-such words. The agent might very well have good reasons for
intending precisely these details, so that the agent’s “real maxim,” fully
stated, would always be some highly restricted maxim – one that applies
only to this case. Of course the full statement of the agent’s intentions in
such a detailed maxim might not raise exactly the moral issue the Kantian
considers salient. But that is just the problem: FUL and FLN cannot tell us
what is and is not salient. That is why it cannot provide a general criterion of
impermissibility.
Self-styled “Kantians” are sometimes inventive, and usually quite stubborn,
in resisting these objections. But I think theirs is a lost cause. It is also a
thoroughly misguided cause, because if we look at what Kant actually does

with FUL and FLN, we see that he does not use them as a general criterion for the
permissibility/impermissibility of maxims. His employment of them is much
more restricted in its aim. The failed replies just discussed actually allude to
these restrictions, but without admitting it (or perhaps even realizing it).
The four examples to which FLN is applied presuppose specific duties,
which Kant does not pretend to derive from FUL or FLN (though he does
argue a few pages later that these duties are derived as applications of FH).
The four maxims tested are all attempts to rationalize making exceptions to
these specific duties, based on the agent’s particular situation or inclinations:
“My future life promises more pain than pleasure,” “I am in dire need of
money,” “I have no inclination to develop my talents,” “I won’t take any-
thing away from others, but have no desire to help them when they are in
need.” The attempt to evade the duty determines the agent’s “real maxim,”
and also sets the criteria of moral salience. The aim of Kant’s arguments is to
show the agent that these exceptions are unwarranted, because if the maxim
rationalizing these supposed exceptions were universally adopted, either this
would undermine their present intention or else it would involve having to
accept a possible situation that would be unacceptable to them because it
would conflict with their own necessary volitions.
Kant makes it explicit that his use of the formula is so restricted.
Immediately after giving his four examples, he describes the mindset of a
person tempted to violate duty, and explains that the universalizability
formula is meant to expose as bogus the attempt to justify the exception
(G 4:424). Kant also claims that this formula is to be used as a canon of moral
judgment. That is, it serves as a rule in those cases Kant has already identified
as needing one: cases in which we know what our duty is, but are disposed to
cavil with it or make unjustified exceptions to it on behalf of ourselves or our
inclinations (G 4:405). When we consider this restricted use of the univer-
salizability test, we see why it will never need to consider the proposed
maxims that threaten false negatives or false positives. The false negatives
are maxims recognized as dutiful, so they could not possibly be used to
rationalize exceptions to duty. The false positives would be maxims that
might describe in detail the intentions of an agent, but they do not offer any
putative justification for making an exception to a recognized duty.
There might of course be maxims that do offer genuine exceptions to
duties. Kant’s “casuistical questions” in the Doctrine of Virtue typically
involve cases where the application of a duty to a particular case might be
questioned. It would be a serious misunderstanding to think that FUL or
FLN must reject all such maxims. As we have seen, the all too common
280 A. W. Wood
thought that Kantian ethics consists of exceptionless rules is either a mis-

understanding or a malicious caricature. Whether we agree with him about
these particular duties, Kant chooses the maxims he does at G 4:421–25
because they illustrate ways the formulas can be successfully applied to what
he regards as unproblematic examples.
Typically, FUL and FLN are read by philosophers as part of a project in
ethical theory that is utterly alien to Kant’s. This is the project of formulating
moral principles, then testing them against our intuitions, modifying them
and seeking what John Rawls called a “reflective equilibrium” between our
principles and our intuitions. Kant’s four examples are then read as attempts
to provide that sort of confirmation, and counterexamples are taken to
discredit them. For Kant, however, these two formulas of the moral law
have been established by the way they follow from the formal aspect of the
concept of a categorical imperative. They need no confirmation by our
“intuitions.” Kant then proceeds to derive a second formula of the law (FH)
by considering the material aspect of a categorical imperative – the rational
motive for obeying such an imperative. This motive consists in an objective
end or end in itself, which Kant argues is identical with the worth we attribute
to rational nature in persons (G 4:427–29). Finally, Kant combines FUL
(with its variant FLN) and FH into a third formula: FA (and its variant FRE),
which is justified by the way it combines the first two formulas (G 4:431).
The formulas form a system (G 4:436). This system achieves the first main
purpose Kant set himself in the Groundwork, that of searching for the
supreme principle of morality (G 4:392). The main task remaining is to
establish this principle, showing that the idea of morality, or of a categorical
imperative, is not an illusion. First Kant argues that the validity of the moral
law for our wills, when the law is formulated as FA, is reciprocally implied by
the freedom of those wills (G 4:446–47). Then he argues that although
freedom cannot be proven theoretically, it is necessarily presupposed as a
property of every will that takes itself to judge or act for reasons (G 4:447–
49). This provides a deduction of the moral law.
Kant’s procedure, as I have just described it, may not appeal to some. They
substitute for it the “principle-intuition-reflective equilibrium” model
because it has become so fashionable that they are no longer able to take
seriously any alternative. That is very sad. For it is one of the aims of studying
the history of philosophy to make ourselves aware that there are ways of
thinking about philosophical issues, many of them actually adopted by past
philosophers, that fall entirely outside our current fashions and assumptions.
Our minds ought at least to be open to the possibility that past philosophers
might be right and we might be wrong.
Conscience
We have seen that Kant’s moral theory probably has more to say about what
kind of person we should be or try to become than it does directly about what
we should do. We decide what to do based on reasons provided by our ends and
projects and our situation. Our ends and projects are guided in a general way by
morality, but are left largely up to us. When moral principles play a role in our
decisions, they do so through the application of duties – whether strict and
owed duties, or wide and meritorious duties. The application of these duties to
our particular circumstances takes practical judgment. This step of applying the
general to the particular cannot be reduced to any deductive or discursive
process. So if I am a Kantian agent concerned about the morality of my actions,
there simply is no procedure I am supposed to follow in deciding whether this
or that action is my duty, or contrary to duty, or permissible or meritorious. I
do need to pay attention to my strict or owed duties, do the things they require
and omit the actions they forbid. In Kant’s view, this still leaves considerable
latitude in how I lead my life. There is no precise criterion I can employ, or need
to employ, in deciding how much of my life I devote to meritorious actions
displaying my virtues, or deciding the priorities among my various wide or
meritorious duties, or which actions I ought to perform in fulfilling them.
These are all matters for my understanding and my judgment.
Kant also acknowledges, however, that these faculties are fallible in all of
us. There are also hard cases in which we cannot be certain what we ought to
do, or are in danger of deceiving ourselves about it. I may think something is
my strict duty when it is not, or that it is permissible when it is not. I may
make wrong judgments about how duties apply to my situation. A person of
moral virtue and good judgment will make fewer errors than a vicious,
morally weak, or practically foolish person. Kant’s position is that it would
be absurd to think that moral philosophy could offer us a criterion or
decision procedure that could infallibly tell even people of bad character
and bad judgment what they ought to do. But sometimes we need to act,
even in the face of uncertain circumstances and of our own fallibility, both of
judgment and of honesty. The heading under which he considers reflections
on such cases is conscience.
Kant does not think of conscience as a faculty that tells us what to do.
That task falls to our practical reason, understanding, and judgment.
Conscience is instead a form of reflection on our actions, both those we
have done and those we are planning to do, and on ourselves as the doers
of them. Kant describes conscience as an inner court of justice, in which
282 A. W. Wood
we as agents stand trial (MM 6:437–40). Heidegger calls the Kantian

description of conscience as an inner court a “metaphor.”16 That is a
serious misrepresentation. Conscience for Kant is quite literally an inner
forum in which you pass judgment on yourself. You are both prosecution
and defense, and also the judge of yourself. The judge in the inner court
usually wears no black robes; the trial does not typically take place in a
wooden-paneled room with some hypocritically pious slogan adorning
the wall. But the court is as real as it gets, at least if the trial is honest.
(Remember that the Critique of Pure Reason itself is a court of justice in
which reason passes judgment on its own authority and on the proper
limits of its own rightful powers [Axi–xii].)
Conscience is an “inner” court in the Kantian sense in which what is inner
is what relates us to ourselves – e.g., “inner sense” is the sensation of our own
states rather than the intuition of objects distinct from us, and “inner”
(moral) worth belongs to actions whose value is appreciated by comparing
them to the moral law, not to other persons (G 4:397). In the inner court of
conscience, you assume the role of distinct persons (MM 6:438, 440), just as
in a legal court a judge accused of a crime may also be the defendant. In the
court of conscience you are either condemned or acquitted. The inner
rational verdict is accompanied by a feeling – of contentment if you are
acquitted, of disapproval and self-condemnation if you are convicted (MM
6:440). The susceptibility to this feeling belongs to your capacity as a rational
agent, without which you would not be a moral subject at all and could have
no duties (MM 6:400–401). (Feelings or attitudes have an important place
in Kantian ethics, though they must be grounded on reasons.) Conscience
does not punish: Kant thinks the execution of a guilty verdict of conscience is
left to God (MM 6:439). But conscience pronounces guilt or innocence. Its
verdict is just as real as one given in the wooden-paneled court inside a public
building. It will be even more real if you are honest with yourself, since
verdicts in public courts of justice sometimes are not given honestly.
The question before the inner court, as Kant sees it, is not what you
should do, or what you should have done. Those are questions for your
reason and judgment. The issue instead is whether you have done what can
be justifiably demanded of you. Either you followed the best judgment of
which you were capable at the time, or you did not. Your action, if required
by duty, was either done in a dutiful spirit, one which gives priority to duty
16
Martin Heidegger, Being and Time, trans. John Macquarrie and Edward Robinson (New York:
Harper & Row, 1962), 316.
over pleasure or advantage when they conflict, or it was not. If you are
acquitted, you may be content with yourself. If you are convicted, then either
you must change course, or if it is too late for that, you must condemn your
action and yourself for having done it.
Reason and judgment, in beings like ourselves, are always fallible. But
conscience cannot err (MM 6:401). When Kant says that the judgment of
conscience is unerring, I think he believes that if such an inner trial takes
place and its verdict is rendered, then you cannot be mistaken about that.
You can fail to prosecute or judge the case, and you can fail to judge yourself
as you ought. This lack of conscientiousness may of course be accompanied
by the self-deceiving opinion that you have judged yourself according to
conscience. That opinion would be in error. But if there has been no genuine
judgment of conscience, then there has been neither truth nor error (MM
6:401). You can deceive yourself. Kant is clear about this:
There are tendencies, indeed, in the souls of many, to make no rigorous

judgment of themselves – an urge to dispense with conscience. If this lack of
conscientiousness is already, in fact, present, we never get that person to deal
honestly with himself. We find in such people that they are averse to any close
examination of their actions, and shy away from it, endeavoring, on the
contrary, to discover subjective grounds on which to find a thing right or
wrong. (LE 27:616–17)
I think when Kant says conscience cannot err he means it in the same sense in
which, in our public judicial systems, we hold that the court of last instance
cannot be overruled. Its judgment is final, and in that sense, beyond the
possibility of error. The U.S. Supreme Court continually supplies us, how-
ever, with proofs that such decisions can be objectively wrong. I see no reason
why Kant could not admit that the verdict of conscience could be in error in
that way too. But just as the decision of a court of last instance is final, so you
cannot judge the particular action that came before conscience otherwise
than as it was conscientiously judged. The decision of conscience is proce-
durally final.
Conscience demands of us a duty of self-knowledge, which is to be applied
in the inner court (MM 6:441–42). This duty, however, is a wide or
meritorious duty. Kantian ethics does not suppose perfect self-knowledge or
perfect self-honesty are possible for us. Virtue, and the duty to act from the
motive of duty, are also wide or imperfect duties. Kantians should not blame
themselves because they are not as virtuous as they might be, or because they
need incentives other than duty to do what they ought (MM 6:392–93). They
284 A. W. Wood
do, however, recognize that there are good reasons to strive for greater virtue
than they have, and for striving to make duty a sufficient incentive for all
actions required by duty. These standards, not lax but also not overly strict,
are the standards applied in the inner court of conscience.
Summary
If I follow Kantian ethics, then, I decide what to do by deciding what I ought
to do. Kantian ethics offers us no decision procedure telling us what to do.
Kantian ethics always puts the question “What should I do?” in the context
of my situation, what kind of person I am, and what kind of person I should
try to become. It involves no exceptionless rules, but only a taxonomy of
duties – to ourselves (perfect and imperfect) and to others (of respect and of
love). These duties are derived not from FUL or FLN, but from FH, which
also supplies us with the motive for following them. This motive is the
dignity of rational nature, in ourselves and in others, as an end in itself.
Duties are always to be applied to particular situations through judgment,
which cannot be reduced to a set of rules or a decision procedure. But we do
have a way of doing what is morally required of us, which is to subject
ourselves and our actions to the judgment of conscience. Our understanding
and judgment are always fallible. FUL and FLN can sometimes aid us in
making judgments about what to do, and also in unmasking our uncon-
scientious tendency to quibble and adapt the demands of duty to our wishes
or our advantage.
Kantian ethics does not pretend to have all the answers. It acknowledges
our fallibility, requiring only that we do our best to act for objective reasons,
cultivate virtue, avoid self-deception, and follow our conscience.
13
Duties to Oneself
Oliver Sensen
Nowadays, if we think of moral duties, we mainly consider what we owe to

others. We believe that we should not harm others without just cause, and we
feel that we should help them when they are in need. However, is there also
something one owes to oneself? Can one only have a duty to another, or are
there duties to oneself? As a preliminary description, one can think of these
duties as something one should do (or abstain from) even if no one is looking
and no one will find out about it. These are not only duties one has with
regard to oneself, but the obligation to fulfill them is also to oneself only.1
Such duties are put forth throughout the history of philosophy, from ancient
sources onward, and even today we might use phrases such as “I owe it to
myself,” or “be the best you can be.” We believe that these demands hold
independently of what we owe to others, and sometimes even independently
of what will make us happy. But why, then, ought I to fulfill duties to myself,
and what exactly is commanded in this way?
Kant places great importance on duties toward self, and he discusses their
justification as well as the derivation of duties toward self comprehensively.
He even holds that they are more important than duties toward others,
1
For the difference, see Robert N. Johnson, “Duties to and Regarding Others,” in Kant’s “Metaphysics of
Morals”: A Critical Guide, ed. Lara Denis (Cambridge: Cambridge University Press, 2010), 192–93.
O. Sensen (*)
Department of Philosophy, Tulane University, New Orleans, USA
e-mail: sensen@tulane.edu

286 O. Sensen
saying that if there were no duties toward self, “then there would be no
duties whatsoever, and so no external duties either” (MM 6:417). This
raises four questions: (i) How does Kant justify duties toward self? (ii)
Why does he believe that they have primacy over duties toward others? (iii)
How, in general, can one derive concrete duties toward self? and (iv)
Which particular duties are there? In the following, I shall devote one
section to each of these questions. I shall argue that there is a direct path
from Kant’s categorical imperative to duties to oneself, but that Kant uses
anthropological knowledge to derive these duties. Since this knowledge is
empirical, some of these duties change as our anthropological knowledge
evolves.
The justification of duties toward self

Why exactly do I owe something to myself? Examples of duties to oneself
might be the prescription not to destroy one’s mental and physical capacities,
but to improve oneself and develop one’s talents. Maybe one owes it to
oneself not to maim one’s limbs, not to sell one’s organs, but to exercise, and
maybe even to develop new skills, such as learning another language. Why
should one do this? A natural answer would be that these duties serve one’s
happiness. They can be seen as a means to getting what one really wants.
Maybe saving money now will be to one’s benefit in retirement, or regular
exercise will keep one fit and healthy until old age. I will be glad later, one
could argue, if I forsake something now, or if I do something now to which
I have no immediate inclination.
However, Kant points out that such a justification does not yield a moral
duty, but merely a counsel of prudence (see G 4:416). On this view, one
should only save up now if one wants more money in retirement, and so on.
There is no unconditioned necessity. One could give up the aim of saving for
retirement, or adjust the amount of money one feels one will need. Moral
duties, on the other hand, seem to have an unconditional necessity. One
should obey them, independently of what one wants: “Everyone must grant
that a law, if it is to hold morally, that is, as a ground of an obligation, must
carry with it absolute necessity” (G 4:389). Happiness, on the other hand, is
based on one’s desires, and can shift with them: “in what each has to put his
happiness comes down to the particular feeling of pleasure and displeasure in
each and, even within one and the same subject, to needs that differ as this
feeling changes” (CPrR 5:25). In pursuit of one’s happiness there is therefore
nothing one needs to do unconditionally, and there is no necessary impera-

tive: “From this it follows that imperatives of prudence cannot, to speak
precisely, command at all, that is, present actions objectively as practically
necessary” (G 4:418).
Kant, on the other hand, believes that there are unconditional duties to
oneself. But how can one even make sense of them? They are not really
demands of happiness or counsels of prudence. On the other hand, they
are also not rules one consciously sets for oneself, such as a New Year’s
resolution. The reason is that one could always unbind oneself from the
obligation: “the one imposing obligation . . . could always release the one
put under obligation . . . from the obligation . . . , so that (if both are one
and the same subject) he would not be bound at all to a duty he lays upon
himself” (MM 6:417). If on one day I set down a conscious rule to exercise
each day, I can as easily abandon the rule on the next. There would be no
unconditional necessity to the rule. It is therefore hard to see how there
could be duties toward self, and Kant acknowledges that they are para-
doxical (see MM 6:417).
Kant’s solution and justification of duties toward self-focus on the fact that
they should be duties, that is, something we should do independently of what
we want. Something can only be a duty in this sense, Kant argues, if it is
immediately prescribed by one’s own reason. As an approximation that will
be clarified later, one can picture moral commands as stemming from an
innate principle. Because this principle is innate (in a sense), and prescribes
independently of what one desires, there can be genuine duties. In this
section, I will elucidate Kant’s justification of duties toward self by briefly
focusing on four questions: (i) What is duty? (ii) Why does morality have to
lie in reason? (iii) What exactly does reason prescribe? and (iv) Why should
we believe that Kant is right?
(i) The main puzzle of duties toward self is how someone could bind
oneself (see above). Bindingness or obligation is a form of necessitation: “To
have an obligation anywhere, or be bound to something” (LE 27:493)
involves “an ought or a necessitation” (LE 27:488). If one has an obligation,
one should do something that one might not want to do. “Bindingness” or
“obligation” refers to the phenomenon as such, while “duty” expresses the
obligation one has to a particular action: “The objective necessity of an action
from obligation is called duty” (G 4:439; see also 400).2
2
In the Metaphysics of Morals, Kant uses Pflicht (duty) to specify the action to which one is obliged:
“Duty is that action to which someone is bound” (MM 6:222). On Kant’s concept of obligation, see
288 O. Sensen
This is a familiar phenomenon from everyday life and does not seem to be
in need of further explanation: When we think of morality, we believe that
there is a rule or rational consideration that tells us what we ought to do:
“Every concept of duty involves objective constraint through a law (a moral
imperative limiting our freedom) and belongs to practical understanding,
which provides a rule” (MM 6:437–38). Because the law is not conditioned
by any desires we have, we conceive of it as a categorical imperative: “The
categorical imperative, which as such only affirms what obligation is, is: act
upon a maxim that can also hold as a universal law” (MM 6:225). Every
duty, including duties toward self, go back to the categorical imperative,
according to Kant, which in its most famous formulation runs: “act only in
accordance with that maxim through which you can at the same time will that it
become a universal law” (G 4:421).
(ii) Another way of making Kant’s conception of duties intuitive would be
to compare it to our ordinary idea of conscience. You may be inclined to
perform an immoral action, but your conscience tells you that you should
not do it. According to Kant, it is not strictly speaking conscience that tells us
what is right or wrong, but reason.3 However, why must reason be the source
of morality, and how exactly is reason this source?
First imagine the opposite – that moral rules do not stem from one’s own
reason. Instead one may think that we receive moral commands – these are
Kant’s examples – either passively from one’s parents, one’s society, from a
willful God, or actively through our own feelings or considerations of
perfection (see G 4:441–42; CPrR 5:40). In all of these cases, Kant argues,
the moral rule stems from outside. This is the case because for each of these
options one would need a desire to be motivated to comply with the rule: “if
one thought of him only as subject to a law (whatever it may be), this law had
to carry with it some interest by way of attraction or constraint, since it did
not as a law arise from his will; in order to conform with the law, his will had
instead to be constrained by something else to act in a certain way” (G 4:432–
33). For instance, if moral rules merely stem from one’s parents, society, or
an arbitrary God, one would be motivated by the desire for a reward, or fear
of punishment by the “power and vengefulness” of the lawgiver (G 4:443).
Even the idea of perfection ultimately rests on our conception of happiness
Oliver Sensen, “Moral Obligation and Free Will,” in Kant’s “Lectures on Ethics”: A Critical Guide, ed.
Lara Denis and Oliver Sensen (Cambridge: Cambridge University Press, 2015), 138–55.
3
On Kant’s account of conscience, see Oliver Sensen, “Kants Begriff des Gewissens,” in Gewissen
zwischen Gefühl und Vernunft: Interdisziplinäre Perspektiven auf das 18. Jahrhundert, ed. Simon Bunke
and Katerina Mihaylova (Würzburg: Königshausen & Neumann, 2015), 126–38.
(see CPrR 5:41), and, like our feelings, on our desire for it. However, if in all
of these cases one would need a desire to act morally, then one is governed by
one’s desires and their laws. Our desires, however, are part of nature, and so
“it would, strictly speaking, be nature that gives the law” (G 4:444). To be
governed by an external law of nature is what Kant calls “heteronomy.”
The problem with heteronomy is that it cannot yield obligation or ground
genuine duties: “heteronomy of choice . . . not only does not ground any
obligation at all but is instead opposed to the principle of obligation and
to the morality of the will” (CPrR 5:33). We have already seen the reason
why desires cannot ground duties. Ordinarily, we consider duties to be
necessary, but desires are contingent and shifting: feelings “by nature differ
infinitely from one another in degree,” and consequently cannot “furnish a
uniform standard of good and evil” (G 4:442). Another way of putting it is
that we believe duty to be something that one should do independently of
what one desires. It therefore is not something that is grounded in desires. If
one merely follows one’s desires, the conformity with what is moral “is only
very contingent and precarious” (G 4:390), and if one is not so inclined, one
would not be obligated (see LE 29:625). In order to conceive of duties, as we
ordinarily understand them, they have to be grounded in autonomy, that is,
“the property [Beschaffenheit] of the will by which it is a law to itself” (G
4:440). If one believes duties to be necessary, they can only have their source
in an internal law: “By explicating the generally received concept of morality
we showed only that an autonomy of the will unavoidably . . . lies at its basis”
(G 4:445).4
Kant also expresses this idea by saying that the supreme moral law is a
priori. It is not something that is “cognized and proved by experience,”
for example, by learning about the laws of nature that govern our desires
(G 4:444). For if that were the case, then the rule would be “in itself
contingent and hence unfit for an apodictic practical rule, such as moral
rules must be” (G 4:444). Experience can also say that something is the
case, but not that it must be the case. It cannot yield necessity. Only
reason can yield necessity, for instance, in analytic propositions such as
“all bachelors are unmarried” (see OD 8:228–29). Therefore, “Necessity
and strict universality are . . . secure indications of an a priori cognition”
(B4). But what exactly does it mean that morality lies a priori in the
autonomy of reason?
4
For a more thorough discussion of the relation between autonomy and morality, see Oliver Sensen,
“The Moral Importance of Autonomy,” in Kant on Moral Autonomy, ed. Oliver Sensen (Cambridge
290 O. Sensen
At the beginning of this section I said that one can picture it as if the moral
law were an innate principle, something we are just born with. However, I
warned that this account would need to be adjusted, for Kant’s view is slightly
more complicated. He does not believe that it is “an implanted sense or who
knows what tutelary nature whispers to it, all of which . . . can still never yield
basic principles that reason dictates and that must have their source entirely
and completely a priori” (G 4:425–26). If the moral law were an innate
principle, a principle that was given to us by a Creator, or – although Kant
himself does not talk about it – learned over the process of human evolution,
we would have a different principle under different circumstances, and mor-
ality would still “lack the necessity that is essential to their concept.” It would
not have an unconditional necessity, but only a “subjective necessity, arbitrarily
implanted in us” (B168).
Kant’s own expression is that a priori principles are “initially acquired”
(OD 8:222). Reason prescribes the moral law “as soon as we draw up maxims
of the will for ourselves” (CPrR 5:29). It is a spontaneous action: “reason
does not give in to those grounds which are empirically given, . . . but with
complete spontaneity it makes its own order according to ideas . . . according
to which it even declares actions to be necessary” (A548/B576). The moral
law is therefore not something reason discovers, for example, as being the
best means to securing one’s happiness, or a discovery of a moral realm where
laws are written down. Instead, the moral law is an “a priori proposition that
is not based on any intuition, either pure or empirical” (CPrR 5:31); rather,
it is something “our own cognitive faculty . . . provides out of itself” (B1).
Duty, according to Kant, derives out of a spontaneous but necessary activity
of one’s own preconscious reason: “Pure reason . . . gives (to the human
being) a universal law which we call the moral law” (CPrR 5:31). This giving
is called “autonomy,” which describes the source of the moral principle:
In regard to the faculties of the soul in general, insofar as they are considered as
higher faculties, i.e., as ones that contain an autonomy, the understanding is
the one that contains the constitutive principles a priori for the faculty of
cognition . . . ; for the faculty of desire it is reason, which is practical without
the mediation of any sort of pleasure. (CJ 5:196–97)
(iii) So far, I have presented Kant’s argument that our ordinary conception of
duties can only be captured with a necessary moral law that is prescribed by
one’s own reason: “We have therefore shown at least this much: that if duty is
a concept that is to contain significance and real lawgiving for our actions it
can be expressed only in categorical imperatives” (G 4:425). But why does
the law have the content of Kant’s categorical imperative: “act only in
accordance with that maxim through which you can at the same time will that
it become a universal law” (G 4:421)? Could reason prescribe a different law?
We do not have a direct cognition of reason itself and how it operates: “all
human insight is at an end as soon as we have arrived at basic powers or basic
faculties” (CPrR 5:46–47). But there is an indirect way in which one can
discover the content of the moral law. A duty should be able to affect one’s
behavior, otherwise morality would be a meaningless demand. Acting, how-
ever, is a form of causality – it should effect a change in the world – and a
central feature of causality, according to Kant, is that it operates according to
laws: “the concept of causality brings with it that of laws in accordance with
which, by something we call a cause, something else, namely an effect, must
be posited” (G 4:446).
The question about the content of the law therefore becomes: “what kind
of law can that be, the representation of which must determine the will”
(G 4:402)? Kant derives the content of the law from the motivational
requirement he argues for in connection with autonomy (see above).
Duties cannot be grounded on any desires we have. But if one abstracts
from any desires and ends we pursue because of them, only the mere form of
a law, its lawlikeness, remains: “Since I have deprived the will of every
impulse that could arise for it from obeying some law, nothing is left but
the conformity of actions as such with universal law, which alone is to serve
the will as its principle, that is, I ought never to act except in such a way that I
could also will that my maxim should become a universal law” (G 4:402; see
also 420–21).
(iv) Is there a categorical imperative? Before we can discuss how one can
derive concrete duties toward self from this highest moral principle, it is
worth it to pause and consider the justification of the categorical imperative.
At this stage of the argument – even if one grants that Kant analyzes our
common conception of duty correctly – it is still possible to deny “that there
really is such an imperative . . . and that the observance of this law is duty”
(G 4:425). In other words: How can one show that the law of duty exists,
and why should one follow this law rather than one’s inclinations?
Again, the method to show that the law exists has to be indirect. We do
not have a direct cognition of reason. But if there is a law that reason
prescribes a priori, it should be discoverable by its features of necessity and
universality, since these are the sure signs of an a priori lawgiving (see again
B4). If we can find necessity in our practical thought, we can discover its a
priori element: “We can become aware of pure practical laws just as we are
aware of pure theoretical principles, by attending to the necessity with which
292 O. Sensen
reason prescribes them to us” (CPrR 5:30). Kant tries to demonstrate the
necessity with the thought experiment of the gallows. In this scenario, a
monarch wants to execute an innocent and honorable person, and the prince
demands that you give false testimony against this person. You can refuse,
but then you would lose your current position at court, your life, and
everything you hold dear. Since it is a thought experiment, one can construe
it in a way that absolutely no desire speaks in favor of your refusal to
condemn an innocent man: one can assume that you do not believe in an
afterlife, you do not have a desire to be moral and no desire to become a
martyr, and so on.
Kant believes that even in such a scenario, where no desire speaks in favor
of refusing to give false testimony, everyone will judge that it is morally
wrong not to refuse. The man is innocent; it would be unjust to condemn
him merely because a monarch demands it, and my desires for reward and
fear of punishment might incline me to support the false allegations.
However, Kant does not use this intuition as the ultimate justification of
the moral law. Rather, he believes that our judgment that we ought to refuse
the false testimony lets us discover that we could refuse it: “ought implies
can.” The agent “must admit without hesitation that it would be possible for
him” (CPrR 5:30). The ‘ought’ gives one the sense that one could act against
and completely free of one’s inclinations. The moral command justifies a
belief in freedom: “He judges, therefore, that he can do something because
he is aware that he ought to do it and cognizes freedom within” (CPrR 5:30).
Kant concludes the argument with a step that we have encountered before.
Freedom is supposed to be a causality, such that we could act and effect
changes in the world by acting freely. Causality implies that it operates
according to laws. Since freedom is supposed to be the independence from
desires, once one eliminates all desires, only the form of the law remains, and
the mere form of a law is nothing other than the categorical imperative (see
above): “what, then, can freedom of the will be other than autonomy, that
is, . . . to act on no other maxim than that which can also have as object itself
as a universal law. This, however, is precisely the formula of the categorical
imperative” (G 4:446–47; see also G 4:402; CPrR 5:29).5
The justification of the law of duties not only involves the existence of the
categorical imperative, but also the question of why one should give priority
to morality over one’s inclinations. It could be that one is torn between both,
5
For a more detailed defense of this interpretation, see Oliver Sensen, “Die Begründung des
Kategorischen Imperativs,” in Kants Begründung von Freiheit und Moral in “Grundlegung” III: Neue
Interpretationen, ed. Dieter Schönecker (Münster: Mentis, 2015), 233–58.
one’s inclinations and the imperative’s demand. It needs a justification to

follow duty: “why, then, ought I to subject myself to this principle”
(G 4:449)? One can find several answers to this question in Kant’s texts,
and each of them can still be interpreted differently. Kant’s most promising
answer, it seems to me, is that one can recognize that one’s desires are fleeting
and contingent, but that the moral law concerns one always and necessarily:
“those laws apply to him immediately and categorically, so that what inclina-
tions and impulses (hence the whole nature of the world of sense) incite him
to cannot infringe upon the laws of his volition as intelligence”
(G 4:457–58). One recognizes that morality is more important in virtue of
it being necessary, while inclinations are not.
In sum: We have duties, according to Kant, because our own pure reason
places them upon us independently of our inclinations. One’s own reason
prescribes a law, the categorical imperative, which binds one unconditionally
and necessarily.
The primacy of duties toward self

After the previous discussion, it is now easy to explain why Kant believes that
duties toward self seem to have a privileged status and that if “there were no
such duties: then there would be no duties whatsoever, and so no external duties
either” (MM 6:417). As Dieter Schönecker has pointed out,6 the claim is not
that without a particular duty toward self, there would be no particular duties
toward others: for example, if there were no duty against gluttony, one would
not have to abstain from killing others. Rather the point is a more general one,
and focuses on the obligatoriness of any duty. To have any duty at all is to be
bound by the law of one’s own reason. Whether the duty is toward oneself or
others in content, in both cases it is prescribed by one’s own reason, and
therefore this duty to oneself – to follow the law of one’s own reason – is
prior to and more important than any duty toward others:
I can recognize that I am under obligation to others only insofar as I at the

same time put myself under obligation, since the law by virtue of which I
regard myself as being under obligation proceeds in every case from my own
6
See Dieter Schönecker, “Kant über die Möglichkeit von Pflichten gegen sich selbst (Tugendlehre, §§1–
3), Eine Skizze,” in Kant als Bezugspunkt philosophischen Denkens: Festschrift für Peter Baumanns zum 75.
Geburtstag, ed. Hubertus Busche and Anton Schmitt (Würzburg: Königshausen & Neumann, 2010),
235–60.
294 O. Sensen
practical reason; and in being constrained by my own reason, I am also the one
constraining myself. (MM 6:417–18)
In this sense, another can claim a right by reminding the agent of his duty to
follow the moral law of his own reason:
the other, having a right to do so, confronts the subject with his duty, i.e., the
moral law by which he ought to act. If this confrontation makes an impression
on the agent, he determines his will by an Idea of reason, creates through his
reason that conception of his duty which already lay previously within him,
and is only quickened by the other, and determines himself according to the
moral law. (LE 27:521; see also MM 6:239)
Without this one abstract, fundamental duty to obey the moral law, there
would be no particular duty whatsoever, including any duties toward others.
It is in this sense that one’s duty toward self has primacy.7
What binds oneself is therefore not the will of another, but the thought
of how one would be if one were to follow the moral law: “It seems as
though, in duty, the will of a legislator underlies . . . what we do. . . . Yet this
other will is not that of another being; it is only our own will, insofar as we
make it general, and regard it as a universal rule” (LE 29:627). Kant does
not rely upon a metaphysical distinction to explain duties toward self: “The
subject that is bound, as well as the subject that binds, is always the human
being only; . . . we may not think of them as different substances putting him
under obligation” (MM 6:419). There is one human being and one con-
sciousness. But within this consciousness there are different aspects: “the
will stands between its a priori principle, which is formal, and its a poster-
iori incentive, which is material, as at a crossroads” (G 4:400). In one’s
consciousness there is a necessary aspect, the moral law, and contingent
aspects, one’s inclinations. The deciding human being as a whole is bound
by the a priori aspect.
This concludes Kant’s explanation of how there could be duties to oneself,
and how they are justified. But it still needs a much more thorough account
of how one can derive concrete duties from the categorical imperative. This is
the subject of the next section.
7
For a more thorough demonstration of the fact that duties to others depend on duties to self, see Lara
Denis, “Freedom, Primacy, and Perfect Duties to Oneself,” in Kant’s “Metaphysics of Morals,” 170–91.
The derivation of duties

Since Kant’s own time, people have questioned whether the categorical
imperative could yield any concrete duties. For instance, Hegel famously
charged that the imperative is empty,8 and recent commentaries added to
this that the imperative is at best an unreliable guide to derive any content.
According to the standard way that Kant’s test is interpreted, one is to – first
– take a proposed action, for example, to tell a lie in my particular circum-
stances, then – second – ask, “What if everyone did that?” and – finally – see
if a contradiction occurs, that is, that the action would become impossible, or
the outcome impossible to be willed (see G 4:424). One can sum up the
charges by recent commentaries as saying (i) that the imperative rules out too
much as being morally forbidden, (ii) that it rules out too little by failing to
condemn other practices we deem immoral, and (iii) that whatever it rules
out, it does so for the wrong reasons.
(i) The most damaging charge is the demonstration that applying the
categorical imperative seems to rule out the majority of indifferent actions as
immoral. One can take a random proposed action – for example, to play
tennis at 10 a.m. on Sundays9 – but if one asks whether everyone could
follow this as a universal law, the proposed action seems to become impos-
sible. If everyone tried to play tennis at 10 a.m. on Sunday mornings, the
courts would be too crowded and no one could play. This objection is not
limited to this one example, but can be applied to almost anything, such as
choosing a profession, planning an event, or finding a niche for oneself.
(ii) Another famous charge asserts that the categorical imperative also fails to
rule out other things we deem immoral. For instance, if one is offered a bribe,
should one refuse? To do so seems to yield a contradiction in conception, akin
to universalizing a false promise. If everyone had that maxim of making false
promises, the whole institution of promising would vanish, “since no one
would believe what was promised” (G 4:422). Likewise, if everyone were to
refuse a bribe, the whole institution of bribery would vanish, since no one
would bother to offer a bribe.10 Does this make it immoral to refuse bribes?
8
See G. W. F. Hegel, Elements of the Philosophy of Right (1820), trans. H. B. Nisbet (Cambridge:
Cambridge University Press, 1991), §135.
9
See Barbara Herman, The Practice of Moral Judgment (Cambridge: Harvard University Press, 1993),
138.
10
See Franz Brentano, The Origin of Our Knowledge of Right and Wrong, trans. Roderick M. Chisholm
and Elizabeth H. Schneewind, ed. Oskar Kraus and Roderick M. Chisholm (London: Routledge, 2009),
11.
296 O. Sensen
(iii) Even if the categorical imperative could be used as a reliable guide for
the derivation of duties, it is not clear whether it would provide a moral test.
If the failure it is meant to rule out is a contradiction (in conceiving the
universalized action or in willing its outcome), this does not by itself amount
to a moral defect. For instance, to commit a logical contradiction is a
deficiency in rationality, but not a moral fault.
Sympathetic Kantians have not been without responses. If one adds
the restriction that maxims are not to be thought of as very specified
courses of action, for example, to play tennis on Sundays at 10 a.m., but
merely as the underlying aim, for instance, to exercise or develop one’s
talents, then the imperative would not necessarily rule out too much.
One could universalize that everyone should develop their talents or
exercise in some form or other, and the particular implementation of
playing tennis on Sundays would not be part of the moral evaluation.11
This Kantian response might be able to avoid the objections that were
raised under (i) and (ii) above. However, further explanation would be
needed to explain why it is also a moral criterion (iii). In addition, we
are now trying to interpret Kant’s procedure in a way that it yields the
results we wanted, rather than being able to use Kant’s procedure to
discover what is morally right. This warrants a fresh look at Kant’s text. I
believe that it states a different procedure that we immediately recognize
as being moral. What is this alternative procedure?
In the Groundwork, after Kant presents the categorical imperative and
illustrates the derivation of duties with four examples (see G 4:421–24), he
gives an explanation of how the derivation of duties from the imperative is
supposed to work. He first mentions the idea of contradictions, and this is
what has been fueling the famous criticisms of his method. But afterward he
gives a second, slightly different explanation, and this is the one I want to
focus on. Kant says the following:
If we now attend to ourselves in any transgression of a duty, we find that

we do not really will that our maxim should become a universal law, since
that is impossible for us, but that the opposite of our maxim should
instead remain a universal law, only we take the liberty of making an
exception to it for ourselves (or just for this once) to the advantage of our
inclination. (G 4:424)
11
See Onora O’Neill, Constructions of Reason: Explorations of Kant’s Practical Philosophy (Cambridge:
Cambridge University Press, 1989), 129; and Otfried Höffe, Ethik und Politik: Grundmodelle und
-probleme der praktischen Philosophie (Frankfurt am Main: Suhrkamp, 1979), 90–91.
The idea that one should not make an exception for oneself is a much
more recognizably moral idea. It refers to a sense of fairness, that one should
not be a free rider. (Of course, Kant would turn it the other way around: We
ordinarily consider free riding to be immoral because we are already guided
by the categorical imperative that prohibits free riding.) It is therefore
worthwhile to examine whether we will get a more reliable test if we do
not focus on contradictions that might or might not occur, but on the idea of
making an exception for oneself.
One might object that, while the idea of not making exceptions fares
better with respect to the objection (iii) above, it is not clear that it can
answer objections (i) and (ii). Again, many morally neutral actions might be
ruled out if one is not allowed to make an exception (i). The reason why one
might choose to play tennis at 10 a.m. on Sunday is because other people will
be in church. By playing at that time, one tries to make an exception. If one
tries to win at sports, or chooses a profession in which there is less competi-
tion, this too can be construed as an attempt to make an exception, but as
such these intentions do not seem to be immoral. On the other hand, the
prohibition against making an exception would probably also not rule out
many things that we do believe to be immoral (ii). For instance, somebody
who accepts fraud and bribery as a common business practice would not
make an exception for him or herself.
But there is a second part to Kant’s method of ruling out exceptions.
Kant also elucidates his method in the following way: “Consequently, if
we weighed all cases from one and the same point of view, namely that
of reason, we would find a contradiction in our own will, namely that a
certain principle be objectively necessary as a universal law and yet
subjectively not hold universally but allow exceptions” (G 4:424).
What is morally problematic is not any exception one tries to make,
but – first – an exception to – second – a rule one holds to be
“objectively necessary.” For instance, if one drives on the interstate and
there is a traffic jam ahead, it is not immoral to leave the interstate and
drive a roundabout through nearby villages. One does try to make an
exception, but there is no objectively necessary rule that people stay on
the interstate. However, if one were to drive on the emergency lane
instead, this could be deemed immoral. It is objectively necessary that
emergency lanes are kept free for emergency vehicles, and making an
exception to this rule seems morally wrong, all things being equal.
On my interpretation, the categorical imperative only demands that one
not make an exception to an objective necessary rule, but it does not itself say
which rules are objectively necessary. I therefore have to provide an
298 O. Sensen
additional method with which one can establish which laws are objectively
necessary. Many Kantians do not go that route, I believe, because it seems to
add an extra element beyond the categorical imperative. This seems to violate
the purity of Kant’s procedure. Should one not be able to just take the
categorical imperative, and derive duties simply from that? However, Kant’s
texts seem to speak against that. Kant admits that the categorical imperative
by itself is empty: “through the law . . . in genere, no rule of dutiful action can
then itself be determined, because this belongs to the matter” (LE 27:578).
This position should not be surprising because it is very similar to Kant’s
theoretical philosophy. We only have cognition of objects if two elements
come together: a priori elements and sense impressions, or form and matter.
As Kant says, “Thoughts without content are empty, intuitions without
concepts are blind” (A51/B75). It is the same for deriving duties. Form
and matter need to come together. The form is the a priori categorical
imperative, and the matter is anthropological knowledge. The imperative
“needs anthropology for its application to human beings” (G 4:412). This
knowledge of our human nature comes a posteriori from observations and
experiments.
Think about it: it is not the case that Kant just takes the categorical
imperative by itself and analytically derives duties merely out of this
principle. If, for instance, human beings were autarkic and indestructible,
there would be no need to help others. The fact that human beings are
finite and vulnerable, together with the categorical imperative, generates
the duty of giving assistance. However, it cannot be the case that
empirical knowledge by itself will generate objectively necessary laws.
As we have emphasized above (in the section “The Justification of
Duties Toward Self”), experience cannot yield necessity, and this will
also apply here. Necessity has to come from an a priori law of reason (see
B4). Kant formulates this law in the following way: “Whoever wills the
end also wills (insofar as reason has decisive influence on his actions) the
indispensably necessary means to it that are within his power” (G 4:417).
In contrast to the categorical imperative, this law is sometimes called the
“hypothetical imperative.”12 According to my reading, then, it often
needs three elements to generate a duty: (1) an anthropological fact
about human beings, (2) a rational insight that something else is a
necessary means to fulfilling the anthropological need, and (3) the
12
See Thomas E. Hill, Jr., “The Hypothetical Imperative,” Philosophical Review 82, no. 4 (Oct. 1973):
429–50.
categorical imperative which says not to make an exception to this

necessary means. (I will argue below that two kinds of duties to oneself,
duties to oneself as a moral being only and the general duty to follow the
categorical imperative, can do without element [2].)
In the following I shall briefly demonstrate in general how Kant derives
one form of concrete duties, and then in the next section I shall go through
the particular duties toward self. Kant starts out with what he takes to be facts
of human nature (1). He sometimes calls these facts an “end of his nature”
(MM 6:419), or “impulses of nature”: “There are impulses of nature having
to do with man’s animality. Through them nature aims at a) his self-
preservation, b) the preservation of the species, and c) the preservation of
his capacity to enjoy life” (MM 6:420). An example of this impulse is that
“self-love . . . impel[s] toward the furtherance of life” (G 4:422). These are
empirical facts, as Kant understood them, and we have to keep in mind that
he might be wrong about some of them. He relies on the science of his day,
and our scientific insights may have changed. I will come back to this later.
On a second level, reason then comes up with hypothetical imperatives,
the means necessary to fulfill nature’s aims (2). For instance, in order to
preserve myself, it is necessary that I will not be killed, or that I do not kill
myself. In order to fulfill my needs and deal with my finitude, it will be
necessary to receive help from others at some point. These considerations
generate laws we hold to be objectively necessary: for example, not to be
killed, but be helped in need, and so forth. Finally, what is immoral is to
make an exception for oneself to these objectively necessary laws. The
categorical imperative commands not to make an exception (3). Our ten-
dency is to “take the liberty of making an exception to it for ourselves (or just
for this once) to the advantage of our inclination” (G 4:424). In the case of
duties toward others, we try to make an exception for ourselves. In the case of
duties toward self, we try to make an exception “for this once.” But in both
cases, the categorical imperative commands not to make exceptions. This is
how Kant derives concrete duties.
This method is further complicated by the fact that Kant believes that there
are different ways of applying the core idea of the categorical imperative. He
does believe that they are “at bottom the same” and “tantamount,” but they
seem to use a different method to derive particular duties (G 4:437–38).
Maybe one can compare it to ways of determining one’s blood alcohol level.
At a traffic stop, a police officer can ask for a breathalyzer test or order a blood
test. These are different ways to test for alcohol, but they are getting at the
same thing and are testing for the same agent. In like manner, one can try to
establish Kant’s core idea in different ways. The core idea is that one is not
300 O. Sensen
selfish or a free rider. One can test it by asking whether one makes an exception
for oneself to an objective necessary law; one could also test it by asking
whether a contradiction occurs if one’s immoral maxim were held as a
universal law, or one could ask whether one tries to exalt one agent over
another (see MM 6:449), and thereby disrespects one and does not treat this
agent as an end in itself (see G 4:429). But all these different methods try to get
at the same idea that we have an equal and high importance. The categorical
imperative is “the principle of equality” (MM 6:451). But the point is not that
we are all equally low in importance, but that we do have a high importance –
we have an impulse to preserve ourselves – and that the importance is equal
among all.
My interpretation will not be shared by Kant scholars who believe that the
different formulas of the categorical imperative are not equivalent.13 My
interpretation tries to take Kant’s texts seriously. He says that the different
formulas are “at bottom only so many formulae of the very same law.” He
believes that their methods for deriving duties differ, but that these are
subjective differences: “There is nevertheless a difference among them,
which is indeed subjectively rather than objectively practical, intended
namely to bring an idea of reason closer to intuition” (G 4:436). I shall
demonstrate this very briefly using the main formula of the categorical
imperative (the Formula of Universal Law), and the end in itself formulation
of it (the Formula of Humanity): “So act that you use humanity, whether in
your own person or in the person of any other, always at the same time as an end,
never merely as a means” (G 4:429). One might think that this formula
introduces a very different justification from the one sketched above. Does
it not affirm that all human beings have an absolute value that should be
respected?
However, what exactly is this value supposed to be? If it is not merely
something that we do value – based on our desires – or merely something that
we should value – this requires a further justification – then it is something that
really has value, where value is a property or entity, most likely a non-natural
entity one cannot just discover with one’s senses. But Kant argues directly that
morality is not grounded on a hyperphysical property (see G 4:410, 425).
Rather, Kant uses “good” or “value” to describe what “reason independently of
inclination cognizes as practically necessary, that is, as good” (G 4:412).
“Good” or “value,” according to Kant, is a prescription of reason, what reason
13
For the latest advance of an old debate, see Henry E. Allison, Kant’s “Groundwork for the Metaphysics of
Morals”: A Commentary (Oxford: Oxford University Press, 2011), 245–60.
deems necessary: “All imperatives . . . say that to do or to omit something

would be good” (G 4:413). But there are two ways: “if the action would be
good merely as a means to something else the imperative is hypothetical; if the
action is represented as in itself good, hence as necessary in a will in itself
conforming to reason, as its principle, then it is categorical” (G 4:414). Kant
does not believe that we could postulate entities that go beyond the senses (see
A770/B798). To say that something is “good” just is to say that reason deems
it necessary. If it is necessary as a means under some circumstances, it is
relatively good; if it is good under all circumstances, then the action is “good
absolutely” (CPrR 5:62; see also 60).
The point of the Formula of Humanity is therefore that one should not
exalt oneself above others (see MM 6:449). This command is a categorical
imperative as well. It is always and necessarily valid. As such it must be a
direct, a priori command of reason, not the empirical discovery of values:
“This principle of humanity . . . is not borrowed from experience; first
because of its universality, . . . so that the principle must arise from pure
reason” (G 4:430–31). This refers back to the justification that we sketched
above (in the section on “The Justification of Duties toward Self”). Morality
should be necessary, and therefore moral commands must lie a priori in
reason. Because of these considerations,14 I shall explain Kant’s derivation of
duties in terms of “not making an exception,” rather than in terms of
respecting a value. One can use the latter method, and it might be more
intuitive in some cases, but it does not try to express a different idea than the
“strict method” (G 4:436). Rather, it merely tries to detect the same fact
(selfishness) using a different test – as the police officer can detect alcohol
levels in different ways.
What, then, are the concrete duties that I owe to myself?
Deriving concrete duties toward self

Kant divides duties toward self into three main classes. The first class are
perfect duties (I), the second are imperfect duties to oneself (II), and the
third is a meta-duty to “know yourself” and to act morally from the right
motivation (III) (see MM 6:441–42, 446–47). I interpret this third class
as the general duty to follow the moral law for its own sake. I have said a
14
For a much more extensive treatment of the Formula of Humanity and the reason to respect others,
see Oliver Sensen, Kant on Human Dignity (Berlin: de Gruyter, 2011).
302 O. Sensen
little about this duty in the section on the primacy of duties toward self,
and will leave that one aside here. In addition, my aim is not to discuss
each particular duty in detail,15 but to sketch the derivation of each class
of duties and to reflect how plausibly it follows from the categorical
imperative.
Kant distinguishes a perfect from an imperfect duty in that the former
“admits no exception in favor of inclination” (G 4:421n). Kant also char-
acterizes perfect duties as “negative” or “duties of omission,” but it seems that
the first definition is the more important one (MM 6:421). Maybe a perfect
duty, for example, of not killing someone, involves actually doing something,
for instance, to stop the dripping poison; and maybe imperfect duties would
require someone from refraining to do something, for example, if one helps
another by not doing anything.16
Kant further divides perfect duties to oneself into two classes: one in which
the human being is both animal and moral being (Ii), and duties which a
human being only has to himself as moral being (Iii) (see MM 6:420). For
the first class (Ii), Kant refers to the aforementioned three impulses of a)
preserving oneself, b) preserving the species, and c) preserving the capacity to
enjoy life (see MM 6:420). He discusses the corresponding duties in terms of
the vices that relate to each category: suicide and mutilation (a), unnatural
use of one’s sexual inclinations (b), and excessive use of food and drink (c).
There is a clear common reason why (a) and (c) are ruled out by the
categorical imperative. The imperative commands – independently of what
one desires – that one be moral. Yet, killing oneself, maiming one’s capa-
cities, or engaging in excessive drinking amounts to “depriving oneself
(permanently or temporarily) of one’s capacity for the natural (and so
indirectly for the moral) use of one’s powers” (MM 6:421).
This reason is probably the clearest in the suicide case: “To annihilate the
subject of morality in one’s own person is to root out the existence of
morality itself from the world, as far as one can, even though morality is
an end in itself” (MM 6:423). Morality is an end in itself: one ought to
pursue it, independently of what one wants. Yet, one would violate and
permanently undermine this unconditional command if one were to destroy
the possibility of fulfilling it: “it is a contradiction that he should be
authorized to withdraw from all obligation, that is, freely to act as if no
15
See the essays in Andreas Trampota, Oliver Sensen, and Jens Timmermann, eds., Kant’s
“Tugendlehre”: A Comprehensive Commentary (Berlin: de Gruyter, 2013) for such an analysis.
16
I thank Maike Albertzart for prompting me to clarify this point.
authorization were needed for this action” (MM 6:422–23). If there is an

unconditional command to be moral, one is not authorized to end this
pursuit. The same reasoning applies to gluttony and excessive drunkenness.
They do not always permanently damage the capacities to be moral, but
brutish excess at least “restricts or exhausts our capacity” to “use his powers
with skill and deliberation” (MM 6:427). Again, if there is an unconditional
command to use one’s powers in striving to be moral, then undermining
these powers – even if temporarily – goes against that command.
But this argument, as plausible as it might seem in the first two cases if one
accepts the existence of the categorical imperative, does not seem to work as
well in the case of “unnatural” sex where “one is aroused to it not by a real
object but by his imagining it” (MM 6:424–25). It does not seem that the act
itself undermines one’s capacity for morality. Kant, accordingly, admits that
“it is not so easy to produce a rational proof that unnatural, and even merely
unpurposive, use of one’s sexual attribute is inadmissible” (MM 6:425). One
thing Kant might have in mind is the following: it could be that Kant
adhered to a medical view of his time, according to which each sex act (at
least for males) weakens the human life powers, and actually shortens one’s
life.17 In that case, engaging in “unnatural” sex would reduce and shorten
one’s powers to further morality. However, this argument heavily dependents
upon an eighteenth-century science, and it loses all force with opposite
conclusions of modern science.
Kant might rely on the view of his time that there is a strong “aversion” in
reaction to “unnatural” sex that makes it seem like a “defiling” (MM 6:424–
25). (It might be that this reaction is still widely held. Who would want to be
caught in the act by his grandmother?) However, this might be a different
reaction than judging something to be morally wrong. It might be disgusting
to think of one’s parents having sex, for instance, but this aversion does not
make it morally wrong. Jesse Prinz uses the CAD-model to explain different
sources of moral beliefs. We may react with contempt against community
violations such as rules of etiquette (C). But we react with anger against
autonomy violations (A), and the infringement upon rights. Disgust, on the
other hand, is often related to norms that originate out of considerations of a
divine or natural order, and that often pertain to bodily fluids (D).18
17
For the possibility that Kant may have held such a view, see Mark Timmons, “The Perfect Duty to
Oneself as an Animal Being,” in Kant’s “Tugendlehre,” 241–42.
18
See Jesse J. Prinz, The Emotional Construction of Morals (Oxford: Oxford University Press, 2008),
69–73.
304 O. Sensen
To apply this back to the current case: even if we have aversions and
react with disgust to certain sex acts, this by itself does not make it an
autonomy violation, according to the CAD-model, and Kant’s categorical
imperative, as a rule of fairness, seems to be mainly about autonomy
violations. What Kant calls “unnatural” sex would only be ruled out by
his system if it becomes addictive, and if a person “surrenders his
personality” – permanently or to a degree where one is incapacitated
for a long stretch of time – similarly to gluttony or severe cases of
drunkenness.
The second class of perfect duties (Iii) that Kant discusses are duties to
oneself “as a moral being only (without taking his animality into considera-
tion)” (MM 6:420). Rather than protecting the capacities that help leading a
moral life, these concern the moral capacity itself. A being capable of
morality is not causally determined by his or her inclinations, but can act
for the sake of the moral principle instead. This prerogative that distinguishes
human beings from the rest of nature should not be forfeited. These duties
consist “in a prohibition against depriving himself of the prerogative of a
moral being, that of acting in accordance with principles, that is, inner
freedom, and so making himself a plaything of the mere inclinations and
hence a thing” (MM 6:420). Kant lists three vices – lying, servility, and
avarice – but it is at first glance not clear why he focuses on these three, and
whether they are the only ones.
The moral fault in all three lies, as Kant says, in the fact that “they make it
one’s basic principle to have no basic principle and hence no character, that
is, to throw oneself away and make oneself an object of contempt” (MM
6:420). In the course of his discussion, Kant seems to give several reasons
why these vices are morally deficient, and there is room for how to interpret
these arguments. In the following, I will try to see whether they can be
reconstructed from the justification Kant gives for the categorical imperative
and the intention he states at the outset: the categorical imperative com-
mands not only to act on inclinations, but on universalizable maxims. Being
able to act on basic principles against inclinations is the human being’s inner
freedom and prerogative. All three vices must violate this inner freedom.
Here is how they do.
In the case of lying, the problem is, as Stefano Bacin has pointed
out,19 that in using the capacity to adopt principles, one is already
19
See Stefano Bacin, “The Perfect Duty to Oneself Merely as a Moral Being,” in Kant’s “Tugendlehre,”
252–53.
deceiving oneself. Human beings have the prerogative of inner freedom.

We are not causally determined by our inclinations, but can act on
principles. If, however, we deceive ourselves in formulating a principle
– and we will do so because of an inclination (of gaining an advantage,
and so on) – we are not using the capacity of inner freedom but
following an inclination. It is in this way “a renunciation by the speaker
of his personality” (MM 6:429). In the case of greedy avarice, one
subordinates the ability to set principles freely to the maxim of acquiring
“all the means to good living with the intention of enjoyment” (MM
6:432). In the case of miserly avarice, the aim is further contradictory
in that one acquires the means without an intention of enjoyment. One
subordinates one’s capacity of inner freedom to acquire means of good
living, but one does not even use them for this end. Finally, in the case
of servility, one does not revere the moral capacity within oneself, but
compares oneself with others in terms of nonmoral goods, thereby again
subordinating one’s moral capacities (see MM 6:434–36). In all three
cases, one does not put one’s moral capacity in its rightful place. One
puts inclinations (to lie for an advantage, to acquire goods for enjoy-
ment, or to lower oneself in order to get favors from others) above the
capacity of inner freedom, and thereby makes it one’s principle not to
have a basic principle, whereby one violates the unconditional command
to follow the basic moral principle.
The derivation of imperfect duties is more straightforward (II).
Imperfect duties are ones that do allow for exceptions in favor of
inclinations and leave room for judgment (see G 4:421n). They concern
the cultivation of one’s natural powers, such as the powers of spirit, soul,
and body (see MM 6:445). One should develop one’s talents because “as
a rational being he necessarily wills that all the capacities in him be
developed, since they serve him and are given to him for all sorts of
possible purposes” (G 4:423). Human beings are finite and not self-
sufficient. We often need to act in order to fulfill our needs. Therefore,
reason declares it to be necessary that we develop the skills to fulfill our
needs: “Whoever wills the end also wills (insofar as reason has decisive
influence on his actions) the indispensably necessary means to it that are
within his power” (G 4:417). And it follows that it is an objectively
necessary law that human beings develop their talents. The categorical
imperative demands not to make an exception to a law which we cognize
as objectively necessary. Thus we should develop our talents. However,
this duty is still “only a wide and imperfect duty; for . . . it determines
nothing about the kind and extent of actions themselves but allows a
306 O. Sensen
latitude for free choice” (MM 6:446). It is up to the discretion of the

agent since “people have different ends” (MM 6:445).
Conclusion
I have argued that there is a clear, coherent path from Kant’s categorical
imperative to duties toward self. Inasmuch as there is a categorical impera-
tive, Kant has shown that there are duties toward self. All duties, according to
Kant, must originate in a necessary and preconscious law of one’s own
reason. Only in this way can something be commanded independently of
what one desires. The categorical imperative prescribes to follow its dictates
and not to make an exception to a rule one cognizes as objectively necessary.
These rules are based upon anthropological knowledge and are cognized as
being necessary to fulfilling basic (universal) human needs. But they are not
purely a priori, as they are conditioned by empirical needs; only the prohibi-
tion against making exceptions is purely a priori. What this means is that
Kant cannot give us a full and complete list of duties toward self.
Like duties toward others, these duties “cannot be set forth in detail and
classified in the metaphysical first principles of a doctrine of virtue, since this
has to do only with its pure rational principles” (MM 6:468). Our knowl-
edge of human nature may change, as well as our nature itself. Advances in
biotechnology may lead to interventions in the human genome or to the
creation of artificial organs that fundamentally alter what human beings are
like and need. There is no way that a philosopher in eighteenth-century
Königsberg could understand and foresee all of humanity’s needs. But his
account formulates a criterion with which we can determine our duties to
ourselves, and it empowers each of us to work out our duties. The empower-
ment should not be seen as a deficiency of Kant’s account, but as an attractive
feature of it.
14
Demandingness, Indebtedness, and Charity:
Kant on Imperfect Duties to Others
Kate Moran
This chapter will focus on a set of puzzles prompted by Kant’s account of

imperfect duty toward others. Perhaps because imperfect duties are so often
introduced and discussed by way of contrast with perfect duties, a reader of
Kant’s texts is often left with a clear idea of what imperfect duties are not, but a
less than clear idea of what they are, or what precisely they entail. Here,
I address two questions about Kantian duties of assistance. The first is a
question about what is required by Kant’s duty of assistance. Questions like
these are often framed as questions about how demanding Kant’s duties of
assistance are. The second question is prompted by Kant’s repeated observa-
tion that successful performance of duties of assistance puts the recipient of
assistance in a morally precarious situation of dependence and obligation to her
benefactor. Kant even suggests that agents have a duty to themselves to avoid
needing or accepting the help of others. But what, precisely, is the moral
danger or failure associated with accepting help? I consider these questions
together because there is a straightforward sense in which they problematize
one another. On its face, it would seem that the more demanding Kantian
duties of assistance are, the more opportunity there is for problematic relations
of dependence to emerge or develop. It is this tension with which this chapter
is fundamentally concerned. My aim is to point to some important and
perhaps overlooked elements of Kant’s account of duties of assistance that
K. Moran (*)
Department of Philosophy, Brandeis University, Waltham, USA
e-mail: kmoran@brandeis.edu

308 K. Moran
may help to shed light on these strands of Kant’s thought and the way in which
they appear to put pressure on one another.
Two questions about the duty of assistance

Let us begin with an overview of the questions that will help to guide the
discussion in this chapter. Before we set out, a brief note about terminology
is in order: throughout, I will be using the phrase “duty of assistance” to
describe in generic terms the imperfect duties that Kant argues we have
toward other agents. I use this terminology because the more frequently used
term “beneficence” has a specific and often-overlooked meaning in Kant’s
texts. We will explore this detail in the section titled “Adopting the Ends of
Others as One’s Own: Benevolence and Beneficence.”
Duties of assistance and demandingness
Under what circumstances do agents have a duty of assistance? Do virtuous

agents have a duty to help as much or as often as they can, or does Kant’s
duty of assistance merely demand that agents sometimes help others, as they
deem necessary, appropriate, or consistent with other desires and projects?
Insofar as Kant offers answers to these questions in his texts, he tends to offer
them by way of contrasting imperfect duty to others with perfect duty to
others. But, as noted above, this appears to leave room for a range of
interpretations. In this section, I present an overview of the textual and
philosophical points in favor of the main positions in the debate regarding
the demandingness of Kant’s theory – though much more can be and has
been said on the matter.1
As far as textual support for an interpretation is concerned, we can
trace the debate to a handful of remarks that do more, perhaps, to
muddy the issue than to clarify it. In a footnote in the second section
1
Important discussions of the topic include: Marcia W. Baron, Kantian Ethics Almost without Apology
(Ithaca: Cornell University Press, 1995), 21–81; David Cummiskey, Kantian Consequentialism (New
York: Oxford University Press, 1996), 105–23; Thomas E. Hill, Jr., “Kant on Imperfect Duty and
Supererogation,” in Dignity and Practical Reason in Kant’s Moral Theory (Ithaca: Cornell University
Press, 1992), 145–75; Thomas E. Hill, Jr., “Meeting Needs and Doing Favors,” in Human Welfare and
Moral Worth: Kantian Perspectives (Oxford: Oxford University Press, 2002), 201–43; and Jens
Timmermann, “Good but Not Required? – Assessing the Demands of Kantian Ethics,” Journal of
Moral Philosophy 2, no. 1 (2005): 9–27.
14 Demandingness, Indebtedness, and Charity 309
of the Groundwork, Kant notes that he “understand[s] by a perfect duty

the one that allows of no exception to the advantage of inclination”
(G 4:421n).2 An apparent implication of this claim is that he under-
stands by imperfect duty one that does allow for exception for the sake of
inclination.3 We cannot, in other words, refuse to abide by perfect duty
because of some competing inclination; but perhaps we can decide
against assisting another in some particular instance for the sake of
inclination. A similar thesis may appear to find support in Kant’s later
description of the distinction between ethical duties and duties of right
in the introduction to the Doctrine of Virtue. Ethical duties, Kant
explains, prescribe “only the maxim of actions, not actions themselves”
(MM 6:390, emphasis added). As such, Kant concludes, ethical duties
are duties of “wide” obligation that leave “a playroom (latitudo) for free
choice in following (complying with) the law” since “the law cannot
specify precisely in what way one is to act and how much one is to do by
the action for an end that is also a duty” (MM 6:390).4
But what might Kant mean by leaving “playroom” for free choice in
following the law? One available interpretation is that wide duties leave
agents the freedom to choose against helping in some particular case for
the sake fulfilling a competing inclination. But alternative interpretations are
also available. Kant might also mean to suggest that wide duties admit of
some freedom of choice and judgment in deciding whether a moral principle
applies in a certain case. At a minimum, Kantian agents must balance their
various obligations of assistance against each other, and against their imper-
fect duties of self-perfection toward themselves. These competing token
obligations make it the case that agents must consider how best, from their
particular standpoints, to fulfill their imperfect duties toward themselves and
others generally. Further, once an agent decides to adopt the end of assistance
in a particular case, wide duty may allow her the freedom to choose the
means toward satisfying this obligation.5
2
Citations from the Groundwork refer to Immanuel Kant, Groundwork of the Metaphysics of Morals, rev.
ed., trans. and ed. Mary J. Gregor and Jens Timmermann (Cambridge: Cambridge University Press,
2012).
3
But this is not the only available interpretation. See Timmermann, “Good but Not Required,” 16.
4
Strictly speaking, Kant should not help himself to the identity of ethical duty and wide duty. Strict
duties to the self seem also to be ethical duties, and these admit of no latitude. Still, for our purposes, a
looser point stands: insofar as no specific action is dictated by duty, there is some “playroom” for free
choice in determining the action.
5
I take this threefold distinction from Hill, “Kant on Imperfect Duty and Supererogation,” quoted in
Baron, Kantian Ethics Almost without Apology, 30.
310 K. Moran
Thus the passages cited above do not necessarily suggest that Kant thinks
that inclination can provide a counterweight to the demands of the duty of
assistance. Indeed, in a passage that follows immediately upon his discussion
of “playroom” and latitude for free choice, Kant seems eager to warn against
such an interpretation, noting that “a wide duty is not to be taken as
permission to make exceptions to the maxim of actions but only as permis-
sion to limit one maxim of duty by another” (MM 6:390). Here, Kant may
mean here that the only “permission” we have not to perform some bene-
ficent act is that we have made it our maxim to perform some other
beneficent act. But some argue that we need not interpret the passage so
strictly. Thomas Hill, for example, emphasizes Kant’s distinction between
the “maxim of actions” and the “maxim of duty” in this passage and argues
that the former is “an indefinite maxim to make the happiness of others an
important end.” As Hill sees it, Kant simply means to suggest that “we may
never exempt ourselves” from any of these more general obligations.6
Perhaps Kant’s text itself leaves us with no clear answer to questions
concerning the demandingness of his theory, but philosophical considera-
tions regarding his normative and metaethical theory may help move us
closer to a conclusion. Here, it will be useful to approach the question by
considering whether Kant’s theory admits of a category of supererogatory
action – that is, action that is good, but not required.7 If we take Kant’s
theory to have no room for the supererogatory, then we can conclude that his
is a relatively demanding theory: any good or beneficent action an agent
performs will be one whose maxim is required by duty.
Again, the question cannot be settled definitively in this section of the
chapter, but at least two observations argue against the idea that Kant’s
theory admits of the supererogatory. First, as Marcia Baron notes, Kant
repeatedly warns against thinking of moral action in terms of heroic deeds
and “noble, sublime, and magnanimous” action (CPrR 5:84).8 This is a
theme that arises especially in Kant’s discussions of moral education – pupils
should not be led to think that virtuous action is only the realm of those with
uncommon and exceptional moral fortitude (CPrR 5:155). Virtue is a task
6
Hill, “Meeting Needs and Doing Favors,” 222.
7
The concern that there is no room for the supererogatory in Kant’s ethics need not be identical with
the concern that his theory is too demanding. Sometimes, when critics complain that Kant’s theory does
not admit of the supererogatory, they do so in in the context of arguing that it is so minimal that it
excludes important spheres of value. See Baron, Kantian Ethics Almost without Apology, 21–22 for an
overview of these criticisms.
8
Baron, Kantian Ethics Almost without Apology, 38–39.
for every agent. Indeed, most virtuous of all would be for an agent to think of
his imperfect obligations as if they were owed duties: as one brings “the
maxim of complying” with imperfect duty closer to perfect duty, “so much
the more perfect is [one’s] virtuous action” (MM 6:390). If anything, this
would seem to be the opposite of making room for the supererogatory.
A second, metaethical point also argues against the possibility of the
supererogatory in Kant’s ethics. This has to do with what is sometimes
dubbed the “silencing” feature of pure practical reason.9 The point, roughly,
is that moral considerations do not simply compete with nonmoral consid-
erations, in the way that prudential considerations compete with one
another. In the case of prudential reasoning, one may ultimately decide
that one course of action or end is better all things considered, but the
value of other ends and actions remains – it is simply overridden by other
values in an agent’s judgment. But moral reasoning is different: the judgment
that something is good to do silences or normatively undermines other
considerations. Once we judge something the right or good thing to do,
any conflicting consideration loses its value. Among other things, this fact lies
at the foundation of Kant’s famous claim that only happiness that is con-
sistent with morality has any worth (see, e.g., G 4:393). What does this mean
for debates surrounding demandingness and the supererogatory? Simply put,
once we correctly judge that assisting would be good, no nonmoral end or
value can compete. If I judge that sending a donation to an aid organization
is the thing I should do, my desire to spend the money on one of my own
prudential ends no longer exerts any rational force.
It would seem, then, that there is reason to suspect that Kant’s moral theory
is quite “demanding,” and, I think there is a sense in which this is true. Later in
this chapter, however, I will point out some ways in which this claim is often
misunderstood or misrepresented. Before proceeding to that discussion, let us
consider another curious feature of Kant’s account of assistance, namely, his
claim that accepting or needing help is a morally problematic thing to do.
Dependence and obligation
In the Doctrine of Virtue, Kant opens his treatment of duties to others by

pointing to a distinction between the types of duties that one can have
toward others:
9
For this argument, see Timmermann, “Good but Not Required,” 12–15.
312 K. Moran
The chief division can be this: into duties to others insofar as you obligate them
by your performance and duties to others the observance of which does not at
the same time result in obligation on the part of others. – Performing the first is
meritorious (in relation to others); but performing the second is fulfilling a
duty that is owed. (MM 6:448, translation modified)
Kant’s reference is to the distinction between duties that are owed to others
and duties to others that are no less obligatory, but not owed. In the Doctrine
of Virtue, Kant dubs the latter sort of duty a “duty of love,” and characterizes
it as the duty to adopt the ends of others as one’s own.10 Kant notes a few
lines later that “by carrying out the duty of love to someone I put another
under obligation; I make myself deserving from him” (MM 6:450). In
fulfilling or carrying out a duty of assistance toward another, we earn merit
in relation to the recipient and put him under a kind of obligation.
It is worth noting that Kant appears to employ two slightly different
conceptions of “merit” throughout his texts.11 The first of these we might
call “absolute merit” – it is a type of merit earned by or ascribed to an agent
who fulfills obligations of imperfect duty. Kant often refers, for example, to
imperfect or “wide” duty as “meritorious duty” (see, e.g., G 4:424, 430; MM
6:227). Absolute merit is the type of merit associated with the “+1” that Kant
sometimes ascribes to the performance of imperfect duty, where “0” equates
to simply avoiding violations of perfect duty, and “-1” equates to running
afoul of perfect duty (see, e.g., MM 6:390). There is thus a sense in which
absolute merit is non-scalar: one either deserves it or not, depending on
whether one has fulfilled her obligations of imperfect duty.
But Kant also seems to employ a subtly different notion of merit – one
that we might call “relative merit,” or, as he dubs it in the passage above,
merit “in relation to others.” The degree and quality of merit “in relation to
others” relies to some extent upon information about the particulars of an act
of assistance. So, for example, Kant notes that “the greater the natural
obstacles (of sensibility) and the less the moral obstacle (of duty), so much
the more merit is to be accounted for a good deed, as when, for example, at
10
There is some ambiguity in the text regarding what the duty of love is. Kant begins his discussion by
calling active benevolence “the duty of love in particular” and pointing out that the performance of such a
duty puts others under obligation (MM 6:450). But shortly thereafter, Kant classifies beneficence,
gratitude, and sympathetic participation as duties of love (MM 6:452). This is a curious claim, since
gratitude, especially, does not seem to fit the description of a duty whose performance puts others under
obligation.
11
For further discussion of merit, see Robert N. Johnson, “Kant’s Conception of Merit,” Pacific
Philosophical Quarterly 77, no. 4 (Dec. 1996): 313–37.
considerable self-sacrifice I rescue a complete stranger from great distress”

(MM 6:228). And merit in relation to others can become tinged with
bitterness or regret depending on the response of the recipient. Typical acts
of assistance generate a kind of moral pleasure, but “bitter merit” arises in the
benefactor when an act of assistance is met with unappreciativeness or
ingratitude. Interestingly, Kant argues, “in this case the merit would be
greater still” (MM 6:391).
The preceding discussion of merit is important when it comes to under-
standing the type of obligation that Kant thinks is generated by acts of
assistance. As a starting point, we can think of this obligation in terms of
the debt of gratitude owed in response to an act of beneficence. Echoing
some of the preceding observations about merit in relation to others, Kant
remarks that the “degree” of gratitude owed “is to be assessed by how useful
the favor was to the one put under obligation and how unselfishly it was
bestowed on him” (MM 6:456). It would thus seem as though the intensity
of the obligation under which recipients stand to their benefactors is
informed in part by the merit that the benefactor has in relation to the
recipient.
When it comes to the debt of gratitude itself, Kant describes this as an
unending obligation. One cannot rid oneself of the obligation because the
benefactor always has “priority of merit, namely having been the first in
benevolence” (MM 6:455). Crucially, this obligation is also accompanied by
a shift in the relative status of benefactor and recipient. The recipient has a
duty to honor the benefactor, and Kant notes that the feeling attached to this
duty is a feeling of respect (MM 6:454). Of course, it would be far from
virtuous for the benefactor to attempt to wring honor and respect out of the
recipient, or to remind the recipient of his obligations. But such obligation
exists nevertheless.12
The clearest indication that acts of assistance have an effect on the relative
standing between benefactor and recipient might be found in Kant’s discus-
sion of friendship, in particular, Kant’s warnings against needing or offering
assistance within a so-called moral friendship – a type of friendship char-
acterized in part by an equality of love and respect between friends (MM
6:469). Kant remarks that the parties’ willingness to help one another should
never be “put to the test since this is always dangerous” (MM 6:471).
12
Excellent discussions of Kant’s account of gratitude include Barbara Herman, “Being Helped and Being
Grateful: Imperfect Duties, the Ethics of Possession, and the Unity of Morality,” Journal of Philosophy 109,
nos. 5–6 (May–June 2012): 391–411; and Houston Smit and Mark Timmons, “The Moral Significance of
Gratitude in Kant’s Ethics,” Southern Journal of Philosophy 49, no. 4 (Dec. 2011): 295–320.
314 K. Moran
Instead, each of the parties should be “generously concerned with sparing the
other his burden and bearing it all by himself, even concealing it altogether
from his friend.” Kant makes the reason for such caution clear: “if one of
them accepts a favor from the other, then he may well be able to count on
equality in love, but not in respect; for he sees himself obviously a step lower
in being under obligation without being able to impose obligation in turn”
(MM 6:471).
In fact, Kant repeatedly argues that agents have a duty to themselves
generally not to need, ask for, or accept assistance from others. Such remarks
appear as early as the mid-1770s in the lectures transcribed by Johann
Friedrich Kaehler (later copied by Georg Ludwig Collins):
it is already a breach of the duty to oneself if one accepts beneficence

[Wohlthaten]; for he who accepts beneficence creates debts that he cannot
repay; he can never even the score with his benefactor, since the latter first
did him the kindness of his own accord; if he returns the beneficence, he does it
only insofar as the other preceded him in this, and thus remains forever owing
thanks to him; but who will incur such debts? A debtor is at all times under the
constraint of having to treat the person he is obliged to with politeness and
flattery; if he does not, the benefactor soon lets him know of it, and often he
has to circumvent the latter with many false detours and greatly burden
himself. But he who pays promptly for everything can act freely, and nobody
will hamper him in doing so. (LE 27:341–42, translation modified [reading
Wohlthaten as “beneficence”])
In this early lecture, Kant suggests that the agent who does not accept help
from others can act more freely that the person who saddles himself with
such obligation. But this sense of freedom is not the moral freedom, or
autonomy, that emerges as a central piece of his theory later in the
Groundwork. Some of Kant’s concern about being indebted in the Collins
lecture may be attributable to the fact that Kant did not yet have a fully
developed notion of autonomy in the mid-1770s, when the lecture was
initially given. In the same series of lectures, Kant also endorses
Baumgarten’s distinction between active and passive obligation, noting that
the person who is obligated to pay debts stands under passive obligation and
is in this sense less free than the person who can compel himself, or who
stands under active obligation (see LE 27:268–69).
But Kant continues to argue for such a duty to oneself to avoid
accepting assistance even after his conception of autonomy emerges. In
later works, the argument for this duty tends to refer to duties regarding
honor or self-respect.13 In the Doctrine of Virtue, Kant argues that a

human being “regarded as a person” – that is, as an agent capable of
giving herself the moral law – possesses an absolute worth and dignity
which allows her to demand respect from other rational creatures, but
also obligates her to treat her own person with respect (MM 6:434–35).
Kant provides a few examples of maxims that would violate this duty to
oneself, including “not accept[ing] beneficence [one] could do without”
(MM 6:436, translation modified). In the Vigilantius lectures, Kant
similarly argues that in accepting beneficence, “the worth of humanity
in our own person is thereby diminished, and we let ourselves be put by
the other into a state of dependency” (LE 27:696).
As a partial guard against the moral dangers described above, Kant recom-
mends anonymous giving, if at all possible. If this is not possible, a benefactor
must do her best to make it seem as though the assistance is merely something
that is owed, rather than performance of wide or imperfect duty (MM 6:453).
In Kant’s discussion of friendship in the later Vigilantius lecture, for example,
he remarks that “the only way . . . to confer a benefit without injuring the
sense of honour, is to so wrap it up that it would seem a duty of friendship if
the other were to accept it, so that only a duty is being met” (LE 27:697).
The threat to one’s dignity posed by accepting unnecessary assistance leads
Kant to suggest a duty of independence, according to which one ought to
cultivate the material and psychological resources necessary not to need
assistance – except, presumably, in emergencies or other serious cases. A
recurring suggestion toward this end is that agents be on guard against
allowing mere amenities or luxuries from becoming needs. Though Kant is
relatively sanguine about the enjoyment of luxuries that encourage some
activity of mind or body (see, e.g., LE 27:397), he warns that we should
“never . . . seize upon the amenities [Annehmlichkeiten] of life with such
inclination that they can become needful to us,” since, when we do this,
these inclinations become “a burden, and restrict our freedom in the fulfill-
ment of duty” (LE 27:652).
In an illuminating fragment, Kant provides some detail about the pre-
scription to attain and maintain independence:
One prescription to be independent is that one accustom oneself to do without

that which one can expect only from the goodwill of others, e.g., to be invited
13
An extended discussion of these themes is beyond the scope of this chapter. For a helpful and detailed
discussion, see Lara Denis, “Proper Self-Esteem and Duties to Oneself,” in Kant’s “Lectures on Ethics”: A
Critical Guide, ed. Lara Denis and Oliver Sensen (Cambridge: Cambridge University Press, 2015), 205–22.
316 K. Moran
as a guest (one must be happy at home), and at most to need only the services
that one can hope to receive merely from the self-interest of others (their own
needs). This requires resources, be it power or honor or riches, in sum, having
an influence.
The second degree is that one could also do without the cooperation of others
altogether. (For it is not certain that they might need us; just like us they can
shake off their ties.)
The third is that one could do without everything that depends on luck, and
that one place one’s contentment and worth in the fact that one could never see
oneself diminished. (Ak 19:271 [R7198])
Kant’s description of various “degrees” of independence moves from a

suggestion to depend only on the self-interest of others, all the way through
the suggestion to make one’s “contentment and worth” independent of luck.
Notably, Kant does not endorse any of these three “degrees” of independence
over the others, so it is difficult to say, on the basis of this fragment, how,
precisely, Kant understands the ideal of independence. However, one thing
becomes clear in this fragment, and that is the close relationship between
Kant’s notion of independence (Unabhängigkeit) from others and what we
might call self-sufficiency, or Selbständigkeit.14 In order not to rely on the
generosity of others, one needs resources – whether this is understood in
terms of power, wealth, or influence – and ultimately the ability to place
one’s worth only in one’s standing as a creature with dignity.
Kant’s argument(s) for a duty of assistance

Kant appears to argue that we ought to help others as much as possible. But
he also views openly accepting such help as morally problematic, even
contrary to duty. Hence, Kant sometimes recommends anonymous helping,
or helping under the guise of simply doing what is owed. But this does not
resolve the tension between the two claims completely. The remainder of this
chapter will aim to relax the tension between these two claims without
denying one or the other entirely.
It will be helpful to begin by considering Kant’s arguments for a duty of
assistance. Readers may already be familiar with Kant’s “universal law”
14
“Self-sufficiency” is the standard Cambridge translation of Selbständigkeit, but the translation is
imperfect, since Kant also refers to Selbstgenugsamkeit (CPrR 5:118), which is closer to self-sufficiency.
argument for a duty of assistance in the Groundwork of the Metaphysics of

Morals. In the second section of that work, Kant concludes a complex
discussion of various types of willing (the details of which extend far beyond
the scope of this discussion) with his famous assertion that there is “only a
single categorical imperative, and it is this: act only according to that maxim
through which you can at the same time will that it become a universal law”
(G 4:421). Important for our purposes is Kant’s application of this impera-
tive to the four cases that follow, each of which corresponds to a particular
type of duty – perfect duties to self and others, and imperfect duty to self and
others. Let us examine the last of these in detail.
Kant imagines an agent who finds himself “prospering while he sees that
others have to struggle with great hardships” (G 4:423). The agent realizes
that he could help these people, but asks himself if he really has an obligation
to do so. Our agent certainly would not begrudge anybody his happiness, or
try to get in the way of another person’s happiness. He simply does not have
the immediate desire to “contribute” to the “well-being” of others, or their
“assistance in need [Beistande in der Noth].” The agent asks himself, in effect,
if a maxim of non-assistance is universalizable. Kant’s answer, unsurprisingly,
is that it is not:
Even though it is possible that a universal law of nature could very well subsist
in according to that maxim, it is still impossible to will that such a principle
hold everywhere as a law of nature. For a will that resolved upon this would
conflict with itself, as many cases can yet come to pass in which one needs the
love and compassion [Teilnehmung] of others, and in which, by such a law of
nature sprung from his own will, he would rob himself of all hope of the
assistance he wishes for himself. (G 4:423)
A roughly parallel argument appears later in the Doctrine of Virtue. Kant asks
again why it should be the case that we have a duty to adopt the ends of
others as our own. On its face, Kant admits, it is not obvious why there
should be such a duty. Indeed, the more “natural” maxim would seem to be:
“everyone for himself, God [fortune] for us all” (MM 6:452). Again, the
answer that Kant offers appeals to the fact that every rational yet sensible
agent will find himself in need from time to time:
For everyone who finds himself in need [in Noth befindet] wishes to be helped
by others. But if he lets his maxim of being unwilling to assist others in turn
when they are in need become public, that is, makes this a universal permissive
law, then everyone would likewise deny him assistance when he himself is in
318 K. Moran
need [selbst in Noth ist], or at least would be authorized to deny it. Hence the
maxim of self-interest would conflict with itself if it were made a universal law,
that is, it is contrary to duty. (MM 6:453)
Both the Groundwork and the Doctrine of Virtue passages accomplish their
argumentative goal by pointing out a contradiction in the will of a person
who universalizes a maxim of non-assistance. On the one hand, he wills a
universalized principle of non-assistance – that each person should fend for
him- or herself, as it were. On the other hand, he is a finite creature who will
inevitably find himself in need at some point in his existence. As such, he also
hopes or wishes for the help of others – that is, he also wills a principle of
assistance, or a principle that his fellow agents at least sometimes help others
in need.
It is worth saying a word or two about Kant’s terminology in these
arguments, specifically his use of the term Noth (often translated as
“need”). In contemporary German, the term denotes something like a state
of emergency. However, in Kant’s time, the term suggested something like a
state of serious need, but not necessarily dire emergency. Translating the
term as “need” is thus appropriate, though we should certainly not assume
that this need is simply a strong inclination for any object of desire.15
Though Kant’s appeal to the unavoidable neediness of rational yet sensible
agents may be his most well-known version of the argument for a duty of
assistance, one does sometimes encounter another version of the argument in
published texts and lectures. Rather than focusing on the help that any agent
would wish for in times of need, this type of argument focuses instead on
making the agent’s pursuit of happiness consistent with universal law, or
avoiding a kind of solipsism with respect to the pursuit of one’s happiness.
We find an argument like this in the Critique of Practical Reason, in the
context of Kant’s foundational assertion that only autonomy of the will can
be the basis for the moral law. In the comment that follows that assertion,
Kant observes that the distinction between heteronomous willing and auton-
omous willing can be drawn out by attending to the difference between the
object, or matter, of a volition and the ground of a volition. In heteronomous
willing, the object of a volition is also its ground: a desire for satisfaction is
also the motivating ground or basis for the action. In autonomous willing,
however, the object and ground of willing must remain distinct. Kant appeals
15
See Jacob Grimm and Wilhelm Grimm, Deutsches Wörterbuch, 16 vols. (Leipzig: Hirzel, 1854–1961),
vol. 13, columns 905–22.
to our duty to assist others, or to attend to their happiness, as an example of

this fact. Other agents desire their happiness, just as we do, but their desire
can never serve as the ground for a universal and necessary rule to assist
others. For one thing, this would impose a motivational precondition on our
duty to assist others: we would have to presuppose some “natural satisfaction
in the well-being of others” (CPrR 5:34). The obligation to assist others
must, instead, be found in the form of willing itself:
Thus the law to promote the happiness of others arises not from the presup-
position that this is an object of everyone’s choice but merely from this: that
the form of universality, which reason requires as the condition of giving to a
maxim of self-love the objective validity of a law, becomes the determining
ground of the will. (CPrR 5:34)
Though the primary aim of this passage is to elaborate the distinction between
heteronomous and autonomous willing, and not to provide an argument for a
duty of assistance, the contours of Kant’s argument are clear: one cannot
pursue one’s own happiness according to the requirements of pure practical
reason without also making the happiness of others one’s end.
A similar assertion appears later in the Vigilantius lectures on moral philo-
sophy. In that lecture, Kant discusses the notion of philautia (self-love) put
forward by Alexander Baumgarten, the author of the texts that Kant used for
his courses. Kant points out that philautia is ambiguous. It can refer to a kind
of conceitedness or arrogance in which one demands that others grant him
more respect than is due to him. Alternately, it can refer to a kind of egotism or
solipsism in which one expects or demands that others will contribute to his
happiness without his contributing to theirs. As Kant puts it in that lecture,
“there is in all men without restriction a love of well-wishing towards them-
selves, and it only becomes a fault . . . when it excludes others from our love or
inclination towards them” (LE 27:620).16 In one sense, Kant is again delineat-
ing the formal requirements of willing one’s own happiness in this passage. But
there is perhaps an elaboration of that point to be found here, too. Specifically,
we do not pursue our own happiness in isolation. Even when it is not a matter
of need or emergency, our pursuit of happiness makes demands on other
agents, and we cannot consistently make such demands without also making it
our maxim to help them in their pursuit of well-being.
16
Compare MM 6:451: “I want everyone else to be benevolent toward me (benevolentiam); hence I
ought also to be benevolent toward everyone else.”
320 K. Moran
Let us take a moment to compare these arguments. Both begin from the
perspective of a practically rational agent (that is, an agent who rationally
pursues her own interests), and both appeal to universality to move the
agent from recognizing an interest in one’s own happiness to also recogniz-
ing an obligation to have a maxim that adopts the ends of others as her
own. The argument in the Critique of Practical Reason and Vigilantius
lecture accomplishes this by pointing out the solipsism inherent in a
single-minded pursuit of happiness. The famous argument in the
Groundwork does the same thing, but with a middle step that appeals, at
least in hypothetical terms, to instrumental rationality. Specifically, the
argument appeals to the fact that one will almost certainly need the
assistance, and will, as a result, wish for the assistance of others. It is crucial
to note that the argument is not itself an instrumental argument. That is,
Kant’s claim is certainly not that one will be more likely to get help in the
future if he adopts a principle of beneficence now. Aside from appealing
illicitly to empirical premises and self-interest, it is very much an open
question whether such a claim would even be true! The crucial move in
Kant’s argument is, again, the appeal to universality – specifically the
realization that if one’s maxim of non-beneficence were universalized,
then there would be a contradiction between one’s particular will (to be
helped when confronted with inevitable need) and the universalized maxim
(that nobody ever helps anyone in cases of inevitable need).
The appeal to inevitable neediness hints at a subtle shift in standpoint
between the two arguments. The argument against solipsism in the Critique
of Practical Reason and Vigilantius lecture is addressed to an agent who – like
any of us – is attempting to pursue her own happiness. The argument against
non-assistance is also addressed to a practically rational agent pursuing his
own happiness – otherwise there would be no moral question or tension to
begin with. But the argument is more specifically addressed to the agent qua
needy agent. This difference in standpoint naturally prompts a question
about the scope of Kant’s duty of assistance: is the duty a duty to adopt
the end of others’ happiness as one’s own? Or is it a duty to attend to the needs
of others, where these are only a smaller set of the desires that contribute to
happiness generally?
The answer to this question naturally has some bearing on the aforemen-
tioned debates about demandingness. If, for example, Kant’s duty of assistance
is really only limited to cases of relatively urgent need – Notfälle – then perhaps
it is possible to say that helping actions that promote the fulfillment of less
urgent desires do, in some sense, go beyond what is required by duty. But it is
clear that a solution to questions about demandingness will not come this
easily. After all, Kant repeatedly glosses our imperfect duties toward others in
the Doctrine of Virtue as a duty to adopt the ends of others – specifically the
end of their happiness – as our own. To understand fully the scope and content
of our duties to others, we will need to get a better grasp on what it means to
adopt the ends of others as one’s own. A detailed answer to this question will
also help us along the way to resolving at least some of the puzzles about
obligation and dependence posed above.
Adopting the ends of others as one’s own:

Benevolence and beneficence
Often, the duty of assistance is glossed in the secondary literature as a duty of
beneficence, and that description is surely understandable, since Kant often
describes acts of assistance as acts of beneficence, or Wohltun (see, e.g., G 4:399;
MM 6:452). But a closer reading of the text – in particular the Doctrine of
Virtue’s discussion of duties of love – shows that there is more to the story than
this. In fact, acts of beneficence are only the end result of the duty of assistance
for which Kant argues.
How can this be the case? First, it is important to bear in mind that Kant’s
moral philosophy takes agents’ maxims – and not particular actions – as its
subject matter. When it comes to our duties of assistance, this point comes
across very clearly in Kant’s assertion in the Doctrine of Virtue that we have a
duty to adopt the ends of others as our own. The Kantian duty of assistance, in
other words, should not be understood as a duty to perform a certain set of
actions, or as many helping actions as possible, but rather as a duty to have a
helping maxim. Kant elaborates on this point when he describes our duty as a
duty to have a maxim of “active, practical benevolence” (MM 6:452). To
appreciate fully the content of this claim, we will need to place this notion of
“active, practical benevolence [Wohlwollen]” alongside several other closely
related terms – well-liking (Wohlgefallen) and beneficence, or well-doing
(Wohltun).17
Wohlgefallen, or well-liking, is simply a feeling of pleasure brought on by
another person (MM 6:212) or even, in the case of some forms of egoism,
by the thought of oneself (LE 27:620). Crucial to Kant’s understanding of
17
For a similar discussion, see Dieter Schönecker, “Duties to Others from Love,” in Kant’s
“Tugendlehre”: A Comprehensive Commentary, ed. Andreas Trampota, Oliver Sensen, and Jens
Timmermann (Berlin: de Gruyter, 2013), 309–41.
322 K. Moran
well-liking is that there can be no duty to have this sort of feeling toward a
person – he explains that to have a duty to “love that is delight” in someone
would be to “be constrained to take pleasure in something,” which is
impossible (MM 6:402). Kant admits that practicing beneficence toward
someone consistently will often have the effect of generating this type of
love for the recipient in the benefactor; still, Wohlgefallen can never be
commanded directly by duty.
Benevolence, or Wohlwollen – the term with which we are here primarily
concerned – is a trickier term in the text, and Kant does himself few favors by
utilizing the term in at least two very different ways, sometimes within the
span of several pages. We can begin by observing that while Wohlgefallen
(well-liking, taking delight in someone) is rooted in the capacity to feel
pleasure, Wohlwollen (benevolence) is rooted in the faculty of desire. In
particular, Wohlwollen – literally, “well-willing” – has to do with choosing
or wishing. But choosing and wishing are ultimately very different forms of
conative activity, in Kant’s terminology. An act of choice, or Willkür,
requires “consciousness of the ability to bring about” an object of desire. If
consciousness of this ability is absent, however, the conative activity is a mere
wish (MM 6:213). Wishing, in Kant’s terminology, thus suggests a desire for
an end for which the agent has no available means, a fortiori an end toward
which the agent can form no determinate maxim.
The types of benevolence, or Wohlwollen, mentioned in the Doctrine of
Virtue can be classified along the lines of the above distinction between
choice and wish. Thus, Kant sometimes refers to “benevolence in wishes,”
which he describes as the “benevolence present in love for all human beings.”
This type of benevolence is “the greatest in its extent, but the smallest in its
degree” (MM 6:451–52). Benevolence in wishing is a type of mild, ineffica-
cious wish that things go well for others. Since it requires no special dedica-
tion or action of the agent, it can be far-reaching in its “extent” – we can wish
everyone in the world well in this sense. As Kant puts it earlier in the Doctrine
of Virtue, “benevolence can be unlimited, since nothing need be done with it”
(MM 6:393). Indeed, so slight is the interest that benevolence in wishing
takes in others, that Kant at one point compares it to feeling, instead of
desire: it is “actually [eigentlich]” simply “taking delight in the well-being” of
others (MM 6:452, translation modified). Whether we classify benevolence
in wishing as a subspecies of feeling or desire is, for our purposes, immaterial.
Crucial to the discussion here is that benevolence in wishing does not
generate or pursue maxims concerning the happiness of others. In the
Critique of Practical Reason, for example, Kant remarks that it is “beautiful
to do good to other human beings . . . from compassionate benevolence
[aus theilnehmendem Wohlwollen],” but that this does not yet amount to a
moral maxim (CPrR 5:82).18
Kant explicitly contrasts benevolence in wishing with so-called “active,
practical benevolence,” which he describes as “making the well-being and
happiness of others my end” (MM 6:452). Unlike benevolence in wishing,
active practical benevolence is restricted in its extent, since it is impossible for
finite agents to adopt maxims that always contribute to everyone’s well-
being. Active benevolence, crucially, has to do with maxims of actions –
specifically helping actions (MM 6:450).
But even active, practical benevolence is different from beneficence, or
Wohltun. Beneficence describes beneficent action – something also indicated
by the suffix -tun, suggesting a deed or action. Of course, there is a very close
connection between active, practical benevolence (Wohlwollen) and benefi-
cence (Wohltun). Specifically, a sincere benevolent maxim will result in an act
of beneficence in most cases. As Kant describes active, practical benevolence:
“It must . . . be thought as the maxim of benevolence (practical love), which
results in beneficence” (MM 6:449, latter emphasis added). What is perhaps
Kant’s clearest distinction between practical benevolence and beneficence
emerges in the course of his discussion of gratitude in the Doctrine of Virtue.
There, Kant distinguishes between two types of gratitude – active gratitude
and affective gratitude. Active gratitude is owed to a benefactor by the person
who benefits from an act of beneficence. But, Kant points out, “Even mere
heartfelt benevolence on another’s part, without physical results, deserves to be
called a duty to virtue; and this is the basis for the distinction between active
and merely affective gratitude” (MM 6:455, middle emphasis added).
Affective gratitude, in other words, is owed in response to a benevolent
maxim that does not – for whatever reason – result in an act of beneficence.19
What sorts of circumstances might prevent a sincere maxim of practical
benevolence from resulting in an act of beneficence? Certainly any number of
empirical factors might present unexpected obstacles that stand between
benevolence and beneficence. I might have a maxim to drive a neighbor to
a doctor’s appointment, but be thwarted by a flat tire or a snowstorm.
Further, I might find that I have several maxims of practical benevolence
that conflict with one another, or with the demands of perfect duty. In such
cases, at least one of my benevolent maxims will have to cede to another, or
18
Following the translation in Immanuel Kant, Critique of Practical Reason, trans. Werner S. Pluhar
(Indianapolis: Hackett, 2002).
19
For a more detailed discussion of active and affective gratitude, see Kate Moran, “Much Obliged:
Kantian Gratitude Reconsidered,” Archiv für Geschichte der Philosophie 98, no. 3 (Sept. 2016): 330–63.
324 K. Moran
to strict obligation. Finally – and most important for the purposes of our
discussion – an offer of assistance might be refused by its intended recipient.
In such cases, my sincere maxim to help will be prevented from developing
into beneficent action by the proposed recipient’s decision not to accept help.
Toward a solution to the puzzle

In the first section of the chapter, we noted a potential tension in Kant’s
arguments and remarks regarding our duties of assistance toward others. On
the one hand, there is good reason to think that Kantian duties of assistance are,
indeed, relatively “demanding” – at the very least, Kantian duties of assistance
do not appear to admit of exception for the sake of inclination. That Kantian
duties of assistance are relatively demanding may be a contentious philosophical
claim to some; to others, it may simply be an unwelcome or burdensome
conclusion about the scope of our obligations to others. But neither of these
conclusions suggests a paradox or tension in Kant’s thought. A tension does
emerge, however, once we note that Kant also classifies being the recipient of
assistance as a morally precarious situation, at best. But if our duties of assistance
include any sort of success condition, then it would seem that the more of these
we perform, the more moral hazard we create for the recipients of assistance.
One natural solution to the puzzle would be to suppose that a duty to protect
the recipient’s self-respect provides a limiting condition on the duty of assistance.
The suggestion may not be so far-fetched: while a benefactor ought never impose a
conception of happiness upon a recipient (MM 6:454), worries about how an act
of assistance might negatively affect the recipient’s virtue may seem to be legit-
imate concerns when it comes to thinking about our duties of assistance. In the
“fragment of a moral catechism” in the Doctrine of Method of the Doctrine of
Virtue, Kant’s imagined teacher suggests that it would be morally inappropriate to
“give a lazy fellow soft cushions so that he could pass his life away in sweet idleness”
or give a drunkard “wine and whatever else he needs to get drunk” (MM 6:481).
But a few words of caution are in order about this suggestion. First, Kant’s
implied prohibition against giving a lazy man pillows in the moral catechism
is not intended as a point about how or when to perform acts of assistance,
but is rather intended to elicit the pupil’s intuitions about the relationship
between virtue and happiness and a person’s worthiness to be happy. And
even if we extend the example beyond its original context to include a point
about how and when to provide assistance others, the point seems only to be
one about the means we use in adopting the happiness of others as our own:
we should never adopt means of assisting others that demean them or tempt
them to vice. But none of the preceding should be taken to suggest that the
moral danger associated with accepting assistance is a counterweight to duties
of assistance generally. After all, Kant thinks that all beneficence – and even
benevolence – puts others under some form of obligation. If such obligation
were sufficient reason to avoid helping others, it would seem that all – or at
least most – instances of benevolence and beneficence would be ruled out
automatically.
How, then, might our observations in this chapter help resolve the tension
between demandingness and indebtedness? The first important point for us
to note is that any discussion of “demandingness” that emphasizes beneficent
acts is misleading. As we saw in the preceding section, even the most
demanding interpretation of Kantian duties of assistance requires that agents
have a maxim of active, practical benevolence. Having such a maxim implies
more than simply being pleased when things go well for others (this is well-
liking). It is also more than simply wishing others well, or hoping that things
go well for them (this is benevolence in wishing). Instead, Kant describes
active, practical benevolence as satisfying a duty to have a maxim of assis-
tance. A maxim of active, practical benevolence must be a sincere, well
developed, and reasonably efficacious maxim. It is, as Kant puts it, the sort
of maxim that typically results in a beneficent action. We can thus assume
that a general, but ill-defined, principle to help where one can does not meet
the requirements of active, practical beneficence.
Insisting that Kant’s duty of assistance is a duty to have a maxim of active,
practical benevolence – as opposed to a duty to perform beneficent actions –
may seem to be splitting hairs, but the distinction is important when it comes
to better understanding Kant’s duty of assistance and resolving the tension
described above. First, understanding the Kantian duty of assistance as a duty
to have a maxim of active, practical benevolence allows us to see how Kant’s
theory can be “demanding” in the ways described above without also imply-
ing that agents have a duty constantly to act in ways that promote the well-
being of others, or to do as much as possible to make others happy.20
Further, insofar as there is a success condition built into Kant’s duty of
assistance, the condition is that the agent has an efficacious maxim of
assistance, where this includes a sincere intent to carry out the maxim.
Kant’s account thus avoids worries regarding moral luck in one important
20
It therefore provides at least a partial response to the objection that Kantian agents must be “moral
saints.” See Susan Wolf, “Moral Saints,” Journal of Philosophy 79, no. 8 (Aug. 1982): 419–39.
326 K. Moran
respect: an agent is still virtuous if circumstances beyond her control –

including the recipient’s refusal – prevent her benevolent maxim from result-
ing in beneficent action. As Kant notes in his discussion of gratitude, even
having a maxim of active, practical benevolence “deserves to be called a duty
of virtue” (MM 6:455).21
Emphasizing that Kantian agents have a duty of active, practical benevo-
lence also makes a demanding account of imperfect duty to others consistent
with an equally stringent account of the duty that one has toward oneself to
foster independence and avoid accepting unnecessary assistance. Indeed, a
complete account of the Kantian duty of assistance includes both the
standpoint of the benevolent agent and the standpoint of the recipient.
From the standpoint of the former, the scope of obligation is quite broad.
From the standpoint of the latter, however, the set of offers of assistance one
ought to accept is much narrower. Importantly, these duties operate indepen-
dently of one another: the recipient’s duty to refuse unnecessary assistance
does not weaken the stringency of the duty of benevolence, just as a
demanding duty of benevolence does not lessen the duty to cultivate
independence.
Note that the difference in these two standpoints is also reflected in the
subtly different arguments that Kant offers for a duty of beneficence (see the
section titled “Kant’s Argument(s) for a Duty of Assistance”). Recall that in
the type of argument found, for example, in the Groundwork, Kant empha-
sizes the neediness of sensible agents in order to bring to the fore the
universality of the duty of benevolence. Crucially, the thought experiment
at the center of that argument asks the agent to imagine himself in a state of
need, thus taking up the standpoint of a potential recipient of assistance. In
contrast, the argument found, for example, in the Critique of Practical Reason
and Vigilantius lecture appeals solely to the standpoint of the potentially
benevolent agent and points out that any pursuit of his own happiness
cannot be solipsistic: we cannot expect others to participate in our pursuit
of happiness unless we also contribute to theirs. The difference in emphasis
between the two arguments might introduce a worry that each points to a
slightly different set of obligations – help in cases of serious need in the
former case, and a duty to promote happiness generally in the latter case.
However, from the preceding we can see that the arguments simply reflect
the standpoints of benevolent agent and recipient, respectively.
21
However, note that, although benevolence and beneficence both warrant the same objective merit,
successful beneficence warrants a greater degree of relational merit.
Luck, injustice, and charity

At this point in the discussion, an obvious worry remains. The preceding
account of duties of assistance according to which sincere benevolent maxims
are paired, so to speak, with a duty on behalf of the recipient only to accept
help when necessary relies on the assumption that many cases of assistance
are cases in which the recipient is in a position to be able to refuse offers of
assistance. But surely many of the most pressing cases are not like this: they
are, instead, cases in which serious need compels the recipient to accept help.
To make matters worse, precisely because the need in these cases is so severe,
it would seem that the obligation generated by the act of assistance is even
greater than it would be in those cases in which an agent accepts help
unnecessarily (MM 6:228).
We can make a few observations that help to mitigate these concerns,
though they do not erase them altogether. First, when it comes to matters of
subsistence, Kant at times seems to evince a surprising optimism about
nature’s providence and the ability of nature to accommodate the basic
needs of every person. Indeed, in his early lectures on ethics, Kant argues
that taking note of this “universal provision” of nature helps us recognize a
duty to limit our consumption of it: “On seeing . . . that the provision is
universal, I have obligations to limit my consumption, and to bear in mind
that nature has made these arrangements for everyone” (LE 27:414). Indeed,
insofar as some experience need with respect to resources, this is typically the
result of some people taking more than their fair share. As Kant puts it, “The
greatest and commonest of human miseries are due more to men’s injustice,
than to ill-luck” (LE 27:415).
Of course, these remarks appear in lectures given in the 1770s, so one
might suspect that the claims about the “provision of nature” are part of a
teleological view of nature that recedes to a certain extent in later works. This
may be the case, but Kant nevertheless continues to insist throughout the
1790s that most cases of serious need are the result of past injustice, rather
than bad luck. The following passage from the Doctrine of Virtue is typical of
such remarks:
Having the resources to practice such beneficence as depends on the goods of

fortune is, for the most part, a result of certain human beings being favored
through the injustice of the government, which introduces an inequality of
wealth that makes others need their beneficence. Under such circumstances,
does a rich man’s help to the needy, on which he so readily prides himself as
328 K. Moran
something meritorious, really deserve to be called beneficence at all? (MM 6:454;

see also LE 27:415–16; CPrR 5:155n; Ak 19:145 [R6736]; LP 9:490–91)
The passage is also noteworthy for its suggestion that the injustice at the root
of most need is systematic – that is, some people are in a position to be
beneficent because of precisely those injustices that makes others need their
beneficence.
How can this passage and others like it help to fill out the picture of
benevolence and beneficence sketched in this chapter? It may seem as though
Kant is suggesting that what we sometimes take to be instances of benefi-
cence are actually a form of repayment for a debt accrued through injustice.
Though the scope of this chapter prohibits an extended discussion of the
matter, I doubt Kant’s point is so straightforward. Notably, all of his remarks
on the way that past injustice problematizes beneficence appear in the
context of virtue – typically in discussions of self-assessment and self-con-
gratulation – rather than in the context of right and justice. Note, for
example, Kant’s reference to the rich man’s pride in his allegedly meritorious
act in the passage above. Kant’s point appears to be one about ascribing merit
to actions. In certain cases of assistance – especially those involving serious
need – we cannot be sure that our apparently meritorious acts of beneficence
are not problematized to some extent by past injustice. Such an observation
ought to inform the way we conceive of beneficence under such circum-
stances. At a minimum, we should not ascribe an especially high degree of
relative merit to the benefactor. As a result, even an action that helps its
recipient a great deal under such circumstances would not be the sort of
action that generates an especially strong debt of gratitude and obligation.
This would minimize, at least to some degree, the moral danger associated
with being a recipient of assistance in these sorts of cases.
Conclusion
None of the preceding should suggest that every problematic conflict
between Kant’s account of benevolence and his concern with independence
and self-sufficiency has been resolved. For one thing, there are many cases in
which an agent is unable to accept or decline assistance because she is
incapacitated, or because time does not permit such a decision.
Nevertheless, I hope the discussion in this chapter has at least shown that
Kant’s theory can demand relatively strong obligations of assistance while
also warning against accepting help. This is possible, in the first instance,
because the duty of assistance is a duty of benevolence – a duty to adopt

sincere maxims to help, rather than a duty to perform as many helping
actions as possible. As such, the duty of benevolence is wholly consistent with
a simultaneous duty to decline help, or not to need it in the first place.
Indeed, we can even see the standpoints of benevolent agent and responsible
recipient reflected in the two types of argument that Kant sometimes offers
for a duty of assistance.22
22
Work on this chapter was generously supported by a Humboldt Foundation Fellowship.
15
Kant and Sexuality
Helga Varden
Introduction
Why do gay people want to marry? Indeed, why does anybody want to
marry; why do we not just love each other instead? Given the bad history of
marriage, especially for women, should we not simply want to avoid this
institution altogether? And if we do marry, must marriage and sexual love be
only between two persons, and do marriage contracts have to be so all-
encompassing? We might also wonder why good, healthy sexualities are so
hard to realize in societies ridden with various kinds of oppression, such as
racism, sexism, and homophobia. Why is the impossibility of living out one’s
sexuality – whether it is one’s sexual identity or one’s sexual orientation – one
of the very few things that can lead people to suicide? Why is it so difficult to
reduce the levels of domestic violence in (even affluent sections of) society,
and why is the experience of being subjected to sexual violence often so
extraordinarily damaging? Should the sale of sexual services and sexual
images be legally permitted, and should it be permitted also, or even
especially, when the ones providing the services and images are predomi-
nantly poor or otherwise seriously struggling women? Why is being truly
loved, including sexually, often experienced as grounding, in the sense of
H. Varden (*)
Department of Philosophy, University of Illinois, Urbana, USA
Department of Gender & Women’s Studies, University of Illinois, Urbana, USA
e-mail: helga.varden@gmail.com

332 H. Varden
contributing greatly to feeling genuinely at home in the world, and why is

losing a loved one something that can undo one in destabilizing ways? Why
is sexual love so unruly; why can it make us want to do the most wonderful of
things, but also make us do terrible things, to ourselves and to one another?
These are some of the many questions facing anyone working on the
philosophy of sexuality. And most of these folks have not thought, and still
do not think, that Kant – a lifelong bachelor infamous for his many
condemning statements about (especially lustful, non-procreative) sexuality
(outside of marriage) – is the place to go for help to think through these
issues. Simply, most people with genuine interests in philosophical questions
regarding sexuality, including sexual love, are not drawn to thinkers who
hold offensive, negative, and intolerant views about it. Unfortunately, Kant is
one of those in the history of philosophy who expresses such views, and often
in angry ways. In addition, Kant’s philosophy as a whole comprises a
complicated system of interlocking parts, and it takes years of hard work to
master even one or two of them, such as the argument of the Transcendental
Deduction, his account of private property, or his idea of universalizing
maxims. If any or all of this were not challenging enough, understanding
Kant’s approach to the complicated issue of sexuality requires also figuring
out how this complex and complementary system is reflected in what Kant
actually says about sexuality. Only a view of the whole picture puts us in a
position to figure out what is worth preserving in Kant’s approach to
sexuality as well as to criticize and overcome his philosophical mistakes,
especially the ones expressed in his vehement condemnation of non-straight
sexualities. Most non-Kantians, therefore, consider it overdetermined that
Kant provides nothing helpful for thinking about sexuality. Even Kantians
have thought it such a tall order to find anything useful that few have been
keen to even look.
Not for no reason, then, was one of the first brilliant papers written on
Kant and sexuality, by Barbara Herman, titled “Could It Be Worth Thinking
about Kant on Sex and Marriage?”1 It was also not coincidental that a major
claim advanced by Herman is that, in spite of the many deeply troublesome
things Kant says about sexuality, some of his central concerns are shared by
contemporary feminist philosophers, namely problems concerning sexual
objectification and the oppression of women. In this chapter, I join
Herman (and others) in arguing that, despite its many serious failures and
1
Barbara Herman, “Could It Be Worth Thinking about Kant on Sex and Marriage?” in A Mind of One’s
Own: Feminist Essays on Reason and Objectivity, ed. Louise M. Antony and Charlotte E. Witt (Boulder,
Col.: Westview, 1993): 53–72.
mistakes, Kant’s philosophy of sexuality has something to offer. I add to the

existing work by proposing that a particular strength of his approach is the
way in which Kant’s analyses of ethics, right, and human nature are inte-
grated: each element does its own independent, yet complementary, philo-
sophical work. Kant’s mistakes issue from his account of human nature, but
these can be overcome without abandoning any of his insights or the
philosophical structure of his approach.2
Setting the stage

Commonly, students are introduced to Kant’s practical philosophy by read-
ing his Groundwork of the Metaphysics of Morals. The focus tends to be on
how, according to Kant, being virtuous is to act on universalizable maxims
from the motive of duty. Most of the teaching is likely aimed at explaining
what this means, including Kant’s distinction between what he calls “perfect”
and “imperfect” moral duties. Looking only at the Groundwork, we end up
with a few famous puzzles and seemingly unsolvable problems. For example,
if Kant thinks that at all times we must treat each other as equally valuable,
what do we do in emergencies when our loved ones are in danger? Do we first
seek the universalizable rule (save all persons) and then try to justify an
exception to this general rule in order to save a loved one? If so, asks Bernard
Williams, do we not have one thought too many in our philosophical
analysis of these situations? Is it not the case that morally, including emo-
tionally, healthy persons save their loved ones just because they love them
(one thought) and not also because it is morally permissible to do so (two
thoughts)?
More generally, how does Kant understand our embodied, emotional
selves? Some worries here stem from the fact that Kant appears to put so
much emphasis on our rational, self-reflective capacities. For Kant, being
moral appears to be identified with thinking about our actions (identifying
the maxim/rule one acts on and considering whether it can hold as a
2
I have been developing this approach in the following recent pieces: “The Terrorist Attacks in Norway,
July 22nd 2011 – Some Kantian Reflections,” Norsk Filosofisk Tidsskrift/Norwegian Journal of Philosophy
49, nos. 3–4 (2014): 236–59; “Kant and Women,” Pacific Philosophical Quarterly, electronic publica-
tion, Oct. 24, 2015; “Kant and Moral Responsibility for Animals,” in Kant on Animals, ed. Lucy Allais
and John Callanan (Oxford: Oxford University Press, forthcoming); “Kant on Sex. Reconsidered”
(under review); and A Kantian Theory of Sexuality (Oxford University Press, forthcoming). This chapter,
for obvious reasons, draws especially heavily on “Kant and Women,” “Kant on Sex. Reconsidered,” and
A Kantian Theory of Sexuality.
334 H. Varden
universal law, and then acting as motivated by this reflection). From this
point of view, central phenomena of sexuality, such as sexual identity and
sexual orientation, having good sex, falling in love, and grieving and
forgiving loved ones seem not to be within reach in meaningful ways.
After all, on the one hand, we do not simply choose, after thinking about
it, to fall in love, to grieve, to be gay or straight, or to sexually identify in
particular ways. On the other hand, although more sexually experienced
people can relatively easily stop to think about what they are doing while
having sex, and although many times one ought to stop to think about what
one is doing, having good sex does involve giving oneself over to it in some
way. Reflection turns one off rather than on; having sex out of duty is
certainly not the ideal. A theory of sexuality must engage these facts in the
right ways, and surely, advocating self-reflection at all times cannot be the
way to go. On the third hand – being an alien now – if it is true that
thinking (reflection) and choice are not the whole story regarding good
sexuality, then when is choice the crucial, morally determining factor, and
why do we do things like choose to marry? Why do we not just love each
other? On the fourth hand, why does Kant say, as he does in other writings,
that marriage is limited to two persons and that it can only properly exist
between a man and a woman (as many still believe today)? And on the last,
fifth hand, what is the big deal about trade involving sexual services,
especially in cases where both parties choose it and could choose otherwise
(are not somehow forced into it)?
In recent years, as previously mentioned, some Kantians have started to
engage these questions. For example, some have joined Herman in writing
about marriage (including same-sex marriage) and the sexual objectification
and oppression of women.3 In addition, much has been written on Kant on
3
See Christine M. Korsgaard, “Creating the Kingdom of Ends: Reciprocity and Responsibility in
Personal Relations,” in Philosophical Perspectives, vol. 6: Ethics, ed. James E. Tomberlin (Atascadero,
Cal.: Ridgeview, 1992), 305–32; Pauline Kleingeld, “The Problematic Status of Gender-Neutral
Language in the History of Philosophy: The Case of Kant,” Philosophical Forum 25, no. 2 (Winter
1993): 134–50; Onora O’Neill, “Women’s Rights: Whose Obligations?” in Bounds of Justice
(Cambridge: Cambridge University Press, 2000), 97–113; Lara Denis, “From Friendship to Marriage:
Revising Kant,” Philosophy and Phenomenological Research 63, no. 1 (July 2001): 1–28; Allen W. Wood,
“Sex,” in Kantian Ethics (New York: Cambridge University Press, 2008), 224–39; Rae Langton, Sexual
Solipsism: Philosophical Essays on Pornography and Objectification (Oxford: Oxford University Press,
2009); Linda Papadaki, “Kantian Marriage and Beyond: Why It Is Worth Thinking about Kant and
Marriage,” Hypatia 25, no. 2 (Spring 2010): 276–94; Matthew C. Altman, “Same-Sex Marriage as a
Means to Mutual Respect,” in Kant and Applied Ethics: The Uses and Limits of Kant’s Practical Philosophy
(Malden, Mass.: Wiley-Blackwell, 2011), 139–64; Mari Mikkola, “Kant on Moral Agency and
Women’s Nature,” Kantian Review 16, no. 1 (March 2011): 89–111; and Carol Hay, Kantianism,
Liberalism, and Feminism: Resisting Oppression (New York: Palgrave Macmillan, 2013).
(unreflective) emotions, on affectionate love, and on forgiveness.4 In the writings

on marriage, several add to Herman by arguing that, given Kant’s basic philoso-
phical commitments regarding justice, he should have also defended same-sex
marriage. In work on the emotions, Kantians commonly utilize other thinkers,
such as Nietzsche or P. F. Strawson, to rethink Kant on these issues. In recent
work on questions concerning our social, embodied selves more generally, and
sexuality more specifically, I have proposed that we rethink Kant’s philosophy
differently by interpreting his accounts of freedom (of ethics, or virtue, and of
justice, or right) as complemented by his account of human nature, where each of
these three accounts does its own independent, philosophical work in an overall,
integrated, and coherent approach. This tack leads to two advances: First, I can
show how the various existing, relevant papers on Kant can fit coherently into one
somewhat more complex account. Second, I can answer a main philosophical and
interpretive challenge for Kantians by showing that a plausible, reconsidered
account of sexuality is consistent with Kant’s own philosophical framework. I will
begin with Kant’s view of human nature.
Kant’s account of human nature

Kant proposes that human nature consists of three conscious, integrated
structures – what he calls the (threefold) “predisposition to good in
human nature” – namely our predisposition to animality, to humanity,
and to personality.5 The predisposition to animality comprises three
conscious orientations in us: self-preservation, the sex drive, and the
social drive. Although these orientations are conscious, realizing them
as such does not require self-reflective consciousness or abstract reasoning
abilities. Consequently, preserving ourselves, having sex, and being social
I have also written on a number of these issues in the aforementioned “Kant and Women,” as well as
in both “A Kantian Conception of Rightful Sexual Relations: Sex, (Gay) Marriage, and Prostitution,”
Social Philosophy Today 22 (2007): 199–218; and “A Kantian Critique of the Care Tradition: Family
Law and Systemic Justice,” Kantian Review 17, no. 2 (July 2012): 327–56.
4
For example, see David Sussman, “Kantian Forgiveness,” Kant-Studien 96, no. 1 (March 2005): 85–
107; Lucy Allais, “Wiping the Slate Clean: The Heart of Forgiveness,” Philosophy and Public Affairs 36,
no. 1 (Winter 2008): 33–68; Lucy Allais, “Forgiveness and Mercy,” South African Journal of Philosophy
27, no. 1 (2008): 1–9; Lucy Allais, “Elective Forgiveness,” International Journal of Philosophical Studies
21, no. 5 (2013): 1–17; and Ingrid Albrecht, “How We Hurt the Ones We Love,” Pacific Philosophical
Quarterly, electronic publication, Oct. 24, 2015.
5
For Kant’s account of “the original predisposition to good in human nature,” see Rel 6:26–28. Here, I
merely summarize core, relevant features of his account. For more details, see my “Kant and Women”
and “Kant and Moral Responsibility for Animals.”
336 H. Varden
are possible also for other conscious, yet unreflective and non-reasoning
animals. That is to say, human animals have a self-consciousness that is
reflective in nature, whereas nonhuman animals do not. Instead, nonhuman
animals have only simple reflexive self-consciousness.6 Correspondingly,
human beings are capable of abstract conceptual thought (reason), whereas
nonhuman animals are not; nonhuman animal thinking is merely associative
in nature.
To illustrate this, notice that, because these animalistic ways of being
consciously oriented do not require reason (thinking through abstract con-
cepts), they are also available to us as newborns, even if in a rudimentary
form, that is, before we have developed our capacities to think reflectively.
For example, as newborns not yet capable of self-reflective being and
thought, we still seek to survive, and we find it profoundly soothing to be
cared for in affectionate, loving, sensuous ways. Correspondingly, experi-
ences such as grief and affectionate love are strong emotions whose phenom-
enological structure at least in part operates on this unreflective and
existentially fundamental level. This is why we do not choose to grieve just
as we do not choose to love affectionately. Rather, we find ourselves grieving,
and we find ourselves affectionately drawn to someone. This is also why it is
deeply good for us (existentially comforting) to be held close by someone we
love and who loves us. And, of course, although other animals do not share in
this much more complex conscious being, they do have this fundamental
predisposition to animality, which is why we can share this aspect of
ourselves with them in important ways. For example, humans and social
nonhuman animals can comfort each other; when we pick up kittens or
puppies, they and we are both comforted. Holding and being held are ways
that we affectionately affirm one another as loved, the world, and our place in
it together as safe and good – at least our shared bit of it, right now. Our
shared capacity for basic sociality enables this. A human and a nonhuman
animal can form affectionate, loving units – an “us”; they can relate to one
another as “mine” and share homes as “ours.”7
According to Kant, the second level of human being is enabled by the
predisposition to humanity, which enables us to have a sense of self and
6
By “self-reflexive” consciousness I mean an awareness internal to any thinking or doing something,
whereas by “self-reflective” consciousness I mean a thinking about what one is thinking or doing. Self-
reflective consciousness can be seen as a second-order awareness of what I am already self-reflexively
conscious of. I return to these concepts in more detail shortly.
7
I explore the differences between human and nonhuman animals in much more detail in my “Kant and
Moral Responsibility for Animals.”
involves both comparative uses of reason and rational end-setting. Realizing

this predisposition fully is possible only for rational beings, beings who both
have reflective self-consciousness and can reason.8 That is to say, at the level
of animality we are social beings in that it is existentially good for us to be
comforted by another. At the level of the predisposition to humanity, in
contrast, we are oriented by a need to be affirmed by and to affirm others as
valuable (having equal worth). Moreover, having reflective self-consciousness
opens up a range of emotions for us that involves what Kant calls “compara-
tive uses of reason,” namely feelings that reveal the importance for us of
seeing and being seen by others, such as taking joy in each other’s success and
feeling shame, jealousy, and envy. Moreover, this predisposition orients us as
rational beings in that we can set ends of our own. Human beings as such
have no single set of ends to set: because we have reason and self-reflective
consciousness, what ends we want to set is a fundamentally open question for
us as individuals to answer, in general and in particular cases.
Using language from the Groundwork, we can say that as a human being
I can be reflectively conscious of what I am doing (I can think about what I am
doing, or be aware of what Kant calls the “maxims,” or subjective rules,
governing my actions), and I can become aware that I am the one doing it
(I can think about this too, that I am doing it). Therefore, as I set out to do
something – including actions at the animalistic (self-preserving, sexual, and
social) level, such as reaching out to offer my friend a comforting hug – I orient
myself and my causal powers with respect to others by my maxim. My maxim
(the subjective rule of action) orients how I go about things in the world – what
I do and do not do – in ways compatible with obtaining a particular end
(offering my friend a comforting hug). In addition, Kant argues, when I do this
well, I do it in such a way that my action is compatible with and furthering of
rational being (mine and the others’), including by affirming those with whom
I interact as equally valuable.9 In turn, this argument connects up with another
in the Groundwork (and elsewhere), namely that acting morally involves acting
on universalizable maxims that can be both thought as universal laws for rational
8
Whether or not higher cognitively functioning nonhuman animals, such as chimps and elephants, who
are able, for example, to pass the mirror test (have a sense of self), can do some of this is a question I
cannot explore here.
9
For example, in the Groundwork, Kant writes: “A maxim is the subjective principle of acting, and must
be distinguished from the objective principle, namely the practical law. The former contains the practical
rule determined by reason conformably with the conditions of the subject (often his ignorance or also his
inclinations), and is therefore the principle in accordance with which the subject acts; but the law is the
objective principle valid for every rational being, and the principle in accordance with which he ought to
act, i.e., an imperative” (G 4:421n).
338 H. Varden
beings (consistent with our self- and other-regarding perfect duties to respect
one another as ends in oneself) and willed as universal laws for rational beings
(consistent with our self- and other-regarding imperfect duties to develop our
capacities and talents and to assist others in their pursuit of happiness)
(G 4:424). It also connects up with one of the formulations of the categorical
imperative in the Groundwork: “So act that you use humanity, whether in your
own person or in the person of any other, always at the same time as an end, never
merely as a means” (G 4:429). That is, living well for a human being involves
developing an ability to figure out which ends one wants to set given the kind of
person one is – something each of us also has to figure out – and making sure
that the ends one wants to set are compatible with, and insofar as possible
supportive of others pursuing their own ends in a similarly open-ended and
reciprocally respectful, supportive way.
The third and final aspect of the predisposition to good in human nature is
called “personality.” Having personality means that human beings have
“moral feeling,” which is a susceptibility in our nature without which we
could not be morally responsible for our actions. That is to say, I cannot only
think through what I am doing – think about whether or not I am doing
something compatible with respecting and furthering of the rational being of
myself and others – but I can also feel obliged to act in accordance with these
reflections. I can decide that there is something I must do or that there is
something I must not do simply because doing or not doing it is morally
required (right) or morally impermissible (wrong). In Kantianese, I can act as
motivated by my practical reason or “from duty.” When I do this,
I incorporate this motivation into my maxim and thereby add a distinctive,
moral “worth” or value to my actions (G 4:401). Because only human
animals, as far as we know, are capable of acting from duty (as motivated
by their reason), only human beings can be morally responsible for their
actions. Moreover, when we do something just because doing so is the right
thing to do (as determined by our reason), we act virtuously.
Notice what follows from Kant’s theory with regard to human versus
nonhuman realization of one’s nature. Our complex form of consciousness
(reflective self-consciousness) and capacity for reason (capacity to think
through abstract concepts) lengthens the relative amount of time it takes to
master most things such that we can set out to do them intentionally (act
rationally, as enabled by our predisposition to humanity). We need even
more time (and practice) to be able to assume moral responsibility for our
actions (act as motivated by our practical reason, or as enabled by our
predisposition to personality). Hence, developmentally, we begin by feeling
our physical needs for the first time outside the womb: we seek food and we
immediately respond positively to soothing, calming, sensuous displays of

affectionate love. Being comforted reveals how our predisposition to animal-
ity enables us affectionately to orient ourselves to others. We also exhibit
rudimentary exercises of causality (choice) in spontaneous and uncontrolled
movements of our limbs. As already mentioned, these kinds of orientations
and activities as such require only reflexive forms of self-consciousness (and
associative thinking); nonhuman animals can participate in them too.
However, human beings also seek to do them rationally, intentionally, and
responsibly – to love, to eat, to master one’s body, say – which nonhuman
animals do not do. For although other animals orient themselves via reflexive
self-consciousness and can think and choose associatively (or exercise what
Kant calls “animal choice”), human beings also seek to orient themselves
reflectively, which is why human choice can be free (autonomous) (MM
6:213–14; see also 226). Human beings aim to master orienting themselves
by means of abstract concepts (maxims, rules, principles); we strive to
become able to think about what we are doing (reason) by reflectively
evaluating the “form” of these rules (whether or not they can hold as
universal laws for all rational beings), and to be able to act as motivated by
these reflections whenever necessary. Due to having more complex capacities,
the task at hand – realizing one’s nature – is much more complicated for us as
human animals than it is for nonhuman animals. It is a complex process
involving learning to set our own ends in an open-ended, self-conscious way
and to assume responsibility for our choices by ensuring that we act on
principles that could hold as universalizable laws for rational beings. Thus it
takes time.
Consider as a further illustration the way we naturally develop and master
simple bodily actions. Notice that nonhuman animals manage the task of
moving about in the world, including walking, relatively quickly. For exam-
ple, foals typically manage to walk within the first hour or so after birth,
whereas kittens commonly take up to three weeks. In contrast, it takes a
human infant roughly one month just to raise its own head and six months
to have basic control over the movements of the head. Becoming proficient at
just the basics of walking usually takes a human being somewhere between
twelve and fifteen months. Moreover, we can see how humans strive to
employ concepts in learning to walk (that is, to act intentionally or ration-
ally). At first, we stagger along either held or guided by an adult. We may
then be able to do it while being held under the armpits, then by the hands,
and finally just with the aid of someone’s finger. As a natural part of this
process, adults say something like, “Does Emma want to walk to Mommy?”
with the result of suggesting or encouraging her to use this rule to govern
340 H. Varden
what she is doing (to set the end and act on the maxim of walking). Then
there is that moment when she can do all of this on her own – staggering
along to mommy – and the adults say, “Oh, look! Emma is walking!” and
then she becomes conscious of what she is doing and that she is doing it, and
then she falls down. In Kantianese, we would say that in moving from just
walking to thinking about our walking, we switch from acting on a maxim
(acting in rationally directed or intentional ways) to thinking about both the
maxim we are acting on and that we are acting on it (becoming reflectively
conscious of ourselves and our maxim). With more practice, however, we can
both walk (again, act on the maxim) and think about what we are doing as
we are doing it (be reflectively aware of what we are doing and that we are
doing it) without falling down. Indeed, we can even come to set for ourselves
the task of intentionally perfecting our walking performance (perfecting our
ability to act on the maxim of walking), including by figuring out how to
walk in very many different and new ways. These are, then, core reasons why
the task of walking for humans is much more complex than it is for nonhu-
mans. Because we have more capacities contributing to a greater complexity
of our activities, we take more time than nonhuman animals to master even
the simplest activities.
Correspondingly, although all social animals quickly pick up on funda-
mental social emotions, such as being calmed by loving, affectionate care
from others, nonhuman animals typically master all the social emotions
available to them relatively speedily, whereas it takes humans years to learn
how to minimally manage social interaction. Again, the reason is that
nonhuman animals have less complex cognitive capacities and so a more
limited range of social emotions and interactions. For example, cats and dogs
express affection by doing things such as cuddling up and offering and
seeking affectionate company when they are frightened. They engage in
play by using certain social cues; this is how they learn to be social and
affectionate with (to orient to) others. Contrarily, humans not only learn
how to do these things and learn that they are doing them, but they also seek
ways of doing them that are rational, including responsible. We engage in
social interaction by acting on maxims in an open-ended way and in ways
that are sensitive to the particular persons involved. For example, we play-
fully invent infinitely many ways of giving and receiving affection, ways of
revealing our attention to one another’s individuality. So, hurting a loved one
can result simply from interacting in a way that suggests one’s lack of
sensitivity to another’s personality and idiosyncrasies.
It is hard (a cognitively complex endeavor), in other words, for humans to
learn to set and pursue ends of our own. It is hard not only because it requires
us to be conscious in a specifically self-reflective way (by exercising an ability

to think about what we are doing right now), but because it also comes with
learning how to do something so well that we can be creative without losing
control. Being truly creative involves playing with the imagination in new
ways – ways that we do not already understand rationally. Because we are
trying to do things we have neither done nor seen anyone else do, we have to
figure it out. Sometimes we sit down to consider exactly what it is we would
like to do so that we can then imagine ways of doing it. Other times,
especially when we are acting within a framework of action that we have
mastered – such as playing soccer – we find ourselves just doing it, going with
our gut. We seek to set new ends and, ultimately, to be able to do all of it in a
way that is consistent with and affirmative of ourselves and others as free
beings, including by paying truthful attention to how we are in the world,
who we are as individuals. All of these things are hard to learn, let alone do
well. Hence most legal-political systems today assume that it takes about
eighteen years to develop full moral responsibility. And although by this age
we can be morally responsible for ourselves (generally), it is a lifetime project
to become better at it.
The inescapable problem of evil

Kant proposes that the threefold predisposition to good in human nature is
both “original” and “good,” meaning that to have human consciousness is
to have the capacities required for these structures (why they are original)
and that these structures enable lives that feel and are deeply pleasurable for
us (why they are good) (Rel 6:28). Realizing the predispositions to animal-
ity and humanity is to realize a way of being that points toward what will be
affirmed upon moral reflection, that is, from the reflective point of view of
personality. Given this, why do we ever not do the right thing? Because,
says Kant, we can choose – that is, we can choose not only to act well, but
also badly. Trailing the predisposition to good in human nature is the
“propensity to evil.” All humans will choose badly at times – choose what
appears to maximize the sensation of pleasure in the moment – rather than
what is truly good for us. Indeed, not only do we have trouble doing the
right thing, but we also experience a real temptation to set the right kind of
end though with a bad motivation, such as to help someone across the
street (a good end) but only so that others will think well of us (a bad
motivation).
342 H. Varden
In addition, because of our embodied, social natures in combination with

our capacity for choice, we have a deep temptation to do bad things that
make us feel intensely alive, that give us an enlivened sense of self, by being
seen by others as valuable, desirable, important, and powerful. For example,
it is deeply tempting to take advantage of other people’s sexual attraction to
us, to yield to feelings of jealousy, to punish loved ones for presumed lack of
attention, or to ignore people who love us to get more attention (by their
working harder to get our attention). Despite how tempting these actions
may be, our predisposition to good as a whole enables us to avoid or stop this
sort of behavior: our capacity for animality enables us to feel that such
behaviors are incompatible with basic sociality; our capacity for humanity
enables us to identify the maxim of our action and then feel the pain and
irrationality of acting in such ways that treat others as not equally valuable,
rational beings; and our capacity for personality enables us to avoid bad
behavior just because it is wrong to do it and to engage in moral behavior just
because it is the right thing to do. That is, our capacity for humanity enables
us to stop and figure out what we are doing, and our capacity for personality
enables us to do something rather than something else just because doing so
is the right thing to do. Hence, because the threefold predisposition to good
in human nature is original and good, despite what we de facto feel like or
want to do in the moment, we are equipped with unreflective and reflective
emotional and reasoning resources that we can rely on as we seek to do
better. It makes sense, then, that Kant argues that our predisposition to
personality can never be corrupted since it is only good – it simply is an
ability to do what is right because it is right – and, although the emotional
structures as enabled by the predispositions to animality and humanity can
be profoundly off-kilter or messed with (through accidents and the choices of
others and ourselves), they cannot be destroyed since they are original.
Consequently, although we are necessarily emotionally imperfect, we can
revisit old wounds to examine the patterns of our pathologies in order to
heal, with or without the guiding help of others.
This account of human nature is both a happy and an unhappy story.
On the unhappy side, we find that as we grow up and mature, we are
subjected to other people’s behavior and pathologies. By the time we are
capable of moral responsibility (have realized personality), we inevitably
have at least some unhealthy, acquired emotional difficulties. Assuming
responsibility for our emotional shortcomings and our related wrong-
doing is also not something easily or readily done, even if we know we
ought to do it, and even if deep down we want to do it. In addition, we
experience serious temptations to do bad things – to be little monsters in
various ways – and we succumb to these temptations. Given the complex-

ities of our emotions, including the fact that we can feel them as strong
forces, acting as we ought to and as would be truly good for us can be
very hard. On the happy side, insofar as we have developed our cognitive
capacities sufficiently to be rational and morally responsible, we can
assume responsibility for our emotional shortcomings and the resulting
behavior. We can heal, which will feel profoundly good, since what is
healed are those structures enabled by the original predispositions to
good, which are in turn the sources of our deepest pleasures.
Heterosexist assumptions in Kant’s account

of human nature
Before turning to the question of how to overcome the heterosexism in
Kant’s writings, one more aspect of his account of human nature requires
mentioning, namely the twofold role of the imagination in developing the
human sex drive. First, Kant proposes that our sex drive is fundamentally
teleological in that it is oriented by a pursuit of embodied complementarity
with regard to the project of procreation. One person seeks to be comple-
mented by another as we participate in the continuation of the species.
Second, the sex drive is oriented by our employment of the aesthetic
principles of the “beautiful” and the “sublime,” where the female is inti-
mately linked to the beautiful and the male to the sublime. For Kant, these
two normative facts entail that all morally justifiable, and emotionally
healthy, sexuality is procreative, heterosexual, and tracking certain biological
features we are born with. All other forms of sexuality are corruptions of our
nature resulting from our propensity to evil. Thus, marriage should only be
legally recognized between a man and a woman.
The heterosexist nature and sheer implausibility of these arguments in
combination with some of Kant’s rather awful statements about non-straight
sexuality are why most scholars today find his account at best to be deeply
unconvincing. Consequently, they seek elsewhere for inspiration to solve the
philosophical puzzles surrounding sexuality. Yet, I argue below, given Kant’s
own philosophical system, he should have been much more careful in his
statements about sexuality, because granting teleological and aesthetic orien-
tations to the sex drive does not automatically lead to heterosexist
conclusions. Moreover, if we let emotionally healthy, well-lived lives inform
our philosophical investigations, as we should, we find that the more
344 H. Varden
plausible Kantian accounts will not reproduce Kant’s heterosexist (and at

times angry, homophobic) claims about non-straight sexuality.
Morality and sexuality

How, then, do we overcome the apparent heterosexism of Kant’s account,
and how do we accommodate a suitably revised account of human nature
within Kant’s moral framework of freedom (of ethics and of justice)? That is
to say, how do we integrate the accounts of human nature and freedom into a
Kantian theory of sexuality? In this last section, I sketch an answer to these
questions.
Let us begin with Kant’s heterosexism. It is useful, I think, to return to the
predisposition to animality, for that is where we find Kant’s account of our
sex drive. Remember that reason is not necessarily involved in developing the
predisposition to animality. Three important consequences follow. First, the
kinds of emotions that are intimately connected with the animalistic predis-
position – self-preservation, sex drive, and basic sociality – are importantly
unreflective in nature; they do not concern things about which we simply
choose. For example, we do not choose to whom we are sexually attracted or
with whom we fall in love, and we do not choose to grieve, to forgive, or to
heal our broken hearts. In an important, irreducible sense, we plainly find
ourselves having these orientations and emotions. Moreover, how we deal
with them is also not simply a matter of choice. Engaging this part of
ourselves is importantly about tending to our own animality – to what we
need and what is deeply good for us as who we are. We should, in other
words, not moralize these normative orientations and emotions by making
them inherently reflective.
Second, because these orientations (and the emotions they enable) are
unreflective as well as first-personal (subjective) in nature, we cannot obtain
or justifiably assert unconditional or necessary truths about them; any knowl-
edge we have of them is contingent. Kant does think that we can acquire
objective knowledge about the world external to us (for we can give a
philosophical critique of pure reason), and he thinks that we can obtain
objective knowledge about our practical world and how we are morally
obliged to act (for we can give a philosophical critique of practical reason,
or freedom). He does not, however, think the same about claims concerning
human nature. A philosophical account of human nature – including how we
employ the imagination in teleological and aesthetic ways when we realize
our sexuality – cannot reach, and so should not seek, the status of necessary,
objective truths.
Third, it follows that, although we make space for human nature in our
(moral) accounts of freedom, we cannot let human nature set the framework
for freedom: the contingent cannot limit the non-contingent. Consider an
example having to do with gender. Kant thought that he had never seen female
scholarly genius. Neither was he sure that women could be great public leaders.
Both accomplishments he considered generally rare and requiring extraordin-
ary abilities for abstract, principled thought (OBS 2:232). He did argue that if
scholarly genius and public leadership express something true about what both
men and women can do, then our children’s education should take this into
account (OBS 2:231). However, Kant never presented his views about the
traditional genders (men and women) as certain, as if they could be philoso-
phically and unconditionally justified. From the beginning he was unsure
about his own judgments (OBS 2:207). Moreover, he writes in his theory of
right that women are passive citizens while men are active citizens, but he
continues by emphasizing that all citizens must be legally permitted to work
themselves into active citizens (MM 6:314–15). And in his essay on the
enlightenment encourages everyone to use and trust their own reason, including
women (WE 8:35–36). Finally, in the introduction to the Metaphysics of
Morals, Kant makes clear that, although what he calls “moral anthropology”
is central to understanding human life, it must never take the place of our
accounts of freedom (including when we construct ideals for legal-political
institutions to enable and protect freedom) (MM 6:217). The problem is that,
if it does, we risk mistaking what we believe we have or have not seen for what
is possible for free beings like us. And, of course, if we interact on the basis of
such mistaken beliefs, we interact with others in ways that are inconsistent with
respecting their freedom.10
What Kant did not seem fully to realize is that the same thing follows
when it comes to assertions about sexual identity and orientation. In this case
too, our knowledge is inherently contingent. Moreover, in writing about
sexuality, it is central that what we say expresses an awareness that we are
dealing with our unreflective, embodied being at the most fundamental,
existential level; we are writing about ourselves with regard to what we need
in order to be happy, emotionally healthy, and safe as who we are. It is crucial
to draw conclusions in careful ways, fully respectful of all that we are as
10
This way of interpreting Kant is controversial. For more, including further relevant textual references,
see my “Kant and Women.”
346 H. Varden
human beings. To put this point differently, although Kant was fairly careful
when writing about women, he was exactly the opposite when it came to
writing about non-straight sexuality. Kant says many awful things, not just in
insisting on a heterosexist language, but also in adopting a very aggressive and
condemning rhetoric, such as saying that non-straight sexual interactions are
not only “unnatural,” but involve “crimes against nature,” “unmentionable
vices,” and doing “wrong to the humanity in our own person,” so they
should be “repudiated completely” (MM 6: 277). And, as if this is not bad
enough, engaging in “unnatural” sexual behavior is supposed to “defile” and
“debase” one “beneath the beasts,” and is an act worse than suicide, since
committing suicide at least requires courage (MM 6:425).
Privilege is a wickedly nasty thing; it is bad for those who have it and
much worse for those who do not.11 Part of the difficulty of privilege is that
often it brings with it a cognitive and emotional blindness to the ease with
which those who are privileged are able to proceed in some aspect(s) of their
lives. That is, privilege seems to bring with it a real danger of a profound
unawareness of the limitations of one’s own first-personal experience when it
comes to understanding the experience of others. Thus, the privileged often
(unknowingly) proceed quite brutally in relation to those who need to live
differently in order to live well and who do not have the power to resist.
Likewise, when one writes philosophy, or theorizes in general about some-
thing in relation to which one’s own way is privileged, a common risk one
runs is that one simply does not have sufficient first-personal access to the
aspects of life one is writing about, and yet one thinks one does.
Another significant danger is the likelihood that one becomes combative
in defending a theory rather than letting actual lived lives provide an invalu-
able source of correction to what one writes. In this case, if people are
profoundly happy only when living non-straight lives, which a theory cannot
capture, the problem is likely with the theory, not with the lived lives. When
one is aware of one’s own privilege, one realizes that claims about different
kinds of lives (those deviating from one’s own and which suffer much social
condemnation) are not equally plausible. In contrast, if one denies one’s own
privilege, one feels not only justified but also invulnerable in speculating
about the consequences of living a different kind of life, such as whether or
11
Much has been written on privilege in recent years in relation to so-called “implicit bias.” See
especially the pioneering philosophical work by Jennifer Saul, “Skepticism and Implicit Bias,”
Disputatio, 5, no. 37 (Nov. 2013): 243–63; and “Implicit Bias, Stereotype Threat, and Women in
Philosophy,” in Women in Philosophy: What Needs to Change? ed. Katrina Hutchison and Fiona Jenkins
(Oxford: Oxford University Press, 2013), 39–60.
not “the others” are perverted. For example, I cannot count how many times
the first question in a talk I have given on non-straight sexualities (always
from someone whose sexuality is privileged) is this: “What about animals and
children? Is sex with them also morally permissible, according to your
perspective?” Mostly, it is not meant as disparaging, and when it is not it
displays an ignorance that is difficult to explain philosophically without
introducing the kind of cognitive and emotional blindness that privilege
brings. Other times it is put forward quite aggressively, which, I believe, is
only possible to explain on this account as coming from someone’s discom-
fort with her or his own sexuality.
In one of his amazing essays written in prison, “Some Cruelties of Prison
Life,” Oscar Wilde talks about the danger of (privileged) physicians not being
correctible by their patients. Instead of seeing that a particular prisoner is
losing his mind, the doctor responsible for treating the prisoner holds fast to
his theory, according to which the prisoner is faking it. Wilde has been
watching the tragedy unfold for a while and urges the public authorities to
interfere. He reports, “At present it is a horrible duel between [the male
prisoner] and the doctor. The doctor is fighting for a theory. The man is
fighting for his life. I am anxious that the man should win.”12 These same
problems are real for those with non-straight sexualities. According to a
recent survey, the attempted suicide rate among transpersons is 41%,
among gay and lesbians 10–20%, and among people who identify as straight
only 4.6%.13 Such numbers yield a call to proceed carefully when we are
seeking to better understand sexuality. If one’s theory is not able to accom-
modate the link between people’s sexuality, including sexual oppression, and
suicide rates, then what needs correction is the theory, not the people. And
though such statistics were not available to Kant, he did know that he talked
about actual people and actual lived lives, which made it his duty to be much
more careful than he was.
So how can we understand sexuality better? The starting point should
probably be that sexuality is a part of human experience regarding which we
do not have first-personal access to one another’s perspectives. For example,
if one is gay, in a fundamental sense, one does not understand, orient to, or
feel what it is like to be straight. Therefore, learning about others’
12
Oscar Wilde, Children in Prison and Other Cruelties of Prison Life (London: Murdoch, 1898), 15.
13
Ann P. Haas, Philip L. Rogers, and Jody L. Herman, Suicide Attempts among Transgender and Gender
Non-Conforming Adults: Findings of the National Transgender Discrimination Survey (New York and Los
Angeles: American Foundation for Suicide Prevention and the Williams Institute, Jan. 2014), william-
sinstitute.law.ucla.edu/wp-content/uploads/AFSP-Williams-Suicide-Report-Final.pdf.
348 H. Varden
sexualities requires that we listen to what others experience. Second, under-

standing sexualities precludes that we use the scientific, strictly determinis-
tic, third-personal point of view. If the theory I propose is right, we will
never find a “gay gene,” which is also what current scientific studies
confirm.14 Sexualities concern individual embodied ways in which we are
in the world with ourselves and together with others. It is not primarily
about how we think about the world from a shareable, moral perspective or
about ourselves as objects that can be studied through a shareable, strictly
causal, scientific (spatiotemporal) perspective.15 Central to sexuality, Kant
seems exactly right to say, is a way of being oriented that is aesthetic (by
utilizing principles of the beautiful and the sublime) as well as teleological
in that it is about how we view others as able to complement us in
embodied ways, ways that we find emotionally good, sexually arousing,
and deeply fulfilling. None of these considerations entail heterosexism,
however.16
Virtue and sexuality

On Kant’s own account of human nature, sexuality involves both our animal-
ity and our humanity in combination with a teleological and aesthetic employ-
ment of the imagination. It all comes together at the level of personality, which
enables us to assume moral responsibility for who we are and what we do.
What I am proposing to revise is the assumption of heterosexism at the
animalistic level, for it fails to capture the complexity of human sexuality
and so also of human nature. It assumes a limited range of sexuality that
appears impossible to explain except by appealing to the damaging work of
privilege. In addition, such expansion of the theory to include other sexualities
easily accommodates Kant’s account of virtue, or of perfect and imperfect
duties. We have perfect duties not to act on maxims that involve damaging,
14
For one relatively recent newspaper article on this – with links to related scientific research – see
Peter Tatchell, “The Latest ‘Gay Gene’ Study Gives No Comfort to Homophobes,” Telegraph (Oct. 9,
2015), www.telegraph.co.uk/news/science/11922975/The-latest-gay-gene-study-gives-no-comfort-to-
homophobes.html.
15
Of course, I am not denying that we might find common, associated clusters of features. The claim is
simply that we will not find the kind of strict necessity needed to meet what Kant thought science
delivers, namely necessary and universal laws.
16
What is entailed is that exploring sexual aspects of ourselves is an individual enterprise, even though
the theory I have proposed is one suggestion for how to proceed exploring one’s sexuality. I explore these
and other issues much more thoroughly in “Kant on Sex. Reconsidered” and A Kantian Theory of
Sexuality.
aggressive, or destructive behavior toward oneself or others, and we have

imperfect duties to develop our own sexuality and to assist others in their
pursuit of happiness in this regard, too. What sort of sex one should and
should not engage in depends, from a first-personal point of view, importantly
on what one wants/desires. Because sexuality concerns both our animality and
humanity, having kinds of sex we do not want is to act in profoundly
unhealthy and damaging ways. Moreover, virtue demands that each of us
learn to attend to our own embodied sexuality and develop our wants and
desires in ways that feel deeply good, in the sense of pleasurable and safe,
emotionally satisfying, and sexually playful and exciting. Those who are able to
attend to themselves and others in ways that are good, affirming, and devel-
oping of who we all are can be a part of this project. But ultimately it is an
endeavor that remains fundamentally one’s own in the sense that it is some-
thing for which each of us must assume moral responsibility as part of realizing
our predisposition to personality.
Barbara Herman, Christine Korsgaard, and Rae Langton, in some of
the best secondary literature on the topic, argue that one of Kant’s core
insights is that sexuality involves a danger of objectification (treating
ourselves or others as mere objects and not also as subjects with equal
worth). I fully concur that objectification is a real danger when it comes
to sexuality. Given our embodied, social natures and the challenges
involved in assuming responsibility for ourselves as free beings in relation
to other free beings, we are highly susceptible to bad forms of objectify-
ing ourselves and others. It can be the result of force and/or deception,
including self-deception, and it is more likely to occur under conditions
of oppression, as they also centrally involve dehumanization (of vulner-
able social groups). In other words, on this point, Kant’s account is
philosophically in line with core arguments put forward by Simone de
Beauvoir in her groundbreaking work, The Second Sex (1949).17
Right and sexuality

For reasons of space, I will simply sketch how the theory of sexuality as
given here is compatible with much of what has been written recently
regarding Kant’s theory of right. To start, it is important to note that,
17
For a more in-depth analysis, see my “Kant and Women,” “Kant on Sex. Reconsidered” and A
Kantian Theory of Sexuality.
350 H. Varden
for Kant, the law is concerned solely with what can be rightfully
enforced in people’s interactions. Considerations of virtue (how one
ought to act) are beside the point here, since virtue cannot be enforced.
Indeed, trying to do so would not only necessarily fail, but would be
inconsistent with respecting each person’s right to freedom. Rightful
interaction, instead, is always respectful of everyone’s innate right to
freedom, understood as their right to “independence from being con-
strained by another’s choice [Willkür] . . . insofar as [the individual’s use
of freedom] can coexist with the freedom of every other in accordance
with a universal law” (MM 6:237). The basic argument is simple: Kant
follows much of the social contract tradition in arguing that the state is
a distinctly public authority. The state is not another private person, but
a representative legal-political means through which we solve (ideal and
non-ideal) problems that we cannot solve on our own in the state of
nature. Consequently, central to Kant’s argument is the idea that the
public authority is not a private person with private interests. Rather, it
is an artificial person that must represent the will of each, and yet no
one in particular, by exercising public authority through a legal-political
framework based on universal laws of freedom. Three kinds of rights are
protected by such laws of freedom: innate rights (the right to our
bodies, to thought and speech, and to honor); private rights that are
acquired and hold between private persons (rights to private property,
contract, and what Kant calls “status” right); and public rights, which
are systemic rights, meaning claims that citizens have on their public
institutions (rather than against one another as private persons). Let us
see how this account of rights plays out with regard to sexuality,
beginning with our innate right to our own bodies.
Because we are embodied beings, one’s person and one’s body must be
regarded as in an analytic relation, meaning that one’s person and one’s
body are coextensive from the legal point of view (MM 6:249–50).18
Given this and the account of human nature above, we can understand
why, on Kant’s account, there can never be an enforceable right to
sexually interact with another person, which is why there can never be
such a thing as a right to have sex or an enforceable contract regarding
sexual interaction. To have such a right would be to have an enforceable
18
For more on the nature of this relation in Kant, see my “A Feminist, Kantian Conception of the Right
to Bodily Integrity: The Cases of Abortion and Homosexuality,” in Out from the Shadows: Analytical
Feminist Contributions to Traditional Philosophy, ed. Sharon L. Crasnow and Anita M. Superson
(Oxford: Oxford University Press, 2012), 33–57.
right to another’s person, which no one can have. Therefore, authorizing

consent is always necessary for rightful sexual interaction with others.
Nevertheless, one can obtain sexual rights involving another person.
Kant’s account of marriage (as part of his account of private right)
includes, among other things, an account of how we can acquire rights
concerning another person’s sexuality. This argument is simple too: if
two people decide that they want to become a private “us” – that is, a
legally recognized home shared by two persons as equals – then part of
this legal contract (the marriage contract) will include the promise not to
choose unilaterally to engage in sexual interactions with others. When
one marries, one promises that henceforth, sexual interactions are a
shared choice. Moreover, as many have argued in recent years,19 to
have such a right to form a legal, private “us” or home cannot plausibly
depend on one’s sexual identity or orientation in a free society; rather, it
is a basic private right we all have as citizens. To be denied the right to
marry is to be forced to stay in a condition (“the state of nature”) where
sharing a home with someone one loves cannot be legally realized or
legally protected.
Finally, as mentioned, Kant provides a separate account of systemic
justice, or public right – an account that has received much attention in
recent years.20 Particularly interesting for our purposes is this concise
argument: once the state assumes a monopoly on coercion, as it must, it
must also reconcile this monopoly with each citizen’s right to freedom,
meaning each citizen’s right not to have her or his freedom subjected to
another’s private choice but dependent only on public laws. Importantly,
this means that the state must provide unconditional poverty relief for all
citizens. It must secure each citizen’s legal access to both a safe place to exist
and to the means required for survival at all times (MM 6:325–26).21
Without such legal conditions of safety, the state fails to be a public
19
See the sources listed in note 3 for various recent ways of bringing out this point.
20
For overviews of this secondary literature, see Kyla Ebels-Duggan, “Kant’s Political Philosophy,”
Philosophy Compass 7, no. 12 (Dec. 2012): 896–909; as well as my “Immanuel Kant – Justice as
Freedom,” in Contemporary Philosophy: A New Survey, vol. 12: Philosophy of Justice, ed. Guttorm Fløistad
(Dordrecht: Springer, 2015): 213–37.
21
How to interpret Kant on this issue is of course controversial in the secondary literature. For a
criticism of the line of interpretation presented here, see Pauline Kleingeld, Kant and Cosmopolitanism:
The Philosophical Ideal of World Citizenship (Cambridge: Cambridge University Press, 2012). For an
exchange between Kleingeld and myself on these issues, see my “Patriotism, Poverty, and Global
Justice: A Kantian Engagement with Pauline Kleingeld’s Kant and Cosmopolitanism,” and Pauline
Kleingeld, “Patriotism, Peace and Poverty: Reply to Bernstein and Varden,” Kantian Review 19, no. 2
(July 2014): 251–84.
352 H. Varden
authority. That is, it fails to represent the basic rights of each citizen,
because without such conditions the possibility of some citizens being
able to exist somewhere safely and to have legal access to means is subject
to some other citizens’ private choices. That is to say, the possibility of such
basic safety and subsistence means (material resources) for any one citizen
at any one time cannot be subjected to any other citizen’s private choice,
such as another citizen’s choice to employ or provide charity for her or him
(the poor). This is why the state must guarantee unconditional poverty
relief for all its citizens.
In the context of the provision of sexual services and sexual images for
consumption, a special problem arises. A main problem with these indus-
tries is that often those providing the sexual services do so only because
they have no real choices available to them for obtaining an income. It
seems that unless the state provides unconditional poverty relief, Kant’s
position can never authorize the criminalization of supplying such ser-
vices; after all, in this scenario, the state is failing to secure basic rights to
those providing these services. Whether or not it should make the buying
of such services illegal is, I believe, a question to which there is no a priori
answer. Similarly, there is no a priori answer to the question of how a
state that does provide unconditional poverty relief (a safe place to stay
and legal access to survival means) but nothing more than this (in terms of
systemic economic justice) should go about dealing with the problem that
many of its citizens find themselves subject to conditions in which there
are no real income-generating options other than those involving selling
sexual services.
In addition, as discussed in the section on “Morality and Sexuality”
above, especially problematic in this case is that the person is engaging
in activity that is truly dangerous in that it can be very emotionally
damaging if one does not want to be doing it. With regard to this issue
as well as whether or not the buying of sex can be outlawed, I think
Kant’s position maintains the following: as the state builds a system that
protects people’s systemic rights in better ways, it must strive to make
sure that the measures undertaken do not make a safe, rightful existence
even harder for those who are already vulnerable. The aim is to seek
ways of protecting those who are providing these services as well as
possible at any particular moment – whether this involves legalizing the
buying of sex or not. Then, the task is to make sure, over time, to build
better systemic institutions that secure all citizens’ choices, enabling and
securing a condition wherein each citizen has real paths available,
according to which his or her hard work, within a reasonable amount
of time, will secure them the means required for a good life and active
citizenship.22
22
Thanks to Lucy Allais, Sarah Broadie, Rachel Bryant, Ann Cahill, Andrew Cutrofello, Chad Flanders,
Jonathan Garthoff, Carol Hay, Barbara Herman, Sari Kisilevsky, Patricia Marino, Eric J. Miller, Sasha
Newton, Jonathan Peterson, Massimo Renzo, Barbara Sattler, Laurie Shrage, David Sussman, Nicolaus
Tideman, Shelley E. Weinberg, Allen W. Wood, Ekow N. Yankah, Lorenzo Zucca, and Rachel Zuckert
for invaluable feedback on ideas along the way. And a special, big thanks to Krupa Patel and Shelley
Weinberg for helping me rid this paper of many presentational infelicities. All the mistakes that have
survived despite these people’s efforts obviously remain mine.
16
Kant in Metaethics: The Paradox of
Autonomy, Solved by Publicity
Carla Bagnoli
This chapter aims to situate Kant’s account of practical reason in metaethical

debates. First, it explains the reasons why it is legitimate and instructive to
discuss Kant’s relevance in contemporary metaethics, hence addressing some
issues about the intended scope of metaethics and its relation to practical reason
and psychology. Second, it defends an interpretation of Kant’s conception of
autonomy, which avoids some paradoxes traditionally associated with self-
legislation. Third, it shows that constructivism best captures Kant’s conception
of practical reason and of its authority. Kantian constructivism is defended as a
self-standing metaethical theory, which is designed to account for the objectivity
and unconditional authority of practical knowledge, on the basis of standards
congruent with shared subjective experience. This version of constructivism,
adopting a “dialogical” interpretation of the autonomy of rational will, explains
how finite rational agents are obligated and bound by the moral law.
C. Bagnoli (*)
Department of Studies on Language and Culture, University of Modena and Reggio
Emilia, Modena, Italy
Department of Philosophy, Classics, History of Art and Ideas, University of Oslo,
Oslo, Norway
e-mail: carla.bagnoli@gmail.com

356 C. Bagnoli
Why discuss Kant in metaethics?

Whether and how Kant’s ethics have a place in metaethics might be con-
sidered a misplaced question. The distinction between metaethics and nor-
mative ethics has been elaborated within analytic ethics, and it is based on
some presumptions about the domain of ethics and its alleged method of
philosophical investigation that are alien to Kant.1 This is particularly
apparent when metaethics is construed narrowly, as the analysis of language
applied to normative discourse.
However, there are three reasons for rejecting the narrow definition of
metaethics and engaging with larger questions about the nature of practical
rationality and moral experience.2 First, the narrow understanding of
metaethics construed as the analysis of concepts and moral discourse has
been harshly criticized since its inception. Philosophers such as H. L. A.
Hart, Iris Murdoch, and G. E. M. Anscombe questioned the very possibi-
lity of setting ethics apart from philosophical questions about the mind and
its life. These philosophers shared Kant’s claim that the sphere of action
does not narrowly coincide with its outward performance, and thus related
the study of moral concepts to “moral psychology,” broadly understood as
the philosophical study of the mind and its activities.3 This is to say that the
narrow definition of metaethics, confined to the analysis of moral language,
is questionable for reasons internal to the very agenda of analytic philoso-
phy. Or, rather, such an agenda is more nuanced and diversified than
supporters of the narrow definition insist. In fact, arguments for adopting
a broad definition of metaethics are to be situated at the defining core of
early analytic ethics and are the roots of many recent debates.
Second, the narrow definition can be questioned on methodological
grounds. While it may be useful to separate the logical study of moral
concepts from the normative principles of ethics, it has been a longstanding
1
See, for example, John Skorupski, “Aristotelianism and Modernity: Terence Irwin on the Development
of Ethics,” European Journal of Philosophy 20, no. 2 (June 2012): 312–37.
2
A third sort of reason for rejecting the narrow understanding of metaethics is that it is too limiting to
confine the scope of discussion to conceptual analysis, given the growing impact of the cognitive sciences
in the philosophical debates about mind and agency. I will not discuss this sort of consideration because
it does not directly concern Kantian scholarship.
3
H. L. A. Hart, “The Ascription of Responsibility and Rights,” Proceedings of the Aristotelian Society 49
(1948–49): 171–94; Iris Murdoch, “Metaphysics and Ethics,” in Existentialists andMystics: Writings on
Philosophy and Literature, ed. Peter Conradi (New York: Penguin, 1997), 59–175; and G. E. M.
Anscombe, Intention, 2nd ed. (Cambridge: Harvard University Press, 1963). See Carla Bagnoli, “The
Exploration of Moral Life,” in Iris Murdoch, Philosopher: A Collection of Essays, ed. Justin Broackes
(Oxford University Press, 2011), 197–225.
16 Kant in Metaethics 357
matter of contention whether such a separation is tenable. Like others in

the philosophy of science, John Rawls has questioned the claim that
conceptual issues are wholly independent of normative presumptions. In
arguing against the subordination of moral philosophy to philosophy of
language, epistemology, and philosophy of mind, he pointed out that “if
there is any relation of priority, it runs the other way, from moral theory to
moral epistemology.”4 His interest in Kant’s philosophy is importantly
related to these methodological concerns.5
This remark brings us to the third reason for adopting a broad con-
ception of metaethics, which is that the narrow definition excludes from
its scope theories of practical reason, such as Aristotle’s and Kant’s.
Rawls’s legacy is crucial to understand the significance of this question
for current debates in moral philosophy. In his Dewey Lectures “Kantian
Constructivism in Moral Theory”6 and Lectures on the History of Moral
Philosophy,7 Rawls argues for a constructivist interpretation of Kant’s
theory of practical reason, which is supposed to provide new resources
for addressing the issue of ethical objectivity. On this reading, Kant’s
conception of objectivity is neither reducible to de facto intersubjective
agreement, nor to convergence on an independent realm of moral facts. In
particular, Rawls’s constructivist interpretation makes apparent that there
are intrinsic connections between Kant’s account of moral agency and the
objective standards of practical reasoning. This claim has become central
to current debates in metaethics.
Kantian constructivism marks a departure in metaethics, the so-called
“practical turn.” It consists in rethinking “the goodness of rationally chosen
ends as a matter of the demands of practical reason rather than as a matter of
ontology,” as Christine Korsgaards puts it:
4
See John Rawls, “The Independence of Moral Theory,” in Collected Papers, ed. Samuel Freeman
(Cambridge: Harvard University Press, 1999), 291.
5
The Kantian problem of theory-laden observations and moral judgments is particularly vivid in the 1950s
and 1960s. See W. V. O. Quine, “Two Dogmas of Empiricism,” Philosophical Review 60, no. 1 (Jan. 1951):
20–43; and Nelson Goodman and W. V. Quine, “Steps Toward a Constructive Nominalism,” Journal of
Symbolic Logic 12, no. 4 (Dec. 1947): 105–22. Compare Rawls, “Independence of Moral Theory,” 288–91.
Goodman’s constructivism and Kantian constructivism share the claim that constructions are importantly
constrained, and thus they both avoid radical relativism: in some cases there are syntactic constraints
governing the manipulation of symbols, and in others there are normative constraints governing the very
activity of reasoning.
6
John Rawls, “Kantian Constructivism in Moral Theory,” in Collected Papers, 303–58.
7
John Rawls, Lectures on the History of Moral Philosophy, ed. Barbara Herman (Cambridge: Harvard
358 C. Bagnoli
The Kantian approach frees us from assessing the rationality of a choice by

means of the apparently ontological task of assessing the thing chosen. We do
not need to identify especially rational ends. Instead, it is the reasoning that
goes into the choice itself – the procedure of full justification – that determines
the rationality of the choice and so certifies the goodness of the object.8
Interpreters in favor of the practical turn do not deny the role of metaphysical
claims in Kant’s general project, but hold that his main task is to explain the
phenomena of normative authority by arguing for the productive powers of
reason in its practical use. Kant’s argument is that the bindingness of moral
obligations cannot be explicated by invoking a primitive moral ontology, but
constructed according to a constitutive norm. This is where the conception
of autonomy as self-legislation proves crucial.
An important complication of these debates is that there is no scholarly
agreement about the definition of Kant’s constructivism and its scope.9 It is
difficult to determine these issues on purely exegetical grounds. This is partly
because Kant’s philosophy is systematic and complex, and hardly reducible to
bare metaethical claims. Following Rawls’s interpretation, constructivists have
largely emphasized the anti-metaphysical aspects of Kant’s critique of practical
reason. In their view, Kant promises to vindicate reasoning, starting from a
sober conception of moral agency, which is compatible with a broadly natur-
alistic methodology.10 Such starting points are clearly not sufficient to sustain
the ambitions of classical rationalism, but it is also questionable whether they
offer enough support to Kant’s own view of moral obligations. Critics of
8
Christine Korsgaard, Creating the Kingdom of Ends (Cambridge: Cambridge University Press, 1996),
261; Rawls, “Kantian Constructivism in Moral Theory,” 354; and Rawls, Lectures on the History of Moral
Philosophy, 203, 241.
9
For a recent survey of the debate on realism in Kantian ethics, see Karl Schafer, “Realism and
Constructivism in Kantian Metaethics: Realism and Constructivism in a Kantian Context,”
Philosophy Compass 10, no. 10 (Oct. 2015): 690–701; and Patrick Kain, “Realism and Anti-Realism
in Kant’s Second Critique,” Philosophy Compass 1, no. 5 (Sept. 2006): 449–65. On the connection
between constructivism and practical cognitivism, see Stephen Engstrom, “Constructivism and Practical
Knowledge,” in Constructivism in Ethics, ed. Carla Bagnoli (Cambridge: Cambridge University Press,
2013), 133–52; and Carla Bagnoli, “Constructivism about Practical Knowledge,” in Constructivism in
Ethics, 153–82. On the different strands of realism and constructivism in Kant’s ethics, see Oliver
Sensen, “Kant’s Constructivism,” in Constructivism in Ethics, 63–81; Onora O’Neill, Constructions of
Reason: Explorations of Kant’s Practical Philosophy (Cambridge: Cambridge University Press, 1989); and
Onora O’Neill, Constructing Authorities: Reason, Politics and Interpretation in Kant’s Philosophy
(Cambridge: Cambridge University Press, 2015).
10
Arguably, Kant’s definition of autonomy is precisely the conception that a non-reductive but
naturalistic ethical theory should offer: “The ethics of autonomy is the only one consistent with the
metaphysics of the modern world” (Christine M. Korsgaard, The Sources of Normativity, ed. Onora
O’Neill [Cambridge: Cambridge University Press, 1996], 5).
constructivism argue that Kant’s moral theory stands in need of a more robust
foundation, hence leaning toward moral realism.11 Constructivists respond that
the novelty of Kant’s theory is best understood by focusing on the practical
understanding of reason, rather than searching for its ontological foundation.
However, it would be a serious mistake to conceive of the metaethical
relevance of Kant’s constructivism solely in terms of its contribution to the
realism/anti-realism debate. Kant’s constructivism is hardly reducible to any
of the existing varieties of realism and anti-realism. Furthermore, the dispute
over realism is not the sole or the most interesting problem of metaethics.12
Certainly, constructivists of all sorts deny that the objectivity of practical
knowledge can be accounted for in terms of the reality of mind-independent
moral facts. But the main focus of disagreement with realism concerns the
sources of normativity. The constructivist interpretation of Kant’s concep-
tion of practical reason aims to capture the novelty of his proposal, in stark
contrast to voluntarism and rationalism, both of which are associated with
moral realism. A distinctive merit of constructivism is that it shows how
autonomy is necessary to explain the authority of obligations.
This chapter argues for a constructivist interpretation of Kant’s theory of
practical reason, but takes seriously the complexity of Kant’s philosophy. In
contrast to other non-cognitivist varieties of constructivism, my view is that
constructivism vindicates Kant’s basic claim of objective practical knowledge,
which is distinctive of animals endowed with reason.13 My argument
revolves around a dialogical conception of self-legislation, which emphasizes
11
Critics have singled out Christine Korsgaard’s argument of humanity for attack. See William J.
FitzPatrick, “How Not to Be an Ethical Constructivist: A Critique of Korsgaard’s Neo-Kantian
Constitutivism,” in Constructivism in Ethics, 41–62; Robert Stern, Understanding Moral Obligation: Kant,
Hegel, Kierkegaard (Cambridge: Cambridge University Press, 2012); Robert Stern, “Constructivism and the
Argument from Autonomy,” in Constructivism in Practical Philosophy, ed. Jimmy Lenman and Yonatan
Shemmer (Oxford: Oxford University Press, 2012), 119–37; and Robert Stern, “Moral Skepticism,
Constructivism, and the Value of Humanity,” in Constructivism in Ethics, 22–40. A different source of
reservations against constructivism concerns the status and role of the “fact of reason.” See Bernard
Williams, “Practical Necessity,” in Moral Luck (Cambridge: Cambridge University Press, 1981), 124–31.
For a reply on behalf of constructivism, see Carla Bagnoli, “Moral Objectivity: A Kantian Illusion?” Journal
of Value Inquiry 49, nos. 1–2 (March 2015): 31–45.
12
See Carla Bagnoli, “Constructivism in Metaethics,” in Stanford Encyclopedia of Philosophy, ed. Edward
N. Zalta (Fall 2011), sect. 9, plato.stanford.edu/entries/constructivism-metaethics/; David Copp, “Is
Constructivism an Alternative to Moral Realism?” in Constructivism in Ethics, 108–32; and Nelson
Goodman, “On Starmaking,” Synthese 45, no. 2 (Oct. 1980): 213.
13
I defend a fairly robust account of “construction,” as an alternative to a proceduralist definition
that is privileged in metaethics. I side with Onora O’Neill when she writes that “Kant’s repeated use
of metaphors of construction and collaboration in his discussions of reasoning make it natural to
speak of his approach and method as constructivist, and of his aim as the construction of reason’s
authority” (Constructing Authorities, 4). For a more extensive argument, see Bagnoli,
“Constructivism about Practical Knowledge”; and Carla Bagnoli, “Emotions and the Categorical
360 C. Bagnoli
the cooperative nature of reasoning and explains how it binds finite rational
beings.
Reason and autonomy

One of the most distinctive aspects of Kant’s ethical theory is that it takes
practical reason to consist in a self-legislative activity of which all rational
beings are capable insofar as they are endowed with reason. The rational will is
subjected only to laws whose authority does not depend on particular inclina-
tions and preferences (G 4:440). This claim about the autonomy of the
rational will is recognized as Kant’s most original contribution to ethics, but
it is also widely objected to as incoherent or vacuous. To begin with, the very
idea of making laws for oneself sounds paradoxical. If the moral law is created
at will, then it does not really obligate because it leaves one free to invalidate
the law at one’s discretion. If moral obligations bind us prior to and inde-
pendently of the working of the will, the purpose of self-legislation becomes
obscure.
Many hold that to explain away the paradox it is necessary to postulate an
external source of authority. The idea of making laws for oneself seems
contradictory on the assumption that the concept of legislation requires the
legislator to be external, on pain of inefficacy.14 In fact, some critics argue
that Kant’s self-legislation implicitly relies on an external power insofar as it
is based on a transcendental standpoint. On the basis of this attribution, they
further object that the Kantian claim about self-legislation insulates moral
agents from their community, thereby encouraging an ideal of self-sufficiency
and invulnerability which is both preposterous and undesirable.15
In what follows, I argue that these difficulties are due to a mischaracter-
ization of the reflexivity involved in self-legislation. To elucidate the specific
sort of reflexivity at work in rational willing, I propose a “dialogical” model,
Authority of Moral Reason,” in Morality and the Emotions, ed. Carla Bagnoli (Oxford: Oxford
14
Allen W. Wood, Kant’s Ethical Thought (Cambridge: Cambridge University Press, 1999), 156. The
objection has been known since Aquinas, and has been voiced also by Suarez, Pufendorf, and Hobbes.
All of them think that the normative and motivational force of obligation rests on an external legislator.
See Korsgaard, Sources of Normativity, 21–27.
15
Many have objected to the ideal of self-sufficiency. See for instance Alasdair MacIntyre, Dependent
Rational Animals: Why Human Beings Need the Virtues (Chicago: Open Court, 1999); and Eva Feder
Kittay, Love’s Labor: Essay on Women, Equality, and Dependency (New York: Routledge, 1999). For the
objection of uniformity, see, for example, Joseph Raz, The Morality of Freedom (Oxford: Clarendon,
1986), 370, 376.
which is meant to emphasize a distinctive aspect of autonomy, that is, its

intersubjective nature.16 Reflexivity introduces the conceptual possibility of
others, on whose recognition autonomy rests. Articulated in dialogical terms,
the claim of self-legislation is directly related to the requirement that reasons
be universal in scope and address all relevant others as co-legislators. I call
this the “publicity requirement.” I take publicity to refer to a formal attribute
of principles, which defines their universal scope: principles are meant to
govern a plurality of agents, that is, all relevant others (namely all rational
agents). The term suggests that principles that are not universal in scope
cannot be made public, and they tend to be dogmatic and hence they lack
genuine authority.17 In constructing a reason of justification, one is asked to
think of oneself in relation to others as participating in an activity of co-
legislation, rather than as competitors in a game-theoretic fashion. This
requirement constrains the way in which rational justification proceeds,
and ensures that reasons apply authoritatively to all practical subjects, that
is, all embodied rational beings. To produce this result, publicity must work
together with practical attitudes of mutual respect and recognition of others
as equals.
Autonomy as self-legislation
Kant defines autonomy as the characteristic of the will by which it is a law to
itself (G 4:440),18 that is, the capacity of giving universal laws to the faculty
of desire (G 4:431).19 The rational will is said to be the author of the laws to
16
The term “dialogical” is also meant to emphasize that the aspect of intersubjectivity is shared by
Kantians and Hegelians. See Carla Bagnoli, “Respect and Membership in the Moral Community,”
Ethical Theory and Moral Practice 10, no. 2 (April 2007): 113–28. In accordance with the view that
constructivism vindicates the proper sort of authority, I have provided a constructivist account of
practical knowledge in Bagnoli, “Constructivism about Practical Knowledge.”
17
The term “publicity” occurs in various constructivist interpretations of Kant. Rawls refers to “public
recognition” in explaining the requirements of the categorical imperative in terms of the “law of nature”
formulation, and then in the formulation of the kingdom of ends. See Rawls, Lectures on the History of
Moral Philosophy, 171, 208–14. Rawls importantly connects the idea of publicity with the requisites for
moral membership. On publicity as a formal requirement of rational justification, see also Rawls,
“Independence of Moral Theory.”
18
On the importance of various senses of autonomy of the will, see Stephen Darwall, “The Value of
Autonomy and Autonomy of the Will,” Ethics 116, no. 2 (Jan. 2006): 263–84.
19
As O’Neill remarks, this is one rare case in which Kant uses the term Selbstgesetzegebend (“Self-
Legislation, Autonomy, and the Form of Law,” in Constructing Authorities, 123n3). Apparently, Kant
never uses the term Selbstgesetzgebung or Selbstgesetzgeber; instead, he tends to use non-reflexive terms
such as “lawgiver” and “lawgiving.”
362 C. Bagnoli
which it is subject, and subject only to those laws that it has itself legislated
(G 4:432).20 The claim about self-legislation is integral to Kant’s overall
argument for the objectivity of moral obligations. Its role is to identify our
rational nature as the locus of a special value or dignity (G 4:435–36), and to
show that it consists in the capacity to legislate universal laws (G 4:440). To
describe the nature of the rational will, it might seem sufficient to say that it
conforms to the law,21 but Kant’s more demanding claim is that the rational
will is lawlike, hence productive of authoritative claims. The distinctive
feature of the rational will is that it chooses maxims that are “consistent
with the will’s own giving universal laws.”22 Thus, the rational will must be
considered the author as well as the subject of the law, and it bears the same
normative relation as the legislator bears to the law he gives.23 One debated
question is how this can be.
The very idea of self-legislation can be dismissed by calling into question
the conceptual possibility of addressing claims to oneself. In fact, Kant
addresses this very problem in discussing the conceptual possibility of duties
to oneself. In this context, he presents two aspects of self-legislation, which
seem to lead to self-contradiction. The first concerns the constraining aspect
of self-legislation, since the agent is both constraining and bound by her
legislative activity:
If the I that imposes obligation is taken in the same sense as the I that is put
under obligation, a duty to oneself is a self-contradictory concept. For the
concept of duty contains the concept of being passively constrained (I am
bound). But if the duty is a duty to myself, I think of myself as binding and so
as actively constraining (I, the same subject, am imposing obligation). (MM
6:417)
20
Here, Kant speaks of human beings as “subject only to laws given by himself but still universal” (G
4:432).
21
The third formula commands “to do no action on any other maxim than one such that it would be
consistent with it to be a universal law, and hence to act only so that the will could regard itself as at the
same time giving universal law through its maxim” (G 4:434).
22
“The principle of autonomy is, therefore: to choose only in such a way that the maxims of your choice
are also included as universal law in the same volition” (G 4:440). On the connection between the claim
about universality in scope and the principle of autonomy stated above, see also G 4:431. In what
follows I distinguish lawlikeness as a formal requirement from publicity as a scope requirement; they are
both formal and structural requirements.
23
“A rational being belongs as a member to the kingdom of ends when he gives universal laws in it but is
also himself subject to these laws. He belongs to it as sovereign when, as lawgiving, he is not subject to the
will of any other” (G 4:433).
The second aspect of the problem concerns the efficacy of a self-imposed

requirement:
One can also bring this contradiction to light by pointing out that the one
imposing obligation (auctor obligationis) could always release the one put under
obligation (subiectum obligationis) from the obligation (terminus obligationis),
so that (if both are one and the same subject) he would not be bound at all to a
duty he lays upon himself. (MM 6:417)
Critics have focused primarily on the second aspect of the problem to show
that self-legislation is vacuous. But it is probably the first formulation that
captures the more radical problem, regarding the alternate passivity and
activity of the self-legislator. While the passage discusses a special category
of duties, the problem is pervasive, and precisely in virtue of the thesis of
self-legislation. The problem concerns, generally and fundamentally, the
very concept of obligation insofar as it is a rational requirement, generated
by the self-legislative activity of reason. In contrast to the received view that
only an independent power can preserve the authority and efficacy of
legislation, Kant argues that the rational will must find its own law: it
cannot appeal to an external power without losing its unrestricted authority.
Autonomy is thus the constitutive principle of the rational will that
explains its distinctive efficacy.
In the discussion of duties to oneself, Kant uses a strategy of disambig-
uation in order to resolve the antinomy exposed above. His argument is
that the paradoxical nature of obligations to oneself dissolves when we
distinguish two senses in which agents are implicated in the recognition of
their duties to themselves, as intelligible beings (homo noumenon) and as
sensible beings (homo phenomenon). Rather disappointingly, one might
notice, Kant’s solution appears to rest on an objectionable metaphysical
dichotomy. However, to properly understand the purpose and force of
Kant’s strategy of disambiguation, we should pay attention to his insis-
tence on forensic language.24 This is a common rhetorical device that
belongs in the voluntarist tradition, but Kant deploys it in a way that
departs from tradition.
24
The forensic language is used to explicate the nature of reason itself – for example, “This court is none
other than the critique of pure reason itself” (Axi–xii). As O’Neill points out, the pervasive occurrence
of forensic and political metaphors is an indication that Kant opposes the Cartesian and rationalist
methods of inquiry, and proposes a “political,” or public, conception of reason. See O’Neill,
Constructions of Reason, 3–28.
364 C. Bagnoli
My conjecture is that Kant avails himself of the forensic language, with

the distinctive aim of construing the practical problem as one concerning
the source of legitimate authority, in strict analogy with a court of law.25
This analogy is an explanatory device that is supposed to account for how
reason gains unrestricted authority over rational but embodied beings.
The forensic device is not designed to account solely for the case of
rational deliberation, as in the discussion of self-directed duties, but it
extends to all reflexive activities in which persons address themselves. A
case in point is conscience, which is not deliberative and forward-looking,
but judgmental and backward-looking. The problem and the solution
that Kant identifies are perfectly analogous to the deliberative case of self-
regarding duties. The way out is to construct a reflective stance, which
does not coincide with the particular stance of the evaluator:
Conscience is peculiar in that, although its business is a business of a human

being with himself, one constrained by his reason sees himself constrained to
carry it on as at the bidding of another person. For the affair here is that of
trying a case (causa) before a court. But to think of a human being who is
accused by his conscience as one and the same person as the judge is an absurd
way of representing a court, since then the prosecutor would always lose.
(MM 6:438)
The case of conscience shows that Kant’s solution of the paradox consists in
construing reflexivity dialogically. The reflective stance coincides with the
stance of somebody else, and yet it is not an external source of authority. It is
the product of an exercise of self-representation in which the agent “will,
accordingly, have to think of someone other than himself . . . as the judge of his
actions, if conscience is not to be in contradiction with itself” (MM 6:438).
The two examples show that the forensic analogy does not support, and in
fact strongly discourages any further analogy between the moral law and
positive laws. On this reading, it becomes apparent that the analogy is
produced in contrast to the voluntarist account of moral obligation, which
instead tends to make the moral law equivalent to the positive law.
The inherent duplicity of the reflective stance suffices to establish a
stricter requirement, which is put forward more explicitly in the case of
forward-looking rational deliberation. When the reflective agents try to
determine what to do, they are required to conceive of themselves as
25
See Bagnoli, “Constructivism about Practical Knowledge,” §4, esp. 168–70.
“citizens of the kingdom of ends.” Admittedly, this formulation is richer

than the previous one, but my point is simply that self-reflection requires
the agent to undertake the perspective of another. The duplicity involved in
reflexivity thus opens the conceptual possibility of thinking and acting
differently than actual incentives and beliefs suggest. On the dialogical
view I propose, this is to say that rational agents are required to think of
themselves as standing in a relation of mutual respect and recognition with
others. This relation allows them to undertake reasoning as a full activity of
co-legislation.26 In sum, the point here is that, in literal analogy with the
court of law and in contrast to the voluntarist interpretation, Kant identi-
fies the supreme power upon whose authority one judges the matter with
the workings of reason itself. The forensic analogy serves to elucidate the
general operation of practical reason, and is consistent with Kant’s distinc-
tion between authoring and legislating the moral law (see G 4:414, 454–55;
CPrR 5:32–33, 81–82; MM 6:227–28).
This hypothesis can be supported on exegetical grounds, pointing to the
different formulations of the categorical imperative and their relation to the
moral law. But it is compatible with considerably different views of rational
deliberation. For instance, Andrews Reath takes the forensic analogy to
indicate that “the distinction between subject and legislator collapses.”27
This is because the rational will bears the same relation as the legislator to
the law in virtue of the fact that the law is universal in scope. Since the law
is universal, it binds anyone – the legislator as well as the subject – so
anyone is at the same time the legislator and the subject. It is in virtue of
this collapse that practical reasoning can carry on its dual task: it produces
justifications that are offered to others and that bind others.28 I agree that
self-legislation serves the purpose of laying down the structure of practical
rationality, and to this extent it establishes the legislative authority and sets
its limits.29 To explain how this task is achieved, however, I think that the
forensic analogy must be interpreted otherwise, so as to emphasize the
separateness rather than the collapse of subject and legislator. The forensic
26
This formulation is preferable to claiming that the rational deliberator acts as a representative of the
moral community, which identifies the standpoint from which to deliberate. The formulation in terms
of co-legislators does not generate the false impression that there is already a moral community of which
the agent is a member.
27
Andrews Reath, Agency and Autonomy in Kant’s Moral Theory: Selected Essays (Oxford: Clarendon,
2006), 100.
28
For Reath, it is because of the collapse between sovereign and subject of the law that “my reasoning
binds others to recognize its validity” (Ibid., 111; see also 78, 104).
29
Ibid., 112.
366 C. Bagnoli
analogy dissolves the air of paradox surrounding the idea of self-legislation

by separating different offices. The purpose of this explanatory device is to
point out that these functions must be kept distinct in order to make sense
of authority. When the (rational) agent legislates, she legislates for all
relevant others, that is, for all rational agents, regardless where she stands,
what she wants, and what she believes.30 This is legitimate only insofar as
and in virtue of the fact that the activity of self-legislation is ultimately an
activity of co-legislation. The rational agent legislates as a member of a
community of co-legislating equals, rather than as their guardian or repre-
sentative. It is in virtue of this sort of membership that the law she
constructs carries genuine authority.
Furthermore, the dialogical view clarifies that the appeal to universality
does not merely constrain the subject’s actual desires and convictions.
Perhaps more importantly, it requires that what subjects regard as reasons
are authoritative for all relevant others. Thus, in construing such reasons,
reference to relevant others become of crucial importance. The normative
standards that the reflective agent implements are the same as they would
be in a public forum, that is, they must be recognizable by all parties as
legislated by the idea of rational will.
The implication is that the individual agent who happens to be in the
position of the legislator is not the proper source of authority; and,
certainly, he is not the “author” of the moral law, because the moral law
does not admit of an author in the sense in which positive laws do.
Consequently, the individual legislator is not at liberty to invalidate the
ruling at his discretion. Instead, the legislator (guided by reason) gives
laws that serve as normative standards for any agent in conceiving of his
action, and for any judge in forming his judgment. The source of author-
ity is the legislative activity. Thus, the forensic analogy explains away the
paradox by making apparent that the authority of judgments depends on
the unrestricted scope of their audience. In order to be authoritative,
judgments must be justifiable to a universal audience. Yet this is not to say
that reasons derive their authority from contents that are universally
acceptable, as the moral realist or the dogmatic rationalist hold. On the
contrary, the constructivist claim is that for finite rational beings, there
are no such contents independently of and prior to building up principles
30
The third formula of the categorical imperative specifies the scope of universality and the condition of
moral membership, while connecting to the other two formulas: “A rational being must always regard
himself as lawgiving in a kingdom of ends possible through freedom of the will, whether as a member or
as sovereign” (G 4:434).
that others could share as co-legislators. In short, for finite rational beings
to be governed by reason is to enter the practice of rational justification as
a public forum. Finitude and mutual vulnerability are the features inher-
ent to the human condition that rationality addresses by constructing
prescriptive principles.
A dilemma about volitional autonomy

The authority of rational justifications rests on the activity of co-legisla-
tion which pertains to all rational beings. It might be objected that
because of this feature the dialogical interpretation ultimately undercuts
the agent’s authorship of action.31 How are finite agents autonomous if
they are unconditionally constrained by reason? The dilemma about
volitional autonomy builds upon two assumptions: first, that the sort of
reflexivity implied by the idea of self-legislation indicates the capacity to
express a personal preference and determine oneself by identifying with
such a preference; and, second, that in order to grant the objectivity of the
moral law we need to postulate an external source of constraints for the
rational will. Kant regards both of these assumptions as mistaken. The
first one commits us to individualistic willing, and the second one violates
the principle of autonomy.
This formulation of the dilemma equivocates the nature of the reflexivity
implied by the idea of self-legislation. Self-legislation is a reflexive term, but it
does not indicate the mere capacity to determine oneself by expressing or
realizing one’s preference.32 On the contrary, normative principles con-
structed in pursuit of individual desires and wants of the agent fail the test
of universality. They amount to what Kant regards as “a lawless use of
reason” (OT 8:145). To create laws at one’s discretion is an instance of
individualistic willing; it is a paradigmatic form of heteronomous agency
(G 4:441, 443). Likewise, the first horn of the dilemma is not a viable option
for Kant. The sort of volitional autonomy that Kant advocates prevents us
31
“To make my own will the author of my obligations seems to leave both their content and their
bindingness at my discretion, which contradicts the idea that I am obligated by them. If we reply to this
objection by emphasizing the rationality of these laws as what binds me, then we seem to be transferring
the source of obligation from my will to the canons of rationality. The notion of self-legislation becomes
a deception or at best a euphemism” (Wood, Kant’s Ethical Thought, 156).
32
To explicate the difference, O’Neill says that Kant’s sort of reflexivity is “impersonal,” rather than
“personal” (Constructing Authorities, 131–36). But this characterization fails to dissolve Wood’s dilemma
because it concedes the polarity between the impersonal and personal conceptions.
368 C. Bagnoli
from interpreting the constraints of reason as applying externally to the will

of individual agents.
This aspect of Kant’s conception of volitional autonomy distinguishes
it from views that understand volitional autonomy as the desire to
completely adhere to divine commands.33 Since the contents of God’s
commands depend on God’s will and are susceptible to change at God’s
bidding, they are not unconditional in the sense that Kant requires.
Remarkably, desires are not the source of conditionality, in this case.
Even if desires were stable and consistent through time, necessarily
directed to the same objects, their content would still be dependent on
God’s will, hence contingent on it. To avoid this sort of dependency, it is
not enough to hold that God’s will does not change because it is a will
that loves the good insofar as it is good. This qualification corrects one
source of contingency, but it is insufficient to establish autonomy in
Kant’s sense because it makes the human will dependent on God’s will.
The wholehearted identification with an external source of authority
counts as a complete lack of autonomy.
This constructivist interpretation does not deny that the moral law could
be formulated as a divine law, insofar as such a formulation does not require
that God be the source of authority of the moral law. It is compatible with
the claim that what humans construct through reasoning is effortlessly
evident to godlike rational beings. It is also compatible with the view that
humans are bound by moral prescriptions and consider the moral law as a
regulative ideal, while rational beings regard the moral law as a law of nature.
But the constructivist interpretation denies that the moral law could be
formulated as a divine prescription, because this formulation undermines
its claim to unrestricted authority. In the latter respect, constructivism
vindicates Kant’s fundamental claim that neither God nor any individual
persons are authors of the law, since the authority and contents of the law are
grounded on reason alone.34
How is it grounded? According to the realist, human reason is
capable of discerning the good as good. If so, then the autonomy of
reason would amount to conforming to what is recognized as objec-
tively good. In this case, self-legislation is parasitic on some more
33
William of Ockham, De connexione virtutum, in Opera Theologica, ed. G. Etzkorn and F. Kelley, vol.
7 (New York: St. Bonaventure, 1984), 11.
34
Compare Karl Ameriks, “Is Practical Justification in Kant Ultimately Dogmatic?” in Kant on Practical
Justification: Interprative Essays, ed. Mark Timmons and Sorin Baiasu (Oxford: Oxford University Press,
2013), 153–75. See also Karl Schafer, “Realism and Constructivism in Kantian Metaethics.”
fundamental normative fact, and its role is only epistemic.35 The

constraints on the rational will depend on a value independent of
reason, and which reason grasps. Thus, the role of reason is to recog-
nize and respond to an independent value. As a consequence, the realist
explanation makes the authority of reason merely derivative, and hence
unstable. To avoid this result, we should bear in mind the distinction
between finite and infinite rational beings.36
In the case of finite rationality, autonomy is necessary for moral obliga-
tions to have unrestricted authority. It explains how rational human agents
are bound by some claims insofar as they legislate them, that is, through the
reasoning that makes such claims into laws. The constraints on the will are
constitutive of the very activity of rational willing, and the demands of
practical reason become binding for individual agents by attending to the
form of the law. This is a two-part explanation, which comprises the con-
stitutive norm of rationality (lawmaking as co-legislating) and its subjective
counterpart (the moral feeling of reverence for the lawmaking).
Autonomy and the recognition of others’ equal

status
The constructivist reading puts together seemingly different aspects of
rational justification: first, that the rational will is autonomous in that it is
a law to itself and it does not derive those criteria from sources of authority
alien to reason; second, that rational will is autonomous in that its activity
bears the structure of the law, that is, it is lawlike; and, third, that the rational
will is autonomous in that it operates respectfully of others’ equal status. The
task of this section is to elucidate the dialogical nature of self-legislation, in
relation to the third aspect of autonomy mentioned above.
35
This claim follows from standard realist readings. See Wood, Kant’s Ethical Thought, 30; Charles
Larmore, The Autonomy of Morality (New York: Cambridge University Press, 2008), 83–84; and Karl
Ameriks, “On Two Non-Realist Interpretations of Kant’s Ethics,” in Interpreting Kant’s “Critiques”
(Oxford: Clarendon, 2003), 263–82.
36
See Onora O’Neill, “Vindicating Reason,” in The Cambridge Companion to Kant, ed. Paul Guyer
(Cambridge: Cambridge University Press, 1992), 280–308; O’Neill, “Self-Legislation, Autonomy, and
the Form of Law,” 121–36; Thomas E. Hill, Jr., “The Kantian Conception of Autonomy,” in Dignity
and Practical Reason in Kant’s Moral Theory (Ithaca: Cornell University Press, 1992), 76–96; and J. B.
Schneewind, The Invention of Autonomy: A History of Modern Moral Philosophy (Cambridge: Cambridge
370 C. Bagnoli
To begin, let me clarify that self-legislation does not establish the status of
specific moral obligations.37 It establishes that, insofar as there are objective
moral obligations that bind everybody, they are requirements of practical
reason. As the structural requirement of practical reason, self-legislation
delimits the normative space where particular normative questions arise
and different moral demands compete. It establishes the background practice
against which agents legislate authoritatively.
Under the dialogical interpretation, the rational standard of justification
implicates others as having equal standing. All rational beings are members of
an ideal community of agents having equal standing (G 4:425–26).38 Insofar as
they have equal standing, others are entitled to ask for reasons and accept the
burden of offering reasons to us. This is to say that they stand in a relation of
mutual recognition with us as co-legislators. Rational justification as such is based
on respect and mutual recognition of others as equals. This is a conceptual point.
To make this point effective, it is necessary to specify something about the
subjects that stand in mutual relation. Humans are endowed with a distinctive
sensibility, which makes them responsive to the demands of the moral law. Such
moral sensibility is not the origin of practical knowledge, but it plays a crucial role
in accounting for the subjective experience of autonomy as it takes the form of
moral obligation. In fact, the distinctive moral sensibility explains that moral
obligations are felt unconditionally binding and thus accounts for the efficacy of
reason. It does so not by adding up to the many incentives one happens to have,
but by constraining and ranking the incentives one surveys in rational delibera-
tion (CPrR 5:76).39 For this reason, it is appropriate to take respect as the
psychological underpinning of the claim about self-legislation: it is the subjective
experience of autonomy and the sensitive condition that makes it possible for
finite rational agents to be responsive to pure practical reason, despite the pressure
of all sorts of concerns, interests, and needs.
37
The claim about self-legislation leaves open the possibility that there are no moral principles of this
form – that is, principles that all rational agents would accept; this possibility is foreclosed on other
grounds. Even so, it does not follow that no one is bound by principles with which one is not willing to
comply, or that morality does not in itself have the power to bind us unless we really want to. See
Rüdiger Bittner, What Reason Demands, trans. Theodore Talbot (Cambridge: Cambridge University
Press, 1989), 104–10.
38
“The concept of every rational being as one who must regard himself as giving universal law through
all the maxims of his will, so as to appraise himself and his actions from this point of view, leads to a very
fruitful concept dependent upon it, namely that of a kingdom of ends” (G 4:433). (Reich can be translated
as “commonwealth.”) “By a kingdom I understand a systematic union of various rational beings though
common laws” (G 4:433).
39
For a more extended argument, see Bagnoli, “Emotions and the Categorical Authority of Moral
Reasons.”
Reference to respect is necessary to sustain the specific variety of con-

structivism that can be plausibly attributed to Kant, and its task is to show
that universal and necessary normative truths can be binding for finite
rational beings. Such truths may be discernable to infinite rational beings.
But for finite rational beings, the issue of the limitation and inefficacy of
practical knowledge importantly arises. Kantian constructivism responds to
this issue in terms of the autonomous activity of reasoning, which allows
finite beings to bind ourselves to normative truths. It addresses humans as
interested in reasoning, endowed with both the appropriate dispositions to
reasoning and the subjective condition of our susceptibility to duty. Our
responsiveness to normative truths does not rest on receptive sensibility but
requires that our rational capacities be in tune with moral sensibility. In other
words, to sustain this rationalist project, reverence for the law and respect for
others as equals should be thought of as complementary claims. Together,
they explain the categorical authority of the truths of reason and their
subjective efficacy. This is to say that constructivism provides a two-part
explanation of how finite rational agents are bound by the moral law.
Publicity as a scope requirement

Stephen Darwall has objected that Kantians fail to prove that autonomy is an
inescapable assumption of practical reasoning because they fail to acknowl-
edge the role of others in the practice of moral justification.40 According to
Darwall, there is something distinctive about the authority of morality,
which depends on the authority of a person to address another and make
moral demands.41
The dialogical account of self-legislation shows that the standpoint of
practical reason does not have to be understood as transcendent in the
sense that Darwall finds objectionable.
However, the dialogical view and the second-personal approach signifi-
cantly differ as to the role they attribute to “others.” In the dialogical view,
others are constitutively implicated in the claim of self-legislation, and not
40
See Stephen Darwall, The Second-Person Standpoint: Morality, Respect, and Accountability (Cambridge:
Harvard University Press, 2006), 32, 213–42.
41
“I argue therefore that the authority to address practical reasons can take forms that are quite different
from the epistemic authority that is presupposed either by theoretical reason-giving or by other forms of
practical reason-giving, like advice, where the reasons are not second-personal” (Darwall, Second-Person
Standpoint, 59).
372 C. Bagnoli
only when they are addressed second-personally or because they are addressed
second-personally. The role of others in the dialogical account of reflection is
thus more basic than in the second-person account. It is not restricted to the
narrow domain of second-personal reasons.42 We must always assume free-
dom of reason in forming any judgment, and this assumption requires that
we justify our thoughts and actions on the basis of principles that are public,
that is, such that they always implicate others (G 4:448). This makes us
constitutively dependent on and vulnerable to one another. But, then, what
is the normative import of self-legislation?
Autonomy and vulnerability

The Kantian conception of autonomy as self-legislation is often found
perplexing because of its normative implications. One of the main objections
is that it is an unrealistic and misguided model of moral autonomy because it
represents the rational will as invulnerable and self-sufficient.43 Dependency
and vulnerability to others are defeaters of autonomy, on a par with external
determinations of the rational will. In contrast to this claim associated with
Kantian ethics, many advocate the moral value of vulnerability and praise the
“virtues of dependency” alongside the virtues of autonomy.44
The dialogical view that I have defended can still be perceived as retaining
some of the problematic features of more traditional interpretations of Kant’s
autonomy.45 By bringing to the fore the role of respect for others for the
42
The matter is complicated by the ambiguity of the expression “second-personal” as a qualification of a
speech act. Some specific obligations arise when one is explicitly addressed second-personally, for
example, when I ask you to pay attention to what you are reading. But in Darwall’s view it seems
that we incur second-personal obligations any time we are in the presence of somebody who has the
authority to issue moral demands. This is because what makes something second-personal is its
conceptual link to addressing a demand. See Darwall, Second-Person Standpoint, 59.
43
Raz, Morality of Freedom, 370n179.
44
On the virtues of dependency, see MacIntyre, Dependent Rational Animals; and Catriona Mackenzie,
Wendy Rogers, and Susan Dodds, eds., Vulnerability: New Essays in Ethics and Feminist Philosophy
(Oxford: Oxford University Press, 2014).
45
For a more extensive argument identifying the role of vulnerability in a constructivist account of
practical reasoning, see Carla Bagnoli, “Vulnerability and the Incompleteness of Practical Reason,” in
Vulnerability in Context, ed. Christine Strahele (London: Routledge, forthcoming). Compare Paul
Formosa, “The Role of Vulnerability in Kantian Ethics,” in Vulnerability: New Essays in Ethics and
Feminist Philosophy, 88–109; and Catriona Mackenzie, “The Importance of Relational Autonomy and
Capabilities for an Ethics of Vulnerability,” in Vulnerability: New Essays in Ethics and Feminist
Philosophy, 33–59.
exercise of rational activities, the dialogical account recognizes the impor-

tance of this critique. However, in contrast to relational accounts, the
dialogical view of self-legislation explains the relationality of the rational
will as a normative claim about the structure of rational agency rather than
a psychological or an ontological postulation.46 To will rationally is to think
and act on the basis of principles, and this requires that we relate to others as
having equal standing. The basic requirement for gaining membership in the
moral community is that the agents be capable of entertaining relations
marked by mutual respect and recognition. Understood in dialogical terms,
self-legislation does not encourage an ideal of self-sufficiency. On the con-
trary, it exposes humans as constitutively vulnerable to one another and
mutually dependent. But it also shows that practical rationality is a social
enterprise building upon the features of vulnerability and interdependency.
We become rational and autonomous thanks to others, from within practices
marked by mutual respect and recognition.
The recognition of others works as a normative constraint for the rational
agent. However, the dialogical interpretation of self-legislation points out
that the role of others in the practice of lawmaking is not merely constrictive
and restrictive. More fundamentally, mutual respect and recognition of
others structure the very practice of justification, and thus play an active
role at the very early stage of intention and judgment formation. Kant vividly
represents the activity of reason in terms of public scrutiny.47 This shows that
reason has no transcendent authority; it can be vindicated through a rigorous
self-discipline. When we reason, we engage in an activity whose structure is
similar to that of a dialogue among free and equal citizens (A738/B766).48
The dialogical model aims to emphasize that all exercises of reflection are
public insofar as they are governed by a constitutive norm of mutual respect
and recognition.
Reason is public also in the sense that it admits of a public use. The
appropriate setting for the public use of reason is that of an open comparison
46
See O’Neill, Constructions of Reason, 53–57, 66–77.
47
“Reason must subject itself to critique in all its undertakings. . . . Now there is nothing so important
because of its utility, nothing so holy, that it may be exempted from this searching review and inspection. . . .
The very existence of reason depends upon this freedom, which has no dictatorial authority, but whose
claim is never anything more than the agreement of free citizens, each of whom must be able to express his
reservations, indeed even his veto, without holding back” (A738–39/B766–67).
48
This claim should not be confused with another important sense in which reasoning is public, that is,
governed by confrontation and comparison, rather than by the appeal to evidence.
374 C. Bagnoli
and discussion. The capacity to submit our opinion to public scrutiny is the
very same capacity that we exercise when we deliberate about what to do or
judge ourselves in our conscience. Even in such private moments when we
examine ourselves, then, we appeal to public methods and aim to produce
reasons that are universally authoritative. This is to say that rational auton-
omy is always constitutively constrained by others, in a more basic sense in
which it happens for the public use of reason. In public contexts, it becomes
clear that authority is not the individualistic achievement of a solitary will:
the exercise of autonomy is always, to an important extent, a public matter,
as it involves undertaking the legislator’s office in relation to co-legislators. It
is useful to formulate publicity as a structural requirement, as opposed to a
public exercise of reason, and also to distinguish it from the normative
attitude of mutual respect, even though these elements are integrated in
the correct practice of reasoning.49
Since we need others to think correctly, we actually exercise autonomy in a
public fashion. Kant emphasizes that we do not think correctly when we are
deprived of the opportunity to test our reasoning against other people’s
judgment. Applied to spheres of agency, the term “public” does not refer
to social transaction and bargaining. A reason is public when it is not
parochial or used to further one’s own private interests. Second, on the
dialogical model, the defeaters of autonomy are sorts of inconsistencies that
we detect by taking possible others into account, that is, by entering a virtual
public forum. Understood dialogically, self-legislation works as a diagnostic
device for interpreting and signalling failures of autonomy as failures of
recognition. An agent who is not capable of self-legislation and cannot act
on principle is constitutively unresponsive and insensitive to the rational
demands of others.
While Kant’s conception of autonomous action abstracts from the par-
ticular determinations of the individualistic will, it is designed to govern
finite agents, that is, agents who are exposed to “certain subjective limita-
tions and hindrances” (G 4:397). Principled action is intelligible in a way
in which idiosyncratic performances are not. Intelligibility is not deter-
mined by particular social agreements and transactions, and thus it is not
relative to specific webs of practices, but it is based on universal criteria of
49
O’Neill takes publicity to include the public use of reason (Constructions of Reason, 28–50, 33–34,
70–71, 206). The implication of her view is that the authority of moral norms does not depend solely on
the formal features of the norms, but it is ultimately established by public communication, that is,
through practices governed by mutual respect and recognition.
mutual respect and recognition. These criteria warrant the intelligibility of

action while protecting from arbitrariness and guarding against bias. To
serve this purpose the principles of action must be public or universal in
scope.
Concluding remarks: The fecundity of Kantian

autonomy
My argument has been that the constructivist interpretation dissolves the
paradox surrounding the idea of autonomy as self-legislation, by explaining
how self-legislation demands legislating with all agents having equal stand-
ing. I would like to conclude this chapter by showing the fecundity of this
approach to autonomy.
An apparent advantage of the constructivist interpretation of Kantian
autonomy is that it does not build upon an extravagant metaphysics, and
thus it is easily adaptable to contemporary naturalistic accounts of the mind
and of the world. However, this is not a decisive merit in favor of Kantian
autonomy, since many other contemporary accounts of autonomy align with
naturalism by construing self-reflexivity in subjective terms, without com-
mitting themselves to claims about rational self-legislation. They may seem
preferable precisely because they are minimalistic.
In contrast to such views of autonomy, the Kantian view strongly ties
autonomous agency to rational autonomy. Autonomy ultimately amounts to
the capacity to act and think on principles rather than being determined by
the pressure of desires and other natural urgencies or the capacity to endorse
completely one’s actual mental states. This is the feature that sets Kantian
autonomy apart from other contemporary views, and the constructivist
interpretation succeeds in bringing about its fecundity.
On the constructivist interpretation, the appeal to normative principles
is constitutive of the exercise of rationality as a feature of embodied
rational agents. The theory of autonomy is tailored to fit the prospects
and problems of rationality as it pertains to the peculiar human condi-
tion, which is neither determined by animality nor defined by rationality
as such. Human agents face distinctive challenges due to the fact that
they are embodied, mutually dependent, and finite. The argument about
autonomy is supposed to show that humans respond to this condition of
vulnerability by exercising a distinctive sort of practical rationality, which
376 C. Bagnoli
is basically captured by reference to normative principles. Acting on

principles means acting on the basis of considerations that could be
shared by all other agents with equal standing. For this to happen,
reflective agents must be sensitive to the mere possibility that things be
otherwise, and this requires a moral capacity for respect of others as
equals. Differently than in the realist interpretation, the moral feeling of
respect is not a mere response to the value of others, but it is the
subjective aspect of their recognition as others.
This claim has two important implications. First, it indicates that practical
reflection is not a mere intellectual exercise in which, by a stretch of
imagination, we play musical chairs and shift to other possible worlds,
magically overcoming our horizon. Practical reflection importantly involves
an affective aspect of respect. Second, relevant others (that is, those who are
equal in status) play a constitutive role in such an exercise. On the dialogical
view, this means that others are not only sources of constraints or addressees
of our claims, but also count as partners in our deliberation. This is the most
relevant normative aspect of the thesis that self-legislation is ultimately an
activity of co-legislation.
The normative effects of autonomy understood in terms of adherence to
self-reflective rational principles resonate at the individual and at the collec-
tive levels. At the individual level, rational autonomy warrants deliberative
coherence and integrity over time. Legislating with others as having equal
standing commits one to relating to oneself in one’s distant future as having
equal standing. This claim forces us to reconsider the normative impact of
present commitments over time. The Kantian account of autonomous
agency shows that the roots of integrity over time and of shared agency are
the very same, and ultimately amount to the complex capacity to act on
principle. Rational autonomy explains how successful rational agents are in
solving problems. They can be effective in the world by constructing reasons
that could be felt as authoritative and compelling also by all other agents with
equal standing.
The pressure for coherence, both synchronic and diachronic, individual
and collective, is thus practical in nature and exercised by normative princi-
ples. This is where constructivism most sharply diverges from moral realism.
Normative principles are not valid because they track an independent reality
but because they express the self-conception of practical subjects. Insofar as
the conclusions of the activity of reason reflect our own self-conception, they
are distinctively efficacious and serve as a practical guide. Such a practical
guide is therefore not a passive form of conformity to any given procedure,
nor a mere expression of idiosyncratic preferences or a sentimental form of

endorsement. Acting on reasons that others can share is the paradigm of
autonomy, and the locus where both the aspirations of objectivity and
legitimate normative authority meet.50
50
Earlier versions of this paper have been discussed at the Department of Philosophy at the New School
of Social Research in New York, and at the Department of Philosophy at Notre Dame University,
cosponsored by the Notre Dame Institute for Advanced Study, in 2011. I would like to thank these
audiences, Karl Ameriks, Robert Audi, Jay Bernstein, Alice Crary, Mark Lebar, Elijah Millgram, Onora
O’Neill, Christine Korsgaard, Andrews Reath, and especially Stefano Bacin, for their helpful suggestions
on previous drafts.
Part VI
Aesthetics
17
Feeling the Life of the Mind:
Mere Judging, Feeling, and Judgment
Fiona Hughes
Introduction
In Section 9 of the Analytic of the Beautiful, Kant raises a problem that he
says is “the key to the critique of taste,” namely: “whether in the judgment of
taste the feeling of pleasure precedes the judging of the object or the latter
precedes the former” (CJ 5:216). His answer is that judging precedes feeling.
What is less clear is why this is the case, as well as how this conclusion fits
into the overall argument of the Critique of Aesthetic Judgment and thus serves
as its key. The task I take up in this chapter is to explain the relation between
judging and feeling that makes distinctively aesthetic judgments possible. In
order to establish the persuasiveness of Kant’s position it will be necessary to
critically assess as well as develop his account.
The reason the priority of judging is so crucial for the project of the
Critique of Aesthetic Judgment is that, were it the case that the feeling of
pleasure preceded judging, judgments of taste would express only the agree-
ableness of an object and would not qualify as subjectively universal. The
judgment of taste is distinctively aesthetic because “mere judging” – his more
precise title for “judging” and which Kant also refers to as the “free play” of
the faculties, as we will see – is prior to our feeling of pleasure in an object.
F. Hughes (*)
School of Philosophy and Art History, University of Essex, Wivenhoe Park,
Colchester, UK
e-mail: fhughes@essex.ac.uk

382 F. Hughes
Thus, the rationale for the first part of the third Critique – the introduction
of distinctively aesthetic judgments into the critical system – rests or falls on
finding the answer to the problem posed in Section 9.
The broader picture is not, however, to simply demote feeling with
respect to judging, but to establish the distinctive type of aesthetic feeling
on which aesthetic judgments are based. In the earlier sections of the
Second Moment of the Analytic of the Beautiful, Kant has argued that
aesthetic feeling is subjectively universal (CJ 5:211–16). In Section 9 he
establishes the distinguishing criterion for – and the condition for the
subjective universality of – aesthetic feeling, namely, that it has as its
ground mere judging. Establishing the precise nature of the relation
between feeling and judging is required if the “key” is to unlock the overall
argument of the Critique of Aesthetic Judgment.
The feeling operative in an aesthetic judgment is distinctive from other
“relations” to the feeling of pleasure and displeasure – such as the agreeable-
ness of gratification and esteem as moral liking (CJ 5:209–10). This reveals
that for Kant there are other kinds of feeling in addition to the aesthetic
feeling that arises in response to a beautiful object. Throughout this chapter,
however, when I refer to feeling I intend only the distinctively aesthetic
feeling that is dually directed both to the object and to the activity of judging.
I explain this dual-directedness of feeling at the end of the third section,
“Free Play as a Feeling of the Life of the Mind.”
Having mere judging as its ground qualifies feeling as the ground for
judgments of taste. There is a double determination of judgments of
taste, the direct ground of which is feeling, while their indirect ground is,
by extension, mere judging. I will establish the distinction between
judging and judgment entailed by this formula in the next section of
this chapter. Although Kant does not provide an account of double
determination in Section 9, in the third section I will show how the
beginnings of such an account are provided in the First Introduction.
I argue that in the Analytic of the Beautiful Kant introduces an account
of feeling that operates as a non-cognitive and yet reflective form of
awareness. The significance of this for Kant’s critical project cannot be
underestimated. The range of modes of awareness – which hitherto com-
prised sensible intuitions, concepts of understanding and conceptually
determining judgments, but also ideas and principles of reason – is
extended to include a new distinctively aesthetic type of judgments that
have feeling as their ground. This signals a radical development in Kant’s
project. Crucially, Kant views this development as the condition of the
integrity of his critical system (CJ 5:196). Although I cannot address the
17 Feeling the Life of the Mind 383
question of the coherence of Kant’s system here, the emergence of reflective

feeling, it would seem, is the key not only to the critique of taste but also to
the coherence of the critical project as a whole.
Establishing that feeling can be reflective has very wide-reaching ramifica-
tions, stretching well beyond Kant studies. I will briefly mention only two
examples. First, Kant’s introduction of feeling has important implications for
the relations between conceptually based cognition and pre-conceptual aware-
ness. Feeling is usually considered as affective and as, at best, the object of
reflection rather than as a variation on its form. Whether or not it is adequate
to treat emotion as passive – as Kant typically does – he distinguishes feeling
from emotion (CJ 5:226). Feeling is a form of pathos that consequently turns
out to operate not merely passively, but as a reflective form of consciousness. I
show how the feeling of pleasure characteristic of taste qualifies as reflective
insofar as it is a capacity both to differentiate and combine. I also argue that
the intermediary status of feeling – between cognitive spontaneity and recep-
tivity – allows it to operate as a reflection on the possibility of cognition
without counting as an actual cognition. Kant’s account of feeling as reflective
– if it works – puts in question the fork between, on the one hand, cognition
where concepts go “all the way down” and, on the other, a pre-reflective
experience that excludes conceptualization.1
Another way in which reflective feeling has broader philosophical bite is,
as I show, that it affords a self-reflection on mental activity – what Kant calls
“the feeling of life” (CJ 5:204). If this is right, aesthetic feeling offers an
insight into the activity of consciousness. In particular, feeling shows us how
the mind operates through a combination of both cooperative and distinctive
orientations and that these orientations are intentionally directed.
The distinction between judging and judgment:

A development of Kant’s “key”
In this section I argue for a distinction between judging and judgment, which
will allow for the solution of a series of problems, in particular, concerning
the roles played in judgments of taste by feeling and objects, respectively.
The key to the critique of taste – the priority of mere judging over feeling
within judgments of taste – implies a distinction between judging and
1
See, for instance, the debate between John McDowell and Hubert Dreyfus on this issue in Inquiry 50,
no. 4 (Aug. 2007): 338–77.
384 F. Hughes
judgment. Kant makes the same distinction – still unexplained – in the

Deduction of Taste (CJ 5:289). In my ensuing argument I will show how
employing this distinction allows us to understand that the “key” can,
indeed, explain how mere judging and feeling are related so as to make
possible distinctively aesthetic judgments. It is necessary, however, to adapt
and develop Kant’s statement of the “key.” In the first instance, I restate it as
a two-term relation between judging and judgment, before going on to
incorporate feeling in a three-term relation, at which point the continuity
with Kant’s original formulation will be more apparent.
First, the adaptation to the relation between judging and judgment:
judging precedes judgment. Next, the development of the account so as
to bring out the role played by the dependent term: judgment expresses
judging. Taken together: while mere judging is the grounding condition
for aesthetic judgments, in aesthetic judgments we exercise or express
mere judging. The aesthetic judgment “This rose is beautiful” is a
linguistic expression of my mere judging, that is, the free play of my
mental powers in response to a particular flower. Although the linguistic
status of judgments is underdeveloped throughout Kant’s opus, his
account of taste in Sections 9 and 38 implies – and, I will argue, his
overall argument requires – a distinction between having a mental state
of free play and expressing this in a judgment. While he sometimes omits
and never adequately accounts for this distinction, in the next section
I will show that he requires it if his account is to avoid falling into a
vicious circularity where mere judging grounds feeling and feeling
grounds mere judging.
The relation between judging and judgment is reciprocal but asymmetric –
there is no judgment without mere judging as its ground, while judging is
expressed in a judgment. It is important to distinguish between these two
directions of implication, because the distinctively aesthetic judgment qualifies
as such because it is ultimately referred to the activity of mere judging as the
meaning of that judgment. If the direction of grounding were reversed, the
distinctive character of aesthetic judgments as subjectively universal would be
critically compromised. Moreover, the distinction Kant requires can be
defended, for to say that a judgment expresses or exercises mere judging is
not equivalent to saying that judgments ground judging.
In the critical literature, there are a variety of ways of parsing the relation
between judging and judgment. Two important – and highly contrastive –
readings are proposed by Paul Guyer and Hannah Ginsborg. Guyer argues
that reflective judgment comprises two acts: the first is an aesthetic estimate
based on a harmony of the faculties, while the second is the aesthetic
judgment proper which establishes the universal validity permitting the

judgment that the object is beautiful.2 Ginsborg argues to the contrary that
the activity of free play is itself the judgment of taste, which already makes a
claim for universal validity. There is thus no need for “two conceptually
distinct acts of reflective judgment.”3 I agree with Guyer that it is necessary
to make a distinction between a judgment proper and the harmony of the
faculties, which is another cognate term for mere judging. I do not, however,
agree that there are two discrete acts of reflective judgment, but hold, rather,
that judging activity is expressed in an aesthetic judgment. The content
communicated by an aesthetic judgment is, in my view, nothing other
than mere judging – standing in some relation, yet to be established, to
feeling – insofar as it is prompted by something beautiful. Thus, while I insist
against Ginsborg that Kant’s account implies two levels, I agree with her –
against Guyer – that subjective universality arises at the level of mere judging.
My solution is that mere judging is the ground of judgments of taste insofar
as mere judging establishes the universality condition for judgments.
Before moving on to a further development of the “key” in line with the
distinction I have proposed, I should clarify the range of terms used for – as
well as the meaning of – “mere judging.” In Section 9, the judging element
within judgments of taste is referred to in a number of ways, which in this
chapter will be treated as equivalent in their denotation – even if they may
not be identical in their connotation. The judging that is prior in a judgment
of taste is also called “merely subjective (aesthetic) judging of the object [bloß
subjective (ästhetische) Beurteilung des Gegenstandes]” (CJ 5:218), while in the
First Introduction it is referred to as “mere reflection on a perception [bloße
Reflexion]” (FI 20:220). “Mere judging” is additionally referred to as a
“mental state [Gemütszustand] that is encountered in the relation of the
powers of presentation [Vorstellungskräfte]” (CJ 5:217, translation modi-
fied),4 as a “free play” of the cognitive or presentational powers (CJ 5:217),
and as a “harmony of the faculties of cognition” (CJ 5:218). Provisionally,
2
Paul Guyer, Kant and the Claims of Taste (Cambridge: Harvard University Press, 1979), 110–16, 151–60.
3
Hannah Ginsborg, “Reflective Judgment and Taste,” Noûs 24, no. 1 (March 1990): 71–72; and
Hannah Ginsborg, “On the Key to Kant’s Critique of Taste,” Pacific Philosophical Quarterly 72, no. 4
(Dec. 1991): 292.
4
Throughout I adjust the Cambridge translation to render Gemütszustand as “mental state” rather than
“state of mind,” and Vorstellung as “presentation” rather than “representation.” My preference for the
first is on account of the neutrality of its ontological implication, while the second avoids suggesting that
Kant is committed to a representational theory of mind. On the latter issue, see also Werner Pluhar’s
note in his translation of the Critique of Pure Reason (Indianapolis: Hackett, 1996), Bxvii note 73.
386 F. Hughes
the meaning of all these terms is the activity of the subjective capacities
necessary for what can variously be described as cognition, presentation, or
determining judgment, but only where their exercise in a reflective judgment
gives rise to no determination of an object or its constituent intuition (CJ
5:217). I return to the ultimate ground of mere judging in the subjective
conditions of cognition in my final section. As mere judging is what estab-
lishes the judgment of taste as reflective, I will sometimes refer to it simply as
“reflection.”
At this point – in order to address the problem of the role played by feeling
in judgments of taste – it is necessary to develop the “key” into a three-term
relation, comprising mere judging, feeling, and judgment. In short, feeling
operates as the intermediary term between judging and judgment. The
solution I will implement in the next section is the following: mere judging
grounds feeling that, in turn, grounds aesthetic judgments. Meanwhile, aes-
thetic judgments express feeling that, in turn, expresses mere judging. The two
levels of expression are distinctive in that feeling is consciousness of mere
judging, while a judgment is an expression or exercise of the feeling of mere
judging. Nonetheless, both expressive relations are dependent on and thus
contrasted to their corresponding grounds.
The final (principal) problem I address in this chapter is the contribution
of the object to aesthetic reflection. The formulation of the question in the
title of Section 9 makes clear that both judging and pleasure are directed to
an object. I am not concerned in this chapter with whether there are aspects
of objects that make them apt for aesthetic reflection.5 However, I cannot
exclude from the current discussion the way in which the presentation of an
object initiates reflection. Mere judging is “set into play” through a given
presentation (Die Erkenntniskräfte, die durch diese Vorstellung ins Spiel gesetzt
werden . . . ) (CJ 5:217). Moreover, Kant claims that the enlivening of the
mental powers arises through the “prompting” of the given presentation
(vermittelst des Anlasses der gegebenen Vorstellung) (CJ 5:219, translation
modified). Both claims strongly imply that the object contributes to aesthetic
reflection. At the end of the next section I argue that when I make the
judgment that something is beautiful I do so because my feeling about it has
mere judging as its ground.
5
For a discussion of this question, see Fiona Hughes, “Analytic of the Beautiful: Design and the Role of
the Object in Taste,” in The Kantian Mind, ed. Sorin Baiasu and Mark Timmons (London: Routledge,
forthcoming).
Free play as a feeling of the life of the mind

So far we have seen that – in some sense still to be identified – mere judging
precedes feeling. Nonetheless, feeling plays a crucial role as our consciousness of
the free play of the faculties (CJ 5:219). Moreover, Kant claims that the mental
state is a feeling: “Thus the mental state in this presentation must be that of a
feeling of the free play of the powers of presentation” (CJ 5:217, translation
modified). In what follows I will make sense of these seemingly incompatible
claims by showing how in free play the mind is implicitly self-conscious of its own
activity and that such a mental state cannot be separate from consciousness of it.
In this section I argue that the “feeling of life” mentioned by Kant in
Section 1 of the Critique of Aesthetic Judgment refers to the free play of the
faculties, which is the centerpiece of his discussion in Section 9 (CJ 5:204). This
helps explain why free play and feeling may not be distinct. However, Kant
claims that the feeling of life is the basis for judgments of taste, and this leads to
the apparent contradiction that both mental state and feeling qualify as deter-
mining. Building on the distinction I made between judgment and judging in
the previous section, I argue that this problem can be resolved with the help of
the First Introduction’s account of what I call a double determination of
judgments of taste both by mental state (or mere judging) and by feeling. I
next argue that the determination of feeling by the mental state operates
through reasons, not causes. Further building on the extended version of the
“key” outlined in the previous section, I argue that the same mental state
qualifies as mere judging when it is seen as ground and as feeling insofar as it
affords consciousness of – or expresses – that ground. Finally, I explain why a
self-reflective feeling of our mental state is only graspable through the pleasure
we take in the presentation of a beautiful object.
How are we to account for Kant’s claim in Section 9 that the mental
state “must be” a feeling of free play (CJ 5:217)? If feeling provides
consciousness of the mental state, then why would he claim that the
latter is a feeling? This is either incoherent or, as I will argue, puts in
question the view that mental state and feeling are distinct. I propose
that we can make sense of this apparently aberrant claim by linking the
argument of Section 9 to the idea of the “feeling of life [Lebensgefühl]”
introduced in Section 1 of the Analytic of the Beautiful (CJ 5:204).6
6
See Rudolf A. Makkreel, Imagination and Interpretation in Kant: The Hermeneutical Import of the
“Critique of Judgment” (Chicago: University of Chicago Press, 1990), ch. 5, for a discussion of the
importance of the idea of life for the third Critique. Makkreel insists that free play is not connected to
388 F. Hughes
As I read it, the feeling of life is a feeling of the liveliness of the mind,
which in Section 9 is referred to as “the enlivened relation between the
two faculties [die Belebung beider Vermögen]” and which is nothing other
than the free play of the faculties (CJ 5:219, translation modified). We
experience the life of the mind as a feeling of pleasure: “Here the
presentation is related entirely to the subject, indeed to its feeling of
life, under the name of the feeling of pleasure or displeasure” (CJ 5:204,
translation modified). I conclude that the free play of the faculties – as a
feeling of life – is a feeling of pleasure.
A problem arises, however, in what might be called the genealogy of
judgments of taste. In Section 1 Kant announces that judgments of taste are
distinctive in that they have feeling as their ground: the “feeling of life,
under the name of the feeling of pleasure or displeasure . . . grounds
[gründet] an entirely special faculty for discriminating and judging” (CJ
5:204). This claim appears to lead to two problems. First, it may sound as if
feeling is the ground of judging and not the reverse, as is argued in
Section 9. Even if this could be resolved, a further problem arises as to
how both feeling and mere judging can be determining, for this sounds like
a viciously circular claim. This second problem would be dissolved if it
could be established that the mental state and feeling are determining in
different respects.
The first problem could be resolved in the following way: when Kant says that
feeling is the basis of the capacity for judging (Beurteilungsvermögen), the latter
refers to the capacity for making judgments not for the mere judging exercised
in such judgments. Beurteilung is the term Kant uses for mere judging, but
he always does so with a qualifier such as “merely [bloß].” Moreover,
Beurteilung, like “judging,” can be understood as the activity of making judg-
ments, and Kant often uses the term as simply equivalent to Urteil, that is,
judgment. Consistently, Beurteilungsvermögen is used to explain “taste
[Geschmack],” which I understand as the power to make aesthetic judgments
(CJ 5:203n, 211, 240, 295). The proposed solution is, however, tested by Kant
immediately going on to say that the power in question “merely compares,” and
this, I will show in the next section, is a further cognate expression for mere
reflection. This transition in Section 1 thus tends to support Ginsborg’s claim
that there is no distinction between judgment and judging. However – apart
synthesis (94, 106). For a criticism of his position, see Fiona Hughes, Kant’s Aesthetic Epistemology: Form
and World (Edinburgh: Edinburgh University Press, 2007), 156–60.
from the fact that this seems to undersell the expressive aspect of judgments – the
distinction is necessary if Kant is not to fall into the vicious circularity of
claiming in Section 1 that feeling is the ground of mere judging and in
Section 9 that mere judging is the ground of feeling. The distinction of levels
he introduces in Sections 9 and 38 is required in Section 1 even though it is
omitted.
I will now examine a distinction that makes the suggested solutions to
both problems compelling. In the First Introduction Kant distinguishes two
types of precedence operative in respect of aesthetic judgments of reflection
and which correspond to the distinction of levels between judging and
judgment that, I am arguing, his account requires:
the harmonious play of the two faculties of cognition . . . produces [bewirkt]

through this mere form a sensation that is the determining ground
[Bestimmungsgrund] of a judgment which for that reason is called aesthetic
and as subjective purposiveness (without a concept) is combined with the
feeling of pleasure. (FI 20:224)
In this passage Kant claims both that mere judging produces – or brings
about – sensation (Empfindung) and that sensation determines a judgment
(ein Urteil). Despite the causal-sounding terminology of the first claim, I will
argue that both claims refer to grounds or reasons, not causes.
However, it is first necessary to establish the identity of the “sensation”
that determines judgments. We have already seen that feeling plays this role
in Section 1. Meanwhile, in Section 9, Kant, strictly speaking, says that we
can only become conscious of free play through sensation (Empfindung) (CJ:
219). It is not remotely likely Kant is suggesting that we become aware of our
mental state through the type of sensation that in the Critique of Pure Reason
is defined as “the effect of an object on the capacity for presentation” (A19–
20/B34, translation modified). Immediately prior to the passage just quoted
from the First Introduction, Kant provides the premise suppressed in
Section 9 with the following definition of feeling: there is “only one so-called
sensation that can never become a concept of an object, and this is the feeling
of pleasure and displeasure” (FI 20:224). This sensation uniquely qualifies as
a feeling in that it is not determinable under a concept, unlike the usual sort
of sensation. This is why, earlier, I reported Kant as claiming in Section 9
that feeling affords consciousness of mere judging. Feeling is a unique
sensation.
Seen in this light, the First Introduction’s account reveals the following
double determination: first, the harmony of the faculties (or mere
390 F. Hughes
judging) produces or brings about – or, as Kant puts it in Section 9,

“precedes” – the sensation (the feeling of pleasure), and, second, as Kant
also says in Section 1, feeling is the determining ground of judgments of
taste. The First Introduction’s formulation does not entail that feeling is
the ground of mere judging and only that the former determines judg-
ments of taste, or as he puts it in Section 1, our capacity for such
judgments.7 In Section 9 Kant adds a further layer to his account: “the
judgment of taste . . . determines the object, independently of concepts,
with regard to liking and the predicate of beauty” (CJ 5:219, translation
modified). Taking up the distinction between grounding and expressing I
introduced in the previous section, the feeling of pleasure and displeasure
is determining of the judgment, which expresses our feeling that the
object is beautiful. As a consequence of its determination by feeling, the
aesthetic judgment non-cognitively determines the object to be beautiful.
Thus, there is no vicious circularity, for while mere judging determines
feeling, feeling determines aesthetic judgments. This means that there is a
double determination of aesthetic judgments, which, in turn, determine
the object. The unique so-called sensation or feeling thus has the peculiar
status of being both undeterminable and the indirect ground for a non-
conceptual determination.
So far I have argued that the feeling of life is the pleasurable awareness of
mental activity in the particular guise of the enlivened activity of the
faculties. Consequently, mere judging and feeling are not distinct mental
events, though I will argue that they are distinguishable aspects of the same
mental event. In order to show that the account I have given so far is viable, I
now need to explain how it is consistent with the priority of mere judging
with respect to feeling.
In what sense, then, does free play “precede,” “determine,” “ground,”
“produce,” or “bring about” the sensation or feeling within the same mental
event? Is the precedence that of cause to effect? Guyer reads the relation
between the harmony of the faculties and the feeling of pleasure as causal.8
Henry Allison, meanwhile, insists that the feeling of pleasure is intentionally
7
This is in contrast to an aesthetic judgment of sense, where the intuition of the object determines the
judgment (FI 20:224).
8
Guyer, Kant and the Claims of Taste, 106–10; see also 205.
9
Henry E. Allison, Kant’s Theory of Taste: A Reading of the “Critique of Aesthetic Judgment” (Cambridge:
Cambridge University Press, 2001), 53–54; and Henry E. Allison, “Pleasure and Harmony in Kant’s
Theory of Taste: A Critique of the Causal Reading,” in Kants Ästhetik/Kant’s Aesthetics/L’esthétique de
Kant, ed. Herman Parret (Berlin: de Gruyter, 1998), 466–83. See also Richard E. Aquila, “A New Look
at Kant’s Aesthetic Judgments,” in Essays in Kant’s Aesthetics, ed. Ted Cohen and Paul Guyer (Chicago:
directed to free play.9 For Allison, at least in Kant’s Theory of Taste, the
intentional status of feeling is compatible with its being caused by mere
judging.10 In his earlier, more extended discussion of this issue, Allison insists
that the feeling of pleasure is intentionally and not causally linked to the
harmony of the faculties.11 My view is that the precedence of mere judging
cannot be causal in any sense. For Kant, a causal relation requires determina-
tion under a concept (A189/B234). As the unique so-called sensation cannot
be determined under a concept, it cannot be caused by the mental state of
free play or by anything else. Moreover, a causal relation would require the
distinctiveness of cause and effect, and if the mental state is a feeling, then
they are not distinct. The peculiarity of the intentional relation between mere
judging and our consciousness of it through feeling is that the mental act
operates as a form of self-reflection in which the constituent terms cannot be
distinct, although they are distinguishable. I propose that mere judging
rationally grounds rather than causes feeling despite Kant’s causal-sounding
term bewirken.12 Correlatively, feeling grounds rather than causes judgments.
Kant explicitly uses “determining ground [Bestimmungsgrund]” and “grounds
[gründet]” for the second level of determination by feeling (FI 20:224; CJ
5:204), and his account entails that, although the latter is distinguishable
from judgment, they are not distinct, as would be required for a causal
relation. Thus Kant asks how we become conscious in a judgment of taste of
mere judging (CJ 5:218).
The solution I am arguing for is that mere judging precedes feeling
insofar as the former is the reason for – not the cause of – the latter. The
free play of the faculties explains why a distinctively aesthetic feeling of
pleasure is subjectively universal. Free play secures subjective universality
because it is a particularly harmonious exhibition of the coordination of the
subjective faculties necessary for any cognition (CJ 5:218). Feeling, mean-
while, is the respect in which the play of the faculties expresses, that is,
qualifies as consciousness of – not just the fact that there is – mental
activity. It is thus not that first there is a harmonious relation between
the faculties and subsequently a pleasure, but rather that the distinctively
University of Chicago Press, 1982), 87–114; and Hannah Ginsborg, “Aesthetic Judging and the
Intentionality of Pleasure,” Inquiry 46, no. 2 (June 2003): 164–81.
10
Allison, Kant’s Theory of Taste, 53–54.
11
Allison, “Pleasure and Harmony in Kant’s Theory of Taste.”
12
When I say that mere judging is the reason for our feeling pleasure, I do not mean that feeling is the
result of a prior rational principle. Rather, when contingently we find something beautiful, then we feel
pleasure that has as its ground mere judging.
392 F. Hughes
aesthetic pleasure arises as one in the harmony of the faculties. The same
mental state can be regarded both as ground insofar as it exhibits the
relation between the subjective faculties of cognition necessary for any
judgment and also as consciousness of that mental state in a feeling of
pleasure.13 This dual status is characteristic of a mental state in which the
mind is conscious through feeling of its own mental activity: in aesthetic
judgments of reflection “the mind becomes conscious in the feeling of its
state” (CJ 5:204).
We now can understand how Kant can claim both that the mental state is
a feeling of free play and that the mental state is the ground for the feeling of
pleasure in that free play (CJ 5:218). Not only would we have no access to
the mental state of free play if we had no consciousness of it through feeling;
there is free play of the faculties only insofar as we are conscious through
feeling of that playful relation. The distinctively aesthetic feeling of pleasure
is the life of the mind in its free play. When we consider the mental state in
respect of its subjective universality we see it as mere judging, but when we
consider the same mental state as a form of consciousness we see it as feeling.
The same mental state qualifies as ground in the first case and as an
expression of that ground in the second.
The reciprocal implications of mere judging and feeling are not
viciously circular because I have distinguished the grounding term –
mere judging – from the expressing term – feeling. The difficulty, rather,
is how we get hold of – and hold onto – a reflection on our own reflective
activity. This is where our contemplation of a beautiful object plays a
necessary role in aesthetic reflection. As I mentioned in the previous
section, the presentation of a beautiful object sets the faculties in play
or prompts them into an enlivened and accordant (einhellig) relation (CJ
5:217, 219). When the object prompts or sets off free play, the former is
neither the cause nor the occasion of the latter. It cannot be the case that
the object causes mere judging, for free play could not be caused and
remain free. If the object, on the other hand, simply occasioned reflection,
this would result in a merely random correlation between object and
mental state, not a determination through feeling. As a result, anything
could and nothing would necessarily count as beautiful. This is not Kant’s
position, as I have argued elsewhere.14
13
Thus I disagree with Allison’s distinction between reflection (or judging) and feeling (Kant’s Theory of
Taste, 70). See Ginsborg, “Aesthetic Judging and the Intentionality of Pleasure,” who also criticizes
Allison on this issue.
14
See Hughes, Kant’s Aesthetic Epistemology, 284–90; and Hughes, “Analytic of the Beautiful.”
If the object neither causes nor merely occasions mere judging, then
what does it mean to say that it “prompts” free play? Although it is
contingent whether we find something beautiful, once we do so the
ground for our liking of the object is that it gives rise to a free play of
our faculties.15 This non-causal reading allows for an asymmetric but
reciprocal relation between mental state and object. While free play
supplies the motivation for our finding something beautiful, it is only
through the beautiful object that we get purchase on free play. It will be
apparent that I am offering the same structural account of pleasure
whether it refers to free play or in response to the object. This is because
the pleasure we take in the object is aesthetic only insofar as we take
pleasure in the object’s prompting a free play of the faculties.
Mere judging not only precedes the feeling of pleasure in the object,
but also is the basis (Grund) of a pleasure in the harmony of the cognitive
powers (CJ 5:218). My solution is that pleasure in the object is at the
same time pleasure in mere judging.16 This approach allows us to make
sense of a fine – and, I believe, significant – detail of Kant’s account,
namely, that when he first introduces the pleasure we take in the play of
the faculties, he refers to it as “this pleasure [dieser Lust],” which suggests
that it has just been discussed (CJ 5:218, emphasis added). Immediately
beforehand, the discussion is, however, of pleasure in the object. Through
the use of the demonstrative article, Kant shows that the two pleasures are
one and the same, in the one case viewed in its intentional directedness
toward the object, while in the other in its intentional directedness
toward our mental activity in response to that object. There is, thus, a
dual intentional directedness of one and the same pleasure, rather than
two pleasures. An advantage of my account – in addition to making sense
of the letter of Kant’s text – is that it avoids the difficulty that, were there
two pleasures, it would be necessary to explain how they are connected.
15
See Hughes, “Analytic of the Beautiful” on the contingency of judgments of taste.
16
Melissa Zinkin argues that Allison is committed to two pleasures, one in the object and one “directed
to the purposiveness of the mental state in reflection on an object” (“Kant and the Pleasure of ‘Mere
Reflection,’” Inquiry 55, no. 5 [Oct. 2012]: 439–40). She refers to Henry Allison, “Reply to the
Comments of Longuenesse and Ginsborg,” Inquiry 46, no. 2 (June 2003): 182–94, where Allison
himself criticizes Longuenesse for claiming that there are “two orders of pleasure involved in the
judgment of taste: a first-order pleasure in the apprehension of the object and a second-order pleasure
in the shareability or universal communicability of the first-order pleasure” (185–86). Allison refers to
Béatrice Longuenesse, “Kant’s Theory of Judgment, and Judgments of Taste: On Henry Allison’s Kant’s
Theory of Taste,” Inquiry 46, no. 2 (June 2003): 153.
394 F. Hughes
This difficulty falls away once they are recognized as two aspects of one
and the same pleasure.
Feeling as a power to reflect

In the previous section I argued that mere judging and feeling are two ways
of describing the same mental event. We have also seen that feeling is the
basis for aesthetic reflective judgments. The question arises: how can feeling
qualify as judging such that it provides the basis for judgments of taste? A
feeling, we might think, belongs to the affective side of experience, with the
consequence that it could only contribute to the content of judgment. In this
section I argue that feeling qualifies as mere judging insofar as it is reflective.
The relevant type of reflection only arises as consciousness of that reflection
and thus must be expressed in feeling, the condition of that consciousness.
I first establish that the reflection in question is a capacity for comparison and
then show how feeling is reflective insofar as it compares the presentational
capacities of imagination and understanding. These faculties are compared
both with regard to what they have in common and their difference. Both
faculties contribute to presentation, but they carry out different roles: imagi-
nation combines, while understanding unifies.
Kant defines reflection as the ability to compare and hold together: “To
reflect [Reflektieren] (to consider [Überlegen]) . . . is to compare and to hold
together [vergleichen und zusammen zu halten] given presentations either with
others or with one’s faculty of cognition, in relation to a concept thereby made
possible” (FI 20:211, translation modified). The first option covers the logical
reflection operative in the genesis of empirical concepts (JL 9:92; VL 24:905).
Meanwhile, the second captures the transcendental reflection of the Amphiboly
of the Concepts of Reflection in the first Critique (A260–92/B316–49). Both
uses of reflection are preparatory for determining judgment: logical reflection
generates empirical concepts under which the sensible manifold can be deter-
mined, while transcendental reflection establishes a distinction between the
faculties necessary prior to the determination of the presentation of an object
(A269/B325). Neither of these types of reflection, however, is sufficient for
judgment, whether reflective or determining. There is, however, another use of
reflection that, like transcendental reflection, compares a presentation with our
cognitive powers and yet qualifies as judgment. Judgment can be regarded
either as “a mere faculty for reflecting” or as an ability to determine (FI
20:211). I take this to mean that reflective judgment is the ability to make
judgments that express mere reflection. Mere reflection arises when the activity
of reflection is carried out for its own sake and is not directed to determining a
presentation of an object under a concept. As I argued in the previous section,
such reflection is inseparable from our consciousness of it. In reflective judg-
ments of taste, reflection and feeling converge, because the reflection that such
judgments have as their ground is not merely a condition of cognition but also
constitutes an alternative, non-cognitive form of awareness.
In aesthetic judgments – the ground of which is feeling – mere judging
“only compares the given presentation in the subject to the entire faculty of
presentation” (CJ 5:204, translation modified). Thus feeling is the basis for
judgments that express the reflective comparison of the presentation of an
object with our mental activity in determining an object as beautiful. As I
argued in the last section, the play of the faculties is the ground of the feeling
in the sense that mere judging establishes the subjective universal validity of
this feeling. If feeling is the consciousness of free play and the mental state of
free play is a feeling, then we can conclude that feeling is a conscious –
although implicit – reflection that compares the cognitive faculties of imagi-
nation and understanding in their free play.
Feeling as reflective is the capacity to non-cognitively discriminate – that is,
differentiate – so as to make possible judgments as expressions of mere judging
(CJ 5:204). This implies that mere judging – as well as the judgments expres-
sing it – is necessarily bound up with differentiation. Although Kant does not
explicitly develop the idea of discrimination, he does not simply abandon it, as I
will show. The link between comparison and differentiation is evident in Kant’s
Lectures on Logic. In the Jäsche Logic, comparison establishes difference: “I see,
e.g., a spruce, a willow, and a linden. By first comparing [vergleichen17] these
objects with one another I note that they are different [verschieden] from one
another in regard to the trunk, the branches, the leaves, etc.” Reflection
(Reflexion, Überlegung), meanwhile, establishes “that which they have in com-
mon [gemein] among themselves, trunk, branches, and leaves themselves” (JL
9:94–95). Reflection and comparison are distinguished in this discussion,
although Allison and Béatrice Longuenesse are certainly right that they must
be integrated in their use.18 In fact, Kant uses the two terms interchangeably.
The Blomberg Logic states that in logical abstraction we compare concepts in
17
Kant uses the Latinate term Comparation at JL 9:94.
18
Allison, Kant’s Theory of Taste, 22; and Béatrice Longuenesse, Kant and the Capacity to Judge:
Sensibility and Discursivity in the Transcendental Analytic of the “Critique of Pure Reason,” trans.
Charles T. Wolfe (Princeton: Princeton University Press, 1998), 116.
396 F. Hughes
light of what they have in common (BL 24:255).19 In the Blomberg Logic and
the Vienna Logic, comparison is directed both to what differentiates and to what
is held in common (BL 24:274; VL 24:833–34). Finally, in the Amphiboly of
the Concepts of Reflection, transcendental reflection differentiates the faculties
of sensibility and understanding in light of the concepts that fall under them.
This counts as a form of comparison: “all judgments, indeed all comparisons
[Vergleichungen], require a reflection [Überlegung20], i.e., a distinction
[Unterscheidung] of the cognitive power to which the given concepts belong”
(A261/B317). Comparison and reflection are clearly mutually dependent, and
either term can refer to difference or to what is held in common. In what
follows, I will show how the mere reflection characteristic of aesthetic judg-
ments and of which we are conscious through feeling is oriented both to
difference and commonality.
Through feeling we are able to reflect on the presentational powers
relating to one another. My proposal is that this reflection is a compar-
ison. In feeling a free play of the faculties, we are aware both of their
commonality and their difference. Their commonality arises only insofar
as we are also aware of the difference between the respective roles they
play. There would be no playful relation were both faculties occupying
exactly the same roles; there would be no harmony were both faculties
singing the same notes. One way of highlighting their contrasting roles is
through a consideration of the respective relations in which the two
faculties stand to freedom in free play. Mere reflection is directed to
the reciprocal relation between the faculties, but only imagination is
predisposed to freedom from determination under laws (CJ 5:40–41).21
While Kant refers not only to “the free lawfulness of the imagination”
but also to “the free lawfulness of the understanding,” we might say that
in an aesthetic judgment the understanding is forced to be free
(CJ 5:240–41).
The respective roles of the faculties in the relation of free play can be seen
more clearly with the help of the Transcendental Deduction from the first
edition of the Critique of Pure Reason. Imagination’s role within mere
judging is the “combining [Zusammensetzung] of the manifold in intuition”
without a determining concept (CJ 5:217, translation modified). In the first
Critique’s Synthesis of Reproduction in the Imagination, combination
19
Kant also links Comparatio to “the agreement [Übereinstimmung] and identity of things” in the Vienna
Logic (VL 24:907–8).
20
Kant also uses reflexio at A260/B316.
21
See also CJ 5:203 on the priority of imagination in judgments of taste.
(Verbindung) is the activity of the pure transcendental synthesis of imagina-

tion and is a necessary but insufficient condition of cognition (A101).
Meanwhile, the understanding’s role that is compared to imagination within
mere judging would be to provide “the unity (Einheit) of the concept”
(CJ 5:217). This corresponds to the first Critique’s Synthesis of Recognition
in the Concept, where the role of the understanding is to establish the unity
(Einheit) of an object under a concept, without which the manifold would not
constitute a whole (A103). Thus to unify an object is to determine it by
unifying its manifold under a concept. In the play of the faculties mere
judging compares the activity of combining without a concept with the
activity of unification under a concept. Aesthetic reflective comparison
shows up the difference as well as the commonality of these ways of holding
together the manifold in intuition. This comparison establishes the basis for
aesthetic judgment – a mere judging that combines but does not unify and of
which we are conscious as feeling.
Exhibiting the life of the mind

I have explained how in a judgment of taste feeling operates as a form of
reflection that compares the faculties as to their difference as well as their
commonality and thus is not aimed at unification. Still, it may seem that the
harmony of the faculties models imagination’s combining of the manifold on
the unifying function of understanding and, as a result, aesthetic judgments
are oriented toward cognition even though they do not achieve it.22 In this
section I will argue that the commonality of the faculties in free play is not
that they both aim at unification, but rather that they both aim at exhibition
(Darstellung): while the exhibition of a concept of understanding necessarily
entails unity, imagination gives rise to an alternative exhibition that does not.
Kant uses “presentation [or representation, Vorstellung]” to refer to a range of
types of awareness, including intuitions, concepts, and cognitions (see A140/
B179–80, A320/B376–77). Meanwhile, “exhibition [Darstellung]” – when
used as a technical term in the third Critique – refers strictly to the combination
of two presentations, namely, the combination of an intuition with a concept.23
Judgment is the faculty that carries out exhibition. Determining judgments
22
See Avner Baz, “Kant’s Principle of Purposiveness and the Missing Point of (Aesthetic) Judgments,”
Kantian Review 10 (Jan. 2005): 1–32; and my response in Fiona Hughes, “On Aesthetic Judgment and
Our Relation to Nature: Kant’s Concept of Purposiveness,” Inquiry 49, no. 6 (Dec. 2006): 547–72.
23
When Kant uses Darstellung nontechnically, it is equivalent to Vorstellung.
398 F. Hughes
exhibit a concept so as to give rise to cognition. In the First Introduction,

judgment exhibits (Darstellung [exhibitio]) a concept of the understanding by
finding for it a corresponding intuition (FI 20:220). Similarly, in the published
introduction Kant writes: “If the concept of an object is given, the work of
judgment, in its use as exhibiting [Darstellung] (exhibitio) a cognition, is in
placing beside a concept a corresponding intuition” (CJ 5:192, translation
modified; see also RP 20:325–26).24 While Darstellung is not used as a
technical term in the Schematism, we can conclude that when a concept is
exhibited so as to give rise to cognition, the concept is schematized with an
intuition (A138/B177).25
It is not the case, however, that all exhibitions are cognitive. In the
published introduction, Kant affirms that exhibition also arises in aesthetic
judgments: “we can regard natural beauty as the exhibition [Darstellung] of
the concept of formal (merely subjective) purposiveness” (CJ 5:193, transla-
tion modified). The more tentative statement in the First Introduction is that
in the harmony of the faculties imagination “agrees [übereinkommt] with the
exhibition [Darstellung] of a concept of the understanding” (FI 20:220–21,
translation modified). In his later discussion of hypotyposis, symbolic exhibi-
tion is a procedure carried out by the power of judgment (Urteilskraft) and is
analogous with schematizing hypotyposis in terms of “merely the form of the
reflection, not the content, which corresponds to the concept” (CJ 5:351).
My proposal is that an exhibition arises in our consciousness – the feeling –
of merely judging the relation between the faculties in their purposiveness for
one another. Judgments of taste express such an exhibition.
If both cognitive judgments and mere judging exhibit a concept, what is
the distinction between them? Aesthetic exhibition is of a concept, “which
concept this is being indeterminate” (FI 20:221, translation modified). In a
cognitive judgment it could not be indeterminate which concept was exhib-
ited, for only through the determination of an intuition under a specific
concept or concepts can knowledge arise. On the other hand, it is not clear
that an aesthetic judgment can exhibit any concept. I think the solution is
that aesthetic exhibition is not so much of a determining concept indetermi-
nately grasped, but of a principle which, broadly speaking, could be
described as an indeterminate concept. Read in this way, mere judging in
aesthetic judgment exhibits the principle that all judgments require a coop-
erative relation between the faculties of sensibility and understanding as the
24
See also A713/B741 on mathematical exhibition.
25
This conclusion is supported by Kant’s account of schematic hypotyposis (i.e., exhibition) (CJ 5:351).
ground of mere judging. Aesthetic exhibition thus does not exhibit a sche-
matized relation between a concept and an intuition, but rather exhibits the
relation between the faculties that makes the schematisation of intuitions and
concepts possible. The published introduction sums this up in saying that
what is exhibited is the “concept of formal (merely subjective) purposiveness”
(CJ 5:193; see also FI 20:224). Aesthetic exhibition is of the mutually
purposive coordination of the faculties that is, in principle, required for
finding a corresponding intuition for a concept, whether or not that combi-
nation is schematized.
Aesthetic exhibition takes place in “the form of the reflection,” because it
is mere reflection, or as I have argued, reflective feeling that exhibits the
relation between the faculties (CJ 5:351). Thus the activity of imagination in
its freedom within an aesthetic judgment is distinct from the exhibition of a
concept of understanding both with regard to what is exhibited – an
indeterminate concept or principle, rather than a determining concept –
and in how it exhibits – through feeling rather than through the determina-
tion of an intuition under a concept of understanding. While it is certainly
the case that the exhibition brought about by a determining judgment entails
unity, it is not the case that all exhibition aims at unification. An exhibition
operating through feeling and aiming to exhibit the indeterminate principle
of our mental activity is not apt for determinate unification or even for
aiming at – and failing to achieve – unification. In the next section I will
develop this argument further, showing how an exhibition carried out by
feeling resists cognitive resolution. What can be concluded at this stage is that
imagination harmonizes with the exhibition of a concept of understanding,
for they have in common the coordination of the subjective faculties neces-
sary for any judgment. Nonetheless, feeling carries out a distinctive form of
exhibition not directed to unification of an intuition under a concept.
I have argued that reflective feeling exhibits the subjective powers of cogni-
tion in their free play: both natural beauty and beautiful art affect us in such a
way that the mind “feels its own state.” In particular, I argued in the section
“Free Play as a Feeling of the Life of the Mind” that the beautiful object allows
us to hold onto conscious activity in its free play. Thus, although aesthetic
exhibition is a form of self-consciousness, it is one that arises through our
response to the presentation of a beautiful thing.26 We can draw a parallel with
the Refutation of Idealism, where Kant says that “the determination of my
26
Thus, although I agree with Ginsborg that judgments of taste are self-referential, I deny that they are
“purely self-referential.” See Hannah Ginsborg, “On the Key to Kant’s Critique of Taste,” Pacific
Philosophical Quarterly 72, no. 4 (Dec. 1991): 306.
400 F. Hughes
existence in time is possible only by means of the existence of actual things that
I perceive outside myself” (B275–76). According to Kant, beautiful things set in
play judging activity; that is, they prompt a response in judgments based on mere
judging. In doing so, an artwork, for instance, offers a focus through which I can
become aware of the activity of my mind. This is why Kant says that the aesthetic
judgment – strictly speaking, mere judging – is referred to the subject and that
“the subject feels itself” (CJ 5:204). But, although taste is subjective, it is not
introspective and is, rather, an awareness of how “it [the subject] is affected by the
presentation” (CJ 5:204, translation modified). Such self-awareness can only arise
indirectly – as in a sidelong glance – when, for instance, I am absorbed by an
artwork. My claim is not that artworks operate so as to make possible a peculiar
epistemological analysis of consciousness, but that they afford an existential
insight into conscious activity.27 Such insight will always operate at the limits
of our experience, for it is an experience of the subjective conditions of that
experience. For Kant, judging something to be beautiful involves a consciousness
of the playful activity of my mental powers. Thus to make an aesthetic judgment
is, for Kant, to be implicitly aware of the life of the mind.
An artwork prompts such implicit insight into the activity of conscious-
ness when, for instance, it offers patterns that are not determinable under a
unifying concept or within a conceptual scheme. When we respond to such
emergent order it is possible to glimpse the activity of the mind responding
in free play. Playfulness visible in the presentation of the artwork is apt for a
playful response on the part of the mind.28
I will now illustrate how an artwork can present an alternative order that
initiates consciousness of mere judging. I will discuss briefly how the lighting
effects of a particular painting display combination but not unity.
Rembrandt’s The Night Watch is a figurative portrayal of a group of influen-
tial Amsterdam citizens (see illustration below). Some of these men paid a
commission for their inclusion in the mass portrait and, no doubt, the
painting captures real historical relations in other ways. However, the paint-
ing offers its own vision of reality. A principal way in which it does so is
through its use of lighting effects. The Night Watch shows figures appearing
into the light or receding from view. The painting is very dark, so it is even
more striking that certain – although by no means all – figures in the
foreground are highlighted by their luminescence. This is most obviously
27
Kant clearly did not use this terminology, but this does not preclude that his insight can now be
identified in this way.
28
I have argued that this amounts to a “dual harmony.” See Fiona Hughes, Kant’s “Critique of Aesthetic
Judgment”: A Reader’s Guide (London: Continuum, 2010), 21, 149–53.
true of the lieutenant, the brightness of whose light-colored uniform is

intensified by the contrastive shadow of the captain’s arm falling upon it.
The girl to the left of the captain, the captain’s face, his decorative ruff, and
the drum to the right of the lieutenant are also illuminated, even though their
positions make it improbable that they share the same light source. Nor does
it seem plausible that the figures left only partially lit in the surrounding
darkness owe their darkness solely to their spatial positions. Rather than
unitary lighting – either natural or artificial – the painting deploys combina-
tions and configurations of light and dark – typical of chiaroscuro – which
construct a series of relations that emerge only on the canvas and do not offer
a reproduction of a determinate external reality (Fig. 17.1).
When we contemplate this painting, that is, when we allow ourselves to
become absorbed by it, we take up its combinations in our response to it. The
painting’s lighting effects encourage – in a non-causal sense, they require – a
response that is free or open. Yet it is not the case that there is no order in the
painting, just that the order is not determined by a unified external reality.
Fig. 17.1 Rembrandt van Rijn, Militia Company of District II under the Command
of Captain Frans Banninck Cocq, commonly known as The Night Watch, 1642. Oil
on canvas. Rijksmuseum, Amsterdam.
402 F. Hughes
Correlatively, the mind is not chaotic and, rather, makes sense without requir-
ing unified sense. If, on the contrary, we tried to unify what is strictly a
combination, we would experience neither the painting nor free mental activity.
Aesthetic judgments, such as those I have outlined about the lighting effects
in Rembrandt’s The Night Watch, express feelings – not cognitions – about the
painting. Such judgments make explicit implicit discriminations and connec-
tions that I make in looking at the painting. I explore the painting reflectively
without coming to a definitive conclusion. Such judgments are not directed to
determining a quality in the painting, nor to establishing something internal to
my mind. I give myself over to the painting – I am taken up with it – and, in
return, I receive an awareness of my own mental activity through my awareness
of the painting. Thus aesthetic judgment is not either a self-reflection or an
observation of something external to the mind: it is, rather, an expression of a
feeling of the life of the mind in response to a beautiful object.
Feeling as mediating reveals both the possibility

and the limitation of cognition
I have argued that feeling affords an exhibition of the subjective conditions of
cognition as a reflection on their difference as well as their commonality.
I have also shown how such an exhibition does not aim at the unity necessary
for cognition. I will now argue that the exhibition afforded by feeling offers
an insight into the possibility of cognition, while simultaneously revealing its
limitation. Only feeling can achieve this on account of its peculiar status as a
unique sensation that is undeterminable and yet is the ground for non-
conceptually determining judgments. I conclude that feeling is neither
wholly active nor merely receptive.
As we have seen, reflective feeling compares the subjective conditions of
cognition, yet precludes the determination of an intuition under a concept
required for cognition. I think it is helpful to consider this as the distinction
between possible and actual cognition. In the Critique of Pure Reason, Kant
discusses these in the first and second postulates, respectively. The possible is
“whatever agrees [übereinkommt] with the formal conditions of experience (in
accordance with intuition and concepts).” The actual is “that which is bound
up with [zusammenhängt]29 the material conditions of experience (of
29
Although literally this means “connected,” and that is how Guyer and Wood translate it in the
Cambridge edition, Kemp Smith’s “bound up with” better captures the strength of the connection
sensation)” (A218/B265–66, translation modified). How might this illuminate

what Kant calls “cognition in general”? There are three main references to this
notion in Section 9: (i) in mere judging the presentational powers “relate a given
presentation to cognition in general” (CJ 5:217, translation modified), (ii) “the
mental state in this presentation must be that of a feeling of the free play of the
powers of presentation in a given presentation for a cognition in general” (CJ
5:217, translation modified), and (iii) in free play the faculties “agree with each
other as is requisite for a cognition in general” (CJ 5:218). While the first of
these may only claim that free play and “cognition in general” are compatible,
the others make clear that free play entails “cognition in general.” As it is evident
that free play cannot be cognitive, “cognition in general” cannot be equivalent
to cognition. I propose that “cognition in general” is the subjective formal
condition of experience, with which any cognition must agree. The first
postulate states the formal conditions of experience are intuitions and concepts.
The subjective grounds of their possibility exhibited in aesthetic judgments are
the faculties of imagination (for the combination of the manifold in intuition
[CJ 5:240]) and understanding (for the unification of the manifold under
concepts). Thus, “cognition in general” is the relation between these subjective
formal conditions of experience with which any possible cognition must agree
and which are exhibited or expressed in free play, that is, reflective feeling.
“Cognition in general” – as the ground of mere judging – is, thus, the ultimate
ground and the vindication of the distinctive status of aesthetic judgments.
Meanwhile, “cognition in general” cannot be – just because it is free and not
determined under a concept – “bound up with” the determination of a material
sensation under a concept. “Cognition in general” is a necessary condition of
the possibility of cognition, but cannot give rise to any actual cognition.
Claiming that “cognition in general” exhibits the subjective cooperation of
the faculties necessary for the possibility of any cognition thus does not imply
that aesthetic judgments are subordinated to the general project of cognition.
Aesthetic judgments do not simply fail to become actual cognitions: they are
incompatible with actual cognition. The subjective conditions of cognition
operating in default of determination under a concept amount to an alter-
native form of awareness, which although not, in principle, opposed to
cognition is, in fact, incompatible with it. This is because aesthetic judg-
ments are based on feeling, the unique “so-called sensation” that cannot be
determined under a concept. Feeling is a form of reflective awareness that
(Immanuel Kant’s “Critique of Pure Reason,” trans. Norman Kemp Smith [London: Macmillan, 1933],
A218/B266).
404 F. Hughes
resists cognitive resolution. The unique so-called sensation thus exhibits the
mere possibility of cognition only because it is undeterminable, that is, only
because it cannot become an actual cognition. Distinctively aesthetic judg-
ments with feeling as their ground serve as a demonstration that not every
form of consciousness is cognitive, nor is oriented toward determination
under a concept. Feeling and cognition coexist within a plurality of orienta-
tions that are, in principle, compatible with and yet irreducible to one
another. It is thus not the case that aesthetic judgments belong to the general
project of determining objects under concepts.30
In resisting cognitive resolution, feeling operates as a distinctive form of
awareness that is distinct from cognition and emotion (CJ 5:226). Feeling
can only serve as a simultaneous reflection on the possibility and the limita-
tions of cognition insofar as this form of awareness has something in
common with, while being distinct from, cognition. Feeling is neither wholly
spontaneous like understanding, nor is it merely receptive like intuition.
Feeling is, nonetheless, allied to spontaneity insofar as it operates through
mere reflection and, at the same time, allied to sensibility in that through
feeling we are receptive to beautiful objects. Feeling allows the mind to feel
its own state – the relation between spontaneity and sensibility – only
because it is neither entirely active nor merely receptive.
For Allison, feeling is active. His initial account of feeling is that it both
appraises and discriminates. We appraise “the capacity of a representation to
occasion an enhancement or diminution of one’s cognitive faculties in their
cooperative activity.”31 Discrimination, on the other hand, operates “in the
reception or acceptance of representations conducive to free play and the
rejection of those that are not.”32 Feeling is thus not merely receptive – in
Allison’s terms, not merely discriminatory – which leads him to conclude
that feeling is an “active faculty.”33
While Allison’s characterization of feeling as more than merely receptive is
suggestive, I do not conclude that feeling is active rather than receptive.
Instead, feeling has the peculiar intermediary status of combining the activity
of mere reflection with receptivity to an object. If feeling were an active faculty,
as Allison claims, it would be incapable of carrying out the role of mediating
30
Aesthetic judgments thus do not appropriate the imagination into the project of understanding, which
Longuenesse refers to as Vermögen zu urteilen. See Longuenesse, Kant and the Capacity to Judge, 7, 208.
31
Allison, Kant’s Theory of Taste, 69.
32
Ibid. Allison’s account of discrimination is puzzling. Since judgments of taste are singular, a judgment
comparing beautiful objects is not aesthetic (CJ 5:215).
33
Allison, Kant’s Theory of Taste, 69.
between receptivity (imagination as the capacity to combine what is given) and

spontaneity (understanding as the capacity to unify under a concept). Instead,
it would ultimately act in the interests of spontaneity, that is, understanding’s
capacity to determine under a concept so as to give rise to cognition.34 If, on
the other hand, feeling were merely receptive, it would not qualify as a “unique
so-called sensation” and would, like all other sensations, be determinable under
a concept (FI 20:224). A merely receptive feeling would also be incapable of
serving as the ground of the non-cognitive determination of objects as beauti-
ful in aesthetic judgments (CJ 5:219). Only a form of consciousness that is
neither entirely active nor merely receptive is capable of relating our sponta-
neous and receptive capacities, that is, of comparing them in their free play
with one another. In doing so, feeling as the unique so-called sensation that is
neither a concept nor a mere sensation is an alternative form of reflection,
operating at the limits of sensibility and understanding.35
34
Makkreel remarks that aesthetic harmony qualifies as “pure spontaneity” (Imagination and
Interpretation in Kant, 92). Kant, admittedly, says – although does not develop the idea – that
imagination, insofar as it is productive, is spontaneous (CJ 5:240). However, it must be the case that
the productivity of imagination is distinct from the spontaneity of understanding, for the productivity of
imagination in aesthetic judgments arises when there is no determination by the understanding.
35
I would like to thank Fabian Freyenhagen for helpful comments.
18
On Common Sense, Communicability,
and Community
Eli Friedlander
Introductory considerations
The four moments of the Analytic of the Beautiful (CJ 5:203–40) form a
unified grammar of the judgment of taste. The first three moments, and
some of their guiding terms, “disinterested pleasure,” “the universal voice,”
and “purposiveness without purpose,” have received much scrutiny in read-
ings of Kant, whereas the Fourth Moment, with its peculiar discussion of
“common sense,” has remained one of the more obscure parts of Kant’s
account.
“Common sense [Gemeinsinn]” itself has different meanings in philosophy
prior to Kant, mostly independent of the consideration of aesthetic judg-
ment. This is particularly evident if one takes into account the identification
Kant makes of the term with the Latin term sensus communis. To point to at
least two such apparently different uses:
1) We mean by common sense the kind of untutored yet healthy judgment,

which is ordinary.1
1
The usual term in German is Menschenverstand. Kant speaks of Gemeinsinn and contrasts it with
Gemein Verstand. But in §40, in which he returns to the issue, he clearly identifies Gemeinsinn with a
non-vulgar idea of common sense.
E. Friedlander (*)
Department of Philosophy, Tel Aviv University, Tel Aviv, Israel
e-mail: frieli@post.tau.ac.il

408 E. Friedlander
2) The term sensus communis is used to characterize the unity of the data of
the different senses.2
We thus have two distinct directions in which we can trace the origin of the
term “common sense,” the one that leads to a characterization of a form of
sociality or of community that shares the same sensibility as it were at the
most basic or natural level, and the other that leads to a characterization of a
unity of the sensory that should not be accounted for by way of a conceptual
synthesis but rather that is itself of a sensible nature.3
On the face of it Kant’s use of the terms Gemeinsinn and sensus communis
to relate to a dimension of the judgment of taste is different from either of
these meanings. This is particularly the case if we emphasize that the sensible
dimension is primarily manifest in pleasure. So, the guiding question we
might raise here at the outset is whether Kant means to provide a wholly new
meaning for the term, or whether he extends the traditional uses of the
concept. That is, is “common” characterizing community in the sense of a
public sensibility, say one that is apparent in cultural agreement, or is
“common” to be understood as characterizing a dimension of unity in the
sensuous, or perceptible presence, that is inherent to our pleasure in the
beautiful? The possibility that Kant relates himself in fact to both meanings
would for sure be in line with the way in which the third Critique has a
tendency, apparent in numerous crucial junctures, to bring together the
dimension of human community and the dimension of the unity of nature.
Exemplary necessity: Idea and ground

The account of the common sense in the Fourth Moment of the Analytic of
the Beautiful falls under the heading of modality in the table of the cate-
gories. Kant characterizes such modality as one of exemplary necessity: “as a
2
Aristotle uses the Greek term koinē aísthēsis to denote a sense in which the data of the different senses are
related to one and the same object of perception. Koine is also the term used for ordinary or low – e.g., in
reference to common dialect (as opposed to a high language). It is significant to note in this context that,
in the tradition of interpretation of Aristotle, the sensus communis is often related or even identified with
the imagination (phantasia).
3
There are several other important points in the transmission of the term “common sense” in the
philosophical tradition up to Kant. Thus, for example, Descartes takes the common sense to effect the
transition from the bodily or sensory to the unity of the mental. For an account of the transformations of
the concept, see, for example, Hans-Georg Gadamer, Truth and Method, trans. Joel Weinsheimer and
Donald G. Marshall, rev. ed. (London: Bloomsbury, 2004), 17–27.
necessity that is thought in an aesthetic judgment, it can only be called

exemplary, i.e., a necessity of the assent of all to a judgment that is regarded
as an example of a universal rule that one cannot produce” (CJ 5:237).4 The
notion of the example appeared already in the account of judgment in the
Critique of Pure Reason. Lacking rules, judgment in general depends on
natural talent, “so-called mother wit, the lack of which cannot be made
good by any school,” as well as on its, albeit poor substitutes, examples which
are “the leading-strings of the power of judgment” (A133–34/B172–73). But
in the field of aesthetics, exemplification relates to the way a subject judges
rather than directly to properties of the object one would wish the example to
single out. In passing judgment I speak for others; I take myself to be
representative. The aesthetic judgment is pronounced with a universal
voice as the judge of taste “judges not merely for himself, but for everyone,
and speaks of beauty as if it were a property of things” (CJ 5:212).
Our judgments of beauty have an exemplary quality. We have to conceive
of a judgment of taste as an act of exemplification, and therefore as involving
in principle expression toward others, or what Kant calls already in the
Second Moment, the communicability of such judgment. For it does not
make sense to take a judgment to be exemplifying, if whatever “takes place”
in judging is not turned to others, or has no communicable dimension to it. I
must make my encounter with the object meaningful, even if only to myself,
in my judgment, thus make it potentially available to others, or communic-
able, if it is to be called “exemplary.”
But this stress on an aesthetic judgment being exemplary only raises the
following question: what is added by the Fourth Moment to the account of
the universal voice given in the Second Moment? Indeed, one might wonder
whether the logic of exemplification is not already laid out, when Kant
conceives of an aesthetic judgment as “speaking with a universal voice,” or
as speaking for all. Is it the case, then, that the idea of the universal voice and
the norm of common sense are one and the same thing? They may indeed be
part of the same grammar, but there must nevertheless be something essential
4
A question that lies beyond the scope of the present chapter is whether the rule that is to be exemplified
is to be taken to be the rule of that specific instance of beauty or something like the rule or principle of
beauty in general. Deciding on this question would make a lot of difference in the interpretation of
Kant’s account. In the first case, we think of a kind of gradual convergence upon the judgment of a
specific work of art. In the other case, one may for instance think of the idea as given in the systematic
interconnection of the instances of art. This latter model is taken up by the Romantics in their notion of
the Idea of Art. In this context, see Walter Benjamin, “The Concept of Criticism in German
Romanticism,” in Selected Writings, vol. 1: 1913–1926, ed. Marcus Bullock and Michael W. Jennings
(Cambridge: Belknap, 1996), 116–200.
410 E. Friedlander
that is added by the Fourth Moment to the idea of exemplary necessity. I

wish to suggest that considering the difference between the universal voice
and common sense, as well as their interrelatedness, clearly brings out a
duality or polarity that pertains to the character of exemplification. Call it the
duality of ground and idea.
In the Second Moment, Kant conceives of the aesthetic judgment as
speaking for a universal voice that is only an idea. It is what one postulates
as the regulative focus which one’s judgment strives to exemplify: “Now here
it can be seen that in the judgment of taste nothing is postulated except such
a universal voice [Hier ist nun zu sehen, daß in dem Urteile des Geschmacks
nichts postuliert wird, als eine solche allgemeine Stimme]” (CJ 5:216). The
language of postulation is also similarly used in relation to universal commu-
nicability, as Kant writes in §9: “The animation of both faculties . . . through
the stimulus [Anlasses] of the given representation, in unison, namely that
which belongs to a cognition in general, is the sensation whose universal
communicability is postulated by the judgment of taste” (CJ 5:219). But
when he speaks of the common sense in §21, Kant uses the language of
presupposition (Voraussetzung) rather than postulation: “the universal com-
municability of a feeling presupposes a common sense” (CJ 5:239).
Presupposition is not something like prejudice, not something that
comes before judgment, but rather is to be seen as internal to the
complete form of reflective judgment as self-guiding. In aesthetic judg-
ment, feeling arises in making one’s experience communicable. One
could think of it also as a feeling that comes with making explicit
what is initially darkly sensed. The two poles of that movement are the
internally related common sense and universal voice. They are respec-
tively the presupposition of a natural ground of what makes itself felt
and the postulation of the possibility of its ideal or complete expression.
Kant, indeed, uses the term ground (Grund) to refer to the presupposi-
tion of common sense in the judgment of taste. To refer oneself to a
ground here does not mean to have a justification, reason, or argument.
(Kant often uses the term “ground” to refer to what cannot be encom-
passed in a natural law of a causal form, or a practical law. Thus,
phenomenal nature rests on a noumenal ground. The noumenal is a
ground, but not a reason or a cause of the phenomenal.) Initially, then,
we may say that what is postulated in the idea belongs to the order of
subjective acts, that is, to the order of the articulation of a possibility for
which it is claimed that it exemplifies what is universal, that is, what
holds for all subjects. But such universality is referred to a ground that is
sensed in our experience of the beautiful object judged. That ground is
not yet objectively characterizable (for that which is beautiful is not

judged determinately), but it nevertheless belongs to what we relate to
experientially. It is through what I, as it were, perceive with my own eyes
that I relate to a ground, which is given articulation by the subjective act
of exemplification. Universal communicability and common sense are to
be seen as belonging to one another insofar as the postulation of the
former rests on the presupposition of the latter to which it purports to
give expression. The intending to give expression to an idea requires the
presupposition of a real ground which is to be expressed. One may also
say that the aesthetic judgment has an expressive structure, meaning that
its subject is to make explicit or communicable that which is not yet
commonly available, and therefore only presupposed as an objective
ground.5
Indeed, in §19 Kant points to the way that there is something like a
structure of conditional necessity to universal communicability: “The
subjective necessity that we ascribe to the judgment of taste is condi-
tioned” (CJ 5:237). The subjective necessity that Kant refers to is that of
universal communicability, namely the postulation of the universal voice.
It demands the presupposition of the ground in nature for the ideal
agreement of subjects. That condition is further identified in §20 with
the idea of a common sense. It is only that which allows Kant to argue
in §22 that the subjective necessity is being represented as an objective
necessity under the presupposition of common sense.
Insofar as we think of a striving toward an idea, we must assume a ground
to be given. This does not mean that such a ground is articulated or present
to us clearly, but it is attributed to what is already there in the perceived,
namely to what we have access as nature. The communicable expression one
gives to such ground in aesthetic judgment is that subjective or ideal active
relation of our faculties involved in that which is given to be articulated. This
ground, in its internal relation to what one strives to express, is certainly only
sensed in feeling. But it would not be enough to say that one rests on the
feeling of pleasure to claim the judgment as exemplary. For the question is
precisely: What does this feeling point to? What is it that is only presupposed
as given, yet to be exemplified?
5
Compare the necessity that pertains to the moral law to what we think of as exemplary necessity that is
conditional. When we think of the “ought” of the moral law, we have as a ground a law in relation to
which one can show why everyone ought to act in such and such a way. But in taking my judgment as an
example, when I cannot appeal to a principle or law, I take my feeling to testify that I speak for a
universal agreement of subjects.
412 E. Friedlander
Feeling pertains to the natural side of our being human. In exempli-

fication I rest on my sense that it would be natural to respond in the way
I do to the object judged. This does not mean that I can point to
something specific in my experience that justifies this sense of the
naturalness in my response, but I feel that my response to beauty is
attuned to something that is assumed to be independent of my subjective
intention or interest, as if I see beauty with my own eyes, or as if it has
definite perceptible qualities. In the judgment of taste I seek to articulate
that which is given to me perceptibly, but is initially left implicit.6
In other words, we have a conception of something belonging to the
order of what is accessible to us as nature, which is yet to be actualized
in a communicable form, but which one can nevertheless sense in feeling,
when taking one’s judgment to exemplify, through its subjective neces-
sity, the universal voice. Put differently, the idea of a universal voice is
formulated as the ideal agreement between subjects, but this idea of
agreement must be based on a presupposition that it is some norm in
nature that is thus brought to expression by the attunement of the
judgments of the different subjects. The question then is how to refer
to a sense of what is implicit or potential in nature, and more crucially in
what way that is something that can be felt.
The systematicity of cognition and the sense of

harmony
One would be tempted to conceive of the indeterminate rule that is exem-
plified in one’s judgment of taste as the rule underlying the specific beautiful
thing one judges: I judge as if we could all agree on a rule that justifies my
response to this specific beautiful thing. But this way of putting things, while
certainly correct at one level, does not sufficiently bring out what is the
character of the indeterminacy at issue. Indeed, why can we not just at some
point hit on the rule that would be universally agreed upon? Why is that
6
This is importantly the structure that Kant articulates in explaining the possibility of the work of art as
a product of genius. Genius is possible only as natural talent – that is, as a condition in which nature in
an individual gives the rule, without the individual being capable of articulating this rule as one that
consciously guides their production. In other words, in genius we must assume a potential of intellig-
ibility of nature, as a ground of production. My discussion of common sense would suggest that this
notion of ground is to be recognized also as being at work in the understanding of the structure of taste
more generally.
impossible in principle? One could conceive of the indeterminacy of the

universal voice in two ways: either in terms of the mere lack of a concept for
that specific object, or in terms of the absence of a systematic place for the
judgment upon that object. If the aesthetic judgment is taken to be funda-
mentally non-conceptual, then one would have to understand indeterminacy
in the first way. But it would be hard to see how the judgment of taste can
have at all a communicable dimension to exemplify. But if what is lacking is
something of the order of systematicity, one could both encompass the
understanding that no determinate concept would do, by itself, and recog-
nize that there is a space of expression that is opened by acts of
exemplification.
Consider that, both in the case of cognition and in that of morality,
systematicity is characteristic of the role of ideas of reason. Our particular
knowledge is assumed to have a place in a systematic order of nature, and
similarly a dutiful act has as its ultimate object the kingdom of ends,
which is a systematic ordering of ends of communal existence. That
systematic order is of course not present, but rather yet to be fully
recognized or produced. The indeterminate norm that is at issue in the
judgment of taste should similarly refer us to an idea of systematicity
specific to the beautiful. What could be the relation of exemplification
and common sense to the idea of systematicity, and is there a possibility
of assessing systematicity in feeling?
This presence of the consideration of systematicity in Kant’s account
of the common sense can be initially recognized in the way in which he
suggests a connection between the aesthetic and the domain of cognition.
Indeed, at the beginning of §21, he claims that the issue of communic-
ability of the judgment of taste bears on our response to a skeptical
threat to our claims to empirical knowledge: “Cognitions and judgments
must, together with the conviction that accompanies them, be able to be
universally communicated, for otherwise they would have no correspon-
dence [Übereinstimmung] with the object: they would all be a merely
subjective play of the powers of representation, just as skepticism insists”
(CJ 5:238). At the end of the same section it is common sense that is
adduced as the necessary presupposition that constitutes a response to the
threat of skepticism:
since the universal communicability of a feeling presupposes a common

sense, the latter must be able to be assumed with good reason, and indeed
without appeal to psychological observations, but rather as the necessary
condition of the universal communicability of our cognition, which is
414 E. Friedlander
assumed in every logic and every principle of cognitions that is not skep-
tical. (CJ 5:239)
The skeptical threat that Kant has in mind here, the countering of which would
make necessary the presupposition of common sense, is probably to be read back
to the problem Kant raises in the introduction to the Critique of the Power of
Judgment, concerning the systematic connection of our knowledge of nature at
different levels:
For it may certainly be thought that, in spite of all the uniformity of things in
nature in accordance with the universal laws, without which the form of an
experiential cognition in general would not obtain at all, the specific diversity of
the empirical laws of nature together with their effects could nevertheless be so
great that it would be impossible for our understanding to discover in them an
order that we can grasp, to divide its products into genera and species in order to
use the principles for the explanation and the understanding of one for the
explanation and comprehension of the other as well, and to make an intercon-
nected experience out of material that is for us so confused (strictly speaking, only
infinitely manifold and not fitted for our power of comprehension). (CJ 5:185)
On the face of it the discussion of common sense and the problem of the
systematicity of cognitions are wholly different issues. But I want to
suggest that it would be fruitful to take into account the more general
problem of systematicity in further elaborating the question of the
relation of the universal voice to the presupposition of common sense.
In other words, I wish to argue that what is to be made communicable
or expressed in the aesthetic judgment is the feeling of a space of
concepts that belong together and are as it were concentrated in and
through what we experience as beautiful.
The communicability of a cognition is not to be viewed merely as the
capacity to assert a judgment applying a concept to an object. The
applicability of a concept depends on a broad and on the whole implicit
agreement that bears on the possibility of using the concept. To recog-
nize this presupposition of agreement is tantamount to appreciate how
much our concepts hang together or depend on one another. In the case
of empirical cognition, the range of interconnected concepts that is
involved in a specific cognition is not to be conceived as something
given to us in judging in an explicit form. But we can assess that we have
the proper subjective disposition of judgment appropriate to what is left
implicit, in terms of a fit in the use of our faculties, or in terms of
appropriateness in the subjective disposition of our faculties.7 Kant refers

to this sense of appropriateness by way of a figure of proportion between
imagination and understanding:
But if cognitions are to be able to be communicated, then the mental state, i.e.,
the disposition of the cognitive powers for a cognition in general, and indeed
that proportion which is suitable for making cognition out of a representation
(whereby an object is given to us) must be also be capable of being universally
communicated; for without this, as the subjective condition of cognizing, the
cognition, as an effect, could not arise. (CJ 5:238)
The elaboration of this relation of the cognitive powers that makes the
cognition communicable is initially understood as a proportion that differs
according to the different objects judged. This can be roughly understood as
how much imagination is required to a given concept, or put differently, how
much of a systematic order of internally related concepts (yet not belonging
to the explicit definition of the concept we seek to apply in cognition) is to be
appealed to in giving a complete justification for that application. For the
concept “dog,” for example, we would not need much imagination, but for
the concept “love” we may need more than we have readily in mind in our
unreflective existence. When someone asks us, “How do you know that this
dog is not a disguised cat?” we can dismiss this skeptical query as irrelevant.
But when one is asked, “How do you know that the other’s love is not merely
feigned?” (indeed, how do I know that my love is true?), this may require
relating to a broad expanse of meaning to deal with doubt.
But, beyond this idea of different proportions of imagination and under-
standing adequate for this matter or another, one can now further raise the
question of the ground of our sense of systematicity as such, which is
presupposed in every case in which we think the communicability of cogni-
tions. On my view, it is this that Kant is appealing to with the presupposition
of common sense. In order to account for this dimension, Kant shifts from
the language of proportionality to that of harmony and from the
7
In the Anthropology, in his discussion of common sense cases in which one appeals to mother-wit rather
than the wit of schools in judgment, Kant suggests that they are cases in which empirical judgments
depend on “determining grounds of masses of judgment that lie in the obscurity of the mind. One could
call this logical tact, where reflection on the object is presented from many different sides and comes out
with a correct result, without being conscious of the acts that are going on inside the mind during this
process” (An 7:140). The term “tact” is striking. It relates to the discussion of the social dimension of the
sensus communis to which I turn later in this chapter.
416 E. Friedlander
communicability as the subjective disposition of the faculties to its ground,

namely common sense.
In attempting to bring out this progression in Kant’s text, we should note
his use of the term Übereinstimmung, translated as “correspondence,” to refer
to the fit of our (empirical) cognitions with the object. It relates back to the
use of Stimmung, or attunement, throughout the discussion of common
sense in §20. The question of the relation to the object can perhaps be better
translated as “consonance” than “correspondence.” The notion of conso-
nance is fitting here in part because of Kant’s appeal to harmony in this
moment. Indeed, I take it that harmony is not to be identified with a specific
proportion between imagination and understanding.8 This is clear in the
following passage in which Kant moves from the account in terms of specific
proportionalities to the recourse to harmony:
But this disposition [or attunement, Stimmung] of the cognitive powers has a
different proportion depending on the difference of the objects that are given.
Nevertheless, there must be one in which this inner relationship is optimal for
the animation of both powers of the mind (the one through the other) with
respect to cognition (of given objects) in general; and this disposition cannot be
determined except through the feeling (not by concepts). (CJ 5:238–39)
Note that Kant speaks of harmony as a dynamic state, for it is a movement of

the faculties that gives the feeling of attunement. It is an animation, an
enlivening of one by way of the other, that constitutes the harmonious state.
This temporal dimension is what allows us to figure the unity of a movement
8
The characteristics of objective systematicity in nature are homogeneity, variability, and affinity. The
last one has a counterpart in the characterization of the activity of the imagination in the Anthropology:
By affinity I understand the union of the manifold in virtue of its derivation from one ground. . . .
The word affinity (affinitas) here recalls a process found in chemistry: intellectual combination is
analogous to an interaction of two specifically different physical substances intimately acting upon
each other and striving for unity, where this union brings about a third entity that has properties
which can only be produced by the union of two heterogeneous elements. Despite their dissim-
ilarity, understanding and sensibility by themselves form a close union for bringing about our
cognition, as if one had its origin in the other, or both originated from a common origin; but this
cannot be, or at least we cannot conceive how dissimilar things could sprout forth from one and the
same root. (An 7:176–77)
It is particularly interesting that Kant thinks of affinity as that mode of activity of the imagination in
which one has a sense of the common ground of imagination and understanding. The term affinity in its
connection to chemistry has a significant place in aesthetics after Kant through Goethe’s novel Elective
Affinities (1809) and Walter Benjamin’s famous essay, “Goethe’s Elective Affinities” (in Selected Writings,
297–360).
that is not identified through any specific proportion of imagination and

understanding in an act of cognition.
In other words, proportion is a quantitative figure. The theory of harmony in
music appeals indeed to the concept of proportionality. It is originally used to
refer to the system of proportional divisions of a vibrating string that produces
the order of tones constituting an octave. But the question is whether an
“optimal inner relationship” is something that can be accounted for quantita-
tively as a determinate proportion. Put differently, is the quantitative system of
proportions something that ultimately would not need to be referred to as
sensed or felt harmony? If so, then when Kant moves to the disposition that is
assessed in feeling alone, we also go beyond the notion of proportionality that is
quantitative in character. The attunement would precisely express the possibi-
lity of a subjective fit or harmony that has no specifiable determinate proportion
but is thereby the manifestation of total harmony in feeling.
The assessment in feeling points to a quality of the meaningful space
opened by the beautiful thing hanging together as a whole harmoniously. It
would express what we might call an immeasurable use of imagination. This
immeasurable is not that which we identify with excess, but rather with
something like precision, with assuming a heart of the matter where all hangs
together precisely; not merely right but just right, and that precision which is
not a matter of calculation is what can only be sensed or felt.9
Insofar as systematicity in our cognitions of nature is not a given, it is some-
thing to be presupposed as a principle of judgment. But acting on that presup-
position must find an echo or a justification in the achievement of the subjective
state that would correspond to objective systematicity, namely the harmony of
our faculties. The harmony of the faculties is judged qualitatively rather than
quantitatively. It is the intensification of feeling that constitutes the criterion of
harmony, rather than the model of proportion that relies on number.
Common sense: Natural or artificial?

The duality in exemplification, the relation of the idea of universal commun-
icability to the ground of common sense, is explicitly raised by Kant in §22.
The title of the section is “The necessity of the universal assent that is thought
9
This can be compared to another term that is important to Kant, namely “orientation.” In his essay
“What Does It Mean to Orient Oneself in Thinking?” Kant identifies orientation as a fundamental
capacity of reason (OT 8:133–46). Orientation presupposes a relation to the totality of the needs of
reason in feeling.
418 E. Friedlander
in a judgment of taste is a subjective necessity, which is represented as objective

under the presupposition of a common sense” (CJ 5:239). The universal voice,
as I have interpreted it, is a subjective focal point. It is something we strive to
realize in an ideal community of taste. But we, as it were, experience it already
when we judge something to be beautiful, feeling that higher meaning to be
already present – that is, represented as objective, as belonging to that which
we perceive to be beautiful – through the presupposition of the common sense.
It is, in other words, referred to something in our present experience of what is
beautiful. I say that there is something implicit in my intuition of what is
perceptually present, that justifies my drawing from it a communicable mean-
ing that pertains to an idea that is to be realized. Here Kant returns to explain
the notion of exemplification and expressly relates it to the act of judgment:
Thus the common sense, of whose judgment I here offer my judgment of taste
as an example and on account of which I ascribe exemplary validity to it, is a
merely ideal norm, under the presupposition of which one could rightfully
make a judgment that agrees with it and the satisfaction in an object that is
expressed in it into a rule for everyone. (CJ 5:239)
In other words, the presupposition of common sense makes what we have a

sense of – that which is accessible to us as nature, or which is present to us in
a sensuous form – to be that which receives a rational form in exemplifica-
tion. Conversely, what is exemplified is represented as pertaining to the
sensuous manifold by being referred back to that natural ground presup-
posed. It is as though we are justified in speaking for all in our judgment,
because we share something “naturally.” One solicits assent from everyone
else because one has a ground for it that is common to all, Kant writes (CJ
5:237). This is not merely having the same senses or the same concepts, but
belonging to the same nature, which is yet to be fully articulated. We are
“suitors” for the agreement of others. This again feels related yet different
from the emphasis of the Second Moment in which Kant speaks of demand-
ing agreement. Whereas the demand recalls the “ought” of morality (yet,
without having any law to ground it), the idea of being a suitor is more a way
of speaking to one’s inclinations (yet, without appealing to any interest or
purpose). Common sense then would be pointing to this naturalness silently
at work in the heart of aesthetic judgment.
We can bring out the importance of thinking of this natural dimension of
the ground of common sense by asking ourselves how the postulation of the
universal voice is to avoid the realization of an agreement that is conventional
or artificial. Artificiality is a serious threat to the idea of exemplification, for it
is precisely the lack of a hold on objective criteria that raises the question
whether what I find communicable in my experience of beauty is not really
something I invented.10 The presupposition of common sense thus rests on
the assumption that my response to the object is natural, and it is for this
reason that I take it to be universally communicable.
But what would it mean to say that we assume something belonging to
nature to be at the ground of our judgments of taste? Is the ground “really
there,” or do we just project back onto nature something whose true source is
a regulative idea of reason? In other words, should we think of the presup-
position realistically or idealistically? Kant appears to leave the question open:
This indeterminate norm of a common sense is really presupposed by us: our

presumption in making judgments of taste proves that. Whether there is in fact
such a common sense, as a constitutive principle of the possibility of exper-
ience, or whether a yet higher principle of reason only makes it into a regulative
principle for us first to produce a common sense in ourselves for higher ends,
thus whether taste is an original and natural faculty, or only the idea of one that
is yet to be acquired and is artificial, so that a judgment of taste, with its
expectation of universal assent, is in fact only a demand of reason to produce
such a unanimity in the manner of sensing . . . this we would not and cannot
yet investigate here. (CJ 5:239–40)
How are we to decide whether common sense is a “constitutive principle of

the possibility of experience,” “an original and natural faculty,” or an
artificial faculty that is to be acquired through a yet higher principle of
reason? One way to answer this difficult question is to emphasize how, all
through the Critique of the Power of Judgment, what Kant seeks is a stand-
point in judgment through which the regulative ideas of reason receive a
constitutive role in the character of experience. Indeed, this would follow
from the way in which a problematic that appears in the Critique of Pure
Reason in relation to the task of reason, namely bringing about systematic
10
This should be related to the Second Moment:
Whether someone who believes himself to be making a judgment of taste is in fact judging in
accordance with this idea can be uncertain; but that he relates it to that idea, thus that it is
supposed to be a judgment of taste, he announces through the expression of beauty. Of that he can
be certain for himself through the mere consciousness of separation of everything that belongs to
the agreeable and the good from the satisfaction that remains to him; and this is all for which he
promises himself the assent of everyone: a claim which he would also be justified in making under
these conditions, if only he were not often to offend against them and thereby make an erroneous
judgment of taste. (CJ 5:216)
420 E. Friedlander
order into our cognitions, is relegated to the power of judgment in its

relation to particular experience, namely our experience of organized natural
beings, or organisms.
Even if we ultimately think of common sense as produced by the needs or
demands of reason, it is still to be produced as something that would appear
to us natural, that is, something that would be referred to our natural
existence. In other words, it would not be produced as an order of law to
which our human nature is submitted (as would be the case with the ideal of
the realm of ends in morality). Yet, neither is the common sense to be
identified with our natural inclinations, for these can only produce an
interested judgment. Making explicit the ground of common sense as some-
thing natural in us, yet fully human, would be a fundamental dimension of
what we call culture. Common sense is internal to the development of
human culture. The natural ground of culture is what we might call a higher
idea of the ordinary. In order to bring out this side of Kant’s discussion of
common sense, we need to consider how he returns to this notion in §40 of
the Critique of the Power of Judgment.
Community, the public sense, and the ordinary

When Kant treats the capacity to judge in its empirical actualization in the
Critique of Pure Reason, he emphasizes how good judgment is something that
is almost to be viewed as a natural talent. This kind of natural ground is
further characterized by Kant as “mother-wit [Mutterwitz],”
the lack of which cannot be made good by any school; for, although such a
school can provide a limited understanding with plenty of rules borrowed from
the insight of others and as it were graft these onto it, nevertheless the faculty
for making use of them correctly must belong to the student himself, and in the
absence of such a natural gift no rule that one might prescribe to him for this
aim is safe from misuse. (A133/B172)
The natural gift that Kant speaks of here is not that of genius; that is, it is
not the creative talent. It is so to speak the natural gift of precision in
judgment. To elaborate this capacity thematically leads us to the account
of common sense in the third Critique. Indeed, the term Mutterwitz in the
German can be used to mean “common sense” (seinen Mutterwitz erkennen
lassen = to display one’s common sense). In the Anthropology this
connection between the capacity to judge in concreto and common sense is

made more explicit:
In order to judge human beings according to their cognitive faculty (under-

standing in general), we divide them into those who must be granted common
sense (sensus communis), which certainly is not common (sensus vulgaris), and
people of science. The former are knowledgeable in the application of rules to
cases (in concreto); the latter, in the rules themselves before their application (in
abstracto). (An 7:139)
While we have now an initial understanding of the relation of the common

sense to our sense of the ground of nature that is to receive systematic
articulation, we still need to account for the kind of communality that is
internal to the notion of common sense. Through the discussion of common
sense, Kant suggests that there is a form of natural communality in the
aesthetic judgment. Now, we can certainly conceive of something like a
community of taste that is formed as a space of culture. But what is it to
think of the common natural ground in that space?
This is not immediately evident in the Fourth Moment, but becomes
apparent as Kant returns to the issue of common sense later in the work. This
is not to say that there is a common sense that must be understood in relation
to cognition and another capacity, say one called sensus communis, that is
turned to the social or the communal. Rather, the two dimensions of our
knowledge of the world and of our relation to others are brought together by
way of Kant’s unified understanding of the common sense. This is indeed
one of the ways in which we can appreciate how the third Critique is a bridge
between nature and freedom.11
In §40, Kant emphasizes what we call the public, social, or communal
character of common sense:
By “sensus communis,” however, must be understood the idea of a communal

sense, i.e., a faculty for judging that in its reflection takes account (a priori) of
everyone else’s way of representing in thought, in order as it were to hold its
judgment up to human reason as a whole and thereby avoid the illusion which,
from subjective private conditions that could easily be held to be objective,
would have a detrimental influence on the judgment. (CJ 5:293)
11
Hannah Arendt takes Kant’s Critique of the Power of Judgment to be the blueprint for his political
philosophy. In this context she takes his discussion of common sense in §40 to be of particular
importance. See Hannah Arendt, Lectures on Kant’s Political Philosophy, ed. Ronald Beiner (Chicago:
University of Chicago Press, 1992), 63–73.
422 E. Friedlander
Of course, a simple expedient would appear to be available here – namely, to

assume that as a matter of nature we all have the same capacities; therefore,
what feels natural to one under appropriate conditions can be also inferred to
be what we have in common and can be called therefore a “public sense.” But
this does not take into account the necessity, in judging, to put into play the
faculties, which we share in a way that is not determined by, yet is fully
attuned to, the singular case we are judging. Indeed, what is precisely posing
the need for representativeness in judgment is the problem raised by relating
our capacities in general to the singular case. Representativeness must rely on
a ground which is sensed, yet not altogether made explicit or realized. To be
truly a speaking for others, it must take itself to actualize a certain natural
potential of rightness at the heart of our judgment: the common sense.
Kant seeks to distinguish the idea of taste as a common sense from the
identification we often make of the common with the vulgar, or the average,
that is the lowest common denominator upon which people would agree. As
Kant puts it, the latter “is regarded as the least that can be expected from
anyone who lays claim to the name of a human being.” He further points out
that “to possess [that which is encountered everywhere] is certainly not an
advantage or an honor” (CJ 5:293). The vulgar or the merely popular is
precisely what lacks the fine precision of an authentic notion of common
sense. It is merely resting on prejudice and hasty generalizations that seem-
ingly receive authentication from their being strongly felt rather than
thought through.
But Kant warns against a parallel mistake of identifying common sense with
what can be called “common understanding.” Kant lists three criteria for
attributing to someone such common understanding: “1. To think for oneself;
2. To think in the position of everyone else; 3. Always to think in accord with
oneself” (CJ 5:294). These are, so to speak, formal characterizations of the use
of the understanding: The first avoids passivity, that is, relying on the judgment
of others, or what one calls prejudice (its most dangerous form being super-
stition). The second avoids narrowness of the mind and encourages the mental
habit of making a final use of our faculties. The third maxim of consistent
thought is, according to Kant, “the most difficult to achieve, and can only [be]
achieved through the combination of the first two and after frequent observance
of them has made them automatic” (CJ 5:295).
Common understanding is indeed an important model in thinking what is
characteristic of common sense, but it cannot be identified with it, precisely
because the former is understanding and the latter must be based on sense,
that is on feeling. The question then becomes how to translate the three
features of common understanding into characteristics of taste, or judgment
resting on feeling. There is no difficulty in translating into aesthetic terms the

requirement to rely on oneself. It is, namely, to judge always based on one’s
own feeling, rather than adopting the judgment of others. The second
requirement for enlarging the use of our faculty is identified in aesthetic
terms with judging with a universal voice. One’s judgment must not rely on
one’s own narrow purpose or interest, but rather incorporate within itself the
finality belonging to the idea of universal agreement. But the most difficult
question is how to account in aesthetic terms for the requirement of con-
sistency. Consistency seems so much to be a matter of conceptual relation-
ships of dependence, of logic, that the notion of aesthetic consistency might
appear itself to be inconsistent. It is here that one must seek to bring in our
earlier characterization of common sense as the feeling that belongs to
systematicity in a space of meaning and whose subjective manifestation is
the harmony of the faculties.
The naturalness of common sense is not nature outside and opposed to
human conventions, but rather that which is nature or ground of our convic-
tions in our existence together. It is to be understood as the dimension of life in
our conceptual space. It is striking to realize how many of the central concepts
that Kant uses in relation to the account of common sense in the Fourth
Moment are indeed pertinent to our ordinary notion of having common
sense. It is having a balanced standpoint on things. In judging by relying on
common sense, the imagination, our mediating capacity, is equally weighing
disparate aspects of our experience, and brings them into balance. If there is
something against which common sense is appealed to, it is excess of one side of
our nature (usually our intellectual excesses). Common sense would be under-
standing that the aesthetic involves a higher idea of the ordinary.12
12
In his early essay “Aesthetic Problems of Modern Philosophy,” Stanley Cavell relates the ordinary
language practices of Austin and Wittgenstein to Kant’s account of the aesthetic judgment, writing that
“Kant’s ‘universal voice’ is, with perhaps a slight shift of accent, what we hear recorded in the
philosopher’s claim about ‘what we say’” (“Aesthetic Problems of Modern Philosophy,” in Must We
Mean What We Say? [Cambridge: Cambridge University Press, 1976], 94). Lacking any empirical
grounds, the conviction in the philosopher’s claim depends in part on his ability to make what is thus
exemplified appear surprisingly natural to us, testifying to our form of life, as if a “natural ground of our
conventions” (Stanley Cavell, The Claim of Reason: Wittgenstein, Skepticism, Morality, and Tragedy [New
York: Oxford University Press, 1979], 125). In thematizing this investigation of the grammar of
language, Cavell appeals once more to a Kantian term pertaining to judgment, to the schematism of
judgment. The grammar of ordinary language reveals the lines of projection of the imagination. In
ordinary language, we can present the intuitive design of the space of possibilities of our concepts. Cavell
takes Emerson (and Wittgenstein) to be broadening the Kantian idea “so that it speaks not alone of
deducing twelve categories of the understanding but of deriving – say schematizing – every word in
which we speak together” (Stanley Cavell, Conditions Handsome and Unhandsome: The Constitution of
Emersonian Perfectionism [Chicago: University of Chicago Press, 1990], 39).
424 E. Friedlander
We might find some support for this reading by considering a passage

from the Anthropology in which Kant sets up the opposition between com-
mon sense and madness:
The only universal characteristic of madness is the loss of common sense (sensus
communis) and its replacement with logical private sense (sensus privatus). . . . For
it is a subjectively-necessary touchstone of the correctness of our judgments
generally, and consequently also of the soundness of our understanding, that
we also restrain our understanding by the understanding of others, instead of
isolating ourselves with our own understanding and judging publicly with our
own private representations, so to speak. . . . He who pays no attention at all to
this touchstone, but gets it into his head to recognize private sense as already
valid apart from or even in opposition to common sense, is abandoned to a play
of thoughts in which he sees, acts, and judges, not in a common world, but
rather in his own world (as in dreaming). (An 7:219)
It is crucial to realize here that no criterion of consistency, understood in

merely logical terms, will suffice. For madness is utterly consistent (in a way
too consistent). It is a “logical private sense.” To allow the attunement to
others to be a subjectively necessary touchstone of judgment, just is to have
common sense. Though it is a position in conceptual space, it is something
to be assessed in feeling, and its highest manifestation is that which occurs in
aesthetic judgment.
19
Immediate Judgment and Non-Cognitive
Ideas: The Pervasive and Persistent
in the Misreading of Kant’s Aesthetic
Formalism
Jennifer A. McMahon
Introduction: Three apparent dichotomies

The key concept in Kant’s aesthetics is “aesthetic reflective judgment,” a
critique of which is found in Part 1 of the Critique of the Power of Judgment
(1790). It is a critique inasmuch as Kant unravels previous assumptions
regarding aesthetic perception. For Kant, the comparative edge of a “judgment”
implicates communicability, which in turn gives it a public face; yet “reflection”
points to autonomy, and the “aesthetic” shifts the emphasis away from objective
properties to the subjective response evoked by the object. Contrary to the view
he had inherited, Kant did not treat aesthetic judgments as confused percep-
tions or inferior cognitions, but rather put them in a category of their own.
For Kant, aesthetic reflective judgments implicated not only the processes
involved in understanding the empirical world but also the processes
involved in ascribing ideas or meaning to experience (which go beyond
what is given in perception/cognition). This “going beyond cognition”
may have been referred to by Kant as non-cognitive, but it is conditioned
on a way of cognizing the object. Kant describes the features of such
judgments in the Analytic of the Beautiful before addressing their possibility
in subtle ways throughout the Deduction of Pure Aesthetic Judgments.
J. A. McMahon (*)
Department of Philosophy, The University of Adelaide, Adelaide, Australia
e-mail: jenny.mcmahon@adelaide.edu.au

426 J. A. McMahon
Kant’s critique gives rise to three apparent dichotomies: immediate intui-

tion and prior learning; autonomy and communicability; and non-cognitive
response and aesthetic ideas. An interpretation of Kant’s theory can be
deemed correct to the degree that it succeeds in showing that the contra-
dictions are only apparent. I will argue that the resulting positive account of
Kant’s aesthetic theory offers an understanding of aesthetic reflective judg-
ment that is quite different from the standard formalism that the critics of
formalism usually target and by which they misrepresent Kant’s aesthetic
theory. When I refer to Kant’s formalism, then, I mean something quite
different by this than what is meant by “standard formalism.” The latter is
the representation given Kant’s formalism by its contemporary critics and is
the typical account of formalism found in the art theory and philosophy of
art literature, as will be demonstrated.
Intuition: Immediate judgment and prior learning

Clement Greenberg (1909–1994) was an influential American art critic who
explicitly referenced Kant’s aesthetics and thereby contributed to the stan-
dard interpretation of Kantian aesthetics in the Western art world.
Greenberg’s notion of the “immediacy” of aesthetic judgment is a case in
point. He writes:
Esthetic judgments are given and contained in the immediate experience of art.
They coincide with it; they are not arrived at afterwards through reflection or
thought. Aesthetic judgments are also involuntary: you can no more choose
whether or not to like a work of art than you can choose to have sugar taste
sweet or lemons sour.1
The typical interpretation given to Greenberg’s approach as expressed here is

that aesthetic judgment involves immediate intuition and, in turn, immediate
intuition is an exercise of perception that does not draw upon background
knowledge. This is a tenet of standard formalism. One can immediately counter
with the fact that, given that our intuitions develop and change over a lifetime,
they must supervene on knowledge, experience, and training. This is considered
a commonsensical view now, but it was already embedded in Kant’s critique of
our capacities for reason, morality, and sociable feeling.
1
Clement Greenberg, “Complaints of an Art Critic,” in Modernism, Criticism, Realism: Alternative
Contexts for Art, ed. Charles Harrison and Fred Orton (New York: Harper & Row, 1984), 4.
19 Immediate Judgment and Non-Cognitive Ideas 427
Nonetheless, according to standard formalism, aesthetic judgment involves

an “immediate communion with a work of art without recourse to auxiliary
structures of knowledge,” in the words of Roxie Davis Mack.2 This is the view
attributed to Greenberg and understood as originating in Kant. However,
Kant’s explanation of the basis of our engagement with a work of art sets his
view clearly apart from this tenet of standard formalism, as we will see.
Greenberg, however, does attempt to avoid the false dilemma between
subjective reception based on irreducible intuitions (immediate) and objec-
tive reception grounded in knowledge and arrived at through reasoning
(formulaic). According to Greenberg, the critic responds subjectively with
irreducible intuitions to objective aesthetic properties, but due to her super-
ior taste and sensibility, her irreducible intuitions constitute an objective
aesthetic judgment. According to Deane Curtin, Greenberg conceives of
objectivity as un-analyzability.3 Greenberg sometimes writes as though the
perceptual object is merely given, as though he holds a naive theory of
perception. Disagreements are implicitly treated as due to one critic having
a better eye, being more sensitive and perceptive, as though what is appre-
hended in aesthetic judgment is simply there in the objective properties of
the work of art but only perceived by the superior critic. The notion
promoted is that the correct interpretation would be self-evident to the critic,
and should be adopted by the ordinary person based on the critic’s authority
alone. This approach sets up a tyranny of taste and explains the accusations
of elitism often aimed at standard formalism.
There is an important difference between this view and the notion of the
ideal critic found in Hume’s Of the Standard of Taste.4 Hume provides the
means whereby we might approximate an ideal critic and in turn judge the
expertise of actual critics by whether they are able to enhance our aesthetic
engagement. Kant develops this self-legislation further through his notion of
autonomy. One should note that aesthetic autonomy for Kant is not a case of
idiosyncrasy, whimsicality, or private impressions but neither is it a subjec-
tion to another person’s judgment. One must find the basis of the judgment
for oneself.
2
Roxie Davis Mack, “Modernist Art Criticism: Hegemony and Decline,” Journal of Aesthetics and Art
Criticism 52, no. 3 (Summer 1994): 341–48.
3
See Deane W. Curtin, “Varieties of Aesthetic Formalism,” Journal of Aesthetics and Art Criticism 40,
no. 3 (Spring 1982): 323.
4
David Hume, Of the Standard of Taste (1757), in Essays Moral, Political, and Literary, ed. Eugene F.
Miller (Indianapolis: Bobbs-Merrill, 1987), 226–49.
428 J. A. McMahon
When a critic uses metaphor, analogy, and prior example to demonstrate

her perception of a work of art, she is providing critical reasons to inform the
judgment of others. This is not mere coercion or deference to authority. The
test for the aptness of the critic’s reasons is the resulting experience one has of
the work in question. We might judge the critic’s expertise by whether re-
perceiving the work in the way she demonstrates augments the significance or
insight that the work affords us, in our experience of it. Where aesthetic
reflective judgments are concerned, critical reasons operate in just this way.
Critical reasons used to communicate an aesthetic judgment do not aim
at reaching a theoretical conclusion but instead aim to have an impact
upon our perception or experience of an object. As such, they are not
grounded in private intuitions (or private reason – about which, more
later). Consequently, an aesthetic reflective judgment may be immediate
in the sense that it engages one’s feeling, but this is not to say that it is
immediate in a temporal sense. In most English translations of Kant’s
critique of judgment, unmittelbares is translated as “immediate,” but
Werner Pluhar explains that he translates it as “direct” in order not to
mislead the reader into thinking that it refers to temporal immediacy.5
One’s aesthetic reflective judgment of a particular object is directly felt but
is not on this basis quarantined from the effects of past or (in subsequent
encounters with the object) future experience, education, and training.
Kant describes how our aesthetic judgment changes over time:
a young poet does not let himself be dissuaded from his conviction that his
poem is beautiful by the judgment of the public nor that of his friends. . . .
Only later, when his power of judgment has been made more acute by practice,
does he depart from his previous judgment of his own free will, just as he does
with those of his judgments that rest entirely on reason. (CJ 5:282)
Kant draws attention to the role of previous experience when he writes that a
person whose aesthetic judgment is not shared by others “does not allow
approval to be internally imposed upon himself by a hundred voices who all
praise it highly . . . [even though] he can even begin to doubt whether he has
adequately formed his taste by acquaintance with a sufficient number of
objects of a certain kind” (CJ 5:284). And Kant acknowledges the role of the
critic in developing our taste by providing the right kind of reasons through
metaphor, analogy, and prior example when he says that “I cannot be talked
5
Immanuel Kant, Critique of Judgment, trans. Werner S. Pluhar (Indianapolis: Hackett, 1987), 49n15
(at CJ 5:208).
into [aesthetic pleasure] by means of any proofs. . . . However, . . . critics

nonetheless can and should reason . . . in a way that is useful for correcting
and broadening our judgments of taste” (CJ 5:285–86). In other words,
arriving at any particular aesthetic judgment may have taken years.
Obviously, the process proceeds in an incremental fashion over a lifetime.
As our perceptions evolve, we might find an artwork insightful where
previously we had found it boring, expressive where previously we had
found it pompous. Aesthetic judgments are not immediate in being all or
nothing on the spot, so to speak. We can re-construe, reconfigure, and
reconstruct our perception to arrive at a new perceptual object. We may be
looking in the same direction as before, listening to the same sounds, but
perceiving something quite different from previous experiences of the same
ostensible object.
Kant recognizes that when we aesthetically judge an object we have a
direct response. We do not consciously reason our way to our response.
However, this is not incompatible with prior reasoning, reflection, and
associated valuing being responsible for the direct response I have at any
given time. The way an object is received can change through acculturation,
which includes hearing how others construe it. In contrast, standard form-
alism implies a notion of perception according to which perceptual objects
are simply given and our perception of them infallible. For such a formalist,
“there is no arguing about taste” means that taste is incorrigible. Contrast
this standard formalism with Hume, who famously made this claim about
taste, yet who nonetheless thinks that taste is susceptible to acculturation.6
He discusses passions generally as a part of a constellation of associations
whose configuration might change in response to social norms. When he
writes that there is no arguing about taste, this is in the context of setting up
the apparent opposition between sentiment and reason in normative judg-
ment.7 He solves the opposition in part by way of a social context of norms
and conventions involving recognition of expertise, the latter involving
knowledge and experience of an appropriate range of objects within the
relevant category.
In a similar vein, when Kant writes that one cannot dispute about taste, he
also does so in setting up an apparent antinomy of taste that he sets out to
6
See David Hume, A Treatise of Human Nature, ed. L. A. Selby-Bigge (Oxford: Clarendon, 1973), bk.
II, pt. iii, §§ 4–5 (pp. 418–24). See also Paul Guyer’s analysis in “Reason, Desire, and Action,” in
Knowledge, Reason, and Taste: Kant’s Response to Hume (Princeton: Princeton University Press, 2008),
173.
7
For Hume’s dialectical approach to arriving at a standard of taste, see his Of the Standard of Taste.
430 J. A. McMahon
solve (CJ 5:338). Instead of Hume’s sociological solution, Kant identifies the
peculiar kind of rule-based judgment involved (CJ 5:341). The rule is
implicit in our attempts to find the basis of our judgments within ourselves
and to communicate this basis to others. Kant makes it clear that commu-
nicability is a key feature of the very possibility of aesthetic reflective judg-
ments when he writes that “taste is thus the faculty for judging . . . the
communicability of the feelings that are combined with a given representa-
tion (without the mediation of a concept)” (CJ 5:296). The aspect of art
which exercises aesthetic reflective judgment is that aspect by which it
communicates feeling.
Many art critics associated with standard formalism, such as Clive Bell,
Greenberg, and Michael Fried, draw attention to the fact, implicitly or
otherwise, that what characterizes an object as art is not literal narrative or
the representational aspect of objects. However, their lack of clarity over the
way that aesthetic judgment is acculturated leaves them vulnerable to accusa-
tions of elitism, as noted earlier. This was the cast that, once set, was carried
forward into subsequent years of art theorizing and philosophizing in the
name of Kant.
Kant’s approach to aesthetic reflective judgment, however, can be seen to
have developed in a very different direction from the standard formalism of
Bell, Greenberg, and Fried, which has implications for the interpretation of his
account. According to Kant, aesthetic reflective judgment is grounded in
knowledge, experience, and training, which are acquired within a society and
which allow for open discussion and critique. This directs our aesthetic judg-
ment out from a more private or idiosyncratic impression. This is because
engagement with the judgments of others allows us to edge our judgments
toward being informed by a basis available to everyone. So the choice is not
between subjectivity and objectivity, a dilemma that the formalist art theorists
attempt to subvert by developing terms which position their subjective
responses as evidence of certain objective aesthetic properties.8 Instead, the
issue for Kant is a choice between the private use of reason and the public use of
reason. Private reason is ultimately a case of idiosyncratic impressions, dogma-
tism, or ignorance, with “ignorance” defined as the state of taking one’s
impressions or dogma to be infallible and not revisable. By contrast, public
reason is based on considerations accessible to everyone, so it is continually
8
For a description of the subjectivism-objectivism debate, see Brian Watkins, “The Subjective Basis of
Kant’s Judgment of Taste,” Inquiry 54, no. 4 (Aug. 2011): 315–36.
open to revision and is the way of enlightenment. We turn now to how,

according to Kant, public reason is reflected in aesthetic reflective judgments.
The “form of purposiveness”: Autonomy and

communicability
Kant’s aesthetic theory is very ambitious. With it he is attempting to explain
how communities of rationally grounded sentiment are possible. He is in
effect dissolving the sentiment-reason dichotomy that he inherited from his
intellectual forebears. But for this he needs to develop a complex apparatus
which can seem unwieldy at times. In this section we address how Kant
envisaged autonomy of sentiment and communicability (which implicates
rational structure and conventional forms) as complementary components of
aesthetic reflective judgment. This will involve a discussion of aesthetic
standards, Kant’s theory of genius, the historicity of aesthetically relevant
properties, and his most difficult but important concept, the “form of
purposiveness,” in order to demonstrate how autonomy and communicabil-
ity ground aesthetic reflective judgment in what Kant calls the sensus
communis.
Turning first to the question of standards, implicit in Kant’s aesthetics
is that there is no objective mind- or culture-independent standard out
there that we can use to judge or interpret art or nature. Art and nature
will always be contextualized relative to a culture – the understanding,
knowledge, experience, and training available to a people at a given point
in time. If we did actually have a set formula for deciding which aesthetic
evaluations and interpretations were correct, we would lose the power to
facilitate critical thought, increase understanding, and calibrate feeling to
newly emerging societal norms. However, there is another way to ground
standards. A standard may be considered objective depending on the
degree to which it approaches an ideal of some sort. Kant’s conception
of the sensus communis, which is an expression of his account of public
reason, is the objective basis he provides for a standard of aesthetic
judgment. The standard is ideal to the extent that it is identified by
engaging the sensus communis.
Kant writes that we assume a common sense “as the necessary condition of
the universal communicability of our cognition, which is assumed in every
logic and every principle of cognitions that is not skeptical” (CJ 5:239). The
common sense (sensus communis) to which Kant refers is constituted by three
432 J. A. McMahon
interconnected maxims: (1) think for yourself; (2) think from the standpoint
of everyone else based on principles that are open to others; and (3) think
consistently (CJ 5:293–96). Kant writes of the maxims that they are of “a
faculty for judging that in its reflection takes account (a priori) of everyone
else’s way of representing in thought, in order as it were to hold its judgment
up to human reason as a whole” (CJ 5:293). The comparative element of
judgment operates within all reason, but in aesthetic reflective judgment the
grounds of its validity are brought out into the open.
Onora O’Neill argues that the passages on the sensus communis in the
Critique of the Power of Judgment “distinguish different aspects of reason’s
task more sharply” than in the Critique of Pure Reason, including that “reason
presupposes plurality-without-preestablished-harmony.”9 O’Neill writes:
Taken in conjunction, the three maxims define constraints for a dynamic

process in which the demand for lawfulness without a lawgiver is realized
among a plurality. Reason is here [in the third Critique] sketched not as
abstract principle, but as the lawlike guidance of thinking and doing in a
dynamic process that neither submits to outside control nor fails to acknowl-
edge differences of opinion and practice, and which treats resulting contra-
dictions and tensions as an indefinitely extended demand for revision.10
In other words, in order for public reason to operate within the individual, it
requires an exchange of different perspectives.
With the sensus communis, reason is demonstrated as operating to establish a
shared basis of communicability of feeling without the mediation of a concept.
Standards are objective not in virtue of objective properties that an individual can
name, but in virtue of the three maxims of the sensus communis that encompass
autonomy and objectivity in reasoning and judgment. According to O’Neill,
“Autonomy, as Kant understands it, is not mere self-assertion or independence,
but rather thinking or acting on principles that defer to no ungrounded ‘author-
ity,’ hence on principles all can follow.”11 Ungrounded authority refers not only
to dogma or deference, but also idiosyncrasy or compulsion. Autonomy is an
exercise of our freedom, and, for Kant, freedom is thinking and acting in ways for
which we can be held responsible. Aesthetic autonomy, in a similar vein, is a
matter of a “liking” and valuing for which we can be held responsible. This is
9
Onora O’Neill, “Vindicating Reason,” in The Cambridge Companion to Kant, ed. Paul Guyer
(Cambridge: Cambridge University Press, 1992), 301.
10
Ibid.
11
Ibid., 299.
among the crucial differences between Kant’s formalism and standard formalism.
Elitism, of which the latter is often accused, precludes the autonomy and
communicability that characterize Kant’s formalism.
The sensus communis provides the standard for aesthetic reflective judgment.
Through the role he gives autonomy, principles, and consistency in what
constitutes the sensus communis, Kant argues that judging or acting on first
impressions which are unaccountable, idiosyncratic, and hence not communic-
able, is not an example of aesthetic reflective judgment. Bearing in mind that
the principles of aesthetic reflective judgment are communicated through
metaphor, analogy, and prior example, Kant shows how aesthetic reflective
judgment can engage one subjectively but focus one upon aspects of objects that
cause a response that one would think everyone ought to share.
Jane Kneller argues that the formal structure of aesthetic judgment, as
identified by Kant, involves reflection that broadens the mind to entertain
possibilities beyond one’s own goals.12 This is partly to do with the inde-
terminacy – yet principled nature – of it, which continually prompts one to
compare one’s judgment to the judgment of one’s peers or what we take to be
judgment in general, lest we miss something significant. Such a judgment
can be understood as an appropriate model for the kind of judgment which
would facilitate communication between those who hold varying cultural
beliefs. That is, aesthetic reflective judgment models the kind of judgment
for which a plurality of perspectives is an invitation to reflect rather than a
threat. This is in line with O’Neill’s view that a plurality of perspectives is
Kant’s starting point for considering the nature of reason, and her claim
(discussed above) that he develops a clearer formulation of this in the third
Critique through the sensus communis. Kneller concludes that aesthetic
reflective judgment, given that it is grounded in feeling but acculturated
through communication, is the kind of experience of alternative viewpoints
that facilitates empathy and communication.
Kant demonstrates how aesthetic autonomy operates in detail through his
account of artistic genius. He explains how the artist learns her craft from the
experts but in a way that does not undermine her autonomy. Kant explains
that novices who want to know how to create art model their procedures on
12
Jane Kneller, “Aesthetic Reflection and Community,” in Kant and the Concept of Community, ed.
Charlton Payne and Lucas Thorpe (Rochester: University of Rochester Press, 2011), 260–83. There is a
very helpful discussion of Kant’s conception of a socially grounded reason in Jane Kneller, “Introducing
Kantian Social Theory,” in Autonomy and Community: Readings in Contemporary Kantian Social
Philosophy, ed. Jane Kneller and Sidney Axinn (Albany: State University of New York Press, 1998),
1–14. See also Hannah Arendt, Lectures on Kant’s Political Philosophy, ed. Ronald Beiner (Chicago:
Chicago University Press, 1982), 40–46.
434 J. A. McMahon
the genius, taking what the genius produces as a rule. However, mere copy-
ing must give way to emulation. Emulation involves recognizing a rule
(inadvertently or otherwise), internalizing it, and producing a work of art
based on it, rather than actually copying any external representation of it. If
produced in this way, the resulting artwork could never have been explicitly
inferred from previous rules, principles, or examples even if in retrospect one
can detect the direction of influence (CJ 5:316–17). It will seem different to
the artwork it emulates and in some cases might produce a new rule that
others strive to emulate. And so it goes.
There is a further point we can draw from this. While Kant does not explore
the consequences of the methods of genius for the question of the history of
aesthetically relevant properties, it is clearly the case that there is scope for
aesthetically relevant properties to develop and change over time given the way
genius operates within his account.13 This is in line with standard formalism in
one respect but not in others. A standard formalist position is that art and
aesthetically relevant properties have their own trajectory historically, and they
develop independently from politics and other social realities.14 In response,
one could question how anyone could bracket out the influences of the various
ideas and understandings that arise within one’s culture. One need hardly be
aware of those influences for one’s concepts and ideas to be shaped by them and
in turn for one’s feelings in relation to specific objects to be impacted. Another
tenet of some versions of standard formalism, such as the version proposed by
the influential British art critic Roger Fry (1866–1934), is that art reflects an
order that we would impose on nature and life; and as part of this, art’s
development toward abstraction is built into its very nature.15
Since Fry’s lifetime, however, art has turned back from abstraction and
splintered into a myriad of tendencies, among which is a more direct and
explicit interface with the politics and social mores of the day. The contem-
porary French art critic Nicolas Bourriaud has termed this trend “relational
aesthetics.”16 So, aesthetically relevant properties have developed and changed
13
Paul Crowther takes the opposite view, arguably because he does not acknowledge the relevance of
Kant’s theory of genius to the question of the historicity of aesthetically relevant properties. See Paul
Crowther, “Kant and Greenberg’s Varieties of Aesthetic Formalism,” Journal of Aesthetics and Art
Criticism 42, no. 4 (Summer 1984): 445. For an in-depth analysis of the concept of “aesthetically
relevant properties,” see Bence Nanay, Aesthetics as Philosophy of Perception (Oxford: Oxford University
Press, 2016).
14
For example, see Roger Fry, Vision and Design (London: Chatto & Windus, 1920).
15
Ibid.
16
Nicolas Bourriaud, Relational Aesthetics, trans. Simon Pleasance and Fronza Woods (Dijon: Les
Presses du réel, 2002).
over time, but not according to the predetermined and necessary trajectory
predicted by Fry. Interestingly for our concerns, “relational aesthetics” is
represented as a victory for anti-formalism, contra standard formalists such
as Fry but also including Kant’s aesthetics among the defeated. However,
Kant’s theory of genius does not only implicate a historicity for aesthetically
relevant properties; it also implicates the genius’s embeddedness in culture
more broadly. The thought here is that a community could only recognize
the insight of an artwork if the artwork’s creator shared aspects of her
experience, background, knowledge, and training with her community.
Giving the rule to art and recognition of that rule by others involves
practice, acquaintance with a sufficient number of relevant objects, and
having had one’s judgment of taste corrected and broadened by aesthetic
reasons (CJ 5:282–86). Unlike standard formalism, for Kant aesthetic
autonomy does not mean art’s separateness from the other aspects of
culture. In fact, Kant explains his notion of “form” in terms of aesthetic
ideas. The “form” given to aesthetic ideas for the purposes of communica-
tion may draw upon all manner of topical content (as we will see).
So far in this section, we have discussed standards, genius, and the
historicity of aesthetically relevant properties in the course of uncovering
Kant’s particular concept of aesthetic autonomy. Now I turn to his concept
of communicability. We saw in the previous section that the critic commu-
nicates her aesthetic judgment through metaphor, analogy, and prior exam-
ple, but we have yet to discuss what is being judged and communicated in
aesthetic reflective judgment apart from “the communicability of the feel-
ings” (CJ 5:296). Unlike standard formalism, which treats the content of
aesthetic judgment in terms of formal relationships, Kant discusses the
content of what our aesthetic reflective judgments communicate in terms
of aesthetic ideas. The fact that Kant’s aesthetics can still be treated by some
as standardly formalist in spite of this confirms one’s impression that such
interpretations are based on the Analytic of the Beautiful and ignore the
Deduction of Pure Aesthetic Judgments, which is where the doctrine of
aesthetic ideas, genius, and the sensus communis are set forth. The Deduction
takes up half of the sections that constitute the “Critique of the Aesthetic
Power of Judgment.” An accurate account of Kant’s aesthetics cannot be
given without it.
Consider that beauty for Kant involves an experience of the harmony of
imagination and understanding. Whether the harmony is of a degree that
would warrant an experience of beauty depends not only on the object
presented but the concepts with which the understanding is equipped.
Aesthetic reflective judgment operates “over and above that harmony with
436 J. A. McMahon
the concept” rather than independently of that harmony with the concept
(CJ 5:317). Aesthetic reflective judgment is conditioned on such a harmony,
which must involve, in particular or in general form, the concept of the
relevant object, even if the concept is of an unidentified or purposeless object.
For Kant, an experience of beauty involves experiencing an object as
expressive of one’s own deeply felt ideas. Yet “ideas” (as opposed to concepts)
are conceived by Kant to explain how subjectivity can be expressive of
community norms and public reason. Kant writes:
Beauty (whether it be beauty of nature or of art) can in general be called the

expression of aesthetic ideas: only in beautiful [schön] art this idea must be
occasioned by a concept of the object, but in beautiful nature the mere
reflection on a given intuition, without a concept of what the object ought
to be, is sufficient for arousing and communicating the idea of which that
object is considered as the expression. (CJ 5:320)
Kant distinguishes concepts from ideas (CJ 5:342–43) in order to distinguish

the processes involved in understanding the empirical world from the pro-
cesses involved in ascribing ideas or meaning to experience (which go beyond
what is given in perception/cognition). An experience of aesthetic ideas
involves the latter, but importantly, it is an experience of objects, not mere
daydreaming.
One could think of aesthetic ideas as a loose collection or constellation
of fragments, feelings, ideas, and facts brought together merely by being
associated with a concept, though not unified or subsumed under it.
Nonetheless, there is a certain “point” to them, and this gives one an
experience of a quickening of the mental powers, so to speak, and is also
the basis of their communicability. Although this does not lend itself to
easy characterization, Kant refers to this “point,” as experienced by us, as
“purposiveness without purpose [Zweckmäßigkeit ohne Zweck].” In the
Analytic of the Beautiful, Kant uses design as a demonstration of the
principle of “purposiveness without purpose,” but this is a limit case, a
demonstration of the principle, not a recommendation for art (CJ 5:221,
228–29, 236; translation modified). Yet his meaning is distorted when
seen in terms of standard formalism’s emphasis on elements like shape,
line, and texture. Consider that for Kant aesthetic reflective judgment
“does not bring to our attention any property of the object, but only the
purposive form in the determination of the powers of representation that
are occupied with it” (CJ 5:228). Hence we attend to the way we are
engaged by the object, which in turn we interpret as indicative of some
property of the object. What aspect of objects might direct our attention
to the “way” they engage us? The aspect that we experience as expressive
of aesthetic ideas, is Kant’s answer.
Standard formalism conceives of “form” relative to the self-reflexive aspect
of existing art forms. By contrast, according to Kant’s formalism, the artistic
genius in finding a communicative form for aesthetic ideas maintains a
dynamic relationship between our experience and our concepts. This is
what he means when he writes:
the aesthetic idea is a representation of the imagination, associated with a given

concept, which is combined with such a manifold of partial representations in
the free use of the imagination that no expression designating a determinate
concept can be found for it, which therefore allows the addition to a concept of
much that is unnameable, the feeling of which animates the cognitive faculties
and combines spirit with the mere letter of language. (CJ 5:316)
The long and short of it is that we experience beauty when we find the world
or objects in it conducive to our own sense of greater purpose. Finding the
same things beautiful as our peers is a matter of calibrating aspects of our
experiential stock with theirs. Achieving this level of communicability is a
complicated process and requires embeddedness in a community. Kant calls
the experiential stock “aesthetic ideas,” and their form is of crucial concern to
him. Kant calls this form, the “form of purposiveness” (CJ 5:221).
The “form of purposiveness” is the means by which a shared basis of
subjective responses is possible. In other words, aesthetic reflective judgment
focuses upon that aspect of objects by which they express their meaning and
significance for us. It is an aspect which goes beyond the prosaic purpose of
the object and implicates those elements of personal experience for which we
might expect to find common ground with others.
In this section, we have found that aesthetic reflective judgment
operates in a way analogous to public reason in virtue of being grounded
in the sensus communis. The three maxims of the sensus communis incor-
porate autonomy as thinking for oneself and communicability as think-
ing from the standpoint of everyone else based on principles that are
open to others. However, more needs to be said regarding the feeling or
experience that is structured to community norms (and is hence com-
municable) while still remaining autonomous. Kant qualifies the relevant
feeling further by establishing a unique category for it. How and why
will be addressed in the next section.
438 J. A. McMahon
Disinterest: Non-cognitive engagement and

aesthetic Ideas
Kant’s notion of disinterest draws out a particular feature of aesthetic
reflective judgment that has become and remains a highly contentious issue
in the art theory and aesthetics literature. For example, Christine Battersby
adopts the standard formalist view that disinterest refers to the independence
of aesthetic judgment from the ethical and political, a view she attributes to
Kant, and on the basis of which she claims that feminists showed little
interest in the nature of aesthetic judgment or beauty because it was con-
nected too irrevocably with Kant.17 On the contrary, we have seen above in
discussing aesthetic ideas that aesthetic judgment for Kant could be con-
nected to world themes in complex ways, through the meaning and signifi-
cance that we ascribe to objects expressive of aesthetic ideas. Being in excess
of the concepts with which they are associated, aesthetic ideas introduce
much from experiences felt to be peculiar to the individual. Yet through
aesthetic reflective judgment, art and nature can be found to be expressive of
these ideas. Like many critics of standard formalism, Battersby overlooks the
work that the notion of disinterest does in Kant’s account, which is to shift
the perspective of the judger away from private (idiosyncratic) to commu-
nicable. This is an aspect that is clearly represented in the writing of Hannah
Arendt, notwithstanding Battersby’s claims to the contrary.18
Many aestheticians’ understanding of disinterest is hampered by an argu-
ably anachronistic notion of “non-cognitive.” For example, a non-cognitive
judgment for Emily Brady is one that is not determined by knowledge of the
object.19 Her notion of disinterest is explained through an analogy with legal
reasoning. Brady explains that, like the way judgment is engaged in legal
reasoning, disinterested aesthetic judgment satisfies the conditions for a
comparative judgment.20 However, she endorses an interpretation of
17
Christine Battersby, The Sublime, Terror and Human Difference (New York: Routledge, 2007), 45–
46. Sheila Lintott supports Battersby’s view in “Feminist Aesthetics and the Neglect of Natural Beauty,”
Environmental Values 19, no. 3 (Aug. 2010): 324.
18
Hannah Arendt discusses the significance of aesthetic judgment for community in a way that
resonates with O’Neill’s argument for the significance of the third Critique for understanding Kant’s
conception of reason. See Arendt, Lectures on Kant’s Political Philosophy, esp. 27–40. Battersby dismisses,
without justification in my view, Arendt’s well-grounded interpretation of Kant’s aesthetics. See
Battersby, Sublime, Terror and Human Difference, ch. 10.
19
Emily Brady, Aesthetics of the Natural Environment (Edinburgh: Edinburgh University Press, 2003),
34–35, 72–73, 129.
20
Ibid., 33. For Brady’s analogy between aesthetic disinterest and legal reasoning, see 134–35.
Kant’s concept of aesthetic judgment that is both non-cognitive and disin-

terested. Yet, it is not clear how a non-cognitive judgment, as she under-
stands it, is compatible with a comparative judgment, given the latter surely
engages cognition.
The best way to categorize “aesthetic ideas” in Kant’s aesthetic theory is as
an object of the imagination, and as such, dealing with possibilities rather
than beliefs about how things actually are. That is, it would be an anachron-
ism for us to employ the term “non-cognitive” as Kant himself did. In Kant’s
own words, ideas are not cognized. They are
representations related to an object in accordance with a certain (subjective or

objective) principle, insofar as they can nevertheless never become a cognition
of that object. . . . An aesthetic idea cannot become a cognition, because it is an
intuition (of the imagination) for which a concept can never be found
adequate. . . . [But in saying this] it is presupposed that they are not entirely
groundless, but rather (in accordance with the above explanation of an idea in
general) are generated in accordance with certain principles of the cognitive
faculty to which they belong ([aesthetic ideas] according to subjective princi-
ples, [rational ideas] to objective ones). (CJ 5:342)
And a few paragraphs later Kant says:
Just as in the case of an idea of reason the imagination, with its intuitions,
never attains to the given concept, so in the case of an aesthetic idea the
understanding, by means of its concepts, never attains to the complete inner
intuition of the imagination which it combines with a given representation. . . .
Both sorts of ideas, the ideas of reason as well as the aesthetic ideas, must have
their principles, and indeed in both cases in reason, the former in the objective
and the latter in the subjective principles of its use. (CJ 5:343–44)
Although we can see why, in Critique of the Power of Judgment, Kant under-
stands aesthetic ideas and hence aesthetic judgment as non-cognitive, the
same implication is not suggested by contemporary theories of mind.
According to the latter, imagination is an exercise of cognition as can be
seen by the way it is engaged in problem solving, for example.21
According to recent work in the philosophy of mind, to be non-cognitive,
an aspect of experience would be deemed idiosyncratic, impressionable, and
21
See Peter Langland-Hassan, “A Puzzle about Visualization,” Phenomenology and the Cognitive Sciences
10, no. 2 (June 2011): 145–73.
440 J. A. McMahon
not answerable to reason, on the one hand, or foundational and hence

unanalyzable on the other. Neither sense captures what Kant intends by
“non-cognitive.” Kant means to represent aesthetic reflective judgment as
distinct from the kind of logical judgments that are demonstrated by proofs.
He also means to represent aesthetic reflective judgments as not subsumed by
or limited to determinate concepts, because then they would answer to
explicit purposes. According to Kant, a proof or formula is impossible for
aesthetic reflective judgment, and neither can an aesthetic reflective judg-
ment be determined by a purpose (like an object of design can be judged
perfect in a predetermined way by whether it serves a particular purpose).
This is why for Kant, aesthetic reflective judgment cannot be cognitive. It is
also what Kant means when he writes that aesthetic reflective judgments are
immediate.
While Brady claims that, for Kant, aesthetic judgment is non-cognitive
and disinterested, it is clear from her account of interpretation and imagina-
tion, which she says grounds aesthetic judgment, that she wants to develop a
cognitive theory of aesthetic judgment. Cognition is employed in imagining
ideas in an object-centered way, in attributing meaning and significance to
objects, events, and experience, and hence allows aesthetic judgments to be
open to reasonable justifications (even if only, like practical reasons,
expressed through metaphor, analogy, and prior example).22 This is not so
much an update or revision of Kant, as Brady suggests, but a realization of his
account in a contemporary context. This would be further consolidated if
Brady included the relation between the individual and her community in
her account of aesthetic judgment, rather than limiting the relation to the
perceiver’s encounter with the object. The three-pronged relation between
subject, object, and community is perhaps what Brady is after when she
writes: “Disinterestedness characterizes attention directed outwards rather
than inwards.”23 In spite of her stated intention to update or revise Kant, in
my view this nicely captures Kant’s motivation in introducing the concept of
disinterestedness in relation to aesthetic judgment.
The reason for introducing disinterest to the tradition Kant inherited is to
reconnect the individual mind with other minds where feeling or subjectivity
is concerned. Because Kant envisages the architecture of the mind in terms of
an individual mind, he needs a way to explain how the isolated mind avoids
solipsism or alienation. Paul Guyer argues that Kantian texts written after the
22
Brady, Aesthetics of the Natural Environment, 70–72 and ch. 6.
23
Ibid., 140.
third Critique support conceptions of cognition and value in terms that

would explain how the individual is able to reason and feel in concert with
others in a community.24 Kant believes that, unless we can compare our
thoughts with the thoughts of other people, we would lose the ability to
think. One might say that a notion of community is demonstrated by the
very exercise of reason. Furthermore, Guyer argues that a conception of some
kind of community implicitly grounds what we value.25 Extending this to
aesthetic reflective judgment, unless we develop means by which feeling can
be communicated (not just purged), our capacity for feeling would be
diminished or would remain infantile. Kant’s conception of aesthetic reflec-
tive judgment, like his conception of reason, positions the individual as part
of a community.
This approach can be compared with certain Japanese traditions. According
to the latter, the primary unit of humanity is a community of minds, such that
feeling is not properly human until it is subject to the norms and conventions of
a community. Because of this, a concept of disinterest in relation to aesthetic
judgment is not needed, nor is the dominance of the intellect. According to
such traditions, aesthetic response can remain in the realm of the sensory
because even the sensory properly so called has a conventional form.26 For
Kant, however, whose tradition holds that sensations are somehow founda-
tional and un-acculturated, disinterest does the work of identifying a subjectiv-
ity that is object-directed and subject to cultural norms. In virtue of disinterest,
one can compare one’s response to that of others in terms of aptness. As such,
feeling is indirectly calibrated with community norms that in turn can be
developed within the context of public reason.
So far we have seen that to experience an object aesthetically, we experi-
ence its purposiveness: that is, we experience a quickening of our mental
powers in relation to the ideas that go beyond the concept we have of the
object. As suggested earlier, the purposiveness can be translated as some point
we feel the object has for us, and furthermore, this point goes beyond the
mere concept we have of the object. Going beyond the concept of the object
is what Kant refers to by purposiveness “without a purpose,” and he refers to
24
For example, The Metaphysics of Morals (1797), Religion within the Boundaries of Mere Reason (1793),
and “On the Common Saying: That May Be Correct in Theory, But It Is of No Use in Practice” (1793).
25
Paul Guyer, “Kantian Communities: The Realm of Ends, the Ethical Community, and the Highest
Good,” in Kant and the Concept of Community, 88–120.
26
See Tomoe Nakamura, “Nishi Amane’s Response to European Dualism,” Postgraduate Journal of
Aesthetics 10, no. 3 (Winter 2013): 24–35; and Tomoe Nakamura, “The Philosophical Structure of the
Evaluation of Aisthesis: Comparative Aesthetics between Europe and Japan” (PhD diss., Monash
University, 2015).
442 J. A. McMahon
it as non-cognitive to reflect this “going beyond.” This way of characterizing

aesthetic reflective judgment is clearly not meant as an indication that
aesthetic reflective judgment is aimless and trivial.
Allen Carlson has carved out a very influential position in contemporary
analytic aesthetics, writing on environmental aesthetics. He presents Kant’s
notion of disinterest as referring to a vacant and unemployed state of mind
and one that excludes cognitive dimensions.27 According to Carlson, Kant’s
concept of disinterestedness “stripped and objectified nature.”28 Carlson’s
own position, which he calls “scientific cognitivism,” seeks to establish an
objective basis for aesthetic standards by limiting what are aesthetically
relevant properties to scientifically explainable ones. However, for Kant,
the purpose of developing an account of aesthetic reflective judgment is
not to find a basis upon which any particular aesthetic reflective judgment
might be found to be absolutely correct for all time. If aesthetic response
referred to a judgment that lent itself to this kind of infallibility, we would
have stopped making art and aesthetic judgments in Kant’s sense. Culture
would have stagnated long ago.
In contrast, for Kant aesthetic reflective judgment involves seeking the basis
of standards within oneself: “in the judging of beauty in general we seek the
standard for it in ourselves a priori” (CJ 5:350). Kant is interested in the way
that this seeking for the standard brings our subjectivity under the constraints
of a comparative judgment and hence into community. This is interesting
because, just as he tries to vindicate reason by the processes in which we
engage rather than in terms of its objects (see discussions above referencing
O’Neill), he also treats aesthetic reflective judgment as vindicated in virtue of
the demands placed upon feeling by attempting to communicate those judg-
ments. In attempting to communicate our feeling, we in turn make our
feeling communicable. In other words, attempting to communicate feeling
inadvertently influences the aspects of feeling that we subsequently cultivate.
As such, Kant rejects the rationalist view, which bases aesthetic judgment
on a determinate concept. According to the rationalist view, aesthetic judg-
ment is objective, but given that one cannot state clearly what beauty is, this
suggests to the rationalist that aesthetic judgment is an example of a confused
(and inferior) sensory perception. If this were the case and the rationalist was
right, there really would be no point in arguing about taste.
27
Allen Carlson, Aesthetics and the Environment: The Appreciation of Nature, Art and Architecture
(London: Routledge, 2002), 24–26.
28
Ibid., 4.
The implication of Kant’s account of aesthetic reflective judgments is that

their basis is continually changing, developing, and perhaps progressing.
There are no proofs, formulas, or set criteria for what counts as an aesthe-
tically relevant property. Nonetheless, aesthetic reflective judgment exempli-
fies our capacity to establish a common sentiment through its dynamic of
demanding communicability or agreement.
When arguing that aesthetic reflective judgments are not merely sensations
(agreeable) nor merely a matter of measuring an object against the purpose it is
meant to serve (the good or perfection of design), Kant points out that if aesthetic
reflective judgment was restricted to these two alternatives, we could simply learn
from nature what is beautiful. There would be a fact of the matter regarding taste
and we would not need to seek standards within ourselves.
If beauty were a fact which did not lend itself to be clearly stated and
identified, it would hold little philosophical interest. However, for Kant,
aesthetic reflective judgment is different from both sensory perception and
judgments involved in ascertaining facts. He treats as a false dilemma that
aesthetic judgment is either subjective and confused or objective and rational.
It is in the context of this rejection of the rationalist view that Kant writes: “a
judgment of taste is not a cognitive judgment and beauty is not a quality of
the object considered for itself” (CJ 5:347). If it were a characteristic of the
object, it would be a cognition and, according to Kant’s nomenclature,
because it did not lend itself to explicit representation, a mere confused
cognition. Kant avoided this implication by construing aesthetic reflective
judgment as non-cognitive but not mere sensation. It was conceptual, albeit
indeterminate, hence subjective and rational:
That nature has the property of containing an occasion for us to perceive the
inner purposiveness in the relationship of our mental powers in the judging of
certain of its products, and indeed as something that has to be explained as
necessarily and universally valid on the basis of a supersensible ground, cannot
be an end of nature [a property of the object], or rather be judged by us as such
a thing: because otherwise the judgment that would thereby be determined
would be grounded in heteronomy [determined by objective conditions
outside us] and would not, as befits a judgment of taste, be free and grounded
in autonomy. (CJ 5:350)
Kant first mentions the non-cognitive nature of aesthetic judgment when

discussing how the object is received by us (a merely contemplative judg-
ment): “this contemplation itself is also not directed to concepts; for the
judgment of taste is not a cognitive judgment (neither a theoretical nor a
444 J. A. McMahon
practical one), and hence it is neither grounded on concepts nor aimed at

them” (CJ 5:209). In other words, aesthetic reflective judgment involves the
state of mind by which experience is something to be made sense of. It is a
capacity which raises us above the other animal species. It is an exercise of our
freedom. For Kant, beauty and aesthetic ideas are not objects of cognition
because he treats objects of cognition as given in experience rather than
created or generated by us in experience. Aesthetic ideas are generated by us,
ascribing meaning and significance to objects.
The significant point is that Kant does not characterize aesthetic reflective
judgments as non-cognitive in order to suggest that they are mindless,
idiosyncratic, or whimsical. Kant’s classification of aesthetic reflective judg-
ment as non-cognitive must be read in the context of his larger purposes and
the positions he opposes.
A positive account: Public sentiment to parallel

public reason
We have seen that Kant’s aesthetic theory presents a complex notion of
immediacy or directness, which expresses the idea of a subjective judgment
rather than delimiting the temporality of the judgment. Furthermore, the
aspect of subjectivity involved is susceptible to reason: practical reasons as
appropriate to aesthetic reflective judgment. To understand this, one needs
to consider Kant’s distinction between private and public reason. Unlike
private reason, public reason takes into account what others would judge; it
applies constraint to thought and feeling in a manner that would be trans-
parent to everyone. Only those considerations that would be communicable
to others are relevant to a pure aesthetic judgment.
This does not imply that our first take is the most apt or the only one. Nor
does it imply that our aesthetic reflective judgments can be revised based on
testimony alone. We can reconstrue, reconfigure, and re-perceive the object
based on metaphors, analogies, and prior examples, so as to experience the
object anew. Objectivity is certainly not reached when one’s judgment can
no longer be revised, as suggested by Greenberg and others. In fact, the
objectivity of judgment involves being open to revision. In short, an aesthetic
reflective judgment is direct or immediate only in the sense that it engages
one’s subjectivity, but one’s subjectivity at any particular moment in one’s
life has a history. It is the result of previous reasoning, reflection, knowledge,
experience, and training; and if it continues to operate as a genuine aesthetic
judgment in the Kantian sense, it will continue to develop its history. The
form of aesthetic ideas did not change for Kant – they were ideas not
exhausted by a concept – but the nature of the objects that evoked aesthetic
reflective judgment and the referents of their associated ideas might change
over time, and would change, as long as we remained living members of a
community of culture.
A standard formalist and hence misinterpretation of Kant’s aesthetic
theory has proven pervasive and persistent in much philosophy of art, but
nowhere more than in attempts to establish the grounds for objective
standards of aesthetic judgment, for which the misinterpretation of Kant’s
theory proves to be a useful foil. What many such accounts share is that
they remain somewhat at a loss to explain the point of our capacity for
aesthetic judgment. Without some sense of this, the basis of any objective
standard for aesthetic judgment will always be arbitrary. For example,
unless aesthetic judgments are concerned with accruing more information
about an object for some distinct purpose, why would scientifically
informed concepts of an object be the relevant supervenience base for
them? In contrast, for Kant the basis of a correct aesthetic response to
objects might be variable over time. The relevant constraint is that, while
we seek standards of taste within ourselves, this involves an awareness of
what others would judge, and it is this process that makes a community
of sentiment possible.29
Many commentators have approached Kant’s aesthetic theory by asking
why we should be interested in taste. The answer common to those who
draw from deep within Kant’s writing is that aesthetic reflective judgment
involves cultivating the conditions for our sociability and morality.30 Or we
could say that community develops with the development of reason, feeling,
taste, and morality. And taste is the conduit between reason and morality.
Kant puts it this way:
since taste is at bottom a faculty for the judging of the sensible rendering of
moral ideas (by means of a certain analogy of the reflection on both [these ideas
29
Jacques Rancière points out that Kant’s third Critique was written just after the French Revolution. As
such, the Critique is an attempt to answer the question “through what means can an equality of
sentiments be brought about that gives the proclaimed equality of rights the conditions of their real
exercise?” Kant’s answer is: through the formal universality of the judgment of taste. See Jacques
Rancière, The Philosopher and His Poor, trans. John Drury, Corinne Oster, and Andrew Parker, ed.
Andrew Parker (Durham: Duke University Press, 2004), 198. Arendt also discusses Kant’s intellectual
involvement in the French Revolution. See Arendt, Lectures on Kant’s Political Philosophy, 40–46.
30
See Watkins, “Subjective Basis of Kant’s Judgment of Taste.”
446 J. A. McMahon
and their renderings in sensibility]), from which, as well as from the greater
receptivity for the feeling resulting from the latter (which is called the moral
feeling) that is to be grounded upon it, is derived that pleasure which taste
declares to be valid for mankind in general, not merely for the private feeling of
each, it is evident that the true propaedeutic [preliminary learning] for the
grounding of taste is the development of moral ideas and the cultivation of the
moral feeling. (CJ 5:356)
An implicit conception of an ideal community provides the comparative edge

to the individual’s cognition, values, and judgment. This conception is not
expressed necessarily in shared beliefs but in modes of communicating. The
various elements that constrain the possibility of communication is nowhere
more evident than in aesthetic reflective judgment.31
31
This chapter was supported by an Australian Research Council Discovery Grant (DP150103143).
20
Sublimity and Joy: Kant on the Aesthetic
Constitution of Virtue
Melissa McBay Merritt
Introduction
One of the best-known passages in Kant’s entire corpus concerns the sub-
lime. It comes not from the aesthetic theory propounded in the 1790
Critique of the Power of Judgment, but from the concluding remarks of the
discussion of moral education in the 1788 Critique of Practical Reason1:
Two things fill the mind with ever new and increasing admiration
[Bewunderung] and reverence [Ehrfurcht], the more often and more steadily
one considers them: the starry heavens above me and the moral law within
me. . . . The first view of a countless multitude of worlds annihilates, as it
were, my importance as an animal creature, which after it has been for a
short time provided with vital force (one knows not how) must give back to
the planet (a mere speck in the universe) the matter from which it came. The
second, on the contrary, infinitely raises my worth as an intelligence by my
personality, in which the moral law reveals to me a life independent of animal-
ity and even of the whole sensible world, at least so far as this may be inferred
1
Hereafter the Critique of the Power of Judgment will be referred to as the “third Critique,” and the
Critique of Practical Reason as the “second Critique.”
M. M. Merritt (*)
School of Humanities & Languages, The University of New South Wales, Sydney,
Australia
e-mail: m.merritt@unsw.edu.au

448 M. M. Merritt
from the purposive determination [zweckmäßige Bestimmung] of my existence

through this law. (CPrR 5:161–62, translation modified)
The passage anticipates Kant’s mature view of the sublime in all of its
complexity: above all, it raises difficult questions about the relation
between the appreciation of the sublime in nature and the sublimity of
morality. These questions will occupy me in this chapter. The passage
also marks a departure from Kant’s earlier view in the 1763 Observations
on the Feeling of the Beautiful and Sublime, where he contends that
appreciating the sublime is a “tiring” affair that “cannot be enjoyed as
long” as the beautiful (OBS 2:211). By contrast, the “starry heavens”
passage points to an appreciation of the sublime that is not inherently
exhausting, but instead gathers strength the more that it is sustained.2
Let me begin by sketching the significance of this in order to set the
stage for this chapter.
We must first acknowledge a psychological point about the enjoyment of
the sublime that remains constant throughout Kant’s career, and that helps
to explain the early view about the exhausting effects of the sublime. We do
not enjoy the sublime through pure, unadulterated attraction; an element
of aversion belongs to it as well. The representation of the sublime in
nature, Kant contends in the third Critique, gives rise to a state of mind
that can be likened to a “vibration” since it consists of “a rapidly alternating
repulsion from and attraction to one and the same object” (CJ 5:258). Still
later, in the 1798 Anthropology from a Pragmatic Point of View, Kant
presents the sublime as a “reverence-arousing [ehrfurchterregende] greatness
(magnitudo reverenda)” that both “invites approach” and yet is “at the same
time a deterrent” (An 7:243, translation modified).3 It is not hard to see
why such a tussle back and forth might be tiring. But Kant also suggests
that the state of mind in appreciating the sublime is like a vibration
“especially in its inception [vornehmlich in ihrem Anfange]” (CJ 5:258),
hinting at the possibility of its settling into a frame of mind that is
governed, more stably, by the attraction. And when we linger as we look
out over the Grand Canyon, when we are held rapt by torrential rains from
2
I conclude with this observation in Melissa McBay Merritt, “The Moral Source of the Kantian
Sublime,” in The Sublime: From Antiquity to the Present, ed. Timothy M. Costelloe (Cambridge:
Cambridge University Press, 2012), 37–49. Thus I am to some extent picking up where I left off,
focusing now on its particular implications for Kant’s ethics of virtue per se.
3
I have modified Robert Louden’s translation here, both to follow the cue of Kant’s Latin gloss and to
show that Ehrfurcht occurs both here and in the “starry heavens” passage.
20 Sublimity and Joy 449
the safety of our porches, when we immerse ourselves in the pounding force
of ocean waves as we perch, more or less untouched, on the rim of a Sydney
rock pool, we are held by an attraction of some kind. The possibility of this
attraction depends upon a person’s commitment to moral ends: this, I will
argue, is the overarching thesis of the third Critique’s Analytic of the
Sublime. My first aim in this chapter is to draw attention to this line of
thought in the Analytic of the Sublime, and to elaborate on the conception
of human moral psychology that supports it.
Thus we need to understand some Kantian moral psychology – above all
concerning the cultivation of good moral character, or virtue – in order to
come fully to terms with his aesthetic theory of the sublime. At the same time,
we can draw on that aesthetic theory to address a standing puzzle about the
Kantian conception of moral virtue. In brief, that puzzle stems from the place
that virtue has in the duty-based, or deontological, ethics that Kant developed
in his critical-period works in moral philosophy, Groundwork of the Metaphysics
of Morals (1785) and the second Critique. There Kant denies the supereroga-
tory any place in ethics: anything good in human action can only ever be a
matter of duty. And duty is necessitation, implying the standing possibility of
being inclined to do otherwise (G 4:412–13). Kant accordingly conceives of
virtue as “moral disposition in conflict” (CPrR 5:84),4 and he rejects an ancient
conception of virtue as a perfect harmony of the soul whereby one unavoidably
acts well because one is no longer so much as subject to incentives to act
otherwise.5 This line of thought has led many commentators to suppose that
what figures as virtue for Kant is really something closer to what Aristotle
deemed mere “continence”: an ability to act well, but only by muscling
through countervailing urges to act otherwise. Friedrich Schiller lodged an
early version of this complaint in his essay “On Dignity and Grace” (1793);
Kant replied to Schiller in Religion within the Boundaries of Mere Reason
(1793), insisting that he had been misunderstood. The “aesthetic constitution”
of virtue is “joyous [fröhlich],” not dejected (Rel 6:24n); thus there should be no
confusing it with the distemper of mere continence. In the same vein, he later
invokes the “cheerfulness [Frohsinn]” that accompanies genuine virtue (MM
6:485, also 6:484). But how could these remarks collectively yield a coherent
4
Kant says little about virtue in the 1785 Groundwork, and offers just a few passing remarks about it in
the 1788 second Critique (like the one just quoted, which simply distinguishes human virtue from
holiness). He begins to develop an account of virtue in the 1793 Religion and elaborates at length in the
1797 Metaphysics of Morals.
5
Kant thinks that such a conception of virtue opens the door to special forms of moralistic Schwärmerei
and self-conceit (CPrR 5:84–86).
450 M. M. Merritt
conception of virtue? The clue, I will suggest, lies in the idea that the
appreciation of the sublime is a special type of contented attraction. My second
aim in this chapter, then, is to draw on the foregoing account of Kant’s
aesthetic theory of the sublime in order to show how he is plausibly entitled
to claim that true virtue is essentially joyful.
The aesthetic judgment of the sublime in nature

To understand the transition to Kant’s mature view of the sublime, we will
need to understand how he links the appreciation of sublimity in nature to
the sublimity of morality. One clue for this comes with his repeated denial
that any physical entity is in fact sublime: what we deem sublime in nature
stands in for a sublimity that can only be found in our own minds (CJ
5:245–46, 256, 257, 264). We speak of the sublime as that which is
“absolutely (not merely comparatively) great” (CJ 5:254), but no physical
entity can in fact be absolutely great. Whatever greatness of size or power that
may obtain is always only comparably so. The explosion of an atomic bomb
is comparable to – because measurable in terms of – the power of some
number of horses.6 (Quite a lot of horses, of course; but still the comparison
can be made.) So how can we speak of the sublimity of nature at all? Kant’s
answer, as we will see in more detail below, is that what we call sublime in
nature figures as absolutely great in our representation of it. This representa-
tion of absolute greatness echoes a sublimity, or absolute greatness, that can
only be found in our own minds.
It is tempting to fill out this last line of thought by pointing to Kant’s
views about the absolute – incomparable – worth we possess as persons
(G 4:434). And we would be right to connect the dots in that way. But
we would also risk oversimplification if we stopped there, without further
elaboration. Consider the German term for the sublime, das Erhabene: it
comes from the verb erheben, to raise or lift up. So, Kant is telling us
that what we call sublime in nature stands in for a sublimity – an
upliftedness – of our own minds. The pressing question is what exactly
this might mean. To unpack it, we will need to see how Kant’s aesthetic
theory of the sublime relies upon background claims about human
nature – claims Kant develops in his anthropological writings, and
6
Kant’s example for this same point concerns greatness of size, rather than power: the breadth of a
galaxy (CJ 5:250).
invokes passim in his ethical works. Kant accepts the classical definition
of the human being as rational animal. Yet he underscores that our
rational essence is neither an endowment received through birth nor a
development of subsequent physical growth: it is a perfective to be
realized in what we make of ourselves through what we do and how
we live. We are called to make ourselves fit for our own essential
rationality. Ultimately, for Kant, we do this by making ourselves fit for
moral ends, which just is to cultivate virtue in ourselves. A person’s
commitment to this calling – to lift oneself up, as it were – is precisely
the true sublimity that Kant contends can only be found in our own
minds. Thus Kant’s aesthetic theory of the sublime is tied particularly to
his ethics of virtue, and not simply to his more general claims about the
ontology of persons.
All of that is overview for a case that still needs to be made. We will
begin by looking into Kant’s account of the sublime in the third
Critique. Why are we attracted to natural sublimity? How does it please
us? Uncovering Kant’s answers to these questions will then lead us to the
conception of human nature that drives his account of the sublime from
behind the scenes.
Clues from the aesthetic judgment of reflection
Kant distinguishes logical and aesthetic judgments. The sublime in nature is

appreciated through a certain kind of aesthetic judgment. Locating that
judgment in Kant’s wider taxonomy of judgment will yield a preliminary
account of why and how the sublime pleases.
Logical judgments, for the most part, determine representations (whether
intuitions or concepts) under concepts, in order to make cognitive claims.7
Aesthetic judgments do not determine representations under concepts;
instead they relate given representations “to the subject and its feeling of
pleasure or displeasure” (CJ 5:203). Aesthetic judgments are expressed in the
register of feeling (pleasure or pain), which Kant distinguishes from sensation
(e.g., of color, sound, and so forth) (FI 20:224; CJ 5:206). While sensation
plays a legitimate role in empirical theoretical judgment – that is, in the
determination of how, in fact, things are – feeling does not. Aesthetic
judgments are, accordingly, non-cognitive (FI 20:224; CJ 5:203). Kant
7
I have just glossed “determining” logical judgments; there are also “reflecting” logical judgments, which
concern the objective purposiveness of nature (FI 20:211).
452 M. M. Merritt
then distinguishes aesthetic judgments of sense from aesthetic judgments of

reflection,8 depending on the basis of the pleasure, or liking, that is involved
(FI 20:224). Aesthetic judgments of sense express immediate liking for what
is agreeable, or disliking for what is disagreeable, owing to contingent facts
about the judging subject’s physical constitution. Aesthetic judgments of
reflection express immediate liking as well, but one that is not based upon
contingent facts about the judging subject’s physical constitution: for such
judgments make potentially legitimate claims to “universal validity and
necessity” (FI 20:225).9 What then is the basis of the pleasure that is
involved? The aesthetic judgment of reflection is expressed in a liking that
answers to the rational nature of the human being.
There are two kinds of aesthetic judgment of reflection: one concerning
the beautiful (the judgment of taste), and one concerning the sublime.10
Both modes of reflective aesthetic judgment are “singular” since they con-
cern a this – this beautiful rose, this sublime seascape – but at the same time
both are “judgments that profess to be universally valid in regard to every
subject, although they lay claim merely to the feeling of pleasure and not to
any cognition of the object” (CJ 5:244). The basis of these judgments’
universal validity, Kant elaborates, is that they express a liking for what is
regarded “to be in accord with the faculty of concepts of the understanding
or of reason, as promoting the latter” (CJ 5:244). Now, the particular
concepts that Jane, a nuclear physicist, has at her disposal may be somewhat
different from those that Jack, a saddlemaker, has at his. They have set
themselves different purposes, and their minds have been cultivated in
accordingly different ways. Thus when Kant says that the aesthetic judgment
of reflection involves a liking based on an appreciation of a thing’s accor-
dance with “the faculty of concepts,” he means what is necessarily and
8
This locution – “aesthetic judgment of reflection” – appears in full only in the First Introduction to the
Critique of the Power of Judgment (see FI 20:221, 224, 230, 231); in the published third Critique, Kant
calls them “aesthetic reflective judgments” (e.g., in the heading at CJ 5:266).
9
While aesthetic judgments of sense can be either pleasurable or painful, it appears that aesthetic
judgments of reflection can only be pleasurable. Commentators have disputed this, at least as regards
negative judgments of taste (i.e., concerning the ugly). I cannot weigh in on that debate here; I simply
note that it is hard to see what an aesthetic judgment of non-sublimity could possibly be. Perhaps it
would be a pained response to a representation of relative greatness. But that would be nothing other
than a contingent felt response to a theoretical determination of magnitude, which simply does not fit
the bill of an aesthetic judgment of reflection.
10
When Kant introduces the aesthetic judgment of reflection in the First Introduction, he elaborates by
making reference to the specifics of taste. However, Kant eventually makes fully explicit that the
aesthetic judgment concerning the sublime and the judgment of taste are alike species of the aesthetic
judgment of reflection (CJ 5:244).
universally constitutive of this faculty: for only then could he claim that such
judgments have universal subjective validity.11
Now we need to consider more carefully what Kant means by the
“faculty of concepts” in the above remark. The aesthetic judgment of
reflection is non-cognitive, since it does not determine a particular this
under any concept. And yet such judgments involve a relation to the faculty
of concepts, and hence involve a relation “to concepts, although it is
indeterminate which” concepts (CJ 5:244). I will return to this indetermi-
nacy, at least as regards judgments concerning the sublime, in the next
section. Our immediate task is to get to grips with what Kant means by the
“faculty of concepts” at all. For Kant, the intellect is a capacity for general
(as opposed to singular) representation: concepts are general representations
(A320/B377). Thus the intellect is, most basically, the faculty of concepts.
But Kant’s terminology for the intellect is confusing, since he speaks of
understanding and reason each in both wide and narrow senses. At times
Kant prefers to call the intellect as such “understanding” (see A131/B169;
FI 20:222; An 7:142, 196–97),12 while at other times he refers to it as
“reason” (see A835/B863). These are wide senses of understanding and
reason. In the remark at issue, Kant says that the aesthetic judgment of
reflection involves the “faculty of concepts,” where this faculty can be
further specified depending on whether concepts of understanding, or of
reason, are at issue. Concepts for determining phenomena are concepts of
the understanding. Concepts for determining what ought to be done are
concepts – ideas – of reason.13 Ideas are concepts to which no sensible
object can ever be adequate (A320/B376–77). The judgment of taste
involves a liking for what accords with the faculty of concepts specified as
the understanding, while the judgment concerning the sublime involves a
liking for what accords with the faculty of concepts specified as reason.14
11
But only judgments of taste – not judgments of the sublime – require a separate “deduction” to
establish their universal subjective validity (CJ 5:279–80). To elaborate on this would require an account
of Kant’s notion of a deduction and his handling of the issue for the judgment of taste, which lies well
beyond the scope of this chapter.
12
Kant regularly invokes the wide sense of “understanding” in contexts where he begins with the
distinction between singular and general representation (intuition and concept), which then aligns with
a distinction between sensibility and understanding – the latter as the “faculty of concepts,” meaning the
intellect as such. Consider A126; JL 9:36; VL 24:806, 846.
13
Other ideas of reason play a role in theoretical knowledge of nature, but they are not constitutive of
reason as a capacity for practical knowledge. I return to this below.
14
The beautiful object appears to be “the presentation of an indeterminate concept of the under-
standing,” whereas the sublime appears to be “that of a similar concept of reason” (CJ 5:244).
454 M. M. Merritt
The faculty of concepts, broadly construed, is the higher cognitive faculty.

Understanding and reason – in the narrow senses – are elements of the higher
cognitive faculty, each specified in terms of the concepts constitutive of its
cognitive exercise.15 Thus when Kant says that the aesthetic judgment of
reflection is a felt appreciation for what accords with the faculty of concepts, he
indicates that the pleasure draws from what is proper to us as creatures capable
of knowing.16 More often than not, Kant casts these claims in terms of the
“subjective purposiveness” of the aesthetic judgment of reflection.17 This
liking, we should underscore, does not follow upon a discursive recognition
that the beautiful or sublime object is good for one’s own capacities
(CJ 5:244). The liking for the beautiful rose, or the “sublime” seascape, is
itself the felt – non-discursive – appreciation of its amenability to proper
human mindedness.18
The specifics of the sublime
So far we know that any aesthetic judgment of reflection involves the felt
appreciation of what accords with the “faculty of concepts”; we also know
that in judgments concerning the sublime, this means reason. Concepts
are general representations; and reason’s concepts are ideas. No sensible
object can ever be adequate to an idea of reason: this, indeed,
15
The higher cognitive faculty is divided into understanding (in the narrow sense), reason (in the
narrow sense), and the power of judgment (A130–31/B169; An 7:196–97). In the third Critique, Kant
points out that the power of judgment has no objective principles of its own, whereas the understanding
has concepts of nature and reason has the idea of freedom articulated as a substantive principle through
the moral law. See FI 20:202, 208.
16
Kant indicates that the capacity for first-personal thought, and thus the faculty of concepts, most
fundamentally accounts for our status as persons rather than mere things (as he takes all other living
creatures to be) (An 7:127). He identifies the capacity for first-personal thought with understanding in
this passage – in the broad sense, as the faculty of concepts tout court.
17
The felt appreciation of purposiveness (i.e., of amenability to the faculty of concepts) just is the
pleasure involved in the aesthetic judgment of reflection (CJ 5:221; FI 20:228, 249 – and especially
250). On subjective purposiveness in the judgment concerning the sublime, see CJ 5:247, 268, 269; and
in the judgment of taste, see CJ 5:222, 227, 242.
18
Note also Kant’s distinction between the formal subjective purposiveness “contained” in the aesthetic
judgment of reflection versus the material subjective purposiveness “contained” in the aesthetic judg-
ment of sense (FI 20:224, 230). Some readers may be tempted to read this talk of “formal” purposive-
ness as tracking the fact that no determinate concept is involved in this assessment of purposiveness. But
Kant makes that point by noting that subjective purposiveness is recognized through pleasurable feeling
(see, e.g., CJ 5:247). Material subjective purposiveness is recognized by feeling that (again) has its basis
in one’s physical constitution, whereas formal subjective purposiveness is recognized by feeling that has
its basis in one’s “form” – that is, in the cognitive constitution one shares with any other rational animal
(FI 20:225; note also 215n regarding this sense of “form”).
distinguishes ideas of reason from concepts of the understanding. Now let

us turn to the specifics of Kant’s account of the sublime. Here we will
need to appreciate certain complexities about the sense in which it
involves a liking for what is amenable to reason in the narrow sense, as
the faculty of ideas. Kant distinguishes two modes of natural sublimity:
the representation of absolute greatness of size (mathematical sublime),
and the representation of absolute greatness of power (dynamical sub-
lime). This division, as we will see, tracks different sorts of ideas of reason.
We should not assume from Kant’s systematic tidiness, however, that the
two categories of the sublime are mutually exclusive. The great surf at
Nazaré, a vast stretch of inhospitable terrain, even the starry heavens
themselves – all can surely occasion sublime feeling inflected in both
ways at once.
The mathematical sublime is a representation of absolute greatness of
magnitude. Since no physical entity is in fact absolutely great, what counts
is how an entity is taken in by the senses – how it is “apprehended” in the
imagination, as Kant says in this context (FI 20:220, 223, 233; CJ 5:190,
217, 232, 244, 251–52) – so that it figures as absolutely great. We must not
stand so close to the huge entity that we take in only a mass that spreads on
all sides, consuming our focal and peripheral vision together; and we must
not stand so far off that the boundaries of the thing easily register in a single
glance. We need to stand at just the point where we will approach having a
comprehensive view of the thing, but always just fail (CJ 5:252). It is
precisely this struggle, and failure, to attain a comprehensive view of the
thing that allows for it to figure, in our representation of it, as absolutely
great: for if we had a comprehensive view of the thing, we could proceed to
compare it in size against something else. This failure of the imagination,
Kant contends, “awakens the feeling of a supersensible faculty in us”
(CJ 5:250) – namely, the power of reason to conceive, in pure thought,
of what can never be given to us through the senses.
The rational idea at issue in the mathematical sublime is that of an
infinite, given totality. Kant implicitly connects the mathematical sublime
with the First Antinomy discussed in the Critique of Pure Reason (CJ
5:254–55), which concerns the cosmological representation of the world
as a whole – whether it is bounded, or infinite, in space and time (A426–
33/B454–61). But the cosmological representation of the world as a whole
is a mere idea of reason to which no sensible experience can ever be
adequate; so it ought not serve as a basis for making determinations
about the real magnitude of the world (A517–23/B545–51). Thus the
First Antinomy exposes the limits of reason in theoretical cognition; it
456 M. M. Merritt
exposes reason’s incapacity in this regard. But the pleasures we take in

natural sublimity turn on a felt appreciation of amenability to the rational
capacity. Hence the idea of an infinite, given totality must play some
legitimate role in theoretical cognition. And indeed, Kant reminds us
here of our need “to think nature itself in its totality” (CJ 5:268) – to
think of it as “the sum total of appearances,” as we must do in order to
cognize it as a law-governed whole (see Pro 4:318). This is why Kant links
the failure of imagination in the mathematical sublime to reason as “the
faculty of cognition” – i.e., to reason in its role in theoretical cognition of
nature (CJ 5:247).
The dynamical sublime involves a representation of absolute greatness
of power. Kant is less clear here about how the representation of the
dynamically sublime should involve a struggle and failure of the imagi-
nation. It will involve taking in sensible particulars (e.g., booming
sounds, sensations of brute impact, visual cues of swiftness); so in that
regard it must involve the imagination as the capacity for sensible
representation. These sensible representations will moreover need to
contribute to a representation of the physical force at issue as absolutely
great. Thus I will need to have a sense of a power that could overwhelm
and crush me. The modality of this is important: I cannot recognize an
actual threat on my life – if I did, I would not be inclined to stick
around to soak up the attendant sensations (CJ 5:261). Yet I need to be
fully immersed in these sensations, so that the power figures as immea-
surable and unmanageable (CJ 5:269). Normally, my apprehension of
particulars informs the grip I have on my bearings in the world and my
ability to move about as I choose within it. Now, Kant says that the
dynamical sublime relates the failure of the imagination to “the faculty
of desire” (CJ 5:247), which he elsewhere glosses as “a being’s faculty to
be by means of its representations the cause of the reality of the objects of
these representations” (CPrR 5:9n). Thus the dynamical sublime gives one
a feeling of the power of rational agency. But rational agency can be
determined either by natural instinct and sensible inclination, or by “laws
of reason independently of natural instincts” (G 4:459). The representa-
tion of the dynamical sublime assaults the former and promotes the
latter. But why, exactly, should this be pleasurable? For the answer, we
need to turn to Kant’s general account of the pleasure involved in our
appreciation of natural sublimity.
When Kant introduces the sublime, he claims that the pleasure we take in
sublimity is “very different in kind [der Art nach . . . sehr unterscheiden]” from
the pleasure we take in beauty, which “directly brings with it a feeling of the
promotion of life” (CJ 5:244, emphasis added). The feeling of the sublime,
by contrast,
is a pleasure that arises only indirectly, being generated, namely, by the feeling
of a momentary inhibition of the vital powers and the immediately following
and all the more powerful outpouring of them; hence . . . the satisfaction in the
sublime does not so much contain positive pleasure as it does admiration
[Bewunderung] or respect [Achtung]. (CJ 5:245, emphasis added)
The contrast that Kant means to draw is not that the appreciation of beauty
involves a “feeling of the promotion of life” while the appreciation of
sublimity somehow does not. For both types of the aesthetic judgment of
reflection involve a liking for what is amenable to our cognitive constitution,
and so as promoting the cognitive life of the mind. Rather it is that the
appreciation of beauty engages this feeling directly, the appreciation of
sublimity indirectly.19 There are, moreover, two conceptions of life or vitality
at work in the remark just quoted. The representation of natural sublimity
assaults vital powers of one order, while it quickens vital powers of another.
With this in mind return again to the famous “starry heavens” passage.
The incomparable vastness of the night sky “annihilates my importance as an
animal creature”: this particle of flesh is nothing compared to all of that; it is
imbued with “vital force” for just a moment; it will return to the dust from
which it came. These are dark thoughts, but we find ourselves drawn to the
vistas that occasion them. And finding ourselves so drawn, we recognize in
ourselves “a life independent of animality and even of the whole sensible
world” – a vitality proper to our status as persons, not things (CPrR 5:162).
The incomparable vastness that checks the feeling of life of the one kind
promotes the feeling of life of the other kind.20
The large or mighty entity in nature would simply repel us, Kant elabo-
rates, were our minds not cultivated along certain lines:
what is properly sublime cannot be contained in any sensible form. . . . Thus

the wide ocean, enraged by storms, cannot be called sublime. Its visage is
horrible; and one must already have filled the mind with all sorts of ideas if by
means of such an intuition it is to be put in the mood for a feeling which is
19
Cf. Reflexion 565 (c.1776–78): “The whole feeling of life is a joyfulness. . . . Liberation following from
a pain is a positive movement of the organ to life” (Ak 15:245).
20
See CPrR 5:89 for the same dual invocation of “life”; cf. CPrR 5:88.
458 M. M. Merritt
itself sublime, in that the mind is incited to abandon sensibility and to occupy
itself with ideas that contain a higher purposiveness. (CJ 5:245–46)
The ideas at issue are specifically moral ideas – namely, those needed to conceive
of oneself in terms of a “higher purposiveness,” of having an end apart from
sustaining oneself in animal life. Accounting again for the indirect pleasures of
the sublime, Kant remarks that “the very same violence that is inflicted on the
subject by the imagination is judged as purposive for the whole vocation
[Bestimmung] of the mind” (CJ 5:259). What I dislike owing to its lack of
amenability to my physical existence I nevertheless enjoy owing to its amen-
ability to the complete picture of what I am called to be.
The term Bestimmung in this context has a special sense. In its most basic
usage, the term Bestimmung means “determination” – what makes something
what it is, or gives it a certain character. But the term takes on a teleological
sense where the Bestimmung of the human mind is at issue, and is therefore
typically rendered “vocation” in English translation.21 The human being is
the rational animal; rationality is the distinguishing difference – the form – of
our kind. But this rational essence is not itself a given endowment; it is
something to be realized in what we make of ourselves through what we do
and how we live. In the anthropological writings, Kant is concerned with the
Bestimmung of the entire species, and he claims that it consists in human
“perfection” (An 7:322; CB 8:115; LP 9:445; see also LE 27:470–71). No
individual can fulfill this Bestimmung (An 7:324; LP 9:445): and that is not
only because the perfection of our rational nature is an ideal toward which we
can only strive, but also (and perhaps chiefly) because we are collectively
responsible for establishing the practices of communication, education,
justice, and so on that stand to make the realization of our rational nature
possible at all.
That said, self-perfection plays a central role in Kant’s ethics of virtue.
In the Metaphysics of Morals, Kant argues that virtue can be cultivated
only with the free adoption of two morally obligatory ends: (a) one’s
own perfection and (b) the happiness of others (MM 6:385). On this
21
Kant alludes to a public debate between Thomas Abbt and Moses Mendelssohn, who were themselves
responding to the considerable popular influence of Johann Joachim Spalding’s 1748 Betrachtung über
die Bestimmung des Menschen. See Reinhard Brandt, “The Guiding Idea of Kant’s Anthropology and the
Vocation of the Human Being,” in Essays on Kant’s Anthropology, ed. Brian Jacobs and Patrick Kain
(Cambridge: Cambridge University Press, 2003), 85–104; Allen W. Wood, “Kant and the Problem of
Human Nature,” in Essays on Kant’s Anthropology, 38–59; and Manfred Kuehn, “Reason as a Species
Characteristic,” in Kant’s “Idea for a Universal History with a Cosmopolitan Aim”: A Critical Guide, ed.
Amélie Oksenberg Rorty and James Schmidt (Cambridge: Cambridge University Press, 2009), 68–93.
basis he divides his account of virtue into what one owes oneself and
what one owes others. Immediately relevant for us here is the end of self-
perfection. It calls for “cultivating one’s faculties (or natural predisposi-
tions), the highest of which is understanding, the faculty of concepts”;
and since this faculty (here broadly construed) includes “concepts that
have to do with duty,” the end of self-perfection therefore “at the same
time” also calls for the cultivation of “one’s will (moral cast of mind)”
(MM 6:386–87). Our Bestimmung, most fundamentally, is to make
ourselves fit for our own essential rationality; this centrally requires
cultivating our cognitive capacities, which for Kant includes the will,
toward the standard of virtue that is thought through the moral law.
Kant contends that a background commitment to moral ends is required
in order to be able to enjoy natural sublimity at all. Without that background
commitment, one would simply feel assaulted, terrified, or (at the very least)
painfully disoriented. One has to be able to have the thought that there is
more to one than one’s physical (animal) existence. To have such thoughts
requires a certain facility with moral ideas. But since the judgment is
aesthetic, it cannot draw on any determinate thought of the good, on any
particular moral ideas.22 Kant walks a fine line here. We might recall the
ambiguous rider in his general gloss of the aesthetic judgment of reflection,
applying it now to the specifics of the sublime: the judgment is the felt
appreciation (the liking) for what accords with our faculty of concepts (here:
of moral ideas) “as promoting” it (CJ 5:244). Natural sublimity attracts us in
the particular way that it does because of (and perhaps to the extent that) we
are attuned to the “whole vocation of the mind” (CJ 5:259). What counts as
sublime must not simply stretch the imagination in ways that make palpable
the power of reason; the movement must indeed “produce a disposition of
the mind” that is “compatible with that which the influence of determinate
(practical) ideas on feeling would produce” (CJ 5:256).23 Kant stops just
short of claiming that the enjoyment of natural sublimity promotes the
cultivation of the moral disposition. He does, however, effectively say that
the enjoyment of natural sublimity must harmonize with this disposition in
some cognitively indeterminate register of feeling.
We can return now to our starting point, that nothing in nature is, strictly
speaking, sublime: “true sublimity must be sought only in the mind of the
22
The aesthetic judgment of the sublime is, after all, not a logical judgment.
23
Or later: “In fact a feeling for the sublime in nature cannot even be conceived without connecting it to
a disposition of the mind that is similar to the moral disposition” (CJ 5:268, emphasis added).
460 M. M. Merritt
one who judges, not in the object in nature” (CJ 5:256). Nature is called
sublime, Kant elaborates, only when it occasions a movement of the mind
that makes “palpable [fühlbar] to itself the sublimity of its own vocation [die
eigene Erhabenheit seiner Bestimmung] even over nature” (CJ 5:262; see also
264, 268). But what exactly does this mean, to say that the Bestimmung of
the human mind is itself sublime? At the very least it must mean that we are
called to raise ourselves up from a default condition to perfect the rational
nature that is proper to us.24 However, if we left it at that we would overlook
one of the chief lessons of the aesthetic theory of the sublime: namely, that
sublimity is appreciated through feeling.
The emotional register of the sublime

We have considered Kant’s account of the indirect pleasures of the sublime: a
representation of absolute greatness first assaults “the vital powers” of our
animal being, before we revel in it for its accordance with what we are called
to make of ourselves as essentially rational beings. Thus Kant contends that
we enjoy the sublime not through straight pleasure, but through “admiration
or respect” (CJ 5:245; quoted at greater length above). Although Kant gives
some impression here of taking admiration and respect to be interchangeable,
in fact he clearly distinguishes between the two. This distinction is muted in
the third Critique, where Kant says (for example) that we appreciate natural
sublimity with “a kind of respect [eine Art von Achtung]” (CJ 5:249), and
later elaborates that “the feeling of the sublime in nature is respect for our
own vocation” (CJ 5:257, emphasis added). But this is not the copula of
strict identification. The usage is something more like the copula in “Frank
Oz is Yoda.” For the respect we feel for the large or mighty entity in nature is
not literally the respect we feel for our own vocation as rational beings; the
former stands in for the latter, and enacts it in a morally indeterminate way.
Strictly speaking, we admire the sublimity of nature and respect the sub-
limity of morality. There are important clues to this usage in the second
Critique, where Kant underscores that respect “is always directed only to
persons, never to things,” and therefore should not be confused with the
“admiration [that] can be directed to things also, for example, lofty moun-
tains; the magnitude, number, and distance of the heavenly bodies, the
24
In Religion Part I, he links this default condition to the fallen state of humankind (i.e., original sin);
see, e.g., Rel 6:38–43.
strength and swiftness of many animals, and so forth” (CPrR 5:76). Kant
then provides an extensive account of respect for the moral law: the moral law
is of course not itself a person; but it is the principle of our personality. It is
owing to the fact that practical reason is constituted by this law, Kant thinks,
that we are imputable agents, and thus are persons rather than things (as
Kant supposes the rest of the animal kingdom to be) (CPrR 5:95–98; MM
6:223). In this context, Kant argues that the basic capacity to feel respect for
the moral law is proper to us as embodied rational beings.25 A quick sketch of
this argument is needed before we can track Kant’s distinction between
admiration and respect.
The argument that concerns us here itself assumes the core result of the
first chapter of the second Critique: that the moral law is constitutive of
reason in its practical capacity (CPrR 5:19–57). This (quite controversial)
claim entails that a rational being must have some grasp of what morality
requires – no matter how obscure – just as soon as she is able to think about
what she has reason to do at all. Kant’s argument establishing that the moral
law is constitutive of practical reason abstracts entirely from the fact that
reason is embodied in the rational animal. A purely rational creature (a
disembodied “holy will”) does not have any physical needs or desires, and
thus is not subject to any sensible incentives for action; it can act on no other
incentive than that of pure reason alone, and hence for it morality could not
be a matter of duty. But a human being does have physical needs and desires,
is subject to sensible incentives for action; our practical thought is guided, by
default, by principles of self-love; thus we are subject not to the moral law full
stop, but to the moral law as a categorical imperative (CPrR 5:32; see also
G 4:414). So, while we necessarily have some grasp (no matter how obscure)
of moral requirement just as soon as we are able to think about what we have
reason to do, at the same time we are also always subject to nonmoral
incentives on action and are liable to answer their call even when doing so
conflicts with duty.
From these premises, Kant argues (in the third chapter of the Critique of
Practical Reason) that the recognition of moral requirement necessarily has a
certain effect on us: the feeling of respect. Respect is a feeling of the sublimity
of the moral law. The moral law checks the claims of self-love, which are in
25
Later Kant underscores that there can be no duty to have this capacity, only to cultivate it (MM
6:400). Thus when Kant says that the appreciation of natural sublimity both “requires culture” and yet
“has its foundation in human nature . . . , namely in the predisposition to the feeling for (practical) ideas,
i.e., to that which is moral” (CJ 5:265), he clearly means that it depends upon the cultivation of moral
feeling that is itself central to the development of virtue.
462 M. M. Merritt
themselves perfectly reasonable (CPrR 5:73): it is morally permissible to answer

the claims of self-love, as long as this does not conflict with moral requirement.
But the moral law humiliates the claims of self-conceit (CPrR 5:73, 74, 75, 78).
Self-conceit is the condition in which self-love is taken to be unconditionally
legislative on action, so that the claims of morality are subordinated to it (rather
than the other way around), and one takes one’s own happiness to be the
objective good and proper end of anyone’s action (CPrR 5:74).26 This humilia-
tion is the painful aspect of respect, in which we are repelled by the claims that
morality makes on us. But inasmuch as we recognize that the law comes from
our own reason and therefore is not imposed upon us from without, we are
thereby also impressed by its command. A gloss on the sublime from the third
Critique is perfectly apt here: “That is sublime which pleases immediately
through its resistance to the interest of the senses” (CJ 5:267). Respect is the
felt appreciation of the sublimity of morality; it is how we revel in our own
power of self-determination as rational beings through the moral law, despite
the possibility of its commanding us counter to the claims of self-love. This is
how Kant comes to speak of the sublimity of duty,27 of certain maxims,28 and
(again) of the human Bestimmung itself.29 It is because the moral law is
constitutive of practical reason that we are disposed to be pleased in this way;
but the stability of this pleasure, and our readiness to feel it, will depend upon
the ends we set ourselves.
This last point is crucial to distinguishing respect (Achtung, Ehrfurcht30)
from admiration (Bewunderung). To do this, we need to return to the idea
that the moral law makes substantive claims about what we ought to care
about: a person ought to care about perfecting herself and promoting the
26
Kant does not take self-conceit to be unusual or monstrous depravity: for if it is essentially a matter of
prioritizing the claims of self-love over those of morality, thereby taking the former to be uncondition-
ally legislative, then it is nothing other than the default human condition after the fall, as sketched in
Religion, Part I (6:3–53).
27
“Duty! Sublime and mighty name that embraces nothing charming . . . and yet does not seek to move
the will by threatening anything that would arouse natural aversion or terror . . . but only holds forth a
law . . . before which all inclinations are dumb” – and so on (CPrR 5:86).
28
“It is just in this independence of maxims from all such [sensible] incentives that their sublimity
consists” (G 4:439).
29
“This idea of personality, awakening respect by setting before our eyes the sublimity of our nature (in
its vocation)” (CPrR 5:87); “The majesty of the law . . . rouses a feeling of the sublimity of our own
vocation” (Rel 6:23n).
30
Kant uses Ehrfurcht more or less interchangeably with Achtung in moral contexts (e.g., at Rel 6:23n,
where Achtung is apposite to Ehrfurcht – though this is obscured in the Cambridge translation of this
passage, where both terms are rendered “awe”). Kant occasionally uses the Latinate term Respect (e.g., at
Rel 6:23n), which he elsewhere glosses as “respect coupled with fear [mit Furcht verbundener Achtung]”
(MM 6:438).
happiness of others. These are the morally obligatory ends around which
Kant organizes his account of virtue, as noted above. My own reason requires
me to adopt these ends, and yet I might still choose not to adopt them as
morality requires; I might in fact care about my own happiness above all else.
But they are not discretionary ends that I am at leave to adopt (or not), as I
am at leave to adopt (or not) the end of being a scientist or a gardener.
Suppose that I myself do not care about gardening: people give me plants,
the plants die, and I shrug. Suppose, too, that my friend Julia has an
exuberant garden; she can make anything grow with luscious abandon. I
can admire Julia for her gardening, if it strikes me as something great. But she
sets no example for me unless I am resolved, and somewhat prepared, to do
better with my wan seedlings. Only then could I respect her (for her garden-
ing). Misplaced respect, moreover, is a sign of delusion. I should be mocked
if I said that I respect Stephen Hawking for his scientific acumen, since it
suggests that I seriously share his end of unlocking the secrets of the universe
– and I am not so much as equipped to set myself this end, since I am clueless
about what it would require of me. But I can perfectly well admire him for
his particular intellectual powers. Respect is exhortative: it calls me to certain
courses of action. Admiration does not.
The point that I have just made concerns the role that ends play in
distinguishing respect from admiration. A further point concerns the role that
morally obligatory ends play in distinguishing respect for a person’s good
character from respect for discretionary skills and talents. Where we are one
and all called to the same essentially human ends, it becomes possible to respect
someone as a good person full stop – as when Kant speaks of the respect that is
wrung from him by “a humble common man in whom I perceive uprightness
of character in a higher degree than I am aware of in myself” (CPrR 5:76–77).31
Since admiration can be felt either for persons or things, it is presumably the
register in which we stand to appreciate natural sublimity. At the same time, my
attraction for the sublime vista is not unrelated to practical thought, to my
conception of myself as a rational agent – although it does not directly engage
this agency. I am not exhorted by it to do anything. Thus it is appropriate to
characterise the feeling involved as admiration rather than respect in the strict
sense.32
31
See Kant’s more tangled account of the two points at issue (in this and the previous paragraph) in his
example of Voltaire and his “common” admirers, at CPrR 5:78.
32
In the third Critique, Kant also speaks of our admiration for natural beauty (CJ 5:301, 482n). But
these remarks are puzzling in light of Kant’s comment that we enjoy beauty through pure unadulterated
liking, whereas we enjoy sublimity indirectly, in a pleasurable attraction that goes by way of painful
464 M. M. Merritt
The temperament of virtue

At the outset of this chapter, I pointed to a curious transition in Kant’s
thinking about the sublime. While Kant initially thought of the sublime as
something exhausting, in the critical period he comes to suppose that sublimity
is enjoyed in a way that gains strength the more that it is sustained. Quite
obviously, however, we are not awestruck in perpetuity by sublime vistas in
nature. Sure, I may linger as I look out over the precipice of the Grand Canyon
– the awe is sustained – but still it is an episode that comes to an end. And
likewise, I may feel respect for an exemplary person’s goodness, but this too is
an episode that comes to an end. Sublime feeling is episodic (as feelings tend to
be). Moreover, the feelings of admiration and respect noted here are ways of
being struck by something that stands out from the ordinary run of things.
They may punctuate normal lived experience from time to time, but they are
not constantly online. What then are we to make of Kant’s hints about
sublimity as an abiding temperament? Humility, I suggest, fits the bill.
Humility is an abiding temperament or attitude, rather than a passing
feeling. Humility is also “a sublime state of mind, that of voluntarily sub-
jecting oneself to the pain of self-reproach in order gradually to eliminate the
causes of it” (CJ 5:264). It is a sublime state of mind because it involves an
openness to the pains of self-reproach, with a contented attraction to the
standard against which one compares oneself. In the Vigilantius lectures on
ethics, Kant invokes the same notion of humility, evidently correcting his
students’ liability to associate it only with Christian ethics, noting that it is
aversion – and therefore enjoy sublimity through admiration and respect (CJ 5:245). Moreover, while
Kant elaborates at length on why respect should be a feeling of this sort, he never explains how
admiration fits that bill. But Kant also mentions how we tend to admire nature at the discovery of
new order within it – namely, the “unifiability of two or more empirically heterogeneous laws of nature
under a principle that comprehends them both” (CJ 5:187). Here we are admiring what we have
presumably first struggled to comprehend. Thus an example like this accords with the point he wants to
make about admiration at CJ 5:245. But how, in Kant’s terms, can we admire the beauties of nature?
A dash of common sense might help here: we admire what stands out from the common run of
things. This begins to account for how we could admire such diverse things as a beautiful flower, an
impressive surf, and a friend’s gardening skills. Further, Kant takes admiration (Bewunderung) to be a
kind of astonishment (Verwunderung) or surprise (Erstaunen). Astonishment and surprise are assaulting:
it is how we respond when we are presented with something enough out of the ordinary that, at least for
a second, we do not quite believe it. There is something unpleasant about such a shock, inasmuch as it
rattles one’s grip on the order of things (though this admittedly sounds overblown for something like
admiring a friend’s gardening skills). At any rate, admiration is a type of astonishment since it involves
this rupture in one’s sense of the order of things, but where we continue to marvel even after this shock is
gone (CJ 5:365; see also An 7:243, 261). We continue to marvel because there is something pleasing
about how the thing stands out from the normal run of things. It lies beyond the scope of my work in
this chapter to elaborate any further on these issues.
found in the “humilitas animi of the Stoic” as well: “the sublimity of mind
whereby, conscious of his inner strength, he took his humanity calmly [sich
über seine Menschheit beruhigte], and determined himself in dutiful obedience
to the law, in that he unceasingly endeavoured in his actions to approximate
to its holiness” (LE 27:610).33 What is Kant saying here? The Stoic accepts
his humanity: he accepts the problem of being human, the calling to make
ourselves adequate to our own essential rationality. And yet, “conscious of his
inner strength,” has some confidence in his ability to heed it: to hear, and be
moved by, the call. This is not a downtrodden, self-denigrating humility; it is
not the “Carthusian” frame of mind, “weighed down by fear and dejected,”
that Kant dismisses in his reply to Schiller (Rel 6:23–24n).34
Humility is, at least intuitively, an abiding temperament, a general attitude
of mind. Humility brings with it a readiness to reconsider how things strike
one in the situations in which one finds oneself: to reconsider what imme-
diately seems to matter most of all, and what one may have neglected –
perhaps not even noticed – as not mattering much at all. It checks blind,
unreflective, confidence in oneself. It is a self-directed attitude, a way of being
on the lookout for the pretensions of self-love, always on the make to assert
its claims as “primary and originally valid,” as Kant puts it in the second
Critique (CPrR 5:74). It is a temperament because it brings with it an
openness to the pains of self-reproach. And inasmuch as it is a manifestation
of a person’s moral commitment, it is a stable and abiding temperament – at
least to the extent that this commitment is itself stable, and strong.
True humility is also not blind, unreflective denigration of oneself. That is
the “false humility” that Kant decries as engendering moral passivity, an
33
This last point – that the standard at issue is one of holiness rather than virtue – is one of Kant’s central
criticisms of Stoic ethics; see CPrR 5:127n.
34
See Jeanine Grenberg, Kant and the Ethics of Humility: A Story of Dependence, Corruption and Virtue
(Cambridge: Cambridge University Press, 2005) for an account of the importance of humility to Kant’s
ethics of virtue – an overarching thesis I support, though I would contest many of the details of her
account. However, Robert Louden (“Kantian Moral Humility: Between Aristotle and Paul,” Philosophy
and Phenomenological Research 75, no. 3 [Nov. 2007]: 632–39) doubts the overarching thesis, pointing
out that Kant more often refers to moral virtue as a kind of “courage” and only rarely explicitly links it to
humility. That may be right, but it does not settle much if humility plays a deeper conceptual role in
Kant’s thinking about virtue. Moreover, I think that Louden errs on some points of interpretation that
are relevant here. He bases his account of Kantian moral humility on the idea that the moral law
invariably “humiliates every human being when he compares with it the sensible propensity of his
nature” (CPrR 5:74). But humiliation by the moral law is not the same thing as humility. Humiliation is
painful, and we naturally respond to pain with aversion. When Kant says that moral humility is a
“sublime state of mind,” he indicates that it is a temperament more stably governed by an attraction to
the standard of goodness thought through the moral law. As such, it should already include the positive
feeling of moral self-esteem that Louden points to as evidence against Grenberg’s claims about the
importance of humility in Kant’s ethics (“Kantian Moral Humility,” 637).
466 M. M. Merritt
unwillingness to make oneself adequate to moral ends (CJ 5:273).35 Thus true
humility is an attitude of willingness to make oneself adequate to moral ends.
To will an end (as opposed to idly wish for it) is to muster all means within
one’s control to bring it about (G 4:394). And where the end is self-perfection,
the means are ultimately capacities of judgment and dispositions to be moti-
vated in certain ways. To muster these resources is to cultivate them, to bring
them to the ready. We do not, after all, acquire virtue by incantation, simply
pronouncing our commitment to moral ends. We commit ourselves through
what we do and how we do it – in the attitude of our willing, and in the
principles of our action. We commit ourselves (or not) through what we take as
a reason for doing what, situation by situation, over and over again.
Humility is the temperament of virtue, we might then conclude. And,
having taken account of Kant’s view that humility is a sublime state of mind,
we should see that it must be governed by a pleasurable attraction of some
kind – even as it involves painful feelings of self-reproach. But these conclu-
sions, nice as they are, still leave us at something of a loss to understand how,
really, the virtuous temperament should be joyous. To address this final
puzzle, we need to elaborate a bit more on Kant’s conception of virtue.
As we have seen, Kant thinks that virtue can only be cultivated with the
free adoption of morally obligatory ends. Relatedly, Kant makes something
of (what he takes to be) the etymology of Tugend, the German word for
“virtue,” from taugen, “to be fit for” (MM 6:390). This move has roots in
classical (but non-Aristotelian) ideas about virtue as a certain sort of skill.
And in the Metaphysics of Morals Kant explicitly allows that virtue might be a
certain skill, Fertigkeit (MM 6:383–84) – though in a certain qualified sense:
it is both like, and yet in certain respects unlike, normal skills. The ins-and-
outs of this is a topic unto itself.36 But we can say, quite generally, that a skill
is an acquired fitness for a certain end. For the most part, skills are acquired
fitnesses for discretionary ends; but virtue, Kant indicates, is fitness for morally
obligatory ends. These ideas surface in the Vigilantius lectures on ethics as
well, where we find an intriguing clue to the puzzle about the joyfulness of
virtue. This remark occurs, moreover, as Kant is in the process of hashing out
his debate with Schiller about the temperament of virtue:
35
See the related passage at Rel 6:159–60, and note Kant’s citation of Matthew 6:16.
36
I discuss Kant’s qualified endorsement of the skill model of virtue in Melissa McBay Merritt, “Kant on
Virtue as a Skill,” in Natur und Freiheit: Akten des XII. Internationalen Kant-Kongresses, ed. Violetta L.
Waibel and Margit Ruffing (Berlin: de Gruyter, forthcoming); and I elaborate on the issue at greater length
in my book manuscript, Kant on Reflection and Virtue (Cambridge University Press, forthcoming).
It is true that one can find enjoyment in virtue and in the contemplation of it,
but only when and for the reason that one has become skilled to carry out one’s
duties,37 so that it becomes easy to follow the prescriptions of reason; from this
one attains a contentment about one’s actions and about the strengthening of
one’s will for the prescriptions of reason; and one regards the future with
cheerful courage. (LE 27:490–91, translation modified)
Virtue is enjoyable because – in the ideal – one has fully cultivated the
required resources. One readily appreciates what does and does not matter,
morally speaking, in the situations in which one finds, and places, oneself;
and one is fully impressed by one’s recognition of what morality requires,
and hence readily disposed to act accordingly. Without humility, this skill
would be unreflective and falsely confident. Thus Kant rejects any idea that
to attain virtue is to transcend human nature. This is where the Stoics erred,
Kant contends, despite correctly appreciating that the path to virtue lies in
humilitas animi (LE 27:609–10; see also CPrR 5:86, 127–28n).
Perhaps, for Kant, the virtuous person is something like the bird whose
joyful song announces a contentment with his own existence (CJ 5:302). For
the virtuous person, too, must be content with his own existence, but not
because he is sure of his own good character. His existence as a rational being
consists of a calling to make himself adequate to an ideal that can only be
conceived in pure thought, through the moral law. Though this calling can
never be definitively answered, it may be possible to become attuned to it in a
way that reverberates through one’s entire frame of mind, giving shape to an
abiding temperament. That is why virtue is not only a sublime state of mind,
but also – just plausibly – a joyful one, too.38
37
The phrase is insofern es uns schon zur Fertigkeit geworden, Pflicht zu erfüllen. The LE translation
renders this “being equipped to fulfil duties” – which is fine, except that it obscures the notion of skill,
and the ancient debate about virtue as a skill, that is tacitly at issue here. I have modified the Cambridge
translation of this passage also to reflect the fact that the German is mostly phrased in the third person
singular (one can find enjoyment in virtue, etc.), and only uses the first person plural once, in the phrase
quoted above (an indirect usage, moreover).
38
Research for this chapter was supported by a grant from the Australian Research Council
(DP130100172).
Part VII
Philosophy of Science
21
“Proper Science” and Empirical Laws:
Kant’s Sense of Science in the Critical
Philosophy
John H. Zammito
For literally millennia, the ideal of knowledge of the physical world was
captured in the concept of scientia, the claim to timeless, universal, and
necessary truth. Natural philosophy, from at least Aristotle, laid claim to
such knowledge of the physical world, at least as a cognitive ideal.1 It was a
crucial talisman of the elevation of philosophical truth above all mere
contentions – the claim to episteme, not doxa – rendering metaphysics
“the queen of all the sciences,” as Immanuel Kant put it (Aviii). But
modern natural science has deflated such expectations. The merely prob-
able character of knowledge-claims in modern science – indeed, the realiza-
tion that any current theory is likely to prove fallacious in the long run, and
that our inferences to best explanations are always tentative models of a
physical world that they never entirely fit – stands a far cry from necessary
certainty.2 That may be one of the greatest sea changes in the history of
thought.
1
On traditional natural philosophy, see, for example, Ann Blair, “Natural Philosophy,” in The Cambridge
History of Science, vol. 3: Early Modern Science, ed. Katherine Park and Lorraine Daston (Cambridge:
Cambridge University Press, 2006), 365–406.
2
As an examplar, see Ronald N. Giere, Scientific Perspectivism (Chicago: University of Chicago Press,
2006).
J. H. Zammito (*)
Department of History, Rice University, Houston, USA
e-mail: zammito@rice.edu

472 J. H. Zammito
The watershed of the transition – one that entailed a change not only in the
“style of scientific reasoning” but also in its disciplinary organization, that is, the
“calving” of natural science away from the overarching sway of philosophy and
then itself into an array of “special sciences” – can be situated in the early
nineteenth century.3 That has been captured in the title of a recent collection:
From Natural Philosophy to the Sciences.4 To be sure, this shift, as with most
historical transitions, took some time – indeed, most of the eighteenth century. It
entailed both the invention of a new way of organizing knowledge – the
elaboration of probability as a method and a standard – and also the critique of
the grander claims to absolute knowledge, especially in relation to the natural
world, in which the British empiricists, Locke and Hume, figured prominently.5
The question that all this aims to raise is where to place Kant in this epochal
transition. I wish to suggest that Kant’s critical philosophy embraced an idea of
“proper science” that partook substantially of the ideal of scientia. Consequently,
we can at best rescue from beneath this overarching schema an ambiguous notion
of empirical natural science that resonates with our own concerns.
Setting Kant’s philosophy of science in the retrospect of a long-run natural
philosophy in the West, I believe, is crucial. It tempers all-too-presentist
appropriations of Kant with a sober dose of historicism.6 We need a clearer
sense of what Andrew Cunningham has called the “Big Picture” in the
history and philosophy of science, namely the persistent power of natural
philosophy as a mode of reasoning at least through the eighteenth century.7 In
3
Ian Hacking has elaborated the idea of “styles of scientific reasoning” (“‘Style’ for Historians and
Philosophers,” Studies in History and Philosophy of Science 23, no. 1 [March 1992]: 1–20), suggesting historical
accumulation and shifting weights of approaches to scientific practices. Thomas Kuhn has offered some
important insights into the question of “calving” of disciplines (“Afterwords,” in World Changes: Thomas Kuhn
and the Nature of Science, ed. Paul Horwich [Cambridge: MIT Press, 1993], 311–41). Together, Hacking and
Kuhn have dramatically enhanced the conceptual resources for the history and philosophy of science.
4
David Cahan, ed., From Natural Philosophy to the Sciences: Writing the History of Nineteenth-Century
Science (Chicago: University of Chicago Press, 2003).
5
Ian Hacking, The Emergence of Probability: A Philosophical Study of Early Ideas about Probability,
Induction and Statistical Inference, 2nd ed. (Cambridge: Cambridge University Press, 2006).
6
I juxtapose presentism and historicism, not as mutually exclusive but as mutually dependent approaches
to the history of philosophy. Of course we appropriate earlier ideas for our present purposes; but we
should do so self-consciously cognizant that this is precisely appropriation, not legitimate historical
ascription – at least not until we have established the latter by historical reconstruction. Historicism, by
the same token, is no simple antiquarianism: it must be motivated by present problems and interests.
7
Andrew Cunningham and Perry Williams “De-Centering the ‘Big Picture’: The Origins of Modern
Science and the Modern Origins of Science,” British Journal for the History of Science 26, no. 4
(Dec. 1993): 407–32. See also Andrew Cunningham, “Getting the Game Right: Some Plain Words
on the Identity and Invention of Science,” Studies in History and Philosophy of Science 19, no. 3 (Sept.
1988): 365–89; and Andrew Cunningham, “How the Principia Got Its Name; or, Taking Natural
Philosophy Seriously,” History of Science 29, no. 4 (Dec. 1991): 377–92. All of these are reprinted in
Andrew Cunningham, The Identity of the History of Science and Medicine (Ashgate: Variorum, 2012).
his interesting essay, “Kant’s Conception of Proper Science,” Hein van den Berg
situates Kant’s philosophy of science in the context of “The Classical Model of
Science,” as explored in two special issues of the journal Synthese.8 That “classical
model” was characterized by Willem de Jong and Arianna Betti, in a theme-setting
article in the first of the two special issues, as “a millennia-old model of scientific
rationality.”9 They characterize “the ideal of scientific explanation philosophers
must have had in mind for a very long time when thinking about science” as a
foundationalist cognitio ex principis – “a more or less strictly axiomatized system”
of natural knowledge from first priniciples.10 The model maintained that true
science was an explanation, not merely a description or classification, of what existed:
why, not simply what.11 It distinguished between the order of the real (ordo
essendi) and the order of discovery (ordo cognoscendi), but precisely to insist that
science had to know, in a full (ontological, not simply conjectural) sense, the real
order of nature.12 De Jong and Betti stress the continued sway of this model over
early modern (sixteenth- through eighteenth-century) science, notably the era of
the so-called “Scientific Revolution,” when a now somewhat unfashionable
historiography presumed much of it had been overthrown.13 Thus, they maintain
that it preoccupied (in a somewhat random order) “Newton, Pascal, Spinoza,
Descartes, Leibniz, Wolff and Kant.”14 Universality and necessity proved core
features, from Aristotle all the way through Kant, who considered these essential
elements of the a priori.15 This is the ideal I denote by the concept scientia, and
what I take to be animating Kant’s sense of “proper science.”
In his essay, van den Berg identifies three core criteria in Kant’s notion of
“proper science”: systematicity, objective grounding, and apodictic cer-
tainty.16 Via the “Classical Model,” van den Berg endeavors to explain and
justify the rather “restrictive conception” of proper science in Kant and
especially his “infamous claim that any proper natural science must be
mathematical.”17 Van den Berg recognizes that we must distinguish between
8
Hein van den Berg, “Kant’s Conception of Proper Science,” Synthese 183, no. 1 (Nov. 2011): 7–26.
9
Willem R. de Jong and Arianna Betti, “The Classical Model of Science: A Millennia-Old Model of
Scientific Rationality,” Synthese 174, no. 2 (May 2010): 185–203.
10
Ibid., 186.
11
Ibid., 190.
12
Ibid., 186.
13
Katherine Park and Lorraine Daston, “Introduction: The Age of the New,” in Cambridge History of
Science, vol. 3, 1–17.
14
De Jong and Betti, “Classical Model of Science,” 187.
15
Ibid., 191.
16
Van den Berg, “Kant’s Conception of Proper Science,” 7.
17
Ibid.
474 J. H. Zammito
Kant’s articulation of “proper science” and his “active interest in the experi-
mental sciences.”18 Much of my treatment takes up this same ground,
especially the differences between “proper” and empirical natural science,
as well as how (or, indeed, whether) they can be reconciled. I will prove less
sanguine on these matters than van den Berg.19
The critical turn and a priori natural science

Kant’s commitment to a priori natural science never wavered across his
famous “critical” turn, yet the terms of his commitment had to change,
since his notion of metaphysical legitimacy underwent a fundamental shift in
response to the challenge to scientific knowledge-claims radicalized by David
Hume but already intimated by John Locke.20 This was one of the precipi-
tating motivations behind the Critique of Pure Reason, as Kant explained
most straightforwardly in his Prolegomena of 1783. Insisting that, Hume
notwithstanding, a priori knowledge in natural science could not be denied,
Kant argued that the proper philosophical question became: “how is pure
natural science possible?” That is, how could there be synthetic a priori
judgments about the “existence of things, insofar as that existence is deter-
mined according to universal laws” (Pro 4:294)? His answer, in capsule form,
went as follows: “a system of nature . . . precedes all empirical cognition of
nature and first makes it possible, and can therefore be called the true
universal and pure natural science” (Pro 4:306). One could never reach
such a vantage by mere empirical induction: “lawfulness in the connection
of appearances, i.e., nature in general, we cannot come to know through any
experience, because experience itself has need of such laws, which lie a priori
at the basis of its possibility” (Pro 4:318–19). Only via mathematics was it
possible to “construct all of its concepts in concreto yet a priori” (Pro 4:281).
Through application of the “mathematical categories” to sensible intuitions,
18
Ibid., 8.
19
But see his more extended treatment, with special reference to biology and in explicit opposition to
my views: Hein van den Berg, Kant on Proper Science: Biology in the Critical Philosophy and the “Opus
postumum” (Dordrecht: Springer, 2014).
20
“I am apt to doubt, how far soever human industry may advance useful and experimental philosophy
in physical things, scientifical will still be out of our reach: because we want perfect and adequate ideas of
those very bodies. . . . we are not capable of scientifical knowledge; nor shall ever be able to discover
general, instructive, unquestionable truths concerning them. Certainty and demonstration, are things we
must not, in these matters, pretend to” (John Locke, An Essay Concerning Human Understanding, ed.
Kenneth P. Winkler [Indianapolis: Hackett, 1996], bk. IV, ch. iii, §26 [p. 247]).
“appearances, as mere intuitions that fill a part of space and time, are
subject to the concept of magnitude, which synthetically unifies the
multiplicity of intuitions a priori according to rules” (Pro 4:309). That,
in turn, enabled the application of the “dynamical categories,” which “do
not concern the appearances and the synthesis of their empirical intuition,
but merely their existence and their relation to one another with regard
to this their existence” (A178/B220). This, ultimately, provided the basis
for Kant’s claim that there could only be so much proper (i.e., a priori)
science in an inquiry as there was mathematics in it (MFS 4:470). In
Kant’s words, “experience can be an objectively valid cognition of appear-
ances and their sequence in time only insofar as the antecedent appearance
can be connected with the subsequent one according to the rule of
[causality]” (Pro 4:312).
Yet Kant insisted this still left an indispensable role for the empirical: “We
must . . . distinguish empirical laws of nature, which always presuppose par-
ticular perceptions, from the pure or universal laws of nature, which, without
having particular perceptions underlying them, contain merely the condi-
tions for the necessary unification of such perceptions in one experience”
(Pro 4:320). He upheld the contingency of empirical law: “my proposition,
urged many times above: that experience, as a posteriori cognition, can
provide merely contingent judgments” (Pro 4:305n). Kant never doubted
that concrete empirical laws would need to be found, not simply made.
More, he took seriously the finitude of human intellect and suspected that
the establishment not only of particular empirical laws but especially also
their integration into a higher-order system would be a task that might well
exceed forever a human grasp. At the same time, however, Kant did insist
that genuine natural science would have to claim some kind of universality
and necessity, those traits he associated with the a priori. Somehow, empirical
laws needed to be grounded – or, in Gerd Buchdahl’s more tolerant formula-
tion, “nested” – in the transcendental a priori principles through which alone
experience was possible for humans. Still, these empirical laws could not
simply be deduced from a priori principles of reason.21 The problem of the
empirical was not resolved with this affirmation. Kant acknowledged with
21
“The pure faculty of understanding does not suffice, however, to prescribe to the appearances through
mere categories a priori laws beyond those on which rests a nature in general, as lawfulness of
appearances in space and time. Particular laws, because they concern empirically determined appear-
ances, cannot be completely derived from the categories, although they all stand under them” (B165).
“To be sure, empirical laws, as such, can by no means derive their origin from the pure under-
standing. . . . But all empirical laws are only particular determinations of the pure laws of the under-
standing, under which and in accordance with whose norm they are first possible” (A127–28).
476 J. H. Zammito
increasing candor over the course of the 1780s, and most explicitly in the two
introductions to the Critique of Judgment, that the first Critique had only
grounded an experience of nature in general, but not the existence or
coherence of empirical laws (FI 20:193–251; CJ 5:171–97).
Metaphysics of nature: “Proper science”

The preface to Kant’s Metaphysical Foundations of Natural Science is one of the
densest and most “architectonic” formulations in the critical philosophy, as well
as the most important general discussion of his notion of natural science.22 It
begins with very careful definitions of nature and of science, then proceeds to a
set of discriminations about their relation. Nature is discriminated into a
“formal” and a “material” sense. The “formal signification” identified “the
primal, internal principle” or the essence of “everything that belongs to the
existence of a thing,” but it only articulated possibility.23 Existence entailed
actuality, hence this “formal principle” was insufficient to articulate nature.
Only the second, or material sense constituted the sphere for science. Kant
defined nature in this material sense as “the sum total of all things, insofar as
they can be objects of our senses, and thus also of experience” (MFS 4:467). The
existence of objects depended, for us, upon their presence in sensibility.24 Such
objects could be of outer sense or of inner sense, and inquiry into these
produced doctrines (Lehre) of body and of soul, respectively. But not all Lehre
deserved the designation Wissenschaft, according to Kant, and this was one of
his crucial interventions in the semantic articulation of scientific practices of his
time. The generic term for natural science in German eighteenth-century
discourse was in fact Naturlehre, not Naturwissenschaft.25 The shift to the latter
designation would be a feature of the transition into modern natural science
22
This work has come under intense scrutiny in recent times. That began with the collection of essays by
Robert E. Butts, ed., Kant’s Philosophy of Physical Science: Metaphysische Anfangsgründe der
Naturwissenschaft 1786–1986 (Dordrecht: Reidel, 1986). Michael Friedman made it the central focus
first of his Kant and the Exact Sciences (Cambridge: Harvard University Press, 1992) and more recently
in Kant’s Construction of Nature: A Reading of the “Metaphysical Foundations of Natural Science”
(Cambridge: Cambridge University Press, 2013). Konstantin Pollok has worked up a careful commen-
tary: Kants “Metaphysische Anfangsgründe der Naturwissenschaft”: Ein kritischer Kommentar (Hamburg:
Meiner, 2001).
23
It remains to ask why, then, Kant seems to indicate that this “formal signification” could occasion a
plurality of “natural sciences,” each articulating the specific, essential difference of things. He associated
this with distinction in quality. But how this could be natural science was left unexamined.
24
Conversely, as Kant argued in the Refutation of Idealism in the B-version of the first Critique, the
coherence of our consciousness required the existence of objects of outer sense.
25
See the section below on the context of scientific discourse.
after the turn to the nineteenth century, and it remains to establish how crucial
Kant was in this mutation.
What was Kant’s definition of science (Wissenschaft)? He wrote: “Every
doctrine [Lehre] that is supposed to be a system, that is, a whole of cognition
ordered according to principles, is called a science [Wissenschaft]” (MFS
4:467). Here he articulated the essential condition of systematicity for
science.26 Then Kant distinguished between empirical and rational princi-
ples. Kant suggested it might appear there were “historical” as well as rational
natural sciences, but he immediately countered that the “historical” (i.e.,
empirical, which the eighteenth century designated as cognitio historica)
could not satisfy the system requirement because the latter entailed “a
rational cognition of the coherence of things.”27 Now, Kant almost imme-
diately wrote of a “systematic art” as the empirical counterpart of a rational
science. That would seem to suggest that empirical systematicity is possible,
and indeed Kant asserted that in chemistry it was actual (MFS 4:468, 471).
He even conceived of “natural history [Naturgeschichte]” as “a systematic
presentation [!] of natural things at various times and places” (MFS
4:468).28 Thus the denial that systematicity could be possible empirically
seems compromised.29 In any event, Kant proceeded to discriminate between
“historical Naturlehre,” as merely “systematically ordered [!] facts about
natural things,” and proper natural science. This occasioned the key assertion
of the preface: “What can be called proper science is only that whose certainty
is apodictic; cognition that can contain mere empirical certainty is only
knowledge [Wissen] improperly so-called” (MFS 4:468). There were two
aspects to Kant’s discrimination of rational systematicity from historical
systematicity: first, it required apodictic certainty, that is, the universality
and necessity of the a priori; and, second, it required that the system so
constructed be “an interconnection of grounds and consequences,” that is, a
26
This is one of the three essential features that van den Berg identifies with “proper science.”
27
Arno Seifert, Cognitio historica: Die Geschichte als Namengeberin der frühneuzeitlichen Empirie (Berlin:
Duncker & Humblot, 1976).
28
Kant endeavored to redefine the conventional sense of Naturgeschichte (natural history) by discrimi-
nating between Naturbeschreibung, the traditional sense of natural history as classificatory description,
and a new, literal sense of “history of nature” or Naturgeschichte. He eventually gave up on this effort and
suggested alternative terms for this new science: “physiogony” and “archaeology of nature.” See my
“From Natural History to History of Nature: Kant between Buffon and Herder,” in Rethinking Kant,
vol. 5, ed. Pablo Muchnik and Oliver Thorndike (Cambridge: Cambridge Scholars, forthcoming).
29
While the epoch made a distinction between an esprit de système and an esprit systematique (first
articulated by the Abbé de Condillac), it is hard to believe that Kant was making a clear distinction
between system and the systematic in anything like that sense. Without it, however, his language is
ambiguous.
478 J. H. Zammito
specifically determined set of relations informed by the dynamical categories,

above all causality (MFS 4:468).30 Kant claimed “this necessity of laws is
inseparably attached to the concept of nature” (MFS 4:469), that is, “the
word nature already carries with it the concept of laws, and the latter carries
with it the concept of the necessity of all determinations of a thing belonging
to its existence” (MFS 4:468). It is not clear that this was what Kant’s
“material signification” of nature betokened; perhaps nature got its formal
signification automatically with its material actuality.31 In any event, articu-
lating this immanent, necessary lawfulness in nature was what “pure” natural
science achieved, what made it “proper” or authentic science. One had,
according to Kant, to establish first and independently (a priori) this
“pure” metaphysics of nature in order to inform and enable applications in
actual empirical natural science; in short, “properly so-called natural science
presupposes, in the first place, metaphysics of nature” (MFS 4:469).
Metaphysics of nature itself required a transcendental foundation; it
required that the pure concepts of the understanding be applied to objects
of experience, which in turn required presentation in intuition, and thus
mathematical construction: “to cognize something a priori means to cognize
it from its mere possibility. But the possibility of determinate natural things
cannot be cognized from their mere concepts.” It “required that the intuition
corresponding to the concept be given a priori, that is, that the concept be
constructed. Now rational cognition through construction of concepts is
mathematical” (MFS 4:470). And thus, Kant came to his famous stipulation:
“a doctrine of nature will contain only as much proper science as there is
mathematics capable of application there” (MFS 4:470). Kant then advanced
to “special metaphysical natural science” (MFS 4:470), presenting in “absolute
completeness” the “metaphysics of corporeal nature” in the four architectonic
moments that constitute the bulk of his Metaphysical Foundations (MFS
4:473, 472). He stressed completeness as the essence of rational systemati-
city: “a complete analysis of the concept of a matter in general will have to be
taken as the basis [of the doctrine of body]” (MFS 4:472). While such a
construction could achieve totality, Kant averred, the “empirical doctrine of
nature” could only construe stepwise ad infinitum (MFS 4:473).
The questions that pile up in all this have to do with the residual place
allocated to the empirical and its ordered cognition. What is a “systematic
30
These are van den Berg’s other two core features of Kant’s “proper science.”
31
It seems more likely that these assertions derived from principles not expounded in this text but in the
first Critique.
art”? What are “empirical principles”? What is a “systematic presentation” of

empirical things? What is “empirical certainty”?32 What cognitive warrant do
all these carry? How do all these notions link with the “metaphysics of
nature” that ostensibly grounds and authorizes them? Whatever the achieve-
ment of Kant’s Metaphysical Foundations of Natural Science, I submit, these
questions were not adequately addressed. Kant himself acknowledged this
just a few years later in the First Introduction to the Critique of Judgment.
The First Introduction to the Critique of Judgment

and empirical science
As Robert Butts has noted, “It is only in the Critique of Judgment that [Kant]
comes to deal with science, not as a finished system, but as a research
program.”33 That is, Kant came to take seriously the problem of empirical
entailment, in which rather than merely prescribing to nature, human inquiry
had to presume and then seek out an “order” that was somehow amenable to
discovery and comprehension (CJ 5:183). What Kant’s Critique of Pure
Reason proved was that the categories could apply to all objects of possible
experience, but just in the structure of that proof the problem manifests
itself: what can hold for all objects cannot completely determine any one in
particular. While, according to the Transcendental Deduction of the first
Critique, “experience in general in accordance with transcendental laws of the
understanding is to be regarded as a system and not as a mere aggregate,”
Kant wrote,
it does not follow from this that nature even in accordance with empirical laws
is a system that can be grasped by the human faculty of cognition. . . . For the
multiplicity and diversity of empirical laws could be so great that it might be
possible for us to connect perceptions to some extent in accordance with
particular laws discovered on various occasions into one experience, but
never to bring these empirical laws themselves to the unity of kinship under
a common principle. (FI 20:209)
Indeed, each experience could be such that it showed no similarity to any

other, so that each would stand in total isolation and consciousness would be
32
That sounds dangerously oxymoronic.
33
Robert Butts, “Teleology and Scientific Method in Kant’s Critique of Judgment,” Noûs 24, no. 1
(March 1990): 1.
480 J. H. Zammito
confronted with “a raw chaotic aggregate and not the least trace of a system”
(FI 20:209). Under such conditions, consciousness, though a formal unity,
would be faced with a nightmare of particulars, of individual entities for
which no classificatory empirical concepts could be found.
That only the general concept of nature had been secured for certainty,
and that the empirical had no a priori coherence, resulted in a substantial
circumscription of what had seemed the established realm of the under-
standing (Verstand), and cleared a very large space for an expanded notion of
judgment (Urteilskraft) as a theoretical resource.34 The full-fledged language
of judgment was a late innovation in Kant’s critical philosophy. Particularly,
the notion of “reflective judgment” was not worked out in the first Critique
of 1781, or even integrated into the revised version of 1787.35 It was only in
the mid-1780s that Kant began to write a great deal more about the “faculty
of judgment.” The discovery of “reflective judgment,” as Giorgio Tonelli and
others have argued, was the decisive turn which engendered the Critique of
Judgment as the whole we know.36
In making a distinction between the “apodictic” and the “hypothetical”
use of reason in the Transcendental Dialectic of his Critique of Pure Reason,
Kant came closest to anticipating the key distinction of reflective from
determinant judgment which he enunciated in the First Introduction to
the Critique of Judgment. Kant wrote:
If reason is the faculty of deriving the particular from the universal, then: Either
the universal is in itself certain and given, and only judgment is required for
subsuming, and the particular is necessarily determined through it. This I call
the “apodictic” use of reason. Or the universal is assumed only problematic-
ally, and it is a mere idea, the particular being certain while the universality of
the rule for this consequent is still a problem; then several particular cases,
which are all certain, are tested by the rule, to see if they flow from it, and in the
case in which it seems that all the particular cases cited follow from it, then the
34
See Reinhard Brandt, “The Deductions in the Critique of Judgment: Comments on Hampshire and
Horstmann,” in Kant’s Transcendental Deductions: The Three “Critiques” and the “Opus postumum,” ed.
Eckart Förster (Stanford: Stanford University Press, 1989), 177–79. Most modern, and a fortiori most
Anglo-American, philosophers routinely privilege “understanding” in their reading of Kant’s theory of
mental activity. See Gerd Buchdahl, “The Relation between ‘Understanding’ and ‘Reason’ in the
Architectonic of Kant’s Philosophy,” Proceedings of the Aristotelian Society 67 (1966–1967): 209–26.
35
Compare Max Liedtke, “Der Begriff der reflektierenden Urteilskraft in Kants Kritik der reinen
Vernunft” (PhD diss., University of Hamburg, 1964).
36
See, above all, Giorgio Tonelli, “La formazione del testo della Kritik der Urteilskraft,” Revue inter-
nationale de philosophie 30 (1954): 423–48; and Michel Souriau, Le jugement réfléchisssant dans la
philosophie critique de Kant (Paris: Alcan, 1926).
universality of the rule is inferred, including all subsequent cases, even those
that are not given in themselves. This I will call the “hypothetical” use of
reason. (A646–47/B674–75)37
This is Kant’s most explicit consideration of the problem of induction in the

whole first Critique.
Only insofar as we can classify similarities and differences among experi-
ences is there any hope for order in our consciousness, however full each of
our intuitions might be. Thus the concern is with the possibility of laws, that
is, concepts which express uniformities among experiences and thereby
organize empirical knowledge. Yet these laws themselves would be capricious
and unavailing if there were not some prospect of coherence among them,
some guarantee of “system” analogous to that provided transcendentally by
the unity of apperception and the categories. That, in its turn, meant that
nature as an empirical whole had to be a unity, indeed, a system of laws in
strict conformity with the systematic structure of human reason, such that
the unity was accessible to human comprehension. The only way that there
could be a hope for system in empirical laws was if it were possible for
consciousness to presume in empirical nature a propensity to organize itself
according to the principles of logical articulation to which human beings
were necessitated by discursive consciousness.
Human understanding of the empirical world could only find uniformities
or homogeneity – that is, it could only classify – if nature had already actually
specified in a logical order, that is, in hierarchical structure. Kant argued that
it was not human “favor” that nature appeared ordered. Humans had to
presume upon nature’s favor. While it was true that in general natural law
was imposed by the necessary structure of transcendental consciousness, it
was just as true that empirically what law there might be had to be found, not
made by us. While Kant argued against Hume that the concept of causality
was necessary at the transcendental level, he acknowledged at the same time
that Hume had every right to consider any empirical application of that
principle to be contingent. Yet it was impossible for humans to deal with this
empirical experience except on the assumption that nature did so favor us.
When nature manifested orderliness in the empirical sphere, as it were
gratuitously, it appeared as if it were an artist. That is, its result appeared
purposive. According to Kant, human consciousness discerned in nature
37
Here already we encounter the curious notion of empirical certainty. What can this mean beyond
problematic givenness in sensibility, and how determinate can that be?
482 J. H. Zammito
gratuitous design, that is, works of nature (Wirkungen) appeared like works of
artifice (Werke) (Pro 4:145). Nature could hardly be taken literally for an
artist, and what was really at play in the notion of a Technik der Natur was
simply analogy.38 It was not nature to which, in fact, “technic” belonged, but
the faculty of judgment (FI 20:220). Kant wrote:
The reflecting power of judgment thus proceeds . . . not schematically, but

technically, not as it were merely mechanically, like an instrument, but
artistically, in accordance with the general but at the same time indeterminate
principle of a purposive arrangement of nature in a system. . . . But to represent
nature as technical, like a reason (and so to attribute purposiveness and even
ends to nature) is a special concept, which we cannot encounter in experience
and which only the power of judgment introduces into its reflection on objects,
in order to treat experience, following its direction, in accordance with special
laws, namely those of the possibility of a system. (FI 20:213–14, 234–35)
Just by virtue of that gratifying discernment of gratuitous order, Kant

generalized a principle for nature as a whole: the purposiveness of nature in
its empirical variety for human comprehension in empirical concepts, the
transcendental principle of reflective judgment.
One of Buchdahl’s most important contributions to the history and the
philosophy of science was his argument for a crucial “looseness of fit”
between the transcendental and the empirical elements in Kant’s epistemol-
ogy.39 Buchdahl construed Kant’s restriction of determinant judgment (the
constitutive use of reason) in his third Critique as entailing that all empirical
judgments are reflective (or: all empirical knowledge is grounded merely in a
contingent, “subjective principle”). By distinguishing between “transcenden-
tal lawlikeness” and “empirical lawlikeness,” Buchdahl sought to make sense
of this whole problematic in terms of “looseness of fit.” As Michael Friedman
summarizes the outcome, “not only is the understanding entirely powerless
38
In the first Critique, Kant explained that analogy could not, in philosophy, achieve the specificity of
result that it could in mathematics. It could not give the fourth element, when the other three were given
in analogical relation, but it could only validate the relation. Yet analogies were extremely important in
human cognition. They functioned especially in the sphere where determinate concepts failed, that is,
where understanding was not legislative.
39
See Buchdahl, “Relation between ‘Understanding’ and ‘Reason’”; Gerd Buchdahl, “The Kantian
‘Dynamic of Reason,’ with Special Reference to the Place of Causality in Kant’s System,” in Kant Studies
Today, ed. Lewis White Beck (La Salle, Ill.: Open Court, 1969), 341–74; Gerd Buchdahl, “The
Conception of Lawlikeness in Kant’s Philosophy of Science,” Synthese 23, no. 1 (Aug. 1971): 24–46;
and Gerd Buchdahl, “Kant’s ‘Special Metaphysics’ and the Metaphysical Foundations of Natural Science,”
in Kant’s Philosophy of Physical Science, 121–61.
with respect to particular empirical concepts and particular empirical causal

laws, but the search for such concepts and laws lies rather within the purely
regulative province of reflective judgment.”40 Therewith, however, Kant’s
insistent distinctions between objective and subjective principles, and
between validity claims of mechanical and teleological explanation in science,
all seem in peril of dissolving. This cannot be entirely right. In Kant’s
language, we can have warranted knowledge only through “determinant”
judgments. Proof is always deductive. There is a radical incongruity between
the reasoning involved in discovery (or invention) and the reasoning involved
in proof.
Friedman has examined the entire sweep of Kant’s philosophy of science
to resolve this key puzzle for students of Kantian epistemology in general:
how can there be objective empirical laws? Friedman sets out from what
Buchdahl established from Kant, namely that “the law-governedness of
nature under universal transcendental laws of the understanding does not
at all guarantee that nature is also governed by particular empirical laws.”41
But then he reconstructs a peculiar “mixed status” of Kantian empirical laws,
which in virtue of their empirical character have to be contingent, but in
virtue of their status as laws have somehow to claim necessity.42 The crucial
point is that “even empirical laws too must have a more than merely
inductive status.”43 The “mixed status” of empirical laws and the question
of the possibility of their systematicity have become, accordingly, central
issues for Kant’s philosophy of science.44
As Friedman urges, “particular causal laws, for Kant, have a peculiar
kind of mixed status. . . . Insofar as particular causal laws merely record
observed regularities they are contingent and a posteriori, insofar as they
subsume such regularities under the a priori principle of causality, how-
ever, they are necessary – and even, in a sense, a priori.”45 That is why,
40
Michael Friedman, “Regulative and Constitutive,” Southern Journal of Philosophy 30, no. S1 (Spring
1992): 76.
41
Michael Friedman, “Causal Laws and the Foundations of Natural Science,” in The Cambridge
Companion to Kant, ed. Paul Guyer (Cambridge: Cambridge University Press, 1992), 167.
42
Thus, Kant was careful always to write that we view empirical laws as necessary and universal; this in
contrast to simply knowing it. See Pro 4:312.
43
Friedman, “Causal Laws,” 172.
44
Paul Guyer, “Kant’s Conception of Empirical Law,” Proceedings of the Aristotelian Society, supplement
64 (1990): 221–42; Paul Guyer, “Reason and Reflective Judgment: Kant on the Significance of
Systematicity,” Noûs 24, no. 1 (March 1990): 17–43; and Margaret Morrison, “Methodological Rules
in Kant’s Philosophy of Science,” Kant-Studien 80, nos. 1–4 (Jan. 1989): 155–72.
45
Friedman, “Causal Laws,” 174.
484 J. H. Zammito
for Friedman, the Metaphysical Foundations of Natural Science represented

a decisive grounding for empirical natural science. Establishing a priori
the metaphysical concept of matter and deriving from it, with no further
empirical input than the existence of matter as “the moveable in space,”
the necessity of postulating the fundamental forces of attraction and
repulsion in physical theory, this was Kant’s paradigm instance of scien-
tific lawfulness a priori, the standard – mathematical and causal
(“mechanical” in the language of the Critique of Judgment) – for all
natural science.46 However, as Friedman acutely notes, Kant was faced
immediately with a very uncomfortable question: “how was this bril-
liantly successful Newtonian paradigm to be extended beyond astronomy
and celestial mechanics?”47 Kant had to uphold the “claim that particular
empirical laws are somehow made possible by – are grounded in or
determined by – the transcendental principles.”48 The problem is,
Friedman acknowledges, that “we are left quite in the dark concerning
the precise nature of this ‘grounding.’”49
There is no guarantee that the discoveries of empirical science “from
below” will converge neatly with that “top-down” structure established by
Kant’s metaphysics of natural science.50 As Friedman observes,
all the rest of empirical natural science remains solely within the regulative
purview of reason and reflective judgement, the aim of which is to ascend from
lower level empirical concepts and laws towards ever more general empirical
concepts and laws so as eventually (in prospect) to attain a complete classifi-
catory and hierarchical system . . . under the highest level empirical concept and
law already constituted as such in the Metaphysical Foundations of Natural
Science.51
Kant faced a massive issue in integrating his a priori derivation of physics

with the empirical lawfulness in the balance of nature that he himself
established as a need of reason. From this point onward Kant saw himself
46
Ibid., 189–90; and Friedman, Kant and the Exact Sciences, 213–42.
47
Friedman, Kant and the Exact Sciences, 240.
48
Friedman, “Causal Laws,” 171–72.
49
Ibid., 174.
50
“So what assurance do we have that the regulative operation of reason and reflective judgement will,
proceeding from the bottom up, actually converge in the direction of this already constituted higher
level?” (Friedman, “Regulative and Constitutive,” 94).
51
Ibid., 90–91.
caught in the toils of a “gap in the critical system” which became the
obsessive theme of his Opus postumum.52
More drastically still, there was something in the very “top-down”
constitution of the metaphysical principles that threatened in principle the
“necessary convergence of constitutive and regulative procedures . . . abso-
lutely essential to Kant’s entire project.”53 Kant purchased the determinacy
of his metaphysical principles at a significant cost. The binding constraint of
the concept of matter he adopted in Metaphysical Foundations of Natural
Science was that the laws he generated could apply only to outer sense. This
“has the effect of restricting our attention to nonliving material substances,” a
dramatic restriction in scope not only of transcendental principles but of
actually existing nature.54 Henry Allison, among others, has noted the
serious problems for the coherence of Kant’s overall philosophy of science
involved in isolating the lawfulness of “inner sense” from that of “outer
sense.”55 Moreover, by defining matter as essentially lifeless, Kant excluded all
actually living things in principle from conformity to the metaphysical
foundations of natural science. Given their prominence among the actualities
of the world for natural science, Kant’s segregation of the realm of living
things from the theory of physical sciences precluded by definition any
empirical (“bottom-up”) integration of empirical concepts and laws in
these domains which could converge with the “top-down” foundation of
his science.56Any science involving “internal purposiveness” becomes irre-
concilable with “proper” science.
52
Friedman, Kant and the Exact Sciences, 214–15. On this idea of the “gap,” see Eckart Förster, “Is
There a ‘Gap’ in Kant’s Critical System?” Journal of the History of Philosophy 25, no. 4 (Oct. 1987): 533–
55; Burkhard Tuschling, Metaphysische und transzendentale Dynamik in Kants Opus postumum (Berlin:
de Gruyter, 1971); and Burkhard Tuschling, “Apperception and Ether: On the Idea of a Transcendental
Deduction of Matter in Kant’s Opus postumum,” in Kant’s Transcendental Deductions, 193–216.
53
Friedman, “Regulative and Constitutive,” 95.
54
Friedman, “Causal Laws,” 185. As Friedman notes, “Thus, the metaphysical principles of pure natural
science apply only to the activities and powers of nonliving, nonthinking beings: beings represented
solely through predicates of outer sense. The transcendental principles of the understanding, by contrast,
apply to all beings without distinction – where, for example, inner principles of causality (appropriate to
living beings) are just as permissible as external causes” (“Causal Laws,” 182).
55
Henry E. Allison, “Causality and Causal Laws in Kant: A Critique of Michael Friedman,” in Kant and
Contemporary Epistemology, ed. Paolo Parrini (Dordrecht: Springer, 1994), 291–308.
56
“The inertia of matter is, and means, nothing else than its lifelessness, as matter in itself. Life is the
faculty of a substance to determine itself to act from an internal principle, of a finite substance to change,
and of a material substance [to determine itself] to motion or rest, as change of its state. Now we know no
other internal principle in a substance for changing its state except desiring, and no other internal activity
at all except thinking” (MFS 4:544).
486 J. H. Zammito
This brings into glaring salience the problem of reconciling biology at

all with Kant’s prescriptions for science. Organisms rupture the “top-
down”/“bottom-up” integration of Kant’s scientific system. I think it is
essential to dwell on Kant’s suggestion that there is a radical incongruity
between his notion of organic form as “intrinsic purposiveness” and his
principle of “proper” natural science: “it is necessary that its form not be
possible in accordance with mere natural laws, i.e., ones that can be
cognized by us through the understanding, applied to objects of the
senses, alone” (CJ 5:370). In short, biology, for Kant, could only be a
“transcendent science.”57 It fell beyond the pale of natural science –
“proper” or even empirical.58 That, I submit, is more than contemporary
biology or its philosophy can or should swallow, and Kant’s notion of
“proper science” needs to be consigned to the pre-history of modern
natural science.
The context of actual natural science in late

eighteenth-century Germany
One way to situate the issue of the historicist versus the presentist reconstruc-
tions of Kantian philosophy of science is to relate it to the scientific community
of Kant’s own day. In German eighteenth-century discourse, the general term
used for natural philosophy or physical science was Naturlehre. A classic
illustration of this was the Göttingen scholar Johann Christian Polycarp
Erxleben (1744–1777), the short-lived prodigy whose famous Anfangsgründe
der Naturlehre (1st ed., 1772) served as the dominant textbook in physical
science in German universities for the last quarter of the eighteenth century.59
Erxleben’s textbook illustrates how German scientific culture of the eighteenth
57
This is Clark Zumbach’s revealing title. See Clark Zumbach, The Transcendent Science: Kant’s
Conception of Biological Methodology (The Hague: Nijhoff, 1984).
58
I have pressed this argument in several essays: “‘This Inscrutable Principle of an Original
Organization’: Epigenesis and ‘Looseness of Fit’ in Kant’s Philosophy of Science,” Studies in History
and Philosophy of Science 34, no. 1 (March 2003): 73–109; “Teleology Then and Now: The Question of
Kant’s Relevance for Contemporary Controversies over Function in Biology,” Studies in History and
Philosophy of Science, Part C: Studies in History and Philosophy of Biological and Biomedical Sciences 37,
no. 4 (Dec. 2006): 748–70; “Kant’s Notion of Intrinsic Purposiveness in the Critique of Judgment: A
Review Essay (and an Inversion) of Zuckert’s Kant on Beauty and Biology,” Kant Yearbook, vol. 1:
Teleology (Berlin: de Gruyter, 2009), 223–47; and “Organism: Objective Purposiveness,” in Kant: Key
Concepts, ed. Will Dudley and Kristina Engelhard (London: Acumen, 2011), 170–83.
59
See William Clark, “German Physics Textbooks in the Goethezeit,” History of Science 35, no. 2 (June
1997): 219–39; and no. 3 (Sept. 1997): 295–363.
century articulated its “semantic field.”60 Of even greater significance are some
emergent mutations in this semantic field.61 An early exemplar of these muta-
tions was the inaugural address delivered in Jena, 1762, by the professor of
medicine Ernst Gottfried Baldinger (1738–1804). In Grenzen der Naturlehre,
Baldinger drew three distinctions in terminology.62 First, he distinguished
Naturlehre from applied mathematics by virtue of its empirical, as opposed to
a priori, element.63 Second, he distinguished Naturlehre from more particular
fields of investigation, such as chemistry or mineralogy, by virtue of its universal
scope. Finally, he distinguished it from traditional Naturgeschichte (natural
history in the sense of classificatory description), because it had causal and
explanatory ambitions. Baldinger’s statement reflected the moment of transi-
tion of the semantic field of German natural science in all its openness to the
future. What he still termed Naturlehre would come to be called
Naturwissenschaft, but, as in his account, it would stress the empirical as against
the mathematical in its ambition to extend its scope to embrace all the emergent
domains of natural-scientific inquiry.64
Stuart Strickland, drawing on Rudolf Stichweh, observes that
“eighteenth-century writers tended to use the terms Physik and
Naturlehre interchangeably,” but there was an emergent tendency that
“Naturlehre tended to be experimental sciences whose phenomena had
thus far eluded mathematical representation.”65 Summarizing later, he
writes:
60
The idea of a “semantic field” was developed by Rudolf Stichweh, Zur Entstehung des modernen
Systems wissenschaftlicher Disziplinen: Physik in Deutschland 1740–1890 (Frankfurt am Main: Suhrkamp,
1989).
61
Volker Hess, “Das Ende der ‘Historia naturalis’? Die naturhistorische Methode und Klassifikation bei
Kielmeyer,” in Philosophie des Organischen in der Goethezeit: Studien zu Werk und Wirkung des Naturforschers
Carl Friedrich Kielmeyer (1765–1844), ed. Kai Torsten Kanz (Stuttgart: Steiner, 1994), 172.
62
Ernst Gottfried Baldinger, Die Grenzen der Naturlehre warden bestimmt: Eine akademische Abhandlung
(Torgau: Rüdel, 1762). On Baldinger’s lecture, see Paul Ziche, “Von der Naturgeschichte zur
Naturwissenschaft: Die Naturwissenschaften als eigenes Fachgebiet an der Universität Jena,” Berichte
zur Wissenschaftsgeschichte 21, no. 4 (1998): 253.
63
The general idea of physical science was itself blurred on two fronts – between the “applied
mathematics” of astronomy and mechanics, on the one side, and “experimental” fields such as chemistry
and mineralogy, on the other (Olaf Breidbach and Paul Ziche, “Einführung: Naturwissen und
Naturwissenschaften: Zur Wissenschaftskultur in Weimar/Jena,” in Naturwissenschaft um 1800:
Wissenschaftskultur in Jena-Weimar, ed. Olaf Breidbach and Paul Ziche [Weimar: Böhlaus, 2001], 9).
We can trace this back to two trajectories out of Newton, one from the Principia, the other from the
Opticks. See Bernard Cohen, Franklin and Newton (Philadelphia and Cambridge: American
Philosophical Association and Harvard University Press, 1966), for the classic formulation.
64
Ziche, “Von der Naturgeschichte zur Naturwissenschaft.”
65
Stuart Strickland, “Galvanic Disciplines: The Boundaries, Objects, and Identities of Experimental
Science in the Era of Romanticism,” History of Science 33, no. 4 (Dec. 1995): 452.
488 J. H. Zammito
Naturlehre and Physik began the eighteenth century as indistinguishable global

terms. By mid-century or slightly later, a division had opened between the
sciences of the organic and those of the inorganic, a division that seems to have
been based on a distinction drawn from natural differences in the objects and
reinforced by the methods applied to those objects.66
This bespeaks a more general proliferation of “experimental” physics into a

wide range of research programs (Fachgebiete) in eighteenth-century science
more in keeping with Newton’s queries to the Opticks than with his math-
ematical agenda in the Principia. That creates a very important, indeed
disruptive context for Kant’s notion of “proper” science.
The University of Jena was the first decisive center of reception of Kant’s
philosophy after 1786.67 Over the 1790s at that university, the rubric
Naturwissenschaft gradually displaced Naturlehre as the umbrella term for
natural scientific course offerings.68 But that was only in the context of a
general flux in the semantic field. According to Olaf Breidbach and Paul
Ziche, “the concepts of ‘Naturkunde’ [nature studies], ‘Naturgeschichte’ [nat-
ural history], ‘Naturforschung’ [natural inquiry] and finally, as the most
general concept, that of ‘Wissenschaft’ [science]” all appeared in the course
listings simultaneously, “without the attempt at any theoretical clarification.”
Gradually, Naturlehre started to be displaced by Naturwissenschaft.69 While,
in general usage, Wissenschaft in German can be quite encompassing, one
impetus in the late eighteenth-century recourse to Naturwissenschaft was
toward stringency. Along those lines, Kant’s Metaphysical Foundations of
Natural Science (Naturwissenschaft) was certainly a force in the semantic
flux.70 Yet his highly restrictive characterization of Naturwissenschaft, which
only admitted systematic, a priori, and above all mathematical fields within
its scope, failed to win wide acceptance (at least in this epoch).71
66
Ibid., 463. See Stichweh, Zur Entstehung des modernen Systems wissenschaftlicher Disziplinen. See also
Hess, “Das Ende der ‘Historia naturalis’?” esp. 154, 157.
67
Norbert Hinske, Erhard Lange, and Horst Schröpfer, eds., Der Aufbruch in den Kantianismus: Der
Frühkantianismus an der Universität Jena von 1785–1800 und seine Vorgeschichte (Stuttgart-Bad
Cannstatt: Frommann-Holzboog, 1995).
68
See Ziche, “Von der Naturgeschichte zur Naturwissenschaft.”
69
“Among the most striking developments in intellectual history of the time around 1800 belongs an
intensive discussion of the concept of ‘science,’ which was conducted in Jena at a very high level”
(Breidbach and Ziche, “Naturwissen und Naturwissenschaften,” 16).
70
Kant sought the eminent physicist Lichtenberg as the reviewer of this work, but Lichtenberg declined,
for a variety of reasons, one of which may have been this impulse toward stringency.
71
Breidbach and Ziche, “Naturwissen zur Naturwissenschaften,” 19–20.
That leads us to the question how Kant’s philosophy of “proper” science

related to the actual practices of science – in particular, life science – in
Germany at the turn from the eighteenth to the nineteenth century. There is
no doubt that naturalists (Naturforscher) of the day reached out to Kantian
philosophy for legitimation of their methodology.72 Kant provided them
epistemological comfort in their pursuit of life forces (Lebenskräfte), especially
given the thinness of their analogy to Newtonian gravity.73 But this was more a
convenient ploy than an epistemological commitment. German naturalists of
the era took vital forces for actual, not regulative. Their project was to specify
their effects through the mechanisms – formative forces (Bildungskräfte) – they
set in motion. Kant’s regulative/constitutive distinction proved useless for them
in that pursuit. There is perhaps no more widely accepted historical finding
about the life sciences in the German 1790s – even, or especially when they
invoked Kantian terms – than that they slid one and all from a regulative into an
unmistakably constitutive conception of organism.74 This was indeed an
inevitable consequence of their commitment to the empirical practice of a life
science, which Kant’s philosophy in fact rendered extremely problematic.
To accentuate this point, it is not at all clear that the naturalists of the
era 1790–1820 could come to a clear determination of exactly what
Kant’s philosophy meant. Kantianism proved starkly polyvalent. To be
sure, Kant was invoked – ubiquitously.75 But there was nothing unequiv-
ocal about that invocation.76 He proved a bewilderment, not a panacea.
72
Timothy Lenoir, “The Göttingen School and the Development of Transcendental Naturphilosophie
in the Romantic Era,” in Studies in the History of Biology, vol. 5, ed. William Coleman and Camille
Limoges (Baltimore: Johns Hopkins University Press, 1981), 111–205. See my critique of Lenoir’s
conception: “The Lenoir Thesis Revisited: Blumenbach and Kant,” in Studies in History and Philosophy
of Science, Part C: Studies in History and Philosophy of Biological and Biomedical Sciences 43, no. 1 (March
2012): 120–32.
73
On Newtonian analogies in life science in the eighteenth century, see Thomas S. Hall, “On Biological
Analogs of Newtonian Paradigms,” Philosophy of Science 35, no. 1 (March 1968), 6–27; A. E.
Gaissinovich, “Le rôle du Newtonianisme dans la renaissance des idées épigénetiques en embryologie
du XVIIIe siècle,” in Actes du XIe Congrès International d’Histoire des Sciences (Paris 1968) (Paris:
Blanchard, 1971), 5:105–10; and more recently, Charles Wolfe, “On the Role of Newtonian
Analogies in Eighteenth-Century Life Science: Vitalism and Provisionally Inexplicable Explicative
Devices,” in Newton and Empiricism, ed. Zvi Biener and Eric Schliesser (Oxford: Oxford University
Press, 2014), 223–61.
74
Perhaps the most direct articulation of this point is James L. Larson, “Vital Forces: Regulative
Principles or Constitutive Agents? A Strategy in German Physiology, 1786–1802,” Isis 70, no. 2
(June 1979): 235–49.
75
Brigitte Lohff, Die Suche nach der Wissenschaftlichkeit der Physiologie in der Zeit der Romantik: Ein
Beitrag zur Erkenntnisphilosophie der Medizin (Stuttgart: Fischer, 1990), 42, 44.
76
As Lohff so well states, “Kant’s influence is more frequently assumed than specified” in the secondary
literature on the period (Ibid., 14).
490 J. H. Zammito
That is why, as Dieter Henrich has observed with harsh accuracy,

Kantianism dissolved almost instantly into a host of “post-Kantianisms”
over the decade of the 1790s.77 Kant’s own publications contributed to
the confusion. In 1800, the Jäsche Logic appeared, offering to a wider
readership Kant’s general introduction to the field of philosophy as he
had presented it to students at Königsberg University for a generation.
But this text was an amalgam of pre-critical and critical formulations,
not a guide to the critical philosophy in its finished form. Almost
simultaneously, three new editions of Kant’s Universal Natural History
and Theory of the Heavens (originally 1755) appeared in 1797, 1798, and
1808, to say nothing of its inclusion in several collected editions of
Kant’s work that were brought to market at the very end of the eight-
eenth century.78 The Universal Natural History was Kant’s boldest effort
toward a history of nature (cosmogony), but it also contained his most
imaginative speculations about life on other planets and so on. Here
“method” and “manner” were blurred in a way that was hardly consistent
with Kant’s critically policed natural science, yet it was widely and
enthusiastically received.79 Even the critical Kant seemed to point in a
myriad of directions.80 Peter McLaughlin has made the very important
point that the message of the Metaphysical Foundations of Natural Science
was not the same as the message of the Critique of Judgment.81 It was the
so-called “dynamic” physics of the former, though not its rigid stipula-
tion for “proper” science, that caught the attention of natural scientists
in the 1790s. With regard to the Critique of Judgment, McLaughlin
77
“How does one explain the explosive development of those six to seven years from 1789 to 1795/6 – a
development initially away from Kant, which at once led to the situation that Kant was virtually totally
dropped from discussion within it, and then in reaction against Fichte as well, which began already in
the year 1794?” (Dieter Henrich, Konstellationen: Probleme und Debatten am Ursprung der idealistischen
Philosophie (1789–1795) [Stuttgart: Klett-Cotta, 1991], 223).
78
Stanley L. Jaki, introduction to Immanuel Kant, Universal Natural History and Theory of the Heavens,
trans. Stanley L. Jaki (Edinburgh: Scottish Academic Press, 1981), 1–76.
79
See my “‘Method’ vs ‘Manner’? – Kant’s Critique of Herder’s Ideen in Light of the Epoch of Science,
1790–1820,” Herder Yearbook, ed. Hans Adler and Wulf Koepke (Stuttgart: Metzler, 1998), 1–25.
80
“Kant’s contemporaries . . . knew from the obscurity of such doctrines as that of the ‘inner sense,’ or
the ‘a priori schematism’ and from the ambiguity of Kant’s position on the ‘thing-in-itself,’ that a reform
of his premises was needed. Moreover, Kant’s official position was complex (and confused) enough to
make each believe, at least in the beginning, that in adopting his respective line of thought he was only
making explicit the real position of the master” (George di Giovanni, “Kant’s Metaphysics of Nature
and Schelling’s Ideas for a Philosophy of Nature,” Journal of the History of Philosophy 17, no. 2 [April
1979]: 206).
81
Peter McLaughlin, “Kants Organismusbegriff in der Kritik der Urteilskraft,” in Philosophie des
Organischen in der Goethezeit, ed. Kai Torsten Kanz (Stuttgart: Steiner, 1994), 100–110.
correctly notes that it posed uncongenial objections to the inquiry into

life forces that was the driving concern of physiological investigation in
the era.
German naturalists sought to grasp the implications of philosophy of science
for their practices; more specifically, they sought rules for integrating con-
crete empirical results into a warranted theory. We can distinguish at least
three levels of articulation: first, philosophy of science, as general epistemo-
logical warrant of scientific claims; second, philosophy of nature, as the
systemic unity of nature in its empirical lawfulness; and finally, determinate
theoretical frames in which specific research fields (Fachgebiete) organized
their experimental-observational results. For the era after 1795, those who
theorized about medicine as a science had to wrestle with Kant’s critical
characterization of the relationship between the rational and the empirical,
between the a priori and the a posteriori. Physiology had to be an empirical
science, but what would raise it to the level of valid knowledge? To be sure,
Kant “was explicitly and frequently invoked in the physiological texts,” but
was the critical philosophy cautionary for the practical pursuit of medical
knowledge, testing it against ideals of systematicity and certainty, or was the
critical philosophy, in Moses Mendelssohn’s striking phrase, alles-zermal-
mende82 – that is, did it crush the very possibility of “scientific” medicine?83
Kant’s rigorous stipulations of systematicity and apodictic certainty left
physiologists and medical professionals generally feeling inadequate – caught,
in the words of one of the key spokesmen of the profession, “between system-
fever and clueless empiricism [zwischen Systemsucht und zügelloser
Empirie].”84 The challenge of Kant’s rigorous notion of “proper science”
82
Moses Mendelssohn, Morgenstunden oder Vorlesungen über das Dasein Gottes, in Gesammelte Schriften:
Jubiläumsausgabe, ed. Fritz Bamberger et al., vol. 3, pt. 2 (Stuttgart-Bad Cannstatt: Frommann-
Holzboog, 1971–98), 3.
83
In fact, Lohff makes clear, “never was the complete philosophical system of doctrine assimilated.”
Instead, only “philosophical slants [Tendenzen] and turns of phrase” got taken up (Die Suche nach der
Wissenschaftlichkeit, 36). Similarly, Reinhard Mocek observes: “Reil and his medical colleagues of the
time had by no means studied this philosophy in all its ramifications, that is in a sense developed
themselves into philosophical competitors.” Instead they oriented themselves to a few lines of thought
and emphases, “categories, high-profile definitions, fundamental determinations” (Johann Christian Reil
[1759–1813]: Das Problem des Übergangs von der Spätaufklärung zur Romantik in Biologie und Medizin
in Deutschland [Frankfurt: Lang, 1995], 101). For one thing, Kant’s very terminology was daunting,
presenting, in Lohff’s delicate phrasing, “difficulties for comprehension [Verständnisschwierigkeiten]” so
substantial that lexica were required to render his technical terms accessible for ordinary readers (Die
Suche nach der Wissenschaftlichkeit, 42). The most prominent such lexicon of the day was published by
Carl Christian Schmid. The same author tried to establish what Kant’s critical philosophy prescribed for
physiology – in three volumes: Physiologie philosophische bearbeitet (1798).
84
August Wilhelm Hecker, editor’s introduction (Einleitungsartikel), Journal der Erfindungen, Theorien
und Widersprüche 7 (1798): 1–22; cited by Lohff, Die Suche nach der Wissenschaftlichkeit, 171.
492 J. H. Zammito
discomfited them concerning both their public authority as well as their

fundamental mission. Kant offered all too little guidance for the empirical
practices of inquiry and all too scrupulous a standard for construing their
results. Thus, both from a historicist and from a presentist vantage, there are
grounds to find Kant’s notion of “proper science” interfering too much with
his conception of ongoing empirical natural science to be of much use either
for his contemporaries or for us.85
85
See my “Should Kant Have Abandoned the ‘Daring Adventure of Reason’? The Interest of
Contemporary Naturalism in the Historicization of Nature in Kant and Idealist Naturphilosophie,” in
International Yearbook of German Idealism, vol. 8: Philosophy and Science, ed. Jürgen Stolzenberg and
Fred Rush (Berlin: de Gruyter, 2010; actually 2012), 130–64; but in rebuttal, of course, see van den
Berg, Kant on Proper Science, and the plethora of works celebrating Kant’s influence, e.g., Michael
Friedman and Alfred Nordmann, eds., The Kantian Legacy in Nineteenth-Century Science (Cambridge:
MIT Press, 2006).
22
From General to Special Metaphysics
of Nature
Michael Bennett McNulty and Marius Stan
It is telling that Kant’s Metaphysical Foundations of Natural Science (1786)

coincided with the centenary of Newton’s Philosophiæ Naturalis Principia
Mathematica (1686–1687). In the preface of the Metaphysical Foundations,
Kant praises Newton’s mathematical method and quotes from his forerun-
ner’s opus: “Geometry is proud of the fact that with so little derived from without
it is able to produce so much” (MFS 4:478). Yet Kant objects to an aspect of
Newton’s approach, namely the latter’s opposition to the incursion of
metaphysics within his natural philosophy. Thus Newton wrote the follow-
ing in a draft preface to the Principia:
Natural philosophy should be founded not on metaphysical opinions, but on

its own principles. . . . In all philosophy we must begin from phenomena and
M. B. McNulty (*)
Department of Philosophy, University of Minnesota, Twin Cities,
Minneapolis, USA
e-mail: mcnu0074@umn.edu
M. Stan
Department of Philosophy, Boston College, Chestnut Hill, USA
e-mail: marius.stan@bc.edu

494 M. B. McNulty and M. Stan
admit no principles of things, no causes, no explanations except those which

are established through phenomena.1
Kant protests that such a solemn desire to avoid metaphysics is ultimately

self-defeating. The very application of mathematics to nature at the heart
of Newton’s physics presupposes metaphysical principles (MFS 4:475).2
Moreover, Kant thinks of natural science as an appendage of metaphysics.
Thus in the Metaphysical Foundations, he explains how the “special meta-
physics of body,” or “rational physics,” emerges from the general metaphysics
of the Critique of Pure Reason. By doing so, he purports to place natural
science upon a certain, evident, and firm foundation.
In this chapter, we detail the transition from the general metaphysical picture
developed in the Transcendental Analytic of Kant’s first Critique to the special
metaphysics of body in his Metaphysical Foundations. In “Nature and its
Metaphysics,” we explain Kant’s account of nature and its presuppositions in
order to clarify his conception of natural science. In “Matter, Motion, and
Apriority,” we examine the particulars of Kant’s transition to rational physics in
the Metaphysical Foundations. We especially concentrate on both the concept of
matter, which serves as the book’s cornerstone, and the specific role that
metaphysics plays in grounding physics. Finally, in “Fruits and Limits,” we
present some of the positive results of Kant’s physics and outline the limitations
of his top-down, metaphysical approach to natural science. This section con-
cludes by situating Kant’s achievements within the broader context of eight-
eenth-century mechanics beyond Newton.
Nature and its metaphysics

Kant opens the Metaphysical Foundations by defining its object of study –
nature – and by distinguishing two senses of the term3:
1
Isaac Newton, “Unpublished Preface to the Principia,” in The Principia: Mathematical Principles of
Natural Philosophy, trans. I. Bernard Cohen and Anne Whitman (Berkeley: University of California
Press, 1999), 53–54.
2
Moreover, Kant believes that the attempt to avoid metaphysics can leave one tacitly endorsing
improper opinions. For more on the specific, improper metaphysical assumptions of Kant’s natural
philosophic predecessors, see Konstantin Pollok, Kants “Metaphysische Anfangsgründe der
Naturwissenschaft”: Ein kritischer Kommentar (Hamburg: Meiner, 2001), 113–22.
3
For an in-depth discussion of the technical notion of nature in Kant’s theory (albeit one that differs
from our own on a few interpretive details), see Peter Plaaß, Kant’s Theory of Natural Science, trans.
Alfred E. Miller and Maria G. Miller (Dordrecht: Kluwer, 1994), 216–28.
If the word nature is taken simply in its formal meaning, where it means the first
inner principle of all that belongs to the existence of a thing, then there can be as
many different natural sciences as there are specifically different things, each of
which must contain its own peculiar inner principle of the determinations
belonging to its existence. But nature is also taken otherwise in its material
meaning, not as a constitution, but as the sum total of all things, insofar as they
can be objects of our senses, and thus also of experience. (MFS 4:467)4
Formal nature corresponds to the adjectival use of the term. In this sense, for
any type of thing in the world, we may speak of and investigate its nature.
The chemist may study the nature of heat or chemical dissolution, while the
biologist may explore the nature of, say, Escherichia coli. As Kant notes in the
Critique of Pure Reason, a description of the formal nature of a sort of thing
involves an account of its causal powers and relations (A418–19n/B446n).5
So we may say that it belongs to the nature of sodium nitrate to dissolve in
water or of heat to flow to colder bodies.6
Material nature, conversely, corresponds to the substantival use of the
term and refers to the whole of objects of the world. Nature, according to this
employment, does not refer to any particular sort of thing, but rather to the
entirety of all things. Yet Kant is clear that nature is no mere aggregate of
things: material nature, in virtue of being nature, also involves things in
“thoroughgoing connection through an inner principle of causality” (A419n/
B446n). Thus Kant distinguishes “nature” from “the world.” The world is
the mere agglomeration of objects, the “mathematical whole of all appear-
ances,” whereas nature is the aggregate of appearances “insofar as it is
considered as a dynamic whole” and concerns the “unity in the existence
of appearances” (A418–19/B446–47). The world and nature are distin-
guished along the same lines as mathematical and dynamical syntheses
(B201–2). Mathematical syntheses are those that combine homogeneous
4
This distinction regularly appears in the critical corpus: see also B163–65, A418–19n/B446n. In the
Prolegomena, Kant distinguishes formal and material nature somewhat differently (see Pro 4:294–96,
318). Their material nature is understood more literally as the matter of nature – that is, intuitions –
which are made possible by the Transcendental Aesthetic. Conversely, formal nature is described as the
form of nature – the principles that make possible dynamic connection.
5
In a footnote to the preceding passage, Kant distinguishes between formal natures and essences (MFS
4:467n). As we understand this distinction, formal natures refer to the dynamical (causal) qualities of a
kind, whereas essences are non-causal. Hence, Kant claims that geometrical figures have an essence, but
not a nature.
6
As Pollok explains, this investigation of the specific diversity of matters – the variety of formal natures –
does not fall under the purview of the physics and the metaphysical foundations of natural science,
according to Kant (Kants “Metaphysische Anfangsgründe der Naturwissenschaft,” 46). See the General
Remark to Dynamics (MFS 4:523–35).
parts – those belonging to a common genus – into whole quantities (A560/

B588).7 Hence a world-concept, generally, is the quantitative whole of all
appearances insofar as they are homogeneous. One such world-concept is the
synthesis of all the homogeneous parts of space into a unified quantity.8 In
contrast, material nature is the dynamic totality of objects, where dynamic
syntheses combine inhomogeneous parts, paradigmatically appearances of
different sorts (A530/B558). Kant claims that the categories of relation
realize dynamic syntheses, so the combination of accident with substance
or effect with cause belongs to this class. Kant also explains that dynamical
syntheses concern “the combination of the existence of the manifold”
(B201), or the “existence of appearances” (A419/B447). So, for example,
the existence of an effect necessitates the existence of some cause, and the
existence of an accident necessitates the existence of a substance.
Since the categories of relation (substance, cause, and community) concern
dynamical syntheses, so too do the associated principles of the understanding,
namely the Analogies of Experience; thus Kant writes, “These principles have
the peculiarity that they do not concern the appearances and the synthesis of
their empirical intuition, but merely their existence and their relation to one
another with regard to this their existence” (A178/B220). Whereas the Axioms
of Intuition and the Anticipations of Perception, which present the principles
corresponding to the categories of quantity and quality, deal with the mathe-
matical syntheses of appearances, the Analogies deal with relations of existence
among appearances, that is, their dynamical syntheses (A161–62/B201–2).
Thus the First Analogy shows that through all change something – substance
– persists and that the change is a mere modification of its way of existing.9 The
Second Analogy demonstrates that any alteration to or coming-to-be of an
object requires the existence of a temporally preceding or coeval cause.
According to the Third Analogy, all things that simultaneously exist stand in
dynamical community, that is, they are constantly causally interacting.
Hence material nature concerns not the mere aggregate of appearances,
but appearances insofar as they stand in dynamical unity (interconnected by
means of the categories of relation). Nature, insofar as it involves such
relations, is made possible by the Analogies of Experience:
7
Quantities, for Kant, are syntheses of homogeneous units (A242/B300).
8
Such a concept, however, transcends the possibility of experience (it is an idea of the unconditioned),
and it therefore gives rise to intractable paradoxes, as Kant explains in the Antinomies of the first
Critique.
9
The A-edition formulation of the First Analogy emphasizes this point. That which changes – ceases to
exist – is a “mere determination, i.e., a way in which the object exists” (A182).
By nature (in the empirical sense) we understand the combination of appearances

as regards their existence, in accordance with necessary rules, i.e., in accordance
with laws. There are therefore certain laws, and indeed a priori, which first make a
nature possible; the empirical laws can only obtain and be found by means of
experience, and indeed in accord with its original laws, in accordance with which
experience itself first becomes possible. . . . Thus together [the Analogies] say: All
appearances lie in one nature, and must lie therein, since without this a priori unity
no unity of experience, thus also no determination of the objects in it, would be
possible. (A216/B263)
As Kant claims in the passage, the Analogies are a priori because they are
conditions for the possibility of experience of objects. To be an object of
experience involves standing in dynamical relations with the whole of
appearances.10 The Analogies therefore make nature – the entirety of objects
in thoroughgoing dynamical interconnection – possible; they guarantee that
all appearances are interrelated according to the categories of relation.
Moreover, the Analogies comprise the first, most basic results of the meta-
physics of material nature. They describe the very conditions of something
belonging to material nature and hence codify the necessary conditions
applying to all objects of nature. Nevertheless, although the Analogies
make possible the dynamical syntheses of appearances, they do not yield all
of the information about such interconnections of existence. For example,
the Second Analogy requires that there exists some cause for the booming
noise that just occurred outside the window. However, it does not entail
anything about the specific content of this cause, whether it was, say, thunder
resulting from a nearby lightning strike or the explosion of a celebratory
firework. The particular laws governing the empirical, dynamical connections
of appearances are discoverable only through experience.
The general dynamical interconnectedness of things prescribed by the
Analogies is among the claims belonging to the branch of metaphysics
dubbed by Kant “immanent physiology” (or, alternatively, the general meta-
physics of nature).11 Rational physiology, in general, “considers nature, i.e.,
the sum total of given objects” according to its a priori rules, and its
10
See also the A-edition version of the general principle of the Analogies of Experience, which sums up
many of our points nicely: “As regards their existence, all appearances stand a priori under rules of the
determination of their relation to each other in one time” (A176).
11
For more on the position of natural science in the Kant’s architectonic, see Plaaß, Kant’s Theory of
Natural Science, 207–10.
immanent specification examines objects of possible sensible experience.12

Immanent physiology hence “considers nature as the sum total of all objects
of the senses, thus considers it as it is given to us, but only in accordance with
a priori conditions, under which it can be given to us in general” (A845–46/
B873–74). The a priori laws that apply to all objects of the senses are those
that are necessary for the possibility of experiencing objects. Therefore, the
laws presented in the Analytic of Principles collectively exhaust immanent
physiology.13 This doctrine is also called the general metaphysics of nature
because it contains the most general, fundamental a priori speculative judg-
ments about nature.
Kant claims in both the first Critique and the Metaphysical Foundations that
material nature – and, along the same lines, the general metaphysics of nature –
admits of division according to the distinction between inner and outer sense:
Now nature, taken in [the material] meaning of the word, has two principal
parts, in accordance with the principal division of our senses, where the one
contains the objects of the outer senses, the other the object of inner sense. In
this meaning, therefore, a twofold doctrine of nature is possible, the doctrine of
body and the doctrine of soul, where the first considers extended nature, the
second thinking nature. (MFS 4:467)
Thus Kant’s distinction between two sorts of objects of nature (between

those of outer sense and those of inner sense) corresponds to the species of
immanent physiology: the doctrine of corporeal nature, or “physics,” and
the doctrine of thinking nature, or “psychology” (A846/B874). As their
genus, rational physiology, is a type of metaphysics, these sciences too seek a
priori judgments applying to their objects.14
The kinds of immanent physiology (including physics, the primary object
of our present concern) can therefore be specified according to material or
formal nature.15 The sum total of appearances (material nature) can be split
into those of the inner and those of the outer sense. Alternatively, we can
12
Transcendent rational physiology examines objects beyond the possibility of experience – the world-
whole and God – though these doctrines are ultimately dialectical.
13
Indeed, Kant explicitly calls the principles of the understanding “physiological” (Pro 4:303).
14
Empirical psychology and physics inhabit a very different branch of the hierarchy of doctrines. They
are types of applied philosophy (A848/B876), which concern rational cognition from empirical princi-
ples (A840/B868).
15
In contrast, Plaaß (Kant’s Theory of Natural Science, 218) claims that the formal sense of nature is
primary for natural science, while Pollok (Kants “Metaphysische Anfangsgründe der Naturwissenschaft,”
46–47) argues that the material sense is the most fundamental.
conceive of physics and psychology as studying the principles concerning the

existence of objects falling under particular concepts, that is, the formal
nature of bodies or thinking things, respectively.
At this juncture we have shown that natural sciences, for Kant, are kinds of
metaphysical doctrine. In virtue of this heritage, natural sciences pursue a
priori knowledge of their objects. Unlike immanent physiology (or the
general metaphysics of nature), which consists of principles regarding the
possibility of experiencing any object whatsoever, the special metaphysical
doctrines of physics and psychology seek a priori principles regarding their
respective objects: those of outer and inner sense. Additionally, since the
objects of nature are dynamically interconnected, physics and psychology
aspire to the laws governing the relations of existence among their objects.
Whereas the general metaphysics of nature consists of the most general
principles of the relations among appearances as such, physics and psychol-
ogy seek more restricted principles that apply especially to the appearances of,
respectively, outer and inner sense. So whereas the Second Analogy (a law of
immanent physiology) concerns the causal lawfulness of appearances in
general, physics involves laws governing the actual, particular causal relations
among extended appearances.
Matter, motion, and apriority

In the preface to the Metaphysical Foundations, Kant explains that a systematic
doctrine achieving a priori knowledge of nature is a “proper” natural science and
that, in contrast, one that contains laws justified by experience is only an
“improperly so-called” natural science (MFS 4:468).16 Thus the sorts of imma-
nent physiology – physics and psychology – are proper sciences insofar as they
are capable of producing a priori cognitions of their objects. In this section, we
explain how it is possible to come to a priori knowledge in physics. (Psychology,
in contrast, is incapable of a priori cognitions, as we touch on below.)
Kant conceives of such a proper science as a special metaphysical doctrine
that rests on “the transcendental part of the metaphysics of nature” (MFS 4:469).
16
Chemistry is such an improper science, for Kant (MFS 4:468, 470–71). For varying views on the
status of this claim, see Martin Carrier, “Kant’s Theory of Matter and His Views on Chemistry,” in Kant
and the Sciences, ed. Eric Watkins (Oxford: Oxford University Press, 2001), 205–30; Michael Friedman,
Kant’s Construction of Nature: A Reading of the “Metaphysical Foundations of Natural Science”
(Cambridge: Cambridge University Press, 2013), 234–58; and Michael Bennett McNulty, “Kant on
Chemistry and the Application of Mathematics in Natural Science,” Kantian Review 19, no. 3 (Nov.
2014), 393–418.
This transcendental part of the metaphysics of nature is just the general meta-
physics of nature, or immanent physiology, that is described above. It is con-
cerned with the laws governing the relations of the existence, or dynamical
syntheses, of objects as such. The objects of the general metaphysics of nature are
“undetermined” in the sense that they are not further specified with respect to
additional concepts.17 That is, the laws of this metaphysical doctrine apply to all
objects regardless of whatever further concepts they fall under. This general
metaphysical doctrine is the foundation for the more specific metaphysics of
body and of mind – physics and psychology – which examine objects that fall
under particular empirical concepts. Thus in physics we are concerned with
achieving a priori knowledge of objects that fall under the empirical concept of
matter. Matter, at this juncture in the Metaphysical Foundations, simply means
an object of outer sense. Throughout the chapters of the book, Kant describes
further features of the concept, but when he introduces the concept in the
preface, it is somewhat austere.
Kant claims not only that the special metaphysical doctrines of nature are
species of general metaphysics but additionally that they bear a kind of
dependence on the latter. That is, special metaphysics of nature involves an
application of the “transcendental principles” of general metaphysics to “the
two species of objects of our senses” (MFS 4:470). The claimed linkage of
physics and metaphysics, however, gives rise to three interpretative issues, the
resolution of which is essential to Kant’s account of the proper natural
science of physics, which we consider in turn.
1) How is it possible to achieve a priori cognition of concepts determined by

an empirical concept, such as matter?
2) Relatedly, in what sense is such cognition a priori?
3) What, precisely, is the connection between the general metaphysics of
nature and the special metaphysical doctrine of body?
Motion and mathematics
Kant not only holds metaphysics to be a necessary component of natural

science, but also claims that mathematics is required to achieve a priori
natural scientific knowledge. He therefore famously (or notoriously) insists
17
Determination is a technical notion that refers to the application of a concept to an object (A261/
B317, A571/B599).
that “in any special doctrine of nature there can be only as much proper
science as there is mathematics therein” (MFS 4:470). Briefly, and glossing
over various textual and conceptual difficulties, Kant’s argument proceeds
along the following lines. As we explain above, in special metaphysics of
nature, one is concerned with objects determined by some empirical concept:
matter or thinking being. To cognize things falling under an empirical
concept, both the concept and an intuition are required. To cognize them
a priori, an a priori intuition is required. It is just mathematics that provides
such a priori intuitions (A712–38/B740–66) and therefore makes possible
proper natural science.18
Two claims rest at the theoretical basis of the mathematizability of physics.
First, the fundamental determination of matter is its movability. Second,
motion admits of mathematical construction. Together, these theses make
possible the application of mathematics to the doctrine of body.
We first reflect on the connection between matter and motion. Kant writes
that “the basic determination of something that is to be an object of the outer
senses had to be motion, because only thereby can these senses be affected”
(MFS 4:476). One may, of course, conceive of other possibilities: for
example, for Descartes, the essence of matter is its extension, or filling of
space. Why then does motion possess this foundational character?
Physics is the proper science of the objects of outer sense, hence it seeks a
priori knowledge of them. We explain this below in more detail, but the a
priori cognitions of physics are the conditions for the possibility of experien-
cing objects of the outer sense. Here in the Metaphysical Foundations, Kant is
thus claiming that motion is a condition of the possibility of the experience
of matter, and hence a condition of the possibility of outer experience as
such. Only through motion can outer objects affect us, so motion is a
prerequisite for our experience of them. For this reason, motion is the
basic determination of matter.19
Throughout the Metaphysical Foundations, matter is conceptualized as the
movable in space, because motion is a constructible concept. In the
18
For varying readings of this argument, see Friedman, Kant’s Construction of Nature, 26–33; and
Michael Bennett McNulty, “Chemistry in Kant’s Opus Postumum,” HOPOS: The Journal of the Society
for the History of Philosophy of Science 6, no. 1 (May 2016): 79–81.
19
For interpretations of this claim, see Eric Watkins, “The Argumentative Structure of Kant’s
Metaphysical Foundations of Natural Science,” Journal of the History of Philosophy 36, no. 4 (Oct.
1998): 578–82; Friedman, Kant’s Construction of Nature, 34–46; and Henny Blomme, “Kant’s
Conception of Chemistry in the Danziger Physik,” in Reading Kant’s Lectures, ed. Robert Clewis
(Berlin: De Gruyter, 2015), 495–97. Watkins analyzes Kant’s assertion that all matter is movable,
whereas Friedman and Blomme focus on his appeal to empirical affection.
Phoronomy, Kant shows that from given motions, conceived of as line

segments, one can mathematically construct their composite motion through
producing the diagonal of the parallelogram that arises from the given
motions (MFS 4:491–93). Whatever is mathematizable of matter – and
hence that which is a priori about the concept – must therefore be linked
with motion. It is hence in this vein that Kant writes, “natural science,
therefore, is either a pure or applied doctrine of motion” (MFS 4:477).
Apriority
This, however, raises a vexing question about Kant’s approach to natural science –
namely, in what sense could the principles of physics presented in the Metaphysical
Foundations be a priori? After all, the central concepts of physics, to wit, those of
matter and motion, are empirical. Judgments about such empirical concepts, the
thought would go, cannot be a priori because they are, unlike the principles of
general metaphysics, not necessary for the possibility of experience. There are
roughly main four schools of thought on this topic in the literature.
1) Full-Blown A Priori. According to this interpretation, the propositions

of the Metaphysical Foundations – the principles of rational physics – are,
despite their initial appearance, necessary for the possibility of experi-
ence as such (similar to those of the understanding from the Critique of
Pure Reason). Such a view falls out of Eckart Förster’s interpretation,
according to which the Metaphysical Foundations was meant to complete
the proof of the objective validity of the categories.20 Crucial for this
account is the Refutation of Idealism added in the second edition of the
Critique (B274–79).21 In this section, Kant explains that, according to
his transcendental idealism but in contrast to the material idealisms of
Descartes and Berkeley, outer experience is necessary for the possibility
of inner experience. Thus, the line of thought would go, conditions for
the possibility of outer experience must likewise be conditions of the
possibility of experience in general.
2) Necessary for Science. According to this interpretation, the propositions
of the Metaphysical Foundations – the principles of rational physics – are
20
Eckart Förster, Kant’s Final Synthesis: An Essay on the “Opus postumum” (Cambridge: Harvard
21
Friedman also sees a connection with the Refutation of Idealism, though he denies the above-
described conception of apriority (Kant’s Construction of Nature, 3–11).
preconditions for the intelligibility of natural science. They are a priori in

the sense that they are constitutive of the object of physics and give
empirical meaning to the terms of the theory. Michael Friedman suggests
this interpretation by claiming, for instance, that the apriority of Kant’s
mechanical laws consists in their making possible a universal measurement
procedure for the quantity of matter in Newtonian physics.22 The thought
is that the mechanical laws are thereby a priori insofar as any empirical
determination of quantity of matter depends for its very meaning upon
these laws. (That said, nowhere does Friedman explicitly endorse this
account of apriority.)
3) A Priori Content, Empirical Proof of Reality. Peter Plaaß claims that the
concept of matter is empirical only in the sense that its objective reality
cannot be proven a priori.23 That is, to confirm that matter exists, one
requires experience. Nevertheless, the content of the concept is determined
a priori by the process of “metaphysical construction,” to which Kant
allegedly alludes in the preface to the Metaphysical Foundations (MFS
4:473). The principles of the book are therefore a priori insofar as they
follow from this a priori content.
4) A Priori Procedures. According to Philip Kitcher, the principles of the
Metaphysical Foundations can be understood as neither fully a priori nor
empirical, despite the fact that they are about objects falling under the
empirical concept of matter.24 For Kitcher, the concept of matter is
through-and-through empirical, and what makes the principles of the
Metaphysical Foundations a priori is that no further empirical content is
necessary for their derivation.25 Rather, “a priori procedures” – such as
22
Friedman, Kant’s Construction of Nature, 569. For an elaboration of this point, see Jeremy Heis,
review of Kant’s Construction of Nature, by Michael Friedman, Philosophical Review 123, no. 3 (July
2014): 342–54.
23
Plaaß, Kant’s Theory of Natural Science, 285.
24
Philip Kitcher, “Kant’s Philosophy of Science,” Midwest Studies in Philosophy 8, no. 1 (Sept. 1983):
387–407.
25
In Kant’s Theory of Science (Princeton: Princeton University Press, 1978), Gordon G. Brittan, Jr.
conceptualizes the situation according to the semantics of modal logic. The principles of the under-
standing are necessary for the class of worlds that includes beings with spatiotemporal intuitions and
discursive intellects. The propositions of the Metaphysical Foundations hold in a subset of such worlds –
those in which the concept of matter obtains. Although this understanding is anachronistic, it shares an
insight of Kitcher’s interpretation: once one fixes the empirical concept of matter, the propositions of the
Foundations hold necessarily.
conceptual analysis and mathematical construction – are utilized to derive

the principles of rational physics.26
Plaaß’s and Kitcher’s proposals are quite similar, but they bear some subtle
differences.27 In particular, for Kitcher but not for Plaaß, the content of the
concept of matter is empirical.
On these grounds, we prefer the Kitcherian account of apriority in the
Metaphysical Foundations. Kant consistently refers to matter and motion as
empirical concepts, without a hint of the mixed status that Plaaß describes.
However, by applying a priori procedures to the concept, we derive the a
priori principles of rational physics. In service of this aim, Kant first provides
a “complete analysis” (MFS 4:472) of the concept of matter – thus present-
ing the determinations that belong to the concept – and then brings its
determinations under the four headings of the categories (MFS 4:475). That
is, he establishes the quantity, quality, relation, and modality of matter.28
Subsequently, in rational physics we utilize mathematical construction to
produce further a priori cognitions about the objects falling under this
concept (see “Fruits and Limits”).
The judgments resulting from such a priori procedures are not free from
any empirical content – the concept of matter saturates the whole of physics
– but they are nevertheless a priori. Kant distinguishes a priori and pure
judgments in the introduction of the Critique of Pure Reason: pure cognitions
are those in which “nothing empirical is intermixed,” whereas a priori,
impure cognitions can involve empirical concepts (B3). For example, the
judgment “Every alteration has its cause” is impure, insofar as it involves the
empirical concept of alteration, although it is a priori. The propositions of
the Metaphysical Foundations, in virtue of regarding the empirical concept of
matter, are likewise a priori but impure. Indeed, Kant’s example concerning
alteration is quite apt in the present context. In the Transcendental Aesthetic,
Kant claims that, just as the pure intuition of space grounds geometry, time
26
Kitcher, “Kant’s Philosophy of Science,” 394. Kitcher calls the principles that result from a priori
procedures applied to an empirical concept “quasi a priori.” However, as we show below, it is natural to
think of these principles as thoroughly a priori, albeit not pure.
27
Charles Parsons draws this connection and compares the positions (“Remarks on Pure Natural
Science,” in Self and Nature in Kant’s Philosophy, ed. Allen W. Wood [Ithaca: Cornell University
Press, 1984], 218–20).
28
Pollok’s view is similar to ours. He argues that there are analytic and synthetic projects in the
Metaphysical Foundations: first Kant analyzes the concept of matter, then synthetically develops its
determinations according to the categories (Kants “Metaphysische Anfangsgründe der Naturwissenschaft,”
111–13).
too rests at the foundation of an a priori science, one which, however,

requires the inclusion of the concept of alteration:
Here I add further that the concept of alteration and, with it, the concept of
motion (as alteration of place), is only possible through and in the representa-
tion of time – that if this representation were not a priori (inner) intuition, then
no concept, whatever it might be, could make comprehensible the possibility of
an alteration, i.e., of a combination of contradictorily opposed predicates (e.g.,
a thing’s being in a place and the not-being of the very same thing in the same
place) in one and the same object. Only in time can both contradictorily
opposed determinations in one thing be encountered, namely successively.
Our concept of time therefore explains the possibility of as much synthetic a
priori cognition as is presented by the general theory of motion, which is no less
fruitful. (B48–49)
The theory of motion referenced here is a priori but, due to its dependence
upon the empirical concept of alteration, impure. Physics, which rests on the
doctrine of motion, handles empirical concepts and is thus incapable of the
purity of the general metaphysics of nature. Kant puts this quite clearly in the
Prolegomena: “But indeed there is also much in [natural science] that is not
completely pure and independent of sources in experience, such as the concept
of motion, of impenetrability (on which the empirical concept of matter is
based), of inertia, among others, so that it cannot be called completely pure
natural science” (Pro 4:295). So the propositions of the special doctrine of
matter are a priori in the sense that they are derived through a priori procedures
and require no further empirical content beyond these concepts.29
29
That said, there are some complications to our picture: namely, Kant refers to the pure doctrine of
motion (cf. MFS 4:477, 496) and claims that a proper natural science requires a pure core (MFS 4:468–
70). We suggest the following interpretation of these claims. As Kant explains, the pure part of proper
natural science contains its mathematical principles. However, motion, in virtue of being an empirical
concept, cannot be mathematically constructible. So, we contend that the pure part of natural science
consists of the mathematical principles for constructing line segments at the basis of Kant’s kinematics
developed in Proposition 1 of the Phoronomy (MFS 4:490–93). The construction of such line
segments makes use of no empirical concepts and is hence pure. To say, however, that these line
segments represent motions, and that the constructed line segment represents a composite motion,
introduces an empirical concept. The doctrine of motion, as a whole, is not pure, although its basis – the
mathematical construction of line segments – is. Friedman helpfully distinguishes mathematical and
empirical motion (Kant’s Construction of Nature, 83–90). Mathematical motion, the motion of a mere
mathematical point, is the topic of the pure doctrine of motion at the basis of the proper natural science
of physics.
The application of metaphysics
As we saw above, Kant conceives of general metaphysics of nature as an

essential propaedeutic and foundation for the special metaphysics of body.
Why is this, and in what precise manner are these two doctrines linked?
Fortunately, Kant is clear about the answer to the second question in the
preface to the Metaphysical Foundations. He writes, “All determinations of
the general concept of matter in general must be able to be brought under the
four classes of [pure concepts of the understanding], those of quantity, of
quality, of relation, and finally of modality,” and, furthermore he claims that
this process of determining the concept of matter via the categories exhausts
“all that may either be thought a priori in this concept, or presented in
mathematical construction, or given as a determinate object of experience”
(MFS 4:475–76). Thus, in each chapter of the Metaphysical Foundations, one
categorial determination of matter is considered. In the Phoronomy,
Dynamics, Mechanics, and Phenomenology, Kant respectively considers
the quantity, quality, relation, and modality of matter (conceptualized as
that which moves in space). This process purportedly yields all of the a priori
knowledge that may be had of the concept, and thus the completeness of
rational physics is guaranteed.
It is natural, though mistaken, to think of the connection between the
categories of the first Critique and the propositions of the Metaphysical
Foundations to be mediated through an instantiation of the pure principles
of the understanding. This thought is reasonable, given the similarity of some
of the respective principles. The principle of the B-edition version of the First
Analogy, which governs the application of the category of substance to
experience, reads: “In all change of appearances substance persists, and its
quantum is neither increased nor diminished in nature” (B224). Its counter-
part, the first mechanical law, which governs the application of the category
of substance to matter, asserts: “In all changes of corporeal nature the total
quantity of matter remains the same, neither increased nor diminished”
(MFS 4:541). One may sensibly think that the latter can be derived from
the former through a simple universal instantiation: the First Analogy applies
to all objects of experience, and matter is an object of experience, so the first
mechanical law – an instance of the First Analogy – applies to matter. There
are, however, fairly straightforward difficulties facing this account, as is
shown by Eric Watkins.30 For instance, although the relation between the
30
Watkins, “Argumentative Structure,” 571–77.
First Analogy and the first mechanical law resembles the universal-instance
relation, this is not so for other principles that are supposed to correspond.
Take, for example, the Third Analogy – “All substances, insofar as they can
be perceived in space as simultaneous, are in thoroughgoing interaction”
(B256) – and the third mechanical law – “In all communication of motion,
action and reaction are always equal to one another” (MFS 4:544). Although
both regard community in their respective domains, they clearly do not stand
in the universal-instance relation. Furthermore, Kant explicitly denies that
there could be such a direct deduction from the principles of the under-
standing to those of physics (B165).
In contrast, we understand the connection between the categories and
principles of physics along the lines of Watkins’s “Transcendental Argument
Interpretation.”31 In the Transcendental Deduction, Kant shows that in order
for something to be an object of possible experience the categories must apply to
it. Subsequently, in the Analytic of Principles, he explains how the categories
apply to constitute experience. In the Metaphysical Foundations, Kant is inter-
ested to show that matter, now conceived of as the movable in space, is a possible
object of experience. To prove this fact, he must therefore demonstrate that and
how the categories apply to objects falling under the concept. This is precisely
the aforementioned task of the chapters of the Metaphysical Foundations: in each,
Kant demonstrates that and how the category at issue applies to matter. For
example, the aforementioned phoronomical construction of composite motions
establishes that motions are the quantity of matter. The proposition of the
Phoronomy shows that motions are homogeneous and provides a method for
combining them into another homogeneous quantity (such are criteria for being
a quantity). Collectively the chapters thus demonstrate that matter is a possible
object of experience and make explicit the conditions of the possibility of
experiencing matter. These conditions furthermore constitute the whole of a
priori facts about matter. As Kant writes at the beginning of the Metaphysical
Foundations, in natural science, we seek a priori knowledge concerning objects
determined by some empirical concept; that is, we are interested in that knowl-
edge belonging to the object’s possibility (MFS 4:470). Only the conditions for
the possibility of experience of matter are a priori. Any other judgments about
matter require additional empirical content. Kant is confident that the
Metaphysical Foundations constitutes the complete a priori science of body, for
all that can be known a priori of matter are the conditions for the possibility of its
experience issuing from matter’s quantity, quality, relation, and modality.
31
Ibid., 577–87.
Fruits and limits

In the body of the Metaphysical Foundations, Kant thus successively presents
the categorical determinations of the concept of matter, thereby showing that
it is an object of possible experience. Further, he extracts substantive physical
theses from this metaphysical account. In the Phoronomy, Kant explains that
motion is the quantity of matter, and he presents a construction procedure
for combining motions into a composite motion (an analogue to the paral-
lelogram of forces).32 In the Dynamics, Kant demonstrates that the quality of
matter is its filling of space. In the course of this argument, he concludes that
matter everywhere fills space (the universe is a plenum) but to different
degrees in different objects. Kant thus rejects the so-called mechanical
philosophy – according to which the universe consists of impenetrable
material corpuscles in a background empty space – and maintains the
repulsive and attractive forces to be essential to the existence of matter.
Indeed, Kant argues that an attractive force acting immediately at a distance
is necessary for the possibility of matter’s very existence in space, thereby
legitimating Newton’s gravitational force in the face of Leibniz’s deeming it
miraculous.33 In the Mechanics, Kant claims that the relation of matter is its
communication of motion and presents his mechanical laws, including those
of the conservation of mass and equality of action and reaction. Finally, in
the Phenomenology, Kant defines possible, actual, and necessary motions in
his system. This completes Kant’s account of the inertial structure for
Newton’s Principia without resorting to “transcendentally real,” or absolute,
space and time as Newton had done.34
The dominant view, at least in North American scholarship, is that the
foundational value of Kant’s project lies in its placing Newton’s gravitation
theory on a solid footing, inter alia. While that is true and important, and
aspects of this grounding are on display in the preceding, fresh research is
now beginning to uncover another dimension to Kant’s foundational success.
We close this paper with an outline of this new evaluative dimension.
32
To show that motion is a quantity (according to Kant’s conception of quantities), he must demon-
strate that we can combine motions into greater motions. So, the construction procedure he develops in
the Phoronomic proposition is necessary and sufficient for showing that motion is the quantity of
matter.
33
See G. W. Leibniz and Samuel Clarke, Correspondence, ed. Roger Ariew (Indianapolis: Hackett,
2000), 17–18, 27.
34
For extended argument, see Michael Friedman, Kant and the Exact Sciences (Cambridge: Harvard
University Press, 1992), 136–64; and also his Kant’s Construction of Nature.
To see it, however, we must look beyond Newton’s Principia to modern

natural philosophy more broadly. By the late 1600s, a twofold rift had begun
to separate metaphysics and empirical mechanics. Specifically, seventeenth-
century doctrines of material substance omit – or simply fail to connect with –
the key element in the new science, namely dynamical laws of proven expla-
natory power. On the other hand, as mathematized mechanics grows, its basic
concepts become highly determinate and precise. That made it increasingly
problematic whether substance metaphysics, with its vaguely worded appeals
to matter, motion, action, cause, and the like, could still play a robust
foundational role for exact science. The early 1700s thus attest to a widening
gulf between metaphysics and advanced mechanical theory. Paying attention
to this divide offers a vantage point that sheds a new light on Kant’s achieve-
ment: we may read his project in the Metaphysical Foundations as a bridging
enterprise across a rift that kept widening in his time.
There are two aspects to this bridge-building. For one, Kant argues
that his metaphysics of matter entails dynamical laws essential to
Enlightenment theory, for example, those of the conservation of mass,
the equality of action and reaction, and the parallelogram of forces.35
Thus Kant succeeds where most early moderns had failed, namely, in
ensuring that the link between ontology and kinetics is inextricable and
robust. For another, this linkage is quite determinate, and thereby Kant
succeeds again in a novel respect. What enabled Kant to couple his
metaphysics with mathematized mechanics was his reliance on two
resources internal to transcendental idealism broadly construed. First,
the table of categories yield the aforementioned ultimate aspects of any
material thing: the phoronomic, dynamical, mechanical, and phenomen-
ological aspects.36 Second, from his injunction that proper science be
mathematized – by construction in intuition – he derived some very
determinate foundational results. Chief among them are that motion is
quantifiable as (vector-additive) linear velocity, resistance to compression
is linear-continuous, mass and inertia are conserved scalars, and interact-
ing bodies induce objective accelerations.
Seen in this light, Kant’s achievement cannot be overstated. His robust
explanatory coupling of force-substance metaphysics with a quantitative
theory of mechanics had not been matched since Descartes’s Principia
35
See Friedman, Kant’s Construction of Nature.
36
In our terms: fundamental kinematics, or degrees of freedom; material constitution; causally-efficient
mechanical agency; and true motion, namely observer-independent kinematic behavior.
Philosophiae. Standing as he did at a crossroads in the evolution of

mechanics, Kant’s synthesis is more advanced and empirically adequate
than Descartes’s was. And Kant’s sophisticated engagement with mechanics,
and fine nose for the foundational questions that really matter, have become
a model of philosophical reflection on exact science well into our time, as
Friedman and the scholarship he inspired have proven at length.
However, any metaphysics with substantive links to an empirical theory
becomes inevitably exposed to the latter’s shifting fortune and emerging
limitations. That fate befell Kant’s doctrine of body too.37 The received
wisdom is that his doctrine became hobbled by the forms of space and time
to which he had wedded it – in particular, by the Euclidean affine-metric
structure and the Galilean kinematics entailed by them. However, Kant’s
foundations have limits at yet another level, having to do with his metaphysics
of body – his notion of a material object, that is – rather than the space- and
time-framework that constrains it. Briefly put, for the purpose of grounding
classical mechanical theory, Kant’s “dynamical” theory of matter was neither
necessary nor sufficient. To see the former, consider a landmark success that
preceded Kant’s mature work. By 1749, Jean le Rond d’Alembert had worked
out a mechanics that reached father and deeper than Newton’s force-based
theory. D’Alembert was able to treat mechanical phenomena that Newton
could not tackle from his laws (e.g., constrained fall and the vibration of elastic
continuous strings); and to quantify certain celestial phenomena – the very
realm of the Principia – that had defeated Newton or were wholly unknown to
him (the precession of the equinoxes and terrestrial nutation, respectively).
Significantly for our topic, d’Alembert succeeded without Kant and Newton’s
key concepts. D’Alembert’s mechanics is a neo-Cartesian theory based in
matter and motion, not force and action, which it lacks.38 His theory likewise
is built up from entirely un-Newtonian laws, which d’Alembert couched in
kinematic terms, as statements about motions. And when he mentions force at
all (as in his reliance on vis viva to solve a range of problems) he means it
merely as a kinematic quantity – rather like Christiaan Huygens, another
Cartesian, before him – devoid of causal-metaphysical import. The laws that
bear the real explanatory-predictive load in his mechanics are the Principle of
37
This reading of Kant’s metaphysics of nature as closely related to a determinate mechanical theory is
not unanimous, though it is the mainstream. Gerd Buchdahl (Kant and the Dynamics of Reason: Essays on
the Structure of Kant’s Philosophy [Oxford: Blackwell, 1992]) and others have advocated for a much
looser link between the two.
38
By “matter” he means any rigid body, and by “motion” the accelerations impressed to, actually
acquired by a mechanical system, or “lost,” that is, suppressed by constraints. See Jean-Baptiste le Rond
d’Alembert, Traité de dynamique (Paris: David L’aîné, 1743).
Virtual Velocities, and a powerful general heuristic called nowadays

“d’Alembert’s Principle.” In Kant’s time, this theory outdid Newton’s
approach both in problem-solving success and in influence. Specifically,
Joseph-Louis Lagrange in 1788 made d’Alembert’s two laws above into the
core of his “analytic” mechanics, an axiomatic-deductive formulation that
bears important Kantian hallmarks of proper science, namely thorough math-
ematization and systematic unity. And yet, the d’Alembert-Lagrange laws are
not even remotely Newtonian. Thus, while Kant’s foundation may have been
enough for Newton’s seventeenth-century theory, it does not seem necessary
for fundamental science in his time.
Nor does it seem sufficient. Even apart from the advent of relativistic
physics, mechanics after 1785 went through several watershed changes. For
one, then-local theories, like linear elasticity, hydrostatics, vibration theory –
the sort of doctrines on which Kant mused, in a speculative-heuristic mood,
in a General Remark to Dynamics (MFS 4:523–35) – have become a
constitutive, deductively integrated area of classical theory, qua continuum
mechanics. This raises the question whether mechanics enlarged by the
addition of these new domains can still be grounded in Kant’s foundation.
We have some reasons to be cautiously optimistic that it can.39 And yet,
d’Alembert’s approach above only grew stronger and more influential after
Kant – as Lagrangian mechanics, a general theory of constrained systems.
Moreover, in the 1840s it acquired a competitor, namely Hamilton-Jacobi
theory, just as alien to Kant’s framework as it was. Together, these two parts
have evolved into the so-called first and second formalism of analytic
dynamics. Some have tried to link it to a Kantian basis of sorts, by loosening
rather greatly the connection that Kant originally placed between metaphy-
sics and dynamics.40 Overall, however, Kant deserves our understanding. It is
not clear that modern classical mechanics is a single, monolithic theory, so
we cannot expect him to have grounded it as one.
39
For an attempt to show that, see Marius Stan, “Metaphysical Foundations of Neoclassical
Mechanics,” in Kant and the Laws of Nature, ed. Michela Massimi and Angela Breitenbach
40
For example, see Michael Stöltzner, “Can the Principle of Least Action Be Considered a Relativized A
Priori?” in Constituting Objectivity: Transcendental Perspectives on Modern Physics, ed. Michel Bitbol,
Pierre Kerszberg, and Jean Petitot (Berlin: Springer, 2009), 215–27.
Part VIII
Philosophy of Religion
23
Kant on Faith: Religious Assent
and the Limits to Knowledge
Lawrence Pasternack
In the preface to the 1787 edition of the Critique of Pure Reason, Kant
explains that one of the aims of his critical project is to establish the limits to
knowledge “in order to make room for faith” (Bxxx). Religion, he presses,
must be freed from the “disadvantageous influence” of the metaphysical
tradition, wrested from the “monopoly of the schools,” and set on a footing
suitable to the “common human understanding” (Bxxxi–xxxii). He main-
tains that this is necessary in order to ensure that our faith in God and
immortality is duly guarded against the “pretension” of speculative reason
(Bxxx).
To many, these statements are utterly baffling, or at least powerfully at
odds with the common image of Kant as a secular philosopher. Yet, such
affirmations of religious faith can be found throughout the critical corpus.
Despite the all-too-common conception of Kant as hostile to religion, we
find in all three Critiques, in numerous works of the 1790s, as well as
throughout his lectures and Reflexionen, frequent presentations of faith as a
legitimate mode of holding-to-be-true (Fürwahrhalten) and as the mode of
assent proper to religion.1
1
There has been a recent flood of publications on Kant’s understanding of assent, particularly religious assent.
Representative articles include the following: Andrew Chignell, “Belief in Kant,” Philosophical Review 116,
L. Pasternack (*)
Department of Philosophy, Oklahoma State University, Stillwater, USA
e-mail: l.pasternack@okstate.edu

516 L. Pasternack
It is the purpose of this chapter to explore Kant’s understanding of faith,

particularly moral/pure rational faith and historical faith, the two forms most
significant to his philosophy of religion. We will begin with a brief discussion
of Kant’s various conceptions of belief/faith (Glaube) before turning to an
overview of its place within Kant’s broader taxonomy of propositional
attitudes. We will then consider what lies behind Kant’s dramatic statement
on moral/pure rational faith in the B-preface to the first Critique. Lastly, we
will look more closely at historical faith, exploring its internal structure and
contents, and will consider whether it too may be a form of holding-to-be-
true. Throughout, we will attend to how best to fit Kant’s conception of faith
within the epistemic strictures of transcendental idealism.
The emergence of moral/pure rational faith

Until quite recently, there has been very little work done on the normative
elements of Kant’s epistemology. Nevertheless, we find in many of his
published works as well as throughout his logic lectures, frequent and quite
detailed accounts of knowledge (Wissen), opinion (Meinung), and belief/faith
(Glaube)2 as our three fundamental modes of holding-to-be-true
(Fürwahrhalten).3 While Kant’s treatments of knowledge and opinion are
no. 3 (July 2007): 323–60; Andrew Chignell, “Kant’s Concepts of Justification,” Noûs 41, no. 1 (March
2007): 33–63; Lawrence Pasternack, “Kant’s Doctrinal Belief in God,” in Rethinking Kant, vol. 3, ed. Oliver
Thorndike (Newcastle upon Tyne: Cambridge Scholars Press, 2010), 200–218; Lawrence Pasternack, “The
Development and Scope of Kantian Belief: The Highest Good, the Practical Postulates and the Fact of
Reason,” Kant-Studien 102, no. 3 (Jan. 2011): 290–315; Lawrence Pasternack, “Kant on Opinion: Assent,
Hypothesis, and the Norms of General Applied Logic,” Kant-Studien 105, no. 1 (April 2014): 41–82;
Lawrence Pasternack, “Kant on Knowledge, Opinion, and the Threshold for Assent,” in Rethinking Kant,
vol. 4, ed. Oliver Thorndike (Newcastle upon Tyne: Cambridge Scholars Press, 2014), 55–74; Leslie
Stevenson, “Opinion, Belief or Faith, and Knowledge,” Kantian Review 7 (March 2003): 72–101; and
Joseph S. Trullinger, “Kant’s Two Touchstones for Conviction: The Incommunicable Dimension of Moral
Faith,” Review of Metaphysics 67, no. 2 (Dec. 2013): 369–403.
2
The German Glaube can be used either in the mundane sense of the English “belief,” or as the religious
honorific “faith.” While Kant sometimes intends its more mundane use, we will be focusing primarily
on those contexts where the latter is operant. Through this section of the chapter, we will use both
“faith” and “belief” for Glaube, varying the English relative to whether or not the sense of Glaube in use
is closer to the former or latter. However, later in the chapter, as we look more focally at Kant’s religious
use of the term, we will primarily use the English “faith.”
3
While there is nothing particularly unique about this triad of propositional attitudes, it is still worthy
of note that Kant’s selection of the three as our fundamental modes of holding-to-be-true carries with it
the convention common to Christian philosophers of distinguishing opinion from faith. The roots of
this triad are often traced to medieval philosophy and can be found, for instance, in Aquinas’s Summa
Theologica, II/II Q.1. art 2 and Q.2 art. 1.
23 Kant on Faith 517
relatively stable, there is a much more complicated developmental story

surrounding his use of Glaube. In my 2011 article, “Development and Scope
of Kantian Belief,” I presented this story in considerable detail.4 But it would
still be helpful here to mention a few of its highlights in order to gain some
clarity with regards to Kant’s many and often confusing uses of Glaube before
we move on to its more specific use in the second Critique and beyond.
To begin, we may make significant progress in disambiguating its use by
considering how it is employed through four distinct periods. The first of these
is pre-critical, where, prior to the Critique of Pure Reason’s pronounced need to
find “room for faith,” it has for Kant relatively less philosophical significance.
Besides an ample nonphilosophical use (akin to the English “belief”), it is given
a somewhat more technical application in the lectures on logic.
As seen most clearly in the early Blomberg Logic, Kant’s use of Glaube through
this period tended to follow Georg Friedrich Meier’s Auszug aus der Vernunftlehre,
the textbook from which he lectured on logic from 1756–1796. Hence, as Meier
defines Glaube as an “acceptance [of] something on the basis of testimony” (Ak
16:497) and evaluates witness statements in terms of what he calls our “moral
certainty” in their personal credibility (Ak 16:510),5 so Kant likewise, at least
prior to the first Critique, uses both Glaube and moralischer Glaube (moral belief)
in a like manner. Hence we find in the early Blomberg Logic lectures, for example,
a use of Glaube that has little to do with its later meaning and much more to do
with Meier’s preference to use the term in relation to trust and testimony. As
such, Kant at that time used “moral belief [moralischer Glaube]” in relation to an
assent based upon “a particular trust toward a person,” such as, for example, the
moral belief a servant has in his master’s judgment (BL 24:242).6
With, however, the advent of the critical period, Kant repurposes mor-
alischer Glaube, using it instead to refer to an assent that is (a) grounded in
pure practical reason and (b) concerns primarily (or perhaps solely) the
highest good and its postulates. Thus we see from the first Critique onwards
“moral faith” as among the terms more commonly used to represent Kant’s
core conception of religious assent.
4
Pasternack, “Development and Scope of Kantian Belief.”
5
For an English translation of Meier’s Auszug, see Georg Friedrich Meier, Excerpt from the Doctrine of
Reason, trans. Aaron Bunch, vol. 1 of Kant’s Sources in Translation, ed. Pablo Muchnik and Lawrence
Pasternack (London: Bloomsbury, 2016).
6
Moralischer Glaube is likewise used by Kant as term for trust in various pre-critical correspondences of
the 1770s. See for example, his 1775 letters to Johann Caspar Lavater, a Swiss pastor and theologian (C
10:175–80).
518 L. Pasternack
However, rather than more thoroughly overhauling his broader use of

Glaube (that comes later), Kant still follows Meier in other regards, and thus
continues to use the term both in relation to religious assent as well as in a
variety of additional ways. Noteworthy among these are two that have
garnered some interest, though are mentioned only in passing in the
Canon of the Critique of Pure Reason.
The first of these is “pragmatic belief,” which is otherwise just found with
passing mention in the lecture notes (A824/B852; BL 24:244). In brief, Kant
uses the term to refer to the adoption of practical hypotheses in order to
guide action. Examples include: a physician’s diagnosis used to direct a
patient’s treatment (A824/B832), a businessman’s assessment of the merits
of a deal (JL 9:68), and a general’s evaluation of enemy strength (DWL
24:750). However, as one might surmise, Kant comes to acknowledge that
this mode of assent is not substantially different from mere opinion (assent
based upon probability), though as an assent relevant to action, he still
during this period associates it with his relatively loose treatment of Glaube.
Second, also in the Canon, Kant makes brief mention of what he there
calls “doctrinal belief” (A825–26/B853–54), a form of assent that has to do
with a strong commitment to matters considered only through speculation or
conjecture. While in one of his early papers, Andrew Chignell sought to use
this as a wedge term to support,7 as we will later discuss, a more metaphys-
ically friendly treatment of transcendental idealism, it is dangerous to make
too much of it. Kant not only acknowledges that it is an “unstable” notion
(A827/B855), but seems to withdraw it thereafter. Not only is the term never
used again in the corpus, but each of the three examples given by Kant are
recast as other modes of assent. For instance, instead of a “doctrinal belief” in
extraterrestrial life (A825/B853), Kant later places the assent within the scope
of opinion (CJ 5:467; see also RPT 8:396n).8 Hence, despite Kant’s explora-
tions of various additional forms of Glaube during this early phase of the
critical period, nearly all drop away in the years that follow.
We may mark then the third period of use with the second Critique’s
robust exploration of Glaube, or as he there often calls it from this text
forward, reiner Vernunftglaube. As we shall discuss in later sections, it is very
much through this text in particular that Kant sets out the framework for the
core mode of religious assent employed through the remainder of the critical
7
Chignell, “Belief in Kant,” 345–54.
8
For a more thorough discussion of doctrinaler Glaube, see Pasternack, “Kant’s Doctrinal Belief in God.”
period. Additionally, where the second Critique is primarily devoted to

Kant’s positive account of religious assent, the third Critique has much to
say about what Glaube is not.
Kant thus writes, for example, that the highest good and its postulates are
the sole “matters of faith (res fidei), and are indeed the only ones among all
objects that can be so designated” (CJ 5:469). For with the model of religious
assent set out in the second Critique and reaffirmed in the third, nothing can
be “intrinsically a matter of faith” that is not an object of pure practical reason
(CJ 5:469). Hence, “everything else that it is at least possible for us to know
given the constitution of our cognitive faculties, belong not among matters of
faith but among facts” (CJ 5:469). Accordingly, by 1790, Kant has come to
equate Glaube simpliciter with reiner Vernunftglaube, shifting the remains out
from his earlier, broader use and (for the most part) into opinion.
While we could potentially here end our survey of how Kant uses Glaube
over time, it is more accurate to add one further phase of use, or at least a part B
to the third phase discussed above. On the one hand, by 1790, Kant has set the
boundaries for his core conception of religious assent; yet he still offers – or at
least ponders – one more religiously significant use of Glaube. Although in
earlier works, Kant uses historischer Glaube much as did Meier, as again a mode
of assent based upon trust and testimony, and even in the third Critique,
explicitly foreswears this earlier term, it is nevertheless, like moralischer Glaube
before it, eventually repurposed, used to refer to religious traditions and their
particular doctrines. Hence, in Religion within the Boundaries of Mere Reason,
Christianity and Islam are referred to as “historical faiths” and, similarly, Kant
therein uses “historical faith” to indicate assent in their particular doctrines (e.
g., Rel 6:109, 119, 164, 166n) – ranging from claims about miracles and
revelation to tenets of piety and ecclesiastical policy.
So, while, on the one hand, Kant established moral/pure rational faith as the
mode of assent for the religious principles that can be derived from pure
practical reason, through the Religion, he also considers the extent to which
these core principles can find expression in the more “mystical cover” offered by
the doctrines of historical faith (Rel 6:83). As for whether or not Kant permits
this repurposed “historical faith” to have its own distinct character, or if it is
merely a more ornate way of affirming in the mode of moral/pure rational faith,
that is a topic we will discuss toward the end of this chapter. But now, having
hopefully offered an adequate overview of the various periods of Kant’s use of
Glaube, let us begin to explore its philosophical structure, beginning with the
general principles that govern Kant’s taxonomy of propositional attitudes.
520 L. Pasternack
Glaube in Kant’s taxonomy of propositional

attitudes
To begin, the primary differentia used by Kant to contrast our various modes
of holding-to-be-true, are what he (again, following Meier) calls objective
and subjective “sufficiencies [zureichenden]” (A822/B850; OT 8:141; JL
9:70, 73). There is some disagreement in the literature about how Kant
uses these terms, but the textual evidence favors the former being taken as an
epistemic notion on par with or equal to “objective” certainty – that is,
grounds sufficient to guarantee truth.9 Subjective sufficiency, rather than
being about the grounds for assent, is best seen in terms of our psychological
orientation toward a proposition. More precisely, when an assent is subjec-
tively sufficient, we may understand this to be a holding-to-be-true with
maximal firmness of commitment or confidence, what we sometimes also
refer to as “certainty,” though in its subjective, psychological sense.
Despite other changes within his analysis of assent, Kant continued through-
out the critical period to distinguish between knowledge, opinion, and faith by
way of the above. Knowledge is, unsurprisingly, rendered as both objectively
and subjectively sufficient. That is, an assent with the status of knowledge is one
that has gained certainty in both its objective sense, as maximally epistemically
justified, and in its subjective sense, as maximal psychological confidence in its
truth.10 Opinion, by contrast, is presented as neither objectively nor subjec-
tively sufficient – although to opine still involves one’s having some grounds
favoring the proposition and so some degree of commitment to its truth. And
then faith, quite distinct from the preceding, has the peculiar structure of
lacking objective sufficiency while still being subjectively sufficient.11
Hence, where the “subjective” side of the assent tracks the “objective” in
both knowledge and opinion, we see in faith their asymmetry: we can be
firmly committed to the proposition’s truth while at the same time lacking
cognizance of any epistemic justification in its favor. Such an asymmetry
9
Leslie Stevenson holds that Kant believes that knowledge requires certainty. See his “Opinion, Belief or
Faith, and Knowledge,” 99–100n21. Chignell, however, thinks otherwise. See his “Kant’s Concepts of
Justification,” 43; and his “Belief in Kant,” 325–26. See also my discussion of the issue in “Kant on
Opinion.”
10
The achievement of (objective) certainty, for Kant, is no small task. He holds not only that the
grounds of assent must be infallible, but that all competing “scruples” must be addressed. See my
discussion of the issue in “Kant on Opinion.”
11
For a discussion of these distinctions, see also Chapter 6 of this volume.
might suggest that faith is actually a form of error, for confidence in truth,
absent grounds for that confidence, is often seen as the hallmark of cognitive
bias. Yet this is not at all Kant’s position.
A further division crucial to Kant’s taxonomy of propositional attitudes is
between what he calls “persuasion [Überredung]” and “conviction
[Überzeugung].” Whereas the former is his term for those instances of assent
that are spurious, lacking the sort of grounds that would meet with universal
agreement, the latter term refers to those modes of assent that are “valid for the
reason of every human being” (A820/B848). Faith, for Kant, just like knowledge,
falls under “conviction,” and thus it is not, as are the forms of persuasion, based
upon one’s own particular biases and psychological interests. In fact, at one point
in the Critique of Practical Reason, Kant explicitly distinguishes faith from wishful
thinking, as he argues that it instead has its basis in a universal “need of [practical]
reason arising from an objective determining ground of the will” (CPrR 5:143n).
Persuasion, by contrast, “has only private validity,” for it “has its ground only in
the particular constitution of the subject” (A820/B848). This may include, as
noted, wishful thinking whereby one has an inclination favoring the proposition’s
truth, or the all-too-common delusion of “logical egoism” whereby one takes the
mere fact that one holds a proposition to be true as sufficient grounds for holding
it to be true (see BL 24:151; Ak 24:551; VL 24:874).12
Although faith has its grounds in a practical need, that does not, for Kant,
either relegate it to persuasion or otherwise deprive it from being a legitimate
mode of assent, one whose grounds are “universally,” “necessarily” and
“objectively valid” (BL 24:202; A820/B848; CJ 5:461).13 Moreover, instead
of being some sort of pseudo-assent, an “as-if” where there is no genuine
holding-to-be-true, Kant not only presents faith (again like knowledge) as a
holding-to-be-true with certainty (e.g., A822/B850, A829/B857; JL 9:67),
but also shows how it allows, as framed in the second Critique, for the
“extension of pure reason for practical purposes” into the supersensible
(CPrR 5:134). Hence, it should be clear that Kant’s position is hardly akin
12
I discuss the table of biases Kant draws from Meier in “Kant on Opinion.” See also Lawrence
Pasternack, “Kant’s Touchstone of Communication and the Public Use of Reason,” Society and
Politics 8, no. 1 (2014): 78–91.
13
While one with an opposing bias may, of course, refuse to accept a properly justified claim, Kant
nevertheless renders conviction under the ideal that “the grounds that are valid for us [should] have the
same effect on the reason of others” (A821/B849). Similar to his portrayal of our expectation of
agreement in aesthetic judgment, the expectation tied to conviction pertains “not so much to the actual
as to the merely possible judgments of others” (CJ 5:294). That is, the “universal” agreement of
conviction is normative rather than factive (i.e., de jure versus de facto).
522 L. Pasternack
to the reductionism promoted by those who take his religious claims to be

merely symbolic, mere depictions our moral ideals in “imaginatively
enhanced or pictorial form.”14 Quite the contrary – for passage after passage
presents faith as (a) a form of holding-to-be-true; (b) a legitimate mode of
assent – that is, conviction; (c) a holding-to-be-true with certainty; and,
further, (d) that the grounds which justify the assent at the same time
establish the objective reality of its referents (e.g., CPrR 5:134, 143n; CJ
5:469; RP 20:299).
We will more fully consider Kant’s views on faith and its grounds below.
But here, let me offer this diagram as summary of the core structure of his
taxonomy of propositional attitudes15:
Assent (Fürwahrhalten)
Conviction Persuasion Opinion

(Überzeugung) (Überredung) (Meinung)
Logical conviction Moral/practical conviction
Knowledge (Wissen) Faith/belief (Glaube)
The limits to knowledge and the room for faith

Until very recently, Kant’s discussions of faith have been routinely over-
looked. Most philosophers and theologians, not even aware that such dis-
cussions exist, have instead followed the prevailing view that Kant is
essentially a secular philosopher. Even those who have had some encounter
14
James J. DiCenso, Kant’s “Religion within the Boundaries of Mere Reason”: A Commentary (Cambridge:
Cambridge University Press, 2012), 28.
15
One may compare this to Chignell’s diagram of propositional attitudes on page 333 of “Belief in Kant.”
Chignell also includes what he calls “mere conviction.” This is not obviously part of Kant’s taxonomy, but it
fits well enough. The main difference between our respective taxonomies is that there is not just logical but
also a moral/practical form of conviction, under which falls Glaube. Kant also states in a variety of texts that
conviction comes in two forms, “logical” and “moral.” The former pertains to knowledge and the latter to
faith (A829/B857). He sometimes uses “practical” instead of “moral.” See Rel 6:62; JL 9:72; Ak 28:1082.
Evidence that Kant views faith as an instance of conviction is abundant. For example, in the Critique of the
Power of Judgment he writes that when a proof is “based on a practical principle of reason (which is thus
universally and necessary valid), then it can make a sufficient claim of conviction from a purely practical
point of view, i.e., moral conviction” (CJ 5:463). See also, e.g., Rel 6:103; OT 8:142; NF 16:373 [R2450],
16:375–76 [R2454], 16:511–12 [R2789]; JL 9:72; BL 24:148–49.
with his religious writings have tended to dismiss them as “wobbles,” mere
symbolism, or even capitulations to the Christian powers of his day.16
Fortunately, though, we have begun to see a reassessment of Kant’s reli-
gious views. While various defenses of Kant’s philosophy of religion have been
with us for some time, including Allen Wood’s Kant’s Moral Religion and
Stephen Palmquist’s Kant’s Critical Religion,17 we have seen over the past few
years a significant rise in support for the positive elements of Kant’s philoso-
phy of religion in general, as well as a dramatic increase in work on the
Religion.18 However, while there are ample grounds to support the growing
movement of “affirmative” interpreters, there is no need to qualify for its sake
the epistemic strictures set by transcendental idealism. Hence, despite the fact
that many affirmative interpreters tend toward a “metaphysically friendly”
reading of Kant, the “room for faith” advanced by Kant in the B-preface to the
Critique of Pure Reason is, as we shall discuss, already part of his critical turn.
According to the “metaphysically friendly” reading currently in vogue,
various emendations can and should be made to transcendental idealism’s so-
called restriction thesis (i.e., that knowledge [Wissen] is limited to the scope
of possible experience). For it is thought that without them, there would be
no way to bridge the gap between Kant’s theoretical and practical philoso-
phy. Hence, in opposition to the standard anti-metaphysical interpretation
of transcendental idealism, one that denies the possibility of knowledge (as
well as cognition [Erkenntnis]) of the supersensible, the “metaphysically
friendly” interpretation does not merely give positive ontological status to
things in themselves, but further maintains that at least in some specialized
cases, we may cut through the restriction thesis in order to secure knowledge
of the supersensible. We see this, for example, in Desmond Hogan’s appeal
to an alleged theoretical knowledge of the freedom of the will, Andrew
Chignell’s claim that Kant permits a specialized “doctrinal” or “theoretical”
assent to God, Christopher Insole’s adoption of Hogan’s and Chignell’s
“metaphysically friendly” stratagems in order to address Kant’s stance on
the problem of concurrence, and even the metaphysical moments of Chris
16
These views refer respectively to Gordon Michaelson, Adina Davidovich/James DiCenso, and Johann
Wolfgang von Goethe.
17
Allen W. Wood, Kant’s Moral Religion (Ithaca: Cornell University Press, 1970); and Stephen R.
Palmquist, Kant’s Critical Religion (Burlington, Vt.: Ashgate, 2000).
18
Although the second half of Chris L. Firestone and Nathan Jacobs’s In Defense of Kant’s “Religion”
(Bloomington: Indiana University Press, 2008), where the authors seek to defend Kant as a Christian
philosopher, is more questionable, the literature survey which comprises its first half is well crafted, and
may very well be part of the catalysis for the recent increase in work on Kant’s philosophy of religion.
524 L. Pasternack
Firestone and Nathan Jacobs’s work, particularly their turn to “transcenden-

tal Platonism” in order to explain Kant’s views on the Incarnation.19
While few would deny that Kant does affirm various supersensible claims, it
does not follow that these affirmations require any exceptions to the restriction
thesis. This is because the positive elements of Kant’s philosophy of religion are
not, for him, matters of knowledge, that is, instances of assent with warrants of the
sort that contribute to “objective sufficiency”; rather, they are quite explicitly
presented in contrast to such, depicted instead as matters of faith, absent the sort of
epistemic grounds that are employed in cases of knowledge (and opinion). This
point is clearly broadcast at Bxxx, where Kant famously writes that he had to
determine the “limits to knowledge” in order to “make room for faith.” Likewise,
we see in the first Critique’s Canon of Pure Reason, through much of the second
half of the Critique of Practical Reason, through §90–91 of the third Critique, in all
the logic lectures, in numerous essays of the 1790s, and more, Kant repeatedly
claiming that it is not knowledge but faith that is the proper mode of assent for the
supersensible claims found within his philosophy of religion.
Somehow, this surfeit of passages has been ignored or overlooked by the
“metaphysically friendly” interpreters.20 Hence, despite the fact that Kant
explicitly sets out a distinct mode of assent for the practical and religious,
explicitly contrasting it with knowledge, many interpreters have simply
sidelined these passages rather than analyzing what Kant means by faith
and why, for him, it is faith rather than knowledge that is the mode of assent
proper to our practical and religious interests. But now, having established
that faith is, for Kant, a distinct mode of assent separate from knowledge, let
us turn to why this mode is of such significance to him and why he advances
it as the mode of assent proper to religion.
To begin, let us briefly consider the significance of Kant’s claim that the
“room for faith” depends upon the limits to knowledge. Less attentive readers
may not take due note of the logic of this statement, and may incorrectly take
its claim to be that because of transcendental idealism’s restriction thesis, we
must make some allowance for faith. This, however, presents Kant’s actual
19
Desmond Hogan, “How to Know Unknowable Things in Themselves,” Noûs 43, no.1 (March 2009):
49–63; Chignell, “Kant on Belief”; Christopher J. Insole, Kant and the Creation of Freedom: A
Theological Problem (Cambridge: Cambridge University Press, 2013); and Firestone and Jacobs, In
Defense of Kant’s “Religion,” 155–70.
20
The same criticism can be laid against DiCenso, whose reading of Kant was formerly offered by Adina
Davidovich in Religion as a Province of Meaning: The Kantian Foundations of Modern Theology
(Minneapolis: Fortress, 1994). The reductive/symbolic interpretation they promote fails to take account
of the ample discussions of the nature of faith found throughout the critical corpus.
logic backwards, for what he actually states is that the limits to knowledge
must be sought in order to make room for faith (Bxxx). That is, his commit-
ment to faith, and its safeguarding against the encroachments of metaphysics,
was at least part of what motivated transcendental idealism from the start.
Unfortunately, however, this point has been overlooked by most “metaphy-
sically friendly” interpreters, for rather than recognizing the mode of assent
explicitly offered by Kant, they instead advocate for exactly what Kant
decries. Amendments to the restriction thesis thus offer religion no service
but rather place it in jeopardy.
The reasons Kant gives for his views on the relationship between faith and
knowledge may be divided into two main clusters. The first of these is
primarily negative – focusing on what is wrong with religious knowledge.21
The second cluster is positive, addressing what it is about faith that makes it,
rather than knowledge, the proper mode of assent for religion.
Perhaps the most persistent negative arguments presented by Kant, one found
from the first Critique’s B-preface through the Religion and Conflict of the Faculties,
has more of a social-political quality than epistemic, for Kant time and again
expresses his opposition to religious oligarchies, or as he puts it in the B-preface,
the “monopoly of the schools” (Bxxxii). Echoing both the Lutheranism of his
youth as well as Enlightenment egalitarianism, Kant challenges “the arrogant
claims of the schools, which would gladly let themselves be taken for the sole
experts and guardians of such [religious] truths” (Bxxxiii). Scholastic and ration-
alist theologians attain to such religious authority by their mastery of the arcane
metaphysical reasoning that allegedly lies behind religious doctrine. They then
reinforce their social authority by elevating the importance of ideological purity,
making religion into more a matter of doctrinal adherence than life-practice.
We find this line of criticism further developed in various texts beyond just
the B-preface to the Critique of Pure Reason – most notably in Part Four of
Religion within the Boundaries of Mere Reason, where Kant challenges the
religious despotism that maintains its control over the laity by training them
to believe that salvation depends upon doctrinal purity rather than a “change of
heart” through which one gives morality priority over self-interest. Instead of a
religion focused on inward growth, what is demanded of the laity is that they
profess as true (and as certain) what they cannot in good conscience genuinely
21
Of course, one may say that transcendental idealism itself provides the core negative argument against
religious knowledge. However, while most would accept this point, it would seem question-begging to
the metaphysically friendly interpreter. Thus, besides the technical reasons for limiting knowledge to
possible experience, we should explore Kant’s further arguments.
526 L. Pasternack
regard as so justified (Rel 6:171, 189). Hence, instead of a religion set on a

footing suitable to the “common human understanding” (Bxxxii), the priestly
class have long promoted metaphysical arcana, for in doing so, they have been
able to maintain their dominance over the benighted masses.22
Of course, there are numerous other negative arguments that could be
drawn from transcendental idealism’s general grounds for rejecting super-
sensible knowledge. From the more semantic objections found in the first
Critique’s Phenomena/Noumena section, to the overdetermination argu-
ments of the Antinomies, to, of course, the Ideal of Pure Reason, Kant
clearly employed numerous stratagems to block claims of supersensible
knowledge, religious or otherwise. While our flights of metaphysical fancy
have long illustrated that we can indulge in such speculative thought, in the
end, absent experience itself, we are, for Kant, unable to make any theoret-
ically grounded determinate judgments about the supersensible.
Reason, isolated from experience, however creative it can be, cannot
determine which supersensible theses are true. Concepts, abstracted from
experience, lose their relations to space and time, and as a result, become far
vaguer (e.g., what is “cause” without a before and after?). And while many
have looked to revelation as a source of supersensible knowledge, Kant
argues: (a) there are no markers that can allow us to distinguish between
authentic revelation and illusion (Rel 6:109, 114)23; (b) because of the
seemingly epistemically privileged character of revelation, an assent based
upon it cannot rise to knowledge, for it lacks the sort of grounds that could
be universally communicated (A821/B849); and, lastly, (c) no doctrines that
come from revelation alone could be taken as essential to our religious
vocation, for if doctrines of such significance were not equally accessible to
one and all, a violation of “ought implies can” would result (Rel 6:103,
154–55).
22
We can also see here a response offered by Kant to the well-known criticism of Bishop Butler.
According to Butler, Enlightenment theologians ultimately promote a religious elitism, for they too rely
upon specialized texts and advanced instruction to disseminate their rendering religious doctrines.
Hence, Butler contends, the most egalitarian path to religious truths is via the Gospels, which are
more accessible both in terms of their contents and as physical artifacts, than the writings of his
contemporary theologians. As Allen Wood likewise discusses, Kant’s alternative takes religious truths
to be nascently available to one and all, rooted in our common moral needs. See Allen W. Wood,
“Kant’s Deism,” in Kant’s Philosophy of Religion Reconsidered, ed. Philip J. Rossi and Michael Wreen
(Bloomington: Indiana University Press, 1991), 1–21.
23
This is a slight overstatement, for Kant does point out that no “theistic” miracle or revelation could
violate the moral law (Rel 6:87). Nevertheless, were “demonic” miracles or revelation possible, Kant
notes that of course they would not be similarly restricted (Rel 6:86).
Hence, Kant rejects, on multiple grounds, both reason and revelation as a

basis for supersensible knowledge. But this is hardly where the issues under-
lying Bxxx come to an end, for as discussed above, Kant does not turn to
faith because of the limits to knowledge, but rather asserts that it is for the
sake of faith that we must place limits on knowledge. Accordingly, rather
than dwell on the familiar, let us look into what underlies these arguments
and objections, what it is for Kant that makes faith the proper mode for
religion assent.
Part of this positive defense of religious faith can be found in the section of
the second Critique titled “On Assent from a Need of Pure Reason” (CPrR
5:142–46), where Kant presents faith as a “voluntary determination” of the
will (CPrR 5:146), a “choice, in which a free interest of pure practical reason
decides” (CPrR 5:146). Such voluntarism, however, is unique to faith, for in
the case of knowledge (and opinion), “convincing proofs” move the will to
assent, and though perhaps an overstatement, we find in the Jäsche Logic: “the
will cannot struggle against convincing proofs of truths that are contrary to
its wishes and inclinations” (JL 9:74).
Hence, one positive feature of faith, a feature that is, moreover, distinct to it, is
that it engages with the will in a way that is not possible for other modes of assent.
With knowledge and opinion, theoretical judgment determines the assent. Even
with persuasion, where one’s subjective interests shape cognitive bias, it is still
inclination that provides the basis for the assent. But not so with faith, for assent
in its case has its grounds in the will itself, since it is “moral interest” that “turns
the scale” (CPrR 5:145). In fact, given that faith is grounded in pure practical
reason, one may even see it as the assertoric expression of Kantian autonomy, in
contrast to the heteronomy of knowledge, opinion, and persuasion, each of
whose grounds lie outside the will itself.
This brings us to a second important component of Kant’s positive
defense of faith. Since faith can, unlike knowledge, be chosen, in faith one
achieves a cognitive state as the result of one’s moral interest. While this
point was operant in Kant’s thoughts from the mid-1780s onwards,24 it is
best understood by way of the argument for the highest good found in the
24
In the first Critique, Kant of course rejects the possibility of religious knowledge. However, Kant does
not yet present our commitment to the authority of the moral law as the basis for our faith in the
postulates. Even though the highest good and the postulates are discussed in the Canon of Pure Reason,
the composition of the A-edition nevertheless indicates that Kant used the highest good in order to
motivate our moral interest through our interest in happiness (see A813/B841). It is not until the second
Critique that Kant begins to clarify faith’s roots in pure practical reason.
528 L. Pasternack
Religion’s first preface. As Kant explains there, “it is one of the inescapable
limitations of human beings and of their practical faculty of reason . . . to be
concerned in every action with its result” (Rel 6:7n). We have, in other
words, a “natural need” to “consider in every action, besides the law, also an
end” (Rel 6:7n). For many, this “natural need” is easily satisfied by way of
our equally natural interest in our own happiness. Yet, this is not the end
prescribed to us by the moral law. It, rather, proffers the highest good as the
end that we ought to adopt in the satisfaction of this need.
The problem, however, is that the world as we experience it tells us that the
highest good is not possible. In the Critique of Pure Reason, for example, Kant
writes that the distribution of happiness in accordance with moral worth can
be “determined neither by the nature of the things in the world, nor by the
causality of actions themselves and their relation to morality” (A810/B838).25
That is, even if “everyone do[es] what he should” (A810/B838), such a
distribution is still beyond our capabilities. Likewise, in the Critique of
Practical Reason, Kant claims that “no necessary connection of happiness
with virtue in the world, adequate to the highest good, can be expected
from the most meticulous observance of moral laws” (CPrR 5:113); and in
the Religion, he writes that “human capacity does not suffice to effect happi-
ness in the world proportionate to the worthiness to be happy” (Rel 6:8n).26
While one could disregard this problem if only concerned with acting
from duty episodically, it cannot be circumvented when taking up the
highest good as one’s actual end, as the ultimate result toward which one
aims. Hardly could we take on the highest good as the “special point of
reference for the unification of all [our] ends” (Rel 6:5), hardly could we
accept it as what the moral law proffers for us as the principle to satisfy the
25
As an empirical thesis, we see that the laws of nature do not follow the laws of justice. In addition,
Kant’s earlier analysis of causality on a priori grounds excludes norms from nature: “In nature the
understanding can cognize only what exists, or has been, or will be. It is impossible that something in it
ought to be other than what, in all these time-relations, it in fact is. . . . We cannot ask at all what ought
to happen in nature, any more than we can ask what properties a circle ought to have” (A547/B575). Of
course, the Critique of the Power of Judgment turns to the question of a final end of nature, but even
there, the judgment, as reflective, is a projection on our part onto nature. Moreover, there too we do not
expect justice to be realized in nature and thus Kant repeatedly appeals to our “future life” (see, e.g., CJ
5:460, 469, 471n).
26
Although some interpreters have sought to defend a “secular” version of the highest good, even
claiming that this version can be found in the corpus, especially the third Critique, Kant actually never
deviates from the “theological” version of the doctrine. The textual basis for the “secular” version is
spurious and based upon weak scholarship. See the case against it in Lawrence Pasternack, Kant’s
“Religion within the Boundaries of Mere Reason”: An Interpretation and Defense (London: Routledge,
2014), ch. 1.
“natural need” through which we are “concerned in every action with its
result” (Rel 6:7n), if it were not something we could take to be possible.
It is in this way that our commitment to morality ultimately depends upon
faith. Yet, it is not simply that faith provides what knowledge cannot. Faith is
not just a fallback in light of the limits to knowledge. Rather, our acceptance
of the highest good and its postulates make possible a commitment to the
moral law that, if not for faith, we could not have. This is not simply about the
objects of commitment, but about the nature of the commitment itself: the
moral law calls upon us to use it as a practical principle through which our
individuals actions are chosen; the moral law further calls upon us to accept
the highest good as the “special point of reference for the unification of all
[our] ends”; but further, we have, as a result of the “limits to knowledge,” the
opportunity to devote ourselves to the moral law so fundamentally that it
determines not just our conduct but how we see the world.
While the limits to knowledge are such that the postulates could only
possibly be affirmed by faith, they also ought to be affirmed by faith, for in
doing so, one all the more fundamentally aligns oneself with the moral law,
subordinating not only one’s interest in happiness to it, but even the interests
of theoretical reason. This is, for Kant, what it means for practical reason to
have “primacy” over the theoretical (see CPrR 5:119–21). It is not merely
that practical reason can go where the theoretical cannot, but that the former
has a “prerogative” to do so (CPrR 5:119).
Accordingly, in the “change of heart,” where we fully align our funda-
mental disposition (Gesinnung) with the moral law, we do not merely accept
it as a formal principle, a “supreme maxim” that gives morality priority over
self-interest, but we invest ourselves in a life-project as well as the worldview
upon which it depends. In all these ways, we subordinate other competing
interests to the moral law, granting it authority over not just our wills but our
cognition. This is why “morality inevitably leads to religion” (Rel 6:8n).27
With the above analysis of Kant’s commitment to the “room for faith” as
our foundation, let us now move from pure rational faith to historical faith,
and explore how the latter as well has a role to play in our commitment to the
highest good. While the former mode of assent is certainly the foundation for
Kant’s philosophical theology, there is, as we shall see, still much more to it,
27
For another treatment of the role of choice in faith and its moral significance, see Wood’s discussion
of what he calls the “absurdum practicum” in Kant’s Moral Religion. More recently, Trullinger has sought
to elevate Kant’s discussions of betting as it relates to faith in “Kant’s Two Touchstones for Conviction.”
530 L. Pasternack
and thus much more to Kant’s conception of religious assent than what is
conveyed by pure rational faith alone.
Religion’s “experiment” and the contests

of historical faith
In the second preface to the Religion, Kant describes two concentric spheres,
with the smaller or more “narrow” one containing the Pure Rational System
of Religion (hereafter “PRSR”), and the “wider” sphere containing the much
more ample body of doctrines of historical faith28:
Historical/
ecclesiastical
faith
Pure rational
system of
religion
He then raises the question as to how much of the wider sphere overlaps with
the smaller one. This question, often (misleadingly) referred to as the
Religion’s “Second Experiment,”29 is at first explored incrementally, as
Kant develops the rational parallels to such core Christian doctrines as
original sin and salvation. Then, toward the start of Part Three, as he begins
to lay the foundations for his ecclesiology, he advances his more general
thesis regarding the overlap between the spheres.
Following his commitment to the principle of “ought implies can,” Kant
maintains that whatever is required of us in order to become “well-pleasing
to God” must be available to all. Accordingly, he contrasts the PRSR, which
is available through the “plain rational faith which can be convincingly
communicated to everyone” (Rel 6:103), with historical faith, which “can
28
Note that I am not here differentiating between “historical faith” and “ecclesiastical faith.” Kant uses
both terms and, while they may not be fully interchangeable, space does not allow me to examine the
subtle ways in which they differ.
29
I have recently argued that the standard First/Second Experiment distinction is based upon a
translation error. See: Lawrence Pasternack, “The ‘Two Experiments’ of Kant’s Religion: Dismantling
the Conundrum,” Kantian Review 22, no. 1 (March 2017): 107–31.
extend its influence no further than the tidings relevant to a judgment on its
credibility can reach” (Rel 6:103). So, while the former, given its universal
availability, is adequate to the theoretical principles guiding his soteriology,
the non-universality of the latter, for Kant, entails that its contents are
instead “intrinsically contingent” (Rel 6:105) – mere “arbitrary precepts”
(Rel 6:106) without any inherent significance to our salvation.
As such, their value may be no more than mere symbols or “vehicles” for the
PRSR, human constructs that provide merely an instrumental value in their
service as “mystical cover” through which the tenets of pure rational faith may
become more vivid. While this may seem the outcome of the text’s (so-called)
“Second Experiment,” an outcome quite inauspicious for a robust “affirmative”
philosophical theology, let me instead suggest that the relationship between
historical faith and the PRSR is actually much more complex, involving at least
four distinct ways in which the contents of the wider sphere relate to the narrower.
To begin with the most obvious of these four ways, Kant clearly presents
some (relatively small) portion of historical faith as overlapping with the
PRSR. In such overlap, the “mystical cover” of historical faith functions as
symbol or vehicle for the soteriologically essential tenets of pure rational
faith. We can see this relationship illustrated most prominently in Part One’s
discussion of original sin, whose rational correlate is the propensity to evil
(Rel 6:31), as well as in Part Two’s discussion of salvation, where Kant
employs, for example, the Pauline imagery of the “old man” and the “new” as
symbolism for the before and after of the change of heart (Rel 6:74).
His explanation for why such vehicles are important can be found within the
ecclesiology developed in Part Three of the Religion. While the tenets of historical
faith often precede in time the insights of pure rational faith, they are nevertheless
to be valued not just as cultural artifacts, but as important to a “natural need” that
we have to give to the “highest concepts and grounds of reason something that the
senses can hold on to” (Rel 6:109). Hence, while any particular of historical faith
may, on its own, be contingent and arbitrary, “some historical ecclesiastical faith
or other, usually already at hand, must be used” (Rel 6:109). In this way, we may
modestly raise the importance of these vehicles, which, while still valued for their
instrumentality, are nevertheless instruments that fulfill an ineliminable need.30
30
This is, perhaps, an overstatement to some extent, for Kant looks forward to a time when “religion
will gradually be freed of all empirical grounds of determination, of all statutes that rest on history” (Rel
6:121). However, this hope may be understood as eschatological in nature, for it concerns a time
“infinitely removed from us” (Rel 6:122) when the church has finally realized its goal of making the
species as a whole “well-pleasing to God” (Rel 6:133).
532 L. Pasternack
Beyond the above, beyond the scope of overlap set by historical faith’s
cognitive service to the PRSR, readers might assume that the remainder of the
wider sphere is dismissed by Kant. That is, one might think of the two spheres
as having a simple binary relationship: the inner portion of the wider sphere,
bounded by the limits of the narrower sphere, corresponds to the domain of
historical faith that is to be valued; while the outer portion of the wider sphere
lacks relevance or is even contrary to the tenets of pure rational faith.
This is not Kant’s position, however. While there are some doctrines that
are anathema to the PRSR, they should not be taken as indicative of the
outer domain as a whole. Instead, doctrines such as the biological inherit-
ability of original sin (Rel 6:40), the Incarnation (Rel 6:64), Vicarious
Atonement (Rel 6:72, 116–17), and the sacrifice of Isaac (Rel 6:86, 187),
which are each rejected by Kant for various reasons, are instead just
members of one particular subset among the others that comprise the
outer domain of historical faith. Hence, rather than envisioning the binary
model on the left, the figure on the right offers a more appropriate
representation of Kant’s views, one that has the doctrines of historical
faith that are anathema to the PRSR as instead just on the outside fringe
of the outer domain:
PRSR
PRSR anathema
anathema
vs. ?
PRSR PRSR
vehicles vehicles
Residing between these two domains, between the doctrines of historical

faith that are vehicles for the PRSR and those doctrines that are anathema to
it, the Religion presents at least two further relationships between historical
faith and the PRSR. The first of these may simply be called PRSR neutral, for
there are, as one would surely expect, numerous doctrines of historical faith
that are neither incompatible with nor suitable as vehicles for the PRSR. The
two examples Kant provides for PRSR neutral doctrines are the virgin birth
(Rel 6:80) and the Trinity – particularly as it is portrayed in the Conflict of the
Faculties (CF 7:39). Hence, we may regard at least some portion of the
intermediate space depicted above as filled by such neutral doctrines.
Lastly, we may add, in addition to its vehicles, a second domain of

historical faith that has a positive relationship to the PRSR. These adjunctives
– let us call them – while they do not directly correspond to any particular
doctrine of the PRSR, nevertheless help to reinforce the worldview of the
PRSR. Rather than offering, as PRSR vehicles do, cognitive support for “the
highest concepts and grounds of reason” (Rel 6:109), the latter fulfill our
social and psychological needs, needs including group-identification, catalysts
for personal reflection, strengthening of hope, and so on. Hence, let us further
add a second domain wherein there is a positive relationship between histor-
ical faith and the PRSR: the first contains those symbols that serve as direct
vehicles for the tenets of pure rational faith; the second has a more indirect
relationship, where its practices contribute to our needs as individuals and as
members of a church, but in ways that reinforce our commitments to the
PRSR rather than directly represent them. We may thus represent our find-
ings through the following enhanced portrayal of Kant’s two spheres31:
PRSR
anathema
PRSR
neutral
PRSR
adjunctives
PRSR
vehicles
In their varying ways, these adjunctives – including rules of piety, rituals, and
other matters of practice – bring definition to the visible church as well as
stimulate the religious consciousness of the individual. Baptism and bris
celebrate the initiation of a new member. Communion and bar mitzvah
serve as the instruments through which one demonstrates one’s commitment
to one’s religious tradition. The adhan, tefillin, mealtime grace, and so on
provide a religious framework within which one can pattern the rest of one’s
day. They serve as occasions for commitment or for its reaffirmation. They
31
It is important to note that where any individual doctrine resides will depend upon its interpretation.
For example, Kierkegaard’s reading of the story of Abraham and Isaac may very well place it as a vehicle
rather than anathema. Similarly, as Firestone himself argues, there are ways of understanding the Trinity
whereby it too can be understood as “mystical cover” for the PRSR. See Chris L. Firestone, Kant and
Theology at the Boundaries of Reason (Aldershot: Ashgate, 2009), ch. 7.
534 L. Pasternack
elevate the major events of life, as well as the trivia of one’s daily routine,
linking the mundane with the sacred.
The question, however, is if in these practices, rituals, and rules of piety, or
likewise, in the vehicles of PRSR, there is any assent to matters supersensible.
While interpreters who prefer to regard Kant’s religious views as involving no
assent would be quite content to leave the contents of historical faith as just
useful symbols and customs, hardly do most religious believers see their
commitments in this way. Certainly most of those who lead religious lives
believe that their rules of piety are given to us through revelation, that many
of our rituals celebrate miracles, and that there is a sacred history that has
recorded God’s many gifts to humanity through time. Let us thus ask
whether for Kant his “room for faith” leaves as well some quarter for
“historical faith,” whether despite its primarily symbolic service to the
PRSR, there is something to be said in favor of it as also a form of
holding-to-be-true. It is to this question that we now turn.
Historical faith as Fürwahrhalten

To begin, let us attend to the ample cautions found within Kant’s discussions of
revelation and miracles. He tells us, for example, that any appeal to “supernatural
intervention” is “very risky and hard to reconcile with reason” (Rel 6:191), and
that it is best to leave “the merit of . . . miracles, one and all, undisturbed” (Rel
6:85). He, moreover, staunchly criticizes the church for demanding congregants
to profess belief in its claims of miracles and revelation (Rel 6:171, 186). Such
demands are a “violation of conscience” (Rel 6:188) insofar as the laity are expected
to profess with certainty claims that cannot be asserted with the “unconditional
confidence” proper to pure rational faith (Rel 6:188–89). Hence, quite unlike the
certainty Kant assigns to the tenets of the PRSR, the contents of historical faith,
even if affirmed, cannot be held in like manner.
Yet, despite these many cautions, Kant has a surprisingly affirmative view
toward miracles and revelation. He writes, for example, that the historical
introduction of a new religion may have been “accompanied and as it were
adorned by miracles” (Rel 6:84), and that it would be “arrogant peremptorily
to deny that the way a church is organized may perhaps also be a special
divine dispensation” (Rel 6:105). Although we have no theoretical grounds
to positively affirm that any alleged miracle or revelation is, in fact, of divine
origin, we can at least take this significant initial step, for quite unlike those
who deny Kant any room for the sacred, we see here that he does not see his
philosophical views as standing in its way. That is, the possibility of miracles
and revelation is wholly compatible with transcendental idealism.
This is, in fact, part of the pure rationalism that Kant develops throughout
the Religion, a theological stance he explains at one point by contrasting it
with the alternative positions of naturalism and supernaturalism. Whereas
naturalism “denies the reality of any supernatural divine revelation” (Rel
6:154), and the supernaturalist holds that “faith in divine revelation is
necessary to universal religion” (Rel 6:155), the pure rationalist will “allow
this revelation, yet claim that to take cognizance of it and accept it as actual is
not necessarily required for religion” (Rel 6:154–55).32 Hence, miracles and
revelation are allowed by pure rationalism, so long as they are not taken as in
themselves essential to our salvation. However, this is somewhat ambiguous,
since the “allowance” here has a number of elements.
First, the preceding quotes indicate that Kant “allows” for the possibility
of revelation and miracles in the sense that their existence is compatible with
transcendental idealism. This should not be particularly shocking, for if God
is affirmed through pure rational faith and God is an agent, then claims of
miracles and revelations are, as Kant himself notes (Rel 6:191), no more
peculiar than allowing the freedom of our human will, as non-natural cause,
to affect the natural world.
Second, given our earlier discussion of the roles assigned to vehicles and
adjunctives within Kant’s ecclesiology, allowance is also given to miracles and
revelation insofar as they are inscribed in symbol and ritual. During the
Seder, for example, one recounts the liberation of the Israelites from Egypt, a
liberation as told in the Haggadah as facilitated by God. That is, in this text,
and, of course, throughout the Bible, we see a representation of divine agency
as shaping human history. So, beyond Kant’s allowance for the ontological
possibility of miracles and revelation, and beyond his allowance for the
instrumentality of symbols and rituals, can those who employ such instru-
ments also hold-as-true their origins in the divine?
One important step in answering this question is to take also into account that
Kant does not merely allow for the possibility of miracles and revelation, but
actually warrants a general affirmation in their actuality. That is, while Kant never
deviates from his claim that it is “salutary to keep ourselves at a respectful
32
Although there is an unfortunate history of confusion regarding the above taxonomy, it should be
evident that Kant’s pure rational faith and Pure Rational System of Religion are of a piece with pure
rationalism as quoted here.
536 L. Pasternack
distance” from any particular claim, he nevertheless affirms the “general presup-
position that grace will work in us what nature cannot” (Rel 6:191). Although
most of his comments regarding divine aid at the individual level are given as
hypotheticals (e.g., “Supposing that [Gesetzt . . . sei] some supernatural coopera-
tion is also needed to his becoming good or better” [Rel 6:44]), we nevertheless
see a much more direct, and in fact quite frequent, affirmation of divine
providence.
We find Kant’s support for providence in texts as diverse as the Critique of
Pure Reason (A743/B771), Perpetual Peace (e.g., PP 8:362), and the
Anthropology (e.g., An 7:328). In the last of these, he writes, “The education
of the human race . . . the human being expects [this] only from Providence” (An
7:328); and in Perpetual Peace, toward the conclusion of a very lengthy review of
the numerous modes of providence (“founding providence,” “ruling providence,”
“guiding providence,” and so on), he turns to the “extraordinary providence” of
“divine intervention or collaboration” (PP 8:361n). After his familiar caveats
related to theoretical reason, he writes, “from a morally practical point of view-
. . . the concept of a divine concursus is quite appropriate and even necessary”
(PP 8:362n). Hence, the necessity by which Kant postulates God, while most
frequently articulated in terms of his role within the highest good as the agent
through which happiness is distributed in accordance with moral worth, we can
see here a further role, one involving God as necessary to bring about this ideal
state of affairs. Through a concursus in history, God underwrites our efforts,
such that through appropriately timed miracles and revelation, our eventual
progress toward the highest good is guaranteed.
In this way, understood as an extension of the postulation of God for the
sake of the highest good, we may understand providence as a further element
of the PRSR, and thus another tenet within the spectrum of pure rational
faith. Yet there is something distinctive about this tenet’s relationship to
historical faith. Unlike most other elements of historical faith, there is a
convergence between symbol and signified here, for the sacred history of a
tradition is just the particularization of the doctrine of providence. Hence, if
one were to affirm as a matter of historical faith that God parted the Red Sea
as the Israelites fled from Egypt, this is not merely a “mystical cover” for
some unrelated element of the PRSR, but rather is a particular claim behind
which there is a general principle that is held as true.
This general principle is also of distinct importance to Kant, for it is vital to the
hope we are to have in the highest good. Its representations of a sacred history
depict the highest good as a telos inscribed in history, such that its realization does
not spontaneously come into being but rather emerges as the outcome of a
process. As such, historical faith’s portrayal of history carries a reflective function,
one that inscribes our hope and one through which that hope is fostered by
cultivating a worldview open to the divine. Rituals, in particular, are significant in
this regard, for just as prayer can help enliven “the disposition to a life-conduct
well-pleasing to God” (Rel 6:198), so a ritual celebrating an alleged miracle, or
likewise a rule of piety that is represented as given through a “special divine
dispensation,” elevates the mundane into the sacred. Thereby, we may see our
lives as part of a history advancing toward the highest good, in turn reinforcing
our commitment to this ideal.33
Analogous to the Metaphysics of Morals’s proposition that sexual desire is to
be inscribed within the rationality of a marriage contract (MM 6:278), so ritual
and prayer likewise elevate the mundane. For the result of a meal ought not
merely be one’s nourishment, but insofar as one fully takes upon the highest
good as the “special point of reference for the unification of all ends” (Rel 6:5),
then the significance of the meal, like all else, is so elevated. Historical faith,
thus, takes the general holding-to-be-true in providence and gives us a way to
enliven our commitment to the same general truth by choosing for it a
particular form, one that transforms the abstract commitment of pure rational
faith into a concrete commitment of life-practice. In so doing, historical faith,
particularly its adjunctives, provide the means through which we can reflectively
portray our lives as participating within a sacred history, and in turn, draw from
this projection a strengthening of our devotion to its end.34
Conclusion
In this chapter, I have sought to examine Kant’s conception(s) of faith first
by discussing the core structure of pure rational faith within Kant’s taxon-
omy of propositional attitudes, and then by considering the positive status of
33
In this we see as well the “voluntary determination” that Kant associates with pure rational faith.
Although many people simply inherit the historical faith of their families, this form of faith too allows
for choice insofar as each individual may (and should) choose which tradition’s vehicles and adjunctives
he or she finds most inspiring.
34
The faith at issue here should not be rendered as merely a belief “as-if.” Despite the popularity of this
construal of the postulates in prior decades, the work that has recently been done on Kant’s conception
of pure rational faith should now make it clear that in it there is a full-fledged holding-to-be-true, a
conviction with certainty, rather than some sort of practically grounded self-deception. If such were true,
then morality would not merely be threatened by self-deception, but it would also be a consequence of
it. Such a reading, thus, collapses into absurdity.
538 L. Pasternack
historical faith in Kant’s philosophical theology. In doing so, I have sought to

navigate a course between the Scylla of those reductionist views that treat
religion as just our moral ideals in an “imaginatively enhanced form” and the
Charybdis of the “metaphysically friendly” interpreters who believe that
Kant’s religious views can only be salvaged by granting exception to the
restriction thesis.
As I have argued, both views overlook what Kant actually proposes, as well
as how and why he supports faith as not only a legitimate mode of assent,
but the one proper to religious belief. At minimum, Kant maintains that we
have objectively valid grounds for faith in the postulates and the highest
good. Further, I have argued as well that we may see his commitment to
providence as part of the PRSR and so as further postulated on behalf of the
highest good.
While this does not itself justify a belief in any particular claim of God’s
activity in time, rituals and practices which celebrate a sacred history more
deeply invest us in a worldview that strengthens our commitment to the
highest good and to the hope that it provides. Hence, consonant with Kant’s
portrayal of our duty to promote the highest good as the foundation of his
ecclesiology, so likewise ecclesiastical practices, rituals, and rules of piety, the
adjunctives of historical faith, can likewise be understood as means through
which we strengthen our commitment to this ideal and to the end that it sets
for one and all.
24
The Fate of Religion within the Boundaries
of Mere Reason
Martin Moors
This chapter on Kant’s philosophy of religion intends to fit architectonically

with the two perspectives by which his critical thinking on religion is
internally organized. The first perspective determines the method and
hence the purpose of his inquiry; the second defines the content. Kant’s
critical method by which he intends to define “genuine religion [eigentliche
Religion]” (Rel 6:12) can at the very outset be identified with the expression
that figures in the title of his chief work: “within the boundaries of mere
reason [innerhalb der Grenzen der blossen Vernunft].” Although its meaning is
rather puzzling, this expression still clearly indicates a methodological orien-
tation and a topological location as well. In his dealing with religion, Kant’s
philosophical focus is intentionally oriented toward “mere1 reason’s” grasp of
it. An adequate understanding of the methodology of this grasping has to
1
I prefer the Cambridge edition translation “mere” of the German bloss, despite Pluhar’s and
Palmquist’s arguments for “bare.” See Immanuel Kant, Religion within the Bounds of Bare Reason,
trans. Werner S. Pluhar (Indianapolis: Hackett, 2009); Stephen R. Palmquist, Kant’s Critical Religion
(Aldershot: Ashgate, 2000); and Stephen R. Palmquist, Comprehensive Commentary on Kant’s “Religion
within the Bounds of Bare Reason” (Malden, Mass.: Wiley-Blackwell, 2016). I base my opinion on Kant’s
own literal indication (at Ak 18:90 [R5107]) that bloss does not refer to the boundaries (Schranken) of a
“bare body” (in this case, of rational religion) but rather to the “act of setting boundaries [actus der
Einschränkung],” which exactly corresponds to Kant’s methodological statement: “I only intend to
determine the boundaries of the sensible and empirical in [religious] faith and of those of reason as well”
(Ak 23:91).
M. Moors (*)
Department of History, University Leuven, Leuven, Belgium
e-mail: martin.moors@hiw.kuleuven.be

540 M. Moors
take into account, first, the fact that reason – as the faculty of universal
principles – has paradoxically a limiting (critical) impact upon the truth
claims of what is subjected to it, and, second, the fact that religion will be
transferred to a location (a “within”) that is circumscribed by reason’s
immanent principles. Transferral and limitation, or subjection and criticism,
are the main operative terms characterizing the way (the method) along
which Kant’s thinking on religion proceeds toward a determination of
both the truth and the content of “the pure religion of reason” (Rel 6:12).
Religion under critique, formally (method)

and materially (content)
Esteemed for being a most important aspect of human life2 and thus an
equally important subject matter for philosophical examination, Kant trans-
fers religion to the domain3 where self-enlightened reason autonomously sets
its rules and principles of truth. Subjected to this setting, religion has to
undergo the criticism that separates, in general, truth from illusion, in this
case the truth of “religion proper” from religious delusions, enthusiasm, and
fanaticism. As to content, reason’s critique consequently targets all specific
aspects that, according to Kant’s conception, pertain to the general concept
of religion. These specific aspects can be identified by referring to the four
parts that comprise Kant’s “philosophical doctrine of religion” (Rel 6:17). In
Part One (“Of the radical evil in human nature”) (Rel 6:19–52), religion
enters on the philosophical scene when the issue of moral conversion is at
stake: “the restoration to its power of the original predisposition to the good” (Rel
6:44).4 Here the first specifically religious content comes to the fore, namely
grace or “supernatural cooperation” (Rel 4:44), treated as “a secondary
2
“For a human being it is impossible to enjoy his life without religion” (Ak 19:649 [R8106]).
3
“Domain [Gebiet]” is used as a technical term in accordance with CJ 5:174: the territory in which pure
concepts legislate.
4
It is surprising that Kant is bringing the first explicitly religious item (grace) to the fore in a General
Remark (on conversion) at the end of Part One (Rel 6:44–52), which deals with radical evil in human
nature. At the end of this first General Remark and in a preview to other General Remarks following at
the end of the next Parts, he classifies the issue of grace as a first parergon, “a secondary occupation
[Nebengeschäft]” that “border[s]” on “religion within the boundaries of pure reason” but does not
intrinsically belong to this domain (Rel 6:52). The structural location of this first typically religious item
by which he opens a work on religion might be surprising at first sight. But the fact that, despite this
strange location, the issue itself introduces a first notion of critique (“bordering on” versus “not
belonging to”) is significant because – as we will see – it is Kant’s critical intention that will from
now on dominate the whole work.
24 Fate of Religion within the Boundaries of Mere Reason 541
occupation (parergon)” (Rel 6:52; see also CJ 5:226) that is extrinsic to the
topic, as one of the “extravagant ideas” which “border on” the pure moral
theme of conversion (Rel 6:52). Part Two investigates the religious presenta-
tion of “the personified idea of the good principle” (Rel 6:60), that is, the
issue of “practical faith in this Son of God,” or Kant’s philosophical
Christology (Rel 6:62). Part Three deals with the communal facet of religion,
that is, “the founding of a kingdom of God on earth,” or Kant’s philosophical
ecclesiology (Rel 6:93). Part Four concerns “religion and priestcraft” referring
to service to God (cultus) (Rel 6:151).
Kant’s criticism subjugates these four facets of what pertains to religion in
general, to a veridical judgment passed by “mere reason” according to
reason’s immanent principles. However, examining more closely the meth-
odology by which Kant presents this critique’s impact on (the truth of) each
of these facets, we discover a most significant complication. From the view-
point of the effects, thus with regard to the outcome of reason’s judgment
about truth and illusion in matters of religion, I will argue that the operative
mode of Kant’s critical method must be interpreted in a nuanced way.
Indeed, Kant’s critical dealing with religion is operating equivocally. This
means that his critique takes up two logically different positions for assessing
truth claims with respect to religious representations and actions.
Establishing in a dual critical way the logical possibility

of truth in religion
From the viewpoint of its effects, Kant’s critical method operates in two
logically different modes. In terms of criticism, what is happening, effectively,
once the content of religion is transferred from a naive unenlightened faith to
the “within” of the boundaries of mere reason, must be judged in the following
nuanced way. The nuance is based on the difference that Kant’s critique
effectively installs between an antithetical opposition and an interim reservation.
The truth of “religion proper” can logically be considered to be either a truth
that is antithetically – in principle – opposed to what contradicts the universal
principles set by mere reason, or a truth that – in concreto – can without
contradiction be acknowledged to dwell within given religious creeds and
practices proper to the contingent and particular historical conditions of a
human being’s religious behavior. Kant’s criticism is operating according to
these two modes with regard to the already mentioned four essential compo-
nents of his “philosophical doctrine of religion.” In other words, the principles
that legislate in the “domain” of the “within the boundaries of mere reason,”
542 M. Moors
and that thus determine what must be taken to be “religion proper,” reject
antithetically all that in concrete religion is opposed to them but, importantly,
these principles also allow what in contingent religious practices and creeds can
(for the time being) coexist with them. Hence, from a veridical vantage point,
the significance of the critically operative term “boundary” differs according to
the modality in which religious creeds and practices are positioned vis-à-vis
reason’s limiting immanent principles. Two positions are possible: either – in
principle – these religious creeds and practices homologously (per se) conform
with these principles (as it necessarily should within the critical boundaries set
by mere reason) or – in concreto – they are contingently and transitionally as yet
non-homologous with them.
Consequently, in Kant’s critical judgment on religion, the orthos of
religious orthodoxy and orthopraxy varies according to these logical mod-
alities of the relation that religion entertains with reason’s determinative
power. Either religion (faith and cult) will in a homologous form fit with
reason’s determining principles (reason’s “domain”) and obey its principles
which demand a (to-be-established) necessary conformity, a conformity that
essentially marks “religion proper”; or, in practicing his or her faith and cult,
the religious person resides in a “territory”5 that is contingently designed after
historical or revealed determinants. In the latter case, the person’s religious
faith and cult can – ad interim – very well have the mark of truth (orthos)
within the confines of a contingently established historical or revealed
environment, though, without exception, this contingency remains princi-
pally subjugated to the critique of reason’s necessitating principles.
In sum, in his critical treatment of religion, Kant’s methodology is putting
at an antithetical distance all those (marks of) given religious creeds and
practices that cannot possibly fit with the universality of reason’s immanent
principles. On the other hand, as concrete religious creeds and practices are
in their particularity contingently marked by (historical, revealed) features
and thus are not per se homologous with reason’s principles, Kant is assessing
them from the distance of an interim reservation. If they can stand well the
test of reason’s criticism and hence can (veridical possibility) homologously
be adopted “within” the boundaries of mere reason, they all will in a certain
way episodically6 share the truth of “religion proper.” Kant’s methodology
5
“Territory [Boden]” is used as a technical term (CJ 5:174): the extension of concepts that can possibly
become determined cognition.
6
An interesting case of metaphorics is presented here. Corresponding to the distinction I propose to
make between antithetical opposition and interim reservation, the various metaphorical schemes that are
used by Kant can be a cause of uneasiness. With respect to antithetical opposition, the topological
relates7 the epistemic paradigm – which is a priori – of antithetical opposition

(between truth and delusion) with the topological (or temporal) paradigm of
(the furtherance of) congruent homology (between the empirical and the
pure rational, or from the contingent toward the pure a priori).
Thus far I have scrutinized, from the point of view of logical modality
(necessity and contingency), the methodology of Kant’s critical attitude
toward religion. Our examination has revealed how the logical possibility for
religion to be true is established in two different ways: in principle and in
concreto, referring, on the one hand, to the necessity of reason’s lawgiving and,
on the other hand, to the contingency of historical contexts.
Given this stratified determination of the logical possibility of truth in
religion, and, in addition to this, taking into account the two dimensions
(faiths and practices, representations and actions) by which in general the
object “religion” materially subjects itself to Kant’s critique, it is now
important to investigate how Kant, “as purely a teacher of reason” (Rel
6:12), establishes content-wise the real definition (Realdefinition) of
religion.
Defining the real possibility of truth in religion
In defining something, what counts in the logical respect also applies in

respect of the real. The fact that “mere reason” is considered – logically – to
put limits (around the island of truth) may be interpreted literally as reason’s
metaphor used in the Critique of Pure Reason is very well appropriated to also be used in the current
context of his critique of religion: “[the land of truth] is an island . . . surrounded by a vast and stormy
ocean, where illusion properly resides and many fog banks and much fast-melting ice feign new-found
lands” (B294–95). Applied to religion: “the land of truth” refers to “religion proper” where universal
principles of reason legislate and determine the place for and significance of each religious concept
(grace, Son of God, church, cult). As an island of truth, “religion proper” is surrounded by an ocean of
religious enthusiasm, delusions, superstitious faiths, and counterfeit services. On the other hand, to
represent in the religious context the meaning of interim reservation, we do not think that the other
topological metaphor, used in CJ 5:174, of “domain [Gebiet]” (see note 3 above) as a section of
“territory [Boden]” (see note 5 above) is appropriate. It is also not appropriate to use the metaphor of
the two concentric circles by which Kant locates revelation and scripture (inasmuch as contingencies are
involved) in the wider sphere of faith inside which is critically circumscribed the pure religion of reason
(Rel 6:12–13). The reason why we raise doubts against these (surveyor’s and geometer’s) topological
metaphors is the fact that they do not fit with what Kant explicitly suggests regarding history-based
religion in its veridical move toward “the pure religion of reason” following the dynamics of the
Enlightenment (Rel 6:12; see also OT 8:144–46; WE 8:40–42). In this regard, temporal expressions
such as “transitory,” “episodic,” and “interim” seem more privileged as schemes expressing Kant’s
persisting emphasis on “furtherance of a pure faith of religion” or “gradual transition” at the basis of
which he lets prevail the dynamics of hope (Rel 6:115).
7
Later on we will see how this relation takes the form of superimposition.
544 M. Moors
role of defining (de-finire). By limiting, reason defines. Conversely, by

defining, reason is establishing limits. This also holds with regard to
the real definition. Defining truth in real terms – on the basis of a real
definition – one relies on reason putting its limits around its proper domain
inasmuch as its principles become objectively real. We may assume this on
the basis of Kant’s dealings with “the ancient and famous question”: “What
is truth?” (A58/B82). He argues that establishing the sheer logical possibility
of truth “is far from sufficing to constitute the material (objective) truth of
the cognition”; rather, we must draw on “antecedently well-founded infor-
mation about [objects] from outside of logic” (A60/B85).
This is exactly what Kant in his doctrine on religion is doing in order to
establish a “pure religion of reason” in terms of objective reality (Rel 6:12–13).
Indeed, he will “go outside logic” and draw on the domain of pure practical
reason wherein reason’s principles are (destined to become) objectively real.
From within this domain, Kant will, indeed, obtain “well-founded [in casu
practical] information about [moral] objects” (A60/B85), such as “good will,”
“duty,” “virtuousness,” “highest good,” and “kingdom of ends,” in the wake of
which he will define in real terms the moral truth of religious signifiers such as
“grace,” “Son of God,” “church,” and “service.” Already hinted at in the
Critique of Pure Reason (A819/B847; see also LRT 28:1102) but most expli-
citly raised in the Critique of Practical Reason, Kant gives a stipulative real
definition of religion that in all his subsequent works remains unaltered:
“religion, that is, . . . the recognition of all duties as divine commands” (CPrR
5:129; see also Rel 6:153; MM 6:440, 443, 487; CF 7:36; CJ 5:481; OP8
21:17, 20, 28, 50; OP 22:115–17, 120–21, 125). The full extension of this
real definition of religion – its real content presenting (darstellen) its real
truth – coincides with “all duties” which, in turn, are defined by the imperative
principles of pure practical reason. The initial methodological transferral of
religion, its creeds and practices, into the “within the limits of mere reason”
thus, regarding real content, amounts to religion defined as “moral religion, i.e.
the religion of good life-conduct” (Rel 6:51). The definiendum “religion”
receives in this way its definite real status (“religion proper”) within the
“domain [Gebiet]” of pure practical reason’s legislation (CJ 5:174).
8
Particularly interesting with regard to the Opus postumum is that it testifies how Kant in the last years of
his life (between April 1800 and February 1803) was intensely dealing with the issue of “human duties as
if divine commands.” He does not explicitly treat it as a theme that introduces the real definition of
religion. He rather investigates (making use of analogy) the synthetic link between pure practical self-
positing (Selbstsetzung) (“I am”) and the moral-practical idea of God (OP 22:115–31).
Furthermore, all particular, contingent creeds and actions within the

“territory” of historical religions will be critically checked on their veridical
nature by their homology with these pure practical principles of morality.
Put differently: the fundamental law of pure practical reason as formulated in
CPrR 5:30, “So act that the maxim of your will could always hold at the
same time as a principle in a giving of universal law,” is, by way of real
definition, “regarded as” (CPrR 5:129) “the extremely pure moral law of our
religion” (A817/B845). Kant’s methodology in both its aspects – logical and
real – reaches its full goal in this recognition (“the seeing as”) by which
religion in general is thoroughly defined and, further on, in the subsequent
application of this moral rule of objective truth with respect to all (four)
religious topics.
Morality leads – four times – inevitably to religion
In the following section, I will explain how Kant, under the rule of morality –
being the real principle of truth for “religion proper” – relates to four
moments of his moral theory. The aim is to show how it “inevitably leads
to religion” (Rel 6:6).9 As indicated earlier, these four facets of religion,
which are subjected to the moral principle of truth, are: divine cooperation
(grace), Son of God, church, and cult. The order of these four parts is not
arbitrary. On the contrary, in the preface to the first edition of the Religion,
Kant claims that he will establish “rigorous coherence of the materials in this
work” and consequentially states that “the three essays now to be added
[Parts Two, Three, and Four] [contain] the complete development of the
first [Part One]” (Rel 6:11). This is of utmost importance as a principle
of interpretation. Part One (on the moral issue of “radical evil in human
9
“Morality thus inevitably leads to religion” (Rel 6:6). On my interpretation, the “leading to” is realized
according to two different motives. The first motive emerges from a reflection on “the concept of the
highest good, as the object and final end of pure practical reason” (CPrR 5:129). Through a reflection
on the final object of the moral law, morality leads to religion. We may call this first motive the objective
motive of finality of the moral law. A second motive is reflective too, though this reflection is led by a
different aim. Morality, namely, is reflectively evoked for bestowing upon religion its real mark of truth.
This second mode of “leading to” is performed by the tropic operation of the “recognition/seeing as”
that originates from the real definition of religion (see CPrR 5:129). According to the veridical motive of
the “recognition as,” morality leads to religion. This second motive can be called the veridical motive of
the semantic trope of “recognition/seeing as.” In the section that we are introducing, we do not
investigate the “morality . . . inevitably leads to religion” statement according to the first “objective”
motive of finality. Because I am examining the real content of Kant’s philosophical doctrine of religion, I
must rather explore, along the second mode of reflection, which (four) moments from his moral theory
will “inevitably” lead to the recognition as being religious (Rel 6:6).
546 M. Moors
nature” and the moral obligation of conversion) sketches the initial moral
coordinates – that is, the human condition (“human nature”) in which
religion in all its facets plays its proper role. “The command that we ought
to become better human beings” – through conversion from evil – thus
becomes the key to understand the real signification of all facets of religion in
a human being’s moral life (Rel 6:45).
In order to achieve this command of the moral law, Kant necessarily
presupposes, first, the pure moral guiding principle of “the personified idea of
the good principle,” which represents the central part of Part Two (Rel 6:60–
78). According to the religious trope of the “seeing as,” this Idea[l] of moral
perfection becomes clothed with religious attributes such as “God’s only-
begotten Son, ‘the Word’ (the Fiat!) through which all other things are, . . . ‘the
reflection of [God’s] glory,’” and so on (Rel 6:60). Kant’s philosophical
Christology has no other veridical role to play than helping us “to elevate
ourselves to this ideal of moral perfection, i.e. to the prototype of moral
disposition in its entire purity” (Rel 6:61). But a philosophical Christology,
though internally necessary for conversion from evil on the way to moral
perfection, is not a sufficient condition for reaching this goal.
In Part Three of his book, (“The victory of the good principle over the
evil principle, and the founding of a kingdom of God on earth” [Rel 6:93–
107]), Kant focuses on the necessary external condition. For him, a
human being is not constantly exposed to the assaults of the evil
principle merely from within his inner “natural” propensity to evil, in
this case the frailty, impurity, and depravity of his heart (Rel 6:29–30).
Rather, in this perilous state, every human being is tempted to give way
to the attacks of the evil principle “as soon as he is among human beings”
(Rel 6:94). Not only malevolent people or bad examples awake in
someone the bad intentions of his will; rather, “it suffices that [other
human beings] are there, that they surround him, and that they are
human beings, and they will mutually corrupt each other’s moral dis-
position and make one another evil” (Rel 6:94). Hence, Kant concludes,
the necessary external condition for the furtherance of morality through
conversion amounts to “the setting up and the diffusion of a society in
accordance with, and for the sake of, the laws of virtues” (Rel 6:94).
Such an association is called “an ethical community,” “an ethico-civil
society,” or “an ethical state, i.e. a kingdom of virtue” (Rel 6:94–95). It is,
again, according to the veridical motive of the semantic trope of “seeing
as divine command” that the representation of ethical community
becomes an essentially religious issue, in this case “people of God” who
represent the ecclesiological facet of “moral religion” (Rel 6:99).
Which argument urges us to perform this veridical shift – morality leading

to religion – regarding this particular ethico-religious communal issue? Kant
claims that what ought to be realized, ideally, is the constitution of an ethical
state in which God is “seen as” the lawgiver. Through the intermediation of
such a constitution, a communal will of the people should be brought into
practice as determined by a communal moral legislation. The virtuous
constitution of individual wills in a distributive unity will not suffice in
battling against the evil principle. What is at stake in a communal perspective
amounts to the establishment of a collective unity, an omnitudo collectiva or,
in this form, an “invisible church” (Rel 6:157).10 Such a communal unity has
the peculiar relational form of a community-through-reciprocity, which, for
its possibility, must always be based on the synthesis with a transcendent
principle.11
According to Kant, the constitution of an invisible church includes the
following two elements. With regard to the communal aspect, the con-
stitution of such a state is made up of what is shared by all its members
or citizens of the divine state, that is, pure practical reason’s moral
legislation. Second, regarding the founding transcendent principle, the
constitution depends on the common recognition of the divine legislator.
The invisible church, ideally constituted by these two determinants,
represents the necessary condition for bringing into practice what is
categorically commanded by the moral law, namely the duty (officium
sui generis) of the human race toward itself (Rel 6:97, 151): “the
promotion of the highest good as a good common to all” (Rel 6:97).
This common end can only be achieved by commonly breaking the
power of the invisible enemy, an achievement that necessarily requires
the religious reference to God.
The last chapter that belongs to Kant’s philosophical doctrine of religion
(Part Four) concerns service (cultus) “seen as” service of God. This issue is
10
By omnitudo distributiva is meant, in general, the form of an aggregate of elements which are all
(analytically) marked by a common property (Ak 17:354–55 [R3936]). In terms of religion, it is “the
rational religion of single individuals” who share a “communality of insight [allgemeine Einhelligkeit]”
(Rel 6:157–58). By omnitudo collectiva is meant the form of a community of elements which, for their
communality, is constitutively (synthetically) dependent on a common ground (Ak 17:434 [R4149]). In
the Religion, it is “the union of the believers in one (visible) church according to principles of a pure
religion of reason” (Rel 6:158). For the difference between distributive and collective unity, see also
A582/B610, A644/B672.
11
For an elaborated argument, see my “Die Bestimmungsgestalt von Kants Gottesidee und das
Gemeinschaftsprinzip,” in Proceedings of the Sixth International Kant Congress, ed. Gerhard Funke and
Thomas M. Seebohm, vol. 2, pt. 1 (Washington, D.C.: University Press of America and Center for
Advanced Research in Phenomenology, 1989), 49–65.
548 M. Moors
intrinsically connected with both the Christological Part Two (that is, with
the internal moral condition of becoming “well pleasing to God”) and the
ecclesiological Part Three (that is, with the external societal condition of
realizing, as God’s kingdom, the moral highest good for all). As connected
with the Christological condition, human beings under the dominion of the
good principle and, hence, with respect to the moral character of their wills,
can be “seen as” servants of the pure religion of reason: “the pure religion of
reason will have all right-thinking human beings as its servants [Dienern] (yet
without being officials [Beamte])” (Rel 6:152–53). By the veridical motive of
“seeing as,” morally well-disposed persons become, by definition, identified
as servants of God in a real sense. From this Christological point of view, the
moral service of God in a pure religion of reason is truly “a free and hence
moral cult [moralischer Dienst]” (Rel 6:179).
The principal aim of all religious service – “becoming well-pleasing to
God . . . [in] the striving for a good life-conduct” (Rel 6:178) – can thus
in principle be met in an absolute manner (Rel 6:179). The moral service
of God can unconditionally be a true one if it is based solely upon the
pure moral disposition of the human being’s virtuous heart. The role of
the ecclesiastical community (the invisible church) is to assist in the
increase of this moral disposition. The purely moral establishment of
the external societal condition of all right-thinking human beings is
needed merely to enhance – through conversion – their inner virtuous-
ness. Consequently, the service of the invisible church can only be an
indirect one, namely in creating the communal moral unity of well-
disposed human beings as a necessary condition for combating the
common enemy of evil. This is why Kant states that “service under the
dominion of the good principle in the invisible church cannot be
considered as ecclesiastical service, and that the religion of reason does
not have legal servants who act as the officials [Beamte] of an ethical
community” (Rel 6:152). In principle, no ecclesiastical statutory arrange-
ment is directly needed for the “true servants (ministi)” (Rel 6:165) to
acquire a “true religious disposition” (Rel 6:201) by which God might be
served “in the Spirit of God who guides us into all Truth” (Rel 6:112,
also Rel 6:146n). Moral cult “seen as” pure moral service of God is
practiced in fear and love for his law (Rel 6:182). They are the true bases
and expressions of worship (Rel 6:7). Together with their “effects” in the
moral person (divine blessedness [Gottseligkeit] [Rel 6:182] and grace
[Rel 6:171, 174]), they are, for Kant, in no way principally mediated
by any ecclesiastical public arrangement.
Criticism at work: Grasping the four essentials

of moral religion
The following section will apply the findings that I presented on Kant’s
methodology of critique (in the section “Establishing in a Dual Critical Way
the Logical Possibility of Truth in Religion”) to the four real facets of religion (as
presented in the previous section, “Morality Leads – Four Times – Inevitably to
Religion”). How does Kant’s critique effectively take action with regard to the
content of the four facets of his moral religion: conversion through “super-
natural cooperation” (grace), Christology, ecclesiology, and divine service?
Remember what I argued earlier: the method by which Kant’s critique seizes
this religious content operates according to two modes understood as modes of
critical limitation, antithetical opposition (repudiating all that contradicts the
principles of pure moral religion) and interim reservation (assessing in revealed
historical religion what has a moral element of truth).
Kant’s critique assessing the truth of the religious theme

of grace
The issue of moral conversion from evil concerns “the restoration to its power of
the original predisposition to the good” (Rel 6:44). Assessed as “surpass[ing] every
concept of ours,” the categorically obligated “ascent from evil back to the good”
brings to light human reason’s finitude and incompetence to achieve this
obligated goal (Rel 6:45). Vis-à-vis this incompetence – which is not due to
lack of courage or perseverance, nor because of the frailty of human nature or
the impurity of the human heart – human reason (recta ratio) in its finitude (Rel
6:52) “feels its need”12 to appeal for a “higher assistance” or “some supernatural
cooperation,” that is, divine grace (Rel 6:44–45). In order to expose how Kant’s
critique concretely impinges upon this first religious key-concept of divine
cooperation as related to the major moral issue of conversion, we must note
that this concept cannot possibly be located either within the domain of
objective knowledge (because it transcends the boundaries of the understand-
ing’s use of the category of causality) or within the practical act of maxim-
making of our will (because “assistance” is, by definition, not of our doing).
Consequently, reason’s critique of the religious concept of grace warns of a
double form of enthusiasm (Schwärmerei): the theoretical enthusiasm in idly
12
On das Gefühl des der Vernunft eigenen Bedürfnisses, see OT 8:136–39 and CPrR 5:142.
550 M. Moors
pretending to know “what God does” and the practical enthusiasm in promot-
ing a so-called divine action that would substitute for the free praxis of moral
maxim-making of the human autonomous will. Kant’s critique strictly con-
trasts the concept of grace with these delusions that arise if one too enthusias-
tically dares “to realize” this concept in granting it any (in this case practical)
objective reality. Hence, the noumenal act of conversion from evil may never be
considered to be directly and efficiently an “effect of grace [Gnadewirkung]”
(Rel 6:53). Seen in a larger perspective, reason’s critique only allows a “moral
religion” in which no heteronomous elements play any efficient role.
Notwithstanding this principal expulsion of the concept of grace from any
domain of objective reality (neither the theoretical nor the practical), it can
very well “host” an unalterable religious truth. By Kant’s main principle of
critique, the representation of “divine assistance” for the moral act of con-
version is not suddenly transfigured into an illusory phantom-concept or a
nothing (ens rationis)13 pertaining to religion. As it functions in its multiple
significations14 in the territory of, for example, Christianity, it can without
contradiction, within the context of historical religion, give us genuine
religious truth. The main critical condition for claiming this truth is for
Kant its practical relevance for the furtherance of a moral aim, in this case
conversion. As long as that pure moral focus – its basis of truth – is not yet
fully enlightened, the concept of grace in all its various significations remains
in the critical sphere of interim reservation. Kant allocates its practical truth in
a dimension that specifically characterizes its religious meaning, namely the
dimension of hope15: “only . . . if [a human being] has made use of the
original predisposition to the good in order to become a better human
being, can he hope that what does not lie in his power will be made good
by cooperation from above” (Rel 6:52). Considered from an epistemic
viewpoint, this religious hope is wedded with rational faith. In matters of
faith, the critique discards absolutely all dogmatic faith (Schwärmerei) that
would compete with knowledge regarding objective reality. However, as a
specific mode of assent,16 reason acknowledges that,
13
In the terminology of the Critique of Pure Reason: a “nothing” is an “empty concept without object”
(A292/B348).
14
See Johann Auer and Joseph K. Ratzinger, Kleine katholische Dogmatik, vol. 5: Das Evangelium der
Gnade (Regensburg: Pustet, 1972).
15
The question “What may I hope?” was already assigned to be specifically religious (see A805/B833–
34; JL 9:25).
16
See A820–31/B848–59, especially A828–30/B856–58 on moral faith; CPrR 5:142–46; OT 8:146; JL
9:65–70.
if in the inscrutable field of the supernatural there is something more than [reason]
can bring to its understanding, which may however be necessary to make up for its
moral impotence, reason even counts on this something being made available to
its good will even if uncognized, with a faith which (with respect to the possibility
of this something) we might call reflective [faith]. (Rel 6:52)
As in the act of reflection on the conditions of the possibility of the highest

good through which reason postulates (in faith) immortality of the soul and
the existence of God (CPrR 5:142–46), in the same way reason reflectively
judges about the conditions of the possibility of conversion. And moreover,
completely in tune with Kant’s claim concerning the postulates, that they are
an “assent from a need of pure reason,” reason’s assent can also be applied to
the religious concept of grace (CPrR 5:142–46). Stemming from a felt need
of pure practical reason which implies that reason judges following its
subjective condition (CPrR 5:145), the assent based on reflection is by
Kant called “a pure practical rational belief” (CPrR 5:146). Applied to the
practico-religious issue of grace, the following statement adequately deter-
mines the critical conditions concerning reason’s approval of it by faith:
the principle that determines our judgment about it [in this case, grace], though
it is subjective as a need, is yet, as the means of promoting what is objectively
(practically) necessary [namely, conversion], the ground of a maxim of assent for
moral purposes, that is, a pure practical rational belief. (CPrR 5:146)
These critical conditions – the subjective (if reason “feels its need”) and the
objective (if required for the realization of what is morally necessary) – will be
discounted with respect to all matters of religious faith, exposed either in
moral religion or a historical religion of revelation.
Kant’s critique assessing the truth of representations

of Christology
In Part Two of his Philosophical Doctrine of Religion, the section that deals
with “the personified Idea of the good principle” “seen as” God’s only-
begotten Son (Rel 6:60), Kant subjects various representations of
Christology to his two modes of critique. The various facets of Christology
are: Logos-Christology (Rel 6:60), incarnation-Christology (Rel 6:63–64),
kenosis-Christology (Rel 6:61), Christology of salvation (Rel 6:61, 128–29),
Christology of preexistence (consubstantiality) (Rel 6:60), and Christology of
creation and eschatology (Rel 6:60). As these facets of the Christological
552 M. Moors
issues that he will discuss make clear, Kant significantly restricts de facto his
philosophical Christology to the confines of an (interpretation of the) arche-
type of a human being well-pleasing to God on earth, that is, a jesuology. He
states: “We cannot think the ideal of a humanity pleasing to God (hence of
such moral perfection as is possible to a being pertaining to this world and
dependent on needs and inclinations) except in the idea of a human being”
(Rel 6:61). The reason for this restriction to jesuology already foreshadows
Kant’s critique: “With [Jesus’ death on the cross] the public record of his life
(which can therefore also serve universally as an example for imitation) ends.”
Consequently, “the more esoteric story of his resurrection and ascen-
sion . . . added as a sequel and witnessed only by his intimates, cannot be
used in the interest of religion within the boundaries of mere reason” (Rel
6:129n).
The restriction of his Christology to a jesuology is itself subjected to a
second restriction, namely by detaching within jesuology all historical fea-
tures of Jesus’ life.17 This last restriction exemplifies Kant’s principle aim of
critique: the final veridical motive of the critique of religious representations
of Christology consists in an intrinsically moral assessment of those repre-
sentations for the sake of which – in principle – all historical contingencies
should be put aside.18 That is, the only facts of Jesus’ life that matter are ones
that exemplify the moral Ideal, not the details of his specific historical
circumstances. Kant’s philosophical thinking on the moral Ideal is framed
within “the battle of the good against the evil principle,” and ultimately
conceived in view of the human being’s duty to convert (Rel 6:57). To
win this battle, the philosophical representation of the moral Ideal has to fit
into the following coordinates in order to be relevant: “human beings cannot
form for themselves any concept of the degree and the strength of a force like
that of a moral disposition except by representing it surrounded by obstacles
[of evil nature] and yet – in the midst of the greatest possible temptations –
victorious [in its will to convert]” (Rel 6:61).
17
This second restriction has significant implications for Kant’s theory on scriptural exegesis: “Historical
cognition that has no intrinsic relation, valid for everyone, to this [moral improvement], belongs among
the adiaphora, which each may treat as one finds edifying.” Our searching for historical features of the
life of Jesus will result in “only a barren addition to our historical cognition” (Rel 6:43n; see also 110).
18
To be clear: this “putting aside” does not amount to an abstractive operation which would transform
Kant’s critique into a logical act of abstraction. In that case we would not even be able to consider the
purified/abstracted figure as that of a moral Ideal defined by mere reason. Moreover, by such an
abstractive operation, one would not reach any justifiable ground for claiming universal worth or
unconditional obedience regarding the abstracted idea.
How will Kant’s critique, according to its two modes, assess the
Christological doctrine of religion after having restricted it in the two
indicated ways? What are the critical conditions under which the religiously
formatted representation of “the personified Ideal of the Good Principle”
“seen as” the Son of God, can be affirmed to be true in the real sense? It turns
out that the two conditions of faith – subjective and objective – which I
presented in the previous section, count in the Christological section as well.
In this regard, the subjective condition expressed as “finite human reason
must feel its need” is of a quite particular nature (CPrR 5:145). In view of
representing the moral Ideal of perfection, practical reason has “no
need . . . of any example from experience to make the idea of a human
being morally pleasing to God a model to us” (Rel 6:62). Such a need,
which would orient reason toward experience, contradicts the universality
and practical necessity regarding the law that would command obedience to
that example.
The true need of human reason to acknowledge a religiously defined Ideal
of moral perfection and adhere to it by faith is awakened by two momenta of
finitude. The first refers to reason’s need for schematization: “It is plainly a
limitation of human reason, one which is ever inseparable from it, that . . . we
always need a certain analogy with natural being in order to make super-
sensible characteristics comprehensible to us” (Rel 6:64–65n). Once “this
person or his expression [is portrayed] in human guise” (Rel 6:64n), human
reason cannot design it according to its truly divine disposition of proto-
typical goodness unless “seeing it as” “a supernaturally begotten human
being” (Rel 6:63). The first – subjective – condition of truth regarding the
Christological representation of the good principle is hereby fulfilled: the
representation originates from reason’s own need to put faith in such a self-
made representation.
On the other hand, the objective condition under which the truth of
religious faith in the Son of God is critically corroborated boils down to
the necessity of the role it plays in the moral improvement of our disposition
to the good. Within the act of conversion, imperatively forced by duty, the
religious representation of the Ideal practically strengthens in the human
being’s will this force and elevates it purposefully to a degree of absolute
purity. In this regard, Kant states: “it is our universal human duty to elevate
ourselves to this ideal of moral perfection, i.e. to the prototype of moral
disposition in its entire purity, and for this the very idea, which is presented
to us by reason for emulation, can give us force” (Rel 6:61). The objectivity
of this objective condition of true belief in the Christological ideal is
ultimately rooted in our own original predisposition to the good (Rel 6:83)
554 M. Moors
as belonging to human nature (Rel 6:28), which is called personality (Rel

6:26–28). Under these subjective and objective conditions, Kant’s critique of
religion that operates in the domain of Christology discards any “objective
realization” of this Ideal other than the moral one which is by nature in us.
He thus rejects “its hypostatization in a particular human being” (Rel 6:64).
For Kant, the object of Christological faith can never be a real existing
being who would “stand under” (hypo-stasis) reason’s pure ideal representa-
tion of it. But instead of an improper and unnecessary hypostatization, there
is nothing objectionable about a hypotyposis (Darstellung, subiectio sub adspec-
tum) that would make an idea of reason sensible (Versinnlichung), which in
this particular case occurs through symbolization (see CJ 5:351–52; RP
20:279–80). Whereas hypostatization is an ontological operation, hypoty-
posis is a logical procedure that aims at exhibiting an idea of reason under a
symbolic designation. The symbolization of an idea of reason (in this case,
the ideal of humanity pleasing to God) is based on an analogical exhibition of
relations that formally apply between, on the one hand, the properties of the
supersensible ideal and, on the other hand, properties pertaining to objects of
experience (in this case, examples of good conduct). What in the latter (for
instance, the account of the historical Jesus) would be displayed are not
examples to be emulated straightforwardly (directe). This would falsely turn
out to become idolatry and enthusiasm (Schwärmerei). Rather, instantiations
of good conduct displayed in historical experience have the status of exhibi-
tions of precepts to be followed. If now there does exist an analogical relation
between, on the one hand, these precepts (followed by historical instantia-
tions) and their effects (of good conduct) and, on the other hand, the moral
law (which identifies ideally the divine prototype) and its effects (of good
conduct), the analogy then represents (Darstellung) a religious real truth that
can be acknowledged by critical reason. Through the analogical relation, “we
begin to see in [objects of experience] the symbolic representation and the
means of furtherance of a pure faith of religion” (Rel 6:176).
What is at work then is “a schematism of analogy, with which we cannot
dispense” (Rel 6:65n). But if this procedure of symbolic hypotyposis is wrongly
used as a schematism of object-determination (i.e., hypostatization), we fall into the
trap of anthropomorphism which, for Kant, “from the moral point of view (in
religion) . . . has most injurious consequences” (Rel 6:65n). Kant says that
“anthropomorphism . . . is highly dangerous with respect to our practical relation
to [God’s] will and to our very morality” (Rel 6:168; see also 182–83). In
Christological matters, Kant warns that “we only need to be assiduous in keeping
[the Ideal] free of any impure mixture” (Rel 6:83). In this critical warning, we
locate the second mode of Kant’s critique regarding Christological faith and
hope, namely the interim reservation with respect to a possible anthropomorph-

ism by which – in concreto – this faith would be infected and ultimately falsified.
If the critique firmly states, as its principle, that any religious representa-
tion must, as its criterion of real truth, be of some benefit to our moral
improvement, how then does Kant’s Christology solve the particularly press-
ing question of justification and satisfaction? How should we think that a
human being’s life is justified before divine justice (Gerechtigkeit), given the
fact that this life is, from its beginning, affected by evil and hence that there is
a debt which is impossible for humans to wipe out? This question – which is
perhaps, from a moral point of view, the most central in Christology – is
treated by Kant in a way that is most typical for his critique. Only in a
practical respect can every human being rely on the idea of God’s Son as a
vicarious substitute (Stellvertreter) bearing the debts of sin, and believe that
they are saved (Erlösung) and their cause pled (advocatus, Sachverwalter)
before divine justice (Rel 6:74). For Kant, the truth of religious faith in
justification and the “mystery” (Rel 6:143) of satisfaction (Rel 6:116) is
critically conditioned by one single requirement that must be fulfilled:
“only on the assumption of a total change of heart” (Rel 6:76). It will then
be by grace that “we are relieved of all responsibility for the sake of this good
in which we believe, though fully in accord with eternal justice (because
based on a satisfaction that for us consists only in the idea of an improved
disposition of which, however, God alone has cognition)” (Rel 6:76).
Therefore, we see in the heart of Kant’s Christology the issue of conversion
reappear as the only critical assignment of the truth of Christological faith
and hope: “In the practical [through conversion] faith in this Son of
God . . . the human being can thus hope to become pleasing to God (and
thereby blessed)” (Rel 6:62; see also 61).
Kant’s critique assessing the truth of ecclesiological

representations
As I outlined earlier, the ecclesiological part of Kant’s philosophical doctrine

of religion as to its content deals with the necessary communal conditions for
the victory of the good principle over the evil principle. These conditions bear
on the founding of an ethical community “seen as” the kingdom of God on
earth (also called the invisible church). For this founding, a first necessary
condition (as a priori condition of possibility) amounts to acknowledging a
divine lawgiver. A subsidiary condition (as historical condition of possibility)
expresses – due to an “unavoidable limitation of human reason” (Rel 6:115) –
556 M. Moors
the requirement of a vehicle that must mediate in the realization of the final
moral purpose. Both these representations (lawgiver, statutory organization)
are subjected to a critique in order to assess their grounds of truth, or to
unmask their delusional aspects.
Which real truth does Kant’s critique allow within the limits of mere
reason with respect to the representation of a divine lawgiver? Notes taken
from the conclusion19 of the Doctrine of Virtue in his Metaphysics of Morals
(MM 6:486–88) are illuminating in this regard. In fact, the subjective and
objective coordinates that Kant uses in this text to assess, from a veridical
viewpoint, the representation of a divine lawgiver are exactly the same as
those that he applies throughout the Religion. The subjective ordinate repre-
sents “the need of human reason” for establishing such a religious representa-
tion. Kant expresses this subjective condition as follows: “we cannot very well
make obligation (moral constraint) intuitive for ourselves without thereby
thinking of another’s will, namely God’s (of which reason in giving universal
laws is only the spokesman)” (MM 6:487). The motive for this overstep from
the moral domain of duties to the religious representation of a lawgiver, as a
divine obligator (LE 27:277–78, 282–83), is characterized as “only subjec-
tively logical” (MM 6:487). This means that the truth resides, subjectively, in
reason’s need to perform this overstep, not in any objectively grounded
practical necessity. In a communal or ecclesiological respect, the religious
truth of the recognition of a divine lawgiver (being a synthetic principle of
the universal collective unity of an ethical community or invisible church) is
thus restricted to this merely subjective exigency of finite human reason.
Dogmatically inferring on this basis the objective reality of the object of this
subjective representation would have two ignominious effects. First, such an
inference would directly contradict the autonomous (free) character of moral
lawgiving by pure will. Second, moral duties would be transformed into
statutory “obligation[s] to perform certain services for [erga] another [in this
case, God]” (MM 6:487; see also Rel 6:99). In his ecclesiology, Kant’s
critique discards absolutely these two capital blunders.
The objective ordinate according to which the “subjectively logical” reli-
gious representation of a divine lawgiver can host some real truth indicates
that this representation is relevant in the pursuit of the moral common good
(MM 6:487). In his optimism, Kant is of the opinion that the mere
subjective representation of a moral lawgiver can positively yield an increase
19
The conclusion is titled: “Religion as the doctrine of duties to God lies beyond the bounds of pure
moral philosophy” (MM 6:486).
of moral force: “for the sake of strengthening the moral incentive in our own
lawgiving reason” (MM 6:487). The genuine moral incentive, namely the
feeling of respect for the moral law (G 4:400; CPrR 5:71–89), increases its
force against sensible feelings of gratification or pain (CPrR 5:92) by virtue of
the religious representation of the divinity of the giver of this law. In this
sense, Kant’s critique subscribes to the objective moral truth of this religious
representation only in assuming some moral efficacy on behalf of this
supplementary force. Moreover, this assumption becomes more veracious
in an ecclesiastical setting by taking into account the fact that the divine
legislator is a public (öffentlich) lawgiver of an ethical community. Moral
efficacy in the pursuit of the common good may be hoped for in a communal
setting by entrusting “to a higher wisdom the whole concern of the human
race (as regards its moral destiny)” (Rel 6:100–101). The objective condition
of holding the religious representation of a divine legislator to be really true
is, hence, in both cases (private and public) focused on the purely moral
parameter of “the most intimate parts of the dispositions of each and every-
one” (Rel 6:99). Regarding the founding of an ethical state “seen as” an
invisible church, the single a priori condition of possibility (divine legislator)
is necessarily one that is determined by the principle of morality in its purity.
At the heart of his ecclesiology, the moral law figures as the standard of
Kant’s critique concerning any divine principle that would ground the
ecclesiastic community.
In ecclesiastical matters, Kant’s critique also grasps rigorously the historical
conditions of possibility regarding the founding of an invisible church.
Among these historical conditions, special attention is paid to historical
(revealed) faith (also called ecclesiastical faith) (Rel 6:102–14) and to all
features of the visible church (statutory legislation) that are proclaimed either
by revelation and holy Scripture or by ecclesiastical officials. In this ecclesio-
logical respect, Kant’s critique concentrates on the veridical conditions under
which the religious idea of people of God (a noumenal practical idea) can be
“realized” in the visible form of a church.
With respect to its form, such a realization can be determined in an a priori
manner. On the pure a priori basis of the moral law, namely, the universal
form of “a kingdom of ends” is thoroughly determined (G 4:433; see also
436–37). In the religious domain, there are no specific conditions that would
determine the form – that is, the constitution (Rel 6:152) – of a church other
20
The a priori condition that formally defines the communal end of the moral pursuit of a people is
expressed in Kant’s theory on “the type of the moral law” (CPrR 5:67–71).
558 M. Moors
than the pure moral conditions that formally determine the establishment of
a kingdom of ends.20 In the ecclesiological respect, Kant’s critique, rather,
concentrates on the means by which, under the conditions of sensuous
human nature, this end – “seen as” church – must be realized.
Regarding the issue of means, which are all of historical and contingent
nature, there is no specific condition that would discriminate materially
which of the historical religious arrangements would be apt or not for the
purpose of realizing the religious idea of a church. Considered materially,
ecclesiastical institutions, statutory legislations, and ecclesiastical beliefs do
not have any moral value in themselves (Rel 6:103, 106, 177). The visible
church’s sacred legacy of institutions and beliefs cannot as such (that is, in
itself) be judged as means to be well-pleasing to God. Regarding the eccle-
siastical means that mediate in a communal fashion (as organization and
administration [Rel 6:152]) in attaining this purpose of being well-pleasing
to God, Kant’s critique judges their worth according to the genuinely
objective standard of all moral evaluation in general. Hence, they are all
critically valuated according to the formal facets that determine how they
are internally predisposed.
In concreto, Kant’s critique puts the following two main ecclesiastical issues
to this moral test:
1) In order to be “considered as the mere representative [Repräsentantin] of a

state [ruled] by God” (Rel 6:102), the ecclesiastical organization of the
visible church will be critically tested on the following four facets: univers-
ality (according to quantity: its numerical unity), purity (according to
quality: strengthening moral incentives), the internal freedom of its mem-
bers among themselves (according to relation: organizational laws shaped
after laws of freedom), and the unchangeableness of its constitution (according
to modality: regulation must be internally proportionated to the necessity
that characterizes the constitutional laws of an ethical state) (Rel 6:101–2).
Under these four critical conditions of truth, Kant discards the following
distortions that are twisted out of the church’s original condition: sectarian
schisms (Sektenspaltung), superstition (Aberglaube) and enthusiasm
(Schwärmerei), hierarchical power and illuminatism (Illuminatismus) (a
kind of democracy through individual inspirations), and arbitrary creeds
(willkürliche Symbole) (Rel 6:101–2). On the historical territory of what
concretely shapes the visible church, these distortions formally contradict
the original idea of church, and Kant rejects them as having no intrinsic
value whatsoever as a means (vehicle) for the promotion of the founding of
God’s state on earth. However, under the given critical conditions, the
historical organization of a visible church can incorporate very well
momenta of truth for this promotion inasmuch as there is in its predisposi-

tion a conformity with the pure ideal moral end.
2) A second communal condition in the historical pursuit of establishing God’s
kingdom on earth concerns ecclesiastical faith. It contains faith in revelation
through holy Scripture and faith in statutory laws “seen as” divine. In Kant’s
ecclesiology, historical revealed truth enjoys a particular status. For two
reasons, Kant considers ecclesiastical faith to be necessary in view of “the
constitution of each and every church” (Rel 6:102).
a) First, there is a subjective necessity: “due to a peculiar weakness of

human nature, pure faith can never be relied on as much as it
deserves, that is, [enough] to found a Church on it alone” (Rel
6:103). Revealed faith and the ground on which it is based, revela-
tion – in sum, statutory religion – must come necessarily to the help
of human beings who, by nature, need a vehicle or a schema for the
realization of a supernatural practical end (Rel 6:96). The subjec-
tively necessary schematism is also required – and for the same reason
– with respect to humanly installed religious statutory laws which, by
ecclesiastical faith, are “seen as” instantiations of a divine legislative
will.
b) Second, in their needful adherence to an ecclesiastical faith, humans
also need for this faith to be enshrined in holy Scripture rather than
founding it in a capricious sense of tradition. Kant writes: “as a
revelation to present and future generations, [scripture] must be the
object of the highest respect, for this is what human need requires in
order to be certain of the duty to divine service” (Rel 6:107). The
fact that holy Scripture is considered to be the founding historical
authority in ecclesiastical faith entails that it has an objectively neces-
sary status. Without the strong hold of “Thus it is written,” eccle-
siastical faith would lack the necessary content on which its internal
communal cohesion depends (Rel 6:107).
Kant’s critique subjects ecclesiastical faith to conditions against which we

assess the degree of truthfulness present in it, versus the distortions into
untruth, worthlessness, and delusion. As already mentioned, these conditions
are of a pure formal nature and in their finality they envisage the internal
predisposition of ecclesiastical creeds. By these conditions, the latter are tested
on their potentiality to become a universal religious faith (Rel 6:115) based
on the universal moral duty of humankind (officium sui generis) to become
well-pleasing to God in a genuine moral manner (Rel 6:97, 151).
These same conditions also validate the interpretations and the public
learning of holy Scripture. Consequently, they determine the internal
560 M. Moors
predisposition of those who are publicly in charge of scriptural exegesis and

learning. Tested by the criterion of universality (see Rel 6:115), ecclesiastical
(professions of) faith as to its/their content and to its/their praxis in public
observances can very well present genuine religious truth if, in both respects,
the inner predisposition aims at being homologous to the pure moral law.
This kernel of truth may, in concreto, for the time being, be wrapped up in
historical and contingent, created expressions of creeds (in plural, see Rel
6:107–8). As is the case with respect to the organization of the visible church
in general, ecclesiastical creeds in particular can be validated to be “the
requisite preparatory advances of the public in religious concepts” (Rel
6:106). According to their status as vehicle or means pertaining to the visible
church, they are thoroughly shaped after historical conditions and culturally
variable circumstances. But, from the viewpoint of the duty-bound promo-
tion of the invisible church, this historicity must be interpreted according to a
moral teleology of purification. The latter transforms historicity into provi-
sionality. It is precisely with respect to this “transfiguration [Verwandlung]”
that Kant’s critique is making it most essential point in matters of ecclesias-
tical faith (Rel 6:114).
If the moral purification or transfiguration is not valued in the learning or
the praxis of ecclesiastical faith, three distortions will appear. First, the praxis of
ecclesiastical faith may be internally – as regards disposition – distorted into “a
slavish and mercenary faith [Fron- und Lohnglaube] (fides mercenaria, servilis),”
which absolutely contradicts the “saving faith [seligmachende Glaube] . . . (fides
ingenua)” of the pure religion of reason (Rel 6:115). Second, the correctness of
belief (orthodoxy) claimed by teachers or heads of a church may turn into
despotic brutal orthodoxy (Rel 6:130), which contradicts the liberal disposition
in publicly claiming this correctness (Rel 6:109; see also WE 8:38–39). Third,
scriptural exegesis (Schriftauslegung), assisted by scriptural scholarship
(Schriftgelehrsamkeit) (Rel 6:112–13), could fail to publicly instruct or moti-
vate the congregation with universal principles of pure moral faith, but instead
install “an unceasing guardianship” or “spiritual despotism,” which contra-
dicts – as “a crime against human nature” – “the sacred right of humanity”
to proceed on the way of enlightenment in matters of religious faith (WE
8:38–40). In spite of these extreme cases of ecclesiastical dogmatism, which
absolutely contradict moral religion, Kant’s critique opens a historical
sphere of freedom wherein the transfiguration – the historical “seen as”
the provisional – can take place.
Within this sphere, the second mode of Kant’s critique – the interim
reservation – will be applied by taking into account the rule of homology with
the pure moral law. Inasmuch, namely, as there is a germ of universal rational
faith (in singular) in learning, praxis, and scriptural interpretation proper to

ecclesiastical creeds (in plural), then a gradual reform is effectively taking
place (Rel 6:122). Whereas the transfiguration, noetically considered (that is,
in principle), equals a revolutionary turn (from the empirical to the rational,
or from the contingent to the necessary), in concreto one can without contra-
diction validate the historical particularities whenever one becomes aware of
the duty-bound germ of moral universality that inhabits the particular. To
put it in dynamical terms: the historical momenta in which a believer lives his
or her faith can/must be considered as momenta of a gradual transition by
which he or she moves little by little toward pure faith of the religion of
reason. In matters of ecclesiastical religion, making such a transition repre-
sents a sign and a realization as well of “the coming of the kingdom of God”
(Rel 6:115) and of approaching the main purpose of Enlightenment
(Aufklärung) (WE 8:41).
In sum, if internally led by a pure moral teleology, ecclesiastical creeds
elevate themselves over and against their contingent earthly conditions, and
will – as vehicles and means – be clothed with divine dignity. As Kant puts it,
“the way a church is organized may perhaps also be a special divine dispensa-
tion” (Rel 6:105).
Kant’s critique assessing the truth of ecclesiastical services,

rituals, and practices
Intrinsically connected with his Christology and ecclesiology, Kant focuses in

Part Four on “service and counterfeit service under the dominion of the good
principle” (Rel 6:151–202). From within this double viewpoint, the idea of
service is pointing to two accidentally different commitments (Rel 6:151), the
moral and the ecclesiastical, by which he means service on the basis of pure
moral faith (i.e., service of the pure religion of reason) and service on the
basis of ecclesiastical statutory faith (i.e., service of statutory religion in a
visible church). Kant’s critique focuses on this last mode of religious service
(cultus) in two respects. First, as based on statutes and rubrics, this religious
cult will be practiced by ecclesiastical believers (the congregation [Rel 6:152])
who are guarded by ministers and officials. The last are in charge of organi-
zation and administration of the visible church (Rel 6:152). Kant’s critique
judges under which conditions this practice and guardianship can host some
real religious truth. Second, if religious faith is based on revelation and,
consequently, must be taught in order to be communicated, and in this
respect learned (learned religion [Rel 6:155]), any religious service to
562 M. Moors
interpret the content of revelation is consigned to an ecclesiastical minister-

ium of scholars. Regarding this particular ecclesiastical service, Kant’s critique
marks the conditions under which scriptural exegesis is authentic and valid
for the whole world (Rel 6:114) and how the learning or the teaching of a
revealed doctrine of faith must internally be disciplined in order to be
considered as true service in the promotion of the common moral good.
Regarding both aspects of ecclesiastical service (the ritual and the doctrinal),
Kant’s critique assesses their truthfulness by the one single veridical criterion
of moral teleology. This is the standard inner requirement that ought to
mark the inner predisposition of those who are committed to these services.
As regards the cult aspect, what is opposed to any moral worth “seen as”
pleasing to God are those rites and services that are performed in a faith that
would take its statutory regulations as “the supreme condition of divine good
pleasure toward human beings” (Rel 6:168). In other words, ecclesiastical
service becomes a means to curry favor with God and to be rewarded. Such
a “religion of rogation (of mere cult)” (Rel 6:51) fosters “a delusion of religion
[Religionswahn]” (Rel 6:168). Based on these delusions, service of God is
twisted into “counterfeit service [Afterdienst]” (Rel 6:167). What in fact dis-
torts service of God into counterfeit service is the “highly dangerous” anthro-
pomorphism (Rel 6:168; see also 141–42). If, namely, our relation to God’s
will is not determined by our internal representation of the moral law (“seen
as” divine command) and our obedience to only this law does not entail that
we see ourselves worthy of divine pleasure to us, and if, instead, our representa-
tion of God’s will and his grace is interpreted anthropomorphically as his being
pleased by our efforts, then the distorting effect of anthropomorphism will be
that human beings exhaust themselves in sacrifices, rituals, and festivals in a
vain effort to earn God’s unmediated favor. This inner distortion gives rise to
“priestcraft [Pfaffentum] as a regime in the counterfeit service of the good
principle” (Rel 6:175–80). The “spiritual despotism” (Rel 6:175n) that reigns
in this regime (Regiment) and favors “religious enthusiasm” (Rel 6:174) and
“mere fetishism” (Rel 6:179) is radically opposed to the genuine moral principle
of religion. If ritual performances are accompanied by “the delusion that
through religious acts of cult we can achieve anything in the way of justifica-
tion before God” (Rel 6:174), we are lost in “religious superstition
[Aberglaube]” (Rel 6:174–75). And, Kant writes, “enthusiastic religious delu-
sion is . . . the moral death of the reason” (Rel 6:175).
However, focusing on cult matters, Kant’s second mode of critique –
interim reservation – leaves an open space for truth in his evaluation of religious
cult. Indeed, the acts themselves, qualified as means and not as ends (which
create enthusiasm), can contain at least negative moral worth “to counteract
the obstacles that stand in the way of a disposition well-pleasing to God.” For
Kant, superstitious religion is “only accidentally reprehensible, i.e. only inas-
much as it transforms what can only be a means into an object immediately
well-pleasing to God” (Rel 6:175). In addition to this, a positive moral worth
can be present in cult acts in an indirect way. By the performance of such acts,
namely, the human being can “make himself worthy of a supplement to his
impotence through supernatural assistance” (Rel 6:178). Kant positively
assesses cult acts only as means of making oneself receptive for grace.
With regard to the second facet of service – the doctrinal – the critical
question points to how an ecclesiastical institution (clergy, laity) should be
predisposed in its service of scriptural exegesis and scholarship. In this regard,
Kant’s critique identifies counterfeit service on behalf of “commanding high
officials [Beamte] (officiales)” who “in fact wish to be regarded as the exclusive
chosen interpreters of a holy Scripture . . . and having commanded scriptural
scholarship for use solely in the interests of ecclesiastical faith” (Rel 6:165).
What should be ministerium is thus distorted into domination. Instead of
considering themselves as “members of a[n] (invisible) church which encom-
pass all right-thinking people” (Rel 6:176), ecclesiastical authorities exalt
themselves as “the single authoritative guardian and interpreter of the will
of the invisible lawgiver” (Rel 6:180). Kant’s critique of ecclesiastical doc-
trinal service rejects all modes of despotism (constitutional [Rel 6:180] and
spiritual [Rel 6:109, 131, 175n]) originating from the false predisposition to
“only give orders” (Rel 6:180). Doctrinal service, however, can be authentic
and host truthfulness only if a moral teleology as the highest authority
determines the predisposition in rendering the service of public doctrinal
instruction. For its doctrinal service, the ecclesiastical ministerium in “the
Christian religion as a learned religion” (Rel 6:163–67) is in a disciplinary
way ultimately bound to “the teacher of the Gospel” (Rel 6:162). Kant’s
appraisal of Christianity is fundamentally based on the fact that he recognizes
that the Christian doctrine of morals “alone satisfies the strictest demand of
practical reason” (CPrR 5:128). Christian historical faith did come “from the
mouth of the first teacher not as a statutory but as a moral religion” (Rel
6:167). With regard to doctrinal service of any institutionalized church, what
principally counts is its “closest proximity to reason” (Rel 6:167).
564 M. Moors
Conclusion
On our journey through Kant’s philosophical doctrine of religion, we had to
pass the following critical moments which were – according to him – of
determinative significance for living a religious life “‘in the Spirit of God,
who guides us into all Truth’” (Rel 6:112). As far as critical normativity is
concerned, in religious matters, the following consecutive determinants prevail.
The norm of the truth is the good, interiorized as the good will. The norm
of the good is the law, interiorized as duty. The norm of a duty-full maxim of
the will is the intention, interiorized as the feeling of respect. The norm of
respect is holiness, interiorized as the love of God (Rel 6:182). Religious
representations and teachings on grace, Christ, church, and religious cult
actions (as prayer and liturgy) are, for their truthfulness, tested by this stratified
complex of norms which are all derived from a highest principal meta-norm of
freedom and autonomy. A self-establishing universe of moral concepts and
principles – ideal expressions and manifestations of this autonomy – for Kant
normatively superimposes itself upon all historical, contingent, and objectively
given manifestations of religious faith and ecclesiastical practice. One could
consequently ask: which is, after all, for Kant the specifically religious perfor-
mance? If, for its truthfulness, we should remain “within” the domain of mere
practical reason’s principles which we ought to obey, the only genuine perfor-
mance that is required amounts to “moral cult” on behalf of the servants of an
ethical community (Rel 6:179). Inasmuch as we are determined to remain
within that domain, the act that institutes religion is the act of a semantic
tropism: the “seeing as,” which, in its material intention, is altogether empty.21
If we diacritically keep searching for the content of the specifically religious,
Kant’s Religion brings to light a unique specific performance “over and above
morality” (Rel 6:182): the attitudinal action of hope. This religious attitude of
hope is made objectively real on the various moments (of conversion) when the
moral person does “accept [the] help” of “supernatural cooperation” (Rel 6:44).
By the very act of accepting God’s grace, the moral person assumes a specific
position that overarches all constraining calls of duty. It is indeed this very
assumption that essentially inaugurates a religious life. And – as Kant rightly
remarks – this is “no small matter” (Rel 6:44).
21
For an elaborated argument, see my “Religious Fictionalism in Kant’s Ethics of Autonomy,” in
Cultivating Personhood: Kant and Asian Philosophy, ed. Stephen R. Palmquist (Berlin: de Gruyter, 2010),
475–84.
Part IX
Political Philosophy
25
The Critical Legal and Political Philosophy
of Immanuel Kant
Howard Williams
In the shadow of the Critique of Pure Reason

Kant’s legal and political philosophy is very much like an iceberg: more than
two-thirds of it is hidden from view. What is often invisible when we focus
on its detail, such as its account of property, the theory of punishment, the
definition of citizenship, and the theory of the state, is the underlying
framework provided by Kant’s novel undertaking in the critical philosophy.
My first objective here is to bring that underlying framework into view by
looking at the Critique of Pure Reason from the standpoint of its prodigious
influence on the presentation of Kant’s practical philosophy and so his legal
and political philosophy.
Of course the picture that Kant provides of the condition of metaphysics
in his time at the beginning of the Critique cannot inspire much enthusiasm
for the role that philosophy might play in our social and political lives. He
finds it in a pitiful condition of disarray, rightly disowned by the ordinary
person as a source of any kind of inspiration. Historically for him metaphy-
sics is at an all-time low; it can be justly seen as far too ambitious and leading
to contradictory results. Political reformers might seemingly be better off
looking outside philosophy for new insights as to how the human condition
might be improved. However disappointing this condition, Kant believes he
H. Williams (*)
Department of Law and Politics, Cardiff University, Cardiff, UK
e-mail: williamsh58@cardiff.ac.uk

568 H. Williams
can explain and, ultimately, remedy it. Although “human reason has the
peculiar fate in one species of its cognitions that it is burdened with questions
which it cannot dismiss” and is not able properly to answer (Avii), we can
nonetheless hope that a thoroughly cleansed and suitably limited metaphysics
may again seek to play a key part in realizing the “necessary and essential ends
of humanity” (A850/B878).
From the perspective of the political philosophy, presented in the
first part of the Metaphysics of Morals, Toward Perpetual Peace, and
various other writings, what the Critique of Pure Reason provides is an
extraordinarily original background to a wholly novel enterprise. The
principal focus of the first part of the critical philosophy is the question
“what can we know?” The answer that Kant provided to this question
was to change the course of philosophy entirely. The marked division
that Kant was to make between theoretical and practical philosophy was
not of course new, but the manner in which he separated the two,
giving ultimate priority to the practical over the theoretical, represented
a revolutionary step. Kant proposed very severe limits to the traditional
philosophical undertaking of metaphysics on the basis of his answer to
the epistemological question of what can we know, since our scientific
knowledge has to be limited to what we can take in from experience
through our sense awareness and understanding. Although political
philosophy for Kant must pay some regard to what we can attest
from experience, it primarily falls within the realm of practical
philosophy.
Transcendental thinking in the theoretical realm shapes the material
that is furnished by our senses and our understanding to form a unity
that is constitutive of the objective world we experience. We are able to
claim knowledge of this objective world because our thinking plays a key
part in shaping it, but we cannot claim that this is a knowledge of things
in themselves. We are only able to know through our intuition. We
shape what comes into our consciousness. With the practical realm
things are different. Our practical reasoning deals with our awareness
of ourselves as active beings who can frame objectives for themselves.
Practical reasoning is about our own process of thinking as we are
engaged in action.
The critical philosophy as outlined in the Critique of Pure Reason gives us
Kant’s basic understanding of the human being and the human being’s
condition. We learn from it what finite rational beings can know and how
they might conceive of their acting. The first Critique presents a view of the
25 Critical Legal and Political Philosophy of Immanuel Kant 569
human individual as a being who is to be grasped from two distinct

standpoints:
1) The standpoint of experience of the understanding. This is the standpoint

of scientific observation where the human individual is viewed as just one
other appearance in the phenomenal world. This first, passive standpoint
leads unavoidably to a causal analysis of the human individual. Our
actions here can be understood as subject to natural laws.
2) The standpoint of action or freedom. Here we can also see our actions
from an alternative standpoint, as brought about through our own voli-
tion. From the theoretical perspective it is impossible to prove that we are
legitimately to be seen as uncaused causes that initiate events in the world;
however, from the outlook of practical reason we have to present ourselves
as such. Without such a possibility we should not be able to attribute our
actions to our own initiative. Neither praise nor blame would be possible
since the events we observe as associated with the lives of human indivi-
duals would have to be seen as fully determined.1
The distinction that Kant presents here between these two ways of con-
ceiving the human individual is admittedly very difficult to grasp and express.
Also the distinction at one level appears wholly contradictory. How can we
both regard human actions as free or independent but also legitimately seen
as caused? The one standpoint seems to negate the other. However, Kant is
trying to present to us very complex relations that lie at the heart of human
existence. Capturing these relations in words (even if this is in expertly
crafted philosophical concepts) is very difficult. It is not easy to avoid the
impression of complexity and occasional contradiction. However, my view is
that the relations – or very close to them – pertain.
Political philosophy is about the freely acting human individual, which is
the second of the standpoints highlighted by the critical philosophy and its
results. But as the human individual is both natural and rational (or, put
differently, capable of freedom), we have also to take into account both
human individuals observed empirically as a phenomenon and human
individuals seen from the perspective of action as the originators of their
actions. Political philosophy has then to consider both the considered
1
“A faculty of choice, that is, is merely animal (arbitrium brutum) which cannot be determined other
than through sensible impulses, i.e., pathologically. However, one which can be determined indepen-
dently of sensory impulses, thus through motives that can only be represented by reason, is called free
choice (arbitrium liberum)” (A802/B830).
570 H. Williams
rational free decisions of human individuals and their historically observed

accomplishments and failures. The potential for free, rational action has to be
the first point of departure, but the weighing up of the potentialities for
action has to be set against our historical context.
Despite the doubts that the Critique raises about the limits of human
knowledge and our need to rein in the speculative use of reason, there is
nonetheless a strong vindication of the deployment of reason it its
practical regard. The deployment of reason in our knowledge and actions
is not, from Kant’s perspective, a merely optional matter. Reason, not-
withstanding its overambition, nonetheless provides a coherent structure
to our awareness. The first Critique demonstrates how to a great extent
the objective world which arises in our everyday awareness is a product of
our thinking faculties. This is the Copernican revolution Kant recom-
mends in the comprehension of our knowledge. We are not to see the
problem of knowledge as how our thought can encompass the object, but
rather how the object is already itself a product of our forms of compre-
hension. As the difficulties posed by human understanding and reason
lead to contradictions (as we see in the antinomies), we have to restrict
their scope and ambitions to the knowledge of appearance. We are
required to do this since we are aware that our very own thinking
capacities lie behind what we experience and so know.
The marked differences in the capabilities of pure and practical reason
which the Critique of Pure Reason shows lead to a clear difference in the
spheres of efficacy of the two and in the tasks they can undertake. A critical
pure reason reins in our claims to knowledge, but a critical practical reason
can help legitimate our attempts to lead better and fuller lives. In a para-
doxical manner, Kant proposes a greater objectivity to practical reason than
pure reason may ever claim. The first Critique therefore provides a powerful
underpinning for practical reason and so a dimension within which reason
can come into play in our social and political lives. Reason is both chided for
its theoretical overambition but at the same time it is extolled for its practical
potential. Reason strives for totality. Reason seeks to provide an architectonic
structure both for our knowledge and our actions. Although we cannot
expect the thing in itself to correspond to the architectonic structure we
put forward, nonetheless we can seek to act in the rule-governed way that
reason demands. The architectonic structure which reason generates for
practical philosophy – centering on the moral law – sets the example for
the legal and political spheres. Just as metaphysics in general should not
present a rhapsody, that is, an elegant combination of insights for our
edification and delight alone, so in political philosophy we need a systematic
general structure that determines the whole. Just as there has to be a

metaphysics which precedes a successful natural science, there has also to
be a metaphysics of morals which sets the tone for legal and political
decisions and action.
The importance of open, critical enquiry

Reason for Kant is not solely a personal matter. We have indeed to rely
on our own reason in developing our capacity to think clearly; however,
what we infer from the use of our reason we cannot take to a conclusion
that we infer for ourselves alone. Reason involves an engagement with
others and, more explicitly, with a public sphere in which all thinking
human individuals may join. The interpersonal quality of reason, as Kant
comprehends it, is made clear when he speaks of the necessity for
rational argument to submit itself to the judgment of others: “Reason
must subject itself to critique in all its undertakings, and cannot restrict
the freedom of critique through any prohibition without damaging itself
and drawing upon itself a disadvantageous suspicion” (A738/B766).
Reason cannot be expressed in the form of a personal opinion which
can be held to be subjectively valid, and so not open to the scrutiny of
others. For “there is nothing so important because of its utility, nothing
so holy, that it may be exempted from this searching review and inspec-
tion, which knows no respect for persons” (A738/B766). The proper
deployment of reason presupposes an open debate in which each indivi-
dual is free to express their doubts or affirm their agreement. For without
that freedom of expression and respect for the views of others, how are
we ever to judge the validity of our claims? “The very existence of reason
depends upon this freedom, which has no dictatorial authority, but
whose claim is never anything more than the agreement of free citizens,
each of whom must be able to express his reservations, indeed even his
veto, without holding back” (A738–39/B766–67).
Kant seeks to sustain a tolerant attitude among philosophers and the
public at large. Falsity and error in thinking cannot emerge as a result
of decrees from external authorities, be they religious or political.
Indeed, all such attempts to control our beliefs are destined to fail
since no one can fully know our internal thought processes nor give
them a precise direction. Kant though is concerned not only that no
one should seek to control our beliefs from outside; he seeks more
572 H. Williams
positively to encourage our confident expression of our own views. For

“if you grasp at means other than uncoerced reason, if you cry high
treason, if you call together the public, which understands nothing of
such subtle refinements, as if they were to put out a fire, then you make
yourself ridiculous” (A746/B774). Political authority has to stand to
one side when it is a matter of the beliefs that individuals form and the
conclusions they draw from philosophical debates. Philosophy should
not simply be judged from the standpoint of public utility, since many
of its arguments may have no direct impact on the welfare of the
public, and where they do they can only contribute to informed debate.
Kant was aware of the temptation that might arise in members of the
public authority to put to rest controversy in matters of knowledge and
belief. But such interference represents a mistake. Indeed, Kant claims
that allowing open discussion and the freedom of thought in general
can work to the benefit of rulers: “Thus instead of charging in with a
sword, you should instead watch this conflict peaceably from the safe
seat of critique, a conflict which must be exhausting for the combatants
but entertaining for you, with an outcome that will certainly be blood-
less and advantageous for your insight.” For Kant “it is quite absurd to
expect enlightenment from reason and yet to prescribe to it in advance
on which side it must come out” (A747/B775).
There is a highly progressive and optimistic side to the principal doctrines
of the Critique of Pure Reason. The doctrines presuppose the emergence of a
tolerant and pluralistic society which permits the expression of a variety of
points of view. Kant is fully aware that this may lead to dissent and
differences. But “reason also very much needs such a conflict, and it is to
be wished that it had been undertaken earlier” than his own Critique and
“with unlimited public permission” (A747/B775). “Thus let your opponent
speak only reason, and fight him solely with weapons of reason” (A744/
B772). Kant provides the most eloquent defense of democratic discourse,
despite himself working in a still autocratic, monarchical society. As he puts
it, we enjoy a
freedom to exhibit the thoughts and doubts which one cannot resolve oneself
for public judgment without thereupon being decried as a malcontent and a
dangerous citizen. This lies already in the original right of human reason,
which recognizes no other judge than universal human reason itself, in which
everyone has a voice; and since all improvement of which our condition is
capable must come from this, such a right is holy, and must not be curtailed.
(A752/B780)
The limits of psychology

One of the key lessons of the Critique of Pure Reason is that psychology is an
unreliable discipline in seeking to discover systematic (coherent) principles
on which to ground knowledge and morality. Kant divides psychology into
rational psychology and empirical psychology. Rational psychology deals
with the doctrine of the soul and it is given the closest attention by Kant
in his discussion of the paralogisms (A341–42/B399–400). There are impor-
tant lessons to be learned from rational psychology, but they are not at all
what the previous proponents of the doctrine believed them to be. Indeed,
Kant believes that rational psychology unavoidably leads to a dialectical
illusion which
rests on the confusion of an idea of reason (of a pure intelligence) with the
concept, in every way indeterminate, of a thinking being in general. I think of
my self, in behalf of a possible experience, by abstracting from all actual
experience, and from this conclude that I could become conscious of my
existence even outside experience and of its empirical conditions.
Consequently I confuse the possible abstraction from my empirically deter-
mined existence with the supposed consciousness of a separate possible exis-
tence of my thinking Self, and believe that I cognize what is substantial in me as
a transcendental subject, since I have in thought merely the unity of conscious-
ness that grounds everything determinate as the mere form of cognition.
(B426–27)
Thus for Kant we can conclude from rational psychology neither that the
soul persists eternally nor that it subsists independently of the body. Indeed,
the rational psychology of previous metaphysicians relied upon importing
too readily into empirical psychology inferences that had significance only
from a reflective transcendental point of view. We need to presuppose from a
transcendental standpoint that our experience is drawn together into one
thinking subject. But for Kant this thinking self is not necessarily experien-
tially accessible to us. It is indeed a precondition for our experience but it is
an intellectual precondition and not an empirical one:
That the I of apperception, consequently in every thought, is a single thing

that cannot be resolved into a plurality of subjects, and hence a logically simple
subject, lies already in the concept of thinking, and is consequently an analytic
proposition; but that does not signify that the thinking I is a simple substance,
which would be a synthetic proposition. (B407–8)
574 H. Williams
Empirical psychology deals with our own immediate perception of ourselves

through our senses and our understanding. In Kant’s view, “empirical
psychology must . . . be entirely banned from metaphysics, and is already
excluded by the idea of it” (A848/B876). By this Kant means that empirical
psychology cannot be relied upon to supply any a priori principles of
cognition; rather it has always to be combined with such principles to
provide any knowledge whatsoever. For Kant empirical psychology can
play a part only “on the side of applied philosophy” (A848/B876). What
we observe about human beings and their thinking cannot then for Kant
form a basis for consistent reasoning about the kind of knowledge we may
gain, the kind of action we may be involved in, and the kind of expectations
we should form of ourselves and the human species as a whole.
This is a very powerful conclusion to draw and one which has many
important implications for how political philosophy might proceed. We
cannot, first, look to psychology to provide us with dependable general-
izations upon which to build a political theory. Quite clearly Kant’s critical
system rules out the approach adopted by those political philosophers such as
Thomas Hobbes and John Locke who seek to begin their enquiries by
determining precisely the nature of the human individual. Niccolò
Machiavelli argues in a similar vein from his direct experience of politics in
the Italian peninsula in the fifteenth century.2 From Kant’s perspective, no
such precise determination is possible simply by observing our own inner
experience and the experiences of other human individuals, even if this is the
experience of the human species as a whole. The nature of the human
individual cannot be prescribed from what we discern of our own inner
lives and the lives of others we observe. Second, if psychology is to be
admitted to political philosophy, it can only be so after the main principles
have been set out from the standpoint of human reason. The nature we
observe and so also the human society we apprehend are from a Kantian
perspective already the products of the structuring process of the mind or the
“I think” – the unity of apperception. To construct a political philosophy, we
must step out of the theoretical mode of philosophy into the sphere of
practice (moral philosophy). The a priori principles of political philosophy
must be drawn from pure moral theory. Now this moral theory of politics
cannot at all disregard the results of the enquiries of those who engage in
empirical psychology nor in the observation of the human species in general.
2
He famously concludes that one can make this “generalization about men: they are ungrateful, fickle,
liars and deceivers, they shun danger and are greedy for profit” (Niccolò Machiavelli, The Prince, trans.
George Bull [Harmondsworth: Penguin, 1968], 96).
In seeking to put into effect the requirements that practical reason places
upon us as legal and political beings, we have to take into account what is
already (albeit incompletely) known of human individuals and society.
The importance of the practical point of view

Kant thinks it a humiliating turn of events for pure reason to discover that in
its theoretical use it is led into contradictions. These contradictions (the
antinomies) may leave it with a sense of impotence. The outcome of the
Critique of Pure Reason might appear to be entirely negative. All that can be
drawn from it seemingly is a discipline which “serves for the determination of
boundaries, and instead of discovering truth it has only the silent merit of
guarding against errors” (A795/B823).
It would be a mistake though to give up at this point. For “nevertheless,
there must somewhere be a source of positive cognitions that belong in the
domain of pure reason, and that perhaps give occasion for errors only
through misunderstanding, but that in fact constitute the goal of the stren-
uous effort of reason” (A795–96/B823–24). How otherwise are we to explain
the ambitions of philosophers over the ages?
For to what cause should the unquenchable desire to find a firm footing
beyond all bounds of experience otherwise be ascribed? Pure reason has a
presentiment of objects of great interest to it. It takes the path of mere
speculation in order to come closer to these; but they flee before it. (A796/
B824)
The ambitions of philosophers are not, for Kant, mere inventions or the
product of overactive imaginations. On these grounds reason “may hope for
better luck on the only path that still remains to it, namely that of its
practical use” (A796/B824).
The tensions in Kant’s thought that lead him to emphasize the primacy of
the practical in the Critique of Pure Reason are brought out very well by Allen
Wood when he says that
no philosopher has laid more stress than Kant did on the importance for
human beings of keeping in mind the limited capacity of their reason in all
the affairs of life, especially in the conduct of enquiry and formation of beliefs.
Yet no philosopher ever asserted more ardently the absolute title of reason to
govern human thought and action, or gave us sterner warnings concerning the
576 H. Williams
inherent badness and the disastrous consequences of permitting human pas-

sions, enthusiasms, or inspirations, or the supernatural deliverances of author-
ity or tradition, to usurp the authority of reason.3
Kant advocates a moral theology which does not rest upon a proof of the
existence of God but rather on the necessity for the existence of such a
highest being if we are going to be true to morality:
This systematic unity of ends in this world of intelligences, which, though as

mere nature it can only be called the sensible world, as a system of freedom can
be called an intelligible, i.e., moral world (regnum gratiae), also leads inexorably
to the purposive unity of all things that constitute this great whole, in
accordance with universal laws of nature, just as the first does in accordance
with universal and necessary moral laws, and unifies practical with speculative
reason. The world must be represented as having arisen out of an idea if it is to
be in agreement with that use of reason without which we would hold ourselves
unworthy of reason, namely the moral use, which depends throughout on the
idea of the highest good. (A815–16/B843–44)
We cannot expect to find empirical proof of the existence of God, but we

have to suppose such an existence from a moral point of view in order to
underpin our attempts at moral action. For if we were to think from the
outset that all such attempts would fail, we would thoroughly undermine our
moral motivation. We have to believe that the outcomes of our attempts at
moral behavior contribute to the happiness of the species even though we
should not take that happiness (especially our own) as the aim of our moral
action.
“I call the idea of such an intelligence, in which the morally most perfect
will, combined with the highest blessedness, is the cause of all happiness in
the world, insofar as it stands in exact relation with morality (as the worthi-
ness to be happy), the ideal of the highest good” (A810/B838). In this way
the epistemological enquiries of the first Critique prepare the way for Kant’s
practical philosophy and so his doctrine of right in a manner that it is
intended to give us confidence in the attempt to bring rationality and so
morality to our existing social and political world.
3
Allen W. Wood, Kant (Oxford: Blackwell, 2005), 108–9.
The central precepts of Kant’s political philosophy

We have seen how Kant’s legal and political philosophy is powerfully
grounded in the critical system that he develops first in the Critique of Pure
Reason. Now we have to consider how that influence expresses itself in the
political philosophy which is presented in his occasional writings such as the
essay “What Is Enlightenment?” and the short book Toward Perpetual Peace,
and most comprehensively in his Doctrine of Right which appears as the first
part of the Metaphysics of Morals (1797).
A key question which confronts us first as we look into Kant’s legal and
political philosophy is the nature of the relationship between Kant’s pure
moral theory – outlined most strikingly in the Groundwork to the Metaphysics
of Morals (1785) – and the theory of right, on which his legal and political
philosophy is based. There is no doubting that this is a complex relationship
and it is not always clearly expressed by Kant himself. Indeed, it is possible to
argue that there are different accounts of the relationship that Kant gives
which do not wholly coincide with each other.4 What I shall argue for here is
a view that I take to predominate in Kant’s writings on politics and law and
the account I believe to be the most coherent.
We have always to begin with the pure moral theory. This is spelled out in
the Groundwork and the Critique of Practical Reason. For Kant systematicity
within philosophy has to be gained by adopting a metaphysical approach that
abstracts from all direct experience. We have seen already that Kant does not
put too much faith in psychology as a basis for generalizing about human
awareness and conduct. Legal and political philosophy has to depart from a
priori ideas. The pure moral philosophy offers a very clear starting point in
that it sets out the principles we should adopt if we are to conduct ourselves
decently in our relationships with others and in regard to ourselves. This
implies that the categorical imperative has to play a foundational role in
political philosophy. In politics and legal affairs, we cannot simply abandon
our moral selves. The requirements of the categorical imperative, particularly
as exemplified by the formulas of the universal law and humanity, should
provide a framework for our reasoning about law and politics. But how is this
delicate relationship between morality and law to be spelled out? It is not
possible that rules that are deemed appropriate for our individual conduct
(ethics) can be carried over wholly unaffected into rules that are necessary to
4
For a valuable discussion of the question, see Sorin Baiasu, “Right’s Complex Relation to Ethics in
Kant: The Limits of Independentism,” Kant-Studien 107, no. 1 (March 2016): 2–33.
578 H. Williams
our social and collective conduct (right). Indeed, the principles that govern
pure moral deliberation have to do with our motivation or disposition in
acting, and in the sphere of legal and political deliberation we have need of
rules that will regulate our actual actions and their effects. The rules of the
Groundwork are wholly metaphysical: they are concerned with our motiva-
tions as rational intellectual beings. The rules of the Doctrine of Right have to
do with the external, phenomenal world, even though they take their origins
in our intellectual existence. The rules of right have to be conditioned by the
a priori reflections of pure moral philosophy, but they have also to take into
account matters encountered in experience. There is an impure element to
moral theory as it is deployed in right and politics.5
One striking way of putting this is to say that the moral theory of the
Groundwork has to do with the conditions of human freedom in general,
whereas the Doctrine of Right has to do with the conditions our external
freedom, as beings of sense and intellect in a phenomenal world among other
similar beings. The Groundwork outlines the moral law which provides the
metaphysical basis for our freedom in general. The moral law assumes the
possibility that each rational human individual has the power to determine
its choices in acting. Moral laws are ones that are appropriate for beings who
have this power of choice. But moral laws (which are the laws of freedom)
can be of two kinds:
As directed merely to external actions and their conformity to law they are
called juridical laws; but if they also require that they (the laws) themselves be
the determining grounds of actions, they are ethical laws, and then one says that
conformity with juridical laws is the legality of an action and conformity with
ethical laws is its morality. (MM 6:214)
The ethical standpoint which deals with the virtue of our actions looks
inward to the human individual’s motivation in acting. Such motivation
takes for granted the efficacy of the human will in putting in to effect its
aims. As Kant emphasizes, taken from a purely theoretical standpoint
there is no conclusive proof that such solely internal intellectual motiva-
tion, not determined by circumstances, is possible. Thus the ethical laws
which practical philosophy brings to light apply only to our inner
deliberative processes. For us to live with others in a civil society requires
laws that govern our actions regardless of the reflections of the individual
5
Robert B. Louden, Kant’s Impure Ethics: From Rational Beings to Human Beings (New York: Oxford
leading up to them. For this we require laws seeking to regulate our

outward actions which are coercible. With ethical laws we are to enforce
them ourselves through our deliberations, with external laws they are
enforced by the irresistible power of the state.
Kant therefore distinguishes between the two kinds of moral laws in a
further and significant way. The two kinds of law differ in the incentives on
which they depend in being obeyed. With an ethical law the incentive to
obey is provided solely by the individual’s own sense of duty. With ethical
and external juridical laws coercion in obeying the law – in response to the
powerful pull of our inclinations and self-interest – is required. But coercion
takes on a different form with each. With external juridical laws this coercion
is provided through the threat of punishment, but with ethical laws the
coercion is provided by our sense of duty alone.
As I see it then, the relation between the moral and the legal in Kant is
this: within the framework of the categorical imperative, we see that the
human individual is the kind of being who belongs both to the sphere of
the intelligible and natural world. So the human being is capable of acting
according to the moral law. However, in belonging also to the natural world
it is drawn by its senses and inclinations to aims and actions that run contrary
to moral law. Thus when it comes to external actions, the incentive to
comply with law, which is provided by our freedom, has to be supplemented
by a natural (physical) incentive. Kant puts it in this way:
All lawgiving can therefore be distinguished with respect to the incentives. . . .

That lawgiving which makes an action a duty and also makes this duty the
incentive is ethical. But that lawgiving which does not include the incentive of
duty in the law and so admits an incentive other than the idea of duty itself is
juridical. (MM 6:218–19)
With juridical law the incentive that is not drawn from duty has to be “drawn
from pathological determining grounds of choice, inclinations and aversions,
and among these, from aversions; for it is a lawgiving, which constrains, not
an allurement, which invites” (MM 6:219).
Although external lawmaking introduces this pathological incentive to do
our duty, it is not as a result removed from the sphere of moral theory. For as
Kant makes evident, we retain the ethical duty to obey the law for its own
sake – even with coercive external law – but should we not find that
constraint sufficient, we assent to the additional incentive of the possibility
of pathological coercion. The possibility of the application of coercion to
ensure that we obey juridical laws has for Kant to be seen as an integral part
580 H. Williams
of a metaphysics of morals and not an additional condition that is derived

from a standpoint wholly external to the teaching of the categorical impera-
tive. The ethical and the legal spheres are indeed distinct, but they are not
wholly separate from each other. The doctrine of the categorical imperative,
and the condition it implies, never to use other individuals solely as means
but always also as ends, applies also to external lawgiving or right.
Property and contract

Kant’s interpretation of property right is thoroughly imbued by the approach
that he takes in the foundational works of his critical philosophy. The right
to property is justified in a manner that is neither historical nor empirical.
Although the concepts of the state of nature and the social contract appear in
his political theory in helping derive our rights to things, they are not dealt
with in the same manner as his precursors in political theory, such as
Hobbes, Locke, and Rousseau. They are not conditions that we are to
imagine as a presupposition of a story that leads to the establishment of
property ownership; rather, they are concepts that are necessarily required if
we are to act as property owners. Just as the categories of the transcendental
logic are required so that we can constitute an objective world for our
understanding, so we have to presuppose the concept of property ownership
(of a “mine and thine”) if we are to inhabit a settled social world. We are able
to exchange property because there are fellow human beings who grasp the
idea of property ownership and are prepared to conduct themselves in
relation to others in a manner that realizes it.
To grasp property rights as Kant apprehends them, we have first to
understand his “universal principle of right,” or his depiction of our external
freedom. Social and political freedom are only possible through the accep-
tance of this general principle: “Any action is right if it can coexist with
everyone’s freedom in accordance with a universal law, or if on its maxim the
freedom of choice of each can coexist with everyone’s freedom in accordance
with a universal law” (MM 6:230). Such external freedom is thoroughly
reciprocal. We cannot grant to ourselves more freedom than we are prepared
to grant others. It presupposes our equality. We do not have the right to
subject others to any coercive restraints on their actions which we will not
accept ourselves. Such freedom presupposes the legislative standpoint in
which each has the right to participate in making the laws to which we are
all subject. In practice Kant accepts that it is impossible for each person to act
as a co-legislator, but those who legislate must be seen and see themselves as
representatives of each and every person. For Kant we have ideally to see
ourselves as co-participants in making the laws to which we are subject.
This is implied by our innate freedom, which we cognize as rational beings
who are also natural beings. We recognize that we have to subject ourselves to
the possibility of coercion if we are live in a society with other beings like
ourselves: “Freedom (independence from being constrained by another’s
choice), insofar as it can coexist with the freedom of every other in accor-
dance with a universal law, is the only original right belonging to every man
by virtue of his humanity” (MM 6:237).
Kant’s political philosophy draws upon a priori ideas in setting out its
foundational themes. I have always found this a very difficult approach to
express in words that the person starting in philosophy might correctly
follow. Any attempted explanation is not helped by the fact that Kant
deploys the idea in a variety of senses. Broadly, however, one might say
that there are two distinct senses which predominate.6 In the first place, we
can say he uses the term in an analytic sense to mean that an a priori idea
abstracts from all experience. In other words, it is an idea we use in discourse
which has to do with our thought and not at all with our senses and
observation. An analytic idea is for instance “A is A and cannot be not-A.”
In the second sense, the idea of the a priori is used to describe a generalization
which, though abstracting from any particular experience, may nonetheless
refer to experience in general. In this sense, a priori ideas can provide a
structure for our experience.
The principal concepts of political theory which Kant uses, such as the
state of nature, the original contract, and the original common ownership of
the earth’s surface, are all a priori concepts. He regards them as necessary to
allow us to conceive of a civil society. They are metaphysical propositions
that have to be generally accepted if we are to be part of a rightful society
with others. For Kant “all propositions about right are a priori propositions,
since they are laws of reason (dictamina rationis)” (MM 6:249). Kant thinks
that the need for us to use a priori teachings comes particularly to the fore in
the deduction of our right to own property. We cannot, he thinks, argue that
we have a right to a thing merely upon the basis of our happening to hold it
in our experience. Rather, we require that others recognize that entitlement
to ownership, and so we can maintain a thing is ours even if we are not at the
6
Here I follow the useful discussion by Reinhard Hiltscher in Kant-Lexikon, vol. 1, ed. Marcus
Willaschek et al. (Berlin: de Gruyter, 2015), 1–3.
582 H. Williams
time physically holding it. Kant calls possession which is merely the holding
of a thing “empirical possession.” And it is then “only possession in appear-
ance” (MM 6:249). What is needed to form an effective society with others is
rational possession. This is
a proposition about the possibility of possessing a thing external to myself,

which puts aside any conditions of empirical possession in space and time
(and hence presupposes the possibility of possessio noumenon), goes beyond
those limiting conditions; and since it affirms possession of something even
without holding it, as necessary for the concept of something external that is
mine or yours, it is synthetic. (MM 6:250)
There is an unavoidable metaphysical dimension to property ownership which

governs even the most basic economic transactions of a settled, civil society.
This metaphysical dimension is the rational presupposition that all those
involved in property ownership must hold, namely that it is possible to own
something in the absence of my physical presence (in other words, beyond the
limiting conditions of time and space). This is not an assertion about the world
independently of our thought and action. The assertion of possession is one of
our collective thought and action which affects external things. This is what
Kant means when he says that the proposition of ownership is synthetic.
Here for Kant a remarkable superiority for practical philosophy over
theoretical philosophy comes to light:
No one need be surprised that theoretical principles about external objects that are
mine or yours get lost in the intelligible and represent no extension of cognition,
since no theoretical deduction can be given for the possibility of the concept of
freedom on which they are based. It can only be inferred from the practical law of
reason (the categorical imperative), as a fact of reason. (MM 6:252)
In his view there is no possibility of deducing our right to own property

merely from observation and seeking to draw generalizations with it. Anyone
can testify that we cannot know from observation who owns what. Use does
not confer right. However, from the perspective of practical philosophy,
ownership can potentially be established through the deduction and recogni-
tion of the a priori possibility of an ownership which persists even in the
absence of our presence to hold it. As Kant puts it,
The concept of merely rightful possession is not an empirical concept (depen-

dent upon conditions of space and time) and yet it has practical reality, that is,
it must be applicable to objects of experience, cognition of which is dependent
upon those conditions. . . . Since the concept of a right is simply a rational

concept, it cannot be applied directly to objects of experience and to the
concept of empirical possession, but must first be applied to the understanding’s
pure concept of possession in general. So the concept to which the concept of
right is directly applied is not that of holding (detentio), which is an empirical
way of thinking of possession, but rather the concept of having, in which
abstraction is made from all spatial and temporal conditions and the object is
thought of only as under my control (in potestate mea positum esse). (MM 6:252–
53)
The general recognition of this aspect of the doctrine of right is a vital step in
making possible a settled human society.
Kant accepts that the presence in our minds of the idea of property: a
personal ownership that can be asserted in the absence of our being there, is
not enough to bring property into being in the world. Historically the first
property rights were established by force and had to be maintained primarily
by force.7 However, what makes property possible in a more secure and
persistent way is the existence of the kind of society where the mutual
recognition of property rights is an established rule in the minds and actions
of individuals. Given that we are both rational and natural beings, the actual
maintenance of property rights in a society requires the combination of the
rational acceptance of the normative standing of property rights and external
power to enforce those rights where they are (for whatever reason) violated.
Thus “something can be acquired conclusively only in a civil constitution; in
a state of nature it can also be acquired, but only provisionally” (MM 6:264).
A civil condition comes into existence with the coincidence of the existence
of a general coercive authority and the recognition by all of the a priori
conditions of right. The general coercive authority has to be bound by the
conditions of right if it is legitimately to enjoy full effectiveness. It may then
legitimately punish those who fail to abide by the conditions of right. All this
has to be present if we are to enjoy property rights: “when you cannot avoid
living side by side with all others, you ought to leave the state of nature and
proceed with them into a rightful condition, that is, a condition of distribu-
tive justice” (MM 6:307). In a condition of distributive justice, the law says
what objects “are capable of being covered externally by law, in terms of their
matter, that is, what way of being in possession is rightful” (MM 6:306).
7
See Allen Wood, “Kant’s Political Philosophy,” in The Palgrave Handbook of German Idealism, ed.
Matthew C. Altman (London: Palgrave Macmillan, 2014), 172–73.
584 H. Williams
But even then the property we own within a particular national civil
society is only provisionally so: “The indeterminacy, with respect to quantity
as well as quality, of the external object that can be acquired makes this
problem (of the sole, original external acquisition) the hardest of all to solve.”
No matter how we suppose it might be resolved for one people by referring
to a social contract they may presume to have agreed to, this still leaves the
matter unfinished. For “even if it is solved through the original contract, such
acquisition will always remain only provisional unless this contract extends to
the entire human race” (MM 6:266). A reciprocal recognition of property
rights which pertains to one state alone will not do to establish one’s property
right in a fully peremptory way. This argument is more or less explicit in the
case Kant puts for the support and establishment of the national state:
it is only a will putting everyone under obligation, hence only a collective

general (common) and powerful will, that can provide everyone this assurance.
– But the condition of being under a general external (i.e., public) lawgiving
accompanied with power is the civil condition. So only in a civil condition can
something external be mine or yours. (MM 6:256)
If the contract is viewed as being for the one people alone, it will neither give
rise to a “will putting everyone under obligation” nor to one “general
external . . . lawgiving accompanied with power.” Implicitly, therefore, the
establishing of a fully effective system of property ownership requires a
worldwide civil society. Kant’s idea of an original (problematic) worldwide
community of ownership – “possession of an external object can originally be
only possession in common” (MM 6:258) – would seem to go side by side
with the ultimate goal of a global civil society. And
this rational idea of a peaceful, even if not friendly, thoroughgoing community

of all nations on the earth that can come into relations affecting one another is
not a philanthropic (ethical) principle but a principle having to do with rights.
Nature has enclosed them all together within determinate limits (by the
spherical shape of the place they live in, a globus terraqueus). (MM 6:352)
At the end of the Doctrine of Right Kant allows himself to infer from the need
for agreement to ground the right of ownership in an original community,
that this grounding principle will eventually become acceptable throughout
the world, not as an ethical principle solely expressing good will but rather as
a principle of external, enforceable law. Our physical circumstances as the
inhabitants of a globe oblige us to interact with one another on the basis of
coherent principles with worldwide standing. For us to have exclusive use of

our own possessions we need ultimately the agreement of all members of the
human race. Claiming and maintaining something as our own implicitly
requires the compliance of all subjects and states with that claim.
The social or the original contract

Kant was a deep admirer of Jean-Jacques Rousseau’s philosophy, and it is
very likely that he was familiar with Rousseau’s The Social Contract (1762),
which provides a defense of popular sovereignty.8 Rousseau was deploying a
term that had wide currency in political philosophy. The notion of an
original contract figures prominently in the writings of Hobbes and Locke,
who were significant influences upon political thinking in the seventeenth
and eighteenth centuries. Just as these writers deploy the term in different
senses – in Locke and Rousseau’s case to argue for the possibility of resistance
and, if necessary, overthrow of unjust governments, and in Hobbes’s case for
the absolute authority of rulers – so Kant deploys the idea in his own distinct
sense to convey an understanding of the idea of an original contract which is
somewhat midway between the uses of Locke and Rousseau on the one hand
and Hobbes on the other. With Kant the concept of an original contract
provides the basis for the criticism of unjust governments (as it does with
Locke and Rousseau) but not for their overthrow. In the manner of Hobbes,
Kant argues that the power of the sovereign should not be resisted, and so
even when the sovereign acts badly it should be obeyed.
Kant sees the original contract as providing a popular foundation to
sovereignty, but since it is an idea which we derive from philosophical
thinking about right, it is not an idea that subjects can themselves apply
directly to politics. The reforms that might be required if a society is
measured by the ideal of a social contract have to be carried out by the
already existing sovereign. Reform has to be from the top down and not from
the bottom up. Another way of putting it is to say that the original contract is
a model that all genuine civil societies have to follow and one we have to
assume if we are to act lawfully. It is also a rule for those in government to
follow, but since their task is partly also to enforce (with coercion if
8
See Susan Meld Shell, The Embodiment of Reason: Kant on Spirit, Generation, and Community
(Chicago: University of Chicago Press, 1996), 4–5, 81–82.
586 H. Williams
necessary) the laws of the society, they cannot be coerced into acting
according to the ideal of the social contract.
The idea of the original contract falls in with the a priori standards of
reasoning that Kant has already applied to the emergence of property as an
institution. For property to exist, we have to suppose that there is a synthetic
a priori proposition which permits individuals to regard things as their own
even when they are not physically holding them. Similarly, for a rightful
society to come about, we have to assume that the terms of the original
contract have been accepted by all sides:
Now this is an original contract, on which alone a civil and hence thoroughly
rightful constitution among human beings can be based and a commonwealth
established. But it is by no means necessary that this contract (called contractus
originarius or pactum sociale), as a coalition of every particular and private will
within a people into a common and public will (for the sake of a merely
rightful legislation), be presupposed as a fact. . . . It is instead only an idea of
reason, which, however, has its undoubted practical reality, namely to bind
every legislator to give his laws in such a way that they could have arisen from
the united will of a whole people and to regard each subject, insofar as he wants
to be a citizen, as if he has joined in voting for such a will. For this is the
touchstone of any public law’s conformity with right. In other words, if a
public law is so constituted that a whole people could not possibly give its
consent to it (as, e.g., that a certain class of subjects should have the hereditary
privilege of ruling rank), it is unjust. (TP 8:297)
It is very clear from this that the original contract represents from a
Kantian perspective a normative standard by which to measure, accept,
or criticize the laws of a state and the policies of its rulers. That the
normative standard is not wholly met does not of itself serve to
undermine the authority of the laws or rulers. Rather, it can provide
an opportunity for laws to be improved and rulers to increase the
effectiveness of their administration. Governments and rulers have to
submit themselves to the authority of reason in public debate, but in
permitting and heeding that public debate, they reinforce rather than
undermine their authority. For Kant “the people too has its inalienable
rights against the head of state, although these cannot be coercive rights”
(TP 8:303). Subjects must be allowed to speak up in the hearing of their
ruler: “a citizen must have, with the approval of the ruler himself, the
authorization to make known publicly his opinions about what it is in
the ruler’s arrangements that seems to him to be a wrong against
the commonwealth” (TP 8:304). It appears that Kant is placing a very
severe restriction here upon the power of rulers (not as absolute of course
as Rousseau would wish by always leaving final authority in the power of
the people). They are presented at first blush as the servants of the
public.
However, Kant interprets the scope for the fallibility of rulers in a civil
society, in a very wide sense. Some might argue that he lets them off the hook
entirely when he remarks that, if it is at all possible that “a people could agree
to [a law],” then “it is a duty to consider the law just, even if the people is at
present in such a situation or frame of mind that, if consulted about it, it
would probably refuse its consent” (TP 8:297). Here Kant takes the non-
empirical nature of the idea of an original contract to very great lengths. The
interpretation of the idea of the possible consent of the people to a policy of
law is apparently left almost entirely to the discretion of rulers. Nonetheless,
he does think there are limits to the case that rulers might make for
advancing a new rule. For instance, Kant does not think a ruler can institute
a law that would permit the creation of a caste of people who would enjoy a
privileged inherited status (TP 8:297). Kant’s saving grace is that he does not
allow rulers wholly to regard their role as pragmatic enforcers or prudential
operators. They have always to pay heed to “the public well-being [Heil des
Staates]” and what “must first be taken into account is precisely that lawful
constitution which secures everyone his freedom by laws” (TP 8:298).
Thus Kant’s interpretation of the social contract does favor popular
sovereignty, but in a highly cautious way. The people should never interpret
the original contract in a manner such that it threatens the irresistible power
of the executive in carrying out the laws. Change should be effected by
alterations in the law which authorize the executive to act in a different
manner. And “if the existing constitution cannot well be reconciled with the
idea of the original contract,” the sovereign should be allowed to change it
“so as to allow to continue in existence that form which is essentially required
for a people to constitute a state” (MM 6:340). Effective lawmaking and the
proper execution of laws have to be permitted to continue uninterrupted
within a state, even as it may change its own constitution. In championing
change to bring an existing society more into line with the idea of the original
contract, Kant much prefers the model of metamorphosis to the model of
revolution. The latter implies the very dangerous possibility of the death of
the state at the same time as it is supposed to transform itself radically.9
9
Howard Williams, Kant’s Critique of Hobbes: Sovereignty and Cosmopolitanism (Cardiff: University of
Wales Press, 2003), 160–74.
588 H. Williams
Republicanism and cosmopolitanism

Kant has a very wide-ranging understanding of the role of the social contract
as a normative standard by which to measure the performance of a state.
Adhering to the original contract requires a good deal more than simply
adhering to the principle of popular sovereignty. Popular sovereignty has to
comply with the innate freedom that each individual should enjoy. Kant
defines our innate freedom as “independence from being constrained by
another” individual’s arbitrary will (MM 6:237). Kant puts this definition
in a more striking and radical form a few lines later when he says that this
independence also implies “a human being’s quality of being his own master
(sui iuris), as well as being a human being beyond reproach (iusti), since before
he performs any act affecting rights he has done no wrong to anyone” (MM
6:238). What is most remarkable in this statement of the implications of our
innate right is its universality. In principle Kant’s critical philosophy of right
makes no distinctions of race, gender, and nationality. Of course, Kant does
not always follow through on this universalist dimension in his own more
empirical discussions of European and world politics, but it is there always in
the background as a measure of legal and political advance.
It is implicit in the idea of right itself that we should obey only those laws
to which we could give our consent. Only this form of legislation conforms
to the idea of freedom which requires that we should be free from the
arbitrary constraining will of any other. Now, “the only constitution that
accords with right” is “a pure republic.” This is “the kind of government suited
to the idea of the original contract.” A republic is the only form of govern-
ment “which makes freedom the principle and indeed the condition for any
exercise of coercion, as is required by a rightful constitution of a state in the
strict sense of the word” (MM 6:340). The gradual metamorphosis that
brings this about is essential: “Only it will finally lead to what is literally a
state.” It is “the constitution in which law itself rules and depends on no
particular person” (MM 6:340–41).
Kant’s short book Toward Perpetual Peace (1795) sees him at the height of
his powers as a political philosopher. The book draws together the main
strands of his practical philosophy as it relates to politics and right. It is not a
utopian treatise but rather an attempt to draw out carefully the implications
of thinking in a manner inspired by his critical metaphysics to politics and
right. And despite the occasional area of discontinuity between the two
works, Perpetual Peace is remarkably at one in its approach with the
Doctrine of Right which appears two years later as the first part of the
Metaphysics of Morals. The idea of the state, as Kant presents it, plays a key
part in the argument of “Theory and Practice.” The republic as the embodi-
ment of the idea of the state is highlighted in the first definitive article of the
essay as one of the principal means of establishing world peace. Kant argues
that all constitutions should gradually transform into republican ones. In
such republican states the laws will be made by the people’s representatives,
carried out by their representatives and implemented by independent judges
in the courts. Neither of these three powers should interfere with the power
of their partners. Kant believes that states with republican constitutions will
be predisposed toward peace since the people who carry the main burden
(both financially and in combat) of prosecuting a war will themselves be
asked to give their verdict on the possible declaring of war.
Since these republics will be founded upon right, they will see the need to
join up with other republics to create a federation of peace. Thus the second
definitive article of Perpetual Peace requires all states to enter such a federa-
tion, even if they are not already republics. This alliance
does not look to acquiring any power of a state but only to preserving and
securing the freedom of a state itself and of other states in league with it, but
without there being any need for them to subject themselves to public laws and
coercion under them (as people in a state of nature must do). The practicability
(objective reality) of this idea of a federalism that should gradually extend over
all states and so lead to perpetual peace can be shown. (PP 8:356)
As themselves agents of law, states do not in principle require coercion.

Certainly sovereign states cannot be coerced to act as good international
citizens; they must learn to do so guided by the principles of morality and law
informed by reason.
What sets the seal however for the realization of the practical goals of
Kant’s critical system in politics and right is the emergence of a third
layer of law which Kant calls cosmopolitan right. Cosmopolitan right is
law for the world-citizens who must inevitably emerge through the
growth of republican rule and the dissemination of the rule of law.
Kant tries to define cosmopolitan right in a very simple way by saying
in the third definitive article of Perpetual Peace that such a right “shall be
limited to conditions of universal hospitality” (PP 8:357). By hospitality
Kant means that we should not necessarily always treat visitors to our
borders or shores with generosity but rather that we should accept every
person’s right, regardless of their origins, to present themselves as possi-
ble visitors and that we should not turn down their request of entry if it
590 H. Williams
would lead to their destruction. Kant anticipated the contemporary right

to refugee status that exists since the founding of the United Nations, its
Charter, and its various declarations, though his main purpose was more
likely to criticize the excessive force and violence through which the
European powers had seized and exploited other parts of the world.
Kant found more compatible with right (as dictated by reason) the
manner in which China and Japan had restricted the incursions of
European people, confining their entry and allowing them (in China’s
case) to settle only in enclaves on their coast (PP 8:359). In terms of the
breach of the third definitive article requiring hospitality, the “civilized,
especially commercial, states in our part of the world, the injustice they
show in visiting foreign lands . . . goes to horrifying lengths” (PP 8:358).
Kant demonstrates with his right of hospitality the thoroughgoing nature
of his grasp of law: as an interrelated, interdependent system from the
ground up. The innate right of freedom which all human individuals
enjoy under practical reason translates itself inexorably into a cosmopol-
itan right to be treated with the dignity befalling a human being on all
parts of the globe:
Since the (narrower or wider) community of the nations of the earth has now
gone so far that a violation of right on one place of the earth is felt in all, the
idea of a cosmopolitan right is no fantastic and exaggerated way of representing
right; it is, instead, a supplement to the unwritten code of the right of a state
and the right of nations necessary for the sake of any public rights of human
beings. (PP 8:360)
Here the radical dimension of Kant’s idea of our innate right comes to the
fore.
It is only to be expected that Kant should be drawn into contemporary
disputes about international relations. His comprehensive view of politics
and law makes his philosophy particularly applicable to today’s globalized
world. He sees law as in three interconnected, mutually dependent spheres:
domestic law (which includes both private and public right), international
law (both public and private), and cosmopolitan right. With his notion of
cosmopolitan right, Kant is able to provide a thorough philosophical foun-
dation for the presently burgeoning sphere of individual rights that go
beyond the borders of states.
Kant’s political philosophy has been widely invoked in contemporary
debates within the study of international relations, notably about just war.
Some writers advocate his views as a subtle and advanced account of just war
theory.10 Others – including the present author11 – believing themselves to

be closer to the original vision lying behind Kant’s political philosophy,
emphasize his systematic dislike for war. At best, they argue, war is for
Kant only a dire expedient when all else fails. War signifies a failure of justice
rather an area of its best application. There is value in both sides of the
debate. A war that is pursued with Kant’s injunctions in mind (both those of
right and morality) is likely to be more humanely fought. The thoroughly
anti-war position brings out the moral and legal dangers of resorting to war at
any time. Generally, Kant’s view is that war is more suited to animals as a
way of resolving disputes. Indeed, we reduce ourselves to a level beneath
animals by engaging in war in that we have the capacity to know and to act
better. The critical philosophy teaches epistemological humility and moral
courage in the face of the apparently never-ending complexities and mis-
fortunes of today’s world.
10
See Susan Meld Shell, “Kant on Just War and ‘Unjust Enemies’: Reflections on a ‘Pleonasm,’”
Kantian Review 10 (Jan. 2005): 82–111; and Brian Orend, War and International Justice: A Kantian
Perspective (Waterloo: Wilfrid Laurier University Press, 2000).
11
Howard Williams, Kant and the End of War: A Critique of Just War Theory (Basingstoke: Palgrave
Macmillan, 2012).
26
A Cosmopolitan Law Created
by Cosmopolitan Citizens:
The Kantian Project Today
Soraya Nour Sckell
Introduction
Cosmopolitan law (the rights and duties of individuals independently of their
state) is created by the exercise of cosmopolitan citizenship (the cross-border
association of individuals) and can only be legitimized by this exercise – this is
the central thesis of this chapter. This thesis can be supported by the legal
and political cosmopolitan reflections of Immanuel Kant, which have
become a benchmark in the contemporary debate on cosmopolitanism,
because of his theoretical concepts and his conjuncture analysis of political
issues of the time.1 Kant founded the rational concept of cosmopolitan law,
1
For deep, detailed analyses of the nuanced sense of Kant’s concept of cosmopolitanism, the historical
context of Kant as well as the contemporary debate, see especially the works of Kleingeld and Cavallar,
including: Pauline Kleingeld, Kant and Cosmopolitanism: The Philosophical Ideal of World Citizenship
(Cambridge: Cambridge University Press, 2012); and Georg Cavallar, Kant’s Embedded
Cosmopolitanism: History, Philosophy, and Education for World Citizens (Berlin: de Gruyter, 2015).
Among the most important works dedicated to Kant’s text Perpetual Peace are Georg Cavallar, Pax
Kantiana: Systematisch-historische Untersuchung des Entwurfs “Zum ewigen Frieden” (1795) von Immanuel
Kant (Wien: Böhlau, 1992); Otfried Höffe, ed., Immanuel Kant: Zum ewigen Frieden (Berlin: Akademie,
1995); Volker Gerhardt, Immanuel Kants Entwurf “Zum ewigen Frieden”: Eine Theorie der Politik
(Darmstadt: Wissenschaftliche Buchgesellschaft, 1995); Reinhard Merkel and Roland Wittmann,
eds., “Zum ewigen Frieden”: Grundlagen, Aktualität und Aussichten einer Idee von Immanuel Kant
(Berlin: Suhrkamp, 1996); James Bohman and Matthias Lutz-Bachmann, eds., Perpetual Peace: Essays
on Kant’s Cosmopolitan Ideal (Cambridge: MIT Press, 1997); Luigi Caranti, ed., Kant’s Perpetual Peace:
S. N. Sckell (*)
Centre of Philosophy, University of Lisbon, Lisbon, Portugal
e-mail: snsckell@campus.ul.pt

594 S. N. Sckell
and he saw that it could become real in a kind of link between French
Revolutionaries and their enthusiastic “spectators” abroad, which we could
see as the condition for the possibility of cross-border associations of indivi-
duals: cosmopolitan citizenship in exercise (even if it still does not exist
formally). This allows us to go with Kant, and beyond Kant, to think
about some of our contemporary problems.
Laws of the modern state address, above all, individuals, who are the most
important subjects of law; while modern international law addresses states,
not individuals, who then are not subjects of law. However, in some areas of
international law, which grew over the course of the twentieth century, the
individual is considered to be a subject of law, having international rights and
international duties independently of the state. The term international law is
no longer even appropriate in issues concerning the global scene that have the
individual, instead of the state, as the main subject of rights and duties. These
issues cannot be understood under the idea of an international law governing
relations between states, but under the Kantian idea of cosmopolitan law –
which means the law that considers individuals as citizens of the world
instead of a particular state, supporting individual power against a state or
supporting, in some juridical instances, power against individuals in spite of
their state. Human rights and international criminal law are the two main
areas of cosmopolitan law today, in which individuals became subjects of law.
But the expression subject of law (be it in state, international or cosmopolitan
law) as well as the word “citizen,” used here by Kant, means in fact only to be
subjected to law, the object of law, and to have rights and duties given by law.
The second sense of subject of law and of citizenship means, according to
Kant’s concept of autonomy, that individuals are the authors, creators, and
producers of the laws that they are supposed to follow. Citizenship, in this
sense, has been consecrated as the principle of the democratic state – but not
of the international or cosmopolitan law, supposed to be produced by states.
The difficulty is that if this cosmopolitan law is interpreted as being based
on its indisputable universal value, which occurs in most reconstructions of
the Kantian concept, then citizenship (as the political autonomy of the co-
legislating citizens) becomes something dispensable, irrelevant. However, this
New Interpretative Essays (Rome: Luiss University Press, 2006); Oliver Eberl, Demokratie und Frieden:
Kants Friedensschrift in den Kontroversen der Gegenwart (Baden-Baden: Nomos, 2008); and Oliver Eberl
and Peter Niesen, Immanuel Kant: “Zum ewigen Frieden” und Auszüge aus der Rechtslehre: Kommentar
(Berlin: Suhrkamp, 2011).
26 Cosmopolitan Law Created by Cosmopolitan Citizens 595
chapter aims to show with Kant that cosmopolitan law (Part I) is produced
and can be legitimized only by the political struggle of the world’s citizens,
cosmopolitan citizenship (Part II). Furthermore, it also show that cosmopoli-
tan law and cosmopolitan citizenship can really have a cosmopolitan dimen-
sion only if identity issues are considered (Part III): Kant’s concerns with the
rights of indigenous people, the conjuncture problem that he had in view
when developing his concept of cosmopolitan law, have been and should be
extended to any kind of discriminated group.
Cosmopolitan law: Human rights and

international criminal law
Cosmopolitan law by Kant
Legal doctrine before Kant2 considered two dimensions of public law: the
first level was constitutional law (Rechtsstaat), that is, the internal law of each
state, and the second was international law, that is, law in the relationships
between states as well as between individuals of different states. In Perpetual
Peace, Kant adds a third dimension of public law: the cosmopolitan law, that
is, the law of world citizens, considered not as members of their state, but as
members, beside states, of a universal state of humankind (PP 8:350). This
cosmopolitan law relates to the two previous laws, according to the table of
the categories in the Critique of Pure Reason (B93): the category of unity
corresponds to a single state (constitutional law); the category of plurality
corresponds to a number of states (international law); and the category of
totality corresponds to all human beings and states (cosmopolitan law).3 As
totality subsumes unity and plurality, cosmopolitan law subsumes constitu-
tional law and international law. All three levels of law are founded on the
same premise, namely, the reciprocal “physical influence” of people. The
round surface of the earth is not infinite but limited (MM 6:311). Thus, it is
not possible to prevent spatial proximity to other natural or rightful people.4
The inhabitants of the entire planet therefore constitute a system in which “a
violation of right on one place of the earth is felt in all” (PP 8:360).
2
An earlier version of this section appears in Soraya Nour, “The Cosmopolitans: Kant and Kantian
Themes in International Relations,” in Kant in Brazil, ed. Frederick Rauscher and Daniel Omar Perez
(Rochester: University of Rochester Press, 2012), 246–70.
3
Reinhard Brandt, “Vom Weltbürgerrecht,” in Immanuel Kant: Zum ewigen Frieden, ed. Höffe, 142.
4
Ibid., 143.
596 S. N. Sckell
The third definitive article of Perpetual Peace presents cosmopolitan law as

the third positive condition for peace with a rightful character: “as in preceding
articles, it is not a question of philanthropy but of right” (PP 8:357). Similarly,
in the Doctrine of Right, the third section of Public Law, on cosmopolitan law,
opens with the statement that “this rational idea of a peaceful, even if not
friendly, thoroughgoing community of all nations on the earth that can come
into relations affecting one another is not a philanthropic (ethical) principle
but a principle having to do with rights” (MM 6:352). This is the foundation
on which Kant formulates the third definitive article: “Cosmopolitan right
shall be limited to conditions of universal hospitality” (PP 8:357). According to
this principle, all people originally have the same right to land (MM 6:352),
and therefore “originally no one had more right than another to be on a place
on the earth” (PP 8:358). Kant notes that the right to land is not an “acquired
right” (MM 6:238), as is the case with the rights that we can have over things
(as in Kant’s “Private Right”); this right to land, on the contrary, comes from
the right to freedom and is therefore an “original right.” This right is the
foundation from which stems the right to one’s own body and, since a body
needs a place, the original “community of land” (MM 6:352).
This right is also the foundation of the right to visit (PP 8:358), that is, the
right of every citizen on earth to enter into community with everybody and, to
accomplish this end, to visit all regions of the earth (MM 6:353) without being
treated by foreigners as an enemy (MM 6:352). In this case, this right is injured
when one who arrives at a place is not accepted by those who were already there.
If the two previous articles have positive content, this third one, in contrast, is
formulated with a restrictive character: cosmopolitan right is limited to the right
of hospitality and cannot be more than this. The main problem that Kant has in
view with the concept of an unjust inhospitality is the case in which this right is
injured because foreigners who arrive in a territory extend their own empire
upon it. Cosmopolitan right should not be confused with the right to settle in
foreign territory (accolatus) (MM 6:353); it is its opposite. Having developed
these principles, Kant can now justify his severe criticism of colonialist attitudes
toward peoples on other continents by appealing to a rigorous theory of right,5
denouncing settling procedures that, while alleging to bring the benefit of
civilization to savages, in fact appropriate lands by force or fictitious purchase.
A people can install itself in recently discovered lands only while keeping its
distance from the residence of those who installed themselves there first and by
5
Ernest Hamburger, “Droits de l’homme et relations internationales,” Recueil des cours 97, no. 2 (1959):
316.
not causing them any harm. When this is a group of shepherds or hunters (like
most of the American nations), whose subsistence depends on great extensions
of unpopulated lands, this process of installation can take place only through
contract,6 which should not exploit the lack of knowledge of the inhabitants –
which has not historically been the case. Kant thus contests any assertion that
such violence leads to a better world, condemning the maxim according to
which the ends justify the means: “all these supposedly good intentions cannot
wash away the stain of injustice in the means used for them” (MM 6:353).
Kant sees the behavior of colonialists toward natives as a reduction of the
other to nullity – which makes relationships between people impossible:
“When America, the negro countries, the Spice Islands, the Cape, and so
forth were discovered, they were, to them, countries belonging to no one,
since they counted the inhabitants as nothing” (PP 8:358). This is the
depreciative vision of others that grounds every kind of discrimination.
Kant not only points to the illegitimacy of the conquest, but also to its
devastating effect in the complete de-structuring that directly or indirectly
stems from it:
In the East Indies (Hindustan), they brought in foreign soldiers under the
pretext of merely proposing to set up trading posts, but with them oppression
of the inhabitants, incitement of the various Indian states to widespread wars,
famine, rebellions, treachery, and the whole litany of troubles that oppress the
human race. China and Japan (Nipon), which had given such guests a try, have
therefore wisely [placed restrictions on them]. (PP 8:358–59)
Finally, Kant uncovers the bond between commercial expansion and wars
between European powers: “The worst of this (or, considered from the
standpoint of a moral judge, the best) is that the commercial states do
not even profit from this violence; that all these trading companies are
on the verge of collapse”; they serve only to conduct more wars in
Europe (PP 8:359).
Cosmopolitan rights at the local level
Kant’s cosmopolitanism has to deal with the following paradox of the

contemporary world: on the one hand, in the context of globalization, the
international scene cannot anymore be called a society of states. The logic of
6
Victor Delbos, La philosophie pratique de Kant, 3rd ed. (Paris: PUF, 1969), 564–65.
598 S. N. Sckell
the state is overlapped by non-state actors acting on the basis of economic,

cultural, and religious concerns. Some authors even prefer to talk about
global politics, rather than about international and transnational relations.
On the other hand, if the state was to be overtaken by individual actors in
various fields, some states, directly or through international or governance
bodies, would continue to impose strong changes in the global order. Still,
the exercise of civil, political, social, and cultural rights is ensured only in the
framework of the state.
The first problem, therefore, is: how can we give local democracy a cosmopol-
itan horizon? This is the first implication of Kant’s concept of cosmopolitan law:
a democratic state must not wait for the development of a cosmopolitan law
outside it, but rather a local democracy must make cosmopolitan law real first
within its own borders. In Kant’s sense, even a local, territorially limited
juridical-political order should consider its place in humanity as a whole, that
is, with regard to human beings (including future generations) considered as
such and not as members of a certain state. So the first goal of Kant’s
cosmopolitanism is the cosmopolitization of state democracy. The democratic
government of a group by itself, according to the principle of popular sover-
eignty, should include the whole of humanity in order to be truly democratic. A
cosmopolitan conception of human rights, so Christoph Menke and Arnd
Pollmann tell us,7 does not depend on a world democracy nor on the export
of democracy, but can be realized in a very limited territory. What is crucial is
that a democracy respects the civil rights of the members of a political com-
munity (such as the right to vote), that it respects the fundamental rights of all
those living in their territory and the human rights of all people in the world,
regardless of their citizenship. This is the deep sense in which the interrelation
between constitutional, international, and cosmopolitan law by Kant must be
understood. The three levels of public law by Kant are not independent,
separate instances: cosmopolitan subsumes constitutional and international
law. The democratic state must be cosmopolitan.
The individual as a subject of cosmopolitan law
A further issue of Kant’s concept of cosmopolitan law concerns the

consecration of the individual in international (or cosmopolitan) law –
as a subject of law, especially with regard to human rights and
7
Christoph Menke and Arnd Pollmann, Philosophie der Menschenrechte zur Einführung (Hamburg:
Junius, 2007).
international criminal law, but also in areas such environmental rights

and the common heritage of humanity.8 While Kant, at the end of the
eighteenth century, considered the individual to be the subject of cosmo-
politan law, independently of his or her state, international law doctrine
through the nineteenth and the beginning of the twentieth century did
not accept that the individual could be considered a subject of law: “It
may be readily admitted that according to a well-established principle of
international law, [a treaty], being an international agreement, cannot, as
such, create direct rights and obligations for private individual.”9
This has drastically changed since the second half of the twentieth
century. For many authors, individuals do not have the international
legal status of a subject of law. Treaties conferring rights and duties only
mean that states are required, vis-à-vis other states, to confer such rights
on individuals and to impose obligations on individuals in their home
system. The right to individual petition in human and labor rights
forums is seen as a procedural standing, which does not match a
substantive right or power to enforce compliance with any decision
favorable to the individual. According to other authors, however, given
the dramatic developments in the area in recent years, its complexity
allows for considering in international law the individual as a subject of
law in some situations.10
The notion of international law would even be inappropriate for governing
relations in which the individual becomes the main subject of law: this is not
only an international law governing relations between states, but a cosmopol-
itan law, which gives the individual power against the state, or which confers
power on international forums against individuals despite their states. In
some respects, in international law, the individual’s status as a subject of law
is compatible with state-centered theories of international law and the
transnationalization of rights; but in the case of cosmopolitan rights and duties,
in contrast, it appears necessary to carry out a thorough review of all legal
categories based on state sovereignty constructed in the last two hundred
years.
8
See Andrew Clapham, “The Role of the Individual in International Law,” European Journal of
International Law 21 no. 1 (2010): 25–30.
9
Permanent Court of International Justice (PCIJ), Jurisdiction of the Courts of Danzig, Advisory
Opinion, Reports Series B No. 15, March 3, 1928, [37], pp. 17–18.
10
For further discussion, see Pierre-Marie Dupuy, “Actualité du cosmopolitisme juridique: Revenir à
Kant pour mieux le dépasser?” Revue Quebecoise de Droit International (June 2015): 313–29; and
Angelika Emmerich-Fritsche, Vom Völkerrecht Zum Weltrecht (Berlin: Duncker und Humblot, 2007).
600 S. N. Sckell
Three aspects of cosmopolitan law (two substantive and one procedural)

should be taken into account11:
1) The individual in cosmopolitan law could be considered a subject of
law as bearer of the obligation not to commit international crimes, such as
genocide, crimes against humanity, war crimes, and terrorism. Individuals
worldwide are currently under such obligations irrespective of the adoption
of such obligations in national legislation, and they can be tried by the courts
of any country in the world or, when applicable, an international criminal
court. When Kant developed his concept of cosmopolitan law, the examples
which he gave are now present-day types of international crime, consecrated
in several documents, such as the Statute of Rome. These cosmopolitan
obligations impose individual responsibility, in contrast to the collective
responsibility that prevails in international law.
2) The individual in cosmopolitan law could be considered as a subject of
law as bearer of the rights corresponding to these cosmopolitan obligations
(not to be the victim of international crimes). The question is: who can
enforce such obligations? Some authors declare that only states have the
power to bring such crimes to justice. Others, however, declare that indivi-
duals have the right, which arises directly from international rules, not to be
the victim of international crimes, although individuals do not have the
means to enforce compliance with these obligations. Individuals may insti-
tute criminal proceedings against international crimes in national courts with
universal, territorial, or personal jurisdiction, they may petition international
human rights institutions, or they may forward their complaints to the
prosecutor of the International Criminal Court.
3) The individual in cosmopolitan law could be considered as a subject of
law because of the right to individual petition. The right to petition inter-
national bodies is given to individuals whether or not it is authorized by
national legislation that implements such treaties. This right is granted to
individuals directly by international rules and exists independently of
national legislation. It is a genuinely cosmopolitan law. Therefore, one can
say that individuals possess international status, to a certain extent, with
respect to individual petition.
It is important to note a fundamental difference between international
rules regarding international crimes and human rights treaties. Rules regard-
ing international crimes confer rights and obligations directly on individuals,
even if the national system does not implement or oppose such rights and
11
Antonio Cassese, International Law, 2nd ed. (Oxford: Oxford University Press, 2005).
obligations. Whoever violates such rules incurs individual responsibility.

Human rights treaties require states to confer such rights and enforce these
obligations in the national system. If the state does not confer such rights and
obligations, individuals have the right to individual petition, but those who
violate these norms do not bear individual responsibility on the international
level. There is only an international collective responsibility on the part of the
state to which such individuals belong.
Collective responsibility, a characteristic feature of international society, is a
typical archaism of primitive, rudimentary societies. As Hans Kelsen points out,
“collective responsibility exists in case of blood revenge that is directed not only
against the murderer but also against all members of his family.”12 In the sense of
Kant’s cosmopolitan law, which considers individuals as citizens of the world for
their rights and duties, just as there is the international individual responsibility of
those who commit international crimes, it should also be the international
individual (and not collective) responsibility of those who violate human rights.
The paradox of Kant’s cosmopolitan law
Kant’s conception of cosmopolitan law and its contemporary reconstructions

have to deal with a paradox in the concept of universality, which concerns any
policy based on the assertion, legitimization, and application of principles:
we argue with Kant’s principles of cosmopolitanism to criticize the ideolog-
ical use of the same principles. This paradox results from a deeper paradox
in the concept of the universal. As the sense of a universal statement depends
on who utters it, whom it addresses, and to do what (a universalistic
discourse can eventually legitimize exclusion and a particularistic discourse
can eventually amplify universalism), we must examine the conditions of
universalism, which exist only in historically specific forms, all of which
claim universality, but which are mutually exclusive. This paradox in the
universalism of Kant’s cosmopolitanism can be highlighted by Étienne
Balibar in his distinction between the universal as reality (real cosmopolitan-
ism as the standardization of the world) and the universal as ideality (Kant’s
cosmopolitanism as the universal values of equality and liberty intended to
denounce all forms of discrimination and deprivation of rights).13 The same
12
Hans Kelsen, Principles of International Law (Clark, N.J.: Lawbook Exchange, 2003), 10.
13
Étienne Balibar, “Cosmopolitisme, internationalisme, cosmopolitique,” in Vivre en Europe:
Philosophie, politique et science aujourd’hui, ed. Bertrand Ogilvie, Diogo Sardinha, and Frieder Otto
Wolf (Paris: L’Harmattan, 2010), 19–49.
602 S. N. Sckell
principle that legitimizes order turns out to be the absolute, exclusive

resource against the non-application of the principle, which is the same as
a consecration of the established order.
However, the ideological use of Kant’s cosmopolitanism rights, as with the
use of all legal and moral principles, does not mean that it expresses the ideas
of a dominant power. The universality of human rights and systems of local,
regional, or universal protection is not based on a substantial essence, but on
the contingency of the struggle of victims of inequality, who rebel in the
name of officially recognized principles that are denied in practice.
This presupposes a fusion of freedom and equality as well as human rights
and citizens’ rights. The tradition of natural law considers human rights to be
prior in the logical order. In the historical order, however, the reverse has
occurred. As Balibar argues, the “unconditional” (a “truth,” a “universal”)
has a “conditional” basis that is “given” and “conjectural.” Moreover, the
activity that establishes it is generally an act of civil disobedience or insurrec-
tion. Citizenship, before becoming a status (having rights), is primarily a
political activity that creates rights. The paradox of Kant’s cosmopolitan law
can be solved only through cosmopolitan citizenship. Cosmopolitan law can
become a reality only through the practice of cosmopolitan citizenship,
which Kant saw in the actors and spectators of the French Revolution – an
insurrection that created rights.
The cosmopolitics of cosmopolitan citizenship

The first section of this chapter dealt with the Kant’s concept of cosmopol-
itan law and the contemporary status of this law. This second section is
about how Kant thought that law could become a reality through political
practice: cosmopolitan law is upheld by cosmopolitics, cosmopolitan citizen-
ship, its legitimacy criterion. Cosmopolitics implies the democratization of
international, global, or cosmopolitan relations through the practice of cross-
border, extra-parliamentary associations of individuals. The underlying ques-
tion is how to democratize the global system, how to transpose principles and
practices that have been created within the framework of the nation-state
into this global system, how to develop new forms of democracy with
something other than a territorial foundation, and finally how to go beyond
national citizenship if there is no formal cosmopolitan citizenship. There is
no democratic representation in the most influential international organiza-
tions and global institutions of governance. On the other hand, there are a
number of forms of association in civil society that transcend borders,

creating new forms of citizenship. Kant, as it should now be seen, had
something like this in view.14
Cosmopolitics by Kant15
Kant’s concept of cosmopolitics is founded on the principle of publicity,

which presupposes the exercise of cosmopolitan citizenship. In Perpetual
Peace, Kant celebrates publicity as the supreme principle of justice. This
principle goes beyond the hegemonic arguments of his time regarding the
importance of political-economic liberalism and enlightened regents for the
promotion of peace. Kant nevertheless appropriated certain premises from
liberalism in his philosophy of history: individuals following their own
interests achieve juridical-rational ideas through the mechanism of “unsoci-
able sociability,” which can protect their lives and property, while making the
full development of human reason possible. He also reflected on the role of
regents in carrying through the principles of natural law, showing how a
moral politician acting under rational principles as well as prudence trans-
forms the theory of right into a political practice.
Even if Kant recognizes fundamental progressive elements in both cases,
however, they can only establish provisional peace. Perpetual peace must be
achieved differently. It must be the work of the public according to the principle
of publicity, not a mechanism of “unsociable sociability,” and even less the
result of personal qualities of regents, however moral these might be. Most
readings of Kant aim to demonstrate the accomplishment of natural law above
all through his philosophy of history or through the action of a moral politician,
lacking the democratic force of the principle of publicity. I intend to show how
only through this principle can the ideal of a cosmopolitan law be fully
achieved, and that this principle presupposes a cosmopolitan citizenship in
exercise. First, I will examine the principle of publicity. Second, I will analyze
the agent concerned with the achievement of cosmopolitan law: the public.
14
James Bohman and Pauline Kleingeld, for instance, have developed a Kantian concept of cosmopol-
itan citizenship that is used in global networks. See James Bohman, “The Public Spheres of the World
Citizen,” in Perpetual Peace: Essays on Kant’s Cosmopolitan Ideal, 179–200; and Kleingeld, Kant and
Cosmopolitanism, 90.
15
An earlier version of this section appears in Soraya Nour, “Kant’s Philosophy of Peace: The Principle
of Publicity,” in Recht und Frieden in der Philosophie Kants: Akten des X. Internationalen Kant-Kongresses,
ed. Valerio Rohden, Ricardo R. Terra, Guido A. de Almeida, and Margit Ruffing, 5 vols. (Berlin: de
Gruyter, 2005), 4:573–84.
604 S. N. Sckell
Third, I will show that this public not only has a literary practice, but a
revolutionary one too: only a revolutionary public acting in accordance with
the principle of publicity, as Kant saw in the French Revolutionaries, can fully
achieve “cosmopolitan law.” This public is not local, but cosmopolitan. Kant
finds the indications of a cosmopolitan public in the cross-border link between
the French Revolutionaries and their enthusiastic spectators outside France.
1) The first institutional condition for publicity is “freedom of writing.”
But this fulfills its critical function only after the right has already been
violated. It is only in Perpetual Peace that Kant presents the two “prophylac-
tic” forms of publicity that do not come after an injustice has been com-
mitted, but are immanent moments in all legal claims.16 The abstraction of
the material of the public law – the empirical relations between human
beings and the state – shows that its form is one of publicity (Publizität),
“which is involved in every claim to a right, since without it there would be
no justice (which can be thought only as publicly known) and so too no right,
which is conferred only by justice” (PP 8:381).
The transcendental formula of public right then obtained is: “All actions
relating to the rights of others are wrong if their maxim is incompatible with
publicity” (PP 8:381). A maxim that must be disguised in order to be
successful and provokes the resistance of all when publicly declared can
only be unjust. In the passage from moral to right, Kant moves from a
theoretical criterion – contradiction (Widerspruch) – to a practical criterion –
opposition (Widerstand) – which, although presupposing the logical contra-
diction of concepts, represents an activity.17 Opposition is a criterion for
recognizing what is not just, that is, the disagreement of politics with the
moral as the doctrine of right.
This confers a negative character on the first formula of the principle of
publicity. If maxims incompatible with publicity are unjust, this does not
mean that maxims compatible with publicity are therefore just: a despot, to
whose power there is no opposition, does not have to fear publishing
principles that are not in accordance with the rational principles of right.
Thus, Kant establishes a transcendental and affirmative principle of public
law: “‘All maxims which need publicity (in order not to fail in their end)
harmonize with right and politics combined’” (PP 8:386).
16
Klaus Blesenkemper, “Publice age” – Studien zum Öffentlichkeitsbegriff bei Kant (Frankfurt am Main:
Haag und Herchen, 1987), 342.
17
Gerhardt, Immanuel Kants Entwurf “Zum ewigen Frieden,” 200–201.
In the first formula, the publication of the maxim is a hypothesis, a mental

experiment: if published, it would provoke opposition on the part of the
public. In the second formula, the publication of the principle is real: this
maxim can only reach its objective if it is made known to the public.18
2) The principle of publicity implies, par excellence, not the action of the
individual, but of the public. In his essay “What Is Enlightenment?” Kant
makes the proper individual responsible for remaining in a minority state
since it is more comfortable. But soon the responsibility is attributed to the
person’s tutor who prevents him of her from walking alone. However, the
difficulty an individual faces in leaving a state can be overcome by the public
– although slowly, since it needs “true reform in one’s way of thinking” (WE
8:35–36). Thus, the reference to the Enlightenment is always as a process;
not as a clarified era, but as an era of clarification, as happens historically over
time (WE 8:40–42). It is a learning process that occurs by means of errors –
only after some failings do we learn to walk, says Kant (WE 8:35–36). In this
same sense, in his Religion Kant says that he does not feel well when he hears
that a certain people, or human beings in general, are not “ready” for
freedom, since it is only by asserting our freedom through our own efforts
– even if this is much more laborious and dangerous than when we follow a
tutor – that we become “ready” for freedom (Rel 6:188).
The moral requirement of public participation in the accomplishment
of rights, the honor that it gives to right – even while wanting to escape
it, even if it is mere “rhetoric” – implies a legitimacy criterion: states
must speak in the name of the “law” because there is a moral disposition
in the moral human being (PP 8:375–76). Volker Gerhardt observes that
this does not mean judging the world as it is by how it should be, which
would only be an external subordination of politics to rights. It means an
internal bond between politics and rights, a criterion according to which
something only occurs in practice when it is in agreement with theory:
politics needs to legitimize itself, evoking rights ahead of the public in
order to reach its objectives. Kant gains a transcendental principle of
politics with the positive formula of publicity, which allows politics to
internally bind itself to rights and to morality. If the necessity of public
legitimization was always a basic element of political thought, this was
the first time it had been inserted into a philosophical system uniting
morality, rights, and politics.19
18
Blesenkemper, “Publice age,” 351.
19
Gerhardt, Immanuel Kants Entwurf “Zum ewigen Frieden,” 45–49. See also MM 6:324, 329.
606 S. N. Sckell
We cannot therefore agree with the thesis that for Kant progress is
ultimately achieved by “unsociable sociability,” or that progress is promoted
“from above,” “paternalistically,” by government20 – that moral politicians
dictate the path to the republic to as-yet-immoral citizens. In Perpetual Peace,
the role of the “nation of devils” is first followed by the role of the moral
politician, but this role is finally surpassed by the role of the public acting
according to the principle of publicity. The public is then the determinative
point of view21 in which “the union of the ends of all [is] possible” (PP
8:386). This public follows a rational principle, while the nation of devils is
subject to the mechanism of unsociable sociability; the principle of publicity
is an ever-valid principle that can only lead to peace, while the mechanism of
unsociable sociability leads as much to war as to peace. The moral politician’s
government cannot be a solution, as he or she is prone to the inevitable
corruption of reason by power. It is true that the public is also prone to
perversion. History gives us many examples of this fact. But Kant’s grandi-
osity does not lead him to assert that the perversion of the public should be
corrected through economic liberalism or the political power of regents, but
through an even broader public. It is only the most cosmopolitan public, and
thus the most universal public possible, that can correct the public that
particularizes itself by excluding others.
3) The French Revolutionaries displayed a local struggle for the realization
of rights on an internal level. But also important for Kant is the Teilnehmung
– sympathy, the attitude of taking part in what is happening – of spectators
of the French Revolution. The French Revolutionaries together with specta-
tors throughout the world constitute a transborder cosmopolitan
community.
The enthusiasm of the spectators of the French Revolution is explained by
the aesthetic of the sublime, which is “not an object for taste, but rather an
object for the feeling of emotion” (An 7:243).22 This sublime affection is
enthusiasm, a “vibration [Erschütterung]” (CJ 5:258). The human soul, says
Kant, “hails, with such universal and impartial sympathy, the hopes for its
success and the efforts toward realizing it” (CF 7:87); the pure concept of
right awakened “zeal and grandeur of soul” in the revolutionaries – “and even
the concept of honor among the old martial nobility . . . vanished before the
20
Bernd Ludwig, “Will die Natur unwiderstehlich die Republik? Einige Reflexionen anlässlich einer
rätselhaften Textpassage in Kants Friedensschrift,” Kant-Studien 88, no. 2 (Jan. 1997): 226.
21
Ingeborg Maus, Zur Aufklärung der Demokratietheorie: Rechts- und demokratietheoretische
Überlegungen am Anschluss an Kant (Frankfurt am Main: Suhrkamp, 1994), 127–28.
22
Gérard Raulet, Histoire et citoyenneté (Paris: PUF, 1996), 209.
weapons of those who kept in view the right of the nation to which they
belonged and of which they considered themselves the guardians” (CF 7:86).
This experience is not simply an empirical phenomenon that can be
disproved by failure or progress. It cannot be forgotten, revealing a relation-
ship that is outside the temporal order “(signum rememorativum, demonstra-
tivum, prognostikon)” (CF 7:84).23 Even if afterward everything remains as it
was before, says Kant,
that philosophical prophecy still would lose nothing of its force. – For that
occurrence is too important, too much interwoven with the interest of human-
ity, and its influence too widely propagated in all areas of the world to not be
recalled on any favorable occasion by the nations which would then be roused
to a repetition of new efforts of this kind. (CF 7:88)
The faculty of judgment disregards immediate data and takes up universality,

just as the cause of the enthusiasm of the spectators of the French Revolution
was the right of a people not to be prevented from giving itself a political
constitution, and a republican constitution at that (CF 7:85).
In the enthusiasm of the spectators of the French Revolution, those who
could not gain any advantage from it, Kant sees that human beings do not
seeks rights simply as a means of satisfying their particular interests; this
feeling, that one can have only for an ideal, indicates a second link between
politics and law, contrary even to the philosophy of history. In the philosophy
of history, human beings seek rights as a means to ensure their own interests;
the enthusiasm of the spectators of the French Revolution, on the contrary,
shows that right can be sought as an end: human beings are enthusiastic in the
realization of rights even if no particular interest can be satisfied by this
realization. According to the philosophy of history, human beings seek rights
to protect their lives and property, but when individuals show enthusiasm for
the French Revolution, they endanger their lives and property in order to pay
tribute to the realization of rights. This is the demonstration that human
beings are able, since they are free and rational, to be the authors of their own
history, which would then not be conducted only by self-interest.
Revolutionaries and enthusiastic spectators of the French Revolution
constitute a “cosmopolitan transborder network.” Spectators in other coun-
tries – such as Kant in Prussia – stood next to the revolutionaries even if they
were not fighting. People were connected by means of sympathy and
23
Ibid., 210.
608 S. N. Sckell
enthusiasm, across borders. This showed that what was happening was not
only of local interest, as a domestic issue, but of cosmopolitan interest – an
issue that concerned all humanity. And that this would be the condition for
its attainment, not only at the local level, but also at the cosmopolitan level.
It showed that the same thing could occur in other countries for the
democratization and cosmopolitization of national law. In the community
of the revolutionaries in France and the spectators in others countries, we see
a kind of cross-border community that struggled for cosmopolitan law.
Cosmopolitics today
Kant’s cosmopolitanism becomes then the practice of citizenship as “activity”

and not as status, as a mode of socialization that produces “collective
militants” on a cross-border level and in any form of organization. This is
the idea of understanding the practice of new resistance movements, in
which we can find with Kant the possibility of creating cosmopolitan citizen-
ship – “citizenship in network,” in contrast to “territorial citizenship.”24
Although from the perspective of the theories of representations it may be
difficult to see the legitimacy of these spontaneous forms of extra-parliamen-
tary associations, since it fails the moment of “authorization,” we could
consider that their legitimacy comes from the moment of “control” and
“accountability,” which requires a reformulation of the theory of democratic
representation. This practice not only creates a fusion of freedom and
equality, but mostly aims at the transformation of given conditions of social
formations and question the violence in the production of “identities.”
Cosmopolitics, cosmopolitanism, and identity issues

When Kant presents his concept of cosmopolitan right and condemns with
this concept the crimes of colonization, he observes how these crimes are
grounded in a depreciative image of the enemy: colonizers “counted the
[native] inhabitants as nothing” (PP 8:358).
As formulated in antiquity, cosmopolitanism is the moral ideal of a
universal community of human beings considered apart from their links to
particular communities. Cosmopolitanism, in this sense, has as an ethical
24
Balibar, “Cosmopolitisme, internationalisme, cosmopolitique,” 19–49.
horizon the construction of a cosmopolitan worldview, a cosmopolitan self.

But the construction of a cosmopolitan self has to face within the psyche
exclusive private bonds such as nationalism, racism, sexism, and all forms of
discrimination implied in identity issues.25 An individual constructs an
identity that excludes all others and all traits of otherness in a conception
of “we.” Thus, violence is produced by the imposition of exclusive identities.
In recent years, identity has become central in cosmopolitan law issues of
human rights and international crimes.26
Indigenous peoples
Although the main political problem Kant was dealing with in developing his
concept of cosmopolitan law was colonization, the current dramatic situation
of indigenous people has received insufficient attention in Kantian recon-
structions of cosmopolitan law.27 However, violation of indigenous people’s
rights continues to be one of the most serious problems of cosmopolitan law
in the domestic spheres of former colonies. For example, Brazil’s democratic
laws show that, through various mechanisms, we are still living with the
problem faced by Kant: the colonization of the land of the Indians.28
Brazil’s 1988 Constitution, established after the military regime, has some
articles with emancipatory content, which nevertheless become ineffective by
means of other legal provisions. The Constitution stipulates in Article 4 that
territories originally occupied by the Indians belong to the Union – that is,
they are not owned by the Indians, but rather belong to the Union as if they
were the “land of anyone.” As a consequence of indigenous movements of the
1980s, the Constitution guarantees in Article 231 that Indians have the right
to use and enjoy land, rivers, and lakes in the territories originally occupied
by them. But first it is necessary to define the borders of this original
territory, which has not been done for half of the indigenous population of
the country; and although the expression “originally occupied” is used, the
criterion is “occupation on 25th October 1988,” the date of the
25
Étienne Balibar, La crainte des masses: Politique et philosophie avant et après Marx (Paris: Galilée, 1997).
26
Matthias Kaufmann, Em defesa dos direitos humanos: Considerações históricas e de princípio (São
Leopoldo: Unisinos, 2013).
27
One exception to this is Vicki A. Spencer, “Kant and Herder on Colonialism, Indigenous Peoples,
and Minority Nations,” International Theory 7, no. 2 (July 2015): 360–92.
28
See Francisco de Vitoria, “On the American Indians [De Indis recenter inventis et de Jure belli
Hispanorum in Barbaros relectiones]” (1539), in Political Writings, ed. Anthony Pagden and Jeremy
Lawrance (Cambridge: Cambridge University Press, 1991), 233–92.
610 S. N. Sckell
Constitution. Second, in demarcated territories, the Union, which owns the

property, can build hydroelectric dams, highways, and so forth on the land.
The construction of hydroelectric dams in the Amazon Forest has become a
priority for the state (see the case of Belo Monte). Moreover, in the indi-
genous territories, Indians can enjoy the ground, but not the underground,
which can be explored by authorized mining companies. With electricity
being generated by hydroelectric dams, these companies will be able to settle
in the forest and extract mineral resources on an industrial scale. Still, in
demarcated territories, illegal activities such as logging and mining of mineral
resources are widely ingrained, tolerated, and go unpunished. Kant’s con-
cerns with indigenous people, often called a historical case of a colonial past,
seem very relevant to contemporary sovereign democracies.
The issue of migrants, refugees, asylum-seekers,

and stateless and displaced persons
In the 1990s, Kantian reconstructions of cosmopolitan law centered on the

question of intervention (for instance, Jürgen Habermas’s first reconstruc-
tions of Kant’s cosmopolitan law).29 In contemporary discussions, the major
issue has become defending the welcoming of migrants, refugees, asylum-
seekers, and stateless and displaced persons.30 Some authors, such as Jacques
Derrida, oppose Kant for having restricted cosmopolitan law to the right of
hospitality: welcoming of foreigners, in any condition – as migrant, refugee,
asylum-seeker, or stateless or displaced person – should be considered an
absolute and unrestricted principle of justice that can hardly be exhausted by
a limited Kantian “right to hospitality.”31 Other authors, however, such as
Pauline Kleingeld, in reconstructing Kant’s cosmopolitan law, think it is
possible with Kant to justify the obligation to give foreigners a safe harbor.32
29
Jürgen Habermas, “Kant’s Idea of Perpetual Peace, with the Benefit of Two Hundred Years’
Hindsight,” in Perpetual Peace: Essays on Kant’s Cosmopolitan Ideal, 113–53.
30
See several works of Georg Cavallar, especially The Rights of Strangers: Theories of International
Hospitality, the Global Community, and Political Justice since Vitoria (Aldershot: Ashgate, 2002).
31
Jacques Derrida, Cosmopolites de tous les pays, encore un effort! (Paris: Galilée, 1997).
32
Kleingeld, Kant and Cosmopolitanism, 77–78. For further discussion, see Martha Nussbaum, “Kant and
Stoic Cosmopolitanism,” Journal of Political Philosophy 5, no. 1 (March 1997): 1–25; Seyla Benhabib, The
Rights of the Others: Aliens, Residents and Citizens (Cambridge: Cambridge University Press, 2004), esp. ch.
1; Peter Niesen, “Colonialism and Hospitality,” Politics and Ethics Review 3, no. 1 (April 2007): 90–108;
Rainer Keil, Freizügigkeit, Gerechtigkeit, demokratische Autonomie (Baden-Baden: Nomos, 2009); Garrett
W. Brown, “The Laws of Hospitality, Asylum Seekers and Cosmopolitan Right,” European Journal of
Political Theory 9, no. 3 (July 2010): 308–27; Jürgen Habermas, “La France et l’Allemagne doivent
Kant stated that those who try to save their lives in a foreign country (for
instance, a ship in a storm) cannot be sent back into danger or captured, but
should stay until they have a favorable occasion to leave (Ak 23:173). Even if
Kant wrote little on this question, this interpretation is the only one that is
consistent with the spirit of Kantian cosmopolitanism. Cosmopolitan law
can only make sense if it absolutely guarantees the human rights of each
person as a citizen of the world and not of a particular state.
Identity issues in cosmopolitan institutions
Identity issues have also become a crucial problem in the practice of the
forums of cosmopolitan law, as for example the Inter-American System
of Human Rights (protection of cosmopolitan human rights, even if
with a limited regional jurisdiction) and the International Criminal
Court.
From the beginning of its operation in 1979, the Inter-American
System of Human Rights (ISHR) was mainly concerned with systematic,
mass violations of human rights by the military dictators in Latin
America. In its current phase,33 the priority of the ISHR is to monitor
the demands of excluded groups who are affected in terms of rights to
participation and expression, who suffer social or institutional violence,
and who have difficulty accessing the public sphere, the political system,
and social or legal protection – violence practiced by police against
certain groups, violence against women tolerated by state authorities,34
deprivations of land and the political participation of indigenous peoples
and communities, discrimination against the Afro-descendant popula-
tion,35 and abuse on the part of bureaucracies against undocumented
immigrants. Individual cases reveal structural discrimination (even
prendre l’initiative,” Le Monde, September 9, 2015, www.lemonde.fr/idees/article/2015/09/09/la-france-

et-l-allemagne-doivent-prendre-l-initiative_4750233_3232.html; and Matthew C. Altman, “The Limits of
Kant’s Cosmopolitanism: Theory, Practice, and the Crisis in Syria,” Kantian Review 22, no. 2 (June 2017):
179–204.
33
Víctor Abramovich, “Das violações em massa aos padrões estruturais: Novos enfoques e clássicas
tensões no sistema interamericano de direitos humanos,” SUR – Revista Internacional de Direitos
Humanos 6, no. 11 (Dec. 2009): 7–39.
34
Inter-American Commission on Human Rights (IACHR), Maria Da Penha Maia Fernandes v. Brasil,
Case 12.051, Report no. 54/01, April 16, 2001.
35
Inter-American Commission on Human Rights (IACHR), Wallace de Almeida v. Brasil,
Case 12.440, Report no. 26/09, March 20, 2009.
612 S. N. Sckell
through “neutral practices”) and violence against certain groups.36 It is

important to note that most of the issues are brought to the court by
civil associations and NGOs, and that the action of the court is mon-
itored by these associations.37
Identity issues have also become a crucial problem in the International
Criminal Court (ICC), particularly regarding gender issues. The first deci-
sion of the ICC, condemning Thomas Lubanga, a militia captain in the
Congo, for war crimes – the use of child soldiers in the conflict – was
considered a milestone in the history of international justice and individual
responsibility. However, the trial also proved once again that the violence
suffered by women and girls in armed conflict often remains invisible. Due to
“insufficient evidence,” Lubanga was charged only with conscripting and
enlisting children under the age of fifteen and using them to participate
actively in hostilities. He was not accused of having raped women and girls in
enemy villages, nor of sexual violence and sexual slavery of the abducted girls.
This decision was harshly criticized by several NGOs, such as Amnesty
International and the Congolese Coalition for Transitional Justice.38 The
response of the court to this crisis of legitimacy caused by its first decision
was to establish the gender issue as one of its key strategic goals.39
The developments in these courts show that Kant’s concerns with the
indigenous people in cosmopolitan law have been, and should be, extended
to any type of discriminated group.
Translation and identity issues
One form of resistance to this violence toward discriminated groups, a form

of cosmopolitanism that inspires some contemporary political programs, is
“multiculturalism,” which takes on very different meanings depending on the
different ways of defining the term “culture” (for instance, the definitions of
36
Inter-American Court of Human Rights (IACourtHR), Niñas Yean y Bosico v. Dominican Republic,
September 8, 2005.
37
Abramovich, “Das violações em massa aos padrões estruturais.”
38
International Criminal Court (ICC), Case ICC-01/04–01/06, Decision, Prosecutor v. Thomas
Lubanga, July 10, 2012.
39
International Criminal Court, Office of the Prosecutor, Policy Paper on Sexual and Gender-Based
Crimes, June 2014, www.icc-cpi.int/iccdocs/otp/OTP-Policy-Paper-on-Sexual-and-Gender-Based-
Crimes–June-2014.pdf.
Charles Taylor and Will Kymlicka). Other theories focus on translation40 as a

prerequisite for effective universalism, which depends on the ability to
establish successful communication without having pre-established common
codes. Multilingualism, as Balibar conceives of it, is an essential condition of
nomads who cross traditional boundaries, who are always multilingual: an
idea conceived in the use of multiple languages is as a result of translations
and movements, which is thus a prerequisite for a cosmopolitan existence.41
Translation does not mean a relativism that could accept discrimination and
inequality, but is legitimate only if it amplifies rights and their effectiveness.
If, for instance, an Ashininka – an indigenous person who considers
several animals to be sacred – were to translate the concept of human rights,
he or she would understand the concept of dignity (being an end in itself and
not a means to another end) not only as a human attribute, but also an
attribute of nature. This reveals that human rights are based on a premise –
anthropocentrism – that is not a universalism. This does not undermine
human rights as a construct of Western civilization, which in principle
cannot be a value for an Ashininka, but they must be translated, and this
translation must be made by an Ashininka.
This hypothetical translation of human rights by an Ashininka – a translation
in which anthropocentrism could be corrected – would not be sufficient to
consider human rights as really universal, because there would still be a
dominant “language,” a matrix language from which human rights are trans-
lated. It would now be necessary to go another way: “human rights,” as
translated by this Ashininka, would be the matrix from which further transla-
tions would be made. This would even involve a reformulation of the human
rights of Western civilization. And a nonanthropocentric conception of human
rights (that of this indigenous person) could then be the matrix for a reformula-
tion of human rights in developed industrialized countries that could promote a
deeper environmental and cosmopolitan consciousness and cosmopolitanism.
With Kant, we could consider cosmology and cosmopolitanism as intrin-
sically linked. Even if in his practical philosophy Kant claims that only
human beings are ends in themselves, consecrating an anthropocentric
view of world, in his early conception of nature humanity is not prioritized.42
In Universal Natural History and Theory of the Heavens, for example, the
40
Manuela Carneiro da Cunha, “Pontos de vista sobre a floresta amazônica: Xamanismo e tradução,”
Mana 4, no. 1 (1998): 7–22.
41
Balibar, “Cosmopolitisme, internationalisme, cosmopolitique,” 36–42.
42
See Christina Hoff, “Kant’s Invidious Humanism,” Environmental Ethics 5, no. 1 (Spring 1983):
63–70.
614 S. N. Sckell
cosmopolitan idea of a connection of all human beings to each other –

humanity – is intrinsically linked to the idea of an original link between
humanity and the cosmos. Humanity as a whole is part of another whole, but
this part is not a fundamental one. Nature has existed long before humanity
came into existence, and if humanity one day disappears, it would cost
nothing for nature, says Kant. Since Kant declared that his feelings and
thoughts were devoted to the starry heavens above him and the moral law
within him, the intrinsic link between these two domains is essential to his
philosophy (CPrR 5:161). A cosmopolitan consciousness is in this way also
intrinsically linked to our consciousness of being part of the cosmos, just like
other nonhuman beings.
Conclusion
Kant’s concept of cosmopolitan law, conceiving the individual as a citizen of the
world, is nowadays the principle underlying human rights and international
crimes. With Kant we can think that this should also be the fundamental
principle first of all for a local democracy in a limited territorial juridical-
political-framework: a political system can be considered a democracy only as
a cosmopolitan democracy, that is, one that actively respects the cosmopolitan
rights of all the world’s citizens. At the international and global level, the
comparison between two juridical-political instruments, the individual petition
for human and labor rights and the individual responsibility for international
crimes, shows a fundamental difference between them that must be surpassed.
With Kant we can also think that violations of human rights should carry
individual international responsibility (which is previewed only for interna-
tional crimes) instead of the archaic international collective responsibility.
Kant’s cosmopolitan law implies cosmopolitan citizenship. Indeed, the
cosmopolitization of democracy must be followed by the democratization of
the cosmopolis – which supposes cosmopolitan citizenship, the cross-border
associations of individuals. Kant presented the “principle of publicity,” which
has a cosmopolitan dimension, as the criterion for cosmopolitan justice, and
saw the condition of possibility for the development of cosmopolitan law –
and of a cosmopolitan democracy – in the link between the French
Revolutionaries and their enthusiastic “spectators” abroad. With Kant, we
can say that cosmopolitan law can be created and legitimated only by
cosmopolitan citizenship, which does not exist as a legal and political statute
but as a practice of cross-border associations in political struggles. This
concept differs from those of “world society,” “world public sphere,” or
“collective conscience,” since it focuses on the dimension of the struggle and

not of consensus. It also demands a reformulation of the liberal theory of
democratic representation: transnational civil associations do not have the
moment of “authorization,” but they do have the moments of “control” and
“accountability.” Furthermore, the established content of cosmopolitan law
is in turn, reciprocally, the criterion of legitimacy of the exercise of cosmo-
politan citizenship. This can delegitimize associations that claim a non-
cosmopolitan identity, that oppress their members and exclude others.
It is then necessary to overcome readings of Kant’s cosmopolitan law
centered on its undeniable universal moral value to consider it as a construction,
one that it is produced and legitimized by cosmopolitan citizenship in practice.
As nothing is less obvious than feeling oneself a victim of injustice and having
rights, there is nothing obvious in the idea of having cosmopolitan rights against
one’s own government or any other dominant power. The difficulty is to
reconcile universal discourse on cosmopolitan law with an analysis of its
socio-historical production through cosmopolitan citizenship. Cosmopolitan
law appears as autonomous, neutral, independent from power relationships,
based on the rationality of morality and thus worthy of universal recognition, a
transcendental principle standing above its historical forms. But neither is it a
product of universal reason nor the imposition of a dominant ideology; it results
from a long, cumulative systematization that produces universalization. The
power of reason is not enough to achieve cosmopolitan law: it is only through
political struggle, “cosmopolitan citizenship” in practice, as Kant saw in the
French Revolutionaries and their spectators, that reason is realized in history.
Moreover, Kant’s cosmopolitan law, produced by cosmopolitan citizenship,
can only have a truly cosmopolitan dimension if it considers identity issues, an
analysis of the way in which social roles and stereotypes are constructed, in
order to surpass them. To give local law a cosmopolitan horizon and transpose
the democratic principles of the nation-state onto international and cosmopol-
itan law in a Kantian sense, it is necessary to amplify democracy to material and
social inequalities, such as inequalities between women and men, and between
citizens and foreigners. Finally, a deep comprehension of Kant’s cosmopolitan-
ism should embrace not only the all of humanity, but the whole cosmos.
27
Kant’s Mature Theory of Punishment, and
a First Critique Ideal Abolitionist Alternative
Benjamin Vilhauer
Introduction
This chapter has two goals. First, I will present an interpretation of Kant’s
mature account of punishment, which includes a strong commitment to
retributivism. Second, I will sketch a non-retributive, “ideal abolitionist” alter-
native, which appeals to a version of original position deliberation in which we
choose the principles of punishment on the assumption that we are as likely to
end up among the punished as we are to end up among those protected by the
institution of punishment. This is radical relative to Kant’s mature theory of
punishment, but arguably it conforms better to the spirit of Kant’s first Critique
remarks on imputation and punishment than his mature theory does.
Overview
Much of the interest and frustration of philosophy arises from the discovery
that concepts that we use with easy consensus in everyday life become
puzzling upon reflection. Concepts whose everyday use is fraught with
controversy are even more confusing. This is common in ethics, but the
concept of just punishment is an especially difficult case, since even a bit of
B. Vilhauer (*)
Department of Philosophy, City College of New York, New York, USA
e-mail: bvilhauer@ccny.cuny.edu

618 B. Vilhauer
reflection can raise doubts about its coherence. Though no two schools of
moral philosophy express this point in quite the same way, morality’s
primary orientation seems to be the provision of reasons for us to avoid
harming each other, and to help each other at least some of the time. Arguing
for the possibility of just punishment seems, at least at first blush, to demand
an inversion of that orientation, since it involves marshalling moral reasons
in favor of harming people. The oddity of this prompts some philosophers to
deny that there can be just punishment, and to advocate its abolition.
Among philosophers who do think punishment can be justified, two basic
strategies predominate, both historically and in the contemporary literature:
retributivism and consequentialism. Retributivism is the view that punish-
ment is justified by the fact that it inflicts on criminals the suffering they
deserve for their actions. Retributivists typically acknowledge that punishing
criminals can have valuable consequences, but they hold that these conse-
quences are merely fortunate side effects of punishment, not part of what
justifies it. Consequentialists see the justification of punishment in its valu-
able consequences, such as preventing criminals from reoffending and deter-
ring potential criminals. They regard suffering as intrinsically bad, and as
justified only if it is “outweighed” by its beneficial consequences. There are
also various “mixed” justifications, according to which both retributivist and
consequentialist justifications play roles in justifying punishment.
Kant makes remarks at various points that seem to support abolitionism,
retributivism, consequentialism, and mixed justifications:
Abolitionism
A constitution providing for the greatest human freedom according to laws

that permit the freedom of each to exist together with that of others [is a]
necessary idea. . . . The more legislation and government agree with this idea,
the less frequent punishment will become, and hence it is quite rational to
assert (as Plato does) that in perfect institutional arrangements nothing of the
sort would be necessary at all. . . . Even though this may never come to pass, the
idea of this maximum is nevertheless wholly correct when it is set forth as an
archetype, in order to bring the legislative constitution of human beings ever
nearer to a possible greatest perfection. (A316–17/B373–374)
Retributivism
Even if a civil society were to be dissolved by the consent of all its members
(e.g., if a people inhabiting an island decided to separate and disperse
27 Kant’s Mature Theory of Punishment 619
throughout the world), the last murderer remaining in prison would first have
to be executed, so that each has done to him what his deeds deserve . . . for
otherwise the people can be regarded as collaborators in this public violation of
justice. (MM 6:333)
Consequentialism
All punishments by authority are deterrent, either to deter the transgressor

himself, or to warn others by his example. But the punishments of a being who
chastises actions in accordance with morality are retributive.
All punishments belong either to the justice or the prudence of the lawgiver.
The first are moral, the second pragmatic punishments. . . . All the punishments
of princes and governments are pragmatic, the purpose being either to correct
or to present an example to others. Authority punishes, not because a crime has
been committed, but so that it shall not be committed. (LE 27:286)
Mixed Theory
There can be no penal law that would assign the death penalty to someone in a
shipwreck who, in order to save his own life, shoves another, whose life is
equally in danger, off a plank on which he had saved himself. . . . A penal law of
this sort could not have the effect intended, since a threat of an ill that is still
uncertain (death by a judicial verdict) cannot outweigh the fear of an ill that is
certain (drowning). Hence the deed of saving one’s life by violence is not to be
judged inculpable . . . but only unpunishable. (MM 6:235–36)
It is hard to see how all these remarks can fit together in one coherent theory,
and I will not claim to be able to do so here. Kant’s abolitionist remark can of
course be made consistent with the others, since any theory of punishment
can be combined with the view that it would be good for there to be no crime
to punish and thus no punishment. But it is worth noting that Kant’s
endorsement of abolitionism appears at the beginning of the critical period
and is not a theme of his mature account of punishment, a point to which I
will return. The other remarks are not so easy to reconcile. The point of MM
6:333 seems to be that murderers must get their just deserts even if it is
impossible that their execution will deter anyone, and this is inconsistent
with the last two remarks. The third remark suggests that state authorities
should be focused only on deterrence in punishing. The fourth remark
suggests a mixed theory because the claim that saving one’s life by killing
another is culpable but “unpunishable” seems to suggest that just punish-
ment must be deserved but also deterring to be just.
620 B. Vilhauer
We should probably accept that some of the difficulty in assigning Kant’s

position on punishment to one of the common categories used in the
philosophy of punishment arises from genuine inconsistencies resulting
from inadequate attention to detail on Kant’s part. With a bit of cleanup
work, though, we can extract a consistent and interesting theory which is
grounded in his broader ethical and metaphysical views. He offers a complex
mixed theory according to which we are obligated to punish retributively
whether or not it deters, so long as the punishments do not violate a respect
for persons principle to be described below, but which nonetheless appeals to
consequentialist considerations to explain why punishment is a concern of
the state. Before exploring the details, however, it may be worth pausing to
ask why Kant should have anything to do with consequentialism or retribu-
tivism in the first place.
As has been widely observed, the categorical imperative rules out conse-
quentialist reasons as foundational moral reasons. In order to understand the
basic directive to respect human beings as ends in themselves, “the end must
here be thought not as an end to be effected but as an independently existing
end” (G 4:437).1 The rational nature essential to humanity is the only thing
of absolute value, but that value does not give us reasons to, say, maximize
the number of humans. It is an end to be respected rather than effected. The
point is not that there is no reasoning about outcomes in Kantian ethics –
interacting with others in ways that treat them as ends sometimes involves
planning, for example. But the forward-looking considerations derive from
foundational relations to the wills of others that are not fundamentally
forward-looking. For these reasons, charitable interpreters should assume
that whatever commitment to consequentialism is evident in Kant’s remarks
about punishment is one that derives from the underlying principle of
respect for persons.
Is Kant’s ethics retributive in a fundamental way? This is less clear.
Attributing moral responsibility, or “imputation” in Kant’s terms, is a pre-
condition for retribution. If we take the foundation of Kant’s ethics to be
respect for persons, then it is important to note that in a number of remarks
Kant describes imputability as a fundamental characteristic of personhood.
In the Doctrine of Right, he states that “a person is a subject whose actions can
be imputed to him,” while “a thing is that to which nothing can be imputed”
(MM 6:223). But Kant’s remarks on the fundamental characteristics of
1
For a helpful discussion of this point, see Allen W. Wood, Kantian Ethics (Cambridge: Cambridge
personhood are diverse, complex, and scattered throughout his texts, and do
not always refer to imputation. For example, in the Groundwork, he tells us
that
every rational being, as an end in itself, must be able to regard himself as also
giving universal laws with respect to any law whatsoever to which he may be
subject; . . . this dignity . . . brings with it that he must always take his maxims
from the point of view of himself, and likewise every other rational being, as
lawgiving beings (who for this reason are also called persons). (G 4:438)
Does the idea of seeing ourselves as ends, and giving universal law along with
all other rational beings, imply imputability? Clearly Kant’s view is that it
does, but his argument for this is not as clear as one might wish. Further,
even if we accept that imputability is required for personhood, would we
have to accept that all actions are imputable, or would it be sufficient to
impute just some? Could there be special reasons of justice to avoid imputing
blameworthy actions when serious retributive harm is at issue? I will argue
below that Kant’s first Critique view implies that there are such reasons, and I
will show how this view disappears in his mature theory of punishment.
After offering an interpretation of Kant’s mature account of punishment, I
will propose a first-Critique-inspired reconstruction. Kant’s mature view is
that punishing in the absence of a retributive justification treats criminals as
mere means, but if we emphasize another aspect of treating persons as ends,
that is, respect for rational consent, then I think we can build what we might
call an “ideal abolitionist” approach to punishment which avoids treating
criminals as mere means and gives them an equal voice in the social contract
by choosing principles of punishment in an original position in which we
assume we are as likely to end up among the punished as we are among those
protected by the institution of punishment.
Kant’s mature account of punishment

With bit of charitable cleanup work, we can see that Kant’s mature
account of punishment involves three independent justificatory princi-
ples: a consequentialist principle that says the authorities have a duty to
hinder hindrances to rights-violations with permissible means; a retribu-
tive principle that says we should punish all and only those who deserve
it, with a kind and quantity of punishment equal to the crime; and a
respect for humanity principle that sets limits to the kind and quantity of
622 B. Vilhauer
punishment even in cases where violation of those limits would yield

more perfect retributive equality.
The consequentialist principle

The consequentialist principle in Kant’s account of punishment is drawn out
of his account of the rights of persons and the social contract. On Kant’s
view, our only “innate right” is “freedom (independence from being con-
strained by another’s choice), insofar as it can coexist with the freedom of
every other in accordance with a universal law”; this includes “innate equality,
that is, independence from being bound by others to more than one can in
turn bind them” (MM 6:237). Others are obligated to respect our equality in
the state of nature, and we can hope that they do, but we are also justified in
enforcing it, which we may do by coercively establishing what Kant calls a
juridical state of affairs, or a condition of right:
the state of nature . . . would still be a state devoid of justice (status iustitia
vacuus), in which when rights are in dispute (ius controversum), there would
be no judge competent to render a verdict having rightful force. Hence each
may impel the other by force to leave this state and enter into a rightful
condition. (MM 6:312)2
A rightful condition is structured by what Kant calls the “Universal Principle

of Right”: “‘Any action is right if it can coexist with everyone’s freedom in
accordance with a universal law, or if on its maxim the freedom of choice of
each can coexist with everyone’s freedom in accordance with a universal law’”
(MM 6:230). If my action is right, then “whoever hinders me in it does me
wrong; for this hindrance (resistance) cannot coexist with freedom in accor-
dance with a universal law” (MM 6:230–31). Coercion within a condition of
right is justified when it is a rightful response to wrongful hindrance: “if a
certain use of freedom is itself a hindrance to freedom in accordance with
universal laws (i.e., wrong), coercion that is opposed to this (as a hindering of
a hindrance to freedom)” is right (MM 6:231). What Kant presents in this
account of the justification of coercion is not an argument from premises
about right to a conclusion about coercion but rather an analysis of what he
2
The details of Kant’s explanation of establishing a condition of right appeal to acquired rights to
property (MM 6:312), but this appears to be an expository matter rather than a necessary feature of his
argument, and I think it obscures its basic structure.
takes to be implicit in the concept of right (MM 6:232). In a civil state, we

cede our individual freedom to coercively protect equality to the authorities,
and the authorities take on a duty to do this. This yields a justification of the
authorities’ duty to coerce compliance with law which is consequentialist but
markedly different from utilitarian consequentialism, since it aims not at
happiness but instead at protecting freedom.3
Kant explicitly makes our right to hinder hindrances a right to con-
strain actions, not to constrain goodness of will or maxims of actions.
This is the basis for his distinction between the distinct legislations of
reason in ethical lawgiving, the topic of the Doctrine of Virtue, and
juridical lawgiving, the topic of the Doctrine of Right. Ethical lawgiving
requires an incentive which is itself ethical. But whether we follow the
law because it is right is not relevant for juridical lawgiving. Juridical
lawgiving couples laws with incentives of “aversion” which coerce com-
pliance so that all members of society are motivated to comply irrespec-
tive of the quality of their wills (MM 6:218).
When Kant speaks of coercively hindering hindrances through incentives
of aversion, he seems to have fear of punishment chiefly (perhaps exclusively)
in mind. But he does not argue that the authorities’ duty to coerce through
aversion contributes to the justification of punishment, and I think his
considered view is that it does not. This is sensible, because there is no
necessary connection between coercing through aversion and punishing to
deter. There are many ways of coercing, and many kinds of aversions: for
example, clever engineers may someday offer a roving corps of ticklebots
always on the scene to make would-be criminals collapse in unbearable
giggles before completing criminal acts.
The notion of a duty to coercively hinder hindrances provides no sub-
stantive guide to the means we may permissibly take to this end. When we
focus on particular wrongful acts in progress, we may seem to get clear and
intuitively satisfactory guidance – it may seem that we are entitled to use
whatever means are necessary to prevent that wrong from occurring. For
example, a duty to coercively hinder hindrances can seem sufficient to justify
applying just enough gentle, gradual pressure to the fingers of a potential
apple thief to loosen his grip on my apple if this suffices to prevent the theft.
But it is not intuitively sufficient when we look at the wider range of coercive
means. What if our would-be apple-snatcher is a dedicated but nonviolent
3
This point is emphasized in Don E. Scheid, “Kant’s Retributivism,” Ethics 93, no. 2 (Jan. 1983): 262–
82.
624 B. Vilhauer
thief who is bruising only produce in his thievish tenacity, but the only
means at my disposal sufficient to prevent the theft is to seriously injure or
kill him? If this is a rightful way of stopping the theft, then the fact that it is
necessary to hinder a hindrance to my freedom does not provide an intui-
tively satisfactory explanation of why it is.
The questions become even more perplexing when we turn to coercive
mechanisms that may be imposed on a criminal after the crime, such as
preventative imprisonment or punishments that deter others. These
practices are not aimed at any particular actual crime but instead at a
field of possible crimes in the future. We might aim to prevent or deter
all crimes, but even if we imposed a public, maximally painful execution
on an apple-thief, we could not be sure of deterring all future apple-
thieves. Even if we could be sure to deter them all, the “hindering
hindrance” idea cannot plausibly be thought sufficient to allay moral
concerns about severe punishments.
The consequentialist principle in Kant’s account of punishment is also of
little help in explaining whom we should punish. What if we could not
capture the actual criminal? Could we frame and punish a scapegoat in his
place to generate deterrence? Kant recognizes this as a challenge to purely
consequentalist justifications:
justice does not follow, if it is inflicted to improve the criminal, or as an

example to others. This would simply have to do with its usefulness, and then it
would be merely a means to that intent, for example, if somebody is flogged,
whether guilty or not, in order to frighten people by his outcry, and create an
impression. (LE 27:553)
To rule this out, we must regard punishment as “an immediately necessary

consequence of the morally bad act” (LE 27:552–53), which Kant sees as an
element of the retributive justification to be discussed below.
Kant himself may be tempted by the thought that guidance about
means flows from the idea of reciprocal coercion. He says that the “law
of a reciprocal coercion” can be constructed “by analogy with . . . bodies
moving freely under the law of the equality of action and reaction” (MM
6:232), which may be meant to suggest a necessary connection with the
retributive equality in ius talionis (to be discussed below). But it is far
from clear how an analogy between people interacting in conformity
with duties of right and bodies moving in conformity with natural laws
could provide a morally substantive link between the concept of recipro-
cal coercion and the idea of retributive equality. Presumably Kant is not
making the crude claim that the law according to which, when rock A
hits rock B, rock B hits A back just as hard, helps justify ius talionis.
However, even the more abstract notion of a construction by analogy
endows this natural law with more moral significance than seems
sensible.4
Another point at which Kant may seem to suggest that guidance about
legitimate punitive means flows from the scheme of reciprocal coercion is
when he claims that “the mere idea of a civil constitution among human
beings carries with it the concept of punitive justice belonging to the supreme
authority.” However, he goes on immediately to point out the question of
“whether it is a matter of indifference to the legislator what kinds of punish-
ment are adopted, as long as they are effective measures for eradicating
crime” (MM 6:362), implicitly acknowledging that the answer to this ques-
tion does not follow from the idea of a civil constitution. He then goes on to
refer to both the retributivist principle (ius talionis) and the respect for
persons principle (“crime against humanity as such”) discussed below as
necessary parts of the answer.
For these reasons, the consequentialist dimension of Kant’s account
plausibly generates the principle that the authorities are obligated to use
morally permissible means to the end of coercing compliance with laws that
protect right, but is properly seen as silent about what the morally
permissible means are.5
4
For the view that retributivism can be “constructed” along these lines, see Susan Meld Shell, “Kant on
Punishment,” Kantian Review 1 (March 1997): 115–35; and Jane Johnson, “Revisiting Kantian
Retributivism to Construct a Justification of Punishment,” Criminal Law and Philosophy 2, no. 3
(Oct. 2008): 291–307. For the view that it cannot, see Paul Gorner, “The Place of Punishment in
Kant’s Rechtslehre,” Kantian Review 4 (March 2000): 121–30.
5
This explanation of the consequentialist principle’s contribution to Kant’s justification of punishment
is related to, but distinct from, explanations by some recent scholars that draw on Rawls’s distinction
between justifying a practice and justifying actions based on rules internal to the practice. See John
Rawls, “Two Concepts of Rules,” Philosophical Review 64 (1955): 3–32. Their notion is that, for Kant,
consequentialism explains why it is morally appropriate to have a practice of punishment, and
retributivism regulates the rules internal to the practice. But since Kant’s notion of the duty to coerce
provides no guidance about means, it has no necessary connection to punishment, and it therefore
cannot provide what H. L. A. Hart calls the “general justifying aim” of punishment in any sense in
which it does not also provide the general justifying aim of ticklebots (Punishment and Responsibility:
Essays in the Philosophy of Law, 2nd ed. [Oxford: Oxford University Press, 2008]). For versions of this
approach, see B. Sharon Byrd, “Kant’s Theory of Punishment: Deterrence in Its Threat, Retribution in
Its Execution,” Law and Philosophy 8, no. 2 (Feb. 1989): 151–200; and Thomas E. Hill, Human Welfare
and Moral Worth: Kantian Perspectives (Oxford: Oxford University Press, 2002). For a critique, see Allen
W. Wood, “Punishment, Retribution, and the Coercive Enforcement of Right,” in Kant’s “Metaphysics
of Morals”: A Critical Guide, ed. Lara Denis (Cambridge: Cambridge University Press, 2010), 111–29.
626 B. Vilhauer
The retributive principle

That Kant is committed to a retributive principle in the justification of
punishment is clear from at least the Collins lecture notes (dating from
1784–1785, but probably copied from other students’ lecture notes dating
from the mid- to late-1770s) onward, for example, at LE 27:286 (cited
above). In that passage Kant is unclear about the relationship between
deterrence and retributivism in the justification of punishment of criminals
at the hands of the authorities. On the one hand, he states that “all punish-
ments by authority are deterrent. . . . But the punishments of a being who
chastises actions in accordance with morality are retributive,” seeming to
claim that deterrence has a role in state punishment that retribution lacks.
But on the other hand, he states that “all punishments belong either to the
justice or the prudence of the lawgiver,” clearly implying that justice is about
retribution and prudence is about deterrence, and thus that retribution has a
role in lawgiving too.
By the time of the Metaphysics of Morals in 1797, Kant makes it clear that
consequentialist considerations are morally and legally irrelevant unless a
retributive justification has been established:
Punishment by a court (poena forensis) . . . can never be inflicted merely as a

means to promote some other good for the criminal himself or for civil society.
It must always be inflicted upon him only because he has committed a crime. . . .
He must previously have been found punishable before any thought can be
given to drawing from his punishment something of use for himself or his
fellow citizens. (MM 6:331)
It is important to emphasize that, on Kant’s account, retributivism implies not

only that punishment is permitted when we legitimately impute a crime, but
also that it is required. This is what Kant means in describing punishment as
“necessary” (LE 27:552, quoted above), in his stern admonition that “the law of
punishment is a categorical imperative, and woe to him who crawls through the
windings of eudaimonism in order to discover something that releases the
criminal from punishment” (MM 6:331), and in his command that the last
murderer be put to death even as society is about to dissolve (MM 6:333,
quoted above). This means that the retributive principle is doing a lot more
heavy lifting in Kant’s account of punishment than the consequentialist prin-
ciple is: the only contribution of the consequentialist principle is to explain why
it is appropriate to turn the retributively justified practice of punishment to the
end of coercing compliance with law.
Earlier we saw that Kant’s consequentialist principle is rooted in his

account of right. Is his retributive principle similarly rooted elsewhere in
his broader moral theory? Kant expends little effort in arguing that it is, and
commentators have understandably taken him to task for this.6 In the
Vigilantius lecture notes, Kant remarks that the necessary link between
“moral badness” and punishment implied in retributivism “cannot be dis-
cerned through reason, nor proved either, and yet it is contained in the
concept of punishment,” apparently suggesting that the spade turns here and
no further explanation can be given (LE 27:552).
However, retributivism has deep roots in Kant’s doctrine of transcendental
freedom, a kind of free will which would make it the case that we have
alternative possibilities of action despite the determinism of the phenomenal
world. The indiscernibility of retributivism is part and parcel of the indis-
cernibility of the transcendental freedom it presupposes. In the first Critique,
Kant holds that we cannot know we are transcendentally free: he claims to
prove only that determinism “does not conflict with causality through
freedom” (A558/B586), and he acknowledges that the idea of transcendental
freedom may be “merely invented” (A545/B573). We can only have knowl-
edge of things that happen in experience as events in a deterministic series,
because the empirical world must be constructed deterministically for there
to be an objective order of time. In the second Critique, there is a shift in
Kant’s epistemology and he argues that “the moral law assures us” of
transcendental freedom, but it remains “inexplicable” (CPrR 5:99).
At A448/B476, Kant sets out a view that he holds throughout the critical
philosophy, that the “transcendental idea of freedom” constitutes the idea of
“the absolute spontaneity of an action, as the real ground of its imputability”
(A448/B476). In a “judgment of imputation” about “blame,” “blame is
grounded on the law of reason, which regards reason as a cause that,
regardless of all the empirical conditions . . . could have and ought to have
determined the conduct of the person to be other than it is” (A555/B583). In
other words, the imputation of blame is grounded on the claim that person
blamed could have acted differently than he in fact acted in doing the
wrongful deed. In more contemporary terms, transcendental freedom serves
as the basis of an alternative possibilities account of moral responsibility and
desert, both of which are presupposed by retributivism. The claim that
someone could have done otherwise is of course not sufficient to justify the
claim that he deserves retribution – it is also necessary, for example, that he
6
See Wood, “Punishment, Retribution, and the Coercive Enforcement of Right,” 121.
628 B. Vilhauer
understood what he was doing and that it was wrong. But Kant is certainly
not alone in thinking the alternative possibilities condition is a necessary
condition which, in combination with whatever other necessary conditions
there may be, is sufficient to justify retribution.
To explain what kind and quantity of punishment we should impose,
Kant turns to what he calls the “law of retribution (ius talionis)”:
what kind and what amount of punishment is it that public justice makes its
principle and measure? None other than the principle of equality. . . .
Accordingly, whatever undeserved evil you inflict upon another within the
people, that you inflict upon yourself. If you insult him, you insult yourself; if
you steal from him, you steal from yourself; if you strike him, you strike
yourself; if you kill him, you kill yourself. (MM 6:332)
There are of course many reasons to object to this principle. One objection
that Kant thinks he can meet concerns the notion of “equality” at work here:
how do we take an eye for an eye if the person to be punished has lost one eye
already? Kant appeals to a notion of similarity when a strict notion of
equality is inapposite:
A fine . . . imposed for a verbal injury has no relation to the offense, for someone
wealthy might indeed allow himself to indulge in a verbal insult on some occasion;
yet the outrage he has done to someone’s love of honor can still be quite similar to
the hurt done to his pride if he is constrained . . . not only to apologize publicly to
the one he has insulted but also to kiss his hand. (MM 6:332)
Clearly such fine-tuning cannot be an exact science, and the inevitable asym-
metries between most criminals and victims make this problem ubiquitous. But
perhaps if we are as exact as we can be, this is not a fundamental limitation in ius
talionis as a moral principle in determining the kind and quantity of punish-
ment. There are, however, cases that do point out such a limitation, that is,
“crimes that cannot be punished by a return for them because this would
be . . . itself a punishable crime against humanity as such, for example, rape”
(MM 6:363). Kant also includes torture and execution by torture in this
category: “No punishment should be coupled with cruelty, i.e., it must not
be so framed that humanity itself is thereby brought into contempt” (LE
27:556). He also states that, when “death is judicially carried out upon the
wrongdoer,” it must “be freed from any mistreatment that could make the
humanity in the person suffering it into something abominable” (MM 6:333).
In cases like the verbal insult case, we ought to return like for like, an injury to
honor with an injury to honor, and the limitation in arranging a relation of like
injury for like injury is due to differences between the agents involved, not due
to a moral prohibition on returning like for like. The question of punishments
themselves punishable as crimes against humanity arises in contexts where there
is a moral prohibition on returning like for like because of the respect for
persons principle discussed in more detail below.
The respect for persons principle

It is easy to overlook Kant’s reliance on an independent respect for persons
principle in his account of punishment when confronted with the remarks on
forfeiture scattered through the Doctrine of Right: a criminal is someone who
has “forfeited his personality” (MM 6:358), the only “human being in a state
[who] can be without any dignity . . . is someone who has lost it by his own
crime” (MM 6:329–30), and a criminal can be “condemned to lose his civil
personality” (MM 6:331). He explains this doctrine at greater length in
“Theory and Practice,” where he explicitly connects the innate “equality”
contained in the innate right to freedom (MM 6:237–38, discussed above)
with criminal forfeiture of this right:
whoever is subject to laws is . . . subjected to coercive right equally with all the
other members of the commonwealth. . . . No one of them can coerce any other
except through public law. . . . through which every other also resists him in like
measure; but no one can lose this authorization to coerce . . . except by his own
crime, and he cannot give it away of his own accord, that is, by a contract, and
so bring it about by a rightful action that he has no rights but only duties; for
he would thereby deprive himself of the right to make a contract and thus the
contract would nullify itself. (TP 8:291–92)
Kant tells us that “beings that have only duties but no rights” are “human
beings without personality (serfs, slaves)” (MM 6:241), and prescribes enslave-
ment for thieves: he remarks that it is “by the principle of retribution” that the
convicted thief “must let [the state] have his powers for any kind of work it
pleases (in convict or prison labor) and is reduced to the status of a slave for a
certain time, or permanently if the state sees fit” (MM 6:333). Together these
remarks suggest that Kant’s doctrine of forfeiture involves bringing his retri-
butive principle to bear upon the scheme of universal reciprocal coercion to
justify the criminal’s expulsion from it. That is, he looks to the retributive
principle to explain why some people lose their right to equal coercive power
630 B. Vilhauer
and can instead be used as means to the end of coercion. Since they could have
avoided committing crimes and therefore deserve to suffer, their expulsion
from this scheme and use as deterrence-generators uses them as means but not
as mere means. If this is right, then the retributive principle operates in a sense
at two distinct levels: it explains why criminals deserve to suffer and why
criminals can be coerced but lose the right to coerce in return (have duties but
no rights).
Now, this idea of forfeiture is crucial in Kant’s account of punishment, but
by Kant’s own lights what he says about it cannot be quite right. If criminals
truly had no rights, then there could be no such thing as a punishment which
is “itself a punishable crime against humanity as such” (MM 6:363) – on
Kant’s theory of right, there can only be such a punishment if criminals can
legitimately coerce others through public law not to impose such a punish-
ment. Kant’s considered view seems to be that we cannot forfeit our person-
ality entirely, no matter how we act. He refers to the unforfeitable core as
“innate personality” (MM 6:331). It is on the basis of this innate personality
that he says, “I cannot deny all respect to even a vicious man as a human
being; I cannot withdraw at least the respect that belongs to him in his
quality as a human being, even though by his deeds he makes himself
unworthy of it,” and on this basis he rules out “disgraceful punishments
that dishonor humanity itself (such as quartering a man, having him torn by
dogs, cutting off his nose and ears)” (MM 6:463).7
The upshot of this discussion is that, while it is clearly Kant’s view that by
the principle of retribution we can forfeit a large proportion of what we tend
to think of as key attributes of personhood in Kant’s ethics, it is equally clear
that there remains a much-diminished but nonetheless crucial notion of
innate personhood that we cannot forfeit, and which plays a necessary role
in constraining the principle of retribution, by limiting the kind and degree
of punishment we can impose by the ius talionis.
An objection to the retributive principle

Critics have made various objections to Kant’s retributivism over the years
and I lack space to comment on them here in any detail, so I will focus on an
objection derived from a point that Kant himself makes in the first Critique.
7
Jean-Christophe Merle also emphasizes the importance of respect for persons, in “A Kantian Critique
of Kant’s Theory of Punishment,” Law and Philosophy 19, no. 3 (May 2000): 311–38.
As mentioned above, transcendental freedom is a power which would

make it the case that we have alternative possibilities of action despite the
determinism of the phenomenal world, and would serve as the “real ground”
of imputability (A448/B476, cited above). As also already mentioned,
according to the first Critique, we cannot know that we have transcendental
freedom. This means we cannot know that our imputations have real
grounds. We apply the idea of transcendental freedom as a “regulative
principle” (A554/B583) that we apply when we impute actions, but we
cannot know that these imputations have the sort of ground they require
to be just.
Kant’s theory of transcendental freedom involves a distinction between
agents’ empirical and intelligible characters. In an agent’s empirical character,
“actions, as appearances, would stand through and through in connection
with other appearances in accordance with constant [deterministic] natural
laws,” and “in combination with these other appearances, they would con-
stitute members of a single series of the natural order.” But we represent the
agent as also having an “intelligible character, through which it is indeed
the cause of those actions as appearances, but which does not stand under
any conditions of sensibility,” and this allows us to represent the intelligible
character as exempt from deterministic laws (A539/B567). A human being’s
intelligible character is the nondeterministic noumenon we posit as the
ground of the actions we impute according to the regulative idea of trans-
cendental freedom. If, as Kant holds in the first Critique, we cannot know
that we are transcendentally free, then we cannot know that we have
intelligible characters that provide real grounds for imputation, and this
calls into question the justice of imputation:
The real morality of actions (their merit and guilt), even that of our own
conduct, therefore remains entirely hidden from us. Our imputations can be
referred only to the empirical character. How much of it is to be ascribed to
mere nature and innocent defects of temperament or to its happy constitution
(merito fortunae) this no one can discover, and hence no one can judge it with
complete justice. (A551n/B579n)
The context of this quotation makes it clear that Kant’s worry was not that,
while we could be sure that our imputations sometimes had real grounds, we
could not be sure when they did, or how robust they were, as he is sometimes
taken to mean. The worry is rather that our imputations might never have
real grounds – that is, grounds in the intelligible character through trans-
cendental freedom – at all, and therefore that the real morality of our actions
632 B. Vilhauer
cannot be judged with complete justice. If we think that imputing blame

must be a matter of complete justice, then this would prohibit it. Some may
worry that this concern is overblown. How much confidence do we really
need about the foundations of retributive justification to legitimately justify
punishment retributively in our everyday legal practices?
In the Vigilantius notes, Kant advocates a very strong standard for the
imputation of blame (imputatio demeriti), that is, one of certainty:
The factum itself, for the man who wishes to impute it, and claim that the
accused is its auctor, must have the utmost moral and logical certainty. . . . For
even the highest degree of probability, if it were to play the part of moral
certainty, would be [accompanied by] the risk of injuring the other’s rights; the
imputation would be staked on such a risk. But here the slightest gamble is at
all times disallowed. (LE 27:566)
It may be objected that surely Kant here means to speak of certainty only on
the matter of whether the particular action was done by the particular person
with a relevant motive, not on the matter of the metaphysical foundations of
imputation. However, the Vigilantius notes have Kant painstakingly drawing
a contrast that appears to be the same as the empirical/intelligible character
contrast that he makes in the first Critique, and stating very clearly that
imputatio facti lies “not simply and solely in the fact that [the criminal] is a
rational being” with a “motive to the action” – it is “absolutely necessary in
addition, that he act with freedom, indeed it is only when considered as a free
being that he can be accountable” (LE 27:559).
It may also be objected that Kant discovers in the second Critique that we
really can be certain that we have transcendental freedom after all, and it is
this discovery that allows imputation to meet the certainty standard pro-
pounded in the Vigilantius notes. That is, by the time we arrive at the second
Critique, all doubt about transcendental freedom has been mooted by Kant’s
discovery of an argument that proves the reality of transcendental freedom in
the “absolute sense in which speculative reason needed it” (CPrR 5:3): the
“moral law is given, as it were, as a fact of pure reason . . . which is apodicti-
cally certain” and it “serves as the principle of the deduction of an inscrutable
faculty. . . . namely the faculty of freedom, of which the moral law . . . proves
not only the possibility but the reality in beings who cognize this law as
binding upon them” (CPrR 5:47). This argument turns on the “ought
implies can” principle: we judge that we “can do something” because we
know we “ought to do it” and thereby “[cognize] freedom within” ourselves
(CPrR 5:30). But this argument violates a basic epistemological constraint
that Kant establishes in the first Critique: it is a transcendental argument that

yields a conclusion about things in themselves rather than the form of
experience. It is also a weak argument, because it is not clear why the
“ought implies can” principle cannot be adequately accommodated by
Kant’s earlier view in the Groundwork, which does not rely on knowledge
of transcendental freedom.
In the Groundwork, Kant argues that rational beings necessarily reason and
deliberate “under the idea of [transcendental] freedom,” and that this implies
that “all laws that are inseparably bound up with freedom hold for him just as
if his will had been validly pronounced free also in itself and in theoretical
philosophy” (G 4:448). Part of the guiding thread in this argument is the
idea that we cannot make a decision about what to do unless we assume that
we have alternative possibilities of action, and when confronted with alter-
natives, the question of which one we ought to pursue inevitably arises, and in
reflecting on this question in pure practical reason we see that the only thing
we unconditionally ought to do is act morally. In this argument Kant’s view is
that the alternatives necessarily posited in practical reasoning provide a
robust enough ontological foundation for “oughts,” and it is not clear why
we ought to reject this view in favor of the second Critique view.
Now, in the Groundwork discussion, Kant makes no reference to his first
Critique worry that our lack of knowledge of transcendental freedom leaves
us with a concern about justice. That is, if all laws bound up with freedom
hold for us, then presumably those include the principles governing imputa-
tion and retribution which become central in the Doctrine of Right. But there
would seem to be plenty of conceptual space for views that combine the
Groundwork’s acting-under-idea view with the first Critique concern. We
could hold, for example, that beings who must act under the idea of
transcendental freedom ought to act under all the laws bound up with
freedom except in cases where our lack of knowledge raises concerns of
justice. Further, Kant’s undeveloped first Critique notion of imputation to
the empirical character might be a useful tool in mapping this conceptual
space: it seems quite intuitive to think that while it may be necessary to
regard myself as having alternative possibilities of action when I deliberate
about how to respond to a crime, I can at the same time regard the crime as
caused by the history of the criminal’s empirical character, and stop short of
the imputation to the intelligible character which would provide a real
ground for imputation.
Kant’s shift to his second Critique epistemology of transcendental freedom
may in part be due to the need to overcome his sense of injustice about the
imputation of blame on uncertain grounds. And overcome it he does, in
634 B. Vilhauer
dramatic style: there he remarks on “children” who are taken to be “born

villains” because they show “such early wickedness and progress in it so
continuously.” Since the “moral law assures us” that this “has as its basis a
free causality” of “evil and unchangeable principles freely adopted,” this
makes their behavior “only more culpable and deserving of punishment”
(CPrR 5:99–100). (This must be one of the saddest things Kant ever wrote.)
Astonishingly, Kant shares with us that, in this way, “appraisals can be
justified which, though made in all conscientiousness, yet seem at first glance
quite contrary to all equity” (CPrR 5:99) As we have seen, the appearance of
contrariety to equity is not just at first glance, but upon thorough considera-
tion in the first Critique.
For these reasons, it seems fair to conclude that the second Critique shift in
the epistemology of imputation fails to meet the “utmost moral and logical
certainty” standard set out in the Vigilantius notes. I think we should also
resist the idea that we should accept a compatibilist or priority-of-the-
practical reconstruction of Kant’s theory of freedom in order to resolve
problems in the epistemology of imputation, because I think that would be
to philosophize in bad faith. (This is not to argue that there are no other
reasons to prefer these reconstructions. I am not convinced that there are, but
I cannot argue that here.)
An “ideal abolitionist” original position

reconstruction
How might a reconstruction that does not incorporate transcendental free-
dom look? We would have to reject retributivism and ius talionis, but also the
underlying notion that criminals could have done otherwise when they com-
mitted their crimes – what I will call the “avoidability of crime assumption.”
At many points in Kant’s texts, his strategy for treating people as ends does
not focus on desert, but instead on treating people as they would rationally
consent to be treated. In his philosophy of right, for example, the justice of
the social contract derives from it being one which would receive the
hypothetical rational consent of all parties to it (TP 8:297). But Kant rejects
Beccaria’s social-contract-based critique of capital punishment (no one could
rationally consent to execution and thus capital punishment cannot be
included in the social contract) and suggests that Beccaria is making a
conceptual error in posing the question of whether criminals can rationally
will punishment, since “it is no punishment if what is done to someone is
what he wills.” However, Kant endorses the view that “saying that I will to be
punished if I murder someone is saying nothing more than that I subject
myself together with everyone else to the laws, which will naturally also be
penal laws if there are any criminals among the people” (MM 6:335). So
Kant does think that the social contract involves my rational consent to a law
that prescribes my execution if I become a murderer. Clearly execution is a
very undesirable outcome, and it is rational for us to avoid undesirable
outcomes when evaluating potential laws for the social contract, and to
give or withhold our consent accordingly. Kant thinks that undesirable
outcomes for commoners of laws granting “hereditary privilege of ruling
rank” are reasons to reject such laws (TP 8:297).8 So why is the undesirable
outcome of harsh punishments for criminals not a reason to reject laws that
prescribe them? The idea seems to be that it is not under my control whether
I am a commoner or a hereditary noble, but it is under my control whether I
become a criminal or not, and I must therefore consent to the social contract
under the assumption that I will avoid committing crimes (an assumption of
what Rawls calls “ideal theory”9). If it is rationally incumbent upon us to
regard our consent to the social contract in this way, then, as Kant puts it,
criminals “cannot possibly have a voice in [this] legislation” (MM 6:335).
But if skepticism about transcendental freedom means that we must reject
the avoidability of crime assumption, then it cannot be rationally incumbent
upon us to make the avoidability of crime assumption when we consent to
the social contract.
What would a reconstructed Kantian social contract look like that
rejected the avoidability of crime assumption? One plausible approach is
by way of a version of original position deliberation in which we assume
that we are just as likely to end up among the punished as we are among
the unpunished.10 If we assume this, we would have a strong preference
for a society that did not punish at all – we would necessarily make
abolition our goal. Our first priority would be to pour resources into
8
Matthew Altman also addresses the limits of Kant’s response to Beccaria – he argues that even if
murderers deserve to die, the fact that we cannot be sure we will not be wrongfully accused of murder
gives us reason to exclude capital punishment from the social contract. See Matthew C. Altman,
“Subjecting Ourselves to Capital Punishment,” in Kant and Applied Ethics: The Uses and Limits of
Kant’s Practical Philosophy (Malden, Mass.: Wiley-Blackwell, 2011), 117–38.
9
John Rawls, A Theory of Justice, rev. ed. (Cambridge: Harvard University Press, 1999), 215–17.
10
Here I draw on a longer discussion of this approach to punishment, in Benjamin Vilhauer,
“Punishment, Persons, and Free Will Skepticism,” Philosophical Studies 62, no. 2 (Jan. 2013): 143–
63. Sharon Dolovich proposes a similar approach, but not in the context of free will skepticism. See
Dolovich, “Legitimate Punishment in Liberal Democracy,” Buffalo Law Review 7, no. 2 (Jan. 2004):
314–29.
636 B. Vilhauer
noncoercive preventative strategies, such as free voluntary therapy for

people at risk of crime, drug treatment, publicly funded education and
job programs in areas where unemployment may be an incentive to
crime, and much greater funding for public services.
But it would be reasonable to be concerned that abolishing all coercion of
criminals before we were able to make broader social changes that diminished
the incentives to commit crimes would effectively throw us back into the
state of nature, leaving the freedom of everyone insecure. If we could devise a
coercive scheme sufficient to maintain a condition of right that places a light
enough burden on those convicted of crime to give them more choiceworthy
lives than they would have in the state of nature, then it would be rational to
consent to it. In determining whether there could be such a scheme, a lot
turns on how we understand “choiceworthy.” If we understand this in
utilitarian terms, then it is easy to imagine that there could be such a scheme.
A life on a carefully titrated opioid drip in a utility-maximizing prison would
surely be more pleasurable than the nasty, brutish, and short life one can
anticipate in the state of nature. But from the Kantian perspective, the most
choiceworthy life is the freest life, and it might sound absurd to suggest that
even the lightest criminal coercion scheme could offer criminals greater
freedom than the state of nature. But it must be kept in mind that the
kind of freedom valuable from the Kantian perspective is not the “wild,
lawless freedom” of the state of nature, but rather lawful freedom (MM
6:316). From the perspective of ethics it is the freedom of acting for the sake
of the moral law, and from the perspective of right it is the freedom of acting
without violating the limits of others’ rightful freedom (which corresponds
roughly with the freedom to act permissibly from the perspective of ethics).
The wider scope of lawless freedom, including the freedom for criminals to
steal or kill in the state of nature, would not make such a life worthy of
choice from the Kantian perspective. (It would be Orwellian to place too
much weight on this point, but it is instructive if applied cautiously.) So it is
not absurd to speculate that a social contract with the right sort of coercive
scheme might afford even convicted criminals more worthwhile freedom
than the state of nature would.
What sorts of coercive constraints would be worth considering if we knew
that we were as likely as not to be constrained by them? Preventative coercion
focusing on crimes in progress would take on a new importance, since it
would seem to be rational from the Kantian perspective (and other perspec-
tives too) to prefer being prevented from completing a criminal act to being
subject to coercive constraints after the fact. But the availability of choice-
worthy strategies for doing this is relative to a society’s development of
relevant technology and social practices.11 If we wanted to rely completely on

preventative strategies, then the options readily at hand today would include
combinations of blanket surveillance and militarized policing, which would
protect us by making the whole world a prison. Even ticklebots would not
seem as benign in the hands of a police state as they do in a thought
experiment.
Are there after-the-fact coercive constraints that it would be rational to
endorse while we work on more choiceworthy technological and social
preventive innovations? I think that for nonviolent thefts, it would be
rational to consent to measures such as fines and ankle-monitors for the
thieves we could catch, and a system of social insurance to compensate
victims of theft when we could not catch the thieves. In the original position,
we would not care so much about protecting our property that we would risk
imprisonment for it.
Violent crimes are another matter. It would be rational to risk imprison-
ment in the right sort of prison in order to be protected from murder or
serious injury. But to pass the test of affording greater lawful freedom than
the state of nature, it would obviously have to be a very different sort of
prison that what we have today, one with real opportunities for social
interaction, voluntary therapy, education, meaningful work, and mainte-
nance of relationships with friends and family in the world outside. In the
original position we would not merely pay lip service to these goals as most of
us do today – we would recognize the radical reform needed to achieve them.
We would only consent to imprisonment while offenders were still risks for
violence and would insist on a continuous parole review process leading to
rapid though probably monitored release into the community.
Would it be rational to consent to after-the-fact coercive measures
designed to deter, or just ones designed to prevent reoffending? We would
not consent to harsh measures for the sake of deterrence, and certainly not
the death penalty. But we could not forgo considerations of deterrence
completely, because if after-the-fact responses to thieves or violent criminals
made criminal life too pleasant, then they would become new incentives to
commit crime, and we would be cast back into the state of nature once again.
So we would need to maintain at least the minimum level of aversiveness
necessary to prevent the system of criminal justice from becoming an
incentive to commit crime. But it seems plausible to think that we could
11
This point is central to the objection developed in Dimitri Landa, “On the Possibility of Kantian
Retributivism,” Utilitas 21, no. 3 (Sept. 2009): 276–96.
638 B. Vilhauer
maintain this minimum in a prison and still give prisoners a life with greater
scope for lawful freedom than the state of nature affords. It is important to
keep in mind that aversiveness is a relative matter – we have an aversion to a
less-pleasant option when a more-pleasant option is available, even if the less-
pleasant option is not intrinsically unpleasant. So the minimum level of
aversiveness would be relative to conditions outside prison, and given the
way the original position prioritizes noncoercive preventative measures such
as education, jobs, and social services, life on the inside of prison would not
have to be intrinsically unpleasant for it to be less pleasant than life on the
outside.
In this way, the original position approach to punishment provides guide-
lines about the kind and quantity of punishment we should impose. Can it
tell us whom we should punish? The retributive principle tells us to punish
the deserving, but if we design an original position to rule out considerations
of desert, we need a different reason to limit punishment to actual criminals
in cases where punishing a scapegoat would improve deterrence. It may be
objected that without an appeal to desert there can be no legitimate reason to
prefer punishing the actual committers of crime in such cases. But I think
that our intuition about the wrongness of scapegoating really has two roots,
one having to do with desert, and another having to do with the deception of
the public required for scapegoating to generate deterrence.
Our lack of knowledge of transcendental freedom pulls up the desert root,
but not the deception root. While Kant does not distinguish them, his
philosophy nonetheless provides reasons for rejecting scapegoating which
are grounded in the deception root. Deception manipulates in a way that
uses rational beings as mere means. In the context of right, Kant propounds
the “transcendental formula of public right”: “‘All actions relating to the rights
of others are wrong if their maxim is incompatible with publicity.’ This
principle is not to be regarded as ethical only (belonging to the doctrine of
virtue) but also as juridical (bearing upon the right of human beings),” and
rules out maxims “that I cannot divulge without thereby defeating my own
purpose” (PP 8:381). A publicized principle of punishing scapegoats to deter
crime would be self-defeating because it would actually undermine deter-
rence, since potential criminals are deterred by the worry that they will be
punished for their crimes, not that scapegoats may be punished in their place.
Further, when contemplated from the standpoint of original deliberation
about punishment, a principle that aims to deter by scapegoating is self-
deceiving and in a sense self-contradictory. Since I may well be among the
deceived, I would be volunteering to be deceived about a fundamental
principle of society which I had chosen myself. Self-deception on this scale
undermines one’s status as a rational agent in a way that is similar to

consenting to slavery and should be recognized as self-contradictory for
similar reasons.
It may be objected that the structure of the original position I rely on in
this approach is arbitrary in assuming an equal probability of finding oneself
among the punished and among the unpunished. Certainly it is not as tidy as
the notion of original position deliberation that Rawls applies to distributive
justice. In that context we can apply “maximin” reasoning because there is a
unique candidate for the worst-off position, the poorest.12 In the context of
criminal justice, however, there are two candidates for the worst-off position,
criminals and those they threaten. The purpose of original position delibera-
tion is to proceduralize Kantian fairness, and it seems to me that the fair
thing to do in dealing with a socially necessary practice that benefits some at
the expense of others is to assume equal probability of being on both sides.
It may also be objected that a system of criminal justice that dispenses with
desert cannot be called a system of punishment at all, because punishment
involves desert as a conceptual matter.13 I am not sure that this is right, but if
the dominant usage of “punishment” takes it to imply blame, then I would in
one sense be happy to go along with this, and call this an account of criminal
justice but not punishment, but I fear that there is a hazard of euphemism
here which is more pressing than the hazard of violating norms of usage.
Referring to a practice as punishment highlights that it cries out for justifica-
tion, but calling it criminal justice does not, and philosophers who take
skepticism about imputing blame seriously ought to keep the demand for
justification clearly in view.
It is therefore important to be clear what this approach can and cannot do
as far as justifying punishment goes. It is an approach that weighs conse-
quentialist considerations but makes them morally significant only insofar as
they derive from the rational consent of each party to punishment, and this
means that the consequentialist considerations derive from a non-consequen-
tialist foundation in respect for persons. It differs from the way that Kant
draws consequentialist considerations out of respect for persons, in that it
shows how strict guidelines for whom we should punish, and the kind and
quantity of punishment we should impose, can all rise directly from this
foundation instead of being partially derived from an interjected retributive
12
Rawls, Theory of Justice, 130–38.
See A. M. Quinton, “On Punishment,” Analysis 14, no. 6 (June 1954): 133–42; and more recently,
13
Wood, Kantian Ethics, 208.

640 B. Vilhauer
principle. It also differs from Kant’s approach in that it treats criminals as

ends by respecting their rational consent rather than by imputing blame and
punishing retributively.
But it might reasonably be objected that this can only be a partial
justification of punishment, because nothing but desert could alleviate the
basic unfairness of the fact that some are punished while others are not. I
think this is correct, and it highlights the importance of the fact that the
original position deliberators in this story would set the abolition of punish-
ment as their top priority, and would endorse after-the-fact coercive mea-
sures only as temporary solutions to maintain the coercive protection of right
en route to abolition. I think that this is in the spirit of Kant’s best thought
on punishment, with which we began: that “the more legislation and
government agree with this idea [of a perfect state], the less frequent punish-
ment will become,” and that we must strive “to bring the legislative con-
stitution of human beings ever nearer to [this] possible greatest perfection”
(A316–17/B373–74).
Part X
Anthropology, History, and Education
28
Denkungsart in Kant’s Anthropology
from a Pragmatic Point of View
Patrick R. Frierson
Anthropology from a Pragmatic Point of View, published in 1798 as one of the

last works Kant published in his lifetime, was the culmination of over
twenty-five years of lecture courses on the topic of “anthropology.” From
his first lectures through his published Anthropology, Kant insists on the
uniqueness of his anthropology’s “pragmatic” emphasis. In one early lecture,
he emphasizes that in his course, “human beings are not studied in spec-
ulative terms, but [in] pragmatic [ones]” (LAn 25:470); and in his published
work, he contrasts his own “knowledge of the human being, systematically
formulated (anthropology) in a pragmatic point of view” from “physiologi-
cal” ones (An 7:119). An important letter to his former student Marcus
Herz, from the time that he first began developing his pragmatic anthropol-
ogy, gives a sense of Kant’s goals:
This winter I am giving, for the second time, a lecture course on Anthropologie,
a subject that I now intend to make into a proper academic discipline. But my
plan is quite unique. I intend to use it to disclose the sources of all the sciences,
the science of morality, of skill, of human intercourse, of the way to educate
and govern human beings, and thus of everything that pertains to the practical.
(C 10:145)
P. R. Frierson (*)
Department of Philosophy, Whitman College, Walla Walla, USA
e-mail: frierspr@whitman.edu

644 P. R. Frierson
Later, in his personal notes, Kant reiterates this emphasis: “the historical kind
of teaching is pragmatic, when it . . . is not merely for the school, but also for
the world or ethics” (Ak 16:804 [R3376]; cf. LAn 25:xv). As a course focused
on human beings, Kant’s Anthropology draws heavily from his earlier (and
continuing) lectures on empirical psychology, in which he discusses various
faculties of the human soul. It also relates to Kant’s writings on human
difference and human history, from his Observations on the Feeling of the
Beautiful and Sublime in 1762 through his essays on history and race in the
1770s and 80s, Religion within the Boundaries of Mere Reason (1793), and
Conflict of the Faculties (1798). But from all of these observations about
human beings, Kant narrows his focus to those that can be applied “for the
world’s use,” and particularly for “the investigation of what he [the human
being] as a free-acting being makes of himself, or can and should make of
himself” (An 7:119).
Despite its importance to Kant, his pragmatic anthropology has only
recently become an object of sustained study, and there is still no clear
consensus about how to interpret his all-important claim that anthropology
is pragmatic, particularly insofar as this involves human beings “as free-acting
being[s]” (An 7:119). The relationship between this claim and the clearly
empirical nature of anthropology is a particular challenge. While I have
discussed these and related issues elsewhere, here I focus on laying out
Kant’s general sense of “pragmatic” and then his specific treatment of
humans’ “mode of thought,” or Denkungsart, a concept central to the
empirical investigation of human freedom. In the end, a discussion of this
concept provides answers to the questions of how anthropology studies
human beings “as free” and, more generally, what Kant’s pragmatic anthro-
pology is.
The many senses of “pragmatic” and the

importance of freedom
Kant is famous for being “technical without being precise.”1 In the case of his
notion of “pragmatic” anthropology, we can give some precision to his
imprecise technical lingo by defining five distinct senses of “pragmatic” in
Kant’s works. Each of these senses reflects a way in which Kant sees
1
Lewis White Beck, A Commentary on Kant’s “Critique of Practical Reason” (Chicago: University of
Chicago Press, 1960), 78.
28 Denkungsart in Kant’s Anthropology 645
“pragmatic anthropology” as dealing with “what the human being makes,

can, or should make of himself as a freely acting being,” but in different ways
(An 7:119). “Pragmatic” can mean (1) non-scholastic, (2) non-physiological,
(3) conducive to happiness, (4) relevant to manipulating other people, and/
or (5) relating to any practical concern(s). Inspired by Kant’s frequent
distinction between pure moral philosophy and his moral anthropology, one
might also see “pragmatic anthropology” as the promised moral anthropology
of the Groundwork (G 4:388) and Metaphysics of Morals (MM 6:217–18);
but Kant nowhere identifies pragmatic with moral anthropology.
Nonetheless, as Robert Louden puts it, “Moral anthropology is already
potentially present within pragmatic anthropology”2; there is room for
moral anthropology as a subset of several of Kant’s senses of pragmatic.3
Thus, after briefly explaining each sense of pragmatic and showing how it
relates to anthropology, I argue that we should privilege the last and most
general sense, that of relating to whatever is practical, and I show how this
includes – but is not limited to – moral anthropology. Throughout, I
emphasize how each sense involves taking the human being as a free being
as its object.
Pragmatic vs. scholastic
Kant often distinguishes his pragmatic anthropology from what he calls

“scholastic” (LAn 25:855) or “speculative” (LAn 25:855) knowledge.4 In
one lecture, he colorfully describes scholastic anthropology as a “brooding
science . . . for the school, but . . . not for common life” (LAn 25:853).
2
Robert B. Louden, Kant’s Human Being: Essays on His Theory of Human Nature (Oxford: Oxford
3
There are other senses of “pragmatic” that one might add to my taxonomy, such as Thomas Sturm’s
conception of pragmatic as “empirical knowledge of the conditions that hinder or encourage the shaping
of our reason such that we attain the mode of thought of a cosmopolitan” (Kant und die Wissenschaften
vom Menschen [Paderborn: Mentis, 2009], 523). I agree with Sturm that this is an important pragmatic
goal in which Kant is interested, but reject the view that it exhausts the sense in which pragmatic
anthropology is pragmatic. The cultivation of memory, for instance, need not be in the service of a
greater cosmopolitan mode of thought in order to legitimately be part of pragmatic anthropology. In the
end, I treat such narrower proposals for the scope of pragmatic the way I treat moral anthropology, as a
legitimate subset of pragmatic anthropology as a whole.
4
This sense of “pragmatic anthropology” is particularly emphasized by Holly Wilson and John
Zammito, both of whom rightly relate it to Weltweisheit (worldly-wisdom) and a sympathy between
Kant’s early anthropology and the budding Popularphilosophie movement in eighteenth-century
Germany. See Holly L. Wilson, Kant’s Pragmatic Anthropology: Its Origin, Meaning, and Critical
Significance (Albany: SUNY Press, 2006); and John H. Zammito, Kant, Herder, and the Birth of
Anthropology (Chicago: University of Chicago Press, 2002).
646 P. R. Frierson
Scholastic sciences are not “popular” in that they cannot “be grasped by
common people,” and “he who makes a scholastic use of his knowledge is a
pedant” (LAn 25:853; see also 1209). Scholastic anthropology seeks for
completeness and systematicity in its rules, while a pragmatic one emphasizes
popularity and “gives no other explanations of the rules . . . except those that
can be observed by everyone” (LAn 25:854). Here “pragmatic” seems pri-
marily to refer to the popularity and accessibility of Kant’s anthropology, in
opposition to the esoteric pedantry Kant associates with “scholastic” anthro-
pologies that are merely “for the school” (e.g., Ak 16:804 [R3376]). His
interest in this pragmatic discipline was part of his more general concern with
Weltweisheit, the worldly wisdom that students would need in order to
succeed in the world. From “the beginning of [his] academic career,” he
offered students a course in Physical Geography that among other things
“considers the human being” and that aims to “make good [students’] lack
of experience” in the concrete matters of life (APL 2:312).
Pragmatic vs. physiological
In these early contrasts between scholastic and pragmatic anthropology, Kant

also hints at a distinction that he emphasizes more later, that between
pragmatic and physiological anthropology.5 In his early lectures, he identifies
the anthropology of Ernst Platner as “scholastic anthropology” (LAn 25:856)
and associates this with the knowledge of the human being treated in
“physiology” (LAn 25:855). In his published Anthropology nearly two decades
later, he returns to this contrast between anthropology from “a physiological
or a pragmatic point of view,” where
Physiological knowledge of the human being concerns the investigation of

what nature makes of the human being; pragmatic, the investigation of what he
as a free-acting being makes of himself, or can and should make of himself. –
He who ponders natural phenomena, for example, what the causes of the
faculty of memory may rest on, can speculate back and forth (like Descartes)
over the traces of impressions remaining in the brain, but in doing so he must
admit that in this play of his representations he is a mere observer and must let
nature run its course, for he does not know the cranial nerves and fibers, nor
5
For an excellent discussion of Kant’s anthropology relative to that of Platner and other physiological
anthropologists, see Sturm, Kant und die Wissenschaften vom Menschen.
does he understand how to put them to use for his purposes. Therefore all
theoretical speculation about this is a pure waste of time. (An 7:119)
Platner’s Anthropology had aimed to “study the body and mind together in
terms of their mutual proportions, limitations, and relations.”6 For Platner,
this offered hope of a medical science of mind, one that would “explain
given mental phenomena based on a theory of physiological prerequisites
for mental phenomena.”7 Kant’s Anthropology, by contrast, rejects the
methodological materialism implicit in these approaches in favor of an
approach to human agency that emphasizes “us[ing] perceptions concern-
ing what has been found to hinder or stimulate memory in order to enlarge
it or make it agile” (An 7:119). While physiological anthropology looks at
human beings as mere machines subject to natural influence, pragmatic
anthropology sees them as free beings who use observation and experience
for self-improvement.
In distinguishing pragmatic anthropology from scholastic and physiologi-
cal anthropologies, Kant focuses on what his anthropology is not. But in both
cases, the key to the contrast lies in the fact that pragmatic anthropology puts
knowledge to use. Kant does not object to scholasticism for bringing human
behavior and cognition under rules, but for bringing it under rules that are
irrelevant to practical purposes (LAn 25:856). And he does not even object to
physiological anthropology’s methodological materialism in itself, but only
to the “waste of time” involved in inquiries into physiological causes that one
“does not understand how to put to use” (An 7:119). And this raises the
question: what is pragmatic anthropology put to use for? And here there are
at least three possible responses: for happiness, for influencing others, or for
any practical concern whatsoever.
Pragmatic anthropology, prudence, and happiness
In his Groundwork, Kant distinguishes between pragmatic, technical, and

moral imperatives: “The first kind of imperatives might also be called
technical (belonging to art), the second kind pragmatic (belonging to wel-
fare), the third kind moral (belonging to free conduct as such, i.e., to
morals)” (G 4:416–17). In his anthropology lectures, Kant makes similar
distinctions between “skill” and “prudence” (e.g. LAn 24:1210) and between
6
Quoted in Sturm, Kant und die Wissenschaften vom Menschen, 76.
7
Ibid., 77.
648 P. R. Frierson
what is “pragmatic” because related to “prudence” and what is properly

“moral” (e.g., LAn 25:855–56). Moreover, Kant’s Anthropology and related
lectures are full of information conducive to living a happier life. While the
Groundwork expresses despair over the possibility of “imperatives of pru-
dence” ever “presenting actions . . . as practically necessary” because “the
concept of happiness is such an indeterminate concept” (G 4:418, cf.
CPrR 5:36), the reason for this indeterminacy is that “all of the elements
that belong to happiness are without exception empirical, that is, they must
be borrowed from experience” (G 4:418). In the Groundwork, Kant’s focus is
justifying the moral law’s categorical demands, so he emphasizes the fragility
of happiness, insisting that “it is impossible for the most insightful . . . finite
being to frame for himself a determinate concept of what he really wills here”
and asking seemingly rhetorical questions such as “If he wills a long life, who
will guarantee him that it would not be a long misery?” (G 4:418). But in his
Anthropology and related lectures, Kant not only lays out general discussions
of pleasure and pain, noting for instance that “pain must always precede
every enjoyment” and “no enjoyment can immediately follow another,” but
also gives specific advice about how best to promote lasting happiness (An
7:231). Thus while the Groundwork despairingly asks, “If he wills a great deal
of cognition and insight, that might become only an eye all the more acute to
show him . . . ills that are now concealed” (G 4:418), the Anthropology care-
fully distinguishes cognitive powers and discusses the roles of various of these
for promoting happiness. For example, after distinguishing “attention” from
“abstraction,” Kant notes,
Many human beings are unhappy because they cannot abstract. The suitor
could make a good marriage if only he could overlook a wart on his beloved’s
face, or a gap between her teeth. . . . But this faculty of abstraction is a strength
of mind that can only be acquired through practice. (An 7:131–32)
And after making a general (cognitive) point about the subjective experience
of time passing, Kant puts this to practical use for promoting happiness:
the multitude of stages that mark the last part of life with various and different
tasks will arouse in an old person the illusion of a longer-travelled lifetime than
he would have believed according to the number of years, and filling our time
by means of methodical, progressive occupations that lead to an important and
intended end . . . is the only sure means of becoming happy with one’s life and,
at the same time, weary of life. . . . Hence the conclusion of such a life occurs
with contentment. (An 7:234)
Kant ends the first major part of his Anthropology with specific accounts of
“the highest physical good” and “the highest moral-physical good,” both of
which, despite their use of the terms “good” and “moral,” are suggestions for
how best to be happy given our human nature (An 7:276–77).8 Thus even as
Thomas Sturm rightly points out that “Kant does not explicitly state what
happiness might be and just how to achieve it,” he wrongly infers from this
that “it is most doubtful that Kant intended the practical purpose of
anthropology to be the teaching of general ‘personal prudence.’”9 Against
the Groundwork’s apparent hopelessness about rules of prudence, Kant’s
anthropology provides practical – albeit empirical and limited – advice for
living as happily as possible.
Importantly, and despite Kant’s apparent contrast between “pragmatic”
imperatives “belonging to welfare” and “moral” ones “belonging to free con-
duct” (G 4:416–17), both the need for and possibility of a pragmatic anthro-
pology to help discern prudential imperatives are due to humans’ freedom. In
his “Conjectural Beginning of Human History,” Kant considers the emergence
of human beings from a state of mere animality into true humanity. The first
and most important step in this emergence was the transition from a condition
in which “Instinct, that voice of God which all animals obey, . . . guided the
novice” (CB 8:111) to the emergence of “reason” and the “discover[y] in
himself [of] a faculty of choosing for himself a way of living and not being
bound to a single one, as other animals are” (CB 8:112). But this discovery was
not unambiguously good from the standpoint of happiness. Echoing a point
from the Groundwork (see G 4:395–96), Kant continues:
Yet upon the momentary delight that this marked superiority might have
awakened in him, anxiety and fright must have followed right away, concern-
ing how he . . . should deal with this newly discovered faculty. He stood, as it
were, on the brink of an abyss; for instead of the single objects of his desire to
which instinct had up to now directed him, there opened up an infinity of
them, and he did not know how to relate to the choice between them; and
from this estate of freedom, once he had tasted it, it was nevertheless impossible
for him to turn back again to that of servitude [to] instinct. (CB 8:112)
Pragmatic anthropology, as “the investigation of what [the human being] as a

free-acting being makes of himself, or can and should make of himself” (An
8
I discuss both of these in more detail in Patrick R. Frierson, What Is the Human Being? (Abingdon:
Routledge, 2013), ch. 5.
9
Sturm, Kant und die Wissenschaften vom Menschen, 501.
650 P. R. Frierson
7:119), directly addresses the problem posed by freedom for lasting happiness.
No longer bound by instinct, we must figure out how to live life well on our
own. A pragmatic anthropology that gives guidance about what constitutes true
happiness and how best to achieve it can help us to pursue this natural end.
Pragmatic anthropology and the use of others
While much of pragmatic anthropology specifies means toward happiness,

Kant often invokes a narrower sense of “pragmatic.” In a footnote to a claim
that “skill in the choice of means to one’s own greatest well-being can be
called prudence,” which he identifies with what is “pragmatic,” Kant notes,
The word “prudence [Klugheit]” is taken in two senses: in the one it may bear
the name of “knowledge of the world [Weltklugheit]”; in the other that of
“private prudence.” The first is a human being’s skill in influencing others so as
to use them for his own purposes. The second is the insight to unite all these
purposes to his own enduring advantage. (G 4:416n)
The second sense of prudence is just that emphasis on promoting one’s own
happiness on which I focused in the previous section. But the first introduces
a new emphasis on influencing others. Likewise, in his lectures, Kant explains
that “every pragmatic instruction makes one prudent,” where “prudence is a
proficiency or knowledge in . . . using other human beings for one’s aims”
(LAn 25:1210) and that the “practical knowledge of the human being [that]
makes us prudent . . . is a knowledge of . . . how one human being has influ-
ence on another and can lead him according to his purpose” (LAn 25:855;
see also LAn 25:471–72, 1436; LE 27:358). In his published Anthropology,
he writes,
In an anthropology from a pragmatic point of view, however, the only thing

that matters to us is to present the character of both [kinds of people], as they
are now, in some examples, and, as far as possible, systematically; which makes
it possible to judge what each can expect from the other and how each could
use the other to his own advantage. (An 7:312)
This repeated reference to “his own advantage” (G 4:416n; An 7:312) might

make it seem like prudence in the use of others is solely egoistic.10 But Kant
10
Cf. Sturm, Kant und die Wissenschaften vom Menschen, ch. 8, §8.
makes clear elsewhere that even insofar as pragmatic anthropology is used to

manipulate others, this manipulation need not be solely for selfish purposes.
Precisely in the context of pointing out how anthropology helps one use others,
Kant insists that “morality must be combined with knowledge of humanity” (LAn
25:472, emphasis added); and among “great uses” of anthropology, he explicitly
indicates a use of others for moral ends, an “influence on morality and religion
whereby one can give these duties the power of incentives through this knowl-
edge” (LAn 25:1437). Pragmatic knowledge provides one with means of manip-
ulating others; this knowledge can be put to use for one’s own personal advantage,
but it can also be put to use for moral (or other) ends.
Pragmatic as practical
In the end, however, this influence over others is not and cannot be the
primary use of anthropology. As Kant says in the Groundwork about his two
senses of prudence, “The latter [prudence in promoting one’s overall well-
being] is properly that to which the worth even of the former [skill in
influencing others] is reduced” (G 4:416n). More generally, knowledge of
the ends worthy of promoting must take precedence over the use of others to
promote those ends. Given humans’ social nature and our dependence upon
others, a pragmatic anthropology that teaches how to use others effectively
will be an important part of promoting any ends. But – and here I turn to the
final and, I think, best conception of “pragmatic” anthropology – this
instruction will be only part of a broader study of everything about human
nature that can help us discern the most important ends to pursue and
effectively perfect and use ourselves and others toward that pursuit. Or, as
Kant puts it in that early letter to Herz, pragmatic anthropology will “dis-
close the sources of all the sciences, the science of morality, of skill, of human
intercourse, of the way to educate and govern human beings, and thus of
everything that pertains to the practical” (C 10:145, emphasis added). All
knowledge of human beings that can be put to any sort of practical use
can fit within pragmatic anthropology. Thus one finds guidance about how
to perfect various human faculties like the senses (An 7:165) or memory (An
7:183–84), how to become happy (e.g., An 7:296), how to cultivate that
character that is a precondition of a good will (e.g., An 7:294), and how to
influence others (e.g., An 7:312).
Moreover, as the letter to Herz makes clear, the realm of the practical here
includes, and even emphasizes, human social life, our social “intercourse” and
mutual “molding and governing.” And Kant’s published Anthropology
652 P. R. Frierson
concludes with a strong emphasis on the social dimension of his pragmatic

anthropology. The final section of the Anthropology – “The Character of the
Species” – ends with call to “the human species” to “rise out of evil in
constant progress towards the good . . . by a progressive organization of
citizens . . . into and toward the species as a system that is cosmopolitically
united” (An 7:333). In that sense, one might rightly see the pragmatic
orientation of the anthropology as toward social self-improvement.11
In the end, then, Kant’s pragmatic anthropology serves to present in popular
and accessible ways the knowledge of human nature that will be needed by
individuals and societies in order to achieve together all of the goals that make
up both the perfection and happiness of the species. Because merely scholastic
or physiological knowledge cannot serve these ends, he discusses physiological
claims and the findings of speculative psychology only insofar as these have
practical relevance. Empirical psychology, in that sense, “establishes its own
domicile in a complete anthropology” (A849/B877). Moreover, because the
determination of what will make humans happy is – by virtue of our freedom –
a difficult problem requiring extensive reflection on empirical claims about
human psychology, pragmatic anthropology has an important emphasis on
happiness. And because advancing our ends – particularly our social and
cosmopolitan ends – requires cooperation with other free human beings,
pragmatic anthropology emphasizes knowledge that we can put to use in social
intercourse. But ultimately, because our vocation goes beyond merely happiness
or even social cooperation, pragmatic anthropology truly treats all that “pertains
to the practical” (C 10:146).
Empirical anthropology and the importance

of Denkungsart
Kant’s anthropology is pragmatic in several distinct senses united under
the banner of putting knowledge of human beings to use. In every sense,
the notion of a “pragmatic” anthropology involves human beings as free
11
For discussion, see Sturm, Kant und die Wissenschaften vom Menschen, esp. chs. 7 and 8. Sturm also
very nicely discusses an important passage from the Groundwork that defines “the proper meaning of the
word pragmatic” in terms of “provision for the general welfare” (G 4:417n). This emphasis on “welfare
[Wohlfahrt]” is consistent with the readings of pragmatic in terms of happiness, but the footnote here
clearly broadens the relevant happiness from one’s own to that of society as a whole. Ultimately, though,
it needs to be broadened even further, to include anything of practical relevance for happiness or any
other good, for individuals or for society as a whole.
beings who exercise control over their own destinies. But strikingly, the
knowledge that Kant seeks to put to use is empirical, knowledge gained
through “observation and experience” (LAn 25:7).12 The whole “purpose
of anthropology” is “to observe the human being . . . and to organize
human phenomena under rules” (LAn 25:472; see also 856). When he
remarks on the challenges to anthropology, he focuses on challenges to
observing oneself (An 7:121; cf. Ak 15:660 [R1482]). And when he
mentions anthropology elsewhere in his critical works – such as the
“moral anthropology” that will complement his pure moral philosophy
(G 4:388; MM 6:216–17) or the “anthropology” discussed in the first
Critique, into which empirical psychology will eventually be absorbed
(A484–89/B876–77) – it is consistently treated as something empirical.
But the appeal to freedom seems inconsistent with any notion of anthro-
pology as an empirical science that would treat human beings as objects
of experience, given Kant’s commitment to strict causal determinism
with respect to experience (A189/B232–34, A536/B564) and his com-
mitment to a conception of transcendental freedom that precludes such
empirical determination (see A448/B476, A533–35/B562–64, A803/
B831; CPrR 5:97). To make matters worse, Kant explicitly states in
his Metaphysical Foundations of Natural Science that there can be no
empirical science of humans’ mental states. And in his Critique of Pure
Reason, he seems to preclude the possibility of any science of empirical
psychology by rejecting the application of the category of substance to
the human soul.13 Despite these problems, however, Kant’s pragmatic
anthropology is best understood as a kind of empirical science.
Elsewhere, I have discussed and offered solutions for several problems that
might arise with Kant’s integration of an empirical science of human beings
and a commitment to transcendental freedom. In Freedom and Anthropology
in Kant’s Moral Philosophy,14 I addressed the problem of how empirical
influences on moral development could be taken seriously within pragmatic
anthropology given that human beings are always free to act from respect for
the moral law; there I argued that moral anthropology is an important part of
12
This is from his very first lecture on anthropology, but the same point is reiterated in his published
Anthropology (An 7:119, 321) and in the constant reference to experience in support of general points in
both later lectures (e.g., LAn 25:1416) and his published Anthropology (e.g., An 7:324).
13
See Kenneth Westphal, Kant’s Transcendental Proof of Realism (Cambridge: Cambridge University
Press, 2004); and Corey W. Dyck, Kant and Rational Psychology (Oxford: Oxford University Press,
2014).
14
Patrick R. Frierson, Freedom and Anthropology in Kant’s Moral Philosophy (Cambridge: Cambridge
654 P. R. Frierson
combatting one’s self-wrought radical evil and thereby expressing one’s free
commitment to good principles. In Kant’s Empirical Psychology,15 I showed
in more general terms how, for Kant, empirical investigation of human
beings is compatible with transcendental freedom by virtue of his transcen-
dental idealism, according to which the empirical character of every human
action can be investigated “as with any investigation in the series of deter-
mining causes for a given natural effect” (A554/B582), but this empirical
character is itself grounded in an intelligible character for which individuals
can be held responsible. In that same book, I showed that Kant’s claim that
“empirical [psychology] can never become a science” (MFS 4:471) is based
on a very narrow and technical notion of “strict science,” and that systematic
empirical investigations of human beings that would be recognizably “scien-
tific” by today’s standards are possible for Kant.16 I there also discussed
several other particular problems that recent scholars have raised for Kant’s
prospects of empirically investigating human beings.17
In this short chapter, I want to focus on how an empirical anthropology
can study human beings as freely acting beings. With respect to this issue, I
focus on one particular concept within Kant’s account of anthropology: the
notion of character as “mode of thought [Denkungsart].” Kant’s published
Anthropology18 starts with a discussion of various faculties of soul – cognition,
feeling, and desire – that tracks closely the empirical psychology sections of
Baumgarten’s Metaphysica, which Kant had used as a textbook for his own
metaphysics and empirical psychology lectures. But the last third of the
Anthropology shifts the focus of anthropology in a new direction, to what
Kant calls “character.” Some of this discussion draws from Kant’s earlier
work on various different human characteristics, such as different tempera-
ments, “the character of the sexes,” and “the character of the nations”
(compare, e.g., An 7:286–91 and 303–21 with OBS 2:218–55). But argu-
ably the most important and innovative part of the Anthropology is Kant’s
discussion of what he calls “character simply speaking [Charakter
schlechthin]” (An 7:292). I have discussed the importance of Charakter
schlechthin in other contexts, but I have not emphasized a point that Kant
15
Patrick R. Frierson, Kant’s Empirical Psychology (Cambridge: Cambridge University Press, 2014), esp.
9–18.
16
Ibid., 18–27. See also Frierson, What Is the Human Being? 46–49.
17
Frierson, Kant’s Empirical Psychology, 27–43.
18
For a nice discussion of changes in the anthropology lectures leading up to this structure in the
published Anthropology, see Sturm, Kant und die Wissenschaften vom Menschen.
– and several commentators on his anthropology19 – emphasize: the strong

connection between Charakter schlechthin and what Kant calls Denkungsart.
Kant introduces the term Denkungsart explicitly in anthropology lectures in
the mid-1770s (see, e.g., LAn 25:649, 821), and in his published
Anthropology he defines it as equivalent to Charakter schlechthin, which he
identifies with that character which “is the distinguishing mark of the human
being as a rational being endowed with freedom” and which “indicates what
he is prepared to make of himself” (An 7:285). These important echoes of
Kant’s initial definition of pragmatic anthropology as “the investigation of
what he [the human being] as free-acting being makes of himself” (An 7:119)
strongly indicate the centrality of Denkungsart for Kant’s eventual pragmatic
anthropology.20
But the concept of Denkungsart, and particularly the way in which it is
both indicative of human freedom and a proper object of anthropological
investigation, is subject to some difficulties. Denkungsart is typically trans-
lated as “mode of thought” or “way of thinking,” and G. Felicitas Munzel,
who has offered a particularly detailed defense of her translation,21 translates
it as “conduct of thought.” None of these turns of phrase are particularly
illuminating or even clear, and all of them overemphasize the cognitive
dimension of Denkungsart at the expense of the volitional.22 For this chapter,
I leave the term untranslated and focus on laying out several key features of
Kant’s definition of it. I also make the case that there are two very different
19
See especially Sturm, Kant und die Wissenschaften vom Menschen; and G. Felicitas Munzel, Kant’s
Conception of Moral Character: The “Critical” Link of Morality, Anthropology, and Reflective Judgment
(Chicago: University of Chicago Press, 1999).
20
See too LAn 25:1367, where Kant claims: “As the first part of the anthropology contains the
physiology of the human being and thus, as it were, the elements out of which the human being is
composed, so the practical part of anthropology is the one that teaches us how human beings are
constituted in their voluntary actions.” The implication here is that anthropology only really becomes
pragmatic as opposed to physiological when we get to its second part, which deals with character. This
does not quite correspond to the actual content of the anthropology (even in these lectures), but it does
show how central Kant considers character as Denkungsart within his pragmatic anthropology as whole.
21
Munzel, Kant’s Conception of Moral Character, xv–xvii.
22
Kant often uses Denkungsart in a way that primarily refers to one’s way of thinking in theoretical
rather than practical contexts (see, e.g., Axi note; C 10:269). In these contexts, the same distinctions I
will raise below in the specifically practical contexts arise. Thus Kant’s reference to “our age’s way of
thinking” (Axi note) involves a specifically historical conception of Denkungsart, and also one embody-
ing a sort of responsibility-holding, in that Kant purports in the rest of that note to justify this way of
thinking. But his claim to Herz that “one cannot expect a Denkungsart to be suddenly led off the beaten
track into one that has heretofore been totally unused” and that thus one must take “time” in order “to
stay that Denkungsart little by little in its previous path and finally to turn it into the opposite direction
by means of gradual impressions” treats human Denkungsart as an orientation of reason, to be sure, but
one that susceptible of empirical study, prediction, and even control through empirical influences
(C10:269, translation modified).
656 P. R. Frierson
technical senses in which Kant uses the term: (1) as equivalent to “intelligi-
ble” as opposed to “empirical” character; and (2) as describing the empirical
character of the higher faculty of desire insofar as this is governed by
principles, as opposed to mere temperament or inclination. Getting clear
on the distinction and relations between these two senses is essential for
properly understanding the nature of Kant’s pragmatic anthropology.
I start, though, with some general characteristics of the anthropological
concept of Denkungsart. Kant uses the term Denkungsart from his earliest
published writings (UNH 1:235) and in a wide variety of contexts, but the
uses most directly relevant to his Anthropology all tie the concept to one’s
“character” and share four other key elements:
1) Denkungsart (mode of thought) is contrasted with Sinnesart, or “mode of

sense.”
2) Denkungsart is linked to human freedom.
3) Relatedly, it is something that human beings construct for ourselves; it is
not given simply by nature.
4) It involves commitment to principles (and is thereby a characteristic of
humans as rational).
The first three of these are already present in the quotations from the
Anthropology given above (An 7:285; see also A551/B597; Rel 6:47; An
7:285, 292; Ak 15:396, 763, 865–66, 870; LAn 25:821). Kant makes the
fourth explicit when he defines “character as Denkungsart” as “that property
of the will by which the subject binds himself to definite practical principles
that he has prescribed to himself irrevocably by his own reason” (An 7:292;
see also LAn 25:438, 630, 651–52, 1175, 1384, 1386).
Once we see Denkungsart as having these four features, we might wonder
how it could be an object of investigation within an empirical anthropology.
To highlight this problem, but also to see how to make sense of an empirical
Denkungsart, we need to distinguish between two related senses of “freedom”
in Kant – empirical and transcendental23 – and two corresponding senses of
Denkungsart.
First, Kant explicitly uses the distinction between Sinnesart and
Denkungsart in the Critique of Pure Reason and Religion to distinguish
between what he calls humans’ “intelligible character” and their “sensible”
23
At the end of this section, I bring up a third sense of freedom – “practical” – and discuss how it relates
to both empirical and transcendental freedom.
or “empirical character.” In the Critique of Pure Reason, Kant emphasizes

that “empirical character (in the Sinnesart24) . . . is all precisely determined
and necessary” but this “empirical character is once again determined in the
intelligible character (in the Denkungsart). We are not acquainted with the
latter [Denkungsart] but it is indicated through appearances, which really
give only the Sinnesart (the empirical character) for immediate cognition”
(A551/B579, translation modified). Similarly, in Religion, Kant identifies a
“revolution in Denkungsart” with that “revolution in disposition
[Gesinnung]” whereby a human being becomes “virtuous according to the
intelligible character (virtus noumenon).” This noumenal virtue is evident
only “for him who penetrates to the intelligible ground of the heart (the
ground of all maxims of the power of choice), . . . i.e. for God.” By contrast,
“for the judgment of human beings, however, who can assess themselves
and the strength of their maxims only by the upper hand they gain over the
senses in time, the change is” evident only as a “gradual reformation in the
Sinnesart” that can “be regarded as a gradual reformation of the . . . perverted
Denkungsart” (Rel 6:47–48, translation and order modified).
In these and similar contexts, the distinction between Denkungsart and
Sinnesart is an application to human character of Kant’s metaphysical distinc-
tion between appearances (phenomena) and things in themselves (nou-
mena).25 As one’s “noumenal” character, Denkungsart is beyond the
possibility of human experience, something to which only God has access.
What human beings experience is Sinnesart; we at best imperfectly infer claims
about Denkungsart from our empirical character. And this Denkungsart can
and should be seen as transcendentally free (A551/B579). Thus for
Denkungsart in this sense, “the formation of a Denkungsart . . . is . . . the act of
an unmoved mover.”26 However, just as appearances are appearances of things
in themselves, in these passages, there is also a grounding relationship between
24
Here Sinnesart and Denkungsart are not the German terms translated as “empirical” and “intelligible”
character. Rather, Kant is identifying the German terms Sinnesart and Denkungsart with empirische and
intelligibelen Charackter respectively.
25
For a more general discussion of this distinction, see Chapters 4–6 in this volume.
26
Sturm, Kant und die Wissenschaften vom Menschen, 480. Here I quote from Sturm, but only to
disagree partly with him. Sturm rightly agrees with Emil L. Fackenheim (“Kant’s Concept of History,”
Kant-Studien 48, nos. 1–4 [Jan. 1957], 381–98) that, in his properly anthropological and historical
works, the formation of Denkungsart is “not the act of an unmoved mover” (emphasis added). My point
here is that there is another usage of Denkungsart according to which human Denkungsart – as the term
referring to our intelligible character – must be seen as the activity of an unmoved mover, that is, of a
transcendentally free cause. My overall purpose is to bring out a distinction between different sorts of
Denkungsart, a distinction Sturm did not make.
658 P. R. Frierson
Denkungsart and Sinnesart.27 That is, one’s Sinnesart is “determined by” one’s
Denkungsart, such that we can rightly (albeit fallibly) interpret moral reform in
one’s empirical character as indicative of a revolution in one’s intelligible
Denkungsart.
These two features of this concept of Denkungsart – that it is beyond the
possibility of experience and that it grounds Sinnesart – make it particularly
well-suited to describe humans’ transcendental freedom. Kant emphasizes in
the Critique of Pure Reason that “the thoroughgoing connection of all
occurrences in the world of sense according to invariable natural laws is
already confirmed . . . and will suffer no violation” so that “the only question
is whether, despite this, . . . freedom might not also take place” (A536/B564).
And Kant develops the category of the “intelligible” as “that . . . which is not
itself appearance” precisely to make room for this (transcendental) freedom
(A538/B566). Moreover, in order to play its roles as regulative idea within
transcendental philosophy and as ground of moral responsibility in practical
philosophy, it is essential that the transcendentally free character of human
beings be the ground of our observed behavior.28
But these same two features of Denkungsart make it particularly ill-suited to
be the object of pragmatic anthropology. To start with the second point, the
notion that Denkungsart grounds Sinnesart, which is an essential feature of
Kant’s account of intelligible and empirical characters, is inconsistent with
Kant’s use of Denkungsart in his pragmatic anthropology. There, Denkungsart
and Sinnesart are two distinct explanatory principles for describing human
beings, neither explicable in terms of the other (An 7:285). Strikingly, insofar
as Kant gives either any influence on the other, it is temperament, an aspect of
Sinnesart, that influences character as Denkungsart (see An 7:290; LAn
25:1388). And with respect to the first point (that Denkungsart is not empiri-
cally knowable), I have argued extensively elsewhere that “character” in Kant’s
anthropology refers to an empirically given form of the higher faculty of desire,
subject to various empirical influences and even determining causes.29 Thus,
for example, Kant points out that “the higher faculty is specifically composed and
has its subjective laws, which precisely constitute the character” (LAn 25:483) and
27
For discussion of the grounding relationship between appearances and things in themselves, see Eric
Watkins, Kant and the Metaphysics of Causality (Cambridge: Cambridge University Press, 2005).
28
See Watkins, Kant and the Metaphysics of Causality; Frierson, Freedom and Anthropology; and Patrick
Frierson, “Two Standpoints and the Problem of Moral Anthropology,” in Kant’s Moral Metaphysics:
God, Freedom, and Immortality, ed. Benjamin J. Bruxvoort Lipscomb and James Krueger (Berlin: de
Gruyter, 2010), 83–110.
29
See esp. Frierson, Kant’s Empirical Psychology, ch. 2.
discusses how the firmness of principles essential to Denkungsart is “brought

about” or “produced” (bewirkt) by such influences as education (e.g., An
7:294; LAn 25:1172). But here it is worth adding that even in Religion and
(obliquely) in the Critique of Pure Reason, Kant uses the notion of Denkungsart
in ways that are incompatible with seeing it as referring to one’s transcenden-
tally free intelligible character. Thus, for instance, in Religion Kant discusses
how “the predisposition to the good . . . is cultivated and gradually becomes a
Denkungsart” (Rel 6:48, translation modified, emphasis added); this emphasis
on gradual change is consistent with the “gradual reform” of the empirical
character, but not at all with the instant revolution of intelligible character on
which Kant earlier insisted (see Rel 6:47). And Kant’s discussion of intelligible
and empirical character in the Critique of Pure Reason is situated in the context
of his claim that
every human being has an empirical character for his power of choice, which is
nothing other than a certain causality of his reason, insofar as in its effects in
appearance this reason exhibits a rule, in accordance with which one could
derive the rational grounds and the actions themselves according to their kind
and degree, and estimate the subjective principles of his power of choice.
(A549/B577)
While Kant does not use the term Denkungsart to refer to these subjective
principles of one’s power of choice, this description of the empirical character
of choice corresponds precisely to the definitions of Denkungsart offered in his
anthropology. But Kant emphasizes that unlike with that Denkungsart that is
identical to intelligible character, for this empirical character of choice, “there
is no freedom, and according to this character we can consider the human
being solely by observing, and, as happens in anthropology, by trying to
investigate the moving causes of his actions physiologically” (A550/B578).
This use of Denkungsart may seem to compromise the notion that
Denkungsart specifically refers to freedom, except that Kant also develops
another concept of freedom that fits perfectly with his anthropological use of
Denkungsart. In the Critique of Pure Reason, Kant makes clear that “the
transcendental idea of freedom is far from constituting the whole content of
the psychological concept of that name, which is for the most part empirical”
(A448/B476). Given that the psychological concept of freedom is “for the
most part empirical,” one can carve out a concept of “empirical freedom,”30
30
Cf. Lewis White Beck, “Five Concepts of Freedom in Kant,” in Stephan Körner – Philosophical
Analysis and Reconstruction, ed. Jan T. J. Srzednicki (Dordrecht: Nijhoff, 1987), 35–36.
660 P. R. Frierson
or what Kant often calls “comparative” or “psychological” freedom (see, e.g.,

CPrR 5:97; LM 28:267). Kant explains this sense of freedom in terms of the
ability to govern oneself in accordance with concepts or reason or the
intellect as opposed to mere inclination or sensations or stimuli. For example,
in an early lecture on metaphysics, he explains,
Animals can be necessitated strictly through stimuli <stricte per stimulus>, but
human beings only comparatively. . . . One can be forced by sensuality to act
contrary to the intellect, but one can also be forced by the intellect to act
contrary to sensuality. The more a human being has power, by means of the
higher power of choice, to suppress the lower power of choice, the freer he
is. . . . This practical freedom rests on independence of choice from necessita-
tion by stimuli. . . . However, . . . transcendental freedom . . . will be spoken of
in the Rational Psychology. (LM 28:256–57)
Kant makes explicit that this “psychological freedom . . . is treated in empiri-

cal psychology” (LM 28:267). In keeping with his aforementioned claim that
the “power of choice” and even “reason” has an empirical character, Kant can
define a sort of empirical freedom that distinguishes human beings from
other animals by virtue of humans’ capacity to regulate their actions not
merely by stimuli but also by principles (or maxims) of reason. And in those
cases, “the human being of principles, from whom one knows what to expect
not from his instinct, for example, but from his will” – that is, from “a
certain causality of his [practical] reason” (A549/B577) – “has a character” in
the sense of Denkungsart (An 7:285). Denkungsart is thus a term describing
humans’ empirical freedom, that capacity of human beings to act in ways that
are determined by the exercise of a capacity for reflective formation of
practical principles and choices in the light of those principles.
On my reading, then, the concept of Denkungsart serves two very
different purposes within Kant’s philosophy. Within his transcendental
philosophy, Denkungsart is used to refer to humans’ intelligible character,
the transcendentally free ground of that empirical character which consists
of the sum of all observable (or even empirically inferable) features of
human life. We might call this “transcendental Denkunsgart,” to refer to
the transcendental freedom that lies at its base. But within his pragmatic
anthropology as such, the primary role of the concept of Denkungsart is to
distinguish between two different kinds of empirical character of the
faculty of desire. The lower faculty of desire is governed by our tempera-
ments (and to a lesser extent our natural aptitudes). This determines what
sorts of inclinations one will have and how one’s inclinations and feelings
operate at a pre-reflective level. The higher faculty of desire is governed by

what we might call our “empirical Denkungsart,” the principles one can
observe or infer as having been adopted through processes of reflection and
that in turn control our higher volitions.31 Because our higher faculty of
desire – our “power of choice” – is reflective and governed by a reason (in
the sense of a particular power with a particular empirical character), it is
considered “free” in an empirical sense. But because even this reflection and
reason are objects of empirical investigation, they are subject to the natural
necessity of all appearances and thus not transcendentally free.
These two different purposes are not wholly disconnected, however. Kant’s
ground for ascribing human beings an intelligible character is our capacity for
moral responsibility, a capacity Kant associates with yet another sense of free-
dom: practical freedom. Within the Critique of Pure Reason, Kant is conflicted
about the relation between practical and transcendental freedom. On the one
hand, he explicitly says that “it is this transcendental idea of freedom on which
the practical concept of freedom is grounded” (A533/B561). On the other
hand, when he says that “I will use the concept of freedom only in a practical
sense and set aside . . . the transcendental signification of the concept, which
cannot be empirically presupposed as an explanatory ground of the appear-
ances” (A801/B829) and emphasizes that “we cognize practical freedom
through experience, as one of the natural causes” (A803/B831; see also LM
28:267); Kant implies that empirical freedom is sufficient for (or even identical
to) practical freedom. Whatever ambivalence might be present within the
Critique of Pure Reason, however, by the time of the Critique of Practical
Reason, Kant is quite clear than any attempt to ground ascriptions of moral
responsibility or to defend moral laws of freedom on merely empirical freedom
would be a “wretched subterfuge” (CPrR 5:97).
And yet, Kant’s use of the concept of Denkungsart in his pragmatic
anthropology brings with it the evaluation of others’ characters in moral or
at least quasi-moral terms.32 Thus Kant explicitly calls Charakter schlechthin
31
Here I am ignoring two important complications to this picture: the possibility of a sort of qualified
Denkungsart whereby one acts on the basis of principles adopted through some process of reflection, but
where this reflection is ultimately not a stable and fixed part of one’s character; and the rarity of character
(and hence Denkungsart) in the strict sense that requires consistency and stability. For further discussion
of these complications, see Frierson, Kant’s Empirical Psychology, esp. chs. 2 and 7.
32
Unlike Sturm, I do not see “the concept’s primary function” as being “to evaluate persons and their
actions,” and I certainly disagree (as does Sturm) that “Sinnesart is descriptive and explanatory while
Denkungsart fulfills a purely normative function” (Sturm, Kant und die Wissenschaften vom Menschen,
421). The concept of Denkungsart plays an important descriptive and explanatory function, in that for
one with Denkungsart, we “know what to expect from his will [that is, their practical principles], not
from his instinct” (An 7:285). But more importantly, as noted in §1, it plays a role in an account of
662 P. R. Frierson
also one’s “moral character” (An 7:285) and claims that “character always has
something worthy of respect about it” (LAn 25:823; see also 1169). The
moral and evaluative importance of character is based on an important
relationship between the two sorts of Denkungsart discussed above.
Transcendental Denkungsart is the ground of empirical Denkungsart.
Importantly, when Kant establishes the reality of an intelligible character
underlying humans’ empirical behavior, he does so on the basis of our
ascriptions of moral responsibility:
the concept of an empirically unconditioned causality [that is, the concept of

transcendental Denkungsart] is indeed theoretically empty . . . but it is never-
theless possible and refers to an undetermined object; in place of that, however,
the concept is given significance in the moral law and consequently in its
practical reference. (CPrR 5:56)
Put in terms of Denkungsart, Kant’s point here is that the mere concept of
Denkungsart as intelligible character is empty; this concept is given reality
and content by a moral argument that shows that we must assume such an
underlying intelligible character as the ground of any empirical character that
we hold to be bound by moral obligations. We hold ourselves to be so
bound, so we have a transcendental Denkungsart. But, strikingly, we do not
hold ourselves to be morally obligated with respect to every feature of our
empirical “character” in the broadest sense. In particular, we are not morally
responsible for mere matters of temperament or mere natural endowments
(see, e.g., G 4:398). That is to say that the empirical character for which we
are responsible is precisely our (empirical) Denkungsart. And in that sense,
even our empirical Denkungsart can be said to give “evidence” of practical
freedom and to warrant praise and blame. Insofar as we already ground that
empirical Denkungsart in a transcendental Denkungsart, the concept of
Denkungsart as a whole picks out those features of ourselves for which are
responsible.33
human beings that allow for the sort of prediction and mutual influence that will make possible the
achieving of our ends, both individual and collective, both hedonic and moral. Nonetheless, Kant often
does emphasize that we appraise human beings in terms of their observed Denkungsart.
33
I should make two points here. First, strictly speaking, Kant takes us to be responsible for our
empirical Denkungsart or lack thereof. Insofar as we fail to have a consistent set of principles on the basis
of which we act, we are responsible for that failure. Second, I would just note the irony that on this
account, what Kant calls (empirical) Denkungsart in his anthropological works turns out to be exactly
identical to what he calls Sinnesart in transcendental contexts. Nonetheless, he does – and it makes sense
for him to – use the term Denkungsart in these two different but related ways for different purposes.
Finally, it is important to remember that everything in Kant’s pragmatic

anthropology is designed to provide empirical knowledge about human
beings that can be put to use. Part of putting such knowledge to use involves
noting the valuable purposes to which it can be put. Thus when Kant notes,
in a section on “Character as Denkungsart,” that “character has an inner
worth, and is beyond all price,” he is exhorting his listeners to take the
knowledge of how to cultivate and recognize character and put that knowl-
edge to use to cultivate character in themselves and others (An 7:291–92,
translation modified). And that is to say that Kant is seeking to affect the
Denkungsarts of his hearers. On the one hand, this attempt to affect them
depends upon Kant’s conviction that, as a matter of empirical fact, it is
possible for exhortation, instruction, and more generally education to “bring
about this firmness and persistence in principles” (An 7:294). That is, he
trusts in his knowledge of the possible empirical influences on (empirical)
Denkungsart. But he also addresses his hearers and readers as agents with
transcendental freedom, calling on us to effect a revolution in our (transcen-
dental) Denkungsart. For one listening to anthropology pragmatically, that is,
as knowledge of human beings that can be put to use, information about
human beings is not a mere catalog of empirical causes. Rather, such
information provides one with reasons to act in particular ways. The fact,
for example, that the fear of fashion undermines character because fashion is
a “fleeting, changeable thing” gives rise to a maxim “to moderate our fear of
offending against fashion” (An 7:294).
Conclusion: What is anthropology from

a pragmatic point of view?
For Kant, anthropology is pragmatic in that it pertains to the whole of the
practical; it is meant to be put to use. It is focused on human beings as free
beings, and thus the study of (empirical) Denkungsart has a special and
prominent place within it. This focus is partly justified by the fact that
human beings must work together in society, and thus anthropology ends
up being “pragmatic” in the sense of emphasizing human traits that allow “one
human being [to have] influence on another and lead him” (LAn 25:855). The
emphasis on Denkungsart is also justified by virtue of the special connection –
in fact, the identity – between (empirical) Denkungsart and empirical freedom.
And it is particularly justified given the importance of forming one’s (empiri-
cal) Denkungsart, given each human being’s status as a transcendentally free
664 P. R. Frierson
agent, one with a transcendental Denkungsart that is responsible for our

empirical Denkungsart.
In the end, pragmatic anthropology is the use of empirical knowledge of
human beings, rooted in general concepts that systematize inner experience
but expanded through study of human behavior and motivation through
observation, in order to develop means for pursuing human goods in oneself
and others. These goods are pursued by free beings who can put empirical
knowledge to use, and they are the goods of free beings, capable of cognitive,
affective, and volitional perfections distinctive of free beings (reason, aes-
thetic pleasure, the good will), but also aiming for other goods that are
genuine goods for the finite, embodied, free beings that we are (e.g., rest after
work, self-control against passions, good dinner parties, good memory, etc.).
Of these goods, the formation and perfection of our Denkungsart is of
preeminent importance, not only as the ground of our individual perfection
but also for the gradual perfection of the species.
29
Kant on Emotions, Feelings, and Affectivity
Alix Cohen
The aim of this chapter is to extract from Kant’s various writings an account
of the nature of emotions, their underlying unity and their function – and to
do so despite the fact that Kant himself neither uses the term “emotion” nor
offers a unified treatment of it.1 A number of commentators have put
forward accounts of what they take to be Kant’s position on the nature of
the emotions.
1) Model 1 – the pain model: According to John Sabini and Maury Silver,
“Emotions [for Kant] are brute forces unconnected with higher mental
functions. Pain is the obvious model. Pain is a brute force; it is beyond the
will; it is, or at least typically is, independent of reason.”2 Similarly, Paul
Guyer defines emotions as brute sensation-like states that are opaque in
the sense that they provide no insight into their causal history.3
1
Kant uses the terms Affekt, Leidenschaft, Neigung, Gefühl, Rührung, and Begierde.
2
John Sabini and Maury Silver, “Emotions, Responsibility, and Character,” in Responsibility, Character,
and the Emotions: New Essays in Moral Psychology, ed. Ferdinand Schoeman (New York: Cambridge,
1987), 166.
3
Paul Guyer, Kant and the Claims of Taste, 2nd ed. (Cambridge: Cambridge University Press, 1997),
103–5. See also Justin Oakley’s claim that Kant’s “view seems to be based on a simple sensation model of
emotions as non-cognitive phenomena over which we have little if any control” (Morality and the
Emotions [London: Routledge, 1992], 94).
A. Cohen (*)
Department of Philosophy, The University of Edinburgh, Edinburgh, UK
e-mail: Alix.Cohen@ed.ac.uk

666 A. Cohen
2) Model 2 – the rational model: According to Marcia Baron, “Kant does not
hold that we are passive with respect to our emotions and feelings. . . . It is
a serious mistake to think that Kant’s psychology of emotion even
approximately fits the model of pain.” She suggests that “it would be
more plausible to criticize Kant for attributing to us too much responsi-
bility for our feelings and emotions than to attribute to him the position
that we are not responsible for them.”4
3) Model 3 – the pragmatic model: Wiebke Deimling suggests that emotional
states can only be unified “in that they are of special interest for pragma-
tically and morally rational action.”5
4) Model 4 – the motivational model: Janelle DeWitt argues that Kant con-
ceived of emotions in terms of their function as “action-initiating evalua-
tive judgments.”6 Similarly, Rachel Zuckert emphasizes that feelings have
a functional relationship with other mental states.7
5) Model 5 – the mixed model: According to Maria Borges, “Kant presents us
with a very colorful, wide range of emotions, which cannot be captured in
one model type.”8 Emotions, she argues, “are intentional states as well as
feelings. . . . They have a propositional content, and also bring evaluation
and cognitive elements. . . . Beliefs and desires are constitutive of these
mental states.”9 Patrick Frierson also puts forward a mixed model accord-
ing to which “so-called ‘emotions’ are cognitive, affective, and
volitional.”10
4
Marcia W. Baron, Kantian Ethics Almost without Apology (Cornell: Cornell University Press, 1995),
195, 197.
5
Wiebke Deimling, “Kant’s Pragmatic Concept of Emotions,” in Kant on Emotion and Value, ed. Alix
Cohen (London: Palgrave Macmillan, 2014), 108.
6
Janelle DeWitt, “Respect for the Moral Law: The Emotional Side of Reason,” Philosophy 89, no. 1
(Jan. 2014): 3.
7
Rachel Zuckert, “A New Look at Kant’s Theory of Pleasure,” Journal of Aesthetics and Art Criticism 60,
no. 3 (Summer 2002): 239–52.
8
Maria Borges, “What Can Kant Teach Us about Emotions?” Journal of Philosophy 101, no. 3 (March
2004): 143. See also Kelly D. Sorensen, “Kant’s Taxonomy of the Emotions,” Kantian Review 6 (March
2002): 109–28.
9
Borges, “What Can Kant Teach Us about Emotions?” 151–52.
10
Patrick R. Frierson, “Affective Normativity,” in Kant on Emotion and Value, 172. DeWitt’s account
could also fit in this category since her functional definition of emotion covers an extremely wide range
of affective phenomena. As she notes, “The forms, objects, principles and characteristics of these
judgments will vary across a wide spectrum, from the purely non-rational/non-cognitive analogues of
judgment on the instinctual end, to the purely rational/cognitive judgments on the moral end, and
passing through the usual evaluative judgments of ordinary emotional responses in between” (DeWitt,
“Respect for the Moral Law,” 3).
29 Kant on Emotions, Feelings, and Affectivity 667
These interpretations can be regrouped around two distinct issues. (1)

and (2) focus on the control we have over the emotions: either we have
no control over them (1), or we have enough, and perhaps even too
much, control over them (2). (3) and (4) focus on the function of the
emotions: either as generally relevant to action, and in particular moral
action (3), or as essential sources of motivation (4).11 The final group of
interpretations (5) takes the apparent diversity of affective states at face
value and chooses to emphasize it on the basis that the notion seems to
cover numerous, distinct states that include feeling, affect, passion,
inclination, desire, and even instinct.12 This approach suggests that it
is a mistake to look for a single set of phenomena, whether psychological
or physical, that one could call “emotions.” Instead of attempting to
identify a criterion that can unify them in spite of their apparent
heterogeneity, it settles for a mixed model.
The problem with this model is precisely that its definition of emo-
tions is so broad that it seems to include everything, from pure affective
states such as the feeling of respect to mere bodily reactions, aesthetic
feelings, and perceptions. Thereby, it loses sight of what makes these
states “emotions.” So, by contrast with (5), I will aim to put forward
criteria that define emotions by focusing on what they share intrinsically.
Although I will not be able to discuss this claim here, by contrast with
(1) and (2) I believe that emotions are susceptible to some degree of
control, correction, and justification, but that this control is indirect.13
Finally, by contrast with (3) and (4), I will deny that the motivational
role of the emotions is one of their essential features. They cannot be
defined in terms of motivational force since not all of them give rise to
desires. On this basis, I will aim to provide a definition of emotions that
is broad enough to encompass all the different kinds of affective states
allowed by Kant’s account, but that can also account for their
differences.
11
According to DeWitt, “emotions emerge as the centerpiece of Kant’s motivational account” (“Respect
for the Moral Law,” 4).
12
For instance, Borges notes that Kant “was very attentive to this variety of emotional events. . . .
Emotions refer to a wide variety of states, which call for different philosophical categories” (“What Can
Kant Teach Us about Emotions?” 157–58).
13
See for instance Alix Cohen, “Kant on the Moral Cultivation of Feelings,” in Thinking about the
Emotions: A Philosophical History, ed. Alix Cohen and Robert Stern (Oxford: Oxford University Press,
2017), 172–83.
668 A. Cohen
The distinction between the feeling of pleasure

and pain and the faculty of desire
As exemplified by models (3) and (4), commentators often fail to distinguish
feelings from their motivational function and thus end up assimilating them
with desires.14 However, on Kant’s account, affective states can take one of
two forms, depending on the faculty they originate from: either the faculty of
desire or the feeling of pleasure and pain. There is thus a distinction between
the order of desire and the order of feeling, and we need to clarify what it
consists in.15
There is an obvious sense in which feelings are connected to desire: they
can cause desires, and they can be caused by desires. But they are not
necessarily connected to desires, for they are not always motivational – or
at least not in principle.16 As Kant writes, “a feeling of pleasure which is
independent of the determination of the faculty of desire . . . is incontrovert-
ibly given” (FI 20:207), namely the feeling of the beautiful.17 On this basis,
we can differentiate pleasures according to what Kant calls their “results”: (1)
they can cause a desire (pleasure as cause), (2) they can be the effect of a
desire (pleasure as effect), or (3) they can be merely contemplative (pleasure
simpliciter). (1) and (2) are what Kant calls “practical pleasures,” which
include agreeable and moral pleasures. (3) is a contemplative, aesthetic
pleasure, which Kant describes as follows: “that pleasure which is not
necessarily connected with desire for an object, and so is not at bottom a
14
See Thomas Sturm, Kant und die Wissenschaften vom Menschen (Münster: Mentis, 2009), 395 (“we
may doubt whether for purposes of empirical explanation of action, Kant has good reasons for insisting
upon the irreducible contribution of feelings”); Jeanine M. Grenberg, “Feeling, Desire and Interest in
Kant’s Theory of Action,” Kant-Studien 92, no. 2 (July 2001): 163 (“For the purposes of describing
action, there is, however, little distinction to be made between the possession of a practical pleasure and
that of desire”); and Allen W. Wood, Kant (London: Blackwell, 2004), 142 (“What we call ‘emotions’
usually involve both the affective states that Kant calls ‘feelings’ or even ‘affects,’ but also states of desire
or even intention”).
15
“We can trace all faculties of the human mind without exception back to these three: the faculty of
cognition, the feeling of pleasure and displeasure, and the faculty of desire” (FI 20:205–6; see also
CJ 5:197–98).
16
The opposite may of course well be the case, for it is plausible to think that desires necessarily involve
feelings. But I will not defend this claim here since my focus is solely on the feeling of pleasure and
displeasure. For discussions of this issue, see for instance Sorensen, “Kant’s Taxonomy of the Emotions,”
114; DeWitt, “Respect for the Moral Law,” 7–11; and Patrick R. Frierson, Kant’s Empirical Psychology
(Cambridge: Cambridge University Press, 2014), 103.
17
See also: “We have pleasure or displeasure without desiring or abhorring, e.g., if we see a beautiful
area, then it enchants us, but we will not on that account wish at once to possess it. Pleasure or
displeasure is thus something entirely different from the faculty of desire” (LAn 29:877).
pleasure in the existence of the object of a representation but is attached only

to the representation by itself, can be called merely contemplative pleasure or
inactive delight” (MM 6:212).18 Since there is no necessary connection
between the feeling of pleasure and the faculty of desire, I would like to
suggest that they need to be defined independently of each other, at least to
begin with. While one way of explaining feelings is through their motiva-
tional results, first, it is not the only way of explaining them, and second, I
will show that it is not the best way of doing so.19
If I am correct that emotions are essentially feelings of pleasure or pain, it
follows that to understand Kant’s account of emotions, we need to focus on
Kant’s account of feeling, and to do so independently of its relationship to
the faculty of desire, including inclinations and passions.20 On this basis, I
will use the following distinction as the basis for my taxonomy, noting that,
in each case, the first instance is the normal affective states, while the second
is its pathological form:
• Faculty of desire (Begehrungsvermögen): “a being’s faculty to be by means of

its representations the cause of the reality of the objects of its representations”
(CPrR 5:10n).
a) Inclination (Neigung) is a “habitual sensible desire” (An 7:251).
b) Passion (Leidenschaft) is “a sensible desire that has become a lasting
inclination” (MM 6:408).21
18
Thus aesthetic pleasure does not result in a further desire; or if it does, it is merely the desire to
preserve its pleasurable state rather than the desire to produce its object. For instance, “The conscious-
ness of the causality of a representation with respect to the state of the subject, for maintaining it in that
state” (CJ 5:220). Being merely contemplative, it is not a desire for “the existence of the object” (MM
6:212).
19
This point is well taken by Frierson, “Affective Normativity,” 168–70, especially when he notes:
“Feelings, then, while often connected to cognitions and volitions, are not in themselves cognitive or
volitional” (Frierson, “Affective Normativity,” 170). However, he also chooses to “follow Grenberg in
downplaying the distinction between pleasure and desire” (Frierson, Kant’s Empirical Psychology, 60).
20
An additional feature of Kant’s account that would go beyond the scope of this chapter concerns the
physiological aspect of the emotions. According to Kant, feelings are not reducible to physical reactions
to certain objects or events; rather, they cause them: “the emotion of sadness produces the physical
agitation of sighing, and the emotion of fright produces a scream” (LAn 25:600, translation modified).
In this sense, Kant’s position is the reverse of William James’s account. Bodily feelings are physiological
reactions of the body to the feelings – they are neither reducible them nor their cause, but rather their
effect on the body.
21
Passions are either strong inclinations or deliberate, reflective inclinations that have been incorporated
into the agent’s maxims (for the former, see CJ 5:272, An 7:251; for the latter, see MM 6:408, CPrR
5:72). For a discussion of the distinction, see Sorensen, “Kant’s Taxonomy of the Emotions,” 117. Since
it is irrelevant to my overall argument, I will not discuss it further.
670 A. Cohen
• Feeling of pleasure and pain: “only the receptivity of a determination of the

subject” (FI 20:208). It is “the way in which we are affected by things”
(DWL 24:701).
a) Feeling (Gefühl) is “the effect of a representation (that may be either
sensible or intellectual) upon a subject and belongs to sensibility” (MM
6:211n).
b) Affect (Affekt) is “the feeling of a pleasure or displeasure in the subject’s
present state that does not let him to rise to reflection (the representa-
tion by means of reason as to whether he should give himself up to it or
refuse it)” (An 7:251).
Feelings as non-cognitive affective states

One thing is clear for Kant, namely that feelings have no cognitive import:
“pleasure and displeasure in general . . . is something merely subjective, which
yields no cognition” (MM 6:400).22 In these feelings, “nothing at all in the
object is designated” (CJ 5:204), for they “involve what is merely subjective in
the relation of our representation and contain no relation at all to an object
for possible cognition of it (or even cognition of our condition)” (MM
6:211–12; see also CJ 5:224; MM 6:211). The fact that no knowledge can
be gained through feelings has led many commentators to conclude (some
with regret) that feelings for Kant are non-cognitive.23 This claim is based on
a number of remarks, including the following:
22
See also: “Through sensation, good feeling, pain – one does not cognize an object. This is only a mode
of representation. But the representation is not distinguished by a particular object. In general, the
relation of representation to the subject is called sensation, to the object, cognition” (VL 24:904); and
“Feelings can never produce a cognition” (DWL 24:730).
23
One of the main objections raised against Kant’s account of feeling is that it is unable to explain the
fact that feelings seem to inform us about the world. For instance, Nancy Sherman notes that Kant’s
“claim is that feelings themselves don’t tell us much of anything about the world. They tell us that we
have been affected, but give us no determinate news of those things or of our own state either” (Making a
Necessity of Virtue: Aristotle and Kant on Virtue [Cambridge: Cambridge University Press, 1997], 179).
She concludes, “An intentional or evaluative view of the emotions would better cohere with his
appreciation of their epistemic function” (Sherman, Making a Necessity of Virtue, 180). The argument
defended here will address this objection, although admittedly it will not do so frontally since I will
maintain the claim that, although feelings are intentional, they remain non-cognitive and non-repre-
sentational, and moreover that they do not inform us about the world, or at least not directly. Contrast
with DeWitt, who defends “the cognitive theory implicit in Kant’s works [on the emotions]” (DeWitt,
“Respect for the Moral Law,” 1).
The agreeable, as an incentive for the desires, is of the same kind throughout,
no matter where it comes from and how specifically different the representation
(of sense and of sensation, objectively considered) may be. Hence in judging of
its influence on the mind it is only a matter of the number of the charms
(simultaneous and successive), and as it were only of the mass of the agreeable
sensation; and thus this cannot be made intelligible except by quantity. (CJ
5:266)
This passage is often interpreted as suggesting that there is no qualitative

distinction between different pleasures, since pleasure is only a matter of
degree.24 Objects can be more or less agreeable but agreeableness is always
the same, which suggests that there is no intrinsic distinction between
different kinds of pleasures apart from their intensity. However, on my
reading, this passage has different, albeit just as important, implications for
our understanding of Kant’s account of feeling.
First, we cannot know with certainty what kind of feelings we are feeling
on the sole basis of our feelings, for we struggle to make sense of them in
discursive terms (see MFS 4:471)25: “pleasure or displeasure, since they are
not kinds of cognition, cannot be explained by themselves at all, and are felt,
not understood” (FI 20:232).26 Since they “cannot be explained more clearly
in themselves” (MM 6:212), the only way of making sense of them is to
interpret them through reflective judgment in light of the context of their
occurrence – both their internal context (our past experience, our emotional
history, our personality, etc.) and their external context (what triggered the
feeling, its causal history but also the way of the world more generally).27
Although it cannot give rise to proper knowledge, thereby we are able to
discover, identify, recognize, classify, anticipate, and reevaluate our feelings.
But the nuance we experience in our affective states is the product of the
24
Most famously in Guyer, Kant and the Claims of Taste, 103–5.
25
For a discussion of the claim that there can be no science of inner sense, see Thomas Sturm, “Kant on
Empirical Psychology: How Not to Investigate the Human Mind,” and Rudolf A. Makkreel, “Kant on
the Scientific Status of Psychology, Anthropology, and History,” in Kant and the Sciences, ed. Eric
Watkins (Oxford: Oxford University Press, 2001), 163–201.
26
Contrast with Borges, who writes, “As intentional states, [emotions for Kant] have a propositional
content, and also bring evaluation and cognitive elements. . . . There is no sense in attributing sadness
without the idea that something valuable is lost.” But the problem is precisely the connection between
the feeling of sadness and this idea. What does it mean to say that “emotions express beliefs” (Borges,
“What Can Kant Teach Us about Emotions?” 151–52)? Contrast also with DeWitt, in particular when
she writes that “pleasure is a representation of an object” (“Respect for the Moral Law,” 7–8).
27
On Guyer’s account, for instance, the connection between the pleasure and the object that caused it is
drawn by a mental act that differs from the feeling of the pleasure itself (Kant and the Claims of Taste,
103–5).
672 A. Cohen
work of reflective judgment upon feelings and the web of mental states they
are part of rather than the consequence of the nature of feelings themselves.28
In this sense, what gives emotions their affective color is the fact that they are
feeling, but what gives them their meaning is reflective judgment.
Second, we can know a lot about the causal history of feelings, but we can
only do so from an external empirical perspective. When, where, and why we
feel pleasure is entirely contingent, and any object can in principle trigger a
feeling of pleasure. What object does and what does not is an issue that can
only be known a posteriori.29 So we can formulate laws based on empirical
generalizations about the various causal connections between feelings and the
objects that trigger them. For instance, chocolate causes pleasure in humans
whereas it makes dogs ill; good looking people attract more sympathy than
ugly people. These laws can be anthropological, psychological (based on
outer sense), or biological depending on the nature of the causal explanations
they put forward. They can be formulated on the basis of empirical data
including self-reports, statistics, observations of human behavior, possibly
experimentations.30 But crucially, none of this knowledge is acquired through
the feelings themselves, for they are non-cognitive.
While Kant’s non-cognitivism about feelings has often been interpreted as
entailing that there is no qualitative distinction between different pleasures
since pleasure is only a matter of degree, the next section will argue that it is
not the case. As I will show, Kant’s claim to this effect only applies to their
motivational function, “as an incentive for the desires” (CJ 5:266).
Independently of this function, however, the following section will argue
that they can be differentiated qualitatively on the basis of their intentional
object.31
28
In this sense, this account of emotion can accommodate the fact that we seem to feel a whole palette of
emotions rather than just pleasure and pain. I would like to thank Ken Westphal and Mark Timmons
for pushing me to clarify this point.
29
There are of course at least one or two special cases, namely the feeling of respect and the feeling of
beauty (see, for instance, CPrR 5:73; FI 20:206–7). For a discussion of the a priori status of these
feelings, see Julian Wuerth, Kant on Mind, Action, and Ethics (Oxford: Oxford University Press, 2014),
225.
30
Note that this knowledge allows a working notion of appropriateness, although it is a rather slim
statistical one (i.e., based on empirical, historical, psychological, or anthropological regularities). For
instance, the human norm is to be scared of bulldogs, but a number of people are not and the
appropriateness of the feeling of fear of bulldogs in not grounded on the feeling itself. When “one
also finds with regard to the agreeable that unanimity in their judging of it may be encountered among
people . . . the universality is understood only comparatively, and . . . there are only general rules (like all
empirical rules are), not universal ones” (CJ 5:213).
31
Contrast with Guyer, who argues that, since feelings are brute sensation-like affective states, they do
not have intentionality (Kant and the Claims of Taste, 103–5). For critical discussions of what is
The beauty model of intentionality

The fact that we feel pleasure and pain when apprehending certain repre-
sentations tells us something – not something about the objects represented,
but something about ourselves and our relation to these objects. For through
feeling, we may not know anything about the object and yet become aware of
our relation to it. As Kant writes in the Metaphysics of Morals, “pleasure or
displeasure (in what is red or sweet) expresses nothing at all in the object but
simply a relation to the subject” (MM 6:211–12), and I believe that this is
true of all feelings. Although feeling functions “just as if it were a predicate
associated with the cognition of the object” (CJ 5:191), it is not the failed
cognition of a representation but rather a different mode of access to it, an
indirect, affective access via its effect on the subject: “the subject feels itself as
it is affected by the representation” (CJ 5:204). In other words, although a
feeling may look like it is about an object, it is really about the subject. To
make sense of this claim, I would like to suggest that the right way of
understanding feelings of pleasure and pain is on the model of Kant’s
account of aesthetic pleasure.
On Kant’s account, when I feel aesthetic pleasure as I contemplate a
painting, while the material object of this pleasure is the painting, its
intentional object is my mental state (Fig. 29.1) – what Kant calls the
harmonious free play of imagination and understanding.32 Without getting
into unnecessary detail here, Kant’s point is that an aesthetic feeling is a
response to the representation of an object, it seems to be directed at an
object, and yet through it not only do we not cognize the object, it is not
even about the object in any meaningful sense. If it is really “about” any-
thing, it is about the subject who feels rather than the object that triggers the
feeling.33 Through it we become aware of ourselves, we feel ourselves in a
sometimes called the “opacity thesis,” see Henry E. Allison, “Pleasure and Harmony in Kant’s Theory of
Taste: A Critique of the Causal Reading,” in Kants Ästhetik/Kant’s Aesthetics/L’esthétique de Kant, ed.
Herman Parret (Berlin: de Gruyter, 1998), 466–83; and Zuckert, “New Look at Kant’s Theory of
Pleasure.”
32
The terminology I use is loosely based on Ronald de Sousa, “Emotion,” in Stanford Encyclopedia of
Philosophy, ed. Edward N. Zalta (Spring 2014), http://plato.stanford.edu/archives/spr2014/entries/emo
tion/.
33
This is of course extremely crudely formulated. For an elaboration of this claim, see Joseph Cannon,
who notes that “in [aesthetic] feeling, one is aware of one’s own state while representing an object in
intuition and reflecting on it. This is what Kant means by the judgment being ‘referred’ not to the
object, but to oneself, and it also means that ‘oneself’ here, one’s mental state, is an intentional object of
feeling, though it is not an object in a cognitive sense, nor is it the object upon which one reflects in
intuition” (“The Intentionality of Judgments of Taste in Kant’s Critique of Judgment,” Journal of
674 A. Cohen
Harmony/
disharmony
2 Object
O
3 Mate
Material object
Feeling of
pleasure/pain
Intentional object
Fig. 29.1 The intentionality of the emotions
certain relation with the representation of the object, although again, it is not
knowledge, not even self-knowledge, that we acquire; rather, we become
affectively aware of the effect of the representation on our mind.
Now my aim in this chapter is not to defend an intentionalist inter-
pretation of Kant’s account of aesthetic pleasure but to extend it to all
feelings. On the model just delineated, a painting is experienced as beauti-
ful in virtue of the fact that I feel aesthetic pleasure. The intentional object
of the feeling is not the painting but rather myself, and in particular the
mental state in virtue of which I feel aesthetic pleasure.34 In the aesthetic
case, the mental state in question is the harmonious free play between
imagination and understanding.35 What is the mental state in virtue of
Aesthetics and Art Criticism 66, no. 1 [Winter 2008]: 60). See also Henry E. Allison, Kant’s Theory of
Taste: A Reading of the “Critique of Aesthetic Judgment” (Cambridge: Cambridge University Press, 2001),
51; and Allison, “Pleasure and Harmony in Kant’s Theory of Taste,” esp. 476–77.
34
Contrast with Sherman’s claim that Kant “never develops the view that they [emotions] are inten-
tional” (Making a Necessity of Virtue, 177).
35
Although commentators disagree on whether the feeling of pleasure is caused by, or the same as, the
free play of the faculties, this issue does not impact on the argument I defend here. Contrast, for
instance, Hannah Ginsborg, The Normativity of Nature: Essays on Kant’s “Critique of Judgement”
(Oxford: Oxford University Press, 2015), esp. 42–44, 94–110; with Guyer, Kant and the Claims of
Taste, esp. 110–19, 151–60. If the feeling is the free play, à la Ginsborg, it is interpreted as
which I feel non-aesthetic feelings? The aim of the following section is to

argue that it is the interplay between our mental powers – an interplay that
can be free, unfree, harmonious, or disharmonious. As I will suggest, there
are different kinds of mental interplay for Kant, and they each give rise to
distinct feelings of pleasure and pain. In this sense, feelings make us aware
of the effect of a representation on the interplay of our mental powers, and
what inhibits this interplay is painful whereas what facilitates it is
pleasurable.
Harmony and free play of the faculties

As already noted, the feeling of aesthetic pleasure that defines judgments
of beauty involves what Kant calls a “free play” of imagination and
understanding in which they harmonize with each other. Their harmo-
nious interplay is free since no concept is applied: “The powers of
cognition that are set into play by this representation [i.e., an artistic
representation] are hereby in a free play, since no determinate concept
restricts them to a particular rule of cognition” (CJ 5:217). While an
interplay between imagination and understanding is also “requisite for a
cognition in general” (CJ 5:218), instead of being free as it is in the
aesthetic case, in the cognitive case it is unfree. The interplay is restrained
by the application of concepts, since for Kant knowledge consists in the
determinate application of concepts to intuitions schematized by the
imagination (see, e.g., A19/B33). Thus cognition and aesthetic experi-
ence engage the same faculties, imagination and understanding, albeit in a
different fashion: in the former they are in free play and in the latter in
unfree play. What they have in common, however, is that both interplays
are harmonious, which suggests that we should distinguish between free
play and harmony. To support this claim, let us examine cognition and
show in what sense successful cognition is a harmonious case of unfree
play while unsuccessful cognition is a disharmonious one. This will show
that not all cases of harmony are cases of free play and all cases of free play
are not cases of harmony.
The clearest case of harmonious unfree play between imagination and
understanding can be found in judgments of perfection, which are
self-referential. If, however, the feeling differs from the free play, their connection is either causal
(Guyer) or intentional (Allison).
676 A. Cohen
judgments that assess whether an object is the kind of tether an object is

the kind of thing it is supposed to be:
when the imagination is compelled to proceed in accordance with a determi-

nate law, then how its product should be, as far as its form is concerned, is
determined through concepts; but then . . . the satisfaction is not that in the
beautiful, but in the good (of perfection, in any case merely the formal kind).
(CJ 5:241)
In this case, cognitive harmony is based on the concept of the object, and
the interplay is conceptually determined. By contrast, in cases of cognitive
disharmony, no concept is applied since no cognition is achieved. For
instance, upon encountering a new object, say an unusual type of hous-
ing, I “determinantly” apply the empirical concept of “house,” while
“reflectively” looking for a concept, distinct from that of “house,”
which I do have, that would better fit the object.36 In the first judgment,
imagination and understanding are in pleasurable harmony, but an unfree
one, “not a free and indeterminately purposive entertainment of the
mental powers” (CJ 5:242).37 In the second judgment, however, imagi-
nation and understanding are in disharmony, or at least not in harmony,
and no pleasure arises from their interplay. The reflective use of judgment
in cognition is not harmonious until it stabilizes itself in its determinative
form – what Kant sometimes calls the “lawful agreement” between the
powers of cognition (CJ 5:295). This unfree harmony is in sharp contrast
with the harmonious free play of imagination and understanding, which
is sustainably ongoing in its freedom insofar as the very process is
pleasurable.
Now, the case of subjective disharmony, what could be called a
disharmonious free play, is famously controversial, for some have argued
that there is no room in Kant’s account for negative judgments of
36
“If a savage sees a house from a distance, for example, with whose use he is not acquainted, he
admittedly has before him in his representation the very same object as someone else who is acquainted
with it determinately as a dwelling established for men. But as to form, this cognition of one and the
same object is different in the two. With the one it is mere intuition, with the other it is intuition and
concept at the same time” (JL 9:33–34; see also An 7:191).
37
As Kant notes, “To be sure, we no longer detect any noticeable pleasure in the comprehensibility of
nature and the unity of its division into genera and species, by means of which alone empirical concepts
are possible through which we cognize it in its particular laws; but it must certainly have been there in its
time, and only because the most common experience would not be possible without it has it gradually
become mixed up with mere cognition and is no longer specially noticed” (CJ 5:187). For a discussion
of Kant’s claim that this pleasure goes unnoticed, see Melissa McBay Merritt, “Kant on the Pleasures of
Understanding,” in Kant on Emotion and Value, 126–45.
beauty.38 While it goes beyond the remit of this chapter to defend the
possibility of ugliness for Kant, it suffices to say that, just as there is a free
play between imagination and understanding that is pleasurable and gives
rise to judgments of beauty, we can, at least in principle, posit one that is
displeasurable and gives rise to judgments of ugliness.39 Kant does not
seem to suggest anything else when he talks about “the judging of an
object through taste [as] a judgment about the harmony or discord of
freedom, in the play of the power of imagination and the lawfulness of
understanding” (An 7:241, emphasis added). We can thus distinguish
between two types of free play between imagination and understanding,
one that is pleasurable and gives rise to judgments of beauty, and one that
is displeasurable and gives rise to judgments of ugliness. In the latter, the
free play is displeasurable because the interplay between imagination and
understanding is disharmonious.
Although it is less straightforward, one could extend this conception of
harmony to other faculties, for although all the cases I have examined so far
have to do with the faculty of cognition, I believe that there are other instances
of harmony. For example, the pleasure we feel when acting from duty can be
understood on the model I have spelled out. As Kant notes, acting from duty
restores “the union and harmony of the mental powers” (LE 27:366): the
faculty of desire (in the form of the will) and practical reason (which grounds
the moral law) are in harmony, and this harmony gives rise to a pleasurable
feeling of moral contentment, “an analogue of happiness that must necessarily
accompany consciousness of virtue” (CPrR 5:117).40
This is of course extremely sketchy and unfortunately, there is no room to
develop these thoughts here. What is crucial for my argument, however, is
that cases of harmonious functioning of the faculties, whether objective or
subjective, are pleasurable while cases of disharmony are painful. It suggests
38
See for instance David Shier, “Why Kant Finds Nothing Ugly,” British Journal of Aesthetics 38, no. 4
(April 1998): 412–18; and Garrett Thomson, “Kant’s Problems with Ugliness,” Journal of Aesthetics and
Art Criticism 50, no. 2 (Spring 1992): 107–15.
39
For a defense of the claim that the harmonious interplay of the faculties is not always free, and that
there is room in Kant’s account for pure ugliness, see Alix Cohen, “Kant on the Possibility of Ugliness,”
British Journal of Aesthetics 53, no. 2 (April 2013): 199–209; and Allison, Kant’s Theory of Taste, 116–17.
For a critical discussion of Allison’s claim, see Paul Guyer, “Kant on the Purity of the Ugly,” in Values of
Beauty: Historical Essays in Aesthetics (Cambridge: Cambridge University Press, 2005), 145–47.
40
“Contentment in the moral sense, however, always has reference to a state founded on consciousness
of the law-abiding use of our freedom, and thus on the conformity of our own actions with the moral
law. It relates, therefore, to the agent and the actions he has decided upon as a free being, and is truly a
contentment with himself, since it can only be effected through a state of affairs in accordance with the
moral law” (LE 27:643–44).
678 A. Cohen
that the source of the pleasurable nature of the interplay between imagina-
tion and understanding is not that it is free but rather that it is harmonious.
Some representations trigger a harmonious interplay between imagination
and understanding, sometimes this interplay is free, sometimes it is unfree,
and I become aware of it through a feeling of pleasure – and vice versa for
disharmony and displeasure. In other words, feelings track the nature of the
interplay between our mental faculties – pleasure when this interplay is
harmonious and pain when it is not. However, to make sense of this claim,
we need to account for the connection between harmony and pleasure on the
one hand, and disharmony and displeasure on the other.
Feelings as valenced appraisals of the conditions

of agency
As is often noted, Kant talks about aesthetic pleasure in terms of “a feeling of
the promotion of life [Gefühl der Beförderung derung des Lebens]” (CJ
5:244).41 What has been overlooked, however, is that the connection
between aesthetic pleasure and the conditions of life is not a feature of the
fact that the pleasure is aesthetic but rather of the fact that it is a feeling. For
throughout his works, Kant defines the feeling of pleasure and pain in terms
of the promotion and hindrance of life42: “Pleasure is the representation of the
agreement of an object or of an action with the subjective conditions of life”
(CPrR 5:9n).43 While this statement is not as clear as it could be, I will
defend the claim that pleasure is a feeling of the promotion of life because its
intentional object, the harmony of our mental faculties, furthers agency.
41
See, for instance, Allison, Kant’s Theory of Taste, 69; and Guyer, Kant and the Claims of Taste, 105,
109–110.
42
One noteworthy exception is Frierson, Kant’s Empirical Psychology, 58.
43
See also: “Enjoyment is the feeling of promotion of life; pain is that of a hindrance of life” (An 7:231);
“We experience pleasure and pain as life being promoted or hindered” (LAn 25:167); “Here [with the
sensation of satisfaction] the representation is related entirely to the subject, indeed to its feeling of life,
under the name of the feeling of pleasure or displeasure” (CJ 5:204); “What promotes our life, i.e., what
brings our activity into play, as it were, pleases” (BL 24:45); “Everything increases or furthers the feeling
of life that favors the activity of its powers; and this goes for the knowing as for the performative powers”
(Ak 15:246 [R567]; quoted in Courtney D. Fugate, The Teleology of Reason: A Study of the Structure of
Kant’s Critical Philosophy [Berlin: de Gruyter, 2014], 343); “The feeling of the promotion of life is
pleasure, and the feeling of the hindrance of life is displeasure” (LM 28:247); and “Feeling is the
sensation of life” (Ak 19:187 [R6870]). For a helpful examination of the historical roots of Kant’s
concept of life, see Fugate, Teleology of Reason, 348–53.
To make sense of this claim, it is important to note that, for Kant, what is
distinctive about the human being as a free rational finite being is that his
agency requires a certain form of organization – a systematic relationship
between its various faculties so as to fulfill the ends of reason. As Kant puts it,
mental agency requires the functioning of “all the higher faculties in accor-
dance with their systematic unity” (CJ 5:197), namely cognition, feeling, and
desire. They must be able to take up a systematic unity since, for the
conditions of agency to be realized, they must be able to operate harmo-
niously while always in flux. As a result, anything that inhibits harmony
inhibits agency, and anything that facilitates harmony facilitates agency.
On this basis, we can make sense of the painfulness of feelings of pain and
the pleasantness of feelings of pleasure in light of the fact that they manifest
the negative or positive effect of a representation upon the subject’s state of
mind. Anything that inhibits harmony, and thus agency, is painful while
everything that facilitates it is pleasurable. Kant hints at this function of
feeling in the following passage:
Here the representation is related entirely to the subject, indeed to its feeling of
life, under the name of the feeling of pleasure or displeasure, which grounds an
entirely special faculty for discriminating and judging that contributes nothing to
cognition but only holds the given representation in the subject up to the entire
faculty of representation, of which the mind becomes conscious in the feeling
of its state. (CJ 5:204, emphasis added)44
This function of feeling can be understood in terms of valenced appraisal.45

First, feelings are appraisals of whether a representation promotes or hinders
agency. And second, each appraisal is valenced insofar as a feeling is pleasur-
able when its intentional object furthers agency and displeasurable when it
hinders it: “Life is the inner principle of self-activity. . . . Only active beings
can have pleasure and displeasure. Subjects that are active according to
44
The close connection between feeling and judgment can be accounted for in a number of ways. For
instance, Kant identifies judgment as containing the a priori principle of feeling: “for the feeling of
pleasure and displeasure it is the power of judgment” (CJ 5:196; see also FI 20:245–46). As often
noted, however, this connection is hard to make sense of. On my reading, they are connected insofar as
they perform the same function, that of appraising, although they do so in different ways.
45
It is in this sense that feeling behaves “just as if it were a predicate associated with the cognition of the
object” (CJ 5:191). It functions like a judgment while conveying no propositional content. This is
something Allison notes, albeit briefly, about aesthetic feeling: feeling is “a faculty of appraisal” (Kant’s
Theory of Taste, 69). Wood formulates it in terms of valuation, but this term can be misleading in a Kantian
context: “feelings are themselves valuational” (Allen W. Wood, “Empirical Desire,” in Kant’s “Lectures on
Anthropology”: A Critical Guide, ed. Alix Cohen [Cambridge: Cambridge University Press, 2014], 142).
680 A. Cohen
representations have pleasure and displeasure” (LM 28:247–48).46 Insofar as

the mind’s potential for life is not always realized, it is the function of feeling
to make us aware of the representations that promote and those that hinder
this capacity. Of course, some may object that this account does not make
room for basic feelings such as fear or lust. But on my interpretation, these
feelings retain the same function except for the fact that the conditions of
agency they reflect are those of bodily rather than mental agency. Similar to
mental agency, bodily agency is always in flux and feelings help us appraise its
condition so as to act accordingly.47 Human agency operates at different
levels, and certain feelings correspond to each level. It is in this sense that
feelings are valenced appraisals of the conditions of agency.
Conclusion
This chapter set out to extract from Kant’s various writings an account of the
nature of the emotions, their underlying unity, and their function. As I have
argued, although we do not learn anything about the world through feelings,
we do learn a lot about ourselves, for we feel ourselves in certain relations
with our representations. It is in this sense that emotions enable the affective
awareness of the effect of a representation on the subject’s mental agency. On
this basis, I have suggested that Kant’s definition of pleasure as the feeling of
the promotion of life should be interpreted as the claim that feelings are
affective mental states that are intentional, valenced appraisals of the condi-
tions of agency. When a representation promotes the harmony of our mental
faculties, and thus our agency, we feel pleasure, and when it hinders it, we
feel pain. Some may object that this account of emotion lacks plausibility
since, as mental states, they are more nuanced than mere feelings of pleasure
46
See also: “Freedom is the greatest degree of activity and of life. . . . Whatever harmonizes with freedom
agrees with the whole of life. Whatever agrees with the whole of life, pleases” (LM 28:249–50). To avoid
misunderstanding, note that on my reading of it, life goes beyond moral life, and activity is broader than
moral agency (by contrast with Fugate, Teleology of Reason, 344–45). Life is not a moral concept; it
encompasses all the levels of human praxis, from technical and prudential activity to moral agency and
cognition. Contrary to what is often assumed, activity, including self-determining activity, is not just the
remit of practical reason; theoretical reason, and cognition in general, is not beyond the realm of agency:
“the power to judge autonomously – that is, freely (according to principles of thought in general) – is
called reason” (CF 7:27). All mental agency can be self-determining since our rational capacity underlies
all our activity, including cognition. I have defended this claim in Alix Cohen, “Kant on the Ethics of
Belief,” Proceedings of the Aristotelian Society 114, no. 3, pt. 3 (Dec. 2014): 317–33.
47
See what Kant calls “animal pleasure” and pain, which corresponds to “animal” life as opposed to
human and spiritual life, in LM 28:248.
and pain allow for. But on my reading, the nuance exhibited by our emo-
tional experience does not stem from the nature of feelings themselves but
rather from judgment, which reflects upon our feelings in light of their
context, both internal and external. As a result, emotional states are first
and foremost feelings. They can be accompanied by beliefs, judgments, and
desires, but what makes them emotions is their affective color. What this
suggests for debates on the nature of emotion is that, on the Kantian picture,
while emotions may involve conative or cognitive states, they are distinct
from both in a meaningful way since they involve a third faculty, the feeling
of pleasure and pain, which differs both from the faculty of cognition, which
generates beliefs, and from the faculty of desire, which generates volitions. In
this sense, emotions for Kant are their sui generis irreducible mental states,
and we cannot understand the full Kantian picture of mental agency without
an account of their contribution to the economy of the mind.48
48
I presented earlier drafts of this paper at the Senior Seminar in the Philosophy Department at the
University of Glasgow, at the Visiting Speaker Series in the Philosophy Department at Marquette
University, and at a conference on the Doctrine of Virtue at the University of Keele. I would like to thank
all the participants for their helpful feedback, and in particular David Bain and Michael Brady in
Glasgow, Joe Cannon and Yoon Choi in Marquette, and Mark Timmons, Kenneth Westphal, and Sorin
Baisu in Keele. I would also like to thank Hannah Ginsborg for her generosity in taking the time to go
through some of these issues with me, and Cain Todd for his numerous attempts to explain the
contemporary literature on the emotions to me.
30
The Philosopher as Legislator:
Kant on History
Katerina Deligiorgi
In several of his essays, Kant appears to be defending a progressive view of

history.1 In “On the Common Saying: That May Be Correct in Theory, but
It Is of No Use in Practice,” for example, he writes that history can be viewed
as an “unending progress toward the better” (TP 8:310; see also IUH
8:29–31; RH 8:65; CB 8:123; TP 8:312; PP 8:380; CF 7:88–89, 93–94).
What this “better” is – whether it includes moral as well as political improve-
ment – and how it may be attained are matters of keen interest and
considerable debate among Kant interpreters.2 Although I will be discussing
1
I say “appears” to avoid conflating Kant’s progressive view with what he calls “eudaimonism” or
“chiliasm,” which he explicitly rejects as “untenable” (CF 7:81–82).
2
An extremely valuable collection is Amélie Oksenberg Rorty and James Schmidt, eds., Kant’s “Idea for a
Universal History with a Cosmopolitan Aim” (Cambridge: Cambridge University Press, 2009). For
discussion on the nature of progress, see Arnd Pollman, “Der Kummer der Vernunft: Zu Kants Idee
einer allgemeinen Geschichtsphilosophie in therapeutischer Absicht,” Kant-Studien 102, no. 1 (April
2011): 69–88; Allen W. Wood, “Kant’s Philosophy of History,” in “Toward Perpetual Peace” and Other
Writings on Politics, Peace, and History, ed. Pauline Kleingeld (New Haven: Yale University Press, 2006),
243–62; and Pauline Kleingeld, “Kant, History, and the Idea of Moral Development,” History of
Philosophy Quarterly 16, no. 1 (Jan. 1999): 59–80. On the theoretical assumptions underpinning the
claims about progress, see Katerina Deligiorgi, “Actions as Events and Vice Versa: Kant, Hegel and the
Concept of History,” in Internationales Jahrbuch des deutschen Idealismus, vol. 10: Geschicte, ed. Jürgen
Stolzenberg and Fred Rush (Berlin: de Gruyter, 2014), 175–95; Lea Ypi, “Natura Daedala Rerum? On
the Justification of Historical Progress in Kant’s Guarantee of Perpetual Peace,” Kantian Review 14, no. 2
(July 2010): 118–48; David Lindstedt, “Kant: Progress in Universal History as a Postulate of Practical
Reason,” Kant-Studien 90, no. 2 (Jan. 1999): 129–47; Pauline Kleingeld, Fortschritt und Vernunft: Zur
K. Deligiorgi (*)
Department of Philosophy, University of Sussex, Sussex, UK
e-mail: K.Deligiorgi@sussex.ac.uk

684 K. Deligiorgi
and taking a position on these issues, my primary aim in this chapter is to

examine what Kant’s writings on history are about.
The question arises for Kant’s essays in a way that it does not arise for
works on, say, “Casca’s attempt at Caesar’s life,” “the Great Fire of London,”
or “the age of revolution in Europe from 1789 to 1848,” which are fairly
straightforwardly about historical actions, events, or periods. Kant’s essays
have a very wide compass and treat historical phenomena, such as wars or the
shift from hunting and gathering to agriculture, at a high level of generality,
as kinds rather than individuals.3 Kant explains that he is interested in
“human actions,” which he wants to consider “in the large [im Grossen]”
(IUH 8:17), in order to bring to view the whole they compose; he calls this
whole “universal,” “human,” or “world” history (IUH 8:17; CB 8:107; IUH
8:30; see also CB 8:121; RH 8:65; CF 7:79).
To say that Kant’s essays are about human actions as a historical whole
answers our initial question but also generates new ones. It is not clear, for
example, how such a holistic perspective might be reached and by whom,
what generic features of actions contribute to it, or indeed what the purpose
of the exercise is. Consider, by way of contrast, the resurgence of interest in
world history among contemporary historians. Their aim is to allow
neglected perspectives on historical events to come to light and to foster
“translocal” historical thinking, which is particularistic rather than general-
ist.4 As for contemporary philosophers, they have given up on encompassing
accounts of world history for being mere speculations.5
Undeniably, Kant’s conception of history is of its time. Kant makes a
contribution to a genre of writing that was familiar to his contemporaries and
Geschichtsphilosophie Kants (Würzburg: Königshausen & Neumann, 1995); and Pauline Kleingeld,
“Nature or Providence? On the Theoretical and Moral Importance of Kant’s Philosophy of History,”
American Catholic Philosophical Quarterly 75, no. 2 (Spring 2001): 201–19. Classic treatments include
Dieter Henrich, “Über den Sinn vernünftigen Handelns im Staat,” in Über Theorie und Praxis, ed.
Dieter Henrich (Frankfurt am Main: Suhrkamp, 1967), 7–37; Yirmiyahu Yovel, Kant and the Philosophy
of History (Princeton: Princeton University Press, 1980); Harry van der Linden, Kantian Ethics and
Socialism (Indianapolis: Hackett, 1988); and Sidney Axinn, The Logic of Hope: Extensions of Kant’s View
of Religion (Amsterdam: Rodopi, 1994).
3
Kant mentions war frequently in IUH, TP, and CF. It is only in the latter that he refers to “the present
war,” though it is not clear which specific war he has in mind. One famous reference to a specific
historical event, albeit veiled, is to the French Revolution at CF 7:84.
4
See Dominic Sachsenmaier, “World History as Ecumenical History?” Journal of World History 18,
no. 4 (Dec. 2007): 465–89; and Dominic Sachsenmaier, Global Perspectives on Global History: Theories
and Approaches in a Connected World (Cambridge: Cambridge University Press, 2011), 5.
5
A locus classicus for philosophical criticisms of philosophical universal history is Arthur C. Danto,
Analytical Philosophy of History (Cambridge: Cambridge University Press, 1968). For further references
and a critical discussion of Danto, see Deligiorgi “Actions as Events,” 176–77.
30 Philosopher as Legislator 685
has since fallen into desuetude. At the same time, history plays an important
part internally to the Kantian architectonic. In what follows, I argue that
Kant’s conception of history as a unified whole presents distinctive features
that are illuminating about the critical and moral commitments of his
philosophy, and also conversely, that his conception of philosophy makes
specific demands that his philosophical history aims to fulfill. The argument
is structured around four questions, each of which I take in turn: Why does
Kant believe it important that history be seen as forming a whole? How does
he argue for the unity of the whole? What are the specific claims he makes
about history? And why should anyone care for philosophical history?
Unity: Why?
Kant’s historical writings span the entire critical period, from 1784, the date
of publication of the Idea for a Universal History with a Cosmopolitan Aim,6 to
1798, the date of publication of the Conflict of the Faculties; and they vary
considerably in their focus, tone, and subject matter. What remains consis-
tent is the view of history as the unified whole of human actions. Our first
task is to examine what motivates this view. While not denying the force of
biographical explanations, provided by the intellectual context in which Kant
is writing and contemporary philosophical histories with which he is enga-
ging, we are looking for a philosophical answer to our question.7 To uncover
it, we need to look outside Kant’s essays on history, in works where he
explicitly discusses the importance of unity (Einheit) for cognition and
action. Examining Kant’s arguments for unity in these other contexts can
help to contextualize these essays and orient our expectations regarding the
study of history.
In the published works, the most extensive discussion of unity occurs in
the Critique of Pure Reason when Kant argues that reason aims at establishing
its own unity, “‘the unity of reason’” (A302/B359). In the first Critique, the
6
Although I follow Allen Wood’s Cambridge translation here, the German word that he translates as
“aim” is Absicht, purpose. In what follows, this notion will play an important role in conjunction with
the notion of “end [Zweck].”
7
Kant reviewed J. G. Herder’s 1784 Ideas for the Philosophy of History of Humanity (RH 8:43–66). He
discusses Moses Mendelssohn’s views on history when he addresses Jerusalem oder über religiöse Macht
und Judentum (1783) in “Theory and Practice” (TP 8:308), and he indirectly confronts Mendelssohn
under the label “abderitism” in Conflict of the Faculties (CF 7:81). Mendelssohn was responding to the
progressive views expressed by G. E. Lessing, who wrote a very influential essay on The Education of
Humanity, which appeared in 1777 in part, and in full posthumously in 1790.
686 K. Deligiorgi
object of reason’s attention is the understanding: unity is a “demand of

reason, in order to bring the understanding into thoroughgoing connection
with itself” (A305/B362). This concern with the unity of reason, however,
has both deep roots and wider manifestations than this quote indicates. As
Paul Guyer has shown, this is a concern that shapes decisively Kant’s
thinking in the decade prior to the publication of the Critique.8 “The
business of reason,” Kant writes in 1778–1779, “consists in creating uncon-
ditional unity in the greatest manifold in the employment of the under-
standing” (Ak 18:225 [R5553]). To come to grips with this and to see
whether it has any application to the question of the unity of the manifold
of human actions, we need to look at the work of the understanding first.
The role of the understanding is to unify appearances “by means of rules”
(A302/B359). This unity serves our cognitive needs. It is thanks to the work
of the understanding that we have experience, structured cognitions that
form the basis of our knowledge of objects, including historical objects or
“empirical data” (CF 7:91). Nothing follows from this, however, about
history – or anything else for that matter – as a whole. History as a whole
is not a possible object of experience and so it would seem that any inferences
to a totality of human actions is subject to Kant’s strictures concerning
dialectical inferences of reason, that is, inferences to conclusions about
unity in the object, which is to say, completeness and totality of appearances
(see Ak 17:703–5 [R4757]; Ak 18:225–29 [R5553]).
To this negative argument, cautioning against misguided uses of reason,
there corresponds a positive one about the regulative use of the ideas of
reason, discussed mainly in the appendix to the Transcendental Dialectic.
These ideas serve the demand of reason to bring the understanding into
unity. This demand arises from reason itself (cf. A338/B396, A651/B679,
A665–66/B693–94). The idea of a demand of reason can be seen either as
stating a brute fact about reason or as stating a fact about our cognitive needs,
since, as Kant claims, the need to unify the employment of the understand-
ing, by making our cognitions systematic, ensures that our cognitions are in
good order.9 Whichever way we take it, the demand of reason is satisfied
8
Paul Guyer, Kant on Freedom, Law, and Happiness (Cambridge: Cambridge University Press, 2000),
60–95. Guyer distinguishes between systematicity and completeness, and he argues that the unity of
reason is effectively practical and consists in the idea of attainment of maximally consistent system of
purposes or “systematic happiness” (94). Although I follow Guyer in my discussion of the Reflexionen, I
do not distinguish as sharply between completeness and systematicity, and my argument about the unity
of human actions required for philosophical history takes a different direction.
9
See too: “Progress (progressus) in knowledge (qua science in general) begins with the collection of the
elements of knowledge, then connects them [in the] manner in which they are to be arranged
through the employment of regulative concepts or “ideas.”10 Although there

may be indirect cognitive gains accruing from the regulative employment of
reason, Kant insists that our knowledge is not enlarged through it: “If we
survey the cognitions of our understanding in their entire range, then we find
that what reason quite uniquely prescribes and seeks to bring about concern-
ing it is the systematic in cognition, i.e., its interconnection based on one
principle” (A645/B673). The ideas that Kant sanctions in this section are
discipline-specific ones, such as “pure earth, pure water, pure air” for
chemistry (A646/B674), and the idea of a “fundamental power” for psy-
chology (A649–50/B677–78). Later, he includes ideas that were previously
shown to be the result of illusory ontological inferences: soul, world, and
God. These ideas, he argues, have permissible uses regulatively, on account of
their unifying function. For example, once it is not considered ontologically,
the idea of the world allows us to think of “nature in general, and the
completeness of conditions in it in accordance with some one principle”
(A685/B713). Such an idea expresses the conviction – and so an intellectual
stance against both empiricist fragmentation and rationalist realism – that
different explanatory models of nature as a “single given object” can be
integrated within a whole that can encompass all (A684/B712). While
history is not included in this discussion, in his very first essay on history,
Idea for a Universal History, Kant invites his readers to consider precisely an
“idea” of history. So it would not be unreasonable to extend the regulative
use of ideas of reason in the domain of history and argue that history allows
us to view human actions in a systematic way. But why should such a view of
human actions matter? Why is unity important for history?
To answer this question, we can turn it on its head and look at why lack of
unity might be a problem. Unlike nature, which is the whole of appearances
(see, e.g., Pro 4:295; MFS 4:467), history is a subset of appearances, those
which can be identified as human actions. History then, as many commen-
tators point out, is the domain of the realization of our practical aims, our
(systematically). For the division of this enterprise into a doctrine of elements and a doctrine of method
constitutes the supreme division; the former presents the concepts, the latter their arrangement in order
to found a scientific whole” (OP 21:386).
10
See Paul Guyer, Kant’s System of Nature and Freedom: Selected Essays (Oxford: Clarendon, 2005), 11–
38. A different approach, by Paul Franks, is to show that Kant offers a deduction from an absolute
ground to secure the requisite unity and so answer the Agrippan trilemma. See Paul W. Franks, All or
Nothing: Systematicity, Transcendental Arguments, and Skepticism in German Idealism (Cambridge:
Harvard University Press, 2005), 62–79.
688 K. Deligiorgi
moral ends and political aspirations to justice.11 The introduction of these

practical concerns creates a distinctive need. When well-intentioned people
consider history, Kant argues, they are beset by moral despair:
The thinking human being feels a sorrow [Kummer], one which can even become
a moral corruption, of which the thoughtless knows nothing: namely, discontent
with the providence that governs the course of the world on the whole [im
ganzen], when he estimates the ills that so much oppress humankind, and (as it
appears) leaves it with no hope for anything better. (CB 8:120–21)
Constructing history as a unified whole can have a role in alleviating the

sorrow that can lead to moral corruption, provided the whole is one in which
oppressive ills are neutralized. So one justification for unity in history might
go like this: thinking human beings need hope, hope needs assurances of
progress, and assurances of progress need a progressive view of history as a
whole. Such a history would be contrived to console, embolden, and reassure
the righteous. So constructed, history is not answerable to cognitive needs
but rather to psychological ones.12
While not without foundation in Kant’s writings, this psychological expla-
nation for treating history as a unified whole is quite misleading, because it
treats hope merely as an affect related to the expectation of something good (as
in An 7:255). There is considerably more to the question of hope than this
affect, however. In the first Critique, Kant describes the question “What may I
hope?” as being “simultaneously practical and theoretical” (A805/B833). Note
that, in its general form, the question arises with respect to any prospective
doing: one seeks reassurance about a theoretical matter, concerning facts and
likelihoods, prior to undertaking a practical commitment. Kant, of course, is
interested in our moral commitments here. He analyzes the question by
providing the suppressed conditional, “‘If I do what I should, what may I
then hope?’” (A805/B833, emphasis added). Crudely translated, the questioner
asks: if I am good, what am I to expect? The key fact implicated in all of our
expectations, according to Kant, is our natural desire to be happy. Less crudely,
the question is about the possibility of a “connection of the hope of being happy
with the unremitting effort to make oneself worthy of happiness” (A810/B838).
Importantly then and despite its first-personal formulation, the question of
11
Most recently in Kristi E. Sweet, Kant on Practical Life: From Duty to History (Cambridge: Cambridge
12
Cf. the distinction Kant makes between psychological and moral (e.g., An 7:258).
hope is ultimately conceptual and metaphysical, not personal: it concerns the

possibility of uniting happiness and goodness.
How is the question to be answered? “Not through speculation” is the
brief answer. Assessing the possibility of uniting happiness and goodness
exceeds the speculative use of reason, because it brings into play practical
matters, such as the idea of the determination of the will “with respect to the
final and complete end” (CPrR 5:120). The attainability of this end, which
Kant calls the highest good, is a matter for practical reason and is addressed
in the second Critique in the discussion of the postulates, immortality of the
soul, an intelligible world, and a “highest independent good” (CPrR
5:132).13 This discussion takes us beyond history. The ideas that satisfy
the practical interests of reason and that answer to the conceptual and
metaphysical issues raised by the question of hope are concerned with history
only negatively, in the sense that they are history-transcending. As Kant puts
it in “Theory and Practice,” even though we seek to bring about “happiness
combined with and in conformity with the purest morality,” it is only duty
that is within our power; the rest “exacts from reason belief, for practical
purposes, in a moral ruler of the world and in a future life” (TP 8:279).
To appreciate how the question of hope relates to the topic of history, we
need to look again at reason’s demand for unity. In the Architectonic of Pure
Reason, Kant defines architectonic as “the art of systems.”14 A system is “the
unity of the manifold cognitions under one idea” (A832/B860). The contrast
is with systems that are not really worthy of this name and that are mere
aggregates of cognitions (A835/B863). Philosophy is just the system of
philosophical cognition. Kant introduces here an important distinction
between two different kinds of philosophy. The first, which is the only
philosophy practiced “until now,” is “scholastic.” It seeks “the logical
perfection of cognition,” but it fails to take into account moral ends of
humanity (A838/B866). The second kind, which Kant implies is new and
starts with him, follows the “cosmopolitan concept,” according to which
philosophy is “the science of the relation of all cognition to the essential ends
13
The intelligible world is just what Kant calls the ideal world, obtainable by abstracting all spatiotem-
poral notions. But “for such an intelligible word to be the highest good,” we need God, because this
world as a good contains the notion of the law of freedom through which one is able “to determine one’s
will” (CPrR 5:132). To these postulates of practical reason, we should add the concept of the “ethical
community,” which is introduced in the Religion and designates a union of people under laws of virtue
or divine commands, as “people of God” (Rel 6:98).
14
For a detailed treatment of these passages, see Lea Ypi, “Practical Agency, Teleology and System in
Kant’s Architectonic of Pure Reason,” in Politics and Metaphysics in Kant, ed. Sorin Baiasu, Sami
Pihlström, and Howard Williams (Cardiff: University of Wales Press, 2011), 134–51.
690 K. Deligiorgi
of human reason (teleologia rationis humanae)” (A838–39/B866–67). The

philosopher who follows the cosmopolitan concept is “the legislator of
human reason” (A839/B867). The legislation of human reason, in turn,
has “two objects, nature and freedom.” The former “pertains to everything
that is [was da ist]” and the latter “to that which should be [was da sein soll]”
(A840/B868). Systematic philosophy in the cosmopolitan sense aims to
bring together the theoretical and the practical in the legislation of human
reason. Only through such a systematic conception of the legislation of
human reason can the practical ends of reason be promoted (A832/B860).
The discussion about architectonic grants us a unique perspective from
which to appreciate the importance of unity in history. To see history as a
whole is to have in view human actions. In this context, “history” is a
collective name for human actions in the large, im Grossen (IUH 8:29).
The reason for considering human actions in the large is because we take an
interest in the essential ends of human reason. The question of hope can
focus our attention on history, if we take it to ask about the realizability of
what ought to be within the natural world. Natural phenomena are subject to
natural causal laws. By contrast, what we set out to realize as moral beings are
laws of freedom. The domain of the realization of these practical ends is the
domain of actions. This domain, in turn, is made available for philosophical
consideration through a conception of history as a whole. The this-worldly
and historical question of hope is addressed by the cosmopolitan concept of
philosophy that aims to unite theoretical and practical uses of reason. So the
philosophical motivation for a unified conception of history relates directly
to this more ambitious unity that philosophy, in its cosmopolitan concep-
tion, aims to establish between freedom and nature.
The challenge now is to show how the historical whole is formed.
Minimally, it must be composed of actions. For this, a concept is needed
that is action-specific and can serve as a rule for unity of actions so that they
form a whole. The next task then is to identify the conceptual tools that
enable Kant to adopt a perspective on history as a whole.
Unity: How?
Actions are observable phenomena like any other natural occurrences (cf. IUH
8:18). All appearances, including actions, are subject to causal laws. So causality
cannot be the rule that helps us pick out actions. Basically put: “Matter causes
[wirkt]. Will acts [Willkür handelt]” (OP 21:226). To pick out actions, we need
some rule other than causality. If we look for something that picks out actions
more narrowly and distinguishes them from other natural occurrences, we need
to look at what further characteristics actions have. In the first Critique, Kant
concerns himself with how actions are brought about. He argues that, although
the will (Willkür) has “an empirical character” (A552/B581), reason should be
considered as “the persisting condition of all voluntary actions” (A553/B581).
To this intelligible character of the action, however, Kant says “no before or
after applies” (A553/B581). When we look at history, however, before and
after do apply. For example, Kant writes that a universal cosmopolitan existence
may be achieved “after many transforming revolutions” (IUH 8:28), or that
practical reason may triumph “after many unsuccessful attempts” (TP 8:313).
So we need to look for a mark of actions that is amenable to this kind of
temporal talk but that also sets them apart from the movement of the planets or
a shift in weather patterns.
In the Metaphysics of Morals, Kant describes all actions as being end-
directed: “Every action,” he says, “has its end” (MM 6:385). He gives the
following account of ends: “An end [Zweck] is an object of the choice
[Willkür] (of a rational being), through the representation of which choice
is determined to an action to bring this object about” (MM 6:381). He later
specifies that an end “is an object of free choice” (MM 6:384). So, on this
account, an action involves at least the determination of the will (Willkür) or
exercise of our capacity to choose, which is transcendentally free, and the
representation of the object we seek to bring about, or end. Because Willkür
is determined by the representation, “to have any end of action whatsoever is
an act of freedom on the part of the acting subject, not an effect of nature”
(MM 6:385). This connection of the end of an action and freedom makes
the notion of an end a plausible candidate for a distinctive mark of actions.
The notion of end is attractive because, while connected to freedom, it
admits of temporal talk. For example, one of the ends that Kant recommends
as a duty is one’s own perfection through the cultivation of one’s faculties,
which explicitly involves progressive achievements such as diminishing one’s
ignorance and correcting one’s errors (MM 6:387). What we need now is a
way to unify actions in the large through this notion of end.
Ends (or final causes) are the domain of teleology. Kant discusses teleology
extensively in the third Critique, granting it a legitimate place in the natural
sciences, in particular the biological sciences.15 Our question is whether or to
15
For an excellent account of the early modern discussion of final causes, see Jeffrey K. McDonough,
“The Heyday of Teleology and Early Modern Philosophy,” Midwest Studies in Philosophy 35, no. 1
(Dec. 2011): 179–204.
692 K. Deligiorgi
what extent this discussion is relevant to the history of human actions. Kant
turns to teleology because mechanical causal natural laws do not suffice to
explain certain features of organisms, such as their capacities of self-repro-
duction and self-preservation (CJ 5:371). He argues that we should accept
that there are things, products of nature, which are possible as ends and
which cannot be cognized through the understanding alone, “rather even
empirical cognition of their cause and effect presupposes concepts of reason”
(CJ 5:370). Basically, teleology is important just to cognize these things. The
problem with invoking reason is that it presupposes “acting in accordance
with ends (a will)” (CJ 5:370). Kant then devotes most of his analysis of
teleology to show that this problem is superable. The upshot is his vindica-
tion of the notion of “a natural end [Naturzweck]” (CJ 5:370). The idea of a
natural end helps explain how we cognize beings that are organized and self-
organizing (CJ 5:374).
The basic shape of the difficulty Kant tries to address here is this: even
assuming a dynamic conception of matter composed of attractive and repulsive
forces (see MFS 4:537), it is still a mystery how the formative force of natural
products is communicated to matter so that they look like they are deliberately
put together (CJ 5:374). A Naturzweck is a purposive unit whose parts are
determined by the whole: “Organized beings are those of which, and in which,
each part is there for the sake of the other” (OP 21:184). This “for the sake of”
gives us the telic function contained in the notion of a purposive unit. The
notion of a purposive unit, in turn, comes with a normative conception of
what something ought to be, that is to say, the end for the sake of which the
parts are organized just so it can be realized or not. If the end is realized, the
being in question is a good specimen of its kind; otherwise it is defective in
some way. Importantly for Kant’s argument, from teleological explanations of
parts of nature – “or even of nature as a whole” (CJ 5:397) – no ontological
conclusions may follow; teleology is a guideline or a maxim for the reflecting
power of judgment (see CJ 5:389, 399).16
16
On the normative conception of form, see Hannah Ginsborg, “Kant’s Biological Teleology and Its
Philosophical Significance,” in A Companion to Kant, ed. Graham Bird (Malden, Mass.: Wiley-
Blackwell, 2005), 455–70; and Marcel Quarfood, “Kant on Biological Teleology: Towards a Two-
Level Interpretation,” Studies in History and Philosophy of Biological and Biomedical Sciences 37, no. 4
(Dec. 2006): 735–47. For a more Aristotelian view and contemporary discussion, see Mark Bedau,
“Where’s the Good in Teleology?” Philosophy and Phenomenological Research 52, no. 4 (Dec. 1992):
781–806. Kant’s discussion of the legitimate use of teleology is part of his solution to the antinomy of
teleological judgment, which is presented as an antinomy between mechanism and final causes (CJ
5:386–89 [§§70–71]). In what exactly the antinomy consists is a matter of ongoing controversy; see
Angela Breitenbach, “Kant on Causal Knowledge: Causality, Mechanism, and Reflective Judgment,” in
The discussion of teleology and its applications in biology is not obviously

helpful to the topic of history. If the notion of “end” is fully at home in the
study of organisms, then it cannot serve as a distinguishing mark of human
actions. So we need to examine whether ends of actions are in any way
distinctive. In the first Critique, Kant introduces the notions of “purpose
[Absicht],” “purposiveness [Zweckmäßigkeit],” and “end [Zweck],” arguing
that they can help with the systematicity demand, by enabling us to “regard
every ordinance in the world as if it had sprouted from the intention
[Absicht] of a highest reason” (A686/B714). To have a purpose in pursuing
an end requires a conscious being who acts intentionally in order to realize
the end. One example, in the cosmic context Kant is discussing here, is the
God of theistic arguments, the “world-author” (A687/B715). Authorship is
connected with a representation – or “consciousness” (MM 6:224) or “idea”
(CJ 5:435) – of what one sets out to do, which is characteristic of purposeful
productions. In light of this connection, we can see the teleological explana-
tions that Kant defends in the third Critique as stripped down, aiming
precisely to loosen the link between authorship on the one hand and purpose
and end on the other. So for example, Kant concedes that an organism,
which is just the “product of the parts and of their forces and their capacity to
combine by themselves,” and so the product of “a mechanical kind of
generation,” appears as a purposeful (absichtliche) production even though
it is not (CJ 5:408). In the Opus postumum, Kant even likens organisms to
artifacts, only to emphasize the difference from artifacts, namely the absence
of authorial purpose: “An organic natural body is thus thought of as a machine
(a body arranged intentionally as to its form). Under no circumstance can it
be a property of matter to have an intention [Absicht] . . . ; for all matter (and
every part of it) is composite” (OP 22:548). One conclusion we may draw
from this is that the discussion of teleology in the biological context is not
directly helpful for history precisely because it presents us with a version of
teleological explanation from which purpose (Absicht) is taken out. Natural
teleology is not irrelevant to history, as we shall see presently, since the fact that
human beings are also natural beings plays a role in how Kant construes
history. It is simply not sufficient for the study of human actions as a whole,
for which we need the concept of a purposefully pursued end.
But now a new problem arises. Purpose is fine when applied to
individual actions (cf. MM 6:227). I purposefully pursue an end and
Causation and Modern Philosophy, ed. Keith Allen and Tom Stoneham (New York: Routledge, 2011),
201–19.
694 K. Deligiorgi
as such count as the author of my action, you purposefully pursue an

end and count as the author of your action, and we may share a purpose
and coordinate our actions. By contrast, if we consider human actions in
the large, as Kant invites us to do, it is unlikely that we will find that
there is a single purpose to them. We cannot assume coordination of
intentions for the sake of some end. But if we cannot do that, then we
do not have a teleological whole of history; we only have this action or
that, this purpose or that. Formally we can still of course speak of history
as a collective name for actions, but the material of history will be as
disparate as ever. Short of assuming some supra-historical author whose
intentions are realized in history, nothing so far gives us grounds to
consider, as Kant states in the very title of Idea for a Universal History,
that there is a cosmopolitan purpose (Absicht) to history as a whole.
Here are the elements we need to fit together: natural teleology, the
purposeful teleology of discrete actions, and history understood as a whole of
which something can be said with regard to its overall purpose. Natural
teleology remains relevant to our concerns, because it concerns beings and
thus also human beings. In the Idea essay, which predates the first edition of
the third Critique, Kant already argues that it is reasonable to assume what he
calls the “teleological doctrine of nature” in order to make sense of the relations
between parts of organisms and the organism as a whole (IUH 8:18). Kant
then applies the doctrine to the natural predispositions of human beings,
arguing that they are set to manifest themselves and develop. This move,
from organic parts and organic wholes to natural predispositions and their
manifestation, is not fully argued for, but it is in conformity with arguments
developed later in the third Critique. Still, nothing follows from this with
respect to purposes in discrete actions or actions in the large. Some authors
take this reference to natural teleology as evidence that, in this essay at least,
Kant treats history as a branch of biology.17 One way of understanding this
claim is that history is treated as a being with parts arranged for the sake of an
end. The epistemic problem is that, unlike natural organisms with which we
are acquainted and which give us our notion of the end for which their parts
are arranged, we have no equivalent contact with history as a whole.
The question is this: from where do we get the idea of an overall purpose,
Absicht, served by or through human actions (IUH 8:24)? Kant is not offering
17
See Allen W. Wood, Kant’s Ethical Thought (Cambridge: Cambridge University Press, 1999), 208.
For a different view that still acknowledges, as I do here, that human beings are also natural beings, see
Karl Ameriks, “The Purposive Development of Human Capacities,” in Kant’s “Idea for a Universal
History with a Cosmopolitan Aim,” 46–67.
his readers the fruits of his empirical research into history. We said earlier that
it is unlikely that we find a common purpose to human actions in history, and
Kant would agree to this: history in his sense is not a matter of empirical
discovery. Nor, however, is he claiming that he has achieved a theoretical
vantage point from which he can inspect human history as a whole and tell us
something about how it ends. History is in part a theoretical abstraction,
merely the compositional aggregate of all actions and a collective name for
human actions. Attaining this theoretical perspective is costless and also not
hugely helpful, given our earlier answer to the “why?” question.
We established previously that history speaks to a this-worldly concern
about the practical unity of nature and reason. This role cannot be fulfilled
by a mere compositional aggregate nor by pointing at this or that action. We
need a notion of history as an organized whole. The shape of this whole is not
read off its parts, discrete actions, even while it is the parts alone (it seems)
that contribute to history’s organization. And this is precisely how Kant puts
it in the Idea essay: we seek, he says, “a guiding thread for exhibiting an
otherwise planless aggregate of human actions, at least in the large [im
Grossen], as a system” (IUH 8:29; see also CF 7:83).18 “History,” as Kant
uses the term and variously qualifies with the adjectives “universal” or
“world” or “human,” is already a philosophical term of art: it is the idea
that ends of human actions are organized purposefully.19
What is history?
One way of summarizing the findings of the previous two sections is to say that
history is a whole shaped to fit the cosmopolitan concept of philosophy. While
the employment of such a concept sounds indulgently speculative, Kant
18
The discussion of this “thread” belongs to an argument about chance and the prospect of barbarism,
which forms the rhetorical context for the questions: “whether it is indeed rational to assume purpo-
siveness [Zweckmäßigkeit] in the arrangement of nature in the parts and yet purposelessness [Zwecklosigkeit]
in the whole” (IUH 8:25). See also Kleingeld, “Kant, History, and the Idea of Moral Development.”
19
There is a parallel and a contrast here with cognitive needs relating to what Kant calls “delimitation”
and “outline” that arise with respect to natural phenomena. In the Opus postumum, Kant defends the
need for a metaphysical foundation of natural science in relation to physics, because “without the
former, [it] would be merely an aggregate (farrago) of observations of nature that would permit no secure
delimitation or outline” (OP 21:477–78; see also 183). The metaphysical foundations mentioned here
give the basic dynamic theory of matter that I attributed to Kant earlier. See also Michael Friedman,
Kant’s Construction of Nature: A Reading of the “Metaphysical Foundations of Natural Science”
(Cambridge: Cambridge University Press, 2013), 221–22. In history we seek an outline of a dynamic
whole composed of actions that does not, however, determine actions a priori.
696 K. Deligiorgi
believes that he can offer a critical defense of it. He rejects standard progressive
views of history – “eudaimonism, with its sanguine hopes . . . of progress end-
lessly broadening its course toward the good” (CF 7:82) – and claims on behalf
of his account that it is tenable within the constraints of “the most rigorous
theory” (CF 7:88). Kant’s assurances have not always convinced readers, who
criticize his philosophical history for being dogmatic, necessary but ultimately
incoherent, or, at best, a heuristic depending on a now obsolete natural
teleology.20 The more sympathetic readers view it as a regulative idea or an
instance of the exercise of reflective judgment.21 I doubt that we need these
theoretical tools to ensure the critical acceptability of Kant’s philosophical
history, because, as we shall see, its theoretical claims are limited and, on the
whole, weak. Recall that philosophical history matters because a this-worldly
answer to the question of hope matters. The question of hope, Kant says, is
“simultaneously practical and theoretical” (A805/B833). The this-worldly,
historical portion of his answer to the question “what may I hope?” is equally
hybrid: it contains some relatively undemanding theoretical claims and the
outline of a practical standpoint, which Kant invites us to occupy as readers of
these essays. I discuss the theoretical claims first.
Kant wants to adduce support for claims about the attainability of progress
in various domains of human life. To do this, he turns to general facts about
human behavior, which explain the kinds of things people do: engage in
wars, form societies, and so forth. These general facts of human behavior are
in turn presented using the conceptual tools we identified earlier: natural
teleology and the purposeful teleology of individual actions. When human
actions are viewed in the large, with the help of these conceptual tools, Kant
argues that certain rational ends are attainable, and therefore, one may
attribute purpose (Absicht) to this whole even in the absence of a single
author. Attainability of rational ends is what counts as progress. Different
essays emphasize different sorts of rational ends. Most focus on political
arrangements – the prospects of the establishment of a just constitution
20
Earlier interpreters (e.g., Yovel, Kant and the Philosophy of History) tend toward the “dogmatic but
necessary” view, whereas more recent interpreters (Kleingeld, Fortschritt und Vernunft; Kleingeld,
“Nature or Providence?”; and Henry E. Allison “Teleology and History in Kant: The Critical
Foundations of Kant’s Philosophy of History,” in Kant’s “Idea for a Universal History with a
Cosmopolitan Aim,” 24–45) find a greater range of theoretical resources in Kant’s thought for dealing
with history and progress.
21
See, for example, Rudolf A. Makkreel, Imagination and Interpretation in Kant: The Hermeneutical
Import of the “Critique of Judgment” (Chicago: University of Chicago Press, 1990), esp. 130–31; and
Katerina Deligiorgi, “The Role of the ‘Plan of Nature’ in Kant’s Account of History from a
Philosophical Perspective,” British Journal for the History of Philosophy 14, no. 3 (March 2006): 451–68.
within states and of peaceful relations between states – and some discuss
broader gains in culture.
Despite these differences, Kant’s treatment of history is remarkably consist-
ent: first he identifies action kinds, then he analyses them in terms of general
traits of human behavior, and then he considers whether, given these traits,
right and virtue can be established. In the essays concerned with political
progress – Idea for a Universal History, “Theory and Practice,” and Perpetual
Peace – Kant identifies certain actions, such as people forming social groups,
cultivating the land, getting into conflicts, and so on. He then analyzes these
actions with reference to general anthropological traits, such as antagonism
(IUH) and conflict (CB). Finally, he argues that these traits do not hinder and
can even promote the realization of principles of right. So Kant gives an
affirmative answer to the question: given such and such facts of human
behavior, is a hopeful or progressive perspective on history tenable? On the
basis of this, he then claims that a purposeful shape can be given to the whole of
human history. In short, form is, in part, what the content permits.
I will show how this theoretical argument works by focusing on one
example from the Idea essay. As we saw earlier, here natural teleology applied
to human beings gives the argument that human capacities and talents are
bound to develop (IUH 8:18). Human capacities and talents are displayed in
actions. Individuals purposefully pursue whatever ends they choose, honing
their talents and developing their abilities in doing so. Observing such
matters from afar, or “in the large,” two traits emerge that are constant in
human life, regardless of what talents people have and what ends they
purposefully pursue: sociability, the need to cooperate for success in individ-
ual endeavors, which brings people together in societies; and its opposite,
the desire to be free of interference from others (IUH 8:20–21). This
example, replicated in other works, shows how Kant seeks to identify
dynamic structures that have a basis in natural teleology but also allow us
to take into account the various projects human beings pursue.22 These
dynamic structures, such as “unsociable sociability” in Idea, enable Kant to
say something general about the means by which any ends human beings
have are realized. We do things by pooling our resources together and by
splitting apart. Because this “antagonism” is a natural trait of the species
(IUH 8:20), Kant then examines whether, given this trait, the establishment
22
In later essays, as well as in the third Critique (esp. CJ 5:430–33), Kant makes much more of the
sociability aspect. He gives a highly differentiated concept, “culture,” as the matrix for the development
of human talents not just individually but also across generations.
698 K. Deligiorgi
of a just constitution is possible. To this he gives an affirmative answer

because justice does not require the eradication of antagonism. On the
basis of this, he then argues that nature, understood as the natural forces
that operate in human societies, can be seen as contributing to the realization
of an end, which is not purposely pursued by the individuals concerned, yet is
attainable, “i.e. a perfectly just civil constitution” (IUH 8:22), which ensures
individual freedoms within a whole that permits cooperation.23 The idea of
“purpose” attributed to the dynamic natural/historical whole is simply
another way of speaking of the attainability of (some) rational ends in the
long run, and the attainability of even some rational ends counts as progress.
The practical standpoint outlined in the historical essays comes to view once
we pose the question of why anyone should be tempted by such an ambitious
conception of history in the first place. Granted, the theoretical claims are weak.
Still, nothing like a philosophical history is needed for comparative judgments
about progress. International organizations and think-tanks regularly check
progress over time given certain goals, such as “democracy,” “governance,”
“poverty reduction,” and so on. Philosophical history – the idea of a purpose-
fully, progressively organized whole of human actions – has a practical role that
is revealed once we consider ourselves qua agents.
As a way into the practical argument, I want to use Dieter Henrich’s idea
of what he calls the “moral image of the world.”24 Henrich develops this
argument in the context of a discussion of the postulates of pure practical
reason, and in particular the idea of the final and complete end. We discussed
this earlier, only to conclude that it is a history-transcending goal, and
therefore not immediately relevant to a progressive view of history.
Nonetheless, the way Henrich presents the idea of the moral image is
relevant to our current concerns and in particular the practical perspective
from which Kant is making his argument about history. Henrich does not
define at the outset what this moral image is; he simply says that it is an
image of the world that enables the agent to “possess a perspective not only
on his conduct but also on the constitution of the world he belongs to.”25 In
23
It does not matter for Kant’s argument whether individual agents have the establishment of such a
constitution as their end. In fact, the ends of individual agents may be “directly opposed” to such an end
(TP 8:312). Still, Kant claims, progress is an attainable end in the long run; or, as he puts it in “Theory
and Practice,” progress may “be interrupted from time to time but will never be broken off” (TP 8:309).
24
Dieter Henrich, “The Moral Image of the World,” in Aesthetic Judgment and the Moral Image of the
World: Studies in Kant (Stanford: Stanford University Press, 1992), 3–28.
25
Henrich, “Moral Image of the World,” 14.
the broader practical sense that interests us here, the world presents itself to
the agent as constituted in such a way that the ends of reason are realizable.26
Highly relevant for our purposes is Henrich’s discussion of the role of the
moral image. Starting from the basic Kantian thought that “to act always
means to pursue purposes,” Henrich offers two alternative ways of under-
standing the role of a moral image of the world.27 The first is the equivalent
of what we called earlier the psychological explanation of the concept of
history: as Henrich puts it, a moral image is a “helpful fiction” arising from
the agent’s hope for happiness.28 The second is the moral image as consti-
tutive of agency. We are committed to it by virtue of being moral; the moral
image “arises spontaneously from the good will” and is integral to the moral
perspective. This constitutive role is of great interest and importance.
Unfortunately, Henrich’s discussion of it is rather elliptical. Henrich says
that the moral image of the world arises when conscience becomes endan-
gered and needs to obtain help in order to sustain itself: he states explicitly
that the moral image is “a strategic defense for morality’s integrity.”29 The
moral image as strategic defense, which arises spontaneously when conscience
is endangered, looks very close to the helpful fiction version of the moral
image, which arises to encourage and perhaps console the good. Therefore, I
want to put this strategic role to one side and focus on the idea of an image of
the world as a “constitutive component of the moral perspective itself.”30
Adapting it to our present concerns with history, we can say that there is a
cosmopolitan image of the world summoned by the conscience of an agent
fighting for justice, enlightenment, culture, “the better.” Philosophical his-
tory is then simply this cosmopolitan image of the world; it is history from
the perspective of agency.
There are four ways of understanding what such a practical justification of
philosophical history amounts to. First, following Henrich, we can say that
the agent intending or planning or setting about to pursue right and virtue
presupposes that her goals are realizable; hence, an image of the world in
which success is possible is part of her practical attitude. There are grounds to
26
In line with the overall purpose of his essay, Henrich associates the moral image with the belief that “it
is possible to promote a state of the world in which happiness and merit coincide” (Henrich, “Moral
Image of the World,” 25).
27
Henrich, “Moral Image of the World,” 24.
28
Ibid., 25.
29
Ibid., 13.
30
Ibid.
700 K. Deligiorgi
believe that Kant is addressing such agents both individually and collectively,
seeking to instill a sense of collective doing:
I rest my case on my innate duty, the duty of every member of the series of
generations – to which I (as a human being in general) belong . . . so to
influence posterity that it becomes always better (the possibility of this
must, accordingly, also be assumed), and to do it in such a way that this
duty may be legitimately handed down from one member [in the series of]
generations to another. (TP 8:309)
Second, and following from the above, we are tasked to be such agents. Kant
both addresses us as agents and seeks to recruit us to the cause of right and
virtue. As he puts it in the Groundwork:
Teleology considers nature as a kingdom of ends, morals considers a possible

kingdom of ends as a kingdom of nature. In the former the kingdom of
ends is a theoretical idea for explaining what exists. In the latter, it is a
practical idea for the sake of bringing about, in conformity with this very
idea, that which does not exist but which can become real by means of our
conduct. (G 4:436n)
Third, by identifying us as agents he is also saying that we are the agents

of history. History is not just something that happens; it is something
that people do. The desolation that befalls well-intentioned people when
they contemplate all the horrors of history is a desolation of contempla-
tion that can make us forget that we are responsible for bringing about
the better:
it is of the greatest importance to be content with providence (even though

on this earthly world of ours it has marked out such a troublesome road
for us), partly in order to grasp courage even among our toils, and partly so
that by placing responsibility for it on fate, we might not lose sight of our
own responsibility, which perhaps might be the sole cause of all these ills,
and avoid the remedy against them, which consists in self-improvement.
(CB 8:121)
Finally, Kant is himself aware of his own agency as a writer of history.

Stories about what we do and what we can achieve matter “propheti-
cally” because they encourage or discourage us. Prophets are “authors” of
the fate they foretell (CF 7:80). To conclude, then, philosophical – that
is, purposeful, progressively shaped – history provides us with a
cosmopolitan image of the world and is, as such, a constitutive compo-

nent of practical agency.
The philosopher as legislator; or, why should

anyone care for philosophical history?
“To search for ‘unity’ and ‘system,’ at the expense of truth, is not, I take it,
the proper business of philosophy.”31 This quote, from G. E. Moore, is likely
to resonate with students of philosophical history. Kant’s philosophical
history has a theoretical and a practical aspect. It is theoretically a unity
formally as a mere compositional aggregate of human actions. To give a
progressive shape to this whole we need to be persuaded by the cosmopolitan
conception of philosophy; we may call this the practical formal unity of
history. If we now ask whether the latter is true or at least plausible, we can
look at the theoretical claims Kant makes in order to adduce support for his
progressive view of history. These facts about human behavior are, I argued,
plausible. Nonetheless, they are very weakly supporting the progressive view
of history; this is because no facts can verify or falsify a view of history as a
whole. At best, once one has such a view one can look to see what facts there
are that can possibly support it. Here is where the practical perspective
containing a cosmopolitan image of the world becomes crucial. To illustrate
by means of a contrast: in the Conflict of the Faculties, Kant tells the story of a
patient, who, tiring of his doctor’s optimism and reassurances of imminent
recovery, exclaims, “I’m dying of improvement, pure and simple!” (CF 7:93).
Kant does not present himself as a doctor telling his readers stories of
progress. This is because he does not take his readers to be patients.
History is not to be suffered but to be created. What Kant tries to commu-
nicate to his readers is that what matters is the practical standpoint of the
agent, not that of the patient.
We have now reached a reasonably detailed understanding of what Kant’s
writings on history are about. What remains to be discussed are some of the
broader assumptions Kant brings to the writing of philosophical history,
starting with his notion of practical agency. Kant does not anticipate that
some of his readers can extricate themselves from their cosmopolitan tasks by
claiming lack of interest or indisposition. Irrespective of our circumstances,
31
G. E. Moore, Principia Ethica, rev. ed., ed. Thomas Baldwin (Cambridge: Cambridge University
Press, 1993), 270.
702 K. Deligiorgi
we are, each and all, equally addressed by the “ought” of right and virtue,
because it is an objective practical requirement. Objectivism, which Kant
defends in his moral and political works, goes hand in hand with his
commitment to unity and system. This coupling is an expression of Kant’s
practical, critical rationalism, which states that the ends of pure reason in its
practical employment are realizable in the world, even though we have no
reason, as critical thinkers, to believe that the world is pre-ordered in
accordance with such ends. With respect now to these ends of reason,
some, such as the final moral end, are history-transcending, as we saw earlier.
Others, though, are not. These historical, practical ends include thinking
freely and participating in our and our fellow-citizens’ enlightenment, living
justly and peacefully, and cultivating a virtuous disposition – in short, being
better. In his writings on history, Kant provides us with the cosmopolitan
image of the world that is constitutive of the active pursuit of such ends.
In light of the interpretation given here, it would be quite wrong then to
see Kant’s philosophical history as merely a consolatory fiction or as an
example of callous optimism, an instance of the bonum-through-malum or
“good as a result of bad” pattern.32 Rather, it is an integral feature of his
critical philosophy: the justification of providence that Kant offers in his
historical works fulfills in part the promise of the cosmopolitan concept of
philosophy, to show that practical and theoretical aims fit together. This
cosmopolitan concept in turn would not even be conceivable without Kant’s
practical conception of philosophy. This is what he says about the value of
mathematics compared to the value of philosophy “with respect to the
practical”:
The former [i.e., mathematics] is that of technical-practical reason (skill in the

discovery of means for whatever ends), the latter [i.e., philosophy] is moral-
practical reason and is directed to the final end, which is absolutely (categori-
cally) obligatory, namely to create men of improved character [Gesinnung].
(OP 22:545n)
32
See here the illuminating discussion of the publication context of IUH in Amélie Oksenberg Rorty
and James Schmidt, introduction to Kant’s “Idea for a Universal History with a Cosmopolitan Aim,” 4–5.
The bonum-through-malum expression is from Odo Marquard, “Unburdenings: Theodicy Motives in
Modern Philosophy,” in In Defense of the Accidental: Philosophical Studies, trans. Robert M. Wallace
(New York: Oxford University Press, 1991), 22. See also William Rasch, “The Public of the
Intellectuals – from Kant to Lyotard,” in The Impact of German Idealism: The Legacy of Post-Kantian
German Thought, vol. 2: Historical, Social and Political Thought, ed. John Walker (Cambridge:
This conception of philosophy fits the idea of the cosmopolitan concept

of philosophy from the first Critique. The philosopher who has this
conception of the discipline is a “legislator of human reason” (A839/
B867). The philosopher neither uncovers nor imposes a law. The philo-
sopher is a legislator in the sense of serving human reason: “Him alone
we must call the philosopher,” Kant says, who uses all cognitions as tools
“to advance the essential ends of human reason” (A839/B867).
Philosophical history plays a small but significant role in this task. It
has a claim then to our philosophical respect because the cosmopolitan
concept of philosophy is the only one worthy of the name.
31
Becoming Human: Kant’s Philosophy
of Education and Human Nature
Robert B. Louden
The final destiny of the human race is moral perfection, so far as it is

accomplished through human freedom, whereby the human being, in that
case, is capable of the greatest happiness. . . . How, then, are we to seek this
perfection, and from which point is it to be hoped for? From nowhere else but
education.
– Immanuel Kant, Lectures on Ethics (LE 27:470, 471, translation modified)
In the winter semester of 1776–1777, Kant began lecturing on practical

pedagogy at the University of Königsberg. In part because this particular
course arose from a governmental decree intended to improve the level of
education of students entering the university, the obligation to teach it was
rotated among the philosophy faculty.1 As result, Kant taught the course
This essay borrows a few points from my “Becoming Human: Kant and the Philosophy of Education,”
in Kant’s Human Being: Essays on His Theory of Human Nature (New York: Oxford University Press,
2011), 136–49; an earlier version of which appears under the title of “Afterword” in Philosophy of
Education: The Essential Texts, ed. Steven M. Cahn (New York: Routledge, 2009), 281–92.
1
For further discussion of the background of Kant’s pedagogy course, see Karl Vorländer, Immanuel
Kant: Der Mann und das Werk, 2 vols. (Leipzig: Meiner, 1924), 2:226–27; www.manchester.edu/kant/
lectures/lecturesListDiscipline.htm#pedagogy; and LP 9:439–40.
R. B. Louden (*)
Department of Philosophy, University of Southern Maine, Portland, USA
e-mail: louden@maine.edu

706 R. B. Louden
only three additional times: summer semester 1780, winter semester 1783–
1784, and winter semester 1786–1787. If one compares this number to the
number of times he taught other courses (e.g., metaphysics: 53, physical
geography: 49, moral philosophy: 28, anthropology: 24), one might reason-
ably infer that the philosophy of education was at best an extremely minor
interest of Kant’s.
Additionally, Kant’s major text in the philosophy of education – Immanuel
Kant über Pädagogik, edited by his former student Friedrich Theodor Rink and
first published in 1803, one year before Kant’s death (LP 9:437–99) – is
problematic for several reasons. First, Rink’s cavalier manner of treating Kant’s
lecture notes has been uniformly criticized over the years, and the text itself is
repetitive and poorly organized. Did Rink publish all or only some of the notes
that Kant handed over to him? Did he rearrange the original order of those
notes that he did publish, and were any deletions or additions made?
Unfortunately, definitive answers to these questions are not possible, since
the original notes on which Rink’s text is based have not survived. Second, no
student or auditor transcriptions of any of Kant’s pedagogy courses have
surfaced – an odd fact, given both the strong tradition of detailed note-taking
that existed in Prussian universities during this time as well as Kant’s own
popularity as a classroom teacher. Multiple transcriptions of Kant’s classroom
lectures on ethics, metaphysics, logic, anthropology, physical geography, and
still more topics are currently available, and they are valuable resources both for
tracing the development of Kant’s thinking on each topic and for determining
more precisely just what he thought in each area. Finally, as Manfred Kuehn
remarks, the “lectures on pedagogy were not his own idea”2 – if Kant had not
been required to help teach the course, Rink’s edited version of the Lectures on
Pedagogy would not exist. Here as well, someone reflecting on these facts might
reasonably conclude that the philosophy of education is simply not an impor-
tant subject within Kant’s philosophy.
But this conclusion, however reasonable it may appear at first glance, is false.
Although Kant does not go quite as far as John Dewey, who, in a famous
passage in Democracy and Education, claims that “philosophy may even be
defined as the general theory of education,”3 in what follows I will argue that the
“general theory of education” plays a vital and central role within Kant’s
2
Manfred Kuehn, “Kant on Education, Anthropology, and Ethics,” in Kant and Education:
Interpretations and Commentary, ed. Klas Roth and Chris W. Suprenant (New York: Routledge,
2012), 55.
3
John Dewey, The Middle Works of John Dewey, 1899–1924, 15 vols., ed. Jo Ann Boydston
(Carbondale: Southern Illinois University Press, 1980), 9:338.
31 Becoming Human 707
philosophy. Its importance stems largely from the commanding position that
education holds within his theory of human nature. In Kant’s view, education
is fundamentally about the effort to realize our humanity. As he proclaims near
the beginning of the Lectures on Pedagogy: “The human being can only become
human through education. He is nothing except what education makes out of
him” (LP 9:443). We literally create ourselves through education.
Education outside the Lectures on Pedagogy

One concrete sign of the importance of education within Kant’s philosophy is
the repeated attention he gives to pedagogical questions in many of his
own published writings. For instance, two of his major works in moral
philosophy – the Critique of Practical Reason (CPrR 5:1–163) and the
Metaphysics of Morals (MM 6:203–493) – both end with discussions of moral
education. In the Methodenlehre parts of each book (CPrR 5:151–61;
MM 6:477–85), he sketches a multifaceted theory of moral education designed
to help the student see “that he4 himself can think” (MM 6:478) seriously about
morality rather than just memorize rules and formulas given to him by adults
and to “feel the progress of his power of judgment” (CPrR 5:154) in moral
matters by examining case studies drawn from “the biographies of ancient
and modern times” (CPrR 5:154). The “first and most essential instrument”
(MM 6:478; cf. CPrR 5:154; LP 9:490) in Kant’s theory of moral education is a
moral catechism – viz., a summary of basic moral principles presented in
question-and-answer form. But unlike earlier religious catechisms on which it
is partly based, Kant’s catechism is “purely moral” (CPrR 5:154) – i.e., unmixed
with any religious or theological doctrines.
To be sure, religious instruction is also an important part of Kantian moral
education, but it must come after rather than before moral instruction. Moral
principles must be rationally justifiable independently of religious assumptions,
and religion itself must be grounded in morality rather than vice versa:
If religion is not combined with morality, then it becomes nothing more than
currying favor. . . . One must not begin with theology. A religion which is
founded merely on theology can never contain anything moral. In such a
religion one will have only fear on the one hand and intentions and
4
Kant is not a progressive voice on the topic of women’s education. But in one unpublished note, he
does at least assert: “Both sexes must be educated” (Ak 15:558 [R1265]).
708 R. B. Louden
dispositions geared toward reward on the other, resulting merely in a super-

stitious cult. Morality must therefore come first. (LP 9:494–95)
In his insistence on a rationally justifiable moral education that does not rest
on religious or theological assumptions, Kant reflects not only a specific
position concerning the relationship between morality and religion that he
defends at greater length in other writings (“on its own behalf morality in no
way needs religion . . . but is rather self-sufficient by virtue of pure practical
reason” [Rel 6:3]) but also a wider Enlightenment consensus concerning
both the underlying unity of the historical faiths and the conviction that
religion’s proper purpose is moral rather than theological.5
Similarly, in his third Critique (CJ 5:165–485) Kant discusses aesthetic
education in some detail, particularly as it relates to human moral develop-
ment. “The beautiful prepares us to love something, even nature, without
interest” (CJ 5:267; cf. MM 6:443). In learning to appreciate the beauty of
both nature and human-constructed artworks for their own sakes without
being influenced by extraneous considerations of utility or monetary value,
the Kantian autonomy of the aesthetic helps prepare humans “for the exercise
of freedom in morality itself.”6
Another text outside of the Lectures on Pedagogy that bears on Kant’s
philosophy of education is “M. Immanuel Kant’s Announcement of the
Programme of His Lectures for the Winter Semester 1765–1766” (APL
2:303–13). This early work, as the title suggests, is essentially an advertise-
ment for the courses he would teach as a Privatdozent during the winter
semester of 1765–1766, but several key themes in his overall educational
theory are also present. For instance, early in the “Announcement” Kant
proclaims his intention “to make public education more adapted to nature”7
(APL 2:305) and to teach his students’ understanding “not thoughts but
thinking . . . so that in the future it will be capable of walking on its own,
5
For further discussion and references, see Robert B. Louden, The World We Want: How and Why the
Ideals of the Enlightenment Still Elude Us (New York: Oxford University Press, 2007), esp. 15–25.
6
Paul Guyer, “The Dialectic of Disinterestedness: II. Kant and Schiller on Interest in
Disinterestedness,” in Kant and the Experience of Freedom: Essays on Aesthetics and Morality (New
York: Cambridge University Press, 1993), 96.
7
Although this is not quite the “cult of nature” later associated with Romanticism (Howard E. Hugo, ed.,
The Portable Romantic Reader [New York: Viking, 1957], 379ff.], many Enlightenment authors never-
theless did appeal to nature as a positive norm for human conduct. Kant may be borrowing here from
Rousseau and/or Basedow, but it is not merely Enlightenment educational theorists who praised nature.
Leopold Mozart, for instance, in his famous treatise on the principles of violin playing, continually advises
his readers to follow “nature herself” in determining how best to play their instrument (Versuch einer
gründlichen Violinschule [Augsburg: published by the author, 1756], 193; cf. 238).
and doing so without stumbling” (APL 2:306; cf. WE 8:35; A838/B866).

Ludwig Ernst Borowski, one of Kant’s earliest biographers, had an extremely
high opinion of his “Announcement”: “Under this completely unassuming
title, Kant gives his ideas on school and university instruction in a form
which is very much worth reading. For me, this always was and still is one of
his most important writings.”8
And, as one might expect, there is also some discussion of educational
issues in several of the transcriptions from Kant’s many classroom lec-
tures. For instance, the Methodenlehre part of the Dohna-Wundlacken
logic lectures contains some additional material on Kant’s recommended
catechismal method, in which he tries to align it with Socrates’ famous
claim of being merely an intellectual midwife of students’ own ideas
rather than a dogmatic teacher who tries to cram foreign ideas down
their throats9:
Socrates said: I am the midwife of someone else’s thoughts. He derived every-

thing from their concepts, which they were simply not able to develop
suitably. . . . Catechisms ought to be arranged thus – first, derive everything
from reason and on this build positive religion – for man as he now is. (DWL
24:780; cf. Ak 24:599–602)
And, because of the central role that education plays within his theory of
human nature, pedagogical themes are also prominent in many of Kant’s
anthropological works. For instance, his concluding discussion of “The
Character of the Species” (An 7:321–33), variants of which we also find
in many of the lecture transcriptions of this annual anthropology course
dating from the mid-1770s onward (LAn 25:838–43, 1194–1203,
1415–29), stresses the necessity and central role of education in helping
human beings to achieve their “final destiny [Endbestimmung]” (LAn
25:1429). As he proclaims at the end of the Mrongovius transcription:
“In the end we will perhaps see that, concerning the well-being of the
world, everything [alles] depends on education” (LAn 25:1428). The
central position of education in helping humans to achieve their
Bestimmung – a crucial point that I will return to later – is also reflected
8
Ludwig Ernst Borowski, Darstellung des Lebens und Charakters Immanuel Kants (1804), in Immanuel
Kant: Sein Leben in Darstellungen von Zeitgenossen; Die Biographen von L. E. Borowski, R. B. Jachmann
und A. Ch. Wasianski, ed. Felix Groß (Berlin: Deutsche Bibliothek, 1912), 32.
9
See Plato, Theatetus, 149a–151d.
710 R. B. Louden
in Kant’s most famous summary of pragmatic anthropology toward the

end of his Anthropology from a Pragmatic Point of View (1798):
The sum total of pragmatic anthropology, in respect to the vocation

[Bestimmung] of the human being and the characteristic of his formation, is
the following. The human being is destined [bestimmt] by his reason to live in a
society with human beings and in it to cultivate himself, to civilize himself, and
to moralize himself by means of the arts and sciences. . . . The human being
must therefore be educated to the good; but he who is to educate him is on the
other hand a human being who still lies in the crudity of nature and who is now
supposed to bring about what he himself needs. Hence the continuous devia-
tion from his destiny [Bestimmung] with the always repeated returns to it. (An
7:324–25)
In several of the anthropology transcriptions, Kant also explores different

hypotheses concerning the role of educational reform in creating a better civil
society where all human talents will develop appropriately. Should we start
by improving schools and universities, or should efforts in this area com-
mence only after political and legal reforms have proven successful? In
Friedländer (1775–1776), for instance, he argues that “the education of
every single human being should constitute the beginning” of civil reform,
“for the education of one human being educates many other human beings,
who in turn educate others” (LAn 25:691). However, later in Menschenkunde
(1781–1782), he suggests that meaningful educational reform itself presup-
poses political reform:
But whether a more perfect civil constitution will not some day with time come
into being cannot be hoped for until human beings, and their education, have
improved; however, this improvement does not appear to be able to happen
until governments themselves become better. (LAn 25:1202)
These multiple discussions within the anthropology lectures concerning

the fundamental importance of education have led Kuehn to claim that
“it is the anthropology (and its connection with ethics) that provides ‘the
unifying thought of Kant’s educational theory,’ and not so much his
philosophy of history, as . . . Beck . . . suggests.”10 Kuehn is responding
here to Lewis White Beck’s earlier argument that Kant’s philosophy of
history writings provide the key to his educational theory. According to
10
Kuehn, “Kant on Education, Anthropology, and Ethics,” 57.
Beck, “Kant’s philosophy of history is a much more important base for

his educational theory than his epistemology and Anthropology is.”11
Beck’s main point is that the three primary educational stages of an
individual’s life discussed in the Lectures on Pedagogy – care, discipline,
and formation or cultivation (Bildung) – correspond to the three stages
of the history of the world that Kant discusses at greater length in several
of his philosophy of history essays: education thus recapitulates history.
And Beck is surely right about this. For instance, the first stage, care –
“the precaution of the parents that children not make any harmful use of
their powers” (LP 9:441) – corresponds to the earliest stage of human
existence, when (Kant conjectures) our distant ancestors were merely
instinctual creatures who unreflectively followed the promptings of
their natural inclinations and desires and had not yet become conscious
of “reason as a faculty that can extend itself beyond the limits within
which all animals are held” (CB 8:111–12). And the second stage,
discipline, which “changes animal nature into human nature” (LP
9:441), corresponds to civilization, the stage of world history where
humans have begun “the transition from the crudity of a merely animal
creature into humanity, from the go-cart of instinct to the guidance of
reason” (CB 8:115). Humans at this second stage have begun to replace
the rule of instinct with the rule of law.
However, Beck’s essay was first published in 1978, and, as Kuehn con-
cedes in an endnote, “Beck did not . . . have available to him the two volumes
of the Academy edition” that contain Kant’s lectures on anthropology and
which were first published in German in 1997.12 Reinhard Brandt and
Werner Stark, coeditors of these lectures, argue in their introduction that
Kant’s philosophy of history is itself “a component of anthropology” rather
than vice versa,13 and at present most scholars accept this claim. As Thomas
Sturm notes: “the origin of most of Kant’s assumptions concerning the
historical development of humanity . . . lie in his anthropology lectures.”14
But the question of who is ultimately the victor in this particular scholarly
debate is not crucial for our own purposes at present. For all parties in the
11
Lewis White Beck, “Kant on Education,” in Essays on Kant and Hume (New Haven: Yale University
Press, 1978), 197.
12
Kuehn, “Kant on Education, Anthropology, and Ethics,” 67n10. English translations of some of this
material are available in Immanuel Kant, Lectures on Anthropology, ed. Allen W. Wood and Robert B.
Louden (Cambridge: Cambridge University Press, 2011).
13
Reinhard Brandt and Werner Stark, “Einleitung,” in Vorlesungen zur Anthropologie by Immanuel
Kant, ed. Reinhard Brandt and Werner Stark (Berlin: de Gruyter, 1997), liii.
14
Thomas Sturm, Kant und die Wissenschaften vom Menschen (Paderborn: Mentis, 2009), 355.
712 R. B. Louden
dispute acknowledge that Kant repeatedly stresses the importance of educa-

tion in many of his “noneducational” works.15 And this is our main point:
Kant explores educational issues not only in the Lectures on Pedagogy but in
many other texts as well, and this is one clear sign that education is not
merely a peripheral issue for him.
An age of education
A second reason for surmising that pedagogy might be more than merely a
secondary interest for Kant is that “education was a topic that held a central
place in the concerns of eighteenth-century philosophy.”16 Indeed, as Kant
himself emphasizes, the very term “Enlightenment,” particularly when
applied to society at large rather than a few lucky individuals, implies a
long, complex process of education: “to enlighten an age is very slow and
arduous; for there are external obstacles which in part forbid this manner of
education and in part make it more difficult” (OT 8:146n).
Locke’s Some Thoughts Concerning Education (1693), with its confident
assertion of human malleability (“of all the men we meet with, nine parts of
ten are what they are, good or evil, useful or not, by their education.’Tis that
which makes the great difference in mankind”),17 and Rousseau’s Émile
(1762), in its call for a child-centered education that takes its cue from
nature rather than tradition (“Respect childhood, and do not hurry to judge
it. . . . Leave nature to act for a long time before you get involved with acting
in its place, lest you impede its operations”),18 are the two most influential
Enlightenment texts on education. But in Kant’s case it was the German
educational reformer Johann Bernhard Basedow (1724–1790) who proved
to be the biggest single influence. While Basedow’s educational theory does
owe some debts to both Locke and Rousseau, he seems to have arrived at his
key ideas independently, through his personal experience as a tutor for the
15
One anthology that conveniently gathers together most of the above-discussed texts is Immanuel
Kant, Ausgewählte Schriften zur Pädagogik und ihrer Begründung, 2nd ed., ed. Hans-Hermann Groothoff
(Paderborn: Schöningh, 1982).
16
Geraint Parry, “Education,” in The Cambridge History of Eighteenth-Century Philosophy, ed. Knud
Haakonssen, 2 vols. (Cambridge: Cambridge University Press, 2006), 1:608.
17
John Locke, Some Thoughts Concerning Education and of the Conduct of the Understanding, ed. Ruth
W. Grant and Nathan Tarcov (Indianapolis: Hackett, 1996), §1.
18
Jean-Jacques Rousseau, Émile; or On Education, trans. Allan Bloom (New York: Basic, 1979), 107;
cf. 79, 90, 92. For general discussion of Enlightenment views about education, see Louden, World We
Want, 27–50.
son of Josias von Quaalen in Borghorst, Holstein (formerly part of Denmark;

at present, Germany) in 1749–1752. Here Basedow worked out an educa-
tional method stressing learning through play, the importance of everyday
objects (“realia” – pictures, models, maps, natural history collections) as aids
to instruction, a conversation-based approach to foreign language study,
science-based sex education, and nonsectarian religious guidance, the details
of which he first elaborates on in his Magister thesis, submitted to the
University of Kiel in 1752.19 Additional influential facets of his educational
method, including an emphasis on physical education, the importance of
educating rich and poor children together, less stress on memorization and
physical punishment, and the need for manual and vocational training for all
social classes, were presented later in an extensive series list of publications.20
Kant used one of Basedow’s best-known books, Das Methodenbuch für
Väter und Mütter der Familien und Völker (Altona and Bremen, 1770), as a
text the first time he offered his pedagogy course in winter semester
1776–1777. From 1780 on – at the behest of his superiors – he used his
colleague Friedrich Samuel Bock’s book, Lehrbuch der Erziehungskunst zum
Gebrauch für christliche Eltern und künftige Jugendlehrer (Königsberg and
Leipzig, 1780). Although the structure of the Lectures on Pedagogy does not
follow Basedow’s book at all closely, much of Kant’s specific advice to
teachers clearly stems from Basedow: for instance, “From earliest childhood
the child must be allowed to be free in all matters (except in those where it
might injure himself . . . )” (LP 9:454; cf. 459); “One can either learn
[languages] by formal memorization or by social intercourse [Umgang], and
with living languages the latter is the best method” (LP 9:472–73); “One
must . . . teach some concepts of the highest being to children, so that, when
they see others praying and so forth, they may know to whom and why such
is being done. But these concepts must be only few in number” (LP 9:496).
However, unlike Locke and Rousseau, Basedow was not merely a
theorist of educational reform but also a practitioner. With the support
of his patron, Prince Leopold III Friedrich Franz, Basedow opened a
nonsectarian, experimental school in Dessau on December 27 (the
19
Johann Bernhard Basedow, Inusitata et optima honestioris iuventutis erudiendae methodus [On the best,
but previously unused, method of instructing the children of the upper classes], Magister thesis,
University of Kiel, 1752. For discussion, see Hugo Göring, “Basedow als Privaterzieher,” in J. B.
Basedow’s Ausgewählte Schriften by Johann Bernhard Basedow, ed. Hugo Göring (Langensalza: Beyer
& Söhne, 1880), xxx–xxxiii.
20
Albert Reble lists 108 different works in his bibliography of Basedow’s writings, while noting that
Basedow’s publications contain “frequent repetitions.” See Johann Bernhard Basedow, Ausgewählte
pädagogische Schriften, ed. Albert Reble (Paderborn: Schöningh, 1965), 253, 267–74.
714 R. B. Louden
Prince’s birthday), 1774. Emphasizing not only the new methods of

instruction first worked out in his employment as a private tutor in
Borghorst but above all “care of the natural germ of love of humanity,
virtue, and innocent contentedness,”21 Basedow’s Philanthropin Institute
instantly captured the attention of many liberal intellectuals, including
Kant. In 1776–1777 Kant published two short essays in support of the
new school in the Königsbergische gelehrte und politische Zeitung (EP
2:445–52), praising Basedow’s experiment for the “entirely new order
of human affairs” (EP 2:447) commencing in it and for being the first
educational institution designed in accordance with “nature itself and not
slavishly copied from old habit” (EP 2:449; cf. 447). Working as an
unofficial fundraiser for the school (“Local subscriptions, against receipt,
are given out at Prof. Kant’s in the morning hours from 10 until . . . around
1 o’clock in the afternoon” [EP 2:452]) – an uncomfortable role that he was
eventually able to free himself from in 1778 – he expresses still stronger
admiration for the experimental school at the conclusion of the Friedländer
lecture on anthropology:
The present-day Basedowian institutes are the first that have come about
according to the perfect plan of education. This is the greatest phenomenon
which has appeared in this century for the improvement of the perfection of
humanity. Through it all schools in the world will receive another form, and
the human race will thereby be freed from the constraints of the prevailing
schools. (LAn 25:722–23, translation modified; cf. LE 27:471; LP 9:451)22
Despite the strong support that Kant and many others lavished on Basedow’s
Philanthropin, the school was by no means an unqualified success. It was
subject to persistent enrollment and financial problems, and part of the
blame must be placed on Basedow himself, who, according to the testimony
of many, had an abrasive personality and was not well-suited for adminis-
trative work. For instance, Goethe, in Book 14 of his Dichtung und Wahrheit,
notes that although Basedow’s “great intellectual gifts were admired, he was
not the man, either to edify souls or to lead them,” and while acknowledging
that Basedow could speak “in a lofty and convincing way of his plans” for the
21
Johann Bernhard Basedow, Das in Dessau errichtete Philanthropinum, eine Schule der
Menschenfreundshaft (Leipzig: Crusius, 1774); reprinted in Basedow, Ausgewählte pädagogische
Schriften, 220.
22
For further discussion of Kant and Basedow, see Robert B. Louden, “‘Not a Slow Reform, but a Swift
Revolution’: Kant and Basedow on the Need to Transform Education,” in Kant and Education, 39–54.
new school, his fundraising efforts were frequently hampered by his rude
manners and “the most incomprehensible way he injured the feelings of the
men whose contributions he wished to gain: indeed, he offended them
unnecessarily because he could not suppress his opinions and silly ideas on
religious topics.”23 Similarly, Hamann, in a letter to Herder, after confessing
that he feared he could “do nothing” for his son’s education, added: “I had,
one Sunday, the horrid idea of packing him off, neck and heels, to the
Pontifex Maximus, at Dessau,” to which Herder replied: “as concerns
Basedow, whom I know personally, I would not give him any calves to
educate, much less human beings.”24 And Kant, in a 1791–1792 anthro-
pology transcription, is recorded as saying: “Basedow’s shortcoming was that
he drank too much Malaga” (Ak 25:1538; cf. 1561). Even Basedow himself
confessed in a 1783 autobiographical text that “in business I am or was rash
and argumentative, and in social intercourse often sorrowful and melan-
choly.”25 Despite (and perhaps also because of) significant turnover in the
directorship position at the school, the Philanthropin finally closed in 1793.
But in other respects Basedow’s experiment at Dessau was an outstanding
success. For instance, it served as a highly influential teacher-training insti-
tute. German schoolteachers at this time often received no formal training
at all, and many instructors who initially worked with Basedow in Dessau
later went on to found their own Philanthropin-influenced schools
elsewhere – e.g., Joachim Heinrich Campe in Hamburg, Carl Friedrich
Bahrdt in Marschlins, Switzerland, and Christian Gotthilf Salzmann in
Schnepfenthal. According to historian Frederick Hertz’s estimate, “by the
year 1790 63 other schools of this type had been founded in Germany”
alone.26 The teachers who were trained in Dessau helped to popularize
Basedow’s novel educational methods, and since the late eighteenth century
many of his ideas have worked their way into mainstream educational
curricula all over the world. Kant himself offers the following sober assess-
ment in his Lectures on Pedagogy:
23
Johann Wolfgang Goethe, Dichtung und Wahrheit; quoted in Basedow, Ausgewählte pädagogosiche
Schriften, 239, 240.
24
Johann Georg Hamann, Briefwechsel, 6 vols., ed. Walther Ziesemer and Arthur Henkel (Wiesbaden:
Insel, 1957), 3:236, 251.
25
Johann Bernhard Basedow, Etwas aus dem Archiv der Basedowischen Lebensbeschreibung von ihm selbst
(Leipzig: Crusius, 1783); quoted in Auguste Pinloche, Geschichte der Philanthropinismus (Leipzig:
Brandstetter, 1896), 157n1.
26
Frederick Hertz, The Development of the German Public Mind: A Social History of German Political
Sentiments, Aspirations, and Ideas, vol. 2: The Age of Enlightenment (London: Allen & Unwin, 1962),
385.
716 R. B. Louden
The only experimental school which to an extent made a beginning in estab-

lishing a course was the Dessau institute. We must let it keep this glory
regardless of the many mistakes of which one could accuse it (mistakes found
in all conclusions that come from experiments) – viz., that new experiments are
always required. It was in a certain sense the only school in which teachers had
the freedom to work according to their own methods and plans, and where
they were in contact with each other as well as with all scholars in Germany.
(LP 9:451; cf. LP 9:467; LAn 25:723)
One downside of situating Kant’s education writings within the larger

Enlightenment concern for educational reform is that this part of his philo-
sophy begins to look derivative and less innovative than his better-known
trailblazing efforts in epistemology, metaphysics, and ethics. In articulating
his philosophy of education, is Kant simply an energetic chorus member of
an overly optimistic zeitgeist whose motto was “education can do all”?27
There is a grain of truth to this criticism, but it is not as devastating as
commentators suggest. Yes, Kant wrote during an era when many of his peers
shared transformative ambitions for education. But the contours of his own
philosophy of education differ in several key respects from both Basedow’s
program as well as the educational theories of other well-known
Enlightenment educational reformers such as Locke and Rousseau.
First, as Georg Cavallar has emphasized, “unlike Basedow’s theory, Kant’s
cosmopolitan education is embedded in a comprehensive political theory, a
philosophy of history and a philosophy of religion.”28 Educational reform, in
Kant’s view, plays a necessary and important role in helping human beings to
achieve their destiny (Bestimmung), but it is not a stand-alone project. In
order to be truly effective, educational reform must occur together with legal,
political, economic, and religious reform, all of which jointly contribute to
the moral transformation of the human species. Kantian education, so to
speak, comes wrapped in a cosmopolitan philosophy of history.
Second, as noted earlier, Kant’s educational theory flows from, and is
presupposed by, his anthropology, the most relevant part of which concerns
his position on the character of the human species and its historical devel-
opment. Again, the primary importance of Kant’s philosophy of education
lies in its connection to his theory of human nature. In Kant’s view, we need
27
Claude-Adrien Helvétius, De l’homme, de ses facultés intellectuelles et de son education, 2 vols. (London:
Société Typographique, 1773), 2:334.
28
Georg Cavallar, “Sources of Kant’s Cosmopolitanism: Basedow, Rousseau, and Cosmopolitan
Education,” Studies in Philosophy and Education 33, no. 4 (July 2014): 374.
education in order to become human and to achieve our humanity: “The

human being can only become human through education” (LP 9:443).
A third distinguishing feature of Kant’s educational theory is what I have
elsewhere called its “species perfectionism.” Other Enlightenment educa-
tional reformers focus primarily on promoting happiness and skillfulness in
individual students. But an additional distinctive feature of Kant’s position
“is what may be termed its species perfectionism, that is, its claim that the
ultimate objective of education is to advance not the welfare of the individual
student but rather the moral perfection of the human species as a whole.”29
Each generation depends on the achievements of the preceding one, and
“this course, which for the species is a progress from worse toward better, is
not the same for the individual” (CB 8:115; cf. LP 9:445). Reason does not
operate instinctively in the human species, “but rather needs attempts,
practice and instruction in order gradually to progress from one stage of
insight to the next” (IUH 8:19).
Finally (though I view this last difference more as one of degree than of
kind), Kant insists on a clearer separation between morality and religion than
Basedow and other Enlightenment educational reformers. For instance, in his
discussion of moral instruction in Book Five the Methodenbuch, Basedow starts
by endorsing the biblical injunction that “fear of the Lord is the beginning of
wisdom.”30 But as Cavallar notes, this implies that in Basedow’s view morality
is not autonomous and that humans can “only be motivated by external
incentives.”31 Kant, on the other hand, as we saw earlier, holds that “morality
must . . . come first, theology then follow” (LP 9:495; cf. Rel 6:3).
The destiny of the human species

I asserted earlier that the chief import of Kant’s philosophy of education lies in
the commanding position it occupies within his theory of human nature32:
“we animal creatures are made into human beings only by education”
29
Robert B. Louden, Kant’s Impure Ethics: From Rational Beings to Human Beings (New York: Oxford
University Press, 2000), 37. Cf. Cavallar, “Sources of Kant’s Cosmopolitanism,” 375.
30
Basedow, Das Methodenbuch, in Ausgewählte pädagogische Schriften, 119. Cf. Proverbs 1:7, 9:10;
Psalms 111:10.
31
Cavallar, “Sources of Kant’s Cosmopolitanism,” 374.
32
In this section I have borrowed a few points from my “Cosmopolitical Unity: The Final Destiny of
the Human Species,” in Kant’s “Lectures on Anthropology”: A Critical Guide, ed. Alix Cohen (Cambridge:
718 R. B. Louden
(EP 2:449; see also LP 9:443). On Kant’s view, human beings need education
in order to become fully human. Indeed, he holds not only that educability is
“an essential human trait,”33 but also a uniquely human trait: “The human
being is the only creature [das einzige Geschöpf] that must be educated” (LP
9:441). It is time now to make good on these claims.
First, Kant clearly subscribes to the view that there is a human nature – viz., a set
of common characteristics shared by all normal members of the human species in
different times and places. He differs here from Sartre and others who hold that
“there is no human nature. . . . Man is nothing but that which he makes of
himself.”34 Kant’s commitment to the claim that there is a human nature also
puts him in opposition to Foucault and other historicists who hold that “man is an
invention of recent date.”35 On Kant’s view, man has existed for a very long time.
At the same time, Kant’s theory of human nature is deeply historical. He
acknowledges that human life has changed profoundly over time, but his account
of human nature is also one that seeks to explain why change has occurred.
Because humans are animals, part of Kant’s theory of human nature
overlaps with his larger philosophy of biology. One key aspect of the latter
is a commitment to teleological explanation. All natural organisms, including
humans, are understood as having inherent goals and purposes: “Nothing in
[nature] is in vain, purposeless, or to be ascribed to a blind mechanism of
nature” (CJ 5:376). At bottom, this appeal to teleological explanation in the
study of nature is reminiscent of Aristotelian final causes. Like Aristotle, Kant
urges us to explain natural phenomena in part by inquiring into the “end or
that for the sake of which a thing is done.”36 And once we remind ourselves
that one of the key developments in modern science is Descartes’s banish-
ment of final causes in the explanation of physical change,37 Kant’s
33
34
Jean-Paul Sartre, “Existentialism Is a Humanism,” in Existentialism from Dostoevsky to Sartre, ed.
Walter Kaufmann (Cleveland: Meridian, 1956), 290–91. “Even in Kant,” Sartre notes correctly, “Man
possesses a human nature” (290). I am borrowing a few points here from my discussion of the rudiments
of Kant’s theory of human nature in the introduction to Kant’s Human Being, xvii–xxviii.
35
Michel Foucault, The Order of Things: An Archaeology of the Human Sciences, trans. Alan Sheridan
(New York: Vintage, 1973), 387. Foucault’s critique of the human sciences is in part an outgrowth of
his early work on Kant’s anthropology. See Michel Foucault, Introduction to Kant’s “Anthropology,” ed.
Roberto Nigro, trans. Roberto Nigro and Kate Briggs (Los Angeles: Semiotext(e), 2008). For discussion
of Foucault’s interpretation of Kant’s anthropology, see Robert B. Louden, “El Kant de Foucault,”
Estudos Kantianos 1, no. 1 (Jan./June 2013): 163–80.
36
Aristotle, Physics II.3, 194b32; cf. Aristotle, Parts of Animals I.1, 639b12–21.
37
“I consider the customary search for final causes to be totally useless in physics: there is considerable
rashness in thinking myself capable of investigating the <impenetrable> purposes of God” (René
Descartes, Meditations on First Philosophy, in The Philosophical Writings of Descartes, trans. John
teleological turn begins to look like a retrograde step. However, Kant’s

teleology is much milder than Aristotle’s:
It is self-evident that this is not a principle for the determining but only for the
reflecting power of judgment, that it is regulative and not constitutive, and that
by its means we acquire only a guideline for considering things in nature, in
relation to a determining ground that is already given, . . . without harm to the
mechanism of nature. (CJ 5:379; cf. 360–61)
Nevertheless, this merely regulative “guideline for considering things in

nature” does play a major role in Kant’s overall philosophy of biology as
well as in his theory of human nature. The key concept here is the
German term Bestimmung, which in English can be translated variously
as “destiny,” “vocation,” and “determination.” Kant relies heavily on the
assumption that every living creature can be explained in terms of its
Bestimmung: “one can assume as a principle that nature wants every
creature to reach its destiny [Bestimmung] through the appropriate devel-
opment of all predispositions [Anlagen] of its nature” (An 7:329); and
“Every creature surely reaches his destiny [Bestimmung] in the end, i.e.,
reaches the time in which all of its natural predispositions [Naturanlagen]
are developed and come to maturity” (LAn 25:1417–18). Small wonder
then, that the concept of Bestimmung has been called “the leading center
of Kantian philosophy.”38
All living creatures are to be viewed as having a natural Bestimmung which
is unique to their species. Briefly, the human Bestimmung differs from that of
other animals in the following key ways. First, as noted earlier, in the case of
Homo sapiens only the species as a whole reaches the Bestimmung, not the
individual members:
it must be noted that with all other animals left to themselves, each
individual reaches its complete destiny [Bestimmung]; however with the
human being only the species, at best, reaches it; so that the human race
can work its way up to its destiny [Bestimmung] only through progress in a
series of innumerably many generations. (An 7:324; cf. An 7:329; LP
9:445)
Cottingham, Robert Stoothoff, and Dugald Murdoch, vol. 2 (Cambridge: Cambridge University Press,
1984), 39.
38
Reinhard Brandt, Die Bestimmung des Menschen bei Kant (Hamburg: Meiner, 2007), 7.
720 R. B. Louden
Part of the significance of education within Kant’s theory of human nature

lies in the fact that the transmission of knowledge between generations is
necessary for the species to “work its way up to its Bestimmung.”
Second, and related, there are the issues of conscious effort and free choice.
In the case of humans, both are necessary in order to reach their Bestimmung;
in the case of other terrestrial living creatures,39 not so: “Animals fulfill their
vocation [Bestimmung] automatically [von selbst] and unknowingly. The
human being must first seek to reach his, but this cannot happen if he
does not even have a concept of his vocation [Bestimmung]” (LP 9:445). Here
a slightly different aspect of the significance of education in human life comes
out. It is the “seek[ing] to reach” the Bestimmung that is uniquely human
among the terrestrial living creatures. Humans need to be aware both of what
their Bestimmung is and how to reach it. In line with many contemporary
researchers, Kant acknowledges that some other animals do have the ability
to transmit behavioral patterns through social and observational learning.
For instance, at one point in the Lectures on Pedagogy he notes that “birds do
not sing by instinct, but actually learn [wirklich lernen] it, . . . each species of
bird, through all the generations, retains a certain main song; and the
tradition of the song may well be the most faithful in the world” (LP
9:443). The tradition of each bird song, he adds elsewhere, is “imparted
through instruction [durch Belehrung]” (An 7:323n). However, unlike con-
temporary authors who are quick to move from observations concerning
such “social transmission mechanisms” to claims about nonhuman animals
possessing culture and education,40 Kant reserves the latter terms for activ-
ities stemming from the capacity to set ends for oneself and to choose freely
from among these ends.
The strong assumption of freedom in Kant’s conceptions of culture and
education also implies that his position on human progress is not as rigid and
naive as that of other Enlightenment authors. On Kant’s view, humans are by
39
But what about extraterrestrial rational beings? As Peter Szendy notes, Kant “regularly summoned
inhabitants of other planets, inviting them over and over again into his discourse” throughout his
writing career (Kant in the Land of Extraterrestrials: Cosmopolitical Philosofictions, trans. Will Bishop
[New York: Fordham University Press, 2013], 45). See also my review of Szendy’s book in Philosophy in
Review 34, no. 6 (Dec. 2014): 339–41. Like many Enlightenment intellectuals, Kant clearly believes in
the existence of extraterrestrial rational beings, but – at least in his mature works – his evident
extraterrestrial enthusiasms stop short of speculations concerning the specific Bestimmungen of other
types of rational beings. (See UNH 1:351–67 for some pre-critical speculations on this theme.)
40
See, e.g., A. Whiten et al., “Cultures in Chimpanzees,” Nature 399 (17 June 1999): 682–85; Frans de
Waal, The Ape and the Sushi Master: Cultural Reflections of a Primatologist (New York: Basic, 2001); and
C. K. Catchpole and P. J. B. Slater, Bird Song: Biological Themes and Variations, 2nd ed. (Cambridge:
Cambridge University Press, 2008).
no means causally determined to achieve unilinear cultural progress. Rather, we

pursue progress as free beings who can and do change our minds. Cultural
regress and nonprogressive cultural change are thus always present as distinct
possibilities. As Kant notes in the Conflict of the Faculties (1798):
Even if we felt that the human race, considered as a whole, was to be conceived
as progressing and proceeding forward for however long a time, still no one can
guarantee that now, this very moment, with regard to the physical disposition
of our species, the epoch of its decline would not be liable to occur. . . . For we
are dealing with beings that act freely, to whom, it is true, what they ought to
do may be dictated in advance, but of whom it may not be predicted what they
will do. (CF 7:83)
Also, because the human being “has a character, which he himself creates,
insofar as he is capable of perfecting himself according to ends that he himself
adopts” (An 7:321), there is an inherent flexibility in human nature that is
missing in other animal species. Our mode of life is radically indeterminate –
open rather than fixed – in comparison with that of other terrestrial animals.
Indeed, part of the human species’ destiny concerns the proper realization of
its inherent capacity for freedom:
All animals have the capacity to use their powers according to choice. Yet this
choice is not free, but necessitated by incentives and stimuli. Their actions
contain bruta necessitas. If all creatures had such a choice, tied to sensory drives,
the world would have no value. But the inner worth of the world, the summum
bonum, is freedom according to a choice that is not necessitated to act.
Freedom is thus the inner worth of the world. (LE 27:344; cf. 1482)
At bottom, the Bestimmung of each species is a function of its inherent

capacities, or what Kant calls its Keime (seeds) and Anlagen (predispositions).
The destiny of each species is to properly actualize its good seeds and
predispositions,41 and once we correctly identify these capacities we will
also know what the species’ destiny consists of. But – at least in the case of
Homo sapiens – these inherent capacities do not simply actualize themselves.
Rather, certain favorable environmental factors are necessary, the most
important of which is education. (This is one reason why English translators
frequently render the term Bestimmung as “vocation” rather than “destiny”
41
The bad seeds and predispositions (e.g., within the human species, selfishness) need to be restrained
and controlled.
722 R. B. Louden
when Kant applies it to human beings. Our own conscious effort is needed to
realize our Bestimmung. But because our seeds and predispositions are
genetically inherited, both meanings are implied in Kant’s usage.) Like
Leibniz’s famous example of a “block of veined marble” which requires
“some labor for these veins to be exposed and polished into clarity,”42 our
own inherent capacities also require some labor in order to be polished into
clarity. And education is the primary agent of this necessary labor.
These two key features of Kant’s philosophy of biology – the teleology in his
frequent use of Bestimmung and the quasi-innatism implied by his use of the
terms Keime and Anlagen – are on ample display in the opening pages of the
Lectures on Pedagogy: “The human species is supposed to bring out, little by
little, humanity’s entire natural predisposition [Naturanlage] by means of its
own effort” (LP 9:441). And again, a few pages later: “Many germs [Keime] lie
within humanity, and now it is our business to develop the natural dispositions
[Naturanlagen] proportionally and to unfold humanity from its germs [Keime]
and to make it happen that the human being reaches his vocation
[Bestimmung]” (LP 9:445). The “our” refers above all to teachers. Their
vocation is to help the human species achieve its collective “destiny
[Bestimmung]: humanity” (LP 9:442) – viz., the actualization of its inherent
potential.
However, not all humans are aware of the details of their species’ specific
Bestimmung, and this too is part of the role of education. In a marginal note
to the manuscript (Handschrift) for Anthropology from a Pragmatic Point of
View, Kant writes: “there is a cosmopolitical predisposition [cosmopolitische
Anlage] in the human species, even with all the wars, which gradually in the
course of political matters wins the upper hand over the selfish predisposi-
tions of people” (An 7:412, translation modified). Because we are not all
aware of our species’ predisposition toward “cosmopolitical unity” (see An
7:333), educators must emphasize it in the moral education of students. On
the final page of the Lectures on Pedagogy, Kant notes:
One must also stress to the student love of humanity [Menschenliebe] toward
others and then also cosmopolitan dispositions. . . . An interest in the best for
the world [Weltbesten] must come to pass. One must make children familiar
with this interest so that they may warm their souls with it. They must rejoice
in the best for the world [Weltbeste] even if it is not to the advantage of their
fatherland or to their own gain. (LP 9:499, translation modified)
42
G. W. Leibniz, “Preface to the New Essays,” in Philosophical Essays, trans. Roger Ariew and Daniel
Garber (Indianapolis: Hackett, 1989), 294.
This strong cosmopolitan orientation of Kantian education is particularly

noteworthy. Education everywhere must aim not at national or parental goals
(both of which are usually vocational in nature, both of which treat students
as tools rather than as ends in themselves), but at human perfection. Students
must learn to become citizens of the world, and to respect the inherent
dignity and moral equality of all human beings. As Kant emphasizes earlier in
the Lectures:
Parents usually care only that their children get on well in the world,
and . . . princes regard their subjects merely as instruments for their own designs.
Parents care for the home, princes for the state. Neither have as their final
end the best for the world [Weltbeste] and the perfection to which humanity is
destined [bestimmt], and for which it also has the predisposition [Anlage].
However, the design for a plan of education must be made in a cosmopolitan
manner. (LP 9:448)
Revolution, experimentalism, naturalism

Although the dominant role of education in helping human beings to
become truly human is clearly the most significant element of Kant’s philo-
sophy of education, several noteworthy secondary features also deserve
comment.
Revolution. Kant agrees with Basedow that existing educational institu-
tions must undergo a “total transformation”43 rather than a piecemeal
reform. For instance, in his second essay in support of the Philanthropin
he declares: present-day schools “must be transformed if something good
is to come out of them because they are defective in their original
organization, and even the teachers must acquire a new formation [or
education (Bildung)]. Not a slow reform, but a swift revolution can bring
this about” (EP 2:449). Total transformation is necessary both because
existing schools are “defective in their original organization” and because
“everything in them works against nature” (EP 2:449). Humans’ natural
capacities are not appropriately developed in present-day educational
institutions: “With the present education the human being does not
fully reach the purpose of his existence” – instead, a “plan of an
43
Johann Bernhard Basedow, Vorstellung an Menschenfreunde (Hamburg: s.n., 1768); in Basedow,
Ausgewählte pädagogische Schriften, 28.
724 R. B. Louden
education more suited to the human being’s purpose” must be developed

(LP 9:445). Kant singles out Basedow’s school for praise precisely because
it the “one school which is established in a radically new way according to
the genuine method” (EP 2:449). Kant’s denial of the right to revolution
in the political realm (see MM 6:320–22) and his less-than-progressive
views on gender, race, and ethnicity continue to be a sore point for liberal
Kantian commentators. But in the sphere of education he speaks in an
unequivocally radical voice.
Naturalism. The Enlightenment appeal to nature as a positive norm was
touched on earlier (see note 7), but it deserves special comment here.
Contemporary readers often recoil at the antinaturalist tone of Kant’s phi-
losophy, particularly his moral philosophy, and it is true that on his view the
moral ought
expresses a species of necessity and a connection with grounds which does not
occur anywhere else in the whole of nature. In nature the understanding can
cognize only what exists, or has been, or will be. It is impossible that some-
thing in it ought to be other than what, in all these time-relations, it in fact is;
indeed, the ought, if one has merely the course of nature before one’s eyes, has
no significance whatever. (A547/B575; cf. G 4:389)
However, at least within his philosophy of education, Kant does espouse a

kind of naturalism. Part of his enthusiasm for Basedow’s Philanthropin stems
from the fact that its “educational method . . . is wisely derived from nature
itself [aus der Natur selbst] and not slavishly copied from old habit and
unexperienced ages” (EP 2:449).
Experimentalism. Like other philosophers in the German Idealist tradition,
Kant normally espouses an extremely top-heavy relationship between theory
and practice. Reason and theory must guide practice, not vice versa, and reason
must always “take the lead with principles for its judgments according to
constant laws and compel nature to answer its questions, rather than letting
nature guide its movements” (Bxiii; cf. TP 8:277). But another pleasant
surprise in Kant’s philosophy of education is its strong advocacy of experi-
mentalism. Experiments are always needed in education, and here at least
theory must take its cue from experience rather than the other way around. In
defending Basedow’s school against its conservative critics, Kant writes:
It is even commonly imagined that experiments in education are not necessary,

and that one can already judge according to reason whether something will be
good or bad. But this is very mistaken, and experience teaches that our
experiments often show quite different effects from the ones expected. One sees
therefore that since experiments matter, no one generation can present a
complete plan of education. The only experimental school which to an extent
made a beginning in establishing a course was the Dessau Institute. We must
let it keep this glory regardless of the many mistakes of which one could accuse
it (mistakes found in all conclusions that come from experiments) – viz., that
new experiments are always required. (LP 9:451)
There is a refreshingly open-ended, empirical tone in Kant’s endorsement of

the view that new experiments are always required in education – one that is
unfortunately absent in his better-known philosophical works.
Public education?
In the preceding sections of this chapter I have argued that the chief
importance of Kant’s philosophy of education44 lies in the central role
that education plays in helping the human species to achieve its collective
destiny. Education is what enables humans to become human. However,
there is one fundamental objection to this thesis that should be exam-
ined. Several commentators have argued that Kant’s strong enthusiasm
for education diminishes sharply in his later years. Kant’s fervent belief
that Basedow’s Philanthropin “shall be the nucleus of a revolution of
humanity,” according to Reinhard Brandt, “is officially shelved” in
179845; his early “education optimism,” Werner Stark claims, “stands
in clear contrast to the political perspective . . . of 1798”46; and “the
importance of education” for Kant, Kuehn maintains, “decreased
between 1772 and 1798.”47
The year 1798 is singled out because in The Conflict of the Faculties (first
published in 1798) Kant, in response to the question “in what order alone
can progress toward the better be expected?” writes: “to expect that this will
eventually happen by means of education [durch Bildung] of youth in the
home, then in schools on both the lowest and highest level . . . is a plan which
44
In this section I borrow a few points from my essay “‘Total Transformation’: Why Kant Did Not Give
Up on Education,” Kantian Review 21, no. 3 (Nov. 2016): 393–413.
45
Brandt, Bestimmung des Menschen, 85.
46
Immanuel Kant, Vorlesung zur Moralphilosophie, ed. Werner Stark (Berlin: de Gruyter, 2004),
366n242.
47
726 R. B. Louden
is scarcely likely to achieve the desired success” (CF 7:92, translation mod-
ified). My own view is that Kant remains committed to the necessity and
fundamental importance of education throughout his career. In the passage
quoted at CF 7:92, he does not renounce his earlier faith in education as a
primary vehicle of human progress. Rather, he shifts his support from private
to public education – he argues for more public oversight of, and financial
support for, education. Education itself is thus to be brought under the larger
purview of government.
Admittedly, Kant does not explicitly and unequivocally argue at length in
defense of public education in any of his writings.48 However, there are
several considerations that do mutually support this reading. First, a few lines
later in Conflict Kant writes: “For while the people feel that the costs for
education of their youth ought to be borne, not by them, but by the state, the
state for its part has no money left . . . , since it uses all the money for war”
(CF 7:92–93). Kant, who elsewhere famously defends the position that
“perpetual peace . . . is no empty idea but a task that, gradually solved,
comes steadily closer to its goal” (PP 8:386), clearly thinks it is wrong for
the state to use “all the money for war.”
Second, throughout the latter half of the eighteenth century the German
government itself was becoming much involved in education, and this issue
was extensively debated in public. Friedrich Paulsen, in German Education,
Past and Present, notes:
the principal innovation during this period was the taking over of the primary
schools from the Church by the State, the compulsory attendance of all
children at school being recognised and enforced as a civic duty. . . . At the
end of the eighteenth century [the primary school] was, in all German coun-
tries, no longer an ecclesiastical but a political institution.49
Kant was certainly aware of this fundamental shift in German public policy.
Third, Kant’s own frustration over the ongoing financial and leadership
problems at Basedow’s school – which culminated in his stepping down from
his fundraiser role in 1778 – also strongly suggests that he had become
48
At one point in the Lectures, Kant asks: “to what extent might private education have an advantage
over public education or vice versa? In general, it appears that public education is more advantageous
than domestic, not only as regards skillfulness but also with respect to the character of a citizen. Quite
often, domestic education not only frequently brings forth family mistakes but also reproduces them”
(LP 9:453). But this seems to be primarily an argument against homeschooling.
49
Friedrich Paulsen, German Education, Past and Present, trans. T. Lorenz (London: T. Fisher Unwin,
1908), 136.
convinced that more government oversight and support for education was
necessary. In another relevant passage from Conflict he writes: “the whole
mechanism of . . . education has no coherence if it is not designed in agreement
with a well-weighed plan of the sovereign power, put into play according to the
purpose of this plan, and steadily maintained therein” (CF 7:93). Had
Basedow’s school been “designed in agreement with a well-weighed plan of
the sovereign power . . . and steadily maintained therein” it would not have had
to close its doors in 1793. This is what Kant is implying here.
Fourth, the public education interpretation coheres better with a much
broader range of Kant’s own texts. For instance, if we accept the counter-
claim that his enthusiasm for education sharply diminishes in his later years,
we would also need to set aside all of his philosophy of history essays from the
1780s as well as the 1798 Anthropology. For one of the key messages in these
writings – works which themselves constitute a substantial portion of Kant’s
corpus – is that “the human being must . . . be educated to the good” (An
7:325). Our species’ Bestimmung requires this.
Finally, the public education reading also squares better with the general
thrust of Kant’s political philosophy, which, as commentators past and
present have argued, is not libertarian in orientation but rather serves as
the basis for a strong theory of social welfare.50
Conclusion
Can the human being only become human through education? The
answer to this question depends in part on what is meant by “educa-
tion,” and here we must remind ourselves of Kant’s assessment of
present-day educational institutions: “they must be transformed if some-
thing good is to come out of them, because they are defective in their
original organization” (EP 2:449).51
50
See, e.g., Alexander Kaufman, Welfare in the Kantian State (Oxford: Clarendon Press, 1999); and
Harry van der Linden, Kantian Ethics and Socialism (Indianapolis: Hackett, 1988). As both van der
Linden and Kaufman emphasize, important earlier German scholars such as Hermann Cohen and Karl
Vorländer also read Kant’s political thought in this manner.
51
An earlier version of this essay was presented as a keynote address at the 3rd International Conference
in Kantian Studies (“Kant and Education”), held at Memorial University, St. John’s, Newfoundland, in
May 2016. I would like to thank James Scott Johnston for the invitation to present some of my work in
Newfoundland, and audience members for their insightful questions and comments following my
presentation.
Part XI
The Kantian Aftermath, and
Kant’s Contemporary Relevance
32
Kant after Kant:
The Indispensable Philosopher
Michael Vater
He. You know how it has been for me with Kant. Surely the whole history of
a philosophical system does not belong to its concept. There would be no end
to it in that case.
I. No beginning, you mean.
– Friedrich Jacobi, David Hume on Faith
Many of Kant’s readers feel the urge not only to excise conceptual items
and arguments they find otiose but to plainly revise Kant’s plan and
rewrite his arguments. One example is P. F. Strawson’s effort to read
the extravagance of transcendental idealism out of the first Critique and
put in its place a more “austere” account of the objectivity of experience.1
Others seem to think that Kant stands as a fixed beacon with whose help
one can navigate contemporary philosophical debates. Wilfrid Sellars
maintains that there are enough similarities between problems facing
Kant and the tools he used to solve them and issues and instruments
available to the contemporary philosopher to make Kant relevant, and not
1
P. F. Strawson, The Bounds of Sense: An Essay on Kant’s “Critique of Pure Reason” (London: Routledge,
1975).
M. Vater (*)
Department of Philosophy, Marquette University, Milwaukee, USA
e-mail: michael.vater@marquette.edu

732 M. Vater
just as a means of communication: “In their most general aspect both his
problems and our perplexities spring from the attempt to take both man
and science seriously.”2 Yet it is peculiar to the experience of seriously
reading Kant to ask, perhaps more seriously than Jacobi, whether our
reading takes on Kant at the beginning of the critical task or at its
systematic end.
This chapter will explore three themes that, taken together, secure Kant’s
continuing relevance for both Anglo-American and European traditions of
philosophy. The first is simple: Kant continues to be relevant in the history
of philosophy because he invented the genre. He wrote the script for modern
European philosophy, framing its questions about the reliability of percep-
tions and the origin of conceptual frameworks to match the solutions his
critical idealism supplied. He sorted the disputes and disputants of the age of
new science into “brands” or tribes – dogmatists and skeptics, materialists
and dualists, empiricists and idealists – so he could take the middle ground
and pacify the warring parties. History is not generated by the accumulation
of evidence or documents, but by the superimposition of a narrative or point
of view.
The second contribution is more difficult to formulate. Before Kant,
metaphysics and epistemology were consumed by questions of what lay
outside the bounds of sense and understanding – whether perception and
its psychosomatic components delivered merely what Descartes called
“objective reality” (the being of a mental state) or communicated some-
thing that exists independent of us (Descartes’s “formal reality”) – and
whether our cognitive faculties reached beyond experience to secure knowl-
edge of supersensible items such as God, an enduring soul, and free will.
After Kant, philosophy becomes agnostic about external items – even about
“things” – and restricts its attention to what is experienced. This “inner
space” gets variously named and renamed transcendental subjectivity, sub-
ject-objectivity, phenomenology, intersubjectivity, ordinary language,
communicative action, and social construction. What all the labels try to
figure is the intra-experiential and multi-centric character of human reality.
As philosophy unfolds after Kant, the cognizing/acting subject is viewed
not as the isolated item designated by first-person descriptors, but as a
universal, multiply instantiated across all social domains – ethics, science,
culture, education, and so on. Even the moral law, though formally a
universal that determines a single will, involves confronting other rational
2
Wilfrid Sellars, Science and Metaphysics: Variations on Kantian Themes (New York: Humanities, 1968), 1.
32 Kant after Kant 733
beings, their wills and their welfare. In the wake of Kant’s dawning
realization that the space of reasons is social or multi-centric, contemporary
explorations of cognition, affect, and communication underscore the social
nature of intelligence even as they illustrate its biological history and
neurological complexity.
Kant’s third contribution is less distinctive: he shares in broader cur-
rents of European thought variously called Enlightenment, humanism, or
political liberalism. Although he isolates philosophy from other endeavors
when doing history of philosophy, Kant was an eager participant in a
broader conversation that included scientists, writers, and public intellec-
tuals. His contribution here lies in the way he uncharacteristically avoids
definitions and frames the project of humanism as “the question of the
human being.” What will resonate with philosophers in succeeding ages is
the primacy of the question and the urgency of redesigning the customs
and institutions – pedagogical, cultural, economic, and political – that
make the people who live within them. Pointedly turning his back on
vertical models of religion, Kant saw that humans have the task of
designing human reality and deciding what it shall be. Of course he
harbored essentialist tendencies. He conflated “reason” with abstract
logical values like universality and necessity, which seemed to support
beliefs that human reality is biologically one and its members morally
equal by nature. But in this area, his agenda was pragmatic, not critical; in
any social-political setting less coercive than theocracy or totalitarianism,
one must design institutions to introduce human beings into their
rational and social capacities.
In what follows, I selectively discuss some philosophers and philosophies
that illustrate these contentions. The chapter’s title, with its use of “indis-
pensable,” is meant to be provocative; it suggests we could not understand
ourselves, or at least our philosophies, if we did not deeply study Kant. The
rhetoric suggests a transcendental argument, and one recent development in
Kantian and post-Kantian philosophy is the probing of this form of argu-
ment with its seeming capacity to prove everything on a nonevidential basis.3
I regret that I cannot touch upon late nineteenth-century developments in
Germany, Britain, and America; instead, I will focus on traditions more or
less in play today.
3
See Paul W. Franks, All or Nothing: Systematicity, Transcendental Arguments, and Skepticism in German
Idealism (Cambridge: Harvard University Press, 2003), 201–59.
734 M. Vater
Making history
Kant did not come to his narrative about modern philosophy without
struggle, and his process toward that narrative as well as the path to his
own critical philosophy remains hidden in the “silent decade” devoted to the
composition of the Critique of Pure Reason. Much of what we can infer about
it comes from minor publications such as the Inaugural Dissertation of 1770,
the statement in the 1783 Prolegomena that he was wakened from “dogmatic
slumber” by his reading of Hume, and the attempt to mediate the
Mendelssohn-Jacobi conflict over Spinoza in the 1786 essay “What Does It
Mean to Orient Oneself in Thinking?” In 1770, Kant is an idealist in the
mold of Leibniz whose mistrust of the senses is overbalanced by confidence
that the intellect can employ logical and mathematical concepts to construct
a metaphysics, and also deploy concepts “critically” to prevent the applica-
tion of sensible predicates to supersensible realities.
Let us turn to the Critique and review its resolution of the warring
positions in modern philosophy, the most interesting of which are found
in the earlier version of the Paralogisms, the Antinomies, and the Refutation
of Idealism added in the B-edition. In the first, Kant argues that it is the
nature of reason to assume antithetic positions on matters incapable of
experiential exhibition and inspection, especially on the nature of the subject.
Spiritualism, the untutored conviction that I am a thinking thing, is the
initial or natural position, but Kant calls it a slumbering imaginary convic-
tion, even “the euthanasia of pure reason” (A407/B434). If one does not
dreamily depart for unknown regions, the alternatives of dogmatism and
skepticism awaken – the first attempts to use concepts of reason to argue that
the I which appears in cognition is a thinking substance, to which skepticism
replies that there is no theoretical way of fixing or evaluating the appearance
of the internal narrator and narrative (A406–7/B433–34). Note that Kant
considers two alternatives in this passage: (a) either one will let conventional
notions rule one’s life and let ordinary language rule when reason ought to,
or (b) reason can awaken, but straightaway falls into antithetical positions.
Kant offers his clearest definition of transcendental idealism in the first
version of the Fourth Paralogism. The idealist is not one who denies the
reality of external objects, but one who is unable to overcome all doubts
about external reality and so remains locked in her ideas; it is likely that Kant
is considering Descartes here. The realist is insecure, however, for as much as
one believes that one’s perceptions are the real characters of thing, one
cannot get outside them and to the nature of thing. The transcendental
idealist can admit that the appearances in space and time that she perceives
are dependent on our sensibility (and so cannot be referred to external
objects apart from that form of representation) while they yet function as
actual things, directly perceived and not inferred, and so support empirical
realism. Within representation, one can be certain that I am as real as the
objects that I perceive (A368–70).
The second edition of the Critique adds a Refutation of Idealism, in which
Kant argues that transcendental idealism is the only appropriate response to
Descartes’s problematic idealism – problematic in that Descartes’s thought started
with universal doubt and idealistic in that the doubt was resolved into the self-
certifying idea of the cogito: whatever my mental state, if I have one I know I am.
What problematic idealism really does is pose the demand for a rational warrant
for the reality of external things, that we not just imagine, but experience them.
And “experience” implies that the existence of objects in space outside me is just
as real as my consciousness of them (B274–75). Of course, there are problems
here; in underscoring the correlation between the unity of consciousness and the
objectivity of the object in appearances, Kant may have said too much about the
actuality of objects to continue talking about things in themselves.
This passage is Kant’s sole but oblique reference to Descartes as the originator
of modern philosophy, not the one who promoted the false light of the cogito,
but the one who (on the basis of some slight but manageable doubts about the
veracity of perceptions) decided to call everything into doubt until he could find
an Archimedean point of certainty. It was Descartes who moved the question
from quid facti? to quid juris? – from the correctness of isolated perceptions to
the justification of knowledge as a whole. It was Descartes who first summoned
reason to court to answer for itself, and after him came a procession of
unreliable witnesses who advanced partial answers that were really only
responses to quid facti questions – the clarity and distinctness of ideas, or the
vivacity and force of sense impression. When Kant steps forward and shows that
what is at issue is the normative reach of reason or the legitimacy of our
cognitive frameworks, he finds that he can organize the half-truths of idealism
and realism, skepticism and dogmatism, and spiritualism and materialism into a
complete case for the objectivity of experiential knowledge – if one but grant the
initially counterintuitive (or semi-skeptic) distinction between appearances
conditioned by our form of sensibility and things in themselves that can be
conceived but never experienced. Kant therefore fashioned the history of mod-
ern philosophy in coming to his own philosophy. Reason had been in the court
of its own making for some two centuries before Kant stepped in to make
closing arguments and read the verdict, which includes a partial vindication of
all of the evidential claims the witnesses advanced. But the vindication of the
736 M. Vater
claims to objectivity and necessity advanced by experiential knowledge in toto

was the work of the process as a whole.
We now turn to our second topic, the reception and retrieval of Kant’s
account of cognition in the Critique of Pure Reason.
Justifying cognitive frameworks

First responders: Three works published in the Critique’s first decade illustrate
how Kant’s initial readers struggled with the work. They all omit or revise
elements in Kant’s doctrine, hoping to clarify the argument or eliminate
unthinkable suppositions. Two of the three give Kant’s cognitive paradigm a
fair reading, but none is willing to accept Kant’s conclusion about the
impossibility of a supersensible metaphysics.
Friedrich Jacobi had argued with Moses Mendelssohn for years about the
threat of Spinozism, or the relative competence of faith and reason to assess
religious matters, when Kant intervened in 1786 with his “Orient Oneself in
Thinking” essay. He suggested that a heuristic use of reason to foster rational
belief in matters philosophically undecidable was appropriate, but also repu-
diated Jacobi’s use of quotes from his own pronouncements on faith and
reason. Jacobi’s 1787 David Hume on Faith, or Idealism and Realism brought
Kant directly into the controversy. While Jacobi previously advocated
recourse to a blind faith to resolve doubts that Spinoza raised, he moderates
the wildness of that position (which Kant deemed enthusiasm) and turns to
Hume’s indirect way of resolving skeptical problems by relying on conven-
tion or everyday belief.4 In a supplement, Jacobi raises the issue of Kant’s
transcendental idealism and its reference to a thing in itself. After extensively
citing the Fourth Paralogism and the Transcendental Aesthetic, Jacobi con-
fesses perplexity about how we can know anything about things of which we
are utterly ignorant, but still contend that we have objective knowledge of
sensible things in experience. For his part, he cannot conceive of knowledge
of objects in any way other than through sense impressions from external
objects. He is candid about adopting this supposition without philosophical
inspection, and remarks that with it he can at least enter the Critique, though
retaining it he cannot remain within it.5 George di Giovanni notes that
4
Friedrich Heinrich Jacobi, David Hume on Faith, in The Main Philosophical Writings and the Novel
“Allwill,” trans. George di Giovanni (Montreal and Kingston: McGill-Queen’s University Press, 1994),
262–65, 270–73.
5
Ibid., 332–38.
Jacobi failed to see the systematic importance of the thing in itself as the
marker of the finite or contingent nature of human reason.6
After Kant warmly received his letters on the cultural and religious
significance of the Critique (see below), Karl Reinhold turned to a systematic
exposition of Kantian philosophy in his 1789 Essay on a New Theory of the
Human Capacity for Representation. Not widely studied today, his contribu-
tion is often mentioned as a defense of the thing in itself or a first attempt at a
wholly foundationalist philosophy.7 It is these things, but Reinhold’s inten-
tion was not merely to recast the presentation of the Critique, but to expand
its explanatory basis to hook outlier features such as the thing in itself more
deeply into its fabric. His vehicle is an extended account of “representation,”
a mental state prior to consciousness or cognition. I prefer to translate
Vorstellung as “mental state,” for I fear that we usually think Kant’s “repre-
sentation” is an item (already) in consciousness.
Reinhold argues that we must insert the presence of a mental state at the
beginning of the account of cognition if we care to explain consciousness,
sensibility, understanding, and desire.8 A mental state has rudimentary
properties such as reference to an object and a subject, a capacity to be
affected and spontaneity.9 Consciousness has properties similarly named, but
since consciousness is a mental state that is a relation of mental states, its
higher-level gesture to subject and object and its way of joining spontaneity
and being-affected are sui generis.10 Kant’s theory gives one no right to drag
in the arbitrary supposition that every objector seems to require: that there be
some property-similarity between sensations/perceptions and things in them-
selves. By having a mental state as the primitive item in the account, but one
that is preconscious, a Kantian can maintain that there is an objective
something underneath our cognition, but avoid Jacobi’s mistaken idea that
a representation must copy or re-present external features.11 If it lacks a two-
layered framework of explanation, transcendental idealism deflates into
simple realism and returns us to the skeptical perplexities realism always
faces.
6
George di Giovanni, introduction to Main Philosophical Writings by Jacobi, 101–2.
7
For the argument that the thing in itself is necessary though intrinsically unknowable, see Karl
Leonhard Reinhold, Essay on a New Theory of the Human Capacity for Representation, trans. Tim
Mehigan and Barry Empson (Berlin: de Gruyter, 2011), 116–17.
8
Ibid., 86–87.
9
Ibid., 114, 124–28.
10
Ibid., 150–51.
11
Ibid., 112.
738 M. Vater
Reinhold offers a fairly standard interpretation of the rest of Kant’s

Analytic, except for his explanation of how imagination’s synthesis underlies
the pure concepts of the understanding. Reinhold offers a primitive con-
struction of synthesis that is operative beneath the level of the explicit
formulation of propositions and the distinction between affirmation and
negation, arguing that at the basic level of information-integration
(Ineinsbildung), spontaneity is joined with or impressed upon being-affected
the way unitary form is imposed on a material manifold.12 Fichte incorpo-
rates this feature of Reinhold’s account and makes it the basic unit of the
activity of world-construction depicted by the Wissenschaftslehre: feeling, or
the active register of being-affected.
Though Reinhold gives careful attention to the reconstruction of the
Analytic, there is no Dialectic, since he does not agree with Kant that reason
is inherently dialectical. Reason is intrinsically reliable; there is no reason to
mistrust the extension of concepts into metaphysical domains.13 But it can
fashion erroneous theories, and when mistakes are made – as they were made
by both empiricists and idealists before Kant – they are made by philosophers
who fail to observe carefully, who confuse ontological and epistemic issues,
and jump levels. Pre-Kantian philosophy’s failed attempt to explain cogni-
tion arose from its failure to notice the mental state underneath the sensa-
tion, perception, or concept. Self-undermining versions of realism and
idealism, materialism and spiritualism, and dogmatism and skepticism
resulted from this failure.14 Kant recovered the half-truth of these positions
in the Analytic.
If Reinhold’s effort as a commentator was solely to order and integrate
Kant’s account of cognition, others soon began to revise it on the model of
idealistic and skeptical theories then current. Salomon Maimon was self-
educated. When he moved into the German world of learning, he changed
his patronymic (ben Joshua) to honor Maimonides. Indebted to Kant’s
friends Moses Mendelssohn and Markus Herz for his entry into Berlin
society, his Essay on Transcendental Philosophy was presented to Kant by
Herz before publication in 1790. Replete with mathematical examples, it
challenged some of Kant’s key assumptions from a Leibnizian-Wolffian
perspective.
12
Ibid., 136.
13
See Reinhold’s argument for an objective absolute ground of rationality – at once first cause, absolute
subject, and the ens realissimum (Reinhold, Essay on a New Theory, 258–65).
14
Reinhold, Essay on a New Theory, 268–71.
Maimon understood the principle of Kant’s philosophy from its funda-

mental decision to place the question of lawfulness (quid juris?) of cognition
ahead of subsidiary questions about sensation, perception, concepts, and
judgment, and their correctness. From the start, however, Maimon contests
key elements in the Kantian philosophy:
• He finds the distinction between analytic and synthetic propositions

puzzling, for synthetic propositions are just ones incompletely analyzed
or understood by finite minds.
• Although the question of normative necessity (quid juris?) opens the path
to systematic philosophy and one can conceive how concepts can apply to
empirical intuitions, the relation of a priori concepts to a priori intuitions
is obscure.
• Intuition must attach to a material manifold, so metaphysics must include
a new kind of idea, an “idea of the understanding” – a material counter-
part to Kant’s formal ideal of reason.
• Kant’s answer to Hume’s skepticism is unconvincing.15
Maimon breaks with Kant by placing reality or ontology ahead of episte-

mology. The material of intuitions is not unknowable qualities, but “differ-
entials” or nonextended point-realities, like Leibniz’s petites perceptions in
magnitude, but not miniature minds; Maimon also calls them “ideas of the
understanding” or “noumena.” They are swarming and evanescent in nature,
so understanding must capture them as they arise and transform them into
intuitions.16 One stage of the process is the grouping together of these items
into space and time, which are not (as Kant said) pure a priori intuitions but
unities created from manifolds. Since they are constructed and not given a
priori, space and time are intuitions, but lack self-certifying necessity and
universality.17
Kant responded to Maimon’s criticisms in a letter to Herz on May 26,
1789. By making synthetic propositions variants of analytic ones, Maimon
moves close to ascribing an intuitive intellect to human beings. His difficulty
about how a priori concepts combine with a priori intuitions is easily solved;
such a synthesis obtains only in the context of experience, and its upshot
15
Salomon Maimon, Essay on Transcendental Philosophy, trans. Nick Midgley, Henry Somers-Hall,
Alistair Welchman, and Merten Reglitz (London: Continuum, 2010), 9. Cf. Maimon’s letter to Kant of
April 7, 1789, in Essay on Transcendental Philosophy, 229 (and also C 11:15–17).
16
Maimon, Essay on Transcendental Philosophy, 21–22.
17
Ibid., 34–36.
740 M. Vater
applies only to appearances, not to things in themselves.18 Maimon replied in

a journal article that, although Kant and he agreed that metaphysics was
concerned with knowledge of things in themselves, Kant thinks that it is a
null set and metaphysics is no science, while he thinks that metaphysics is the
science of the elements or end-members of the series of appearances, that is,
of the elemental differentials or ideas of the understanding.19
Constructing freedom: Reinhold’s citadel of Kantian studies at Jena became,
even in Kant’s lifetime, the center of efforts to transform criticism into a
distinctive philosophical system, initially called the Wissenschaftslehre – or
Transcendental Idealism – by J. G. Fichte, later Objective or Absolute
Idealism by F. W. J. Schelling and G. W. F. Hegel. Although each of the
three allowed himself to get ensnared in debates with other figures, at the
core of the movement was a decision to rescue key Kantian insights from the
epistemic preoccupations that occasioned their formulation and to exhibit
them as reason’s positive work in the natural and human worlds. Among the
idealists’ key assumptions are:
• Reason’s normative authority extends beyond logical and mathematical

domains to exert organizing force in nature, science, human society,
morals, and all the branches of learning.
• Rationality is powerful and manifests as non-entropic activity – or an edge
of spontaneity over being-affected or limitation in finite contexts.
Sometimes this is interpreted as the primacy of ethics over cognitive
rationality.
• Reason’s systematic character endows its works with universality and
necessity.
• Philosophy’s task is comprehensive explanation, not the solution of spe-
cific puzzles or the deployment of tightly formulated arguments, but
“world-construction” – a total account of intra-human (or conceptual)
reality.
• Kant’s restrictions on the extent of human knowledge are upheld. Lacking
access to things beyond experience, transcendental philosophy is agnostic
about ontology and unwilling to make existence claims.
Kant himself would recognize the items on this list as Kantian, although
he might find the fourth inflated in its ambition. It is one thing to pose
18
Ibid., 231–33.
19
Ibid., 248–49.
claims that reason is systematic, its structure architectonic, and the goal of its
explanation a totality of conditions, and quite another to consume thousands
of printed pages to fabricate total world narratives.
How do the idealists differ from Kant? Fichte makes three decisive
alterations in arguing with Kant and his surrogate, Reinhold. (1) If
philosophy is to incorporate and argumentatively defend freedom, it
must take its stance in freedom or activity. It cannot therefore take
“thing” as a primitive term; although it may justify certain cognitive states
as not only factual but necessary, there can be no thing in itself. (2) The sort
of compromised activity-and-being-affected that Reinhold used to mark
the mental state contained too much being-affected and too little sponta-
neity. If representation, along with its gesture to an unknowable thing, is
eliminated, there is no longer a basis for any sort of realism, and philoso-
phy becomes phenomenalism. (3) Accordingly, the distinction between
consciousness and self-consciousness collapses. Whereas Reinhold argued
that self-consciousness depends on consciousness, Fichte reverses the
account: consciousness is possible only on the basis of a self-consciousness
that allow direct access to the self-realizing activity that is the I, in a
founding intuition he calls intellectual intuition.20 Schelling and Hegel
make similar moves and their philosophies too become descriptive or
phenomenological, claiming privileged intuition or methodological trans-
parency as anchors in lieu of objectivity.21
It is difficult to see how the resulting philosophies are Kantian, even if they
call themselves Transcendental Idealism, for Fichte and his successors simply
walk away from Kant’s problem of the legitimacy of human cognitive
structures. In defense of a freedom which Kant himself found difficult to
define and defend, they simply jump into a postulated original activity – or
productivity (Schelling) or thinking (Hegel)22 – and consider that they
succeed if from this postulate and a series of limitations or dialectical
20
Fichte succinctly presents the gist of his argument with Kant in the 1798 “Second Introduction to the
Wissenschaftslehre.” Since self-reflection leads to intuition of the I as active, a philosophy that wishes to
explain freedom starts and finishes in “intellectual intuition” of that activity. There can be no original
thing. See J. G. Fichte, Introductions to the Wissenschaftslehre and Other Writings, trans. and ed. Dan
Breazeale (Indianapolis: Hackett, 1994), 41–51.
21
More as a matter of ornamentation than of substance, Schelling and Hegel imitate the structure of
Kant’s Critiques. Schelling closes the System of Transcendental Idealism with a treatment of teleological
and aesthetic judgment, the content of Kant’s final Critique. Hegel’s Logic has rationality return to itself
in self-conscious form as methodology, mirroring the Critique’s Doctrine of Method.
22
This discussion arbitrarily takes Fichte’s Foundation of the Entire Wissenschaftslehre (1794–1795),
Schelling’s System of Transcendental Idealism (1800), and Hegel’s Encyclopedia of Philosophical Sciences
(1817) as these authors’ defining works.
742 M. Vater
reversals, they can finally approximate the situation of a finite human knower/
agent in a world of incompletely known things, among them other entities
claiming also to be free subjects. Yet the strategy of the argument is Kantian –
one colossal transcendental argument.
If the idealists’ strategy is simple – assume activity, productive power, or a
thinking that is creative, not merely reproductive, and whittle it down to
finite size by having it create its own limitations – the task it sets itself is
difficult. Fichte succeeds, to some extent, for in the practical part of the first
version of the Wissenschaftslehre (1794) (we might call it moral psychology
today) he can point to drive and feeling as the first place where the I’s limited
activity (striving) is concretized. Instead of the dry logical clash of self-
positing and not-self-positing which theoretical philosophy presents, the
affective realm reveals a point where spontaneity is directly joined to being-
affected, where will in its primordial self-active state shows up as pre-directed
to a specified state.23 Reinhold observes something similar, for viewing
representation as an indissoluble knot of activity and passivity, as impossible
to observe as is the eye’s seeing,24 he discovers a primordial self-presenting
active state at the point where desire arises as drive, a pre-orientation which
seems to be a choice already made.25 Fichte looks at the same phenomenon
but offers a different analysis, for he finds dynamic phenomena (striving vis-
à-vis counter-striving, drive, and feeling) at the basis of consciousness.
Tensed activity is primary; representation and cognitive states are subsequently
generated by the primordial conflict.26 For Kant and Reinhold, the preemi-
nent human power is cognitive, while willing (the power of for desire) is
secondary and dependent; for Fichte, activity is prior and all phenomena
which arise for us as “mind” are but alterations of underlying
psychodynamics.
I have suggested that the post-Kantians’ basic endeavor is “world-con-
struction,” or the encyclopedic description of human reality. But one can
argue that they had a more limited aim: to confirm the small share of
freedom that Kant believed he had saved after he conceded the universal
sway of causal determination to Spinoza (in the Third Antinomy). Yet this
modest aim is what they seem to achieve, for they typically posit unlimited
activity, productivity, or expressive power as first principle, discover a feature
23
J. G. Fichte, The Science of Knowledge, trans. and ed. Peter Heath and John Lachs (New York:
Appleton-Century-Crofts, 1970), 253–61.
24
Reinhold, Essay on a New Theory, 120–22, 129.
25
Ibid., 275–78.
26
Fichte, Science of Knowledge, 259.
in that activity that undermines or limits it, and deduce all the limitations of
that principle, until the construction accumulates all the conditions that
make human reason a contingent or merely social-historical phenomenon. If
the (transcendental) form of argument is admissible, the argument is correct,
for it returns us to the limited forms of reason and freedom that we
experience in ourselves and our social and natural environments.
But if the idealists’ arguments merely return us to the Third Antinomy,
what is the point? The point is descriptive and stays quite within the Kantian
pattern of articulating the conditions of our experiential cognition and finite
action, although the idealists employ the illusory rhetoric of transcendence to
suggest that the logical motion of dialectic or genetic deduction gets us
somewhere more exalted than the next stop on the railroad home from
work. The logical push-pull of dialectic just gets the reader from one stop
to the next; it takes a conductor (or hermeneutic commentator) to announce
where we are – or to overlay a narrative of human significance upon the
journey. The journey is entertaining, if fictional – Schelling and Fichte
admitted the fictive nature of the start of the system, while Hegel maintained
it was self-starting27 – but what function does it serve? Kant may say that the
narrative compiles the totality of conditions for (limited or human) ration-
ality and action, but it appears as the totality of limitations that finite
existence imposes on the quantum of freedom with which we started. We
have not left the Third Antinomy, though perhaps we learn a Spinozistic
lesson on how to enjoy a rational but limited life.
Kant critiqued: In 1818 Arthur Schopenhauer published The World as Will
and Representation, with an appendix pointing out Kant’s merits and flaws.
Dismissing all that happened after Kant as little more than common sense
inflated by hot air, Schopenhauer selected Kantian topics and, freeing them
from their original context of legitimizing experiential knowledge, combined
them with spiritual themes from Plato, the Vedas, and early Buddhist teach-
ings to form a soteriological system. Logically, Schopenhauer’s boldest move is
to identify four types of relational concepts – ground and consequent in
“things,” subject and object in “cognition,” relations of externality and
succession in “space and time,” and motive and decision in “psychology.”
They are more than analogously related; they are the fourfold root of sufficient
reason, the dream logic that organizes the human experience and the
27
See Schelling, System of Transcendental Idealism, trans. Peter Heath (Charlottesville: University Press
of Virginia, 1978), 24, 33; J. G. Fichte, “Second Introduction,” in Introductions to the Wissenschaftslehre,
37–38, 43; and G. W. F. Hegel, The Encyclopaedia Logic, trans. T. F. Geraets, W. A. Suchting, and H. S.
Harris (Indianapolis: Hackett, 1991), 45–55.
744 M. Vater
appearances or objects featured in that long dream. That logic makes clear
that objects are just appearances and that the rules of experience are little
more than bins set up to classify and collect similar objects, but it leaves their
origin and essence totally unexplained, as is seen most clearly in the subject-
object relation, which is really all about objects and says nothing about the
subject.28
Real causes, subjects, motives, or the basis of space and time cannot appear
in the unfolding dream that these relations govern, and any reality they
might have besides being placeholders in descriptions of appearances would
have to be connected to will, Kant’s thing in itself reconfigured as a principle
of activity. We do not know the dimension of will except in its objectified
form, as body – the real-life correlate of the Kantian dream, with its inter-
laced biological, physiological, mechanical, and behavioral systems. My body
is the representation of will and the vehicle for all other representations, so to
say I know only appearance is to say I know myself only as embodied,
organic, semi-responsible action.29 The ceaselessly productive but not pur-
posive creativity of nature, particularly at lower or inorganic levels, indicates
the chaotic nature of will; while we knowers and everything we know falls
under the principle of individuation (location in space and time), will escapes
all parameters and knows no measure. If what we can cognize must be
characterized by spatiotemporal location, causality, persistence, and change,
will lacks all logic and exhibits basic wildness: striving without goal.30
Schopenhauer compounds two other branches of Kant’s thinking into a
palliative for the eerie, divided state of individual existence: aesthetics and
ethics. Aesthetic works and the pleasures which accompany their creation and
contemplation allow cognitive mind an intuition not bound by the rules of
ordinary experience and furnish a satisfaction different from the use, consump-
tion, or destruction that quotidian enjoyments entail. And ethics indicates what
mollification, if any, is possible for will in its unsatisfactory, multiple, and self-
destroying objectifications. Will is free, that is, active and not bound by rules or
purposes, but my body and its life is quite bound, for I have such and such a
character and such and such a psychology, and I find myself in these circum-
stances, with these desires and fairly limited choices. As embodied knowing, I
can choose to affirm this tranch of will or transcend it by renunciation based on
a clear cognition of the character of life. This leads one to a recognition of
28
Arthur Schopenhauer, The World as Will and Representation, vol. 1, trans. Judith Norman, Alistair
Welchman, and Christopher Janaway (Cambridge: Cambridge University Press, 2010), 23–34.
29
Ibid., 134–35.
30
Ibid., 188–89.
something like the Vedic declaration Tat tvam asi (You are that!) and to a
strategy of love or compassion as a fulfilling way of life.31
In a lengthy appendix, Schopenhauer praises Kant for three moves: his
distinction of appearances and the thing in itself, his assignment of the
former to cognition and the latter to ethics, and the depiction of the
human situation in the Third Antinomy, which consigns all human actions
to rule of cause and effect but bids us look to another domain for freedom.
But Schopenhauer vehemently criticizes the way that Kant got to the thing in
itself, citing an Asian proverb that there is no lotus without a skinny stem and
stinking pond underneath; Kant could have derived the distinction of thing
in itself (= will) and appearance directly by enlarging Berkeley’s claim: no
object without a subject.32 In the Third Antinomy, Schopenhauer claims that
Kant proposed a riddle to which his doctrine of will (the thing in itself) is the
key. But Kant fancied instead that the ground of appearances must be an
object of some sort, forgetting that one has objects only in experience and
that all objects are appearances.33 An even more basic criticism is that Kant
failed to define reason at the start and only surreptitiously introduced the
distinction between cognitive and practical reason late in the Critique (A802/
B830). Kant there speaks of how theoretical reason determines the facts of
experience, while practical reason prescribes oughts; the appropriate contrast
should have been between the logical musts governing how appearances are
shaped and the oughts of the moral realm.34
Other reactions to the conceptual idealism of the post-Kantians are akin to
Schopenhauer’s soteriological system. Søren Kierkegaard borrows concepts
from Kant’s thinking but uses them to craft an apology for religion – not the
universal but hollow theism of the old rational theology, but a vivid way of
life that involves personal encounter with the divine (or flight from it) as one
of life’s options.35 These fundamental options or potencies (Schelling’s term)
are the aesthetic life of pleasure, the ethical life of renunciation, or the
religious life of personal encounter with the infinite – each of them governed
by one of Kant’s modal categories: possibility, necessity, and actuality. What
Kant pointedly excluded from philosophy’s competence – comprehending
existence (while agreeing it is not conceptual) and embracing individuality
31
Ibid., 401.
32
Ibid., 461–64.
33
Ibid., 531–33.
34
Ibid., 553.
35
Perhaps Jacobi’s “faith” had such a personalist or I-Thou cast. See Jacobi, David Hume on Faith,
328–29.
746 M. Vater
(while shunning universal prescriptions) – is what Kierkegaard includes.36

He does this not only conceptually, but makes his life an experiment, writing
many works in different genres, from varying points of view. Sometimes the
author ironically distances himself from his thought by posing as editor or
commentator on his pseudonymous creations. Schelling too tries to start
with existence when he returns to the lecture hall in Berlin in 1841, in a
state-sponsored attempt to lessen the influence of Hegel’s conceptualism. In
the so-called Positive Philosophy, Schelling deliberately intertwines two
themes that Kant wanted to exclude: talk of existence and of positive
(revealed) religion as truth.37
Kant in twentieth-century Anglo-American thought: The sharply different
angles described by Anglo-American theoretical philosophy in the twentieth
century can be traced to the influence of Ludwig Wittgenstein’s major works.
The Tractatus Logico-Philosophicus (1921) marked the high-water of philoso-
phers’ faith in logic to keep philosophical discourse within Kantian parameters,
for the ideal language advanced there had the power to represent empirical facts
and situations, but excluded other human concerns such as ethics and aes-
thetics, which escape formulation into propositions with definite truth values.
The only “historical” philosopher Wittgenstein studied was Schopenhauer, and
the latter’s view of the laws of experience as a relational net which lets facts swim
through and produces only the texture of a dream powerfully influenced his
formalistic view of logic: a network overlaid upon events that describes only the
rules of description.38 The direction is reversed in Philosophical Investigations
(1953), where the hope for all-embracing formal rules as the mark of reason is
abandoned and we are returned to the myriad distinctions and inflections of
ordinary language to make sense of the world. But wander as we must in the
bowels of language, we are still in Kantian territory: the interface between
subject and object where, as Schopenhauer reminded us, each is given only as
affected by the other – a world of experience and reliable cognition, capable of
inspection and verification, but offering us no access to subjects-in-themselves
or objects-in-themselves.39
36
Cf. Søren Kierkegaard, The Sickness unto Death: A Christian Psychological Exposition for Upbuilding
and Awakening (trans. Howard V. Long and Edna H. Long [Princeton: Princeton University Press,
1980], 13–21, 29–74) for his use of Kantian categories to describe despair.
37
See Michael Vater, “Religion Beyond the Limits of Criticism,” in The Palgrave Handbook of German
Idealism, ed. Matthew C. Altman (London: Palgrave Macmillan, 2014), 499–517.
38
Ludwig Wittgenstein, Tractatus Logico-Philosophicus, trans. D. F. Pears and B. F. McGuinness
(London: Routledge & Kegan Paul, 1961), 6.124–6.1251, 6.522–6.54. Cf. J. N. Findlay, Kant and
the Transcendental Object: A Hermeneutical Study (Oxford: Clarendon, 1981), 367–76.
39
Schopenhauer, World as Will and Representation, 461–62, 532–33.
P. F. Strawson advanced several much-discussed conceptual analyses. In

Individuals (1959), he distinguishes between descriptive metaphysics, which
stays within Kantian parameters, and “revisionary” theories, which tend to
expand the world of things (seen and heard) and persons to include things
thought but not experienced.40 Key to descriptive metaphysics is the relia-
bility of perception, which Strawson explains in terms different from the
classic empiricists who appeal to force or vivacity of impression as sensations’
mark of ultimacy. For perceptions to mark out individuals, they must be
identified or located in the common space-time nexus; once located, an
individual can be reidentified or cognitively recognized.41 Having a material
body is the condition for being perceived, so persons can be figured into a
descriptive metaphysics, since they are embodied. Since the “re” in reidenti-
fication connotes some contribution from the subject, perception includes at
least the ability to navigate relations of contiguity and succession.
Strawson explicitly takes on Kant’s theory of cognition in The Bounds of
Sense (1966), showing just what sort of haircut is required to trim back his
revisionary impulses and return him to the descriptive terrain of a “meta-
physics of experience.” What Strawson hopes to fashion is an “austere”
Kant who would meet the current need for a minimal account of experience
without getting involved in fanciful strategies like transcendental argu-
ments: “In order to set limits to coherent thinking, it is not necessary as
Kant . . . attempted to do, to think both sides of those limits. It is enough to
think up to them.”42 Still, Strawson admits, no one has come closer to that
elusive goal.43
Since Strawson’s account of perception focuses on location in space and
time, what it takes for perceptions to count as objects, and a general tendency
to introduce mind-talk sparingly, it is not surprising that in his revision of
the Critique, he emphasizes the arguments of the Transcendental Aesthetic,
the two Transcendental Deductions, and the Analogies supplemented by the
Refutation of Idealism. If one puts all these sections together, Kant’s argu-
ment apparently was that one has a world of objects with episodes of
enduring being, change, and ordered interplay if and only if one is a
40
P. F. Strawson, Individuals: An Essay in Descriptive Metaphysics (Garden City, N.Y.: Anchor, 1963),
xii–xv.
41
Ibid., 5–49.
42
Strawson, Bounds of Sense, 44.
43
Barry Stroud questions whether one can just “think up to the limit” in “The Synthetic A Priori in
Strawson’s Kantianism,” in Understanding Human Knowledge: Philosophical Essays (Oxford: Oxford
748 M. Vater
self-conscious subject: this is full-frontal transcendental idealism, I gather.

Strawson would reduce Kant’s last condition to the necessity of there being
some logical reflexivity sufficient to discriminate objective from subjective
time-orders, so that in general objectivity is correlated with temporal order.44
But does the representation “I think” that accompanies every objective
experience (B132) point to anything more than a representation of logical
unity, that is, to an empirical I? This is the central ambiguity of the
Transcendental Deduction, which, according to Strawson, moves from a
straightforward claim that an empirical I must be able to ascribe all experi-
ences to herself to the need for a transcendental I, or logical reflexivity in
general, that must be posited as a condition of empirical self-consciousness.45
Clearly the reduplication is questionable. What Strawson is at pains to read
out of his account of Kant’s “problematic idealism” is an inflated metaphy-
sical picture of a demiurgic knower somehow fashioning a world out of
privileged intuitions of purely unknowable things in themselves.46
Wilfrid Sellars is quite technical in the close analyses he offers of detailed
subjects, but very much a traditional philosopher in the “big-picture” sense. He
returns time and again to Kant not just because of the enticingly technical,
perhaps faulty, details of Kant’s conceptual scheme but because in the wider
sense he stands with Kant, wishing to endorse the truths turned up by empirical
science but hanging onto a traditional humanistic or “manifest image” when it
comes to the moral and social situation of human beings.47 This need not mean
that we are condemned forever to a stereoscopic vision, for, holding onto the
irreducibility of ought to is, we can formulate the way intentions work to
integrate single persons into a community and thus approach a more holistic
or “field-theory” of human reality. In a more technical vein, he can say that his
distinction between the manifest image and the scientific image is an appro-
priate translation of Kant’s phenomenalism (which he characterizes as esse est
percipi) but at the same time suggest that successor concepts from the former,
especially those describing objects, have by our time crossed the line and are part
of the scientific image.48
In earlier essays, Sellars notes that the Kantian contrast of sensibility and
understanding is not as simple as it first seems, and though the model of an
44
Strawson, Bounds of Sense, 117, 147.
45
Ibid., 108, 110–11.
46
Ibid., 22.
47
Wilfrid Sellars, “Philosophy and the Scientific Image of Man,” in Science, Perception and Reality
(London: Routledge & Kegan Paul, 1963), 38–39.
48
Sellars, Science and Metaphysics, 46–50, 150.
intuition is the presentation of a singular item, some intuitions involve a

complexity and spontaneity, making them more like concepts, while others
are more obviously marked by receptivity. Some originate in productive
imagination, others in a unification of a sensible manifold; both have the
logical structure of judgments.49 In more developed presentations, Sellars
observes that the intuiting of a sensible complex must involve at least all the
pertinent physical categories and that it must be a synthesis effected in
accordance with the laws of nature.50 But he departs from the historically
dominant thread that views Kant as a phenomenalist who would construct a
world by imposing concepts on mere sensations. Kant is a realist in main-
taining that there is an in-itself, a domain not dependent for its being on
being represented; that complex intuitables cohere in one reality or system
depends not just on their being synthesized according to rules, but their
belonging to one complex intuitable individual, nature.51 When his notes
say: “Truth [is] the objectively and intersubjectively valid belonging together
of representables,” I take it he is offering a definition and an account of
knowledge that is both seriously realistic and seriously transcendental.52 In
the background, Sellars depends on the function of the rational community
as language-learners and language–teachers to demonstrate how a norm-
governed community exists, where some in the community deploy “rules
of action” and others “rules of criticism.” Reason’s ultimate imperative is
velvet-gloved: “This is how we say/do that.”53 But this is only a paraphrase of
the opening sections of Wittgenstein’s Philosophical Investigations.54
Finally, we must mention recent thinking on the nature of transcendental
arguments which has thrown light on Kant’s thinking, particularly in the
Refutation of Idealism and the Second Analogy. Robert Stern considers how
such arguments vary, depending on whether they try to justify truth-claims,
in the most strenuous case, or merely facilitate conceptual analysis, in the
least rigorous.55 The skeptical challenge Kant faced is open to a range of
interpretations, from the epistemic skepticism Descartes posed with his
49
Ibid., 1–23, 28–29.
50
Wilfrid Sellars, “Cassirer Lecture Notes,” in Kant’s Transcendental Metaphysics: Sellars’ Cassirer Lectures
Notes and Other Essays, ed. Jeffrey Sicha (Atascadera, Cal.: Ridgeview, 2002), 478–79.
51
Ibid., 480.
52
Ibid., 481, 484.
53
Remarks from the unpublished first part of “Some Remarks on Kant’s Theory of Experience,” cited by
the editor in Sellars, Kant’s Transcendental Metaphysics, 235–36.
54
Ludwig Wittgenstein, Philosophische Untersuchungen (Frankfurt am Main: Suhrkamp, 1967), 13–15.
55
Robert Stern, Transcendental Arguments and Scepticism: Answering the Question of Justification
(Oxford: Oxford University Press, 2000), 10.
750 M. Vater
universal doubt, to Hume’s difficulties, which may merely undercut the

reliability of our knowledge claims or pose a more holistic challenge to
integrity of our empirical cognition. Modern commentators tend to go
with the third alternative, interpreting Kant’s transcendental argument as a
systematic justification of the knowledge claims of experience, or at least as a
strenuous effort in that direction.56 Hume’s attack on causality as neither a
self-evident conceptual relation nor an idea derived from experience is what
motivates Kant’s theory, and the mentalistic suppositions that his transcen-
dental theory introduces are the only way open to Kant to answer the double
challenge posed by both rationalist and empiricist skepticism.57 Remarks
Stern makes in passing on the job of the philosophical commentator shed
some light on our difficulties with Kant: one can abstractly distinguish critical
readings of a writer and charitable ones, but with Kant, apparently, we need
to simultaneously blue-pencil errors and offer charitable rewrites.
Working with “the crooked wood of humanity”58

If the justification of our cognitive frameworks was the organizing question
of the decades devoted to elaborating the Critiques, the question of human-
kind was that of the final decade of Kant’s writing. Rationality may shape
some basic endeavors and even take on an institutional guise of sovereignty
or rule-governed arrangements, but in us humans taken singly, reason is
often lodged with rude companions: triviality, habit, and convention on one
side; superstition, fanaticism, and madness on the other.59 Kant was justly
concerned with human improvement, but hardly naive about the likelihood
of the Enlightenment achieving its agenda. His starting point, the question of
humanity – certainly not a “doctrine of humanity” – provides guidance and
caution for subsequent thinkers. As Michel Foucault remarks, the precar-
iousness of Kant’s anthropological thought stems from the way it presumes
but cannot include critique, because its theme of finitude cannot be posed
either there or in the classical framework of meditations on God, world, and
humankind that was undertaken by Spinoza.60
56
Ibid., 35.
57
Ibid., 199–200.
58
IUH 8:23, translation modified.
59
See Kant’s remarks on the powers of pleasure/displeasure and desire at An 7:230–82.
60
Michel Foucault, Introduction to Kant’s “Anthropology,” trans. Roberto Nigro and Kate Briggs, ed.
Robert Nigro (Los Angeles: Semiotext(e), 2008), 118–21.
My remarks will be slender on this part of Kant’s reception, but one must
mention two works that introduced a simpler version of Kant’s thinking to
the contemporary audience and won for it some degree of acceptance. In
1786–1787, Karl Reinhold presented a popular discussion of Kant’s philo-
sophy in a series of letters in Der Teutsche Merkur, which were later expanded
and put in book form. Kant expressed admiration for the relatively painless
way that they integrated the technicalities of his philosophy with broader
Enlightenment concerns. Letters on the Kantian Philosophy (1790) secured
Reinhold’s appointment to the University of Jena and made that institution
the center of Kantian studies for the next two decades. Unlike the
“Elementary Philosophy” of On the Foundation of Philosophical Knowledge
(1791), which systematically derived Kant’s technical views from a single
principle, the Letters expands points Kant himself made about religion and
the history of philosophy. In Reinhold’s eyes, Kant solved the quandary
about faith and reason by putting the objects of religion beyond philosophi-
cal proof – or disproof. And if in Kant’s own view the Critique settled the
dispute between rationalism and empiricism by negotiating a territorial
compromise, Reinhold enlarges that move into a general technique that
views the history of philosophy as a narrative of warring claims and nego-
tiated settlements, or the resolution of antitheses into a synthesis.61
Reinhold appealed to the “needs of the time” in presenting critical
philosophy as particularly suited to decide the religious situation of the late
Enlightenment, which put philosophical rationalism in contention with
enthusiasm or faith without content.62 The poet Friedrich Schiller makes a
similar claim in a series of letters originally penned in 1793 for a Danish
prince, but enlarged for publication in 1795 in Schiller’s journal Die Horen.
Reacting to the bloody course of the French Revolution but inspired by the
ideal of a harmonious human spirit which he got from his reading of the
Critique of the Power of Judgment, Schiller’s On the Aesthetic Education of Man
explicitly turns readers’ attention to a Kantian anthropology and to its fruits
in education, politics, and morals.63 There are two impulses at work in the
human as she comes from nature: a sensuous impulse and an intellectual one.
Before freedom can emerge, there must be a period of training for freedom
that involves not the subjection of the sensuous to the intellectual, but their
61
Karl Leonhard Reinhold, Letters on the Kantian Philosophy, trans. James Hebbeler, ed. Karl Ameriks
62
Ibid., 1–17.
63
Friedrich Schiller, On the Aesthetic Education of Man in a Series of Letters, trans. Reginald Snell (New
York: Ungar, 1965), 38–47.
752 M. Vater
unrestricted interplay in aesthetic imagination. The aesthetic realm involves

the free creation of appearance, and only through this play at freedom is the
human spirit readied for the real work of moral freedom.64 The poet
exchanges the historical order of aesthetic and moral considerations in the
Kantian Critiques but gains a narrative of human development more inter-
esting than Kant’s dry list of powers: cognition, pleasure/displeasure, and
desire. Humankind comes from nature and is destined to be a free self-
creator or co-creator, but only by means of the conquest of self and appear-
ances involved in aesthetic play. Schiller’s account lets Kant’s preoccupation
with the epistemic and the question of objective grounding drop away, so
that only human, political, and pedagogic concerns remain about the use of
appearances and the prospect for human freedom.
Phenomenology and fundamental ontology: Kant’s epistemic logic is simple:
concepts are empty representations, or complexes of abstract signs; intuitions
directly present content, whether that be sensuous or relational. The propo-
sition is combinatory, either analysis of a complex into its components, or a
synthetic aggregation of simples into something complex. Absent intuitions,
concepts are empty, and absent concepts, sensations are blind (A51/B75).
There is no doctrine of signs, no analysis of signification, and no considera-
tion of meanings apart from the things of appearance. But matters would not
stay this simple. Reinhold’s analysis of consciousness discovered three items:
a subject, joined to an object, and an awareness which both distinguishes and
joins the other two. In 1812 Fichte maintained in his Transcendental Logic
that thinking involves three elements: an image, an object which the image
expresses, and space or matter for an intuition, with correspondingly more
abstract elements at the level of a concept, which is an image of an image.
The term “expression” here indicates a semantical or intentional space
between image and object. The same year, Hegel’s Science of Logic makes
self-thinking, self-specifying universal thought the motor for the develop-
ment of cognitive content and insists that the individual thinker/actor
(Kant’s subject of experience) is carried along on a universal current of
thinking like a leaf on the stream.65
The idealists thus opened the door to a marriage of logic and psychology
that finds the universe of meanings constituted by subjective mental acts as
well as sensory or experiential content. Edmund Husserl’s Logical
64
Ibid., 91–102.
See Michael Vater, “Thought and Being: The Beginnings of the Wissen(schaftslehre) in Fichte and
65
Hegel (1812),” in New Perspectives on Fichte, ed. Tom Rockmore and Daniel Breazeale (Atlantic
Highlands, N.J.: Humanities, 1996), 107–26.
Investigations (1900) explores the difference that consciousness and its modes
of “taking” (e.g., belief, disbelief, expectation, and assurance) make to the
apprehension of entities on all levels, both things and items of thought.
Husserl strikes an anti-Kantian pose, rejecting his designation of sensibility
and understanding as essential human faculties; faculty-talk is just a whisker
awaiting Ockham’s razor, as one can see if one pretends to explain dancing
by a “faculty of dancing.” In his view, Kant’s logic is but rewarmed Aristotle,
and its claimed formality just the lack of objectivity that he and Johann
Friedrich Herbart took to be the hallmark of the conceptual. Husserl puts his
studies of meaning and reference, abstraction, intentional objects, and phe-
nomenology of knowledge under the rubric of pure logic as his way of
distancing himself from empirical psychology. But he is confident that
Kant’s focus on the objectivity of cognition oversimplified the account of
human mental life.66
There is more detail in the accounts of intentionality and phenomenology of
knowledge which close the Investigations than can easily be summarized.
“Intentionality” connotes reference to an object of consciousness; it is spoken
of as an “act of consciousness” to indicate that it is constituted by the subject,
not the object – there is no connotation of mental activity involved.67 Where
Reinhold’s theory of conscious representation simply distinguished mental state
and conscious awareness of it, as if consciousness was an immediate “black-box”
response to the presence of an object, Husserl finds a nest of complex factors on
both sides of the distinction. In any case of cognition, the subject may consider
the contents fully or partially – as occupying the foreground or the background,
as characterized by complete predicates or incomplete ones – and though some
parts of a presentation may carry the feature of objective reference, only as fully
assembled is there an object of reference. On the side of consciousness, Husserl
speaks of many-rayed and single-rayed intentionality; acts which present objects
may be simple or complex, foundational or founded. Cognition is not a single
thing; although there can be simple cases like naming that join perceptions to
linguistic signs, even such simple cases are mediated by judgment or recognition
(Erkennen).68 Everywhere in mental life, complexity is behind the scenes.
Although Kant never approached the topic of linguistic signs, meaning and
reference, incorporation of the findings of semantics or a study of intentionality
can enrich his account of the conditions of objective cognition. Knowing is
66
Edmund Husserl, Logical Investigations, trans. J. N. Findlay, 2 vols. (New York: Humanities, 1970),
2:833–34; cf. 457–59.
67
Ibid., 2:562–63.
68
Ibid., 2:620–35.
754 M. Vater
surrounded by a corona of attitudes (e.g., belief or doubt, expectation or

hesitation) which are general and not just states of the empirical subject. As
Foucault evaluates Husserl’s contribution, the detail of the attitudes discovered
is less important than their cause – the inability to provide the real ground of the
a priori, which if it cannot find berth in simple subjectivity, pours over into “the
density of syntheses and the already there.”69
Two years after publishing Being and Time (1927), Martin Heidegger turned
to Kant’s Critique to elucidate his ideas on fundamental ontology, human
existence, and temporality, sketched against their inauthentic counterparts
that he locates in “metaphysics” and the descriptive anthropology which is
the subject of Kant’s late lectures and notes. Rejecting the dominant tradition of
reading the Critique as legitimizing cognitive frameworks or offering a theory of
experience, Heidegger describes Kant’s realm of interest as “special metaphy-
sics,” or the relation of human reality to transcendence, “a transcendental
‘analytic’ of the subjectivity of the subject.”70 As Heidegger reads it, the
Critique offers a reconstruction of human cognition that positively advances a
third capacity in human reality as the ground of cognition besides sensibility
and understanding, namely imagination; it is in fact the fundamental capacity,
the hidden “common root” hinted at in the work’s first words (A15/B29).71
The three forms of intellectual synthesis in the A-version of the Transcendental
Deduction – apprehension, reproduction, and recognition – correspond to the
three modes of ecstatic temporality, so that time and imagination are virtually
convertible terms.72 The original interweaving of activity and receptivity that is
imaginative synthesis is the temporization of time itself. Imagination is also the
basis of the forms of sensible intuition, since the spatial relations of contiguity
vis-à-vis being-at-a-distance and temporal succession are internally constructed,
not furnished by experience. And the ultimate fit of the categories of relation
(causality, substance/accident, and reciprocal interaction) is determined by their
temporal application, as Kant described the process in the Schematism and the
Analogies of Experience.73
In 1929 Heidegger did not sharply distinguish between ontology, meta-
physics, and transcendence. He thinks that the lived unity of the three
69
Foucault, Introduction to Kant’s “Anthropology,” 107.
70
Martin Heidegger, Kant and the Problem of Metaphysics, trans. James S. Churchill (Bloomington:
Indiana University Press, 1962), 226; cf. the redefinition of “transcendence” as human comprehension
of being rather than “objects of cognition” (20).
71
Ibid., 93–94.
72
Ibid., 184–93.
73
Ibid., 144–54.
temporal dimensions is the meaning of transcendence, and human reality to

be the essential uncovering of meaning. Time, imagination, and human
reality mirror one another in the only possible metaphysics that can be
constructed on Kantian foundations. But ultimately that Kantian foundation
is static; even its sense of time is stationary, lacking the dynamic of move-
ment imparted by futurity of the ever-lapsing present. For Heidegger, human
reality is hyper-metaphysical, marked by a questioning of self and a knowl-
edge that is interrogation, not restatement of fact. Care, anxiety, flight to
futurity, and an ultimate hollowing out of the beings that mere experience
displays are the true touchstone of being.74 These features point to a basic
finitude at the core of being: possibility without guarantee, a recapitulation
perhaps of the spontaneity-receptivity of the tensed framework of being, but
groundless or without permanence. Kant’s final question – What is man? (JL
9:24–25) – resounds in Heidegger in a way that the author of Anthropology
from a Pragmatic Point of View could not anticipate, except perhaps in those
troubling passages where madness and biological frailty are shown to fac-
tually bookend our rationality.
Cassirer’s philosophy of culture: Early in his career, Ernst Cassirer sought
to integrate the discoveries of relativity theory and quantum physics into a
Kantian philosophy of science. From 1923 until his death in America,
Cassirer elaborated a philosophy of symbolic forms. He points to the
embodied intelligence of the human community and the fluid creativity
of the intellectual, social, and communicative structures they embody,
perhaps something akin to Wittgenstein’s “forms of life.” Cassirer’s reflec-
tions on language, myth, religion, art, history, and science assign a
fundamental but plastic creativity to human intelligence and sociality.
The theory incorporates Kant’s view of overlapping human powers of
cognition, pleasure and desire, and the ideal of a subjective harmonization
of understanding and imagination that he posed as the definition of
aesthetic judgment (FI 20:205–8, 245–47, 221–25).
The various symbolic forms that Cassirer investigates have a common
function, expression, or “configuration toward being,” most readily apparent
in the flexible linguistic sign that designates without copying a single item,
and is not a receptive impression but an impressing of the mark of our
spontaneity upon variable contents.75 It is difficult at first for Cassirer’s
74
Ibid., 240–47.
75
Ernst Cassirer, The Philosophy of Symbolic Forms, vol. 1: Language, trans. Ralph Manheim (New
Haven: Yale University Press, 1955), 89, 98–99.
756 M. Vater
reader to discern both the unity and diversity of the “forms” he studies, for
each one distills the whole of the human reach for meaning but is at the same
time constituted by its difference from the others. Each single form posits the
whole of consciousness.76 From the author’s side, one can explain the multi-
plicity from Cassirer’s allegiance to a broader methodology of “Copernican
revolution” that Cassirer sees in the unfolding of the Kantian corpus, where a
progressive identification of different functions of human reality specifies the
order of investigations and works Kant produces, not some preexisting list of
different substances or subject-matters. The critical-idealist project of articu-
lating the human spirit came to Kant gradually, unlike Pascal’s one-night
stand of fiery vision from beyond.77
One can best see how a symbolic form functions if one turns from the
cognitive domain, where we think we understand what we are doing, to
more archaic forms such as myth and religion. It is not the case that we lift
empirical contents from their proper context and drop them into alien
dress. Our whole experience is distilled in the mythical form, and it is not a
question of our capturing the content or the adequacy of the capture, but of
the quality of attention we direct to phenomena. Humans have objects only
insofar as they inhabit these life forms.78 In their very plurality and
irreducibility – what could be more different than the spheres of art and
science, or history and mythology? – they form an interlinked chain, not of
substances, but of functions.79
Public intellectuals: Two recent thinkers who take up Kant’s project of insis-
tently questioning the place of humankind achieved public voices considerably
more influential than academics usually attain. Michel Foucault submitted the
usual works for the doctorat d’Etat in 1964, Madness and Civilization and a
translation of Kant’s Anthropology. His committee recognized the core of an
independent work in the translation’s introduction; after months of meditation
on Diego Velázquez’s Las Meninas, Foucault transformed it into The Order of
Things (1966).80 What is novel in Foucault’s treatment of human culture is the
suggestion that the human subject is absent in the very disciplines and activities that
76
Ibid., 98.
77
Ibid., 78–79.
78
Ernst Cassirer, Language and Myth, trans. Susanne K. Langer (New York: Dover, 1953), 19, 66.
79
Ernst Cassirer, An Essay on Man: An Introduction to a Philosophy of Human Culture (New Haven: Yale
80
In the painting there is a play between representing, viewing, and mirroring: a representation of
representation. See Michel Foucault, The Order of Things: An Archaeology of the Human Sciences, trans.
Alan Sheridan (New York: Vintage, 1970), 10–16.
support human life, as the painter suggests in placing the “real” subject – the
Infanta’s mirrored royal parents or the viewer who should appear in that mirror –
outside of the painting and putting the entourage in its center. Our life is
supported by currents of thought and activity that are impersonal, pre-reflexive,
and impervious to philosophical analysis. Language, biology, and economic
exchange are the ways we currently understand our reality, but in them the
thoughts of individuals and the actions of identifiable communities sink to
intellectual insignificance, for their intelligibility resides elsewhere in autonomous
and unmanageable institutions of knowledge that surround the human stage. As
in the Velázquez painting, the sovereign subject is missing even in the midst of
those academic “maids of honor” who curate our understanding of human
existence. In ceaselessly ordering and reordering things, imagination and under-
standing – or “representation represented” – generate shifting schemata of knowl-
edge, micro-histories that fail to exhibit a center or an enduring reality.81
What is Kantian in this picture? There is no trace of Kant’s constitutive
reason, no insistence that the life of cognition, desire, and pleasure are
regulated by some normative power. Led instead by the Anthropology,
Foucault finds that individuals and their capacities are randomly placed in
situations where a mixture of rationality and irrationality prevails, where
attempts to address human needs are compromised by limiting, if not
antagonistic, tendencies. The peculiarity of human culture, underscored by
both Kant and Foucault, is the contingency of prevailing arrangements or the
historical conditions that have brought them about, and the possibility that
other ways of arranging human affairs might produce better outcomes. That
morals are communicated (and undermined) by situations where manners or
custom are the main educator is almost a strike-through that Kant writes
across his own rigorous prescriptivism. That he prized Königsberg as the
ideal confluence of culture, commerce, and education – almost a second
Athens – brings smiles to the lips of many readers (An 7:120n, 146–53).
Both Kant and Foucault value the way that ephemeral cultural arrangements
transmit normative reason precisely because of the oddity of the arrange-
ment. Perhaps Foucault presents a reprise of Schiller’s vision of an ultimate
harmonization of sensuous and intellectual human capacities, led not by the
individual poet’s (or ideologue’s) imagination, but by the incessant novelty of
mass culture and communication.
While Foucault uses Kant to locate our present reality, Hannah Arendt
uses Kant to look back to the opening of our traditions in Greek thought.
81
Ibid., 70–71.
758 M. Vater
Both as a thinker in her own right and as one of Heidegger’s major heirs,
she casts a critical look at the culture and practice of philosophy and
focuses on the exceptional character of thinking, inevitably viewed in
modern times through the Kantian lens that marks cognition, willing,
and judging as discrete functions of reason. In her view, Kant was aware
of the difference between reason and intellect (Verstand), but chose the
latter and pursued questions that could be answered, deferring the
challenge of reason and its peculiar search for meaning – or thinking.82
Arendt never views thinking as anything utilitarian or narrowly focused
on finite tasks or contingent ends. Early on she distinguished political
action, the individual’s self-defining deed performed in the communal
theater of politics, from administration, recognizing that reason is pecu-
liarly about itself and self-actualizing.83
Arendt accentuates the self-concern and self-interrogation that are the
hallmarks of reason by looking back to the Platonic Socrates and the
peculiarly conscientious way he confronted the sophists’ sense of expe-
diency and relativity. A Socratic self is a “two in one”: an actor in a situation
and an observer endowed with aims and values; to live an honest life is
simply to keep the two yoked or harmonious. The rational conduct of life
demands that a person not just “do the necessary business,” but more or less
be an achieved congruence of possibilities, deeds, and ideals.84 Kant’s
abstract reason imposes the same demands on itself: it decrees that unity
shall be found in difference, harmony in disharmony, universality in
infinite diversity of detail, and singleness of willing amidst a diaspora of
situated choices.
Willing stands opposite mere thinking and remembrance in that its aim is
not merely to reflect the world or unify its image, but to transform it – to
turn a possible future into a “now.” Thinking requires flight from the world,
some refuge from ceaseless happening, whereas willing requires world
engagement.85 Post-Kantian German philosophy turned what for Kant was
a simple difference of jurisdiction in reason’s rule into something momen-
tous, independent, and coercive. From Hegel’s self-moving concept to
Schelling’s freedom as assertion of will in the face of nature’s order, to
Schopenhauer’s principle of irrational facticity, all the way to Heidegger’s
82
Hannah Arendt, The Life of the Mind, 2 vols. (New York: Harcourt Brace Jovanovich, 1978), 1:14–15.
83
Hannah Arendt, The Human Condition (Chicago: University of Chicago Press, 1958), 175–210.
84
Arendt, The Life of the Mind, 1:167–91.
85
Ibid., 2:28–39.
Gelassenheit (the will not to will), recent European thought embraced an

agent of historical movement beyond individual rational control or beyond
the assessment of conscience.86 Arendt died before she could produce a
volume on judging, but it seems clear that her treatment would have gone
beyond the limited cases of reflexive unification under concepts that Kant
considered in the Critique of the Power of Judgment. Thinking is essentially
judgmental, self-reflective, and evaluative. Nowhere did Arendt display her
Kantian bent more clearly than in her characterization of Adolf Eichmann,
the man who thought he was a good Kantian because he followed orders, as
simply lacking the capacity to think.87
Conclusion
In a concise note on the history of European philosophy that Friedrich
Nietzsche includes in Twilight of the Idols (1889) under the title “How the
‘Real World’ at Last Became a Myth,” the author assigns Kant to the
midpoint of the story of how the distinction between “real” and “appar-
ent” worlds evaporated. As Nietzsche views Kant, and especially his moral
philosophy, the essentialist sun of Plato’s idealism has become in him but
a pale light, shining through mist and skepticism, cold, sublime, and
Königsbergian.88 I would argue that Kant belongs at the end of
Nietzsche’s narrative, that the succeeding centuries have as little invali-
dated Kant’s relocation of the space of reasons to the incessant discourse
of the human community as they have invalidated the cosmological switch
from Ptolemaic to Galilean frameworks. Reason is fundamentally our
business, not something lent from elsewhere, but it is an organic process,
not a mechanism, and its maintenance and growth depends on a sort of
horticulture that is now not just dialogical and intergenerational, but
technological as well. As Cassirer frames it, Kant’s “Copernican revolu-
tion” is an ongoing project, not a dated museum specimen. If, as recent
thinkers such as Heidegger and Foucault note, we cannot account either
for the logical transcendence of the a priori or the peculiar finitude that
86
Ibid., 2:149–94.
87
Ibid., 1:3–6.
88
Friedrich Nietzsche, Twilight of the Idols, in Twilight of the Idols and The Anti-Christ, trans. R. J.
Hollingdale (Harmondsworth: Penguin, 1971), 40–41.
760 M. Vater
makes us such poor managers or custodians of human reality, that is

because we are their interface. Although humans find aesthetic satisfaction
in narratives with their closed parentheses of beginnings and endings,
philosophy can but view these items as punctuation, not part of the
smaller story that is its business to tell.
33
Kant, the Copernican Devolution,
and Real Metaphysics
Robert Hanna
Human reason has this peculiar fate in one species of its cognitions that it is
burdened with questions which it cannot dismiss, since they are given to it as
problems by the nature of reason itself, but which it also cannot answer, since
they transcend every capacity of human reason.
Reason falls into this perplexity through no fault of its own. It begins from
principles whose use is unavoidable in the course of experience and at the same
time sufficiently warranted by it. With these principles it rises (as its nature
also requires) ever higher, to more remote conditions. But since it becomes
aware in this way that its business must always remain incomplete because the
questions never cease, reason sees itself necessitated to take refuge in principles
that overstep all possible use in experience, and yet seem so unsuspicious that
even ordinary common sense agrees with them. But it thereby falls into
obscurity and contradictions, from which it can indeed surmise that it must
somewhere be proceeding on the ground of hidden errors; but it cannot
discover them, for the principles on which it is proceeding, since they surpass
the bounds of all experience, no longer recognize any touchstone of experi-
ence. The battlefield of these endless controversies is called metaphysics.
– Immanuel Kant, Critique of Pure Reason (Avii–viii)
R. Hanna (*)
Independent philosopher, Co-Director, The Contemporary Kantian Philosophy
Project, Boulder, USA
e-mail: Robert.Hanna@Colorado.EDU

762 R. Hanna
The central theme of this book is: realism about structure. The world has
a distinguished structure, a privileged description. For a representation to
be fully successful, truth is not enough; the representation must also use
the right concepts, so that its conceptual structure matches reality’s
structure. There is an objectively correct way to “write the book of the
world.” . . .
I connect structure to fundamentality. The joint-carving notions are the
fundamental notions; a fact is fundamental when it is stated in joint-carving
terms. A central task of metaphysics has always been to discern the ultimate or
fundamental reality underlying the appearances. I think of this task as the
investigation of reality’s structure.
– Theodore Sider, Writing the Book of the World ([Oxford:
Oxford University Press, 2011], vii)
Introduction
It is an ironic fact that philosophers who fail to take the history of philosophy
sufficiently seriously are doomed to repeat its errors.
As a striking case-in-point, contemporary Analytic metaphysics, for all its
logico-technical brilliance and its philosophical rigor,1 essentially amounts to
what I will call the Copernican Devolution, a retrograde evolution in philo-
sophy that brings us back, full-circle, to naive, pre-Kantian, pre-critical
conceptions of mind, knowledge, and world that are essentially Baconian,
Cartesian, Spinozist, and especially Leibnizian-Wolffian in nature.2
Characteristic of this contemporary philosophical backsliding are commit-
ments to noumenal realism in ontology, to Conceptualism about the nature
of mental representation, to a heavy reliance on modal logic as providing
1
The leading figures of Analytic metaphysics include David Lewis, David Chalmers, Kit Fine, John
Hawthorne, Theodore Sider, and Timothy Williamson; and some of its canonical texts are David Lewis,
On the Plurality of Worlds (Oxford: Blackwell, 1986); Sider, Writing the Book of the World; David J.
Chalmers, Constructing the World (Oxford: Oxford University Press, 2012); and Timothy Williamson
Modal Logic as Metaphysics (Oxford: Oxford University Press, 2013).
2
The Copernican Devolution also has a mirror reflection in contemporary Kant-scholarship. See, e.g.,
Anja Jauernig, The World According to Kant: Things in Themselves and Appearances in Kant’s Critical
Idealism (Oxford: Oxford University Press, forthcoming); Anja Jauernig, Thought and Cognition
According to Kant: Our Cognitive Access to Things in Themselves and Appearances in Kant’s Critical
Philosophy (Oxford: Oxford University Press, forthcoming); and Nicholas F. Stang, Kant’s Modal
Metaphysics (Oxford: Oxford University Press, 2016). It is tempting to give a philosophical-sociological
explanation of this trend as the natural result of PhD training in departments dominated by Analytic
metaphysicians, especially Princeton.
direct insight into the ultimate structure of noumenal reality, and to a

dogmatic scientific naturalism usually combined with scientific essentialism.
The Copernican Devolution is, in fact, a disastrously regressive turn in
philosophy. More specifically, contemporary Analytic metaphysicians really
and truly need to learn Kant’s eighteenth-century lessons (1) about the
inherent limits of human cognition and knowledge, (2) about the unsound-
ness of all possible ontological arguments from logical or analytic necessity to
actual or real existence, (3) about the essential cognitive-semantic difference
between (3i) mere logical, analytic (a.k.a. “weak metaphysical”) possibility
and (3ii) real, synthetic (a.k.a. “strong metaphysical”) possibility, and (4)
about the essential ontological difference between noumena and phenomena.
For without these insights, they have been, are, and forever will be inevitably
led into the very same “obscurity and contradictions” that beset classical
metaphysics prior to Kant (Avii).
But as they say, it’s an ill wind that blows nobody any good; that is, few
misfortunes are so bad that they do not have some unintended good side effects
for somebody. Hence, seeing the Copernican Devolution for what it really is, a
philosophically disastrous regression, makes it possible for us to provide a well-
focused re-characterization of Kant’s metaphysics in a contemporary context.
In this light, Kant’s critical metaphysics is decisively what I will call a
“real” (or, alternatively, “human-faced”) metaphysics, and correspondingly it
can be illuminatingly presented in terms that specially emphasize what I call
Kant’s “proto-critical” period in the late 1760s and early 1770s and also his
“post-critical” period in the late 1780s and 1790s, both of which are some-
what neglected or undervalued, even by contemporary Kantians. Looked at
this way, Kant’s real or human-faced metaphysics consists, fundamentally, of
the following six commitments:
(i) a strict evidential appeal to human experience, which I call the criterion of
phenomenological adequacy for metaphysical theories,
(ii) a radical epistemic agnosticism about both the nature and existence of
noumenal reality,
(iii) a thoroughgoing diagnostic critique of deep confusions in “ontological
argument”-style (and more generally, noumenal-metaphysical) reasoning
that is driven by modal logic,
(iv) a maximally strong version of Non-Conceptualism in the theory of
mental representation, and correspondingly, a direct argument for trans-
cendental idealism from the nature of human sensibility together with
Non-Conceptualism, that is essentially in place by the time of Kant’s
famous letter to Marcus Herz in 1772 (C 10:129–35),
764 R. Hanna
(v) modal dualism and apriorism (according to which there are two essen-
tially distinct types of necessity, both of which are irreducibly a priori,
combined with a strong commitment to the “necessity if and only if
apriority” thesis), and finally,
(vi) a theory of synthetic a priori truth and knowledge, grounded directly on
Non-Conceptualism.
I will briefly sketch, unpack, and defend each of these six Kantian
commitments.
In freely going back and forth between Kant’s philosophy and contem-
porary philosophy, I am applying the following strong metaphilosophical
principle, for which I have argued elsewhere,3 that I call The No-Deep-
Difference Thesis:
There is no fundamental difference in philosophical content between the

history of philosophy and contemporary philosophy.
In other words, in doing contemporary philosophy one is thereby directly

engaging with the history of philosophy, and in doing the history of philo-
sophy one is thereby directly engaging with contemporary philosophy. There
is no serious distinction to be drawn between the two.
The criterion of phenomenological adequacy for

metaphysical theories
In the B preface of the first Critique, Kant says that “there is no doubt that up
to now the procedure of metaphysics has been a mere groping, and what is
the worst, a groping among mere concepts [bloßen Begriffen]” (Bxv). A “mere
concept” is the same as an empty (leer) concept or noumenal concept, which
in turn is a concept that is minimally well-formed in both a formal-syntac-
tical and sortal sense, and also logically self-consistent, but essentially dis-
connected from human sensibility and actual or possible sensory intuition
and all its apparent or manifestly real natural objects, hence a concept that
does not have objective validity (objective Gültigkeit). Smoothly compatibly
3
See Robert Hanna, “Back to Kant: Teaching the First Critique as Contemporary Philosophy,” APA
Newsletter on Teaching Philosophy 8, no. 2 (Spring 2009): 2–6, http://c.ymcdn.com/sites/www.apaon
line.org/resource/collection/808CBF9D-D8E6-44A7-AE13-41A70645A525/v08n2Teaching.pdf.
with this Kantian critical line of thinking, Peter Unger titled his recent
critique of Analytic philosophy Empty Ideas.4
But according to Kant, real metaphysics must be evidentially grounded on
human experience. Or otherwise put, real metaphysics reverse-engineers its
basic metaphysical (including ontological) theses and explanations in order
to conform strictly to all and only what is phenomenologically self-evident in
human experience. By “phenomenologically self-evident” I mean this:
A claim C is phenomenologically self-evident for a rational human subject S if

and only if (i) S’s belief in C relies on directly-given conscious or self-conscious
manifest evidence about human experience, and (ii) C’s denial is either logically
or conceptually self-contradictory (i.e., a Kantian analytic self-contradiction),
really metaphysically impossible (i.e., it is a Kantian synthetic a priori impos-
sibility), or pragmatically self-stultifying for S (i.e., it is what Kant calls “a
contradiction in willing” in the Groundwork).
This leads directly to what I call the criterion of phenomenological adequacy for
metaphysical theories:
A metaphysical theory MT is phenomenologically adequate if and only if

MT is evidentially grounded on all and only phenomenologically self-
evident theses.
By this criterion, contemporary Analytic metaphysics is clearly phenomen-

ologically inadequate, and so is classical metaphysics more generally, whereas
by sharp contrast, Kant’s real metaphysics of transcendental idealism is,
arguably, fully phenomenologically adequate.
Kant’s radical agnosticism

According to Kant, both the origins and limits of human cognition or
Erkenntnis are determined by the nature of our specifically human sensibility
or Sinnlichkeit (B1, A19–49/B33–73). In particular, there is an inherent
cognitive-semantic constraint on all fully or “thickly” meaningful cognition:
a cognition is “objectively valid,” i.e., fully or “thickly” meaningful, if and
only if it presupposes actual or possible externally-triggered sensory intuitions
4
Peter Unger, Empty Ideas: A Critique of Analytic Philosophy (Oxford: Oxford University Press, 2014).
766 R. Hanna
or Anschauungen of empirical objects (A238–42/B298–300, A289/B345),

presented within the global, framing structures of egocentrically-centered,
orientable (i.e., it contains intrinsic enantiomorphic directions determined
by a subject embedded in the space or time) phenomenal space and time.
Empirical objects in this specific, anthropocentric sense are appearances
(Erscheinungen) or phenomena; by sharp contrast, objects of cognition which,
if they existed, would fall outside the scope of human sensibility, are mere
“entities of the understanding [Verstandeswesen]” or noumena (A235–60/
B294–315, esp. B306). In short, a noumenon, if it were to exist, would be
a non-sensory, non-empirical, non-spatiotemporal, trans-human object, a
supersensible object (A254–55/B355). If, in addition to being a noumenon
in this supersensible-object sense (a.k.a. “a noumenon in the negative sense”),
any noumenal object which, if it were to exist, would also be an individual
Cartesian/Leibnizian substance, whose nature is completely determined by
intrinsic non-relational properties, would be a “thing in itself [Ding an sich]”
(a.k.a. “a noumenon in the positive sense”) (B306–7).
Now I am being very careful about my formulations here, in two ways.
First, I am distinguishing between negative noumena (“supersensible”
objects in the minimal sense of non-sensory objects) and positive noumena
(things in themselves). Why? Because I think it is arguable that Kant held
that there are perfectly legitimate negative noumena, or supersensible/non-
sensory objects (e.g., abstract objects in the formal sciences, especially math-
ematical objects like numbers), that are NOT positive noumena or things in
themselves (e.g., God, immortal souls, etc.), and indeed are at least partially
constituted by (by being in necessary conformity with) the forms of our
sensible intuition, and hence they are thoroughly as-it-were phenoumenal.
Second, I am framing the concepts of a noumenon and of a thing in itself
counterfactually, hence NOT committing Kant to the claim that things in
themselves really exist.
Why? Because I think that we should not automatically assume that Kant
believes that noumena or things in themselves really exist. This of course is
one of the great controversies in Kant-interpretation. And my own view is
that Kant is in fact a “methodological eliminativist” about things in them-
selves (A30/B45, A255/B310, A286–87/B343). But at the very least, we
need to remain open-minded and not be dogmatic about Kant’s supposed
commitment to the real existence of things in themselves, especially given
Kant’s own deep and fully explicit insight about the basic ontological
distinction between (i) logically or analytically defined objects (merely think-
able objects), and (ii) actually or really existing objects (experienceable,
knowable objects).
Now back to Kant’s cognitive semantics. For Kant, a cognition is fully

meaningful if and only if it is empirically meaningful from the human stand-
point. Failing this, a cognition is “empty [Leer]” (A51/B75), and therefore it
not only (i) lacks a directly referential, empirical-intuitional grounding in
actually existing empirical objects, but also (ii) lacks a truth-value (hence it is
a “truth-value gap”) (A58/B83). Incidentally, element (i) is a crucial feature
of Kant’s famous critique of ontological arguments for God’s existence: all
such arguments lack a directly referential, empirical-intuitional grounding in
actually existing empirical objects; hence the predicate “exists,” as deployed
in such arguments, is merely a “logical” predicate, and not a “real” or
“determining” predicate. I will come back to Kant’s worries about ontologi-
cal arguments later.
Now it is important to recognize that, for Kant, “empty” cognition need
not necessarily be wholly meaningless, or nonsense: it can be partially or “thinly”
meaningful if (and only if) it is logically well-formed according to the logical
forms of judgment/categories, and also conceptually and/or logically consis-
tent (Bxxvi note, A239/B298). This is what Kant calls mere “thinking
[Denken],” according to concepts (Begriffen). In turn recognizing our natural
capacity for mere thinking is metaphilosophically important because mere
thinking is characteristic of classical metaphysics, and consequently also of
contemporary Analytic metaphysics.
Thinking about X establishes the logical or analytic possibility of X. But it
does NOT establish the real or synthetic possibility of X. Hence a crucial
mistake in classical metaphysics, and correspondingly a crucial mistake in
contemporary Analytic metaphysics, is to confuse logical or analytic possibi-
lity/necessity with real or synthetic possibility/necessity. As we will see in the
next section, this metaphysical confusion leads directly to deep “obscurity
and contradictions” (Aviii).
In other words, David Lewis, Kit Fine, David Chalmers, John Hawthorne,
Theodore Sider, and Timothy Williamson, for all their logico-technical
brilliance and their philosophical rigor, and even despite their high-powered
contemporary professional philosophical status, are every bit as confused and
wrongheaded as Christian Wolff. They make all the same old mistakes, just
as if they had never been made before. For example, when Sider asserts,
without any doubt, hesitation, or irony whatsoever, just as if the previous
235 years of European philosophy had never happened, that “the world has a
distinguished structure, a privileged description,” that “for a representation
to be fully successful, truth is not enough; the representation must also use
the right concepts, so that its conceptual structure matches reality’s struc-
ture,” and that “there is an objectively correct way to ‘write the book of the
768 R. Hanna
world,’” it simply takes your Kantian breath away.5 Amazing. That is the
Copernican Devolution. On the contrary, Kant holds that real metaphysics is
based fundamentally on reasoning with real or synthetic possibilities/neces-
sities, NOT on reasoning with logical or analytic possibilities/necessities.
In any case, the cognitive-semantic determination of full meaningfulness
of a cognition by sensibility, in turn, sharply constrains the scope of knowl-
edge in the strict sense of “scientific knowledge [Wissen]”: objectively con-
vincing true belief with certainty (A820–22/B848–50). Since strict or
scientific knowledge requires truth, but truth-valuedness requires objective
validity or empirical meaningfulness, then if a cognition is not objectively
valid/empirically meaningful, then it cannot be either true or false, and there-
fore it cannot be strict or scientific knowledge. In particular, it directly follows
from this point that in the strict or scientific sense of “knowledge,” we cannot
know things in themselves, either by knowing their nature, or by knowing
whether they exist or do not exist. In other words, we know a priori, by
reflection on the cognitive semantics of human cognition, that we cannot
have strict or scientific knowledge of things in themselves. This is what I call
Kant’s radical agnosticism – “radical,” because unlike ordinary agnosticism
(epistemic open-mindedness or doxic neutrality about some claim C), it is
strict or scientific a priori knowledge about our necessary ignorance of things
in themselves, and about our necessary inability to know or prove whether
things in themselves (e.g., God) exist or do not exist.
Given the truth of Kantian radical agnosticism, it directly follows that
neither classical Rationalist metaphysics nor contemporary Analytic meta-
physics, since they are based on mere thinking alone, and reasoning from
mere logical or analytic possibilities, is capable of having strict or scientific
knowledge, despite all their highly technically sophisticated, rigorous-
sounding, dogmatic claims about knowledge of things in themselves.
Moreover, it also directly follows from Kantian radical agnosticism that
any claim in speculative natural science that violates the cognitive-semantic
constraints on strict or scientific knowledge (e.g., any natural-scientific
claim about positive noumenal entities belonging to microphysical
essences, e.g., molecules, atoms, quarks, neutrinos, etc., etc.), is a truth-
value gap. Hence any form of metaphysical noumenal realism in natural
science is deeply mistaken.6
5
Sider, Writing the Book of the World, vii.
6
See Robert Hanna, Kant, Science, and Human Nature (Oxford: Oxford University Press, 2006), chs. 3–4.
Kant’s critique of modal metaphysics

Kant’s critique of modal metaphysics is based on his radical agnosticism and
his cognitive semantics. The core of Kant’s critique is that it is a philosophi-
cally tragic mistake to make a naive use of modal logic for ontological and
metaphysical purposes. Indeed, that is the core philosophical message of the
entire Transcendental Dialectic (A293–704/B349–732).
In particular, modal metaphysicians make the following three basic errors:
(i) they confuse the purely logical or analytic sense of the predicate “exists”
with the metaphysically real or synthetic sense of the predicate “exists”
(A592–602/B620–30), (ii) they confuse purely logical or analytic possibility
with real or synthetic possibility and therefore correspondingly also confuse
purely logical or analytic necessity (a.k.a. “weak metaphysical necessity”) with
real or synthetic necessity (a.k.a. “strong metaphysical necessity”) (A226–35/
B279–87), and (iii) they confuse the purely logical or analytic notion of an
object with the real or synthetic notion of an object.
Re (i). The purely logical notion of existence is particular quantification,
construed as a second-order concept: “X exists” is a judgment or proposition
containing a first-order concept X and a second-order concept, “exists,” which
says that some object falls under that first-order concept X, in some domain of
objects, x1, x2, x3, etc. But real existence or actual existence, Realität or
Wirklichkeit, i.e., the schematized pure concept or category of existence, requires
the intuition of the object, and more specifically it requires the empirical
intuition of a causally relevant or efficacious object in egocentrically-centered,
orientable, phenomenal space and time (A225–26/B272–74). But these are
non-logical, non-analytic, and essentially non-conceptual cognitive-semantic
facts. Hence it is a modal-metaphysical fallacy to infer from purely logical,
analytic, or conceptual proofs of existence to real, synthetic, or intuitional
(=essentially non-conceptual) proofs of existence (A258–59/B313–15).
And this is as true of logically necessary existence as it is of logically contingent
existence. Truth of a proposition in all logically possible worlds does NOT
establish real or actual existence in the actual, real world of rational human
experience, unless the objective validity of the constituent concepts in the
proposition has already been established. Hence, for example, all ontological
proofs of God’s existence fail. But the more general point against modal
metaphysics is that ANY kind of ontological proof in modal logic cannot
discriminate between noumenal (supersensible, divine-standpoint) existence
and phenomenal (experiential, human-standpoint) existence.
770 R. Hanna
Now, given Kantian radical agnosticism, we know a priori that noumenal

existence, that is, Cartesian-Leibnizian style substances with intrinsic non-
relational essences, cannot be known in the strict or scientific sense.
Therefore, ALL ontological proofs in modal metaphysics are unsound, and
what is even worse, if they are falsely assumed to be sound, then they lead
directly to hyper-contradictions or paradoxes, that is, antinomies of pure
reason (A405–567/B432–595).
Re (ii). Logical or analytic possibility establishes only the pure logical or
analytic consistency of a set of concepts or properties in what Kant calls pure
general logic, namely second-order intensional monadic predicate logic (A52–
55/B76–79).7 But in order for something to be really possible, or syntheti-
cally possible, it must also be shown to be empirically intuitable – but this
cannot be established through pure general logic or concepts alone.
This in turn indicates a subtle but absolutely fundamental point about
Kant’s theory of logical truth: because a proposition cannot be either true or
false unless all its constituent concepts are objectively valid, even the truths of
pure general logic are not “topic-neutral.” The “generality,” or “formality” of
pure general logic for Kant (A54/B78) does NOT mean that logical truths
are semantically insensitive to domains of truth-making objects, or “models.”
Radically on the contrary, Kantian pure logical generality means that a
logical truth is such that its meaning and truth are necessarily underdeter-
mined by whatever objects occur in a given logically possible world. This is
because every logical truth is objectively valid, hence every logically possible
world contains some possible objects of human experience. Indeed, for Kant,
a possible world is simply a maximally consistent set of different conceivable
ways the actual world of human experience could have been (A571–74/
B579–602). Hence even a logical truth, like all truth (A58/B52), consists in
an “agreement,” conformity, or correspondence between the formal content
of a pure-logical proposition and its proper “objects,” that is, a cross-world
semantic configuration consisting of multiple self-contained, other-contain-
ing, or identical/perfectly overlapping sets of objects in possible worlds,
across all possible worlds.
This is sharply different from the conceptions of logical generality, logical
formality, and logical truth in conventionalist accounts of the foundations of
logic, which still dominate in contemporary Analytic metaphysics, even
7
See Robert Hanna, “Kant’s Theory of Judgment,” in Stanford Encyclopedia of Philosophy, ed. Edward
N. Zalta (Fall 2013), sect. 2.1 and supplement 3, http://plato.stanford.edu/archives/fall2013/entries/
kant-judgment/; and Robert Hanna, “Jäsche Logic,” in The Cambridge Kant Lexicon, ed. Julian Wuerth
(Cambridge: Cambridge University Press, 2014), http://cambridgekantlexicon.com/.
though the conventionalist theory of logical truth was in fact decisively

criticized and refuted by Quine in “Truth by Convention” in the mid-
1930s.8 The problem is that post-Quinean philosophers of logic have simply
not been able to find a more adequate account of the foundations of logic
since Carnap. So for the last eighty years, they have simply rolled along as if
the conventionalist theory were somehow minimally adequate. But if I am
correct, then the only philosophically adequate account of the foundations of
logic, grounds logic on human rationality, in a broadly Kantian way.9
In any case, from a Kantian logical point of view, it is a basic modal-
metaphysical fallacy to infer from purely logical or analytic possibility to real
or synthetic possibility. But since logical necessity is defined in terms of
logical possibility (i.e., “necessarily P” is necessarily equivalent with “not
possibly not P”) this means that it is also a modal metaphysical fallacy to infer
from purely logical or analytic necessity to real or synthetic necessity. More
specifically, a purely logical or analytic proof of necessity cannot discriminate
between noumenal necessity (the necessity of unknowable, unperceivable,
causally irrelevant and causally inert, intrinsic non-relational essences) and
phenomenal necessity (the necessity of knowable, perceivable, causally rele-
vant and causally efficacious, intrinsic relational spacetime essences).
Therefore, ALL proofs of necessity and essence in modal metaphysics – for
example, Saul Kripke’s famous proof for truths of scientific essentialism, via
his modal logical proof of the necessity of identity10 – are unsound, and as
per the Transcendental Dialectic, if they are assumed to be sound, then they
lead to hyper-contradictions or logical paradoxes.
Re (iii). As we have seen already, the purely logical or analytic notion of an
object, as a mere bundle of mutually logically or conceptually consistent
concepts or properties (a.k.a. the thin notion of an object), is essentially
different from the real or synthetic notion of an object, as an empirically
intuitable, causally relevant, and causally efficacious object in egocentrically-
centered, orientable phenomenal space and time (a.k.a. the thick notion of an
object). Hence it is a modal metaphysical fallacy to infer directly from the
purely logical or analytic properties of “thin” objects to the real or synthetic
properties of “thick” objects.
8
W. V. O. Quine, “Truth by Convention” (1936), in The Ways of Paradox and Other Essays, rev. ed.
(Cambridge: Harvard University Press, 1976), 77–106.
9
See Robert Hanna, Rationality and Logic (Cambridge: MIT Press, 2006).
10
See Saul Kripke, “Identity and Necessity,” in Meaning and Reference, ed. A. W. Moore (Oxford:
Oxford University Press, 1993), 162–91; and Saul Kripke, Naming and Necessity, 2nd ed. (Cambridge:
Harvard University Press, 1980).
772 R. Hanna
This modal fallacy has fundamentally important philosophical implica-

tions for twentieth-century and contemporary Analytic philosophy, includ-
ing Analytic metaphysics. For example, Frege’s notion of a set (crucial to his
definition of “number”) based on the naive comprehension axiom (axiom 5
in Basic Laws of Arithmetic) failed to take account of this difference.11 Hence
it permitted the unrestricted formation of sets containing ANY sort of
objects, including noumenal or supersensible objects, and especially including
sets formed entirely of sets that are not members of themselves – but this led
directly to Russell’s Paradox, which says that necessarily, the set of all sets
that are not members of themselves, call it K, is a member of itself if and only
if it is not a member of itself. Russell’s Paradox, in turn, undermines classical
Frege-Russell Logicism, when it is taken together with Gödel’s incomplete-
ness theorems: (i) necessarily, Principia-style logical systems + the Peano
axioms for arithmetic are consistent if and only if they are incomplete (i.e.,
not all tautologies are theorems), and (ii) the truth-definitions of such
systems must lie outside those systems themselves. And this fundamental
negative result in twentieth-century philosophy – that classical Logicism is
false – is of course perfectly in line with Kant’s eighteenth-century anti-
Leibnizian thesis that arithmetic in particular and mathematics more gen-
erally, is NOT explanatorily reducible to logic (+/- polyadic quantification,
and +/- set theory). That is, mathematics is synthetic, not analytic.
Now it is important to see here that I am not saying that all or even most
contemporary Analytic metaphysicians are committed to classical Fregean or
even to neo-Fregean Logicism – on the contrary, all or most of them are not
so committed, or at least not explicitly so committed, although perhaps in
their philosophically romantic moments they may still quietly long for the
heady, good old days of Logicism, wine, and roses. All I am saying is that
Frege’s failed project of Logicism, at the very outset of the Analytic tradition,
is a particularly vivid example of the sort of catastrophic philosophical error
to which contemporary Analytic metaphysics is inherently subject. In short,
the philosophical project of modal metaphysics, whether classical
Rationalism or contemporary Analytic metaphysics, is doomed to tragic
failure.
And this failure is genuinely “tragic” for two reasons.
First, the collapse of modal metaphysics induces a serious meta-philoso-
phical skepticism about the prospects for ANY sort of metaphysics, thus
11
See Gottlob Frege, Basic Laws of Arithmetic, trans. and ed. Montgomery Furth (Berkeley: University
of California Press, 1964).
making philosophy itself seem empty, trivial, and “scholastic” in the worst
sense, nothing but a “glass bead game [Glasperlenspiel]” played by super-
clever professional academic philosophers, which, as Kant clearly saw, is the
death of philosophy (Avii–xii).
And second, the cognitive origins of what we can call “the drive to
modal metaphysics,” the conation of pure theoretical reason (Axii),
which is the fundamental rational human need for absolute grounding –
in effect, the intense desire to be like God and have godlike knowledge,
“intellectual intuition” (B71–73) – is a natural effect of the capacity for
human rationality (A293–98/B349–55). Indeed, this “drive to modal
metaphysics,” at bottom, is nothing but sublimated ethics and subli-
mated religion (Bxxv; Pro 4:350–65), disguising itself as “rigorous
science [strenge Wissenschaft].” Hence the drive to modal metaphysics is
not something that can be merely brushed off or laughed away: in fact, it
is a perpetual threat to the survival of philosophy as a fundamental,
rational human project, stemming from the will-to-philosophy itself, and
from human reason’s failure or refusal to recognize its own “human, all-
too-human” limits.
These deep Kantian metaphilosophical critical insights about the nat-
ure, errors/fallacies, and cognitive sources of modal metaphysics are very
similar to the later Wittgenstein’s critique of philosophy – and his
critique of the philosophical theories of his own earlier self, the author
of the Tractatus – in the Philosophical Investigations.12 Hence it is no
accident that contemporary Analytic metaphysicians normally completely
ignore the later Wittgenstein, and typically do not even consider him to
be a philosopher.
Kant’s Non-Conceptualism, Kantian Non-

Conceptualism, and Kant’s proto-critical argument
for transcendental idealism for sensibility
The four-year period from 1768 to 1772 was a fundamentally important
phase in Kant’s philosophical career – centered on the fully critical period
from 1781 to 1787 – so important, in fact, that I will call it his proto-critical
See Robert Hanna, “Wittgenstein and Kantianism,” in The Blackwell Companion to Wittgenstein, ed.
12
Hans-Johann Glock and John Hyman (Oxford: Blackwell, 2017), 682–98, sect. 3.
774 R. Hanna
period, in order to distinguish it sharply from both his dogmatic-slumber-

filled Leibnizian/Wolffian pre-critical period and also what I call his post-
critical period after 1787.
In this section, I want to demonstrate an essential connection that emerged
during Kant’s proto-critical period between his Non-Conceptualism and his
transcendental idealism, by tracing this line of thinking directly back to his
seminal proto-critical essay of 1768, “Concerning the Ultimate Ground of the
Differentiation of Directions in Space,” a.k.a. “Directions in Space” (DS
2:377–83).
I will begin by defining some technical terms: “transcendental idealism,”
“Conceptualism,” “Non-Conceptualism,” “Kantian Conceptualism,” and
“Kantian Non-Conceptualism.”
In a nutshell, Kant’s thesis of transcendental idealism says that the basic
structure of the apparent or phenomenal world necessarily conforms to the
pure or non-empirical (hence a priori) structure of human cognition, and
not the converse (Bxvi–xviii). In other words, Kant is saying that the
phenomenal world fundamentally conforms to the a priori structure of
the human mind, and it is not the case that the human mind fundamentally
conforms to the phenomenal world, or indeed to any non-apparent or
noumenal world.
And here is Kant’s primary argument for transcendental idealism. If the
human mind fundamentally conformed to the world, whether phenomenal
or noumenal, then since human knowledge of the world would be contin-
gent on the existence and specific character of that world, then a priori
human knowledge of the world would be impossible (C 10:130–31). But a
priori human knowledge of the phenomenal world (e.g., in mathematics) is
already actual and therefore really possible. So the phenomenal world neces-
sarily conforms to the a priori structure of the human mind. And in
particular, the phenomenal world fundamentally conforms to our a priori
representations of space and time, because that is the only acceptable philo-
sophical explanation of the real possibility of mathematical knowledge
(ID 2:398–406; A19–49/B33–73).
So if Kant is correct, then he is saying that the world in which we live,
move, and have our being (by which I mean the phenomenal natural and
social world of our ordinary human existence) is fundamentally depen-
dent on our minded nature, and not the converse. Correct or incorrect,
transcendental idealism seems to me to be a deeply important philoso-
phical thesis. For one thing, if transcendental idealism is true, then we
cannot be inherently alienated from the world we are trying to know, as
global epistemic skeptics claim, and human knowledge – not only
a priori knowledge, but also a posteriori knowledge – is therefore really

possible.13
In general, the thesis of Conceptualism14 says that the representational
content of human cognition is essentially conceptual, and necessarily deter-
mined by our conceptual capacities. Strong Conceptualism says that our
conceptual capacities are not only necessary but also sufficient for determin-
ing the content of human cognition, and weak Conceptualism says that our
conceptual capacities are not alone sufficient but also require a contribution
from some or another non-conceptual capacity (e.g., the capacity for sense
perception) in order to determine the (ultimately conceptual) content of
human cognition. Correspondingly, the thesis of (essentialist content) Non-
Conceptualism15 says that at least some of the representational contents of
human cognition are not essentially conceptual, and not necessarily deter-
mined by our conceptual capacities, and also that these contents, on the
contrary, are essentially non-conceptual, and necessarily determined by our
non-conceptual capacities (e.g., the capacity for sense perception).
Although these distinctions might initially seem Scholastic in the bad
sense and trivial, the opposition between Conceptualism and (essentialist
content) Non-Conceptualism is philosophically important. This is because
what is at issue is nothing more and nothing less than the nature of the
human mind. According to Conceptualism, human minds are basically
13
Robert Hanna, Cognition, Content, and the A Priori: A Study in the Philosophy of Mind and Knowledge
(Oxford: Oxford University Press, 2015), esp. chs. 3, 6–8.
14
See, e.g., Wilfrid Sellars, “Empiricism and the Philosophy of Mind,” in Science, Perception, and Reality
(New York: Humanities, 1963), 127–196; Wilfrid Sellars, Science and Metaphysics: Variations on
Kantian Themes (London: Routledge & Kegan Paul, 1968); and John McDowell, Mind and World
(Cambridge: Harvard University Press, 1994).
15
See, e.g., Gareth Evans, The Varieties of Reference, ed. John McDowell (Oxford: Clarendon, 1982). In
the contemporary debate about Conceptualism versus Non-Conceptualism, it is now standard to draw a
distinction between state (or possession-theoretic) Non-Conceptualism and content Non-
Conceptualism. State Non-Conceptualism says that there are mental states for which the subject of
those states fails to possess concepts for the specification of those states. Content Non-Conceptualism,
by contrast, says that some mental states have content that is of a different kind from that of conceptual
content. In turn, essentialist content Non-Conceptualism says that the content of such states is of a
categorically or essentially different kind from that of conceptual content. For a general survey of Non-
Conceptualism, see José Bermúdez and Arnon Cahen, “Nonconceptual Mental Content,” in Stanford
Encyclopedia of Philosophy, ed. Edward N. Zalta (Spring 2012), http://plato.stanford.edu/archives/
spr2012/entries/content-nonconceptual/. For the distinction between state and content Non-
Conceptualism, see Richard G. Heck, Jr., “Nonconceptual Content and the ‘Space of Reasons,’”
Philosophical Review 109, no. 4 (Oct. 2000): 483–523. And for the distinction between non-essentialist
and essentialist content Non-Conceptualism, see Robert Hanna, “Kantian Non-Conceptualism,”
Philosophical Studies 137, no. 1 (Jan. 2008): 41–64; Robert Hanna, “Beyond the Myth of the Myth
of the Given: A Kantian Theory of Non-Conceptual Content,” International Journal of Philosophical
Studies 19, no. 3 (2011): 323–98; and Hanna, Cognition, Content, and the A Priori, ch. 2.
776 R. Hanna
intellectual in character, having nothing inherently to do with the embodied,

sense-perceiving, affective, desiring, animal side of human nature. By con-
trast, according to (essentialist content) Non-Conceptualism, human minds
are basically bound up with the embodied, sense-perceiving, affective, desir-
ing, animal side of human nature, and are not basically intellectual in
character: on the contrary, the intellectual capacities of the human being
constitutively presuppose, and are thereby grounded on and built on top of,
the non-intellectual capacities. Hence the philosophical debate about
Conceptualism versus Non-Conceptualism is really a debate about whether
an intellectualist or a non-intellectualist conception of the human mind is the
correct one. This has far-reaching implications not only for other parts of the
philosophy of mind, but also for epistemology, metaphysics, ethics and
philosophical anthropology,16 and even political philosophy, to the extent
that it depends on ethics and philosophical anthropology.
Although both Conceptualism and (essentialist content) Non-
Conceptualism are competing theses/doctrines in contemporary philosophy
of mind, their philosophical origins both go back to Kant.17 Hence it is
possible to defend either Kantian Conceptualism or Kantian (essentialist
content) Non-Conceptualism, as competing interpretations of Kant’s theory
of human cognition in particular and of Kant’s philosophy of mind more
generally.
Now according to Kant, our conceptual capacities are located in the
understanding (Verstand), whose operations yield concepts, judgments/prop-
ositions, and inferences, when those operations are also supplemented by our
further intellectual capacities for apperception or self-consciousness, for
judgment and belief, and for logical reason or inference. By contrast, again
according to Kant, our non-conceptual capacities are located in the sensibility
(Sinnlichkeit), which contains both a non-intellectual sub-capacity for sense
perception and also a non-intellectual sub-capacity for imagination, and
whose operations yield material or formal intuitions, material images, and
formal images or schemata. Human sensibility for Kant, it must also be
noted, further contains non-intellectual sub-capacities for feeling, desiring,
and sensible willing or “the power of choice [Willkür].” In other words,
16
See, e.g., Robert Hanna, “Sensibility First: From Kantian Non-Conceptualism to Kantian Non-
Intellectualism,” unpublished ms. (Spring 2015), https://www.academia.edu/11942928/Sensibility_
First_From_Kantian_Non-Conceptualism_to_Kantian_Non-Intellectualism_Spring_2015_version_
comments_welcomed_.
17
See, e.g., McDowell, Mind and World; and Robert Hanna, “Kant and Nonconceptual Content,”
European Journal of Philosophy 13, no. 2 (Aug. 2005): 247–90.
sensibility for Kant is as much non-cognitive or practical, as it is cognitive or

theoretical.
Since Kant believes that the understanding and the sensibility, as capac-
ities, are essentially distinct from and irreducible to one another, and also
that both are required for rational human cognition (and in the case of
human practical reason, a.k.a. “the faculty of desire [Begehrungsvermögen],”
both are required for rational human action and agency), Kant is also a
cognitive capacity dualist.
But is Kant a Conceptualist or a Non-Conceptualist? Or in other words, is
Kant a cognitive content dualist as well as a cognitive capacity dualist? Or in
still other words, is Kant an intellectualist about the nature of the human
mind, or a non-intellectualist?
The intellectualist thesis of Kantian Conceptualism says that, for Kant, the
representational content of human cognition is essentially conceptual, and
necessarily determined by the understanding. And just as there are strong and
weak versions of Conceptualism in general, so too there are strong and weak
versions of Kantian Conceptualism.18 By contrast, the non-intellectualist
thesis of Kantian (essentialist content) Non-Conceptualism says that, for
Kant, at least some of the representational contents of human cognition are
not essentially conceptual and not necessarily determined by the under-
standing, and also that these contents, on the contrary, are essentially non-
conceptual and necessarily determined by our sensibility.19
18
See, e.g., Christian Helmut Wenzel, “Spielen nach Kant die Kategorien schon bei der Wahrnehmung
eine Rolle? Peter Rohs und John McDowell,” Kant Studien 96, no. 4 (Dec. 2005): 407–26; Hannah
Ginsborg, “Empirical Concepts and the Content of Experience,” European Journal of Philosophy 14,
no. 3 (Dec. 2006): 349–72; Hannah Ginsborg, “Was Kant a Nonconceptualist?” Philosophical Studies
137, no. 1 (Jan. 2008): 65–77; John McDowell, “Avoiding the Myth of the Given,” in Having the World
in View: Essays on Kant, Hegel, and Sellars (Cambridge: Harvard University Press, 2009), 256–72;
Stephanie Grüne, Blinde Anschauung: Die Rolle von Begriffen in Kants Theorie sinnlicher Synthesis
(Frankfurt am Main: Klostermann, 2009); Brady Bowman, “A Conceptualist Reply to Hanna’s
Kantian Non-Conceptualism,” International Journal of Philosophical Studies 19, no. 3 (2011): 417–46;
Thomas Land, “Kantian Conceptualism,” in Rethinking Epistemology, vol. 1, ed. Günter Abel and James
Conant (Berlin: de Gruyter, 2011), 197–239; Nathan Bauer, “A Peculiar Intuition: Kant’s
Conceptualist Account of Perception,” Inquiry 55, no. 3 (June 2012): 215–37; Aaron M. Griffith,
“Perception and the Categories: A Conceptualist Reading of Kant’s Critique of Pure Reason,” European
Journal of Philosophy 20, no. 2 (June 2012): 193–222; Jessica Williams, “How Conceptually-Guided Are
Kantian Intuitions?” History of Philosophy Quarterly 29, no. 1 (Jan. 2012): 57–78; John McDowell, “The
Myth of the Mind as Detached,” in Mind, Reason, and Being-in-the-World: The McDowell-Dreyfus
Debate, ed. Joseph K. Schear (London: Routledge, 2013), 41–58; Robert B. Pippin, “What Is
‘Conceptual Activity’?” in Mind, Reason, and Being-in-the-World, 91–109; and Sacha Golob, “Kant
on Intentionality, Magnitude, and the Unity of Perception,” European Journal of Philosophy 22, no. 4
(Dec. 2014): 505–28.
19
See, e.g, Hanna, “Kantian Non-Conceptualism”; Robert Hanna and Monima Chadha, “Non-
Conceptualism and the Problem of Perceptual Self-Knowledge,” European Journal of Philosophy 19,
778 R. Hanna
The classic or standard line of Kant-interpretation in twentieth-century

Anglo-American philosophy simply took it as obvious that Kant is a
Conceptualist and also an intellectualist. So the Non-Conceptualist inter-
pretation of Kant is importantly revolutionary and unorthodox, and even if it
were not correct (although I do think it is correct), nevertheless it has forced
Conceptualist, intellectualist Kantians to rethink, re-argue, and rework their
previously unchallenged view.20
But most importantly for the purposes of this essay, both classical Rationalist
metaphysics and also contemporary Analytic metaphysics alike presuppose
Conceptualism and intellectualism. For example, the very idea of ontological
arguments in modal metaphysics requires the truth of Conceptualism.
Correspondingly, real metaphysics in the Kantian sense requires the truth of
(essentialist content) Non-Conceptualism and non-intellectualism.
Now I can reformulate the main aim of this section more precisely, in four
sub-claims. What I want to claim is
(i) that Kant is an (essentialist content) Non-Conceptualist,

(ii) that there is a specifically non-intellectualist version of Kant’s transcen-
dental idealism that depends inherently on the nature of human
sensibility,
(iii) that Kant’s (essentialist content) Non-Conceptualism is foundational for
any philosophically defensible version of his transcendental idealism, and
(iv) that this line of thinking in Kant can be traced directly back to his proto-
critical “Directions in Space” essay.
In other words, what I want to claim is that Kant’s non-intellectualism about

the human mind goes all the way down into real metaphysics; that it is
defensibly arguable that the apparent world fundamentally conforms to
no. 2 (June 2011): 184–223; Hanna, “Beyond the Myth of the Myth of the Given”; Hemmo Laiho,
Perception in Kant’s Model of Experience (PhD diss., University of Turku, 2012); and Clinton Tolley,
“The Non-Conceptuality of the Content of Intuitions: A New Approach,” Kantian Review 18, no. 1
(March 2013): 107–36. Weaker versions of Kantian Non-Conceptualism are defended by, for example,
Lucy Allais, “Kant, Non-Conceptual Content and the Representation of Space,” Journal of the History of
Philosophy 47, no. 3 (July 2009): 383–413; Colin McLear, “Two Kinds of Unity in the Critique of Pure
Reason,” Journal of the History of Philosophy 53, no. 1 (Jan. 2015): 79–110; Christian Onof and Dennis
Schulting, “Space as Form of Intuition and as Formal Intuition: On the Note to B160 in Kant’s Critique
of Pure Reason,” Philosophical Review 124, no. 1 (Jan. 2015): 1–58; and Peter Rohs, “Bezieht sich nach
Kant die Anschauung unmittelbar auf Gegenstände?” in Kant und die Berliner Aufklärung: Akten des IX.
Internationalen Kant-Kongresses, ed. Volker Gerhardt, Rolf-Peter Horstmann, and Ralph Schumacher,
5 vols. (Berlin: de Gruyter, 2001), 2:214–28.
20
See, e.g., Colin McLear, “The Kantian (Non-)Conceptualism Debate,” Philosophy Compass 9, no. 11
(Nov. 2014): 769–90.
human sensibility even if it does not fundamentally conform to the human

understanding; and that the basic source of all this is Kant’s (initially proto-
critical but later also critical) theory of space and how we represent it.
Kant’s “Directions in Space” essay contains an argument against the
relational or Leibnizian view of space and in favor of the absolute or
Newtonian view of space (DS 2:377–83), but this merely scratches the
surface of Kant’s argument. The relational theory of space says that the
nature of space is necessarily determined by extrinsic relations between
objects in space. By contrast, the absolute theory of space, as Kant under-
stands it, says that the nature of space is necessarily determined by a single
universal framework – a global space-frame – in which physical objects are
inherently embedded or located as filling up and realizing proper parts of the
global space-frame, whose structure necessarily includes certain special
intrinsic relational topological properties that allow for fundamental asym-
metries, in addition to the familiar Euclidean relational topological properties
and relations, which are symmetrical.
According to Leibniz, who was a relationist about space, the objects stand-
ing in extrinsic relations are monads. So space is actually a “well-founded
phenomenon” for Leibniz, and strongly supervenient on the intrinsic non-
relational properties of noumenal monads. Nevertheless, other relationists
about space, including Kant himself in the Physical Monadology (PM
1:475–87), hold that these objects are actually material point-sources of causal
forces in real physical space. So the version of relationism that Kant was
working with in “Directions in Space” is not an orthodox Leibnizian theory.
According to Newton, who was an absolutist about space, the single uni-
versal framework in which physical objects are embedded is itself a noumenal
entity. But as far as I know, Newton was unaware of the idea that the structure
of absolute space contains special asymmetry-allowing intrinsic relational
topological properties. Hence the version of absolutism that Kant was working
with in “Directions in Space” is also not an orthodox Newtonian theory.
According to Kant in “Directions in Space,” space does indeed constitute a
global frame for embedding or locating physical objects, like Newtonian
space, but also (and much more importantly) it is an egocentrically-centered,
orientable space with inherent structural asymmetries such as mirror-reflected
incongruence or “handedness” in qualitatively identical objects (enantiomor-
phy), which Kant also calls “incongruent counterparts” (DS 2:377–83).
“Orientable spaces” are spaces with intrinsic directions, and “egocentric
centering” means that the specific characteristics of an orientable space are
fixed indexically and locally by conscious embodied perceivers who are
themselves actually embedded or located within the total global space-frame.
780 R. Hanna
In “Directions in Space,” Kant discovered that structural asymmetries

such as handedness can be detected and differentiated only by the essen-
tially non-intellectual, non-conceptual outer sensibility of living, embod-
ied, conscious, cognizing subjects like us, who are actually embedded or
located in such a global space, and therefore that there is a necessary
isomorphism between the representational form of the outer sensibility of
such subjects, the abstract structure of that global space, and the material
structure of perceivable objects also embedded or located in that global
space:
Because of its three dimensions, physical space can be thought of as having

three planes, which all intersect each other at right angles. Concerning the things
which exist outside ourselves: it is only in so far as they stand in relation to ourselves
that we have any cognition of them by means of the senses at all. It is, therefore, not
surprising that the ultimate ground, on the basis of which we form our [representa-
tion] of directions in space, derives from the relation of these intersecting planes to
our bodies. The plane upon which the length of our body stands vertically is
called, with respect to ourselves, horizontal. This horizontal plane gives rise to
the difference between the directions which we designate by the terms above
and below. On this plane it is possible for two other planes to stand vertically
and also to intersect each other at right angles, so that the length of the human
body is thought of as lying along the axis of the intersection. One of these two
vertical planes divides the body into two externally similar halves, and furnishes
the ground of the difference between the right and the left side. The other
vertical plane, which also stands perpendicularly on the horizontal plane, makes
possible the [representation] of the side in front and the side behind. . . . Since
the distinct feeling of the right and the left side is of such great necessity for
judging directions, nature has established an immediate connection between this
feeling and the mechanical organisation of the human body. (DS 2:378–80,
emphasis added)
In short, the apparent or phenomenal world must conform to the form of our
embodied outer sensibility; that is, the apparent or phenomenal world must
conform to the form of human outer intuition.
Now for Kant the form of human outer sensibility or intuition is essentially
non-conceptual for three reasons.
First, Kant says explicitly in the Critique of Pure Reason that intuitions of
outer sense or inner sense, which pick out appearances – the undetermined
objects of empirical intuitions (A20/B34) – are possible for us
independently of the functions of our understanding, that is, indepen-

dently of our concepts:
Since an object can appear to us only by means of . . . pure forms of sensibility, i.e.,
be an object of empirical intuition, space and time are thus pure intuitions that
contain a priori the conditions of the possibility of objects as appearances, and
the synthesis in them has objective validity. . . . Objects can indeed appear to us
without necessarily having to be related to functions of the understanding. . . .
Appearances can certainly be given in intuition without functions of the under-
standing. . . . For appearances could after all be so constituted that the under-
standing would not find them in accord with the conditions of its unity,
and . . . in the succession of appearances nothing would offer itself that would
furnish a rule of synthesis and thus correspond to the concept of cause and
effect, so that this concept would therefore be entirely empty, nugatory, and
without significance. Appearances would nonetheless offer objects to our intuition,
for intuition by no means requires the functions of thinking. (A89–91/B121–23,
emphasis added)
That representation that can be given prior to all thinking is called intuition.
(B132)
The manifold for intuition must already be given prior to the synthesis of
understanding and independently from it. (B145, emphasis added)
Second, Kant explicitly claims in some pre-critical writings and critical

writings alike that at least nonhuman animals (e.g., oxen and dogs) (FS
2:59–60; JL 9:64–65) and some nonrational human animals (e.g., ordinary
human infants) (An 7:127–28) are capable of sense perception and thus
capable of inner and outer sensory intuition, but do not possess conceptual
capacities.21
Third, and most importantly for our purposes, our pure or non-empirical
representation of space picks out egocentrically-centered, orientable, asym-
metric structural topological properties of space that cannot be represented by
the understanding and concepts. This is shown by the “incongruent counter-
parts” argument, which, in a nutshell, says:
(1) Incongruent counterparts, like our right and left hands, by hypothesis,
are such that they possess all their conceptually-representable qualities in
21
See, e.g., Colin McLear, “Kant on Animal Consciousness,” Philosophers’ Imprint 11, no. 15
(Nov. 2011): 1–16.
782 R. Hanna
common, yet they still are essentially different because they are
incongruent.
(2) This incongruence and the essential difference between our right and left
hands is immediately and veridically represented by human cognizers,
but only by means of our empirical intuition of real objects in physical
space and also our pure sensory intuition of the structure of space, as
necessarily conforming to the form of our outer sensibility or intuition.
(3) Therefore, our pure or non-empirical (hence a priori) representation of
space is necessarily underdetermined by concepts.22,23
When the conclusion of the incongruent counterparts argument is conjoined

with the first two reasons, then it follows that the form of our outer
sensibility or intuition is essentially non-conceptual and also a priori.
Therefore, in “Directions in Space,” at least implicitly, Kant is saying that
the basic structure of the apparent or phenomenal world necessarily con-
forms to the essentially non-conceptual a priori form of human embodied
outer sensibility or intuition. This line of argument is made even more
explicit in, and furthermore is strongly supported by, Kant’s doctrine of
the nature of space in the Inaugural Dissertation, On the Form and Principles
of the Sensible and the Intelligible World (1770) (ID 2:403); by his argument
for the transcendental ideality of space in the Prolegomena to Any Future
Metaphysics (1783) (Pro 4:286); and by his later discussion of geographical
spatial orientation in “What Does It Mean to Orient Oneself in Thinking?”
(1786) (OT 8:135).
This way of reading “Directions in Space,” however, is confusingly con-
cealed by the way that Kant formulates his main thesis in the essay:
My purpose in this treatise is to see whether there is not to be found in the

intuitive judgements about extension, such as are to be found in geometry,
clear proof that: Absolute space, independently of the existence of all matter and as
itself the ultimate foundation of the possibility of the compound character of matter,
has a reality of its own. (DS 2:378)
22
For more fully spelled out versions of this argument, see Hanna, “Kantian Non-Conceptualism”;
Hanna, “Beyond the Myth of the Myth of the Given”; and Hanna, Cognition, Content, and the A Priori,
ch. 3.
23
For more detailed discussions of these topics, see Robert Hanna, Kant and the Foundations of Analytic
Philosophy (Oxford: Oxford University Press, 2001), chs. 3–5; Hanna, “Kant’s Theory of Judgment,”
sections 2.2.2–2.2.3; and Robert Hanna, “Axioms,” “Synthesis,” and “Synthetic A Priori,” in Cambridge
Kant Lexicon.
In other words, the notion of absolute space, as Kant is using it in “Directions

in Space,” is ambiguous as between
(i) a global space-frame with orientability, egocentric centering, and structural

asymmetries, that fundamentally conforms to the essentially non-concep-
tual representational structure of human outer sensibility or intuition, and
(ii) noumenal space, as in Newton.
However, by the time of the Inaugural Dissertation, and then later in the
Transcendental Aesthetic and throughout the critical period, it is perfectly
clear that for Kant the global space-frame must be transcendentally ideal, and
cannot be noumenal.
So for all these reasons, I want to claim that the central argument in
“Directions in Space” is almost certainly the major proto-critical philosophi-
cal breakthrough that Kant famously reports when he says in one of the
Reflexionen that “the year’69 gave me great light” (Ak 18:69 [R5037]). More
precisely, what Kant had discovered between 1768 and 1772 is what I call
transcendental idealism for sensibility. In 1772, Kant told Marcus Herz that if
the human mind conformed to the world, whether phenomenal or noume-
nal, then a priori knowledge would be impossible (C 10:130–31); but by
1770 Kant already also held that a priori knowledge of the phenomenal
world is obviously actual and therefore really possible in mathematics, hence
the phenomenal world must conform to the non-empirical sensible structure
of the human mind, and more specifically must conform to our a priori
representations of space and time, since that is what makes mathematics
really possible (ID 2:398–406). So transcendental idealism for sensibility says
that the apparent or phenomenal world fundamentally conforms to the
essentially non-conceptual a priori forms of human sensibility, our represen-
tations of space and time.
Kant worked out explicit proofs for transcendental idealism for sensibility
in the Inaugural Dissertation and again in the Transcendental Aesthetic in the
first Critique. The simplest version of the proof, provided in the
Transcendental Aesthetic, goes like this:
(1) Space and time are either (i) things in themselves, (ii) properties of/
relations between things in themselves, or (iii) transcendentally ideal.
(2) If space and time were either things in themselves or properties of/
relations between things in themselves, then a priori mathematical knowl-
edge would be impossible.
784 R. Hanna
(3) But mathematical knowledge is actual, via our pure intuitions of space
and time, and therefore really possible.
(4) Therefore, space and time are transcendentally ideal. (A23/B37–38,
A38–41/B55–58)
There is, of course, much more that can and should be said about this
highly controversial argument. What is most crucial for our purposes
here, however, is that this version of transcendental idealism relies only
on essentially non-conceptual content and the nature of human sensi-
bility, and neither relies on concepts and the nature of human under-
standing, nor does it entail that the phenomenal world necessarily
conforms to our concepts and the nature of human understanding.
Kant’s modal dualism, apriorism, and theory

of the synthetic a priori
Kant’s modal dualism, apriorism, and theory of the synthetic a priori24 can
be boiled down to three basic parts, each of which contains three sub-theses.
First, (i) a belief, judgment, proposition, or statement is analytic if and
only if its meaningfulness and truth or falsity are necessarily determined by
intrinsic conceptual connections, including the intensional “containment” of
a predicate-concept in a subject-concept (e.g., “Bachelors are unmarried”),
conceptual identity (e.g., “Bachelors are bachelors”), and the intrinsic con-
ceptual connections characteristic of pure general logic (e.g., the Principle of
Minimal Non-Contradiction, namely, “Not every proposition and its nega-
tion are both true,” i.e., “~(P)(P&~P)”), alone, no matter what the other
semantic constituents of those beliefs, judgments, and so on, might be.
(Note: I have given the weakest possible version of the universal Principle
of Non-Contradiction here, in order to accommodate Kant’s explicit thesis
in the Antinomy of Pure Reason that if certain deeply confused or outright
false metaphysical assumptions are made, then antinomies, paradoxes, or
“true contradictions” logically follow. In this sense, Kant is the first serious
dialetheic logician in the history of modern philosophy, and thus the first
serious exponent of non-classical logic in post-seventeenth-century philoso-
phy. And obviously Hegel was also deeply influenced by this thesis.)
24
See Hanna, Kant, Science, and Human Nature, ch. 3; and Hanna, Cognition, Content, and the A Priori,
ch. 4
(ii) The universal criterion of analyticity is that the negation of any

analytic proposition entails a conceptual or logical contradiction (A150–
53/B189–93).
And finally (iii) an analytic truth is a necessary truth that is true in every
conceptually and/or logically possible world. Simply put, analytic truth and
knowledge are conceptual truth and knowledge.
Second, (i) a belief, judgment, proposition, or statement is synthetic if and
only if
its meaningfulness and truth or falsity are necessarily determined by the

empirical or non-empirical sensible intuitions that are semantic constituents
of the relevant belief, judgment, proposition, or statement, not by pure general
logic alone, and not by the concepts that must also belong to it.
(ii) The universal criterion of syntheticity is its negation is conceptually

and logically consistent, that is, its negation does not entail a conceptual or
logical contradiction.
And finally (iii) a synthetic truth is true in all and only the possible worlds
that meet the special spatiotemporal and mathematical conditions of human
sensible experience, a.k.a. the “experienceable worlds,” and a truth-value gap
otherwise. Simply put, synthetic truth and knowledge are essentially non-
conceptual truth and knowledge.
Third, (i) apriority entails both non-empiricality and necessity (and con-
versely), and aposteriority entails both empiricality and contingency (and
conversely).
Kant’s “necessity if and only if apriority” thesis is controversial and has
been generally rejected by Analytic metaphysicians, who, following Kripke,
claim that there are necessary a posteriori truths and also contingent a priori
truths. Give the quasi-canonical status of this Kripkean dogma, refuting it
takes some serious cognitive-semantic work, which I have undertaken else-
where.25 But the basic idea of the anti-Kripkean critique is that the necessary
a posteriori and the contingent a priori are both myths (a) because they
presuppose a highly questionable conception of the a priori-a posteriori
distinction, (b) because the inference from empiricality to aposteriority is
fallacious (e.g., does anyone seriously think that “3 beer bottles made by
Budweiser + 4 beer bottles made by Budweiser = 7 beer bottles made by
Budweiser” is a posteriori just because it has significant empirical content?),
25
See, e.g., Frank Jackson, From Metaphysics to Ethics: A Defence of Conceptual Analysis (Oxford: Oxford
786 R. Hanna
and (c) because Kripke’s argument is systematically ambiguous as between the

epistemic and semantic properties of sentences and the epistemic and semantic
properties of propositions (e.g., even if it is true that the same sentence-type “Water
is H2O” has tokens such that one sentence-token under one semantic inter-
pretation is necessary and a priori, and another token under another interpreta-
tion is contingent and a posteriori, it does not follow that there is a single
proposition “Water is H2O” that is both necessary and a posteriori).
(ii) Whereas all analytic beliefs, judgments, etc., must be a priori, there
are nevertheless not only synthetic a posteriori beliefs, judgments, etc., like
“The cat is on the mat,” but also, and most importantly – since this
uniquely semantically characterizes the necessary truths of mathematics
(e.g., “3+4=7”), metaphysics (e.g., “Every event has a cause”), and philo-
sophy (e.g., “Persons are conscious, intentional, free agents”) more gener-
ally – there also really can be, and really are, synthetic a priori truths.
And finally (iii) whereas a synthetic a posteriori truth is a contingent
truth – hence its negation is conceptually and logically consistent – that is
true in some experienceable worlds and false in some experienceable worlds,
and a truth-value gap otherwise, by sharp contrast, a synthetic a priori truth is
a necessary truth that is true in all and only the experienceable worlds, and a
truth-value-gap otherwise.
For example, “3+4=7” is a synthetic a priori truth. This is because its
meaningfulness and truth are necessarily determined by our non-empirical
sensory intuition of the successive, serial, recursive (i.e., generated by
repeated self-applications of the same operation) structure of the moments
of phenomenal time, which provides a unique model of the natural numbers,
not by pure general logic alone, and not by the concepts that must also
belong to “3+4=7.” Correspondingly, “It is not the case that 3+4=7” is not a
conceptual or logical contradiction, because there are conceptually and
logically possible worlds in “3+4=7” is not true, namely, either worlds with-
out phenomenal time per se or worlds without anything isomorphic to the
structure of phenomenal time in them, hence worlds without any natural
numbers in them.
Nevertheless “3+4=7” is never false in any logically possible world, since it
is true in all the experienceable worlds and also a truth-value gap in all the
rest of the conceptually and logically possible worlds.
In other words, a synthetic a priori truth is a “necessary truth with a
human face,” that is, an anthropocentrically necessary truth, hence it is a
necessary truth even though its negation is logically and conceptually consistent.
It does not tell us what an omniscient God or a disembodied thinking spirit
could know by reason alone; instead it tells what only a rational, but also
finite, embodied, sensible creature like us, could ever know.
Indeed, it is precisely this irreducibly anthropocentric semantic and epis-
temic character of synthetic apriority that has seemed, and still seems, most
puzzling and even downright paradoxical to those who cut their philosophi-
cal teeth on Humean Empiricism or Logical Empiricism, including most
contemporary early twenty-first-century professional academic philosophers.
This is because Empiricism presupposes, without argument, that there is one
and only one kind of necessary truth, namely, analytically necessary truths,
that is, conceptual truths or logical truths. Therefore, Empiricism is always
explicitly or implicitly committed, without argument, to modal monism. Not
only that but also both classical Rationalism and contemporary Analytic
metaphysics alike, as well as Conceptual analysis,26 are all committed to
modal monism.
On the contrary, and as against Empiricism, Conceptual analysis, classical
Rationalism, and contemporary Analytic metaphysics, Kant is committed to
modal dualism, and therefore the Kantian doctrine of necessity is that there
are irreducibly two essentially different kinds of necessary truths, analytic
(conceptually necessary) and synthetic a priori (non-conceptually necessary).
Correspondingly, the general idea of a necessary truth in a Kantian, modal
dualist framework is this:
a belief, judgment, proposition, or statement is necessarily true if and only if it

is true in every member of a well-defined, complete class of logically possible
worlds, and never false in any logically possible world.
This formulation encompasses both logically or analytically necessary

truths (true in all logically possible worlds) and synthetically necessary
truths (true in all experienceable worlds and a truth-value gap otherwise)
alike.
It is a sad but true sociological fact of contemporary professional academic
philosophical life that nothing bores distracted introductory students,
stressed-out graduate students, and jaded professional philosophers too,
more quickly and utterly, than discussing the analytic-synthetic and a
priori-a posteriori distinctions. So I also need to answer this question expli-
citly: Do the analytic-synthetic and a priori-a posteriori distinctions actually
matter, and if so, why?
26
See Thomas Nagel, The View from Nowhere (New York: Oxford University Press, 1986).
788 R. Hanna
Yes, they actually matter! In fact, they massively matter.

And they actually and massively matter, precisely because (i) this pair of
distinctions tells us what kinds of truth and knowledge are really possible for
creatures like us, so that having a good theory of them tells us how human
cognitive rationality is really possible, and (ii) this pair of distinctions tells us
what kinds of truth and knowledge are really possible in philosophy, so that
having a good theory of them tells us how philosophy is really possible. In
short, these distinctions actually and massively matter because without them,
it would be the end of the rational and philosophical world as we know it – both
human rationality and philosophy would be really impossible.
Conclusion
Very simply put, the Copernican Devolution in contemporary Analytic
metaphysics, just like classical seventeenth- and eighteenth-century
Rationalist metaphysics prior to Kant – especially Descartes, Spinoza,
Leibniz, and Christian Wolff, together with seventeenth-century scientific
naturalism, especially Bacon – says that metaphysics “writes the book of the
world,” and therefore it knows the world and ourselves from the absolute
standpoint. The absolute standpoint is the same as Thomas Nagel’s “view
from nowhere,”27 whether this is, on the one hand, God’s standpoint, or, on
the other hand, the modern equivalent of God’s standpoint, the standpoint of
all-knowing mechanistic natural science, according to which “science is the
measure of all things.”28 Hence the Copernican Devolution is at once a
metaphysics from nowhere and a metaphysics for “moist robots.”
By sharp contrast, Kant’s real metaphysics, on its negative side, says (i) that
noumenal modal metaphysics leads inevitably and directly to “obscurity and
contradictions” (the critique of pure reason) (Aviii), and (ii) that noumenal
modal metaphysics is phenomenologically inadequate and explanatorily
27
See Francis Bacon, Novum Organum, https://archive.org/stream/baconsnovumorgan00bacouoft#
page/n3/mode/2up; Hanna, Kant, Science, and Human Nature; Robert Hanna, “Kant, Natural Piety,
and the Limits of Science,” unpublished ms. (Fall 2015), https://www.academia.edu/17038961/Kant_
Natural_Piety_and_the_Limits_of_Science_Fall_2015_version_comments_welcomed_; and Robert
Hanna, “Kant, Scientific Pietism, and Scientific Naturalism,” unpublished ms. (Fall 2015), https://
www.academia.edu/18030039/Kant_Scientific_Pietism_and_Scientific_Naturalism_Fall_2015_ver
sion_comments_welcomed_.
28
See, e.g., Jennifer Schluesser, “Philosophy That Stirs the Waters,” New York Times (29 April 2013),
http://www.nytimes.com/2013/04/30/books/daniel-dennett-author-of-intuition-pumps-and-other-
tools-for-thinking.html?emc=eta1&_r=0.
empty. But on its positive side, transcendental idealism, it says that real
metaphysics knows the world and ourselves necessarily and a priori from the
rational human experiential standpoint, and therefore with phenomenologi-
cal adequacy, but also without noumenal dogmatism, fallacy, or paradox.
Hence it is a rationally and phenomenologically adequate metaphysics with a
human face.
The philosophical battle cry of the nineteenth-century neo-Kantians,
against a profoundly wrong-headed Hegelianism, was “Back to Kant!” But
as I have argued, contemporary Analytic metaphysicians really and truly need
to learn the older Kantian lessons in order to make a metaphysics of the
future really possible. Hence for the rest of the twenty-first century and
beyond, the real and true philosophical battle cry, against a tragically wrong-
headed Analytic metaphysics, and toward the real metaphysics of the future,
is Forward to Kant! 29
29
I am very grateful to the members of the seminar on logic, science, mind, and language at the Federal
University of Minas Gerais, Brazil, especially Patricia Kauark-Leite, for their extremely helpful com-
ments on an earlier version of this essay.
34
Contemporary Kantian Moral Philosophy
Michael Rohlf
Kant’s moral philosophy is the part of his overall system that is most widely
defended in contemporary philosophical debates, and his influence on
contemporary moral philosophy is vast. The type of moral theory Kant
pioneered, sometimes characterized (but not uncontroversially) as deontological
or non-consequentialist, is standardly classified today as one of the major
systematic approaches in normative ethics – usually alongside consequentialism
(of which utilitarianism is one variant) and sometimes also alongside virtue
ethics and intuitionism. It is not possible to engage very deeply with contem-
porary debates between defenders of these approaches, or about many other
topics such as moral motivation, the nature of reasons and rationality, or the
metaphysical status of moral properties and values, without encountering posi-
tions designated as Kantian.
There is, however, a subtle and sometimes bewildering distinction between
scholarship on Kant’s moral philosophy, which aims primarily to interpret and
evaluate historical texts by Kant, and positions designated as Kantian in
debates among contemporary moral philosophers. It is not surprising that
philosophers who find some or all of the historical Kant’s moral thought
defensible today would devote some of their attention to engaging with
Kant’s own texts and some of their attention to arguing in contemporary
terms for Kantian views, perhaps with (more or less explicit) modifications to
M. Rohlf (*)
School of Philosophy, The Catholic University of America, Washington, DC, USA
e-mail: rohlf@cua.edu

792 M. Rohlf
the letter of Kant’s actual views but normally at least with a view to preserving
and sometimes articulating better than Kant himself did what they regard as
the spirit of Kant’s moral thought. Yet not all Kantians agree, of course, about
what the spirit of Kant’s moral thought is, and the term “Kantian” is also
deployed by philosophers who are less sympathetic to Kant in order to
designate positions that they reject but that many Kantians regard as carica-
tures. All of this makes it difficult to get a handle on what is or should be
meant by the term “Kantian” when it is used in contemporary moral philo-
sophy. One needs to keep track not only of the different ways one might
interpret Kant’s own texts, and not only of the various ways one might apply
those texts so interpreted to contemporary debates, both of which are difficult
enough to follow on their own; but also one must keep tabs on how and (if
reasons are given) why different scholars depart from Kant’s actual views in
different ways while still portraying their positions as Kantian.
One way to orient oneself in contemporary Kantian moral philosophy is
to remember that it was basically invented by John Rawls just a few decades
ago. In his 1980 Dewey Lectures, titled “Kantian Constructivism in Moral
Theory,” Rawls portrayed himself – justifiably – as the first to develop what he
calls Kantian constructivism as an important moral theory on par with (and,
according to Rawls, in fact superior to) intuitionism and utilitarianism.1
Through this and other publications as well as his Kant lectures at Harvard,
different versions of which Rawls made available in written form to his
students from the 1970s until his retirement in 1991, Rawls was the dominant
influence on the next generation of philosophers who carried on and expanded
the enterprise of Kantian moral philosophy that he started.2 Rawls’s students –
such as Christine Korsgaard, Barbara Herman, Onora O’Neill, Andrews
Reath, and Thomas Hill, Jr. – are among the most prominent figures in the
field today; they and their own students have developed Kantian moral
philosophy in many important new directions. But much or even most recent
work in the field continues to fit broadly within the framework developed by
Rawls himself and named by him Kantian constructivism. To very many
philosophers today, in fact, contemporary Kantian moral philosophy just is
more or less what Rawls meant by Kantian constructivism.
1
John Rawls, “Kantian Constructivism in Moral Theory,” Journal of Philosophy 77, no. 9 (Sept. 1980):
515–72.
2
The latest version of Rawls’s Kant lectures is published in John Rawls, Lectures on the History of Moral
Philosophy, ed. Barbara Herman (Cambridge: Harvard University Press, 2000). On the history of
Rawls’s lectures, see the editor’s forward (xi–xix).
Of course Rawls’s approach to Kantian moral philosophy has not gone

unchallenged. It has been challenged in relatively minor ways by some of
Rawls’s own students, but its more significant challenges have largely come
either from moral philosophers unpersuaded by the philosophical merits of
Rawls’s position or from Kant scholars unpersuaded that Rawls’s position
corresponds with that of the historical Kant. One of the most prominent
figures to have combined both lines of criticism is Allen Wood. Most
recently and forcefully in his 2008 book Kantian Ethics, Wood has developed
a sustained attack on the Kantian constructivist position that stems from
Rawls and has proposed in its place an alternative conception of Kantian
moral philosophy that rivals Rawls’s own in its depth and comprehensive-
ness.3 Wood also sees Rawls as the dominant influence in contemporary
Kantian moral philosophy and gives credit to the Rawlsian tradition for
decisively shaping his own understanding of Kant’s ethical thought, as well as
to Rawls himself for emphasizing that the equal worth of all persons is at
the heart of Kant’s moral outlook (KE x, 3).4 Since Wood claims that
“radical egalitarianism, grounded in the conception of every human being
as a rationally self-governing agent, is the most fundamental idea in Kantian
ethics,” his agreement with Rawls on the spirit of Kant’s moral philosophy
appears to go very deep indeed (KE 94). But this appearance is in tension
with Wood’s statement that he “simply cannot accept” the following three
elements of the Rawlsian constructivist interpretation of Kant:
I have especial trouble with the elements in it that are most familiar to and taken
for granted by moral philosophers – I mean especially [2] the overemphasis on,
and misconstrual of, the Formula of Universal Law. But the constructivist
reading also seems mistaken in [3] the metaethical conclusions it wants to
draw from Kant’s conception of autonomy, and perhaps most of all [1] its
basic conception of the aims and methods of ethical theory – all of which
seem to me deeply at odds both with what Kant himself actually thought
about these matters and also with the best way Kant’s thinking about ethics
can be appropriated by us today. (KE x–xi)
I have enumerated these elements in the order in which Wood actually

discusses them and in which I will summarize his criticisms below.
3
Allen W. Wood, Kantian Ethics (New York: Cambridge University Press, 2008).
4
In this chapter, I use the following abbreviations and cite the works parenthetically: KC for Rawls’s
“Kantian Constructivism in Moral Theory” (note 1), MP for Rawls’s Lectures on the History of Moral
Philosophy (note 2), and KE for Wood’s Kantian Ethics (note 3).
794 M. Rohlf
My goal in this chapter is to review and assess Wood’s criticisms of Rawls’s

Kantian constructivism on these three topics as a way of presenting and
commenting on some core issues in contemporary Kantian moral philoso-
phy. Although Wood directs his criticisms against the entire constructivist
tradition that stems from Rawls, I will consider them primarily insofar as
they target Rawls himself in order to sort out clearly where and why these
two giants in Kantian moral philosophy actually disagree. I will argue that
the most promising approach to Kantian moral philosophy incorporates
elements of both Rawls’s and Wood’s accounts, specifically Rawls’s con-
structivist view of moral judgment and Wood’s metaethical realist interpre-
tation, but in the context of a conception of ethical theory that lies
somewhere between Rawls’s and Wood’s.
Wood’s criticisms
I begin by summarizing Wood’s main criticisms of Rawls’s Kantian construc-
tivism as he interprets it. First, Wood accuses Rawls and his followers of
subscribing to what Wood calls the dominant “intuitional” and “scientific”
conception of ethical theory, which Wood traces back to Henry Sidgwick,
without recognizing either its philosophical limitations or its divergence from
Kant’s own (according to Wood, superior) conception of ethical theory
(KE 4–65). As Wood describes it,
The task of ethical theory, according to this model, is to reconcile initially

conflicting judgments and principles, either by qualifying or modifying them,
or explaining away apparent conflicts, so as to produce the most coherent
overall explanation of our intuitions. The aim is to give the most coherent and
intuitively compelling account of all our moral intuitions, at all levels of
generality, an account that both reconciles our intuitive judgments and also
gives us the most satisfying explanation of why we consider them true. (KE 44)
Wood cites Rawls’s appeals to due reflection and reflective equilibrium

(which I discuss below) as evidence that his Kantian constructivism sub-
scribes to this model of ethical theory (KE 46).
Wood’s main philosophical objection to this model is that it is superficial:
it “aims not at truth but only to systematize beliefs, which are left without
any firm foundation” (KE 51). Since more than one system of moral beliefs
could turn out to cohere equally well, this model threatens us with “an
indecisive result or with a plural moral truth or a form of moral relativism,”
which Wood thinks Rawls at one point welcomed (KE 51). But even setting
this worry aside, “it is not self-evident that the right answer to an ethical
question, or even the best grounded answer, is always the one that would
emerge from achieving maximal coherence among our preexisting ethical
judgments” (KE 49). On the one hand, a maximally coherent set of moral
intuitions could yield a wrong answer to some ethical question if those
intuitions are manipulated by the wording of the question or by artificial
examples (such as trolley problems) (KE 49–50, 284). On the other hand,
some difficult moral questions do not have a single right answer, but the
dominant model neither provides moral judgment with the right sort of
guidance in such cases nor does it enable us to distinguish them from cases in
which there is a single right answer (KE 48–49, 284–85). This latter point is
related to what Wood calls the scientific ambitions of the dominant model,
by which he means that it aims “to settle all moral questions and make all
moral decisions, as far as possible, by a rigorous derivation from precisely
stated principles” (KE 47).
Instead of this dominant intuitional and scientific model, Wood claims
that Kant subscribes to what Wood calls the “foundational” and “philoso-
phical” model of the aims and methods of ethical theory (KE 54–58). Kant’s
actual model is not intuitional, according to Wood, because it does not
attempt to justify its fundamental principle by appealing to intuitions, which
“play at most a secondary role in arguing for its basic value” (KE 55). On
Wood’s interpretation,
Kant’s definitive search for the supreme principle of morality occurs not in the
First Section of the Groundwork, based on appeals to “common rational moral
cognition,” but rather in the Second Section, where it is derived from a
philosophical account of volition that is wholly independent of any appeal to
moral common sense and rests not at all on intuitive judgments either about
particular acts or moral principles or on any “reflective equilibrium” between
such judgments. (KE 54)
Moreover, Kant’s model is also not scientific in the sense Wood attributes to
the dominant model of Sidgwick and Rawls, according to Wood, because
“Kant does not think it is the function of a fundamental principle of morality
directly to tell us what to do in particular cases” (KE 56) but rather “to
provide a basic framework, or value-oriented background, for justifying,
modifying, and applying the more particular rules or precepts of morality
that do tell us this,” which “they can do only to a limited extent” (KE 57).
According to Wood, Kant holds that applying the fundamental principle of
796 M. Rohlf
morality to particular cases involves acts of judgment that are “interpretive or

hermeneutical” because they must take into account “general empirical facts
about the human condition and human nature, perhaps also as modified by
cultural or historical conditions” (KE 60). Even with complete factual
information, however, the application of moral judgment may still involve
indeterminacy and thus insoluble moral dilemmas may remain (KE 61–65,
81–82). So moral judgment for Kant is far from the rigorous, deductive
process that Wood thinks Rawlsian constructivism makes it out to be.
Wood’s second criticism of Rawls and his followers for overemphasizing
and misconstruing the Formula of Universal Law (FUL) flows directly from
the previous criticism. According to Wood, for Rawlsians,
The universalizability tests, namely, are supposed to constitute a universal moral

criterion, a so-called “CI-Procedure,” that is applicable to any conceivable maxim
that might be proposed. It is supposed to be a method for grounding all moral
duties, or even for “constructing” the content of all morality. FUL and FLN [the
Formula of the Law of Nature] are often thought to constitute Kant’s chief (or
perhaps his only significant) contribution to moral philosophy. (KE 70–71)5
So interpreted, much of the work in Kantian ethics involves tweaking the

categorical imperative (CI) procedure in order “to make the test ‘come out
right’ even when applied to the most ravishingly ingenious maxims devised as
counterexamples” (KE 71). The goal of this enterprise, as Wood understands
the Rawlsian interpretation of Kant, is to arrive at a rigorous decision
procedure that yields a complete system of inflexible moral rules covering
every possible situation in human life, with moral intuitions again providing
the touchstone for success or failure (KE 63, 68, 70–71).
Against this reading of Kant, Wood argues that FUL and FLN either are
not for Kant or should not for Kantians be regarded as particularly useful or
informative formulations of the categorical imperative. Here Wood is a bit
difficult to pin down. He appears to say both that Kant did not intend these
formulations to be used for deriving specific duties, and that Kant did express
this intention but only in some “overconfident and unsupported remarks”
that should not be given much weight by Kantians interested in the most
plausible application of Kant’s ideas (KE 74; see 71–74). In any case, Wood’s
deeper point is that the later formulations of the categorical imperative, the
Formula of Humanity (FH) and especially the Formula of Autonomy (FA)
5
Wood attributes this view specifically to Rawls at KE 288.
(along with the Formula of the Realm of Ends [FRE]), are more adequate
expressions of the fundamental value underlying the moral law. He claims
that they, rather than FUL and FLN, both are actually used by Kant and
should be used by Kantians as the basis for arriving at moral rules for
particular cases (KE 74–79, 290). Again, however, the process of applying
the fundamental principle of morality – as expressed most adequately in these
later formulations of the categorical imperative – to particular cases is not
deductive but more loosely interpretive or hermeneutical, so it is impossible
in principle to arrive at anything like a complete system of all moral duties
(KE 77). Moreover, the duties one arrives at by applying these later formula-
tions to particular cases are not exceptionless moral rules that cover kinds of
action but rather cover only the specific maxim under consideration and
therefore may permit and forbid actions of the same type under different
conditions (KE 68, 71).
To bolster this interpretation of Kant’s Groundwork, Wood argues not
only that the various formulations of the categorical imperative are not in fact
equivalent, but also that the almost universally accepted view that Kant
explicitly says they are equivalent is based on a mistranslation. What Kant
actually says in the passage at issue (G 4:436), according to Wood, is not that
each of the formulations unites the other two in itself, but rather that one
formulation – namely FA – unites the other two in itself (KE 80–82).6
Likewise, when Kant says that “one does better always to proceed in moral
appraisal by the strict method and put at its basis the universal formula of the
categorical imperative: act in accordance with a maxim that can at the same
time make itself a universal law” (G 4:436–37), Wood claims that Kant is
actually referring to FA – not, as most other Kant scholars believe, to FUL
(KE 82–84, 289).7
Third and finally, Wood’s criticism of the Rawlsian constructivist approach
to Kantian moral philosophy culminates in his extended attack on the
metaethical conclusions drawn by Rawls and his followers from Kant’s con-
ception of autonomy. In contrast to this approach and in line with his previous
two criticisms of it, Wood’s own approach to Kantian moral philosophy
grounds the fundamental principle of morality in the objective value of
rational nature as an end in itself, as expressed in texts surrounding Kant’s
6
Mary Gregor translates the passage as follows: “The above three ways of representing the principle of
morality are at bottom only so many formulae of the very same law, and any one of them of itself unites
the other two in it” (G 4:436). Wood disputes the translation of this last clause. The German is: deren
die eine die anderen zwei von selbst in sich vereinigt.
7
I quote Gregor’s translation, which in this case is not significantly different from Wood’s.
798 M. Rohlf
articulation of FH in the Groundwork (KE 74–77, 85–95). So Wood finds

it “highly questionable to ascribe to Kant, as many Kantians today do, a
metaethical position that positively rejects the idea that there are objective values,
and denies that it is true (or a fact) that humanity is an end in itself having
absolute and objective worth” (KE 59). But according to Wood, Rawls and
many of his followers either explicitly endorse or are implicitly committed to an
anti-realist position about moral value. He realizes that some constructivists do
not regard their own position in this way but instead as an alternative to both
anti-realism and realism about value, but Wood denies that there is any stable
position in between those two alternatives and regards all Kantian constructivists
as in fact committed to anti-realism (KE 92, 95, 107–8, 283–84, 295–96). He
represents the stark choice between anti-realism and realism in terms of an
unavoidable dilemma regarding which component of “autonomy” one
emphasizes:
If we emphasize the “autos” and mean self-legislation [of the moral law] proceed-
ing from subjective voluntary acts of our will as the whole point of the doctrine,
we will either reject altogether the (natural) lawfulness of the moral law, its
objectivity, and universal validity, or else we will treat the law’s objectivity as
merely a way of considering (or a “legislative” stance on) what results from our
volitional acts (based perhaps on some “procedural conception of the norms of
practical reason” through which we “construct” the norms to which we subject
ourselves). The volitional act of legislation is then the literal content of the
doctrine of autonomy, and the objectivity of the law is merely a way of consider-
ing or regarding the content that we subjectively will. If, however, we emphasize
the “nomos” (the objective validity of the law, its groundedness in the objective
worth of rational nature as an end in itself), then we have to treat “self-legisla-
tion” as just a certain way of considering or regarding a law whose rational content
is truly objective and whose authority is therefore independent of any possible
volitional act we might perform. In either case, half of the doctrine of autonomy
turns out to be meant literally, while the other half is treated as an illuminating
way of considering or regarding the half we really mean. The only question left is:
Which half do we really mean? What wins out: autos or nomos? (KE 109–10)
According to Wood, the constructivist approach to Kantian moral philo-

sophy embraces the first horn of this dilemma, either explicitly or impli-
citly, while Wood himself embraces the second horn. This dilemma “is
nothing but a reappearance of the age-old quarrel between voluntarism and
rationalism that has been with us at least since Plato’s Euthyphro” (KE 110).
But “the choice is easy. Rationalism is the correct view and, incidentally,
Kant’s view. Voluntarism is a false view, though often the view of Kant’s
interpreters” (KE 110).
Wood cites texts from Kant’s published writings and lectures in support of
his reading that, “strictly speaking, our own will is neither the legislator nor the
author of the moral law. (To speak of it in either of these ways is at most
merely an appropriate way of considering or regarding the matter)” (KE 112;
see 111–14). Who then is the author of the moral law? Wood’s answer is: no
one. “The moral law has no author because it is a natural law” (KE 113). It is
a natural law whose content “lies in the nature (or essence) of the rational will
or practical reason” (KE 114). Here Wood does not mean specifically human
reason but rather reason as such; and he returns to Kant’s argument in the
second section of the Groundwork, now characterizing it as a “philosophical
account of the faculty of will or practical reason, and especially of the norms
(technical, pragmatic, and moral) that are constitutive of it (G 4:412–20)”
(KE 115). What is the point of considering or regarding the moral law as self-
legislated, if in fact it is constitutive of practical reason or will as such?
Wood’s main reply to this question invokes Kant’s distinction between
Wille (will) and Willkür (choice) and his view that the latter (choice) is free:
“Will” in the narrow sense is neither free nor unfree, because it does not choose
one thing or another. It simply presents to choice the reason or ground for
choosing this over that (whose highest norm is the moral law). As “choice,” the
rational being is subject to the law. “Choice” is free because it is able to, and this
also means motivated to, follow the law even when it is tempted not to and fails
to do so. What “choice” chooses are maxims, or subjective principles, that may or
may not conform to the laws given by “will.” When “choice” is tempted not to
obey the law but is inwardly constrained to do so, it is constrained by (its own)
will, and that is what makes it autonomous in obeying the moral law. (KE 121)
So we deserve to think of ourselves as fully autonomous only when we

obey the moral law (KE 119). But considering or regarding the moral law
as self-legislated also explains how categorical obligation is possible at all,
even insofar as we do not obey it, by representing the ground of the moral
law as lying in our own will. Since one’s own will or practical reason shares in
the nature of will or practical reason as such, this way of regarding the matter
is literally true: “Nothing beyond our will itself needs to be assumed as an
end or interest that is being satisfied by our obedience to the moral law”
(KE 117). The senses in which this is merely a way of considering or
regarding the literal truth, Wood seems to be saying, are both that no act
of legislation actually occurs since the moral law in fact has no legislator but
800 M. Rohlf
simply lies in the nature of will as such; and that it lies in the nature of any
rational will, not just one’s own.
The formula of universal law

Perhaps the best place to begin assessing Wood’s objections and alternatives
to Rawls’s Kantian constructivism is with his accusation that Rawls – again,
insofar as Wood’s target is specifically Rawls – overemphasizes and miscon-
strues the Formula of Universal Law. Here I think that Wood misrepresents
Rawls’s view, that he and Rawls actually agree more than Wood’s criticisms
suggest, but that where they actually disagree on this issue Rawls is in an
important way closer to capturing the spirit of Kant’s account of moral
deliberation and judgment.
Rawls does characterize the CI-procedure “as a way of generating the content
– the first principles along with the essential rights, duties, permissions, and the
rest – of a reasonable moral doctrine” (MP 163). But he expressly disavows the
sort of view that Wood attributes to him of the CI-procedure as a deductive
algorithm:
It is a serious misconception to think of the CI-procedure as an algorithm

intended to yield, more or less mechanically, a correct judgment. There is no
such algorithm, and Kant knows this. It is equally a misconception to think of
this procedure as a set of debating rules that can trap liars and cheats, scoundrels
and cynics, into exposing their hand. There are no such rules. (MP 166)
Moreover, far from overemphasizing FUL, Rawls writes that “it does not
greatly matter, I think, what specific formulation of the CI-procedure we
adopt, provided that it meets certain essential conditions,” one of which
(the “content condition”) is that it “must . . . have sufficient structure to
specify requirements on moral deliberation so that suitably many maxims” –
not, as Wood claims, “any conceivable maxim,” including “the most ravish-
ingly ingenious” ones (KE 70–71) – “are shown to be fit or unfit to be made
universal law” (MP 162–63; see also 254–55). An important feature of
Rawls’s interpretation of Kant that Wood ignores is that the moral law is
an idea of reason, and that “since an idea of reason can never be fully realized,
neither can the content of such an idea. It is always a matter of approxima-
tion, and always subject to error and correction” (MP 239n; see also 167).
Wood in fact sounds very much like Rawls himself when Rawls emphasizes
that “a moral conception can establish but a loose framework for deliberation
which must rely very considerably on our powers of reflection and judgment”
(KC 560). Nobody is pretending here that a formal procedure can settle all
moral questions on its own.
In his interpretation of Kant, however, Rawls rightly emphasizes FUL and
FLN when describing the procedure by which Kant thinks we both do and
should direct our moral judgment in order to decide what duties we have
(MP 162–76). I say he is right to emphasize these because Kant leaves no
room for doubt about his intentions when he first introduces FUL in section
one of the Groundwork, where he says twice that common human reason
always uses this principle as the norm in its practical appraisals (G 4:402,
403–4), and that “the shortest and yet infallible way” to decide what one’s
duty is in a specific situation is to apply FUL to one’s maxim in the way Kant
details there (G 4:403). If it is not totally clear whether Kant intends this
procedure to apply to every situation in the same way or just to certain
examples where duties we already recognize are at issue, as Wood claims
(KE 72), Kant removes any shred of doubt about this on the following page:
Here it would be easy to show how common human reason, with this compass in
hand, knows very well how to distinguish in every case that comes up what is good
and what is evil, what is in conformity with duty or contrary to duty . . . ; and that
there is, accordingly, no need of science and philosophy to know what one has to
do in order to be honest and good, and even wise and virtuous. (G 4:404)
Neither Rawls nor Wood thinks that FUL is in fact adequate for this purpose
(MP 163; KE 288). But they compensate for what they regard as the
inadequacy of FUL in different ways. As we have seen, Wood denies the
equivalence of the different formulations of the categorical imperative and
wants the later formulations, especially FA, to do the work of guiding moral
judgment instead of FUL and FLN. For my part, I do not find Wood’s
arguments against equivalence persuasive8; and Wood’s preference for using
the abstract philosophical ideas expressed in FA independently of the prin-
ciple Kant attributes to common human reason seems to me inconsistent
with Kant’s deep, Rousseauian respect for the common man, which he
articulates forcefully in the passage quoted above and in the surrounding
text of section one of the Groundwork, and which presumably is (or ought to
be) at least part of what Wood means by calling radical egalitarianism the
most fundamental idea in Kantian ethics (KE 94).
8
Gregor seems to me to have rendered the German clause quoted in note 6 correctly. On this point I
agree with Rawls, MP 183.
802 M. Rohlf
Rawls, on the other hand, compensates for what he regards as the inadequacy
of FUL not by turning instead to a different formulation of the categorical
imperative in the Groundwork, but rather by modifying how the procedure
(which he calls the procedure of construction) that guides moral judgment is
represented so that it better captures Kant’s own view of the requirements of
practical reason while still being accessible to (and authoritative for) any ordinary
person. Again, for my part, I would quibble with some of Rawls’s modifications,
and like Wood I think Rawls is mistaken to regard the duties to be derived from
or constructed by the CI-procedure as inflexible rules instead of situation-specific
requirements.9 Still, once we get past the mistaken idea that Rawls intended the
CI-procedure as a mechanical algorithm that is somehow supposed to replace
moral judgment (any more than Kant intended FUL or FLN that way), then we
are in a position to see just how well the general idea of Rawls’s Kantian
constructivism in fact captures the spirit of Kant’s approach to moral judgment.
On Kant’s own approach, we formulate the maxim of a proposed action,
imagine a world in which it is a universal law of nature to act on that maxim,
look for contradictions between our acting on that maxim and its being a
universal law, and can act permissibly on that maxim only if we find no such
contradictions (G 4:403, 421–24). This procedure does not replace moral
judgment but rather is the natural and appropriate framework within which
moral judgment is exercised, because it begins to draw our attention to morally
relevant features of our situation, and because reflection on why this apparently
artificial procedure is at all relevant to deciding what is morally permissible can
uncover deeper value commitments that are in fact built into the procedure
itself, explicit reference to which can sharpen our moral judgment by guiding
our further use of the procedure (especially in the formulation of maxims).10
As I understand it, the basic idea of Rawls’s own version of Kantian
constructivism, where that departs from his interpretation of Kant, is to
come up with a different artificial procedure as a framework for moral judg-
ment, due in part to inadequacies of the actual procedure employed by Kant
himself, but to bake into this new procedure the same deeper value commit-
ments that Kant actually endorsed (among others), so that in effect the new
procedure reflects Kant’s deepest ideas better than FUL does. (Importantly,
Rawls’s alternative procedure employing what he calls the model-conceptions
of a well-ordered society, a moral person, and the original position have a
9
For details, see Michael Rohlf, “Kant on Determining One’s Duty: A Middle Course between Rawls
and Herman,” Kant-Studien 100, no. 3 (Sept. 2009): 346–68.
10
I defend this interpretation in Michael Rohlf, “Contradiction and Consent in Kant’s Ethics,” Journal
of Value Inquiry 43, no. 4 (Dec. 2009): 507–20.
different and narrower scope of application than Kant’s CI-procedure, since it

applies to the basic structure of society rather than to individuals’ maxims and
aims to yield or construct principles of justice rather than the full content of a
moral doctrine.11) Thus Rawls says that the “main value for us” of the CI-
procedure “is to bring to life and to make intelligible Kant’s characteristic and
deeper ideas . . . by providing a means for their expression” (MP 163–64; see
also 237). For Rawls these deeper ideas can be summarized in Kant’s “con-
ception of free and equal persons as reasonable and rational, a conception that
is mirrored in the [CI-]procedure” (MP 240). Rawls does not locate Kant’s
articulation of this conception of persons in any particular formulation of the
categorical imperative, but rather in “the use Kant makes” of the CI-procedure
in general; and he claims that for Kant it is “elicited from our moral experience
and reflection” and “implicit in our practical reasoning” (MP 240–41).
Rawls might agree with Wood that the later formulations of the categorical
imperative are more adequate expressions of this conception than FUL or FLN
are. But Rawls would – rightly, I think – regard as a disadvantage precisely
what Wood regards as an advantage of FA, Wood’s preferred formulation:
namely, that it “does not pretend to offer us any test at all to discriminate
between maxims that have this rationally legislative property and maxims that
do not.” Instead, according to Wood, FA “tells us to think of ourselves as
members of an ideal community of rational beings, in which each of us should
strive to obey the moral principles by which we would choose that members of
the community should ideally govern their conduct” (KE 78). So Wood
regards it as an advantage of FA that it fails what Rawls calls the content
condition. This seems to me less Kantian in spirit than Rawls’s approach
insofar as it not only abandons the specific procedure that Kant thought does
and should guide moral judgment, as both Rawls and Wood do, but it also
abandons Kant’s view of moral judgment as guided by any “procedural
representation of all the requirements of practical reason,” as Rawls puts it,
that is accessible to and rooted in common human reason (MP 237).
The aims and methods of ethical theory

Although I would characterize some aspects of Rawls’s conception of the
aims and methods of ethical theory differently than Wood does, I think
Wood’s criticisms do point to real problems with Rawls’s account. Kant’s
11
On these model-conceptions, see Rawls, KC 520–22.
804 M. Rohlf
own conception, however, seems to me to lie somewhere between Rawls’s

and Wood’s.
What I disagree with in Wood’s reading of Kant on this topic is again his
downplaying of section one of the Groundwork, where it seems to me clear
that Kant does base his search for the supreme principle of morality on
appeals to common rational moral cognition, which for Kant do provide
important confirmation of the independent, philosophical account that he
then goes on to develop in section two. So I would not characterize Kant’s
conception of the aims and methods of ethical theory as philosophical instead
of intuitional, as Wood does, since for Kant the ideal seems rather to be an
account in which both common sense moral intuitions and purely philoso-
phical arguments align or cohere. There is, then, some role in Kant’s model
of ethical theory for appeals to both intuitions and coherence, with the
intuitional element again reflecting Kant’s Rousseauian commitment to the
authority of common human reason, which here as before I think Wood
underemphasizes.
It is true that appeals to intuition do not play the same role in Kant’s
model as they do in what Wood calls the dominant model of ethical theory
that he traces to Sidgwick. But that is because the intuitions appealed to on
Kant’s model are not initially conflicting judgments that need to be recon-
ciled by qualifying or modifying them or by explaining away apparent
conflicts, as Wood characterizes intuitions on the dominant model. Rather,
they are the moral cognitions of common human reason (where “common”
means both unrefined and shared) about which Kant stipulates consensus –
on such matters as the worth of a good will, the value of acting from the
motive of duty, and the role of intended consequences in giving actions
moral worth – and from which he claims to extract FUL as the underlying
formal principle of common sense morality in the extended argument of
section one of the Groundwork. It seems to me that both Wood and Rawls
part company with Kant on this point, with Wood skipping appeals to
intuition altogether and instead emphasizing Kant’s philosophical arguments
in section two of the Groundwork, and with Rawls retaining appeals to
intuition but rejecting Kant’s stipulation of consensus in advance of critical
reflection (MP 274).
As for Rawls, he criticizes classical utilitarianism for harboring the kind
of scientific ambitions of which Wood accuses Rawls but that in fact Rawls
disavows. Utilitarians hold that “every question of right and justice has an
answer,” according to Rawls, while his own “justice as fairness, as a con-
structivist view, holds that not all moral questions we are prompted to ask in
everyday life have answers” and tries only “to identify the most fundamental
questions of justice that can be dealt with” (KC 563–64). Rawls directs this
criticism toward Sidgwick (among others), whose method Rawls understands
as aiming at “reaching true judgments that hold for all rational minds,”
which could equally well be said of Wood’s philosophical method (KC 555).
For Rawls, by contrast, it is a basic fact about modern, democratic societies
that there are “deep and pervasive differences of religious, philosophical,
and ethical doctrine,” such that “for many philosophical and moral notions
public agreement cannot be reached” (KC 539; see also 536). There are
indeed some “shared notions and principles thought to be already latent in
common sense” or at least “certain conceptions and principles congenial to
its most essential convictions and historical traditions” on which we may
hope for consensus (KC 518). But achieving such consensus is a goal for
Rawls, rather than a starting point as it is for Kant, although Rawls finds
support for his approach in what he interprets as Kant’s view of reason as self-
critical and self-authenticating (MP 242–43, 262, 266–68, 324–25). That is,
Rawls interprets Kant as holding that “neither theoretical nor practical reason
is transparent to itself. We can misdescribe our reason as we can misdescribe
anything else.” As a result, “our conceptions of person and society . . . do not
stand alone and as such constitute the basic moral truths” but instead are
products of reflection in which “we are really using our reason to describe
itself. This is a struggle, for the task of coming to understand what to think
on full consideration continues indefinitely” (MP 243). Rawls regards his
constructivism as Kantian in part because it incorporates this conception
of reason as reflective but not transparent to itself, and also because the
“conceptions of person and society” on which it aims to achieve consensus are
substantially Kant’s own. But for Rawls, achieving “reflective equilibrium” on
these matters after due reflection would show only that our (ultimately) shared
convictions are “the most reasonable for us” given our particular social and
historical circumstances, not that they are true in the sense of conforming to
some independent moral order (KC 534; see also 518–19, 554, 568–72).
So Wood is correct that Rawls leaves room for the possibility of “an
indecisive result or . . . a plural moral truth or a form of moral relativism”
(KE 51). But Rawls does not embrace or think we have reason to expect these
outcomes, since he finds it “hard to imagine realistically any new knowledge
that should convince us that these ideals [of the person and of a well-ordered
society] are not feasible, given what we know about the general nature of
the world, as opposed to our particular social and historical circumstances”
(KC 566). What can sometimes appear as an embrace of relativism in “Kantian
Constructivism” comes off in Rawls’s Kant lectures instead as an openness to
constructivism possibly turning out false and its method of authentification
806 M. Rohlf
failing: “As with any other view, constructivism may be false. It must prove
itself by showing, as Kant says, that there is pure practical reason. Doing that
involves . . . showing that the moral law is manifest in our moral thought, feeling,
and conduct” through “such increasing critical reflection as might be achieved by
a tradition of thought from one generation to the next, so that it looks more and
more as if upon fuller reflection the moral view would be constructivist. There
should be increasing success in formulating the doctrine as a whole” (MP 274–
75). Key here is Rawls’s conception of reason not as timeless but as historically
and socially situated, and of rational reflection as an essentially public enterprise,
views for which Rawls finds support partially in Kant but also in Hegel (MP 171,
208–9, 301–2, 366–71). So while Rawls’s model of ethical theory is not in
fact scientific in the way Wood claims, Wood is right to characterize Rawls’s
model as intuitional in the sense that it seeks the most coherent and intuitively
compelling account of at least a central core of our moral intuitions, and indeed
on a massive scale. This sort of model is inherently open to the kinds of
philosophical criticisms raised by Wood, as Rawls recognizes.
I agree in particular with Wood’s criticism that Rawls’s model “aims not
at truth but only to systematize beliefs, which are left without any firm
foundation,” and I will return to the issue of moral truth in the next section
(KE 51). Indeed, Kant does not hold that the aim of ethical theory is just to
systematize beliefs or moral intuitions; but neither is it Kant’s primary aim to
derive the fundamental principle of morality from a purely philosophical
argument about the nature of volition, as Wood claims. The aim of ethical
theory, as I understand Kant’s view, is rather to defend the moral cognition of
common human reason against what Kant calls its own “natural dialectic, that
is, [its] propensity to rationalize against those strict laws of duty and to cast
doubt upon their validity, or at least upon their purity and strictness, and,
where possible, to make them better suited to our wishes and inclinations, that
is, to corrupt them at their basis and to destroy all their dignity” (G 4:405; see
also G 4:459; CPrR 5:48). At times Kant does seem to betray an ambition to
establish the correctness of common reason’s moral cognition through purely
philosophical arguments, but I do not think that is the general tenor of his
approach to ethical theory. The regress argument on which Wood focuses, for
example, occupies a single paragraph in section two of the Groundwork
(G 4:428–29), and it is a notoriously slippery argument which is not obviously
intended to stand on its own in the way Wood interprets it (KE 90–93).12
12
See also Allen W. Wood, Kant’s Ethical Thought (Cambridge: Cambridge University Press, 1999),
124–32.
It may rather be intended as resting on (perhaps among other assumptions)

Kant’s assertion in the previous paragraph: “Now I say that the human being
and in general every rational being exists as an end in itself, not merely as a
means to be used by this or that will at its discretion” (G 4:428). Of a piece
with this sort of outright assertion is Kant’s doctrine of the fact of reason in
the Critique of Practical Reason, which claims that “consciousness of this
fundamental [moral] law . . . forces itself upon us” and is “undeniable”
(CPrR 5:31–32). These sorts of claims incline me to regard the primary
aim of Kant’s model of ethical theory not as establishing but rather as
defending moral principles that Kant assumes we all know (not just believe)
to be true and authoritative for us. This is once again what I have called
Kant’s Rousseauian assumption of the authority and pre-reflective consen-
sus of common human reason.
The methods Kant employs to achieve this aim, as I interpret him, do
include appeals to intuition and coherence as well as independent philoso-
phical arguments, but I think these have somewhat different roles for Kant
than Rawls and Wood assign to them. Our moral intuitions do not need to
prove themselves by fitting into an account that becomes increasingly more
coherent over generations, as Rawls holds. Rawls incorporates even Kant’s
doctrine of the fact of reason into a coherentist account of how he thinks the
moral law is authenticated for Kant, although he admits that “this inter-
pretation of [Kant’s] thought is not generally shared” (MP 268; see 266–68,
322–25). But Kant’s doctrine of the fact of reason implies that the moral
law neither requires nor admits of any authentification, since it is already
“apodictically certain” and “has no need of justifying grounds” (CPrR 5:47).
I suggest that Kant also extends this status to the particular moral cognitions
that he ascribes to common reason in section one of the Groundwork. Moral
philosophy appeals to these intuitions, formulates the principles underlying
them, and argues that both are grounded in our nature as rational and free in
order, as it were, to keep us honest and true to commitments that we are
assumed already to have on some deep level, because the natural dialectic of
practical reason inclines us to ignore or flout these commitments in order to
accommodate our wishes and inclinations.
Rawls complains that “Kant appears to know in advance of critical reflec-
tion how a constructivist doctrine must look,” and I would add that Kant
thinks everyone else knows this too (MP 274). His model of ethical theory is
dogmatic in that sense and basically defensive. It is true that Kant’s moral
philosophy will seem persuasive to us only if we find it compatible with our
intuitions, but that does not make his approach intuitional in a fundamental
way, because Kant neither aims to authenticate those intuitions nor is he
808 M. Rohlf
willing to countenance qualifying, modifying, or explaining any of them

away for the sake of coherence. Likewise, Kant’s model is foundational in the
sense that it aims to uncover a firm basis for the beliefs of common reason in
the form of principles and fundamental values, but not in the sense that it
seeks primarily to justify or establish these foundational principles and values
through independent philosophical arguments. Rather, Kant asserts that
their truth and authority for us are just a fact, and he assumes that every
person can (I do not say will) acknowledge them if they and their connection
to common sense moral beliefs are presented clearly. As Wood admits, “the
mere claim that human beings have absolute worth as ends in themselves may
in the end be more compelling all by itself than any argument that Kant or
anyone else could ever offer for it” (KE 88–89).
The metaethics of autonomy

On the issue of whether Kant’s conception of autonomy is best understood as
implying realism or anti-realism about moral facts and values, I am definitely
on Wood’s side and agree that Kant is a metaethical realist. But I do not
agree with Wood’s characterization of Rawls as an anti-realist. So in this final
section I concentrate on explaining what I understand as Rawls’s non-
metaphysical approach to Kantian moral philosophy and why I do not regard
that aspect of his approach as in fact Kantian.
For Rawls to embrace anti-realism would be incompatible with his
view that public reasoning about moral and political principles should not
appeal to controversial religious or philosophical doctrines on which there
is and possibly could be no consensus (KC 536, 539; MP xii). Since both
realism and anti-realism in metaethics seem to fit this description, Rawls
cannot consistently base Kantian constructivism on either one or the
other. Wood may be correct that there is no stable metaethical position
intermediate between realism and anti-realism, but that does not imply
that Rawls must embrace any metaethical position at all. As I interpret
Rawls, he wants to conduct ethical theory without embracing either horn
of Wood’s dilemma.
Rawls does use apparently anti-realist language, however, which can be
traced back to his interpretation of what he calls Kant’s doctrine of the
priority of the right over the good:
rather than starting from a conception of the good given independently of the
right, we start from a conception of the right – of the moral law – given by pure
(as opposed to empirical) practical reason. . . . A moral conception is not to

revolve around the good as an independent object, but around a (formal)
conception of the right (together with a [formal] conception of a good will),
as constructed by our pure practical reason, into which any permissible good
must fit. (MP 227)
Rawls argues that Kant develops this doctrine in the context of his critique of
the perfectionist intuitionism of Wolff and Leibniz. As Rawls summarizes
Kant’s critique: “it is only if ends are already given that the concept of
perfection can help to determine the will. In the absence of an antecedent
criterion for appropriate ends, perfectionism is indeterminate, or as Kant says
[G 4:443], empty and indefinite” (MP 228). Yet Rawls claims that Kant’s
actual critique of perfectionism as empty is misleading, because it suggests that
Kant would have no objection “if the required ends were given by an object
specified by a rational idea of a prior moral order. . . . But this is not so.
Whenever the object is prior to practical reason itself, and specifies its content
or an end to be furthered by its principles, Kant holds that heteronomy
results.” Rawls emphasizes:
This point must be understood broadly. It says that there neither exists nor
subsists any such object, whether the Supreme Being, or a given moral order of
values (as exemplified, say, by the relations between Platonic Ideas, or by ideas
lying in the divine reason), or an order of nature, or the constitution of human
nature, or the psychic economy of our natural feelings and the laws of harmony
of our inclinations and needs. Kant’s radical sense of autonomy is that any such
allegedly independent object must first stand trial, so to speak, at the bar of the
supremely authoritative principles of pure practical reason before we can grant
its reasonableness. Otherwise, our pure practical reason loses its sovereign
authority as the supreme maker of law. (MP 228–29)
Rawls concludes that Kant’s conception of autonomy is opposed not just to

the perfectionist intuitionism of Wolff and Leibniz, but also to the moral
sense doctrine of Hutcheson and Hume, as well as the rational intuitionism
of Clarke, Sidgwick, Moore, and Ross (MP 230; KC 556–60).
Despite appearances, these are not metaethical claims about the meta-
physical status of moral values or facts, but normative ethical claims
meant to distinguish Kant’s fundamental principle of morality from
various alternatives. Rawls is claiming that Kant’s fundamental principle
of morality specifies (or constructs) the content of all our duties without
appealing to anything external to the requirements of practical reason
810 M. Rohlf
itself, by which he means Kant’s conception of free and equal persons as

reasonable and rational:
In Kant’s moral constructivism, it suffices for heteronomy that first principles

are founded on relations among objects the nature of which is not affected or
determined by our conception of ourselves as reasonable and rational persons
(possessing the powers of practical reason) and by our conception of the public
role of moral principles in a possible realm of ends. Kant’s idea of autonomy
requires that there exists no moral order prior to and independent of those
conceptions that determine the form of the procedure that specifies the content of
the duties of justice and of virtue. (MP 236–37, emphasis added)
I add italics here to emphasize that Rawls’s point is about the norm that
determines the content of duties, namely that it does not refer to a moral
order independent of our conception of the person. Rawls is clear that what is
constructed in accordance with this norm is only “the content of the doctrine”
understood as the duties specified by the CI-procedure; the CI-procedure
itself, however, is not constructed but “simply laid out” (MP 239). It is laid
out on the basis, Rawls says, of “the conception of free and equal persons as
reasonable and rational, a conception that is mirrored in the procedure” (MP
240). This conception, in turn, is “elicited from our moral experience and
reflection and from what is involved in our being able to work through the
CI-procedure and to act from the moral law as it applies to us” (MP 240).
In Rawls’s justice as fairness this same conception is baked into what Rawls
calls “the Reasonable,” which is itself “incorporated into the background
setup of the original position which frames the discussions of the parties”
(KC 529). In that context, Rawls distinguishes three points of view: “that
of [1] the parties in the original position, that of [2] citizens in a well-ordered
society, and finally, that of [3] ourselves – you and me who are examining
justice as fairness as a basis for a conception of justice that may yield a
suitable understanding of freedom and equality” (KC 533). He describes the
parties as “artificial agents of construction within the original position” who
choose first principles of justice from within a framework of constraints
set by the Reasonable (KC 535). But it is really “the way the Reasonable is
represented in the original position [that] leads to the two principles of
justice” chosen by the parties: “These principles are constructed by justice
as fairness as the content of the Reasonable for the basic structure of a well-
ordered society” (KC 530; see also 532). The key point here is that Rawls’s
apparently anti-realist claims apply only from the point of view of the parties,
not from either of the other two points of view:
The parties in the original position do not agree on what the moral facts are, as if
there already were such facts. It is not that, being situated impartially, they have
a clear and undistorted view of a prior and independent moral order. Rather
(for constructivism), there is no such order, and therefore no such facts apart
from the procedure of construction as a whole; the facts are identified by the
principles that result. (KC 568, emphasis added; see also 560–61, 564)
Citizens in a well-ordered society, who occupy the second point of view,

“are indeed bound by first principles and by what [their] duties are,” as these
are constructed by the parties and as they express the content of the
Reasonable (KC 568). From the third point of view, you and I employ the
test of “reflective equilibrium” to assess “how well the view as a whole meshes
with and articulates our more firm considered convictions, at all levels of
generality, after due examination, once all adjustments and revisions that
seem compelling have been made” (KC 534). This is the appropriate test for
us to apply not because Rawls assumes metaethical anti-realism, but rather
because of what Wood calls Rawls’s intuitional model of ethical theory,
which aims to achieve consensus by organizing our considered convictions
about persons, social cooperation, and principles of justice in a democratic
society into the most coherent and intuitively compelling account (KC 569).
I think that Rawls’s non-metaphysical approach to moral (and political)
philosophy is consistent but in that respect non-Kantian. Its Kantian cre-
dentials are again that it incorporates a conception of the person that is
substantially Kant’s own and that it shares an intuitional element with Kant’s
model of ethical theory. But Kant not only assumes a pre-reflective consensus
about the moral cognitions of common human reason, as I have claimed; he
also holds that moral philosophy’s task of defending these cognitions against
the natural dialectic of human reason requires us “however reluctantly, [to]
step forth, namely into metaphysics” (G 4:426). I would emphasize more
than Wood does that the type of metaphysics Kant embarks on here is the
metaphysics of morals and thus constitutes practical cognition valid only
for practical purposes, but I agree with Wood that Kant grounds the moral
cognitions of common human reason on a realist account of rational nature
as an objective end having absolute worth (G 4:427–28).
Rawls admits that he departs from “the role of the a priori and the formal”
in Kant’s view but claims that these “downplayed elements are not at the
heart of [Kant’s] doctrine,” which lies, Rawls thinks, “in [Kant’s] view of free
constructive reason and the idea of coherence that goes with it” (MP 275). So
Rawls is happy to say that our conceptions of person and society are
“convictions about practical reason itself, about its principles and its ideas,”
812 M. Rohlf
and that formulating the doctrine of constructivism amounts to “laying

out the constitution of practical reason as a whole” (MP 243, 271). But
for Rawls the constitution of reason is not “a unity already given for reason to
discover”; instead, “reason supplies its own unity through a critique of itself:
the aim of critique is precisely to establish this unity” (MP 324). This is true
in different ways both of practical reason specifically and of reason as a whole,
as Rawls interprets Kant: internal coherence is a criterion for authenticating
practical reason in a preliminary way, but then practical reason is finally
authenticated only by “showing its role and place within the constitution of
reason as a whole” (MP 272; see 266–77).
This interpretation is almost diametrically opposed to my own reading
of Kant’s account, according to which the a priori and formal principle of
morality is given as a fact to common human reason and assumed to be valid
(CPrR 5:31). The role of moral philosophy is in part to uncover the
substantive value lying at the basis of our pre-reflective commitment to the
authority of the moral law, and this amounts to discovering what is already
independently true: namely, that rational nature exists as an objective end in
itself with absolute worth. The authority of the moral law and the objective
truth of the substantive value lying at its basis imply that they are grounded
not just in some hoped for consensus of freely constructive human reason,
but in the nature of reason as such, precisely as Wood claims. Kant could not
be more emphatic about this:
it is clear that all moral concepts have their seat and origin completely a priori
in reason, and . . . it is of the greatest practical importance not to make its
principles dependent upon the special nature of human reason . . . but instead,
just because moral laws are to hold for every rational being as such, to derive
them from the universal concept of a rational being as such, and in this way to
set forth completely the whole of morals, which needs anthropology for its
application to human beings, at first independently of this as pure philosophy,
that is, as metaphysics. (G 4:411–12)
Even Kant’s further systematic goal “some day to attain insight into the
unity of the whole pure rational faculty (theoretical as well as practical)”
refers to a unity that is already given for us to discover (CPrR 5:91), and
indeed Kant thinks he has discovered it (CPrR 5:121). Discovering this
higher-level unity is not necessary for justifying the moral law, however, but
rather is part of Kant’s larger project of defending the moral law against
objections challenging its compatibility with various theoretical claims
(CJ 5:450–53, 471n).
Wood’s metaethical realist account of the moral law as the natural law of
practical reason seems to me, however, to fit very well with Rawls’s construc-
tivist account of moral judgment and deliberation, when properly understood.
I think the most promising approach to contemporary Kantian moral philo-
sophy involves combining these elements of Rawls’s and Wood’s accounts in
the context of what I have characterized as Kant’s defensive model of ethical
theory.
35
Conclusion: Kant the Philosopher
Matthew C. Altman
Kant is known as one of the last great system-builders, and this book does
little to dispel that image. As evidenced by its many chapters, Kant
formulated and defended positions in nearly every area of philosophy.
Critics claim that he distorted whole fields of inquiry in order to fit them
into a Kantian framework. On this reading, Kant is not a philosopher who
is driven by uncertainty and a spirit of questioning, but a dogmatic thinker
who lays claim to achieving “proof” and “apodictic certainty” in his
demonstrations, thus brooking no dissent (Axv, Bxxii note; MFS 4:474n;
DWL 24:734–35). In this brief conclusion, I will show that this is a
caricature of Kant (or at best an incomplete assessment), and that instead
we ought to think of him as a lover of wisdom in the classical, Socratic
sense. In fact, along with Montaigne and Descartes, Kant is perhaps the
clearest example of a modern philosopher whose intellectual trajectory was
changed by the spirit of questioning and persistent investigations into even
his most basic beliefs.
The young Kant was a rationalist in the tradition of Leibniz and Wolff, and
like Leibniz, he sought to discover “the general principles of corporeal nature and
of mechanics itself [which] . . . belong to some indivisible forms or natures as the
M. C. Altman (*)
Ellensburg, USA

816 M. C. Altman
causes of appearances, rather than to corporeal mass or extension.”1 By his own

account, Kant’s commitment to rationalist metaphysics was shaken by his
encounter with Hume, who demonstrated that a priori concepts could yield no
knowledge about absolute reality. As Kant famously put it, “the remembrance of
David Hume was the very thing that many years ago first interrupted my
dogmatic slumber and gave a completely different direction to my researches in
the field of speculative philosophy” (Pro 4:260). What followed was Kant’s “silent
decade” from 1771 to 1780, during which he published virtually nothing.
We have some indication of his thinking during that time, when he began to
challenge the philosophical assumptions of his rationalist and empiricist pre-
decessors. In a letter to Marcus Herz in 1772, Kant describes a foundational
question that animated his investigations: “What is the ground of the relation of
that in us which we call ‘representation’ to the object?” (C 10:130). This must
have seemed a strange question. Lockean empiricists had assumed that sensa-
tions are caused by and reflect the primary qualities of substances that exist in
the world mind-independently. Rationalists had assumed that, regardless of
what the senses indicate, we can deduce a priori, from incontrovertible first
principles discovered by reason, what must be the case in fact. Hume had
showed, however, that causal claims in particular cannot be justified by appeal-
ing to matters of fact (which are synthetic and a posteriori) or relations of ideas
(which are analytic and a priori). If, as Kant claims, the concept of causality is
necessary to make sense of the world (B4–5, A112–13) – events do not “just
happen” without reference to prior causes – then there is a problem to address,
namely “how are synthetic judgments a priori possible?” (B19; see also B73;
Pro 4:276). “Every event has a cause” is a necessary and universal claim and it is
not derived from experience, so it is a priori (B2–4); and, as opposed to “every
effect has a cause,” it is not merely definitionally true, but rather increases our
knowledge regarding events, so it is synthetic (Pro 4:266–67). Kant notes that
all of the concepts of metaphysics are like this (Pro 4:260), and that they can be
enumerated in a Table of Categories that includes concepts of quantity, quality,
relation, and modality (A80/B106; Pro 4:303). Rationalists and empiricists had
been unable to explain such judgments – indeed, they had not recognized that
there are such judgments – so how synthetic a priori judgments are possible
becomes the animating question of Kant’s theoretical philosophy.
I want to emphasize two important aspects of this well-worn story. First, people
tend not to appreciate what this says about Kant as a philosopher. Kant had served
1
G. W. Leibniz, Discourse on Metaphysics, in Philosophical Essays, trans. and ed. Roger Ariew and Daniel
Garber (Indianapolis: Hackett, 1989), 51–52.
35 Conclusion 817
as a lecturer from 1755 to 1770, when he received the professorship in logic and
metaphysics at the University of Königsberg. Imagine Herr Professor Kant, who by
that time had established himself as an accomplished proponent of Leibnizian-
Wolffian rationalism, deciding that he had not accounted for the relation between
sensibility and the intellect, and spending the next decade trying to figure it out.
Kant followed the idea where it led rather than trying to fit things into an existing
(rationalist) system – at some personal and professional risk.
The second thing to note is the style of Kant’s approach, and the ways that he
lays bare the process of philosophizing in his philosophy. In this regard, Kant’s
writing is more akin to the work of Nietzsche than to other German Idealists
such as Fichte and Hegel. We see this in the fact that he calls attention to the
questions that motivate him, some of which I have noted already, including in
several titles of his works.2 In fact, Kant defines his work in epistemology and
metaphysics, ethics, and religion by stating that “all interest of my reason (the
speculative as well as the practical) is united in the following three questions”:
1) What can I know?

2) What should I do?
3) What may I hope? (A804–5/B832–33)
He later added anthropology to the list:
4) What is a person? (JL 9:25, translation modified; see also C 11:429)
Revealing the questions that motivate him shows the mind at work and gives an air
of tentativeness to his thinking that is absent from the many “treatises” and
“discourses” that populate the history of philosophy. Kant is attuned to the fissures
and resistances within reason itself (such as the desire of reason to go beyond the
bounds of the understanding – more on that later), but this is the same faculty
involved in the work of critique. Kant’s project is thus inherently reflexive in the
sense that it calls attention to the fractured, uncertain endeavor of philosophy. Or
we might say that he at once expresses the desire for systematicity, symmetry, and
intelligibility, while also taking on Hume’s claim that such aspirations must be
recognized as our own contribution to what we take to be true.
In keeping with this spirit of uncertainty, Kant undertakes his Copernican
turn as an “experiment [Versuch, Experiment]” (Bxvi, Bxxi): “let us once try
2
“An Answer to the Question: What Is Enlightenment?” (1784), “What Does It Mean to Orient
Oneself in Thinking?” (1786), and What Real Progress Has Metaphysics Made in Germany since the Time
of Leibniz and Wolff? (1793/1804). The question posed in the title of Real Progress was actually
formulated by the Berlin Royal Academy for a prize essay competition.
818 M. C. Altman
whether we do not get farther with the problems of metaphysics by assuming

that the objects must conform to our cognition” (Bxvi). The work of the first
Critique is meant to validate this experiment and indeed to establish the
Copernican hypothesis with apodictic certainty. However, to do so Kant
turns from what he calls dogmatism and inquires into the conditions for
the possibility of experience by investigating the activity of knowing.
Metaphysics cannot become a science “without an antecedent critique of
its own capacity” – namely, a critique of pure reason (Bxxxv). Kant discovers
through such a critique that experience is in part a function of judgment and
our way of sensing the world, and thus that there are subjective conditions
that make objective claims possible. The result is transcendental idealism,
which separates appearances from things in themselves and limits the applic-
ability of a priori synthetic judgments to “nature as the sum total of all
appearances” (B163).
Again, here critics of Kant emphasize the certainty that he claims to
achieve rather than the investigatory spirit with which he challenges over
two thousand years of accepted philosophical doctrine. The claim that we
can know only appearances and never things in themselves precludes classical
metaphysics, revealing the errors of Descartes’s arguments for the substanti-
ality of the soul and Aristotle’s argument for the unmoved mover, among
other things. Famously, this led the poet Heinrich Heine to call the Critique
of Pure Reason a “destructive, world-crushing thought,” and to say that Kant
should be feared more than the executioner, who, after all, “only kills people”
rather than deeply held worldviews.3
Kant’s philosophy fulfills a negative function in revealing the limits of
what we know, but it also serves a positive function in validating (or
attempting to validate) our use of the categories in experience. We investigate
the world, in part, by investigating the epistemic conditions under which it
becomes a world for us. Kant is not changing what we know, but only the
conditions under which we know it.4 Although he cannot deliver on the
promise of rationalism (or absolute idealism, for that matter), Kant can get us
knowledge beyond what empiricism can deliver (in the form of Humean
3
Heinrich Heine, On the History of Religion and Philosophy in Germany and Other Writings, trans.
Howard Pollack-Milgate, ed. Terry Pinkard (Cambridge: Cambridge University Press, 2007), 79.
4
Hegel derides Kant for such a claim: “This is like attributing right insight to someone, with the
stipulation, however, that he is not fit to see what is true but only what is false. Absurd as this might be,
no less absurd would be a cognition which is true but does not know its subject matter as it is in itself”
(Georg Wilhelm Friedrich Hegel, The Science of Logic, trans. and ed. George di Giovanni [Cambridge:
Cambridge University Press, 2010], 26). Heine compares Kant’s knower to the prisoners, who only see
shadows, in Plato’s allegory of the cave (On the History of Religion and Philosophy in Germany, 82–83).
35 Conclusion 819
skepticism). Thus his synthesis of empiricism and rationalism amounts to a

more human kind of knowledge: not the godlike discovery of the world in
itself by means of thinking alone, but a world of appearances that we navigate
by means of shared a priori concepts.
Kant expresses the sublimity of philosophical thinking with his famous
line from the second Critique, where he emphasizes both our limitations and
our specialness: “Two things fill the mind with ever new and increasing
admiration and reverence [Bewunderung und Ehrfurcht], the more often and
more steadily one reflects on them: the starry heavens above me and the moral
law within me” (CPrR 5:161). In some ways, Kant is echoing Aristotle’s
claim that philosophy begins in wonder (thaumazein).5 Although Kant here
means something more like awe than perplexity, he gets at the unique
position that human beings have and why philosophy is such a human
activity: the ability to give reasons for our beliefs and actions distinguishes
us from the rest of nature and gives us dignity, even as we occupy what seems
like a very small part of the world. Philosophical contemplation is empower-
ing in the way that it both helps us to transcend what seems natural and
obvious, and develops in us a sense of connection with other reason-givers
and the universe as a whole.
Although our knowledge is limited to things as they appear to us, Kant
says that we are inevitably led to make knowledge claims about what he
calls the “unconditioned” – for example, to discover an uncaused cause to
explain “the absolute totality of the series” of conditioned events (A416–
17/B443–45). In doing so, we wrongly employ the categories and make
either erroneous claims (paralogisms) or seemingly contradictory claims
(antinomies). The fact that this is “a natural and unavoidable illusion”
means that we can never stop making such errors, but Kant insists that his
critique does at least succeed in “protecting us from being deceived by it,”
thus correcting the errors of traditional metaphysics (A297–98/B354). In
his ethics, Kant introduces a similar divide: because we have both a predis-
position to good and a propensity to evil, we are able to apprehend and act
for the sake of the moral law, but we are constrained by it as an imperative
rather than being necessarily determined by it (G 4:413–14; Rel 6:29–51).
The Kantian knower is stretched between knowledge and ignorance, and
the Kantian agent is a site of struggle between goodness and evil. Kant
pictures human beings essentially as philosophical creatures, striving for a
5
Aristotle, Aristotle’s Metaphysics, trans. W. D. Ross, vol. 1 (Oxford: Clarendon, 1958), 982b11
(p. 120).
820 M. C. Altman
wisdom or a holiness that we can never achieve – much like Diotima’s

description of the philosopher in Plato’s Symposium.6
It is because our moral progress is endless, Kant says, that we must commit
ourselves to the soul’s immortality as a practical postulate, in order to make the
highest good possible (CPrR 5:122–24). Similarly, Kant claims that we are
constantly engaged in intellectual progress – a process of enlightenment rather
than being enlightened (WE 8:40) – and he says that we must freely pursue it
through our own efforts: “Enlightenment is the human being’s emergence from his
self-incurred minority [or immaturity, Unmündigkeit].” Refusing to think for
ourselves is the result of a lack of will on our parts and an adherence to
dogmatism as a result of “laziness and cowardice.” Thus Kant calls on his
readers to “have courage to make use of your own understanding!” He cautions
especially against allowing books to think for us (WE 8:35).7
Reflecting on his work as a whole, Kant identifies autonomy as the theme
that unites the three Critiques (CJ 5:196–97). In the introduction to this
anthology, I emphasized the different ways that we hold ourselves to stan-
dards: the categories that we use to make sense of the world are imposed by
us through the activity of the understanding, we constrain ourselves morally
by engaging in practical reasoning, and we make aesthetic judgments based
on our own feeling of disinterested pleasure, undetermined by concepts. In
light of Kant’s rejection of dogmatism and his call on us to subject philoso-
phical principles to continual scrutiny (A838/B866), C. D. Broad may give
the best assessment of Kant’s philosophy:
There is no important problem in any branch of philosophy which is not

treated by Kant, and he never treated a problem without saying something
6
Plato, Symposium, trans. Michael Joyce, in The Collected Dialogues of Plato, ed. Edith Hamilton and
Huntington Cairns (Princeton: Princeton University Press, 1989), 204a–b (p. 556).
7
With his references to freeing us from dogmatism, regaining a sense of uncertainty and wonder, and
achieving a highest good in ultimate unity, Bertrand Russell echoes both Kant and Aristotle in his
famous defense of philosophy:
Philosophy, though unable to tell us with certainty what is the true answer to the doubts which it
raises, is able to suggest many possibilities which enlarge our thoughts and free them from the
tyranny of custom. Thus, while diminishing our feeling of certainty as to what things are, it greatly
increases our knowledge as to what they may be; it removes the somewhat arrogant dogmatism of
those who have never travelled into the region of liberating doubt, and it keeps alive our sense of
wonder by showing familiar things in an unfamiliar aspect. . . . Through the greatness of the
universe which philosophy contemplates, the mind also is rendered great, and becomes capable
of that union with the universe which constitutes its highest good. (The Problems of Philosophy
[Oxford: Oxford University Press, 1959], 157, 161)
35 Conclusion 821
illuminating and original about it. He was certainly wrong on many points of
detail, and he may well be wrong in his fundamental principles; but, when all
criticisms have been made, it seems to me that Kant’s failures are more
important than most men’s successes.8
It is very much in keeping with the spirit of Kant’s philosophy not to adhere
slavishly to his views – not to let his books think for us – but rather to
critique even its foundational assumptions in order to establish our own,
independent positions. This began notably with the post-Kantian idealists
(Reinhold, Fichte, Schelling, and Hegel), who attempted to “complete”
Kant’s project by challenging much of it, and it continues through the
work of Heidegger, Sellars, Rawls, Foucault, Bok, and O’Neill, among
many others who have drawn on Kant’s philosophy for inspiration and
staked out their own unique positions.
Kant stands alongside Plato and Aristotle as one of the greatest philoso-
phers of all time, but when this is said, the emphasis is almost always on the
fact that he was great. What I have emphasized here is the fact that he was a
great philosopher. To be sure, this is only one dimension of someone who
wrote thousands of pages and defended numerous positions in nearly every
area of philosophy, but it is foundational to everything that he wrote, and it
is the aspect of Kant that is most alive today. When his work is adopted,
adapted, rejected, or refined, what remains is not always the letter of his
philosophy – as compelling as much of it is – but the spirit of critical inquiry
that is behind it, and the value of autonomy that Kant promotes throughout
the corpus.
8
C. D. Broad, Five Types of Ethical Theory (London: Routledge & Kegal Paul, 1930), 11.
Erratum to: Apperception,
Self-Consciousness, and Self-Knowledge
in Kant
Dennis Schulting
Erratum to:
Chapter 7 in: M. C. Altman (ed.), The Palgrave Kant Handbook,
Palgrave Handbooks in German Idealism, https://doi.org/10.1057/
978-1-137-54656-2_7
The original version of Chapter 7 was inadvertently published without a
symbol ϕ on page 149. The chapter has been updated.
The updated online version of this chapter can be found at https://doi.org/10.1057/978-1-137-54656-2_7
© The Author(s) 2018 E1

in German Idealism, https://doi.org/10.1057/978-1-137-54656-2_36
Index
A Allison, Henry E., 60n14, 73–75, 79,

Abbt, Thomas, 458n21 93–95, 103n27, 238, 240,
Abstraction, from epistemic conditions/ 390–391, 392n13, 393n16,
forms, 74, 77, 84–85, 212, 395, 404–405, 485,
689n13 674–675n35, 679n45, 696n20
from experience/content, 129, 140, Kant’s Theory of Taste, 390–391
173, 176–177, 183–184, 526, Kant’s Transcendental Idealism, 73, 93
573, 577, 581 Altman, Matthew C., 134n22, 635n8
Actuality, 106, 111–112, 114, 129–130, Ammon, C. F., 26
134, 136, 178, 180n18, 186, Analysis, 65, 179–180, 504
187, 399–400, 402–404, Analytical method, see Synthetic vs.
476–478, 485, 489, 508, analytic(al) method
521n13, 535, 734–735, 745, Analytic judgments, 5, 34–35n42, 54–55,
763–770, 774, 783–784 57, 92, 106, 130, 135, 144,
Aesthetics, see Art, philosophy of/ 182–184, 191, 200, 255–256,
aesthetics; Autonomy, 289, 298, 350, 504n28,
aesthetic; System/systemati- 547n10, 573, 581, 739, 763,
city, in aesthetics 765–772, 784–787, 816
Alembert, Jean-Baptiste le Rond See also Analytic-synthetic
d’, 29n28, 510–511 distinction; Truth, analytic
Allgemeine Literatur-Zeitung, 37, 42n67 Analytic mechanics/dynamics, 511

in German Idealism, https://doi.org/I10.1057/978-1-137-54656-2
824 Index
Analytic philosophy, 2, 93, 356, 442, 214, 218–219, 221, 222, 228,
762–789 230, 237–238, 657, 735,
Analytic-synthetic distinction, 739–740, 745, 766, 818
182–183, 739, 787–788 See also Consciousness, as appearance
See also Analytic judgments; Apperception, vs. inner sense/empirical
Synthetic judgments apperception, 141–142, 147,
Animality, vs. humanity, 9, 126, 149, 155–157, 159
335–336, 339–342, 359, 444, transcendental unity of, 139–161,
447, 451, 454n18, 457–461, 231, 481, 573–574, 776
569n1, 649, 660, 711, 717–721 See also Inner sense;
predisposition to, 299, 302, 304, Self-consciousness
335–337, 339, 341–342, 344, A priori, 5–7, 35, 43, 53–54, 57–63,
348–349, 375, 775–776 66–67, 71–73, 78–84,
Anomalous monism, 222–223 101–104, 114, 125, 127n14,
Anthropology, 156, 298, 711, 716, 128–129, 134, 141–142, 144,
751, 754, 776, 817 145–147, 151–154, 157–159,
empirical, 652, 654, 656, 658 161, 167, 168n5, 171–172,
Kant’s teaching of, 30–31, 33, 643, 173–182, 184, 191–192, 194,
706, 709 200, 206, 215, 217, 228, 237,
moral, 345, 645, 653–654, 812 239, 243–244, 247–248,
physiological, 127–128, 643, 255–256, 257–258, 289–291,
646–647, 659 294, 298, 301, 306, 352, 421,
pragmatic, 127–128, 643–664, 432, 442, 473, 474–475,
709–710 477–478, 480, 483–484,
vs. scholastic, 645–646 487–488, 491, 497–507,
See also System/systematicity, in 528n25, 542–543, 555, 557,
anthropology 574, 577–578, 581–583, 586,
Anthropomorphism, 554, 562 610, 679n44, 695n19, 739,
A posteriori, 4, 5, 129, 153n22, 206, 754, 759–760, 764–765, 768,
247, 294, 298, 475, 483, 491, 770, 774–775, 781–787, 789,
672, 774–775, 785–787, 816 811–812, 816, 818–819
Appearances, 6–7, 35, 71–72, 74–77, See also Cognition (Erkenntnis),
82–84, 89, 95–99, 122, 131, a priori; Intuition, a priori/
133–134, 166, 185, 193–195, pure; Knowledge (Wissen),
198–204, 207–208, 219–220, a priori; Synthetic a priori
226, 228, 231, 233, 237, 239, judgments
240, 456, 474–479, 570, 631, Architectonic, 167, 184–187, 476, 478,
657–659, 661, 686, 687, 690, 539, 570, 685, 689–690,
734–735, 743–744, 780–781, 740–741
819 Arendt, Hannah, 421n11, 438,
vs. things in themselves, 71–90, 445n29, 757–759
91–116, 133, 207–208, 212, Aristotelianism, 718
Index 825
Aristotle, 165, 174, 235, 269, 271, Methodenbuch für Väter und
357, 408, 449, 471, 473, Mütter der Familien und
718–719, 753, 818–821 Völker, 713, 717
Art, 10–13, 43, 171, 399, 409n4, Baumgarten, Alexander Gottlieb,
412n6, 515–538, 550–551 31–32n36, 51, 53–54, 56n9,
philosophy of/aesthetics, 2, 3, 63–66, 155, 169n7, 270, 314,
10–13, 42–43, 50, 64, 66, 319, 654
381–467, 744, 746 Acroasis logica, 169
See also Judgment, aesthetic Metaphysica, 31–32n36, 65, 654
Assent, 408–409, 417–419, 515–538, Beauty, 2–4, 10–13, 43, 66, 187n25,
550–551 343, 348, 382, 384–387, 390,
Assistance, duty of, 298, 307–329 391n12, 392–393, 395,
Association of ideas, 3–5, 200, 336, 398–400, 402, 404–405,
339, 429, 445, 673, 679n45 408–414, 417–419, 428,
Autonomy, 36, 267, 290–292, 435–438, 442–444, 448, 452,
303–304, 314, 318, 355, 453n14, 454, 456–457, 463,
358–361, 362n22, 363, 668, 672n29, 674–677
367–377, 527, 594, 793, natural, 10–12, 398–399, 431, 436,
808–810, 820–821 438, 442–443, 463–464n32
aesthetic, 425–427, 431–433, 435, as symbol of morality, 708
437, 443, 708 Beauvoir, Simone de, 2, 349
defined, 219–220, 289, 361, 432, Second Sex, 349
798 Beccaria, Cesare, 634–635
political, 594 Beck, Lewis White, 710–711
and religion, 44, 46, 527, 541, 543, Belief, see Faith/belief (Glaube)
549, 564, 751 Beneficence, 269–271, 276, 310,
See also Categorical imperative, 312n10, 313–315, 320–328
Formula of Autonomy See also Assistance, duty of
(FA); Freedom Benevolence, 273, 275–276, 312n10,
Ayer, A. J., 226 313, 319n16, 321–329
Benjamin, Walter, 416n8
“Goethe’s Elective Affinities,” 416n8
Bennett, Jonathan, 98, 235
B Berkeley, George, 37, 88, 502, 745
Bacon, Francis, 762, 788 See also Idealism, Berkeleyan
Bahrdt, Carl Friedrich, 715 Berlin, Isaiah, 24
Baldinger, Ernst Gottfried, 487 Bible, 23, 535, 717
Grenzen der Naturlehre, 487 Biology, 43, 486, 495, 691–694,
Balibar, Étienne, 601, 602, 613 718–719, 722, 757
Baron, Marcia, 310, 666 Bird, Graham, 73, 75, 238–239
Basedow, Johann Bernhard, 708n7, Bock, Friedrich Samuel, 713
712–717, 723–727 Lehrbuch der Erziehungskunst, 713
826 Index
Body, 52, 89, 205, 305, 315, 339, 350, Formula of Humanity (FH), 10,
494, 573, 596, 647, 669n20, 275, 279–280, 284, 300, 301,
693, 744, 747, 780 338, 796, 798
doctrine of, 476, 478, 494, 498, Formula of the Law of Nature (FLN),
500–501, 506–507, 510 275–280, 284, 317, 361n17,
Bohl, J. C., 27 368, 796–797, 801–803
Borges, Maria, 666 Formula of the Realm (or Kingdom)
Borowski, Ludwig Ernst, 23n5, 27, of Ends (FRE), 275, 280,
32–33, 709 361n17, 362n23, 364–365,
Brahl, Johann, 46 366n30, 370n38, 796–797
Brandt, Reinhard, 725 Formula of Universal Law (FUL), 8,
Breidbach, Olaf, 488 253, 255, 274–280, 284, 288,
British Idealism, 2 290–292, 295–297, 300,
Broad, C. D., 820–821 316–318, 333–334, 337–339,
Buck, Friedrich Johann, 27, 30n31, 362, 545, 577, 793, 796–797,
33n40 800–804 (see also
Buddhism, 743 Universalizability)
Butler, Joseph (Bishop), 526n22 punishment as, 626
Butler, Judith, 2 See also Moral law, related to
categorical imperative
Category/ies, 6–7, 9, 11, 54, 62,
C 90, 126, 129–132, 134, 141,
Campbell, John, 2 148, 174–177, 181, 184,
Campe, Joachim Heinrich, 715 186, 199, 201–202, 204,
Carlson, Allen, 442 212–213, 221, 232, 234,
Carnap, Rudolf, 771 235, 244, 408, 423n12,
Cassirer, Ernst, 755–756, 759 475n21, 479, 481, 496–497,
Casuistry, 272, 279 502, 506–507, 580, 745,
Catechism, moral, 324, 707, 709 754, 767, 818–820
Categorical imperative, 8, 10, 36, mathematical, 474–475
194, 200, 209, 214–215, table of, 6, 122–123, 187n26, 408,
217, 236, 239, 240, 504, 509, 595, 816
249–250, 251, 267, 280, See also Causality, category of;
286, 288, 290–306, 317, Substance, category of; System/
338, 361n17, 365, 366n30, systematicity, of concepts
461, 577, 579–580, 582, Causality, 74, 87, 133, 198, 202,
620, 797, 801–803 203, 213, 223, 227–228,
CI-procedure, 269, 275–277, 281, 241, 267, 291, 292, 339,
284, 295–298, 796, 475, 483, 495, 528, 659,
800–803, 810 660, 690–691, 744
Formula of Autonomy (FA), 275, category of, 5, 7, 126, 131–133,
280, 796–797, 801, 803 186, 214, 217–218, 220, 221,
Index 827
231–240, 477–478, 481, 549, cosmopolitan, 593–615, 652, 723

754, 816 Civilization, 596, 711
See also Objectivity/objective Civil society, 578–579, 581–582, 584,
validity, of causality 587, 602–603, 618–619, 626,
Cavallar, Georg, 716, 717 710
Cavell, Stanley, 423n12 Clarke, Samuel, 809
Chalmers, David, 762n1, 767 Coercion, 351, 428, 579–581,
Character, see Gesinnung (character, 583, 585, 586, 588–589,
disposition, conviction, 622–626, 629–630,
attitude) 635–640
Chemistry, 416n8, 477, 487, Cogito, ergo sum, 141, 735
499n16, 687 Cognition (Erkenntnis), 3n2, 7, 53–54,
Chignell, Andrew, 108n39, 518, 57–59, 61, 63–66, 71–74, 76,
520n9, 522n15, 523 94–95, 98n14, 99–102,
Chiliasm, 683n1 104–109, 111, 114–115,
Choice (Willkür), 85n16, 196, 202, 203, 117–137, 139, 144n10, 154,
207–208, 222, 224, 226–227, 158n29, 173, 178–185, 192,
230, 233–236, 238–239, 241, 195, 197–199, 204, 215, 221,
289, 339, 342, 358, 527, 229, 236, 240, 289, 298, 383,
569n1, 578, 579, 659–661, 385–386, 397–398, 413–417,
691, 720–721 419–420, 425, 431, 438–444,
See also Will 446, 451–453, 455–456,
Christianity, 22, 44–45, 46n72, 474–475, 477–479, 482n38,
516n3, 519, 522–523, 530, 486, 498n14, 528n25, 544,
550, 563 575, 670–676, 679, 680n46,
ethics, 24–25, 45, 464–465, 563 685–692, 703, 724, 736–747,
See also Jesus Christ (“son of God”); 750, 753–754, 765–768,
Lutheranism; Pietism 775–777, 817–818
Christology, 541, 546–548, 549, a priori, 61, 72–73, 79–80, 82–84,
551–555, 561 167, 177–185, 206, 237, 289,
vs. jesuology, 551–552 478, 499–505, 574, 774
Church, 531n30, 533–534, defined, 118
542–543n6, 544–545, faculty/ies of, 43, 53, 290, 385,
547–548, 555–561, 389, 391–396, 399, 415–416,
563–564 421, 437, 454, 456, 519,
Cicero, 23 648, 654, 668n15, 669n19,
CI-procedure, see Categorical impera- 676–677, 681, 752, 755,
tive, CI-procedure 757–758
Citizenship, 350–353, 373, 547, 571, historical, 552n17
586, 594, 602, 726n48, in general, 402–405, 410, 675
810–811 intuitive, 124–126 (see also
active vs. passive, 345, 352–353 Intuition, intellectual)
828 Index
Cognition (Erkenntnis) (cont.) collective, 614–615

mathematical, 125–126, 129 as inner court, 281–284
of oneself, 140–143, 146–147, Consciousness, 1, 6, 33, 75n8, 118,
155–157, 159–161, 291, 573, 120, 121n10, 124, 127–128,
657, 734 131–132, 168–169, 175, 180,
perfect, 122–123, 133 184–185, 194, 227, 229, 294,
practical, 178, 186, 191–192, 322, 335–336, 383, 386, 387,
194–195, 209, 300, 305–306, 389, 391–392, 394–400,
529, 804, 806–807, 811 404–405, 476, 480–481, 573,
of supersensible objects, 53–54, 61, 679, 693, 735, 737, 741, 742,
193, 207, 212–213, 234, 752–753, 776
237–238, 292, 523, 555, 568, analytical unity of, 144, 151–153,
582, 632, 661, 732 159
See also Knowledge (Wissen), vs. as appearance, 145–147, 156–157,
cognition (Erkenntnis); 159, 169
System/systematicity, of cosmopolitan, 613–614
cognitions moral, 127, 255–257, 677, 807
Cohen, Hermann, 727n50 objective unity of, 159–160
Colonialism, 596–597, 610 transcendental, 142, 481
Common sense (sensus communis), 12, 41, See also Apperception; Inner sense;
407–424, 431–433, 435, 437 Self-consciousness
Community, ethical, 546, 548, 555–557, Consequentialism, 618–620, 623,
564, 689n13 625n5, 791
Compatibilism, 202, 204, 211–212, Conservation of mass, 5, 508–509
221–228, 230, 231, 234, 237, Constitution, civil, 583, 586–589,
239–242, 634 607, 618, 625, 640,
defined, 202, 221, 226 696–698, 710
See also Post-compatibilism See also Law, constitutional
Concepts, empty, 55, 99, 112, 124, Constitutive principles, 56, 140, 182,
129–130, 134, 208, 298, 238, 240, 290, 363, 369, 419,
550n13, 662, 752, 764, 454, 482, 485, 489, 503,
767, 781 568, 719
See also Category/ies; System/ Contract, social/original, 350, 580,
systematicity, of concepts 581, 584–588, 621–622,
Conceptualism, 762–763, 769, 634–636
774–778, 784–787 See also Property rights/rightful
defined, 775 possession
See also Non-Conceptualism Conversion, 540–541, 545–546,
Condillac, Abbé de, 477n29 548–553, 555, 564
Confucianism, 24–25 Copernican revolution/turn,
Conscience, 281–284, 288, 364, 374, vs. Copernican devolution,
534, 699, 758–759 762–763, 767–768, 788
Index 829
in philosophy, 1, 3, 6, 16, 133, problematic; Skepticism,

165–166, 570, 756, 759, Cartesian
817–818 Desire, faculty/ies of, 43, 217–218,
Cosmopolitanism, 589–590, 222, 290, 322, 361, 456, 654,
593–615, 652, 689–691, 655–656, 658, 660–661,
694, 695, 699–703, 716, 668–669, 677, 679, 681, 752,
722, 723 755, 757, 777
See also Citizenship, cosmopolitan; Despotism, 525, 560, 562–563, 604
Consciousness, cosmopolitan; Determinism, 86, 103, 192, 202,
Education, cosmopolitan 203n16, 204–205, 207–208,
Critique, 2, 10–11, 49–50, 53, 56–63, 211, 214, 220–223, 225–228,
66–67, 167, 177–182, 184, 232–234, 236–239, 241, 627,
216, 246, 373n47, 474, 631, 653
540–543, 549–551, 553–555, Dewey, John, 706
558–559, 562, 564, 570–572, Democracy and Education, 706
734, 740, 750, 805, 819, 821 DeWitt, Janelle, 666
Crusius, Christian August, 29n29, Dialectical illusion/inference, 498, 573,
34n41, 54n5, 102 686, 738
as rationalization, 806, 807, 811
See also Kant, Immanuel, work:
D Critique of Pure Reason,
Davidson, Donald, 132, 133n18, 222, Transcendental Dialectic
224, 240 Di Giovanni, George, 736–737
Definition, nominal, 197 Dignity, 284, 315, 316, 362, 590, 613,
real, 543–545 621, 629, 723, 806, 819
transcendental, 85 See also Worth/value, absolute
Deimling, Wiebke, 666 Discursivity, 53, 61, 93–94, 123n12,
Democracy, 558, 572, 594, 598, 124–126, 156, 212, 272, 454,
602–603, 608, 614–615 481, 503n25, 671
Denkungsart (mode of thought), 16, Disinterestedness, 11–13, 43, 407,
644, 654–664 438–442, 820
vs. Sinnesart (mode of sense), 656–658, Divisibility, infinite, 86
661–662n32–33 Dogmatism, see Metaphysics, dogmatic
Derrida, Jacques, 610 Drunkenness, 270, 303–304, 324
Descartes, René, 4, 29, 44, 64–65, 112, Dualism, Cartesian, 205, 766, 770
205, 408n3, 473, 501, 502, See also Modality, dualism vs.
509–510, 646, 718–719, 732, monism
734–735, 749–750, 762, 788, Duties, perfect and imperfect, 271,
815, 818 273, 276, 283, 284, 301–302,
Principles of Philosophy, 509–510 304, 305, 307–329, 333,
See also Cogito, ergo sum; Dualism, 337–338, 348–349
Cartesian; Idealism, See also Talents, duty to cultivate
830 Index
E Erxleben, Johann Christian Polycarp,

Eberhard, Johann Augustus, 37, 53–54, 486–487
56, 67–68 Anfangsgründe der Naturlehre,
Education, 345, 458, 658–659, 663, 486–487
705–727 Ethical community, see Community,
aesthetic, 708 ethical
cosmopolitan, 716, 722–723 Eudaimonism, 626, 683n1, 696
moral, 310–311, 447–448, Euler, Leonhard, 29n29
707–708, 727 Evans, Gareth, 2
public, 726–727 Evil, propensity to, 217, 341, 343, 531,
of women, 345, 707n4 540n4, 546–547, 552, 555,
Emerson, Ralph Waldo, 423n12 653–654, 819
Empirical realism, see Realism, Extension, 85, 99, 113, 130, 227,
empirical 498–499, 501, 782, 815–816
Empiricism, 4–5, 59, 94, 200, 227, Extraterrestrials, 518, 720n39
228, 491, 687, 747, 751, 787,
816, 818–819
British, 50, 472, 787, 816
Logical, 787 F
regarding standards of taste, 3, 445 Fact of reason, 214–216, 248,
See also Berkeley, George; Hume, 255–256, 258, 582, 807
David; Locke, John; Skepticism, Faculties, see Desire, faculty/ies of;
empiricist Feeling(s), faculty of;
“Empty formalism” charge, 277, 295, 298 Judgment, power/faculty of;
Enlightenment, Age of, 26, 43–44, 509, Knowledge (Wissen), faculty/ies
525, 526n22, 542–543n6, 605, of; Soul, faculties of; System/
708, 712, 716–717, 720, 724, systematicity, of faculties
733, 750, 751 Faith/belief (Glaube), 44, 92, 103,
Enthusiasm (Schwärmerei), moral, 449n5 119–120, 122, 135–136,
religious, 540, 542–543n6, 195n5, 214, 221, 224, 231,
549–550, 554, 558, 562, 235, 515–538, 541–542,
575–576, 736, 751 542–543n6, 550–551,
Epistemic conditions, 73–74, 79, 553–555, 560–562, 564,
93–95, 115, 122, 818 736, 745n35, 751
Equality, moral/legal, 34, 300, defined, 517
313–314, 337, 342, 349, historical/ecclesiastical, 530, 531,
351, 361, 366, 369–371, 537n33, 557, 559–561, 563
373, 375–376, 445n29, 580, Feder, J. G. H., 36–37, 88, 88–89n22
601–602, 608, 615, 621, Feeling(s), 3, 5, 10, 11–13, 85n16, 207,
622–623, 629–630, 723, 270, 286, 288–289, 321–322,
733, 793, 803, 809–810 381–405, 410–414, 416–418,
retributive, 621–622, 624, 628 422–424, 426, 428, 430–437,
Index 831
440–442, 444–446, 451–452, Order of Things, 756

454n18, 660–661, 665–681, Foundationalism, 473, 737
738, 742, 776 Francke, August Hermann, 24–25
defined, 383, 389 Frankfurt, Harry, 221–222, 226–227
faculty of, 43, 654, 668 Frankfurt School, 2
moral, 216–218, 224, 269, 282, 313, Freedom, 9–10, 34–44, 103–104, 136,
338, 369, 376, 460–461, 191–197, 202–205, 207–209,
462n29, 463–464, 465n34, 211–242, 255–256, 258, 260,
556–557, 564, 667 267–268, 280, 288, 292,
self-feeling, 140, 146 (see also 304–305, 309, 314–315, 335,
Perception, self-perception; 344–345, 350–351, 366n30,
Self-affection) 372, 373n47, 396, 399, 421,
vs. sensations, 451 432, 444, 454n15, 509n36,
of the sublime, 456–457, 459–460, 523, 535, 558, 560, 564, 569,
464, 466, 606 (see also 571, 572, 576, 578–582,
Sublimity) 587–590, 596, 602, 604,
Fichte, Johann Gottlieb, 2, 47, 134n22, 605, 608, 618, 622–624,
143, 146–147n12, 148, 154, 627, 629, 631–638, 644,
159–160, 490n77, 738, 649–650, 652, 656n23,
740–743, 752, 817, 821 658–663, 676, 677, 680n46,
Foundation of the Entire 689n13, 690–691, 698, 708,
Wissenschaftslehre, 741n22 716, 720–721, 741–743,
“Second Introduction to the 745, 751–752, 758, 810
Wissenschaftslehre,” 741n20 absolute, 103n27–28, 192,
Transcendental Logic, 752 196–197, 203, 207–208, 220,
Fine, Kit, 762n1, 767 222, 226, 236, 627
Firestone, Chris L., 523–524, 533n31 comparative, 222, 225, 231, 240,
Forces, attractive and repulsive, 114n52, 659–660
508, 692 cosmological, 220
Formalism, aesthetic, 425–446 empirical, 656, 659–661, 663
See also “Empty formalism” charge of expression, 350, 571–572, 604,
Forms of sensible intuition (space and 611
time), 6, 53, 71–90, 92, 97, external/outer, 578, 580
99, 101, 104, 115, 166, 201, idea of, 10, 103, 214–215, 225,
205, 212–213, 228, 230–231, 238–240, 242, 454n15
510, 734–735, 739, 744, 754, indiscernibility of, 627
765–766, 769, 771, 774, 781, innate, 581, 588
783–784 internal/inner, 304–305, 558
Förster, Eckart, 502 negative, 219
Foucault, Michel, 2, 718, 750, 754, objective reality of, 216
756–757, 759–760, 821 “of a turnspit,” 9, 34n41, 202
Madness and Civilization, 756 positive, 220, 226
832 Index
Freedom (cont.) Frierson, Patrick, 666

possibility of, 34–35, 44, 103, Kant’s Empirical Psychology, 654
135, 194, 212, 214–217,
219–220, 222, 223, 226,
228, 234–235, 237–238, G
243–260, 268, 627, 658 Garve, Christian, 36–37, 88, 243
practical, 203, 207, 219–220, Genius, 43, 67, 412n6, 420, 431,
221, 226, 268, 660–662 433–435, 437
psychological, 228, 659–660 Geometry, 5, 73, 79–82, 104, 171,
system of, 576 493, 495n5, 504–505, 782
transcendental, 203, 207, 209, Gerhardt, Volker, 605
220–222, 225, 227–228, 232, Gesinnung (character, disposition,
234, 237, 238–239, 627, conviction, attitude),
631–635, 638, 653–654, 222–223, 228–230, 232–234,
656–661, 663–664 239, 269–270, 281, 449, 458,
See also Autonomy; Compatibilism; 463, 467, 548, 651, 654–663,
Determinism; God, freedom 702, 721, 744
of; Incompatibilism; empirical character, 203n16, 230,
Indeterminism; Libertarianism 233–234, 631, 633, 654,
(theory of freedom); Postulates 656–662, 691
of practical reason; Skepticism, intelligible character, 203n16,
free will; Unconditioned/ 218, 226, 230, 232, 234,
unconditional, freedom 631–633, 654, 656–657,
Free play of imagination and 659–662, 691
understanding, 12–13, 381, See also Maxim, supreme, as
384–385, 387–400, Gesinnung
403–405, 673, 674–677, Gibbard, Allan, 264–265, 271
751–752 Thinking How to Live, 264
Frege, Gottlob, 171n9, 772 Ginsborg, Hannah, 384–385, 388
Basic Laws of Arithmetic, 772 God, 24, 34, 42, 44–46, 86, 88,
French Revolution, 445n29, 593–594, 113–114, 136, 181, 185,
602, 604, 606–607, 614–615, 192–193, 194, 282, 288,
684n3, 751 363, 498n12, 515, 523,
Friedman, Michael, 482–484, 530, 531n30, 535–537,
501, 502n21, 503, 541, 544n8, 549–564,
505n29, 510 576, 657, 687, 689n13,
Friedrich II (“the Great”), King (of 718n57, 732, 750, 766,
Prussia), 25–26 768, 773
Friedrich Wilhelm I, King (of Prussia), 24 as author of the world, 693
Friedrich Wilhelm II, King (of Prussia), freedom of, 219n4
43–44 as lawgiver, 547–548
Index 833
people of, 546, 557, 689n13 528–529, 536, 576, 623, 645,
way of knowing, 125, 555, 657, 647–650, 651–652, 677,
786–787 (see also Cognition 686n8, 688–689, 699, 705, 717
[Erkenntnis], intuitive; See also Highest good
Intuition, intellectual) Hawthorne, John, 762n1, 767
See also Jesus Christ (“son of God”); Hegel, Georg Wilhelm Friedrich, 2, 242,
Postulates of practical reason; 277, 295, 740–743, 746, 752,
Revelation, divine 758, 784, 806, 817, 818, 821
Goethe, Johann Wolfgang von, 36, Encyclopedia of Philosophical
416n8, 523n16, 714–715 Sciences, 741n22
Dichtung und Wahrheit, 714 Science of Logic, 752
Elective Affinities, 416n8 Hegelianism, 2, 361
Good, predisposition to, 335, 338, Heidegger, Martin, 2, 282, 754–755,
341–342, 540, 549–550, 553, 758–760, 821
659, 819 Being and Time, 754
See also Personality, predisposition to Heilsberg, Christoph Friedrich, 26
Good will, 8, 218, 544, 551, 564, 584, Heine, Heinrich, 38–39, 818
651, 664, 699, 804, 809 Henrich, Dieter, 140, 145,
Grace, 24, 535–536, 540–541, 180–181n19, 490, 698–699
542–543n6, 544, 545, Herbart, Johann Friedrich, 753
548–551, 555, 562–564 Herder, Johann Gottfried, 26, 28,
Gratitude, 270, 312n10, 313, 323, 685n7, 715
326, 328 Ideas for the Philosophy of History of
Green, Joseph, 39–40 Humanity, 685n7
Green, T. H., 2 Herman, Barbara, 332–333, 334–335,
Greenberg, Clement, 2, 426–427, 349, 792
430, 444 Herz, Marcus, 118, 643, 651, 655n22,
Guyer, Paul, 96, 98, 120n6, 134, 738–739, 763, 783, 816
384–385, 390, 402, 403n29, Heteronomy, 220, 289, 318–319, 367,
440–441, 665, 671n27, 443, 527, 550, 809–810
672–673n31, 674–675n35, 686 Highest good, 55, 136, 517, 519,
527–530, 536–538, 544,
H 545n9, 547–548, 551, 576,
Habermas, Jürgen, 2, 610 689, 820
Hamann, Johann Georg, 715 Hill, Thomas Jr., 310, 792
Hanna, Robert, 2, 118–119, 225–226, Hippel, T. G. von, 23, 25, 33n40, 39
232n24, 238n28 Man Who Lived by the Clock, 39
Happiness, 7–8, 217–218, 266–269, 271, History
273–276, 285–290, 310–311, natural, 477, 487–488
317–326, 337–338, 348–349, philosophy of, 603, 607, 683–703,
458, 462–463, 527n24, 710–711, 716, 727
834 Index
History (cont.) Hume, David, 3–5, 11, 29n29,

sacred, 536–538 34–35, 36–37, 49, 50, 54,
See also Cognition (Erkenntnis), 59, 180n18, 216, 226–227,
historical; Organism, history 230, 244, 265, 427,
compared to; System/ 429–430, 472, 474, 481,
systematicity, in history 734, 736, 739, 749–750,
Hobbes, Thomas, 360n14, 574, 580, 585 809, 816–817
Hogan, Desmond, 101, 102n23, Of the Standard of Taste, 427
102–103n25, 103n27–28, See also Induction, problem of
104, 523–524 (Hume); Skepticism, Humean
Homogeneity, 110n44, 416n8, Humility, 464–467
477n28, 481, 495–496, 507 epistemic, 85, 97, 591
Hope, 533, 536–538, 542–543n6, 550, Husserl, Edmund, 2, 752–754
554–555, 564, 568, 688–690, Logical Investigations, 752–753
696, 699, 817 Hutcheson, Francis, 809
Horen, 751 Huygens, Christiaan, 510
Hospitality, right of, 589–590, 596, Hypothetical imperative, 200, 236,
610 249–251, 298–299, 301
Hudson, Hud, 222–224
Hufeland, Gottlieb, 37
Humanity, 271, 315, 342, 346, I
348–349, 465, 552, 554, Idea(s), 436
560, 568, 581, 613–614, aesthetic, 426, 435–439, 444–445
620, 621–622, 649, 651, of reason, 57–58, 103, 294, 300,
689, 707, 711, 716–717, 419, 439, 454–455, 554, 573,
722–723 586, 739, 800
crime against, 625, 628–630 regulative, 195, 238, 240, 368, 419,
as end in itself, 10, 217, 270, 275, 631, 658, 696
300, 338, 798 of the understanding, 739–740
predisposition to, 335–338, See also Freedom, idea of
341–342, 722 Idealism, absolute, 740, 818–819
See also Animality, vs. humanity; Berkeleyan, 6, 36–37, 133
Categorical imperative, defined, 6–7, 35, 91, 133, 134, 166,
Formula of Humanity (FH) 212, 734–735, 774
Human nature, 298–299, 306, dogmatic, 37
333, 335–345, 348, 350, material, 502
450–451, 461n25, 467, objective, 740
545–546, 549, 553, Platonic, 759, 809
557–560, 649, 651–652, problematic, 734–735
706–707, 709, 711, subjective, 88
716–721, 775–776, transcendental, 1, 35, 71–90, 92, 93,
795–796, 809 104–105, 134n22, 166, 195,
Index 835
201, 202n15, 205, 208, 209, Inner sense, 75, 78–79, 141–142, 147,
214, 219, 221, 224–226, 502, 155–161, 181, 222, 224, 228,
509, 516, 518, 523–526, 535, 230, 282, 476, 485, 490n80,
654, 731, 734–737, 740–741, 498–499, 574, 780–781
747–748, 763, 765, 773–775, See also Apperception, vs. inner sense/
778, 783–785, 789, 818 empirical apperception
See also British Idealism Insole, Christopher, 523–524
Identity, 141–144 Intelligible world, 92, 135, 185n22,
Imagination, 12, 180n18, 342–345, 212, 219, 225, 230–231n20,
348, 394–399, 403, 404–405, 576, 579, 689
415–417, 423, 435, 437, Interest, moral, 217–218, 222, 527
439–440, 455, 456, 458–459, Intuition, a priori/pure, 79, 84, 102, 159,
675, 738, 754–755, 757, 776 501, 504–505, 739, 781, 784
productive, 158, 193n4, 405n34, 749 formal, 82, 776
transcendental synthesis of, 158, intellectual, 85–86, 146–148, 155,
396–397 200n10, 212, 255–256, 741,
See also Free play of imagination and 748, 773 (see also Cognition
understanding [Erkenntnis], intuitive)
Immortality, see Soul, immortality of manifold in (see Representations,
Imperatives, see Categorical imperative; manifold of)
Hypothetical imperative sensible/empirical, 6–7, 53–55, 71–72,
Inclination, 8–10, 186, 195–196, 202, 81–83, 85, 88, 90, 94–95, 99,
204, 215–218, 226, 241, 104–106, 108–109, 111, 114,
257–258, 279, 286, 291–294, 118–119, 123–131, 134,
296, 299–300, 302, 304–305, 145–147, 151–159, 166,
308–310, 315, 318–319, 324, 168–172, 175–182, 184–186,
337n9, 360, 418, 420, 456, 199–201, 205, 212–213,
462n27, 521, 527, 552, 579, 229–230, 236, 255–256, 282,
656, 660–661, 667, 669, 711, 290, 298, 382, 386, 390n7,
806–807, 809 397–399, 402–404, 418,
defined, 669 426–428, 436, 439, 451,
Incompatibilism, 102–104, 226–228, 453n12, 457–458, 474–475,
240 478, 481, 495n4, 496, 501,
See also Libertarianism (theory 503n25, 568, 673–674n33,
of freedom) 675, 676n36, 739, 748–749,
Incongruent conterparts, 779–782 752, 754, 764–767, 769–771,
Incorporation thesis, 219, 338, 669 780–783, 785–786
Indeterminism, 230, 240 See also Forms of sensible intuition
Induction, 474, 483 (space and time)
problem of (Hume), 3, 480–481 Intuitionism, 280, 791–792, 809
Inertia, 485n56, 505, 508–509 Islam, 45, 519
836 Index
J power/faculty of, 11–12, 63n18,

Jachmann, Reinhold Bernard, 38n38, 67, 169, 174, 176n14,
39–41, 47 186–187, 398, 409, 480,
Jacobi, Friedrich Heinrich, 732, 734, 482, 607, 679n44, 692,
736–737, 745 707, 719
David Hume on Faith, 736 reflective, 11–12, 13, 50, 57,
Jacobs, Nathan, 523–524 60–64, 66–68, 384–386,
Jensch, Christian Friedrich, 31, 394–395, 410, 425–426,
31–32n36 428, 430–433, 435–438,
Jesus Christ (“son of God”), 25, 45, 541, 440–446, 452n8, 480,
542–543n6, 544–545, 482–483, 528n25, 671–672,
552–555, 564 676, 696, 719
See also Christianity; Christology
Judaism, 45
Judgment, 3–5, 57, 59, 62–64, 67,
83–84, 106–109, 111, 115, K
118–127, 129–131, 133, Kant, Immanuel, life, 21–47
135–137, 141, 156, work:
174–177, 181, 185, “Announcement of the Programme
197–201, 204–207, 209, of His Lectures,” 708–709
213, 232, 451, 480, 504, “Answer to the Question: What Is
507, 527, 666, 673–674n33, Enlightenment?”, 28n24, 577,
675–676, 739, 749, 753, 605, 817n2
769, 776, 784–787, 818 Anthropology from a Pragmatic
aesthetic, 10–13, 43, 66, 187n25, Point of View, 41, 156, 157,
381–405, 407–424, 425–446, 161, 415n7, 416n8, 420–421,
451–467, 521n13, 665–681, 424, 448, 536, 643–664,
741n21, 755, 820 (see also 709–711, 722, 727, 755,
Empiricism, regarding 756–757
standards of taste; Knowledge Attempt to Introduce the Concept of
[Wissen], in aesthetic Negative Magnitudes into
judgment) Philosophy, 34–35n42,
determinate/determinant, 11–12, 112–113
61–62, 67, 133, 480, Blomberg Logic, 395–396, 517
482–483, 526 “Concerning the Ultimate
logical, 6, 9, 440, 451, 459n22 Ground of the Differentiation
vs. aesthetic, 451, 459n22 of Directions in Space,” 774,
moral, 196, 242, 257–258, 263, 778–780, 782–783
268, 269, 272, 279–284, Conflict of the Faculties, 45, 525,
305, 309, 311, 627, 657, 532, 644, 685, 701, 721,
794–796, 800–805, 813 725–727
Index 837
“Conjectural Beginning of 77, 79–83, 85, 88, 92–93,

Human History,” 649 99n16, 103–104, 106, 113,
Critique of Judgment/the Power 117–118, 124–125, 129–130,
of Judgment, 2, 3, 11, 134–135, 140, 165–167, 170,
42–43, 50, 57–58, 60–63, 175–177, 180, 182, 185,
66, 171, 186–187, 195, 187n25, 192–194, 195n5, 197,
387–388n6, 397, 408, 414, 202, 205–206, 208–209,
419–421, 425, 432–433, 211–212, 214, 221, 230,
438n18, 439–441, 445n29, 230–231n20, 237–238, 243,
447–449, 451, 452n8, 245–246, 254, 259, 282, 389,
454n15, 460, 462, 394, 396–397, 402, 409,
463–464n32, 475–476, 419–420, 432, 455, 474, 476,
479–480, 482, 484, 478n31, 479, 494–495, 496n8,
490–491, 519, 522n15, 498, 502, 504, 506, 515–518,
524, 528n25–26, 691, 523–528, 536, 542–543n6,
693–694, 697n22, 708, 544, 550n13, 567–570,
751, 759, 820; Analytic of 572–573, 575–577, 595, 617,
the Sublime, 449; “Critique 621, 627, 630–634, 653,
of the Aesthetic Power of 656–659, 661, 685–686, 688,
Judgment”/Critique of 691, 693, 703, 731, 734–737,
Aesthetic Judgment, 381–382, 741, 745, 747, 750–751, 754,
387, 435; Deduction of 764, 780, 783, 818, 820;
Taste/Pure Aesthetic A-edition/first edition, 71,
Judgments, 384, 425, 435; 80–81, 396, 496n9, 497n10,
First Introduction, 382, 527n24; Amphiboly of the
385, 387, 389–390, 398, Concepts of Reflection, 99n16,
452n8, 452n10, 479, 480 113–114, 394, 396; Analogies
Critique of Practical Reason, 2, of Experience, 496–497, 747,
3, 32, 33, 34n41, 42, 186, 754; First Analogy, 496,
191, 208, 214–215, 221, 506–507; Second Analogy,
227, 234, 238, 244–247, 131, 203, 208, 213, 496–497,
249–250, 252–253, 255, 499, 749; Third Analogy,
257, 260, 318, 320, 322, 496, 507; Analytic of
326, 447, 449, 460–461, Principles, 182, 207, 498,
465, 517–519, 521, 524, 507; Anticipations of
527, 528, 544, 577, 627, Perception, 496; Antinomy
632–634, 661, 689, 707, of Pure Reason, 86, 496n8,
807, 819–820 526, 734; First Antinomy,
Critique of Pure Reason, 1, 3, 5–6, 455–456; Third Antinomy,
10–11, 32, 35–37, 40, 42, 51, 86–87, 213, 221, 234, 237,
53, 56, 57, 60–63, 67, 71–72, 240–241, 742–743, 745;
838 Index
Kant, Immanuel, work (cont.) Transcendental Deduction,

Architectonic of Pure 37, 53, 57, 61, 130, 139, 142,
Reason, 689; Axioms of 144, 146, 149, 155–157, 182,
Intuition, 496; B-edition/ 207, 332, 396, 479, 507,
second edition, 3n2, 36–37, 747–748, 754 ; Transcendental
42, 77, 79, 83, 85, 142, 144, Dialectic, 60–61, 67, 71, 86,
146, 149, 155–157, 159n31, 109, 166, 167, 182, 213, 480,
165, 178, 180, 187n26, 207, 686, 738, 769, 771;
476n24, 502, 506, 734, 735; Transcendental Logic, 107,
Canon of Pure Reason, 518, 167, 168n5, 180, 182
524, 527n24; Doctrine of Dohna-Wundlacken Logic, 709
Method, 120, 135–136, “End of All Things,” 44
741n21; Ideal of Pure Groundwork of the Metaphysics of
Reason, 526; Metaphysical Morals, 7, 36, 200, 205, 206,
Deduction, 174, 177n15, 209, 214, 215, 221, 238, 242,
179–180, 182; Metaphysical 243, 260, 274–275, 276, 280,
Exposition, 79–81; 296, 308–309, 314, 316–317,
Paralogisms of Pure Reason, 318, 320, 326, 333, 337–338,
140–141, 143–144, 449, 577–578, 621, 633, 645,
159n31, 734; Third 647–649, 651, 652n11, 700,
Paralogism, 77n12; Fourth 765, 795, 797–799, 801–802,
Paralogism, 71, 734, 736; 804, 806, 807
Phenomena/Noumena, 113, Idea for a Universal History with a
526; Postulates of Empirical Cosmopolitan Aim, 685, 687,
Thought, 130, 199; Refutation 694, 697
of Idealism, 6, 37, 157, Inaugural Dissertation (On the
399–400, 476n24, 502, Form and Principles of the
734–735, 747, 749; Schematism Sensible and the Intelligible
of the Pure Concepts of the World), 61, 66, 92, 734,
Understanding, 181–182, 398, 782–783
754; Synthesis of Recognition in Jäsche Logic, 47, 127n14, 165n1,
the Concept, 397; Synthesis of 395, 490, 527
Reproduction in the Lectures on Anthropology
Imagination, 396–397; (Friedländer, Menschenkunde,
Transcendental Aesthetic, 71, Mrongovius), 67, 709, 710,
80–81n14, 85–86, 88, 94, 714
97, 102, 115–116, 166, 180, Lectures on Ethics (Collins,
182, 495n4, 504–505, 736, Herder, Mrongovius,
747, 783; Transcendental Vigilantius), 314–315, 319,
Analytic, 94, 97, 104, 320, 326, 464–466, 626, 627,
115–116, 167, 248, 494, 738; 632, 634, 709
Index 839
Lectures on Metaphysics but It Is of No Use in

(Vigilantius), 44n68 Practice,” 441n24, 589, 629,
Lectures on Pedagogy, 706–708, 683, 685n7, 689, 697, 698n23
711–713, 715, 720, 722 Opus postumum, 484–485, 544,
Metaphysical Foundations of 693, 695
Natural Science, 36, 114n52, Perpetual Peace, 536, 568, 577,
476, 478–479, 483–485, 488, 588–589, 595–596, 603, 604,
490, 493–494, 498–504, 606, 697
506–509, 653; Dynamics, 506, Physical Geography, 47, 646
508, 511; Mechanics, 506, 508; Physical Monadology, 779
Phenomenology, 506, 508; Prolegomena to Any Future
Phoronomy, 501–502, Metaphysics, 36, 72, 81–83, 88,
505–508 174, 176, 185n22, 474, 495,
Metaphysics of Morals, 36, 287n2, 505, 734, 782
345, 441n24, 449n4, 458, Religion within the Boundaries
466, 537, 556, 568, 577, of Mere Reason/Bare Reason/
588–589, 626, 645, 673, 691, Reason Alone, 44, 441n24,
707; Doctrine of Right 449, 460n24, 462n26, 519,
(Rechtslehre), 36, 577, 578, 523, 525, 527–528,
584, 588–589, 596, 620, 623, 530–532, 535, 539n1, 545,
629, 633; Doctrine of Virtue 547n10, 556, 564, 605, 644,
(Tugendlehre), 36, 269, 656–657, 659, 689n13
272–273, 275, 279, 309, “Succinct Exposition of Some
311–312, 315, 317–318, Meditations on Fire,” 29
321–324, 327, 556, 623 Thoughts on the True Estimation of
New Elucidation of the First Living Forces, 27, 29
Principles of Metaphysical Universal Natural History and
Cognition, 29, 34n41, 44 Theory of the Heavens, 29, 45,
Observations on the Feeling of the 490, 613–614
Beautiful and Sublime, 10, 33, Vienna Logic, 109, 111, 168n5,
40n58, 448, 644 396
On a Discovery whereby Any New “What Does It Mean to Orient
Critique of Pure Reason Is to Be Oneself in Thinking?”, 28n24,
Made Superfluous by an Older 417n9, 734, 736, 782, 817n2
One, 37n47, 50, 53–54, 56, What Real Progress Has
57n11, 60, 62 Metaphysics Made in Germany
Only Possible Argument in since the Time of Leibniz and
Support of a Demonstration of Wolff?, 52–53, 83–84, 817n2
the Existence of God, 44, 113 Kelsen, Hans, 601
“On the Common Saying: That Kierkegaard, Søren, 533n31,
May Be Correct in Theory, 745–746
840 Index
Kingdom/realm, defined, 370n38 400, 402–405, 455–456, 485,

of ends, 44, 362n23, 364–365, 499, 515, 522–529, 544, 568,
366n30, 413, 420, 544, 557, 570, 732, 740, 763, 765–768,
700, 810 773, 818–819
of God on earth, 541, 546, 548, vs. opinion (Meinung), 119–120,
555, 559, 561 122, 135–136, 516–520, 522,
See also Categorical imperative, 524, 527
Formula of the Realm (or phenomenology of, 753–755
Kingdom) of Ends (FRE) physiological, 646–647, 652
Kitcher, Philip, 503–504 practical, 101–103, 120, 193–194,
Kleingeld, Pauline, 603n14, 610 195n5, 202, 355, 359,
Kleist, Heinrich von, 1 370–371, 453n13, 650
Knowledge (Wissen), 1, 3–7, 35, 44, pragmatic/prudential, 650–653,
53–53, 55, 57–59, 62–66, 79, 663–664
88, 91–116, 117–124, scientific, 471–492, 499–501,
135–136, 158–159, 171, 193, 506–507, 568, 686–687n9, 768
199, 200–201, 207, 213, 238, transmission of, 720
241, 268, 298, 344–345, 398, See also Epistemic conditions; God,
413–414, 421, 453n13, 477, way of knowing; Humility,
516, 520–529, 549–550, 568, epistemic; Progress, epistemic;
570–574, 627, 633, 638, 643, Self-knowledge
645–647, 651–652, 670–675, Knutzen, Martin, 25–27, 33
686–687, 732, 735–736, 740, Korsgaard, Christine M., 2, 238, 241,
743, 749–750, 762–764, 768, 349, 357–358, 359n11, 792
773, 774–775, 785, 788, 816, Kraus, C. J., 40n59
818–819 Kripke, Saul, 113, 771, 785–786
in aesthetic judgments, 426–427, Kuehn, Manfred, 40n62, 706, 710,
429–431, 435, 438, 444–445 711, 725
a priori, 35, 53, 73, 78–79, 84, Kymlicka, Will, 612–613
101–104, 128–129, 146, 152, Kypke, J. D., 26–27, 30n31
474, 499–501, 506–507, 768,
774–775, 783–784
vs. cognition (Erkenntnis), 95, L
117–123, 136–137, 139n1, Lagrange, Joseph-Louis, 511
154–161, 523 Lambert, Johann Friedrich, 29n29,
faculty/ies of, 57–58, 60–61, 65, 54n5, 75n8
67, 139 Lavater, Johann Caspar, 45, 517n6
limits of, 1, 7, 9, 35, 61–63, 67, Law, 309, 349–350, 577, 579,
91–116, 120, 125, 161, 169, 583–584, 586–590, 594, 598,
171–172, 192–193, 197, 199, 604–605, 607–608, 619, 623,
206, 212–213, 217, 231, 234, 624–625, 629, 635, 711
237–238, 241, 254, 268, 282, constitutional, 595, 598
Index 841
cosmopolitan (see Cosmopolitanism) 431, 440, 451, 481, 503n25,

international, 589–590, 594–595, 552n18, 554, 604, 687, 733,
597–602, 614–615 734, 740, 742–746, 748–749,
See also Moral law 752–753, 762–763, 766–772,
League of nations, 589 776, 784–787
Legislative power, 360, 362–363, defined, 168n5
365–366, 462, 482, 559, modal, 503n25, 762–763, 769, 771
580, 618, 640, 798, 803 subjective, 556
Leibniz, Gottfried Wilhelm, 4, 13, transcendental, 62, 63, 167, 173–174,
29, 37, 49–68, 96–97, 99, 176–177, 179–180, 183–187,
110n46, 114, 155, 473, 508, 197–198, 200, 580
722, 734, 738, 739, 762, universal/general, 173, 197–198,
766, 770, 772–774, 779, 770, 784–786
788, 809, 815–816, 817 See also Judgment, logical;
Essays on Theodicy, 51 Possibility, logical; Science,
Meditations on Knowledge, Truth, logic as; Subject, logical;
and Ideas, 51, 64–65 Truth, logical
Monadology, 51 Louden, Robert B., 448n3, 465n34, 645
New Essays on Human Lutheranism, 22, 24, 525
Understanding, 51 See also Pietism
See also Monad/monadology
(Leibniz); Pre-established
harmony (Leibniz) M
Lessing, Gotthold Ephraim, 685n7 Machiavelli, Niccolò, 574
Education of Humanity, 685n7 Magnitude, 86, 127, 452n9, 455,
Lewis, David, 762n1, 767 474–475, 739
Libertarianism (political theory), 727 Maimon, Salomon, 738–740
Libertarianism (theory of freedom), 90, Essay on Transcendental Philosophy,
211–212, 221, 226, 235, 237, 738–739
240, 242 Manifold, see Representations,
See also Incompatibilism manifold of
Lichtenberg, Georg Christoph, 488n70 Marriage, 331–332, 334–335, 343,
Locke, John, 4, 49, 54, 60, 127, 227, 351, 537
474, 574, 580, 585, 712–713, same-sex, 334–335
716, 816 Materialism, 647, 732, 735, 738
Some Thoughts Concerning Mathematics, 4, 71–73, 79–84,
Education, 712 125–126, 129, 172, 183,
Logic, 6, 9, 50, 53–57, 63–66, 90, 93, 473–475, 478, 482n38, 484,
104, 107, 109, 110n44, 111, 487, 488, 493, 494–496,
113, 130–131, 165–187, 500–504, 505n29, 506, 702,
195–198, 204, 296, 394, 734, 738, 740, 766, 772, 774,
395–396, 415n7, 423–424, 783–786
842 Index
Mathematics (cont.) Metaphysics, 2, 6, 27, 35, 36, 49–50,

applied, 487 52–54, 57, 61, 64, 93, 115,
See also Category/ies, mathematical; 118–119, 127–128, 135–137,
Cognition (Erkenntnis), 183, 237, 358n10, 375, 471,
mathematical; Geometry; 493–494, 500, 502, 506,
Sublimity, dynamical vs. 509–511, 525, 567–568,
mathematical 570–571, 574, 588, 716, 732,
Matter, 114n52, 239, 447, 478, 734, 736, 739–740, 754–755,
484–485, 494, 499–510, 690, 761–789, 812, 816–819
692, 693, 782 of body, 494, 500, 506, 510
of experience, 6, 156, 232–233, 298, descriptive vs. “revisionary”
495n4 (Strawson), 747
Maupertuis, Pierre Louis, 29n29 dogmatic, 49, 53, 55–56, 60, 64,
Maxim, 8, 10, 85, 215, 219, 222, 238, 137, 237, 366, 550, 556,
242, 252–253, 256–258, 734–735, 738, 762–763, 768,
273–280, 288, 290–292, 773–774, 789, 818 (see also
295–296, 300, 304–305, Wolffian rationalism/
309–311, 315, 317–329, 332, metaphysics)
333–334, 337–340, 342, modal, 769–773, 778, 788–789
348–349, 362, 370n38, of morality, 36, 579–580, 811
431–432, 437, 462, 545, of nature, 36, 476–479, 484,
549–551, 564, 580, 597, 497–502, 505, 506, 510n37
604–605, 621, 622, 623, 638, Miracles, 519, 526n23, 534–537
657, 660, 663, 669n21, 692, Modality, 110n45, 112–115, 120,
796–797, 799–803 122–123, 130, 150, 408, 456,
defined, 8, 337 504, 506–507, 542–543, 558,
supreme, as Gesinnung, 529 (see also 745, 764, 816
Gesinnung [character, dualism vs. monism, 784–787
disposition, conviction, See also Logic, modal; Metaphysics,
attitude]) modal; Possibility
McLaughlin, Peter, 490–491 Monad/monadology (Leibniz), 52,
Meerbote, Ralf, 222–224, 242 56–57, 67, 779
Meier, Georg Friedrich, 31–32n36, 63, Moore, G. E., 2, 100n19, 121n10,
64, 67, 167n4, 171, 517–520 701, 809
Auszugaus der Vernunftlehre, Moral law, 4, 8–10, 33, 43, 45, 103n28,
31–32n36, 167n4, 171, 517 194–196, 202, 204–205, 209,
Mendelssohn, Moses, 36, 75n8, 215–217, 219n4, 223–225,
458n21, 491, 685n7, 734, 227–228, 236, 238, 243–260,
736, 738 266–267, 274–275, 277, 280,
Jerusalem, 685n7 282, 289–294, 301, 315, 318,
Menke, Christoph, 598 355, 360, 364–368, 370–371,
Index 843
411n5, 447–448, 454n15, 459, material vs. formal, 478, 494–499

461–462, 465n34, 467, as sum total of appearances, 43,
526n23, 527n24, 528–529, 122, 213, 474–476, 495,
545–547, 554–557, 560, 562, 687, 818
570, 576, 578–579, 614, 627, systematic order of (see System/
632, 634, 636, 648, 653–654, systematicity, in nature)
661–662, 677, 732, 796–797, See also Human nature; Metaphysics,
798–800, 806–808, 810, of nature; Naturalism; Natural
812–813, 819 purposes; State of nature;
as fact of reason, 582 Sublimity, in nature; System/
objectivity of, 195–196, 202–203, systematicity, in nature
204, 207, 246–248, 250–252, Neo-Kantianism, 2, 789
266–268, 274, 280, 283–284, Newton, Isaac, 27, 29, 33–34, 473,
287–288, 297–300, 305–306, 487n63, 493–494, 508, 510,
319, 326n21, 337n9, 355, 357, 779, 783
359, 362, 367, 368, 370, 377, Opticks, 487n63, 488
462, 545n9, 570, 701–702, Principia Mathematica, 487n63,
797–798, 811–812 488, 493–494, 508–510, 772
related to categorical imperative, 194, Newtonian science, 27, 29, 33, 43,
461 45, 225, 484, 489, 503,
See also Categorical imperative 510–511, 779
Moral worth, see Worth/value, moral Nietzsche, Friedrich, 2, 335, 759, 817
Motherby, Robert, 39–40 Twilight of the Idols, 759
Motion, 79, 85, 99, 114n52, 485n56, Non-Conceptualism, 225–226,
489, 501–502, 504–505, 762–764, 769, 773–785, 787
507–510 defined, 225, 775
Motives, 112, 545, 702, 744 See also Conceptualism
Normativity, 59, 119, 129, 133–134,
136, 199, 201, 239–240, 245,
250–251, 254, 264–265, 310,
N 311, 343, 344, 356–359,
Nagel, Thomas, 788 360n14, 362, 366–367,
Naturalism, 224, 358, 375, 489, 491, 368–377, 429, 516, 521n13,
535, 724, 762–763, 788 564, 583, 586, 588, 661n32,
Natural law tradition, 7, 602, 603, 692, 735, 739–740, 757, 791,
624–625 809
Natural purposes, 43, 50, 56, 58–60, Noumena, 35, 61, 90, 91, 98, 114,
64, 67, 195, 299, 327, 412, 134, 145, 185, 208, 211–214,
451, 481–482, 692–698, 718 218–219, 223–242, 363, 410,
See also Teleology 550, 557, 631, 657, 739,
Nature, defined, 124, 389 762–764, 766, 768–772, 774,
in general, 475–476 779, 783, 788–789
844 Index
Noumena (cont.) real and logical, 112–114

ignorance of, 91–116 Organism, 43, 225, 226, 419–420,
See also Kant, Immanuel, work: 486, 488–489, 692–694, 718,
Critique of Pure Reason, 744, 759, 786
Phenomena/Noumena; history compared to, 679, 692–695,
Phenomena-noumena 698
distinction; Thing in itself (Ding Original contract, see Contract, social/
an sich) original
Original sin, 460n24, 530–532, 555
“Ought implies can,” 8, 196n6, 215,
O 220, 292, 526, 530,
Objectivity/objective validity, 3–4, 35, 632–633
53–54, 56, 58–59, 62–64,
66–67, 72, 83, 86, 90, 94n3,
117–137, 139–140, 160, 166, P
175, 180–182, 195, 198, 201, Palmquist, Stephen, 523
227, 228, 229, 241, 244, 246n3, Kant’s Critical Religion, 523
344–345, 416n8, 417, 451n7, Parfit, Derek, 264–265
473, 475, 483, 502–503, 509, Paulsen, Friedrich, 726
520, 568, 570, 580, 671, 677, German Education, Past and
731–732, 735–738, 741, Present, 726
748–749, 752, 753, 764, 765, Perception, 4–5, 52, 55, 84, 96, 123–124,
767–770, 781, 818 126, 128, 131–132, 156, 161,
in aesthetics, 12, 410–411, 206, 385, 408n2, 732, 738, 739,
418–419, 421, 425, 427, 747, 775–776, 781
430–432, 439, 442–445, aesthetic, 425–429, 436, 442, 443
454n15 defined, 124, 128
of causality, 4–5, 7, 34–35, 211, self-perception, 157–160, 574
213, 237, 627 Perfection/imperfection, 9, 65, 66,
defined, 132 103n26, 110n46, 114,
in political philosophy, 589, 702 122–123, 133, 217, 219n4,
of religious faith, 521–522, 524, 267, 268–270, 283, 288, 309,
538, 544–545, 549–551, 342, 440, 443, 449, 451,
553–554, 556–559, 564 458–460, 461–462, 466,
See also Consciousness, objective unity 474n20, 546, 552–553, 576,
of; Freedom, objective reality of; 618, 640, 652, 657, 664,
Moral law, objectivity of 675–676, 689, 691, 698, 705,
O’Neill, Onora, 2, 359n13, 361n19, 714, 717, 721, 723, 809
363n24, 367n32, 374n49, ideal of, 553
432–433, 438n18, 442, 792, See also Cognition (Erkenntnis),
821 perfect; Duties, perfect and
Opposition, antithetical, 541–543, 549 imperfect
Index 845
Personality, predisposition to, 155, real, 112–115, 119, 123, 130–131,

305, 335, 338, 341–342, 134–135, 178–179, 181,
348–349, 447–448, 461, 187, 237–238, 242,
462n29, 553, 629, 630 543–545, 770, 774
Phenomenalism, 133, 741, 748–749 Post-compatibilism, 225–226
Phenomena-noumena distinction, 35, 57, Postulates of practical reason, 42, 44,
61–62, 91, 98, 100, 103, 214, 46, 88, 136, 181, 185,
218–219, 221, 223, 224–226, 192–193, 194, 515, 517, 519,
230–231, 235, 410, 657 527n24, 529, 535–536,
See also Appearances; Kant, Immanuel, 537n34, 538, 551, 576, 687,
work: Critique of Pure Reason, 689, 698, 732, 820
Phenomena/Noumena; Predispositons, see Animality, predis-
Noumena; Thing in itself position to; Good, predisposi-
(Ding an sich) tion to; Humanity,
Philosophisches Magazin, 37 predisposition to; Personality,
Physics, 495n6, 498–507, 511, predisposition to
695n19, 718–719n37, 755 Pre-established harmony (Leibniz), 50,
defined, 498 52, 56–58, 60–61, 67–68
Physiology, 156, 490–491, 498n13, 643, Primacy of practical reason, 191–209,
645, 652, 655n20, 659, 669n20 529, 575–576, 740
rational vs. immanent, 497–500 Principle of Contradiction/Non-
See also Anthropology, physiological; Contradiction, 53–54, 56–57,
Knowledge (Wissen), 102n23, 194, 200, 784
physiological Principle of Sufficient Reason, 53–54,
Pietism, 22–25, 102 56, 56–57n10, 102n23
See also Lutheranism Progress, aesthetic, 443
Pistorius, Hermann Andreas, 80–81, 84 epistemic, 686n9
Plaaβ, Peter, 498n15, 503–504 historical/political, 603, 606–607,
Platner, Ernst, 646–647 652, 683, 688, 691, 695–701,
Anthropology, 646–647 717, 719–721, 725–726, 820
Plato, 3, 618, 743, 758, 818n4, 821 moral, 536, 572, 707, 820
allegory of the cave, 818n4 Property rights/rightful possession, 332,
Euthyphro, 798 350, 567, 580–585, 586, 603,
Symposium, 819–820 607, 622n2, 637
theory of Forms, 809 Providence, 34, 536–538, 688, 700, 702
See also Idealism, Platonic Psychology, 498–500, 577, 666, 687,
Pollmann, Arnd, 598 743, 752
Possibility, logical, 112, 114, 115, moral, 356, 449, 742
123, 130, 178–179, 223, rational vs. empirical, 65–66, 155,
234, 237–238, 267, 277, 169n7, 498n14, 573–575,
541–545, 764–765, 644, 652–654, 659–660,
767–771, 784–787 753
846 Index
Publicity, ethical requirement of, 361, Lectures on the History of Moral

362n22, 374 Philosophy, 357
principle of, 603–606, 614, 638 See also Reasonable (Rawls);
Punishment, fear of/as deterrent, 288, Reflective equilibrium (Rawls)
292, 371–372, 579, Realism, 737–738, 741
618–620, 623–624, 626, and anti-realism, metaethical,
630, 637–638, 713 358–359, 376, 794, 798, 805,
theory of, 567, 583, 617–640 808, 810–811, 813
See also Retributivism empirical, 6–7, 734–735
Purgstall, W. J. G. von, 31–32n36 noumenal, 762, 768
Purposiveness without purpose, 12, rationalist, 60n14, 687
59–60, 407, 436 Reasonable (Rawls), 810–811
Putnam, Hilary, 2 Reath, Andrews, 238, 365, 792
Receptivity, 7, 94–97, 109, 158,
170, 209, 371, 383, 402,
Q 404–405, 445–446, 670,
Quine, W. V. O., 770–771 748–749, 754, 755
“Truth by Convention,” 770–771 Reflective equilibrium (Rawls), 280,
794–795, 805, 811
Regulative principle, see Idea(s),
R regulative
Rancière, Jacques, 445n29 Reinhold, Karl Leonhard, 37, 40–41,
Rationalism, 4–5, 24–25, 34n41, 49, 55, 42–43, 737–738, 740–742,
59–61, 66, 101, 112–113, 115, 751–753, 821
141, 155, 358–359, 363n24, Essay on a New Theory of the Human
366, 371, 442–443, 525, 535, Capacity for Representation, 737
687, 702, 750, 751, 768, 772, Essay on Transcendental Philosophy, 738
778, 787–788, 798–799, Letters on the Kantian Philosophy, 37,
815–819 751
See also Baumgarten, Alexander On the Foundation of Philosophical
Gottlieb; Descartes, René; Knowledge, 751
Leibniz, Gottfried Wilhelm; Representations, manifold of, 131–132,
Metaphysics, dogmatic; 144–146, 149, 151–154, 158,
Realism, rationalist; Spinoza, 160, 168, 182, 187, 197, 204,
Baruch; Wolff, Christian; 209, 229–230, 235, 394,
Wolffian rationalism/ 396–397, 403, 416n8, 418,
metaphysics 437, 496, 738–739, 749, 781
Rawls, John, 2, 280, 357–358, 361n17, unconscious, 127–128, 168
625n5, 635, 639, 792–798, Republicanism, 588–589, 606–607
800–813, 821 Respect, 10, 216–218, 224, 238, 270,
“Kantian Constructivism in Moral 284, 300, 301, 319, 325–326,
Theory,” 357, 792 337–338, 370, 376, 457,
Index 847
460–464, 564, 571, 620–622, Schematism, 423, 490, 554, 559

625, 629–630, 639–640, See also Kant, Immanuel, work:
653–654, 661–662, Critique of Pure Reason,
672n29, 723 Schematism of the Pure
mutual, 361, 365, 370–371, Concepts of the Understanding
372–376 Schiller, Friedrich, 449, 465–467,
See also Self-respect 751–752, 757
Retributivism, 617–640 “On Dignity and Grace,” 449
defined, 618 On the Aesthetic Education of
See also Equality, retributive Man, 751
Reusch, Christian Friedrich, 32, 40 Schmid, Carl Christian, 491
Revelation, divine, 1, 24–25, 45, 519, Schopenhauer, Arthur, 231, 743
526–527, 534–536, World as Will and Representation, 743
542–543n6, 551, 557, 559, 561 Schultz, Franz Albrecht, 22–23, 25, 26
Rink, Friedrich Theodor, 27, 706 Schultz, Johann, 37, 80
Romanticism, 409n4, 708n7 Elucidations of Professor Kant’s “Critique
Rorty, Richard, 2 of Pure Reason,” 37, 80
Ross, W. D., 809 Schütz, C. G., 37, 42–43
Rousseau, Jean-Jacques, 29n29, 33–34, Science, 34–35, 54n5, 122, 170,
580, 585, 586–587, 708n7, 172–173, 177, 193, 195,
712–713, 716 198–199, 224, 348, 356n2,
Émile, 712 357, 421, 442, 445, 568,
Social Contract, 585 569, 571, 643, 645, 651,
Russell, Bertrand, 2, 170, 772, 820n7 653–654, 686–687n9, 691,
Russell’s Paradox, 772 695n19, 718–719, 740,
748, 756, 762–763, 768,
770–771, 773, 788
defined, 477–478
S ethics as, 794, 795, 804, 806
Sabini, John, 665 in ethics, 299, 303
Salzmann, Christian Gotthilf, 715 logic as, 167–171, 173, 182
Sartre, Jean-Paul, 2, 241n36, 718 philosophy of, 471–492, 493–511,
Schelling, Friedrich Wilhelm Joseph 755
von, 2, 740–743, 745–746, practical, 120n7
758–759, 821 of reason, 165–167, 689–690,
System of Transcendental Idealism, 818
741n21–22 scholastic, 645–646
Schema, 90, 130, 132, 181, 184–186, See also Chemistry; Knowledge
201, 205, 213, 231, 235, 240, (Wissen), scientific; Physics;
398–399, 423, 553–554, 559, System/systematicity, in
675, 769, 776 science
848 Index
Self-affection, 158–160 (see also Social contract, see Contract, social/

Feeling[s], self-feeling; original
Perception, self-perception) Socrates, 709, 758
Self-consciousness, 14, 139–161, 197, Soul, doctrine of, 476, 478, 573
199, 204, 229, 230, 337–341, faculties of, 52, 290, 305, 449, 644,
387, 399, 741, 747–748, 776 654
empirical, 141, 748 immortality of, 42, 136–137, 181,
identity of, 141–144 185, 191–194, 515, 551, 689,
See also Apperception; Consciousness; 732, 766, 820 (see also
Inner Sense Postulates of practical reason)
Self-knowledge, 139–161, 251, 283, substantiality of, 147, 155, 653, 687,
673–674 818
Self-respect, 314–315, 324 Space, absolute, 508, 779, 782–783
Self-sufficiency, 316, 328, 360, 373 infinitude of, 79–80, 86
Sellars, Wilfrid, 2, 123–124, 731–732, See also Forms of sensible intuition
748–749, 821 (space and time)
Sensation, 35, 55, 109, 119, 124, Spalding, Johann Joachim, 458n21
126–129, 131, 146, 157, Betrachtung über die Bestimmung des
169–170, 198–199, 227, Menschen, 458n21
232–233, 282, 389–391, Spener, Philipp Jakob, 24
402–405, 410, 441, 443, 451, Spinoza, Baruch, 473, 734, 736,
456, 660, 665, 670n22, 671, 742–743, 750, 788
672n31, 678n43, 737–739, Spinozism, 736, 743, 762
747, 749, 752, 816 Stark, Werner, 711, 725
defined, 124, 389 State of nature, 350–351, 580–581,
Sensus communis, see Common sense 583, 589, 622, 636–638
(sensus communis) Stern, Robert, 749–750
Sider, Theodore, 761–762, 767 Stichweh, Rudolph, 487–488
Sidgwick, Henry, 794–795, 804–805, Stoicism, 464–465, 467
809 Strawson, P. F., 2, 98, 207–208, 335,
Silver, Maury, 665 731, 747–748
Skepticism, 49–50, 413–415, 431, 732, Bounds of Sense, 747
734–739, 759, 772–773, Individuals, 747
774–775 See also Metaphysics, descriptive vs.
Cartesian, 749–750 “revisionary” (Strawson)
empiricist, 49, 60, 750 Strickland, Stuart, 487
free will, 635, 639 Sturm, Thomas, 645n3, 649, 652n11,
Humean, 4–5, 36, 54, 60, 749–750, 657n26, 661n32, 711
787, 818–819 Subject, logical, 144, 146–147n12, 147,
moral, 243–260 152, 156–158, 159n31, 160
motivational, 216 Sublimity, 15, 447–467, 606
Index 849
as absolutely great, 450, 455–456, 460 in nature, 56, 413–414, 416n8, 417,
dynamical vs. mathematical, 419–421, 474, 478, 482, 491
455–456 in philosophy, 36, 42–43, 491, 573,
in nature, 448, 450–451, 455–460, 577, 605, 690, 739, 740–741,
461n25, 463–464 812, 815, 817
Substance, 52, 57–58, 67, 90, 96–97, in political theory, 570–571, 615, 652
104, 143–146, 146–147n12, rational vs. historical, 477–478
170, 180n18, 181, 185, 205, in science, 473, 475, 477, 479, 481,
416n8, 485, 496, 506–507, 483–486, 488, 491, 499, 511,
509, 573, 734, 756, 766, 654
770, 816 soteriological, 743, 745
category of, 126, 131–133, 176, See also Freedom, system of
181, 496, 506, 653, 754
unconditional, 145
See also Conservation of mass; Soul,
substantiality of T
Suicide, 271–272, 275–276, 302, 331, Talents, duty to cultivate, 269, 270,
346–347 274–276, 279, 286, 296, 305,
Sulzer, Johann Georg, 67, 75n8 337–338
Synthetic a priori judgments, 5, 71–73, Taste, judgment of, see Judgment,
79–84, 206, 239, 243–244, aesthetic
247, 255–256, 474, 505, 586, Taylor, Charles, 2, 612–613
739, 764, 765, 784–787, 816 Teleology, 43, 50, 52, 61–62, 327,
Synthetic judgments, 57, 182–183, 343–345, 348, 458, 483,
504n28, 739 560–563, 691–694, 696,
See also Analytic-synthetic 700, 718–719, 722, 741n21
distinction; Synthetic a priori See also Natural purposes
judgments Teske, J. G., 26
Synthetic vs. analytic(al) method, Tetens, Johannes Nikolaus, 67
82–83, 504n28 Thing in itself (Ding an sich), 6–7, 56,
System/systematicity, in aesthetics, 413, 71–90, 91–116, 119, 133–135,
423 157, 159, 185, 193, 206–208,
in anthropology, 645–646, 650 212–214, 218–219, 221–222,
of cognitions, 686–687 227, 230–238, 490n80, 523,
of concepts, 168–169, 175, 177n15, 568, 570, 632–633, 657–658,
179, 181–182, 415 735–737, 739–741, 744–745,
defined, 686n8, 687, 689 748, 766, 768, 783, 818
in ethics, 327–328, 370n38, 413, See also Appearances, vs. things in
791, 794, 796–797, 806, 812 themselves; Noumena
of faculties, 679 Time, see Forms of sensible intuition
in history, 687, 693, 695, 702 (space and time)
850 Index
Transcendental idealism, see Idealism, 156–159, 161, 167–168,

transcendental 169n6, 170–187, 193n4,
Transcendentals, 2–4 197–199, 212–213,
Trendelenburg, Friedrich Adolf, 80, 224–225, 232, 241, 244,
80–81n14, 84 247, 281, 284, 290, 382,
Truth, 1, 3, 7, 9, 62, 98, 102n23, 394–399, 403–405,
103–104, 105–109, 111, 414–417, 421–424,
118–123, 126–127, 435–436, 439, 452–459,
129–130, 183–185, 191, 480, 481–483, 485n54, 486,
193, 195, 197–202, 209, 496, 498n13, 502, 503n25,
236, 265, 283, 344–345, 506–507, 528n25, 549,
371, 471, 520–521, 525, 568–569, 574, 580,
526n22, 527, 537, 539–545, 673–678, 685–687, 692,
548–551, 553–564, 575, 724, 732, 737–740,
602, 746, 748–749, 764, 748–749, 753–755, 757,
767–772, 784–788, 766, 776–781, 784, 817, 820
794–795, 805–808, 812 common, 422–423
analytic, 92, 191, 785 defined, 169, 453n12
defined, 107, 121, 197, 749 See also Free play of imagination and
logical, 770–771, 787 understanding; Idea(s), of the
moral, 544, 557, 794–795, 805–806 understanding
religious, 525, 526n22, 550, 556, Unger, Peter, 764–765
559–561 Empty Ideas, 765
Universalizability, 9, 274, 276–279,
304, 317, 333, 337–339, 796
U Unsocial/unsociable sociability, 603,
Unconditioned/unconditonal, in 606, 697
ethics, 8, 249–250, Utilitarianism, 623, 636, 758, 791,
252–253, 255, 257, 792, 804
286–290, 293, 302–303,
305, 355, 367–368, 370,
462, 548, 552n18, 633 V
freedom, 196–197, 208, 662 Vilhauer, Benjamin, 232, 242
ground, 177n15, 213, 496n8, 602, Virtue, 109, 269–271, 273, 281,
686, 819 283–284, 306, 308, 310–311,
substance (see Substance, 313, 323–326, 328, 333, 335,
unconditional) 338, 348–350, 372, 449–451,
Understanding, 5–6, 9, 12, 35, 43, 458–459, 461n25, 463,
53–64, 66–67, 76–94, 465n33–34, 466–467, 544,
96n10, 103–104, 109–111, 546–548, 578, 638, 657, 677,
124–125, 127n14, 129–131, 689n13, 697, 699–702, 714,
134, 136, 139, 141, 791, 801, 810
Index 851
defined, 269 Wolff, Christian, 24–26, 29n29,

See also Highest good 34n41, 37, 44, 49–68, 114,
Vollmer, Gottfried, 47 155, 473, 767, 788, 809
Vorländer, Karl, 41n63, 727 Deutsche Logik, 65
“On the Moral Philosophy of the
Chinese,” 24–25
W Wolffian rationalism/metaphysics, 4,
Walker, Ralph C. S., 74, 230 13, 27, 34n41, 37, 49–68,
War, 589–591, 597, 600, 606, 612, 738, 762, 773–774, 815–817
684, 696, 726 Wöllner, J. C., 43–44
Watkins, Eric, 232, 506–507 Wood, Allen W., 2, 96n9, 120n6,
Will, 8, 10, 85n16, 191–194, 205, 227, 230, 523, 526,
215–218, 223, 225–226, 234, 575–576, 679n45, 685n6,
250, 255–258, 267, 289–292, 793–808, 811–813
294, 318–319, 360–361, Kantian Ethics, 793
366n30, 369, 459, 521, 523, Kant’s Moral Religion, 523
527, 564, 656, 677, 689, 691, Worth/value, absolute, 300–301, 312,
758–759, 809 315, 455, 620, 798, 808,
good (see Good will) 811–812
as Wille, 9, 85n16, 219–220, 225, moral, 282, 528, 536, 562–563, 804
236, 799 See also Dignity
See also Choice (Willkür)
Williamson, Timothy, 762n1, 777
Wittgenstein, Ludwig, 199–200,
423n12, 746, 749, 755, 773 Z
Philosophical Investigations, 746, Zedlitz, K. A. von, 43–44
749, 773 Ziche, Paul, 488
Tractatus Logico-Philosophicus, 746, Žižek, Slavoj, 2
773 Zuckert, Rachel, 666

Altman, M. C. (2017) - The Palgrave Kant Handbook PDF

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Altman, M. C. (2017) - The Palgrave Kant Handbook PDF

Caricato da

Copyright:

Formati disponibili

Palgrave Handbooks in

More information about this series at

The Palgrave Kant Handbook

Also by Matthew C. Altman

Palgrave Handbooks in German Idealism

Library of Congress Control Number: 2017947736

Printed on acid-free paper

This Palgrave Macmillan imprint is published by Springer Nature

This is the ﬁrst volume on a single philosopher in the Palgrave Handbooks in

topics and also defend their own (sometimes controversial) interpretations.

1 Introduction: Kant the Revolutionary 1

Part I Biographical and Historical Background

3 Kant and His Philosophical Context: The Reception and

Part II Metaphysics and Epistemology

4 Transcendental Idealism: What and Why? 71

5 Noumenal Ignorance: Why, for Kant, Can’t We Know

6 Kant’s Concept of Cognition and the Key to the Whole

7 Apperception, Self-Consciousness, and Self-Knowledge

Part III Logic

8 The Place of Logic within Kant’s Philosophy 165

Part IV Relation between Theoretical and Practical Reason

9 The Primacy of Practical Reason 191

10 A Practical Account of Kantian Freedom 211

11 Moral Skepticism and the Critique of Practical Reason 243

12 How a Kantian Decides What to Do 263

13 Duties to Oneself 285

14 Demandingness, Indebtedness, and Charity: Kant on

15 Kant and Sexuality 331

16 Kant in Metaethics: The Paradox of Autonomy, Solved

17 Feeling the Life of the Mind: Mere Judging, Feeling,

18 On Common Sense, Communicability, and Community 407

19 Immediate Judgment and Non-Cognitive Ideas: The

20 Sublimity and Joy: Kant on the Aesthetic Constitution

Part VII Philosophy of Science

21 “Proper Science” and Empirical Laws 471

22 From General to Special Metaphysics of Nature 493

Part VIII Philosophy of Religion

23 Kant on Faith: Religious Assent and the Limits to

24 The Fate of Religion within the Boundaries of Mere Reason 539

Part IX Political Philosophy

25 The Critical Legal and Political Philosophy of Immanuel

26 A Cosmopolitan Law Created by Cosmopolitan Citizens:

27 Kant’s Mature Theory of Punishment, and a First Critique

Part X Anthropology, History, and Education

28 Denkungsart in Kant’s Anthropology from a Pragmatic

29 Kant on Emotions, Feelings, and Aﬀectivity 665

30 The Philosopher as Legislator: Kant on History 683

31 Becoming Human: Kant’s Philosophy of Education and

Part XI The Kantian Aftermath, and Kant’s Contemporary Relevance

32 Kant after Kant: The Indispensable Philosopher 731

33 Kant, the Copernican Devolution, and Real Metaphysics 761

34 Contemporary Kantian Moral Philosophy 791

35 Conclusion: Kant the Philosopher 815

Matthew C. Altman is Professor of Philosophy and Chair of the Philosophy

Carla Bagnoli is Professor of Philosophy at the University of Modena, and

Andrew Chignell is Professor of Philosophy at the University of

Katerina Deligiorgi is Reader in Philosophy at the University of Sussex. She

Eli Friedlander is Professor of Philosophy at Tel Aviv University. Among his

Patrick R. Frierson is Associate Professor of Philosophy and Garrett Fellow

Paul Guyer is the Jonathan Nelson Professor of Humanities and Philosophy at

Robert Hanna is an independent philosopher and co-director of the

Fiona Hughes is Senior Lecturer in Philosophy at the University of Essex.

Chong-Fuk Lau is Professor of Philosophy at The Chinese University of