Almog, J. - What Am I

W H A T A M I?
Descartes and the Mind-Body Problem
Joseph Almog
OXFORD
U N I V E R S I T Y PRESS
Oxford University Press, Inc., publishes works that further
Oxford University's objective of excellence
in research, scholarship, and education.
Oxford New Y o r k
Auckland Cape T o w n D a r es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Copyright © 20o2 by Joseph Almog
First published in 2002 by Oxford University Press, Inc.

198 Madison Avenue, New Y o r k , New Y o r k 10016
First issued as an Oxford University Press paperback, 2005
www.oup.com
Oxford is a registered trademark of Oxford University Press
A l l rights reserved. N 0 part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data

Almog, Joseph.
What am I? : Descartes and the mind-body problem / Joseph Almog.
p. cm.
Includes index.
ISBN-13 978-0-19-514646-2; 978-0-19-517719-0 (pbk.)
I S B N 0-19-514646-8; 0-19-517719-3 (pbk.)
ι. Descartes, Rene, 1596-1650—Contributions in dualist doctrine of
mind and body. 2. Mind and b o d y — H i s t o r y — 1 7 t h century. I . Title
B1878.M55 A46 2001
i28'.2—dc21 00-067094
ι 3 5 7 9 8 6 4 2
Printed in the United States of America

on acid-free paper
Dedicated to my parents, Lea and Joshua
Preface
The title question is vintage Descartes. I t makes its debut in

his Meditations on First Philosophy, on the second page of Medi-
tation I I . I refer to it from now on as "the primal question."
The text that follows is an attempt to trace Descartes's own
answer(s) to this question. 1
This book has been reconstructed, with minor alterations

(footnotes, references, etc.), from a transcript made in a U C L A
introductory class, locally known as Philosophy 7, which I have
taught every year since coming to U C L A in 1984. The class is
given to a large audience of freshmen and sophomores, up to
350 students. Most of them take the class as a "general educa-
tion" requirement, and it is thus their first, and probably last,
exposure to philosophy.
The idea to teach Descartes came to me by accident. I n my
first year o f teaching at U C L A , I had to find some large-audi-
ence class to teach. I could not teach ethics or history introduc-
tions. By elimination, I was left with Philosophy 7, then a name
without any connotation for me. One day I visited the class to
get a sense of what it is like to teach it. The teacher, Rogers
Albritton, was reading through Meditation I I . I was spellbound
ι. In what follows, I have relied on The Philosophical Writings of Des-

cartes, trans. J. Cottingham, R. Stoothoff, and D. Murdoch (Cambridge:
Cambridge University Press, 1985), vols. I and I I . I refer to the two vol-
umes as CSM I and CSM I I . I also refer to Descartes—PhilosophicalLetters,
ed. A. Kenny (Minneapolis: University of Minnesota Press, 1970). It is des-
ignated in what follows as PL.
viii W H A T A M I?
and decided to imitate him, which was not very sage on my

part, for Albritton is inimitable. But I have found Descartes.
In the first io years, the class consisted only of chapter i of
this book—the proof of the mind-body real distinction. I took
myself to be following Descartes's own methodology in Medi-
tation I I . Having raised the primal question, he answers it in
the next sentence with the phrase "a man." But he goes on to
dismiss this primal answer in the following way:
But what is a man? Shall I say 'rational animal'? No; for

then I should inquire what an animal is, what rationality
is, and in this way one question would lead me down
the slope to harder ones and I do not now have the
time to waste on subtleties of this kind. (CSM I I , 17)
He soon substitutes the primal question by surrogates—What

is Mind? What is Body? I t is thus that mind and body "grow
capitals." The project as Descartes characterizes it technically
is to prove their real distinction. The task is to prove that the
mind and the body—any old mind and body and never mind
whose (A man's mind? This man's mind?)—are (i) real subjects
that (ii) are numerically distinct and (iii) can exist without the
other. The point of the class for many years was to examine
Descartes's proof. After 10 weeks, we would reach a happy
end—Descartes's proof vindicated.
The happy end did not tell the full story. I felt that without
going back to the primal question, we did not come full circle.
Stronger yet, as long as we did not consider Descartes's origi-
nal, quickly dismissed answer (i.e., what-I-am is a man, a hu-
man being), we would not get a full understanding of Des-
cartes. Most discussions of Descartes and of his progeny—the
doctrine of mind-body dualism—started with Mind and Body
as the basic notions. They would add as a sort of appendix that
somehow, out of these two primal elements, Descartes "com-
poses" (sometimes, "constructs") a full-blooded man.
I knew in my bones that Descartes was not involved in such
metaphysical engineering. This was a man who, in the dead of
PREFACE ix
night, bought cadavers in the "street of veal" so that their dis-

section would teach him about human nature. This was " m y "
Descartes: a dissector of human beings, in whose nature it is to
be first alive and then to turn into such lifeless bodies. But to
so ground Descartes in the human condition—the life and death
of the human body—ran against much o f the atmosphere that
radiated from Cartesian studies. I had to keep Descartes out of
my sight for some time.
The class, however, had to go on. A n d so, for a few years
I introduced students to the question "What am lP"(or even
better: "What makes someone one-of-us?"), using quite a dif-
ferent text—the movie Blade Runner, which has its own way
of making the question vivid for beginners. But I knew my
estrangement from Descartes was only temporary.
It was only around 1996, after teaching Descartes anew in a
class in Sweden, that I came to see how he worked his way
back to the primal question. I have convinced myself that the
key to his methodology lies in his remarks to Princess Eliza-
beth, especially the last sentence of the following quotation:
I supposed that your Highness still had in mind the argu-

ments proving the distinction between the soul and the
body, and I did not want to ask her to put them away
in order to represent herself the notion of the union
which everyone has in himself without philosophizing.
Everyone feels that he is a single person with both body
and thought so related by nature that the thought can
move the body and feel the things which happen to it.
(PL, 142)
Driven by these remarks, I have redesigned the class in re-

cent years. It expounds Descartes as pursuing the dual key proj-
ect: a real distinction in a real man. The challenge to the class
in the final exam—a challenge reviewed in the Synopsis, which
follows this preface—is to evaluate whether the real distinction
argument has been successfully embedded in a framework that
respects the integrity of the man as a real subject.
χ W H A T A M I?
Running this class for more than 15 years led me to accumu-

late many debts. I am grateful to the philosophy department of
U C L A and to the university as a whole for supporting such an
"exotic" general education course. I n public education, where
there is a w i l l , there is a way.
I am grateful to many past and present colleagues for discus-
sions of this class and related metaphysical issues: Tyler Burge,
Keith Donnellan, Kit Fine, BarbaraHerman, Tony Martin, the
late Warren Quinn, Abe Roth, Seana Shiffrin, Sean Kelsey,
and Mike Thau. My fellow historians John Carriero and Calvin
Normore have given this book a most thorough reading despite
the fact that its author must have seemed to them an accidental
tourist in the land of Descartes studies. I have learned much
about the mind-body union from interacting in Vancouver with
Catherine Wilson and in Uppsala with L i l l i Alanen. This class
also owes much to extremely dedicated teaching assistants who
struggled in creative ways to make the material sensible to stu-
dents: Andrew Hsu, Michael Thompson, Torin Alter, Keith
Kaiser, Erin Eaker, and Dominik Sklenar. Finally, throughout
the process, I found Peter Ohlin of Oxford University Press a
deft and supportive editor.
The greatest influence over my metaphysical formation, i f
anything could be so called, has come from three philosophers:
the inspiring pedagogue Rogers Albritton; Saul Kripke in his
book Naming and Necessity, a sort of modern-day Meditations,
whose luminous language is responsible for the way we do
metaphysics (and that, in this positivist age, we do it at all);
finally, there are the daily interactions, for two decades, with
my teacher, David Kaplan.
The last word before the reader confronts the text proper
regards the fundamental limits of the present book. Though I
enjoyed the years of teaching Descartes and the mind-body
problem, I feel something is amiss with my work on both.
Regarding Descartes, my readings of his work are far from
"complete." For one example, the leading interpretation I offer
in chapters 2 and 3, what I call integrative dualism, ignores
PREFACE xi
Descartes's seeming promise to leave room for the immortality

of the soul. For another such example, many of my historian
friends have pushed me to trace the thread between Descartes
and his medieval predecessors. Some (not the same) have ar-
gued I should connect him with his rationalist successors, e.g.,
Kant. Now, at times, Descartes does use the metaphysical lan-
guage of the medievals. Also, Descartes does have moods in
which he speaks in a very intellectualized, abstract way about
our ideas and our reason. But what always struck me is how
perceptual, indeed sensuous, a philosopher he is. His explana-
tions o f the workings of human beings are strikingly graphic,
be they of the circulatory system and the pineal gland or be
they, as in his Passions of the Soul, of emotional states such as
love and admiration. Throughout, I view him like a gardener
touching his beloved flowers delicately, as he sketches them on
a piece of paper. W i t h this picture in mind, I have elected to
keep Descartes a philosopher-by-nose rather than a late medi-
evalist or an early rationalist. I am sure this makes the Des-
cartes that follows substantially incomplete.
What applies to the Descartes that follows applies even more
so to what I say about the mind-body problem in general. By
this last, I refer to a casting o f the problem in modern (twenty-
first century) terms, using concepts from the philosophy of lan-
guage, (modal) logic and other such "technology." The incom-
pleteness runs deeper here. There is on the one hand a sense
of local progress on this or that "technical issue." But it is
coupled with a deeper feeling of not getting to the bottom of
the real problem. I track a like-minded recognition of this kind
of "technical sublimation" in the writings of T o m Nagel. W i t h -
out drafting his writings to support the present discomfort, let
me explain my own reservations.
By "local progress" on technical issues, I mean something
like this. I n research lectures (though not in the aforementioned
classes), I project some ideas from what I call Descartes's inte-
grative dualism to the modern version of the dispute between
materialists and neo-Cartesian dualists, of whom I take Saul
xii W H A T A M I?
Kripke to be the leading expounder. I t is thus that I have come

to think, perhaps against Kripke, that we need not assert the
real possibility of mentality (say, my state of pain) without a
brain state, or the possibility of a brain state (say, the firing of
such and such brain fibers) without the corresponding mental
state of pain. I , for one, would see the two—the physical and
mental (type of) states—as bound by modal necessity; stronger
yet, I see them as interconnected by their very nature. Related
to this is the sense I express in chapter 3 that there is no ground
under the various brain-body-swap scenarios fantasized about
in the contemporary personal identity literature. Nor do I see
critics of materialism (like myself) needed to accept the possi-
bility of so-called zombies (beings with our kind of physical
brain structure but no mental experiences). A l l of these imag-
ined possibilities are illusory and not required in a defense o f
(i) the numerical distinctness o f the mind (and its states) from
the body (and its states) and o f (ii) what I describe in the book
as the categorical difference o f matters physical (or physical-
functional) from any thing mental.
So much I feel I can argue for. Nonetheless, there remains
the abovementioned malaise. I do not think that such local i m -
provements have provided me with a clear language with which
to teach what is the mind-body problem. The substance lan-
guage of Descartes—do we have here one or two things (sub-
jects, substances)?—is one philosophical language we all use to
talk about the problem; the modern reference to mental and
physical properties (or for that matter, kinds of states) is another
such language. I believe that within these languages, we can
make useful moves, ascertain what premises need to be asserted
by a given position, what items are identical with what, and
what depends on what. A l l of this leaves the discomfort where
it was, the feeling that these "technical" philosophical languages
do not get to the bottom of the problem. I often say to the
students: many complain that when it comes to the mind-body
problem, we have a fundamental question and no clear answers;
PREFACE xiii
for me, it is the other way round: we have many clear answers
but no fundamental question.
Perhaps we are overly wedded to the standard philosophical
languages. A t one point in his correspondence with Princess
Elizabeth, Descartes, after he himself so ingeniously used the
standard philosophical languages of his time, says to her that
to understand the mind-body union, one needs to bracket away
metaphysical analysis and fall back on "ordinary life and con-
versation." Well, I tried. Such is the aforementioned graphic
language o f the movie Blade Runner. Recently, I have so de-
ployed ih teaching the language of an imaginative novel, The
Golden Compass Qjy Philip Pullman). The author there connects
in a most beguiling way each person—seemingly a kind-mate
of you and me—with what he calls his (her) daemon. Such
graphic aids are invaluable but one still feels the problem has
not been articulated.
The philosopher by nose he was, I do think that Descartes
smelled things right—the insight that is needed will come not
from, but in spite of, technical philosophical languages, by at-
tending to our "ordinary life and conversation." A t this time, I
don't know how to do it.
Saint Jean Du Gard, France

July 2000
Contents
Synopsis: The Project and the Challenge, xvii
ι. The Real Distinction, 3

1.1. The Basis for the Real Distinction, 3
1.2. What Is in a Real Distinction?, 4
1.3. Conceivability, Possibility, Whatness, 5
1.4. The Argument from Possibility—Descartes vs.
Arnauld, 14
1.5. The Primacy of Whatness, 23
1.6. Appendix: Conceivability—Does Epistemology Precede
Metaphysics?, 42
2. Integrative Dualism, 59
2.1. The Prove Too Much/Prove Too Little Dilemma, 59
2.2. Separatist Dualism: Four Basic Annotations, 61
2.3. Integrative Dualism: The Quest for Symmetry, 71
2.4. Subjects in Time: T w o Cartesian Frameworks, 81
3. The Real Man, 99

3.1. The Real Distinction Reviewed, 99
3.2. The Primal Question and the Primal Answer, 115
Index, 125
Synopsis: The Project and the Challenge
Descartes's dual key project targets two objectives. The first is

the separation of the human mind and body, the second their
integration into a human being. The demands—separation yet
integration—seem to make the project impossible. As often,
Descartes himself best articulates the project and the challenge
facing it. I n his fourth replies, his response to Arnauld, he for-
mulates his project (the dual key project) and the challenge—
the prove too much/prove too little dilemma:
Nor do I see why this argument "proves too much."
For the fact that one thing can be separated from an-
other by the power of God is the very least that can be
asserted in order to establish that there is a real distinc-
tion between the two. Also, I thought I was very careful
to guard against anyone inferring from this that man
was simply "a soul which makes use o f a body." . . .
Now someone who says that a man's arm is a substance
that's really distinct from the rest of his body does not
thereby deny that the arm belongs to the nature of the
whole man . . . nor do think I proved too little in saying
that the mind is substantially united with the body, since
that substantial union does not prevent our having a
clear and distinct concept of the mind on its own, as a
complete thing. (CSM I I , 160)
Let me put the dilemma in my own words. We might be so

successful at separating mind and body that when we put them
together again to form a single man, we no longer get a real,
natural unity—a full-blooded human subject—but what Des-
xvii
xviii W H A T A M I?
cartes calls a mere "unity of composition/' an artificial com-

pound. Thus, we have the "prove too much" horn. T o amend
things, we might start by carefully protecting the primality of
the full human being. But then whatever separation-in-our-con-
ception we grant mind and body, it is not enough to establish
them, our conception aside, as distinct, complete subjects, each
of which can exist without the other. Thus we have the "prove
too little" horn of the dilemma. 1
We shall follow Descartes's two-stage defusing of the prove

too much/prove too little dilemma. Chapter i will discuss his
argument for the real distinction, and chapter 2 will expound
his way of integrating mind and body into a single real man.
By this point, we will evaluate how Descartes fares vis-ä-vis
the dilemma.
I do not provide such a resolution. I doubt, in general, the
feasibility of quests for "the one and only interpretation," even
more with so subtle a thinker as Descartes, ever the master of
nuances, depending on who his interlocutor is. More instructive
in my view is to recognize that various forms of dualism may
be distilled from Descartes's texts. I t is then interesting to sepa-
rate the questions for each framework and see how, within a
given framework, an outstanding puzzle is worked out.
To reiterate: I n the following text I have no ambition to
uproot other interpretations in order to leave room for this one
orchid—the "real" Descartes. I would rather let a couple of
such flowers—and more—bloom.
ι. I should say that in teaching this critical dilemma-setting paragraph, I

often suggest replacing, in the example, the arm by the brain. There is then
a much better intuitive sense to the claim that though both items are com-
plete subjects, they existentially depend—by their very nature—on each
other.
What Am I?
ONE
The Real Distinction
i . i . The Basis for the Real Distinction
I n a letter written on January 19, 1642, to Father Gibieuf, Des-

cartes says:
The idea of a substance with its extension and shape is a

complete idea because I can conceive it alone and deny
of it everything else of which I have an idea. Now it
seems to me very clear that the idea of a thinking sub-
stance is complete i n this sense, and that I have in mind
no other idea which is prior to it and joined to it in
such a way that I cannot think of the two together while
denying the one o f the other. (PL, 24)
A similar observation about the completeness o f the idea o f his

mind recurs in Descartes's reply to what strikes me as the most
penetrating objection made to him, Arnauld's (the fourth) ob-
jection. I n what follows, I try my hand at a defense of Des-
cartes's proof by reviewing his back-and-forth exchange with
Arnauld. My sense is that i f Descartes can work his way out of
Arnauld's objection, he is on the homestretch. 1
ι. The readings I take to be relevant to this issue and which are re-
ferred to in detail below are the exchange with Arnauld over "complete
ideas" in the fourth replies (CSM I I , 156-160), related remarks to Caterus
in the first replies (CSM I I , 85-86), and the letters to Mesland (May 2,
1644, PL, 152) and Gibieuf (PL, 24).
3
4 W H A T A M I?
i.2. What Is i n a Real Distinction?
When I speak of "the real distinction o f mind and body/' we

must keep in mind two simple observations. The first concerns
the nature of the intended relation-real distinction; the second,
the nature of the purported relata.
Concerning the first distinction and following Descartes, I
will gloss "x and y are really distinct" as "x and y are sub-
stances and χ andjy can exist without each other." I n this gloss,
"without each other" will be taken "weakly"; that is, it is
enough that at least one of x and y can exist without the other.
What is more, we need to be intentionally primitive about the
"can" in "can exist without the other." Today many readings
offer very sophisticated senses of this "can." Reference is made
to "epistemic," "logical," and "metaphysical" brands o f possi-
bility, as i f such technical notions were as old as the hills; the
language of "possible worlds" is invoked, and it is said that
"there is some possible world where the mind exists without
the body." More generally, much o f the sophistication from
modern studies of modal logic and its model theory (semantics)
shows in analyses of Descartes.
I want us to regain our innocence. Descartes's "can" is just
the "can" of the vernacular. Whether it means an "epistemic
can," a "modal can," or yet a third, hitherto undelineated "can"
is an open question. We should not prejudge the force of Des-
cartes's "can" by infusing sophisticated modern readings.
Second, I now come to the nature of the relata involved. I n
many discussions of Descartes's real distinction proof—his own
Meditation V I discussion is a paradigm in this respect (CSM I ,
56)—there is a gap in proving that (i) his mind and body are
really distinct and (ii) he and his body (in his mouth: I and my
body) are really distinct. Our question in chapter 1 is this: Are
Descartes's mind ( D M ) and Descartes's body (DB) really dis-
tinct? The question about the man Descartes or, more neu-
trally, the referent of his " I " - w h e t h e r it (he) is distinct from
DB—is left for chapter 2.
THE REAL D I S T I N C T I O N 5
1.3. Conceivability, Possibility, Whatness
Existentially Separatist Arguments
There are at least three arguments of Descartes that we shall

be looking at—the argument from conceivability, the argument
from possibility, and the argument from whatness. The three share
a certain form. Each provides (i) a numerical discernibility argu-
ment and (ii) an existential discernibility argument. By (i) I
mean that the argument isolates some genuine property of indi-
viduals F—the discerning property—and states: The mind bears
F but the body does not. Using logic, the argument applies
Leibniz's principle of distinctness of discernibles: Individuals dis-
cernible by any genuine property F are numerically distinct
individuals. I t is concluded that the mind is numerically distinct
from the body; they make two distinct subjects.
In addition, all three arguments have a particular way—the
way of existential separation—of ascertaining the numerical dis-
tinction. Each asserts that the mind can exist without the body.
This last claim is found by analyzing the preceding premises:
The mind bears property F but the body does not. Shortly
we shall discuss how exactly one "gets" existential separability
from the discerning property F. Sometimes F wears the existen-
tial separability on its sleeves; sometimes more analysis is called
for to isolate it. But for the moment let us call arguments that
secure both numerical and existential discernibility existentially
separatist arguments.
In a slightly more canonical form, our three arguments are
as follows:
(1) D M bears F.
(2) I t is not the case that D B bears F.
Now, by the distinctness of discernibles:
(3) D M Ψ DB.
6 W H A T A M I?
It is then added that given the meaning o f the discerning feature

F, premise ( i ) gives us further:
(4) D M can exist without D B .
It is time to verify how this basic mold shows up i n the

three Cartesian arguments.
The Argumentfrom Conceivability
In the argument from conceivability, the discerning property F

is as follows: "Conceivably exists without D B " (or stronger
yet, "without any body whatsoever"). We get
(iC) D M conceivably exists without D B .
(2C) I t is not the case that D B conceivably exists without

DB.
Hence,
(3C) D M Ф DB.
This takes care o f the numerical distinction. But where are

we to get the real distinction o f D M and DB? I t is suggested
that the truth o f ( i C ) provides a locus—a scenario i n our
head—in which D M can exist without D B . W e may thus assert,
" i n a conceived story" sense o f "can," that
(4C) D M can exist without D B .
The Argumentfrom Possibility
In the argument from possibility, the discerning property is

"possibly exists without D B , " that is, has the real possibility o f
existing without D B . W e then get
(iP) D M possibly exists without D B .
(2P) I t is not the case that D B possibly exists without D B .
Hence,
(3P) D M Ф DB.
It is now argued that ( i P ) directly gives us
(4P) D M can exist without D B .
The Argumentfrom Whatness
In the argument from whatness, the target property F relies on

the notion of the kind of thing D M is, that is, the nature of
D M or what-DM-is. I use this vocabulary to formulate F as "by
whatness thinking":
(iW) D M is by whatness thinking.
(2W) I t is not the case that D B is by whatness thinking.
(3W) D M Ф DB.
The premises of the whatness argument are much more pru

dent. Indeed, its very point is to proceed from a denial of the
preceding possibility argument: Though D M and D B may of
necessity coexist (in union), they are nonetheless ascertainable
as numerically distinct. So unlike the first two arguments, the
whatness argument does not assert directly the existential sepa
rability of D M and D B . There thus remains the question o f
whether D M and D B are not only numerically but also really
distinct, that is, existentially separable. Establishing such exis
tential separation will not be as easy as it was before; this is,
after all, the virtue of the present argument. But those accus-
8 W H A T A M I?
tomed to witnessing a numerical distinction by existential sepa-

rability will insist on posing the following question: Is there no
way to save, as Descartes wanted, a logically objective sense of
"can" according to which D M can exist without DB? The ques-
tion calls for some careful discussion.
THE QUEST FOR WHATNESS AND

COHERENT STORIES
We are walking a tightrope here. The logically objective sense

of "can" that we are looking for must be weaker than a real
possibility yet stronger than a merely "epistemic can": "For all
the information available to D M that is immuned to Cartesian
doubt, D M may exist without D B . " T o get such a weaker than
real possibility yet objective existential separation, we may ar-
gue as follows.
The claim that what-x-is is an F is equivalent to this proposi-
tion: It is not logically consistent with what x is that x is not
F. We may now add this: T o every such consistency claim,
there is a corresponding "coherent story" (model) about x. The
intended idea is that there may be consistent stories about x
that nonetheless are not genuine possibilities for x. Our quest
is for a story that is logically consistent with what D M is but
one that is subtracting D B .
Consider all the coherent stories about D M . Either DB is pres-
ent in them all or it is not. I f it is not present in at least one of
them, we are done—DM can exist without DB. But what if, story
by story, we find that whenever D M is present, DB is?
A t this juncture, readers of Descartes's Meditation I I may
well be thinking of a variety of growingly strong premises, all
of which I call subtraction principles. The weakest is this:
(Si) Let the specification of what-#-is, for example, an F,

not involve some particular subject y; also, what-j-is is
not an F; then, there exists a coherent story where x ex-

ists but y does not.
Some such premise might rid D M of the specific body D B it is

in union with, while still demanding that every coherent story
about D M makes it embodied. This kind of argument is very
popular with theories of personal identity that come from John
Locke and are embraced by many in our time. We shall have
the occasion to dwell on such theories later. Such (neo-)Lock-
ean theories sustain the sense of a story or "case" about D M
without D B .
A principle stronger than (Si) and deniable by the neo-Lock-
ean tradition seems extractible from Descartes:
(52) Let the specification o f what-x-is, for example, an F,

not involve any object of another kind (whatness) G;
then, there exists a coherent story where x exists but
no G objects do.
According to this principle, there exists a coherent story with

a totally disembodied D M . Both (Si) and (S2) assume that no
reference to extendedness or materiality has been made in speci-
fying what D M is ("a thinking thing").
Finally, some read Meditation I I as providing a scenario in
which Descartes's mind is all by itself. They point to the coher-
ence o f the supposition that an evil genius has been feeding
Descartes's mind illusions about his body, other men, and the
rest of the external world. The coherence of such a story would
provide the basis of yet a stronger principle, a superposition of
the foregoing two:
(53) Let the specification o f what-x-is, for example, an F,

not involve any specific objectjy and any other kind
(whatness) o f objects G; then there exists a coherent
story where x exists but (i) no G objects exist and (ii)
y does not exist.
io W H A T A M I?
According to (S3), there is a coherent story about D M in which

nothing but it (and the creator of all things) exists, not even
other soul mates, that is, other thinking things.
To reiterate, none of the above (Si)-(S3) arguments may
represent a genuine possibility for D M . But, by the lights of
Meditation I I , each submits a coherent story, one witnessing
what strictly belongs in DM's nature; what does not belong
may be coherently subtracted. Such subtractive stories, weak or
strong, justify our drawing from the whatness argument the
further result of a real distinction:
(4W) D M can exists without D B .
Descartes's Projection Method
The arguments from conceivability, possibility, and whatness

seem autonomous, and we can consider the merits of each sepa-
rately. A modern eye, trained in the metaphysics of possibility
and essence (whatness), will certainly take the arguments from
possibility and whatness as standing on their own. Thus, many
modern metaphysicians have been happy to assert that (i) Rich-
ard Nixon might have lost the 1968 elections; (ii) it is necessary,
though not pertaining to what he is, that Nixon was the actual
winner in 1968; and (iii) his being human does pertain to what
he is. These truths have been standardly submitted by modern
metaphysicians without an accompanying theory of how we
know them to be true. I am surely one such "modern."
However, in the context of Descartes's arguments, it is clear,
most of the time, that this is not the order of ideas (chapter 2
will stress the few times when this is not so clear). I n a natural
first reading of Descartes, the argument from conceivability is
the fundamental one, serving as the basis for the remaining
two. Driven by this fact about Descartes's affinities, we focus
here on the possible justification of his conceivability argument.
As mentioned, this goes against my metaphysical grain. None-
theless, I feel that the focus on conceivability has its benefits,
THE REAL D I S T I N C T I O N II
in addition to the already mentioned virtue of staying close

to Descartes. We are thus forced to ponder this intriguing
question: What makes conceivability such a metaphysically (i)
sound and (ii) complete tracking method? 2
As emphasized, Descartes's starting point is his argument

from conceivability. The two remaining arguments are projec-
tions from conceivability to something more objective, some-
thing that is meant to be conceivability free. I n the first projec-
tion, we are finding a real possibility for D M . I n the second,
we are finding what is logically consistent with what D M is,
what is not excluded by its objective essence. Either way, our
conceivability experiments have been left behind. We now have
the objectual targets proper—DM and D B . What discerns them
is something in their own objective (i) potentialities and/or (ii)
essence (the kind of thing they are).
Common to both projections is the following two-step mold,
what I'd like to call Descartes's basic projection mold. Step i
starts with some direct data, and so I call it data articulation.
We consider what we can conceive of D M and of DB. The
results are available to us—to our intellect—directly; further-
more, the results are known to us with supreme certainty.
Given this data articulation, we are ready to move on to step
2, the projection step. Sometimes, our objective is the projec-
tion of a real possibility of D M without D B . We rely then on
the following scheme:
(Real possibility projection) Whatever is (clearly, distinctly

and completely) conceivable about a given subject x is re-
ally possible for χ.
2. The notions of a sound and complete tracking method will be intro-

duced in detail later; see i.6. At this stage, it is enough to say that by the
metaphysical soundness of the method, I mean that what is conceivable
about χ is indeed possible for x (consistent with its nature). By complete-
ness, I mean the converse: I f something is possible for x (consistent with
x's nature), then it is conceivable about it; that is, conceivability about x
completely exhausts what is possible for x (consistent with its nature).
i2 W H A T A M I?
Alternatively, our target may be to establish the weaker claim,

that whether it is really possible or not, it is logically consistent
with DM's essence—being a thinking thing—that it exists with-
out D B , indeed without any extended thing around. T o so pro-
ject, we call on this scheme:
(Coherent story projection) Whatever is (clearly, distinctly

and completely) conceivable about a subject x is logically
consistent with what it is and attributable to x in a coher-
ent story (model) about it.
Given the soundness of such projections, we are given a real

possibility, or a coherent story, in which (i) D M exists and D B
does not. I t follows that (ii) in that possibility (coherent story)
D M is distinct from D B . But then it follows that (iii), in reality,
D M is numerically distinct from D B . 3
However, by this very argument, D M is not only distinct

in reality from DB but also "really distinct"—in the technical
sense—from it: I t can exist without it. The "can" may mean
either " i n a real possibility" or merely " i n a coherent story,"
depending on which argument has been followed.
3. This last step, from distinctness in a possibility (coherent story) to

distinctness in reality, is due to the logical stability of identity and distinct-
ness relations: I f x is (not) identical to y, then in all possibilities (coherent
stories) χ is (not) identical to y. Many in the first half of the twentieth cen-
tury, doubted these principles, because they confused the coincidence of
two individual concepts (presentations of individuals in our thinking) with
the relation of objectual identity between the individuals proper. Others
have doubted the principles because they confused the objectual relation
with its linguistic representation: a = b sentences with proper names a and
b. In contrast, Descartes and Arnauld—very much like modern (modal) lo-
gicians—were crystal clear about the logic of identity and assumed the sta-
bility principles as a matter of course. They did not confuse the relation
proper with how we present it to ourselves, in thought or in language.
With Descartes and Arnauld, we assume the logical principles throughout
this work. The emphasis on the distinction between representations of iden-
tity and the objectual relation will concern us later in this chapter, in the ex-
tended discussion of conceivability, de dicto, and de re.
THE REAL D I S T I N C T I O N ij
In reconstructing Descartes's order of argument, we first

have to take notice of the primacy of the conceivability argu-
ment. Descartes abides by a distinctive flow diagram:
Conceivability > - possibility (coherent story)

> - reality
Second, in contrast, the most ubiquitous projective argument

considered by critics of Descartes—for example, Arnauld in the
fourth objections—is the possibility argument. The objections
are formulated against it. I t is only in the subsequent Cartesian
reply that the attenuated (weaker and more sophisticated) pro-
jective argument from whatness and coherent separation
emerges. I n this sense, the possibility argument is the "first
blush" projection argument, and the whatness and coherent
separation argument is a subsequent correction; that is, when it
is realized, we can obtain our desired real distinction from
weaker premises.
In the reconstruction below, I w i l l follow the dynamics of
the Descartes-Arnauld exchanges. We shall examine first the
possibility argument, look at Arnauld's objection to i t , and then
move to Descartes's repartee. This w i l l lead us to Descartes's
argument from whatness and coherent separation. 4
4. It may seem that we have ignored, on behalf of Descartes, an obvi-

ous third path. Why not use the argument from conceivability without any
projection, either of possibility or coherent separation? After all, i f D M has
a property—conceivable without DB—that DB lacks, by Leibniz's law (dis-
tinctness of discernibles), it seems that the game is over, that D M and DB
are numerically distinct. And it may be claimed that we have here a real
distinction—we can conceive of D M without DB. Thus, in this sense of
"can"—can in the understanding—DM can exist without DB. What, i f any-
thing, is wrong with such a direct, projection-free argument? Perhaps noth-
ing. The fact remains that Descartes (and Arnauld) fixed on projection ar-
guments. I , a reconstructor of both, will follow them. By the time we have
distilled the most defensible of the projection arguments—the end of chap-
ter ι—I shall return to the theme of an altogether projection-free argument
and ask (i) has it been followed by Descartes and (ii) could it justify the
real distinction?
i4 W H A T A M I?
1.4. The Argument from Possibility—

Descartes vs. Arnauld
We need to fix on the sense of possibility intended in the argu-

ment. The possibility targeted is meant to be real. This is in
contrast to the merely formal or combinatorial possibilities that
were to be so attractive subsequently to Hume and more mod-
ern philosophers (most of whom I regard as skeptical about
real possibility). I n the merely formal sense, any combinatorial
possibility is a possibility. Thus consider Princess Elizabeth. Some
pertinent facts about her are these: What she was—the kind of
being she was—was a human being; she lived in the seventeenth
century; she was not the daughter of Pablo Picasso and Lana
Turner. Also, she was an astute mathematician, an intriguing
correspondent, and interested in the passions.
The combinatorialist sees these traits as all of a kind when
it comes to delineating what is and is not possible for the prin-
cess. I t is possible that she would not be good in mathematics
and not care about the passions. But it is equally possible that
she would live in our time or Aristotle's; be the daughter of
Picasso; and still in the same breath, be a flower, a river, or a
book about the passions. A l l are possibilities for Elizabeth—as
indeed for any other object—because none are formally contra-
dictory. The combination of this bare subject—Elizabeth—with
any of the properties does not produce a formal, logical contra-
diction. I n the same vein, classed as impossible are only her
being identical to Descartes (assuming identity belongs in logic),
her being both a flower and not a flower, and perhaps her being
both married and single.
The Descartes-Arnauld idea of possibility is not the combi-
natorialist idea. Some will insist that both Descartes and A r -
nauld allude, sometimes, to "what God can create (do)." Is
this not a step away from "real possibility" and toward more
extraordinary possibilities? Perhaps so. But Arnauld would not
allow God to create a right triangle that does not abide by
Pythagoras' theorem, make the sum of 5 and 7 not be 12, make
a body that is not extended, and so on. Alas, Descartes, though

not in his work on mind and body, sometimes seems to allow
God to create the uncreatable (the " t r u t h " that 5 and 7 make
13). I ignore here such uses of "creativity" and "possibility."
Both Descartes and Arnauld use a nonformalist, real-individ
ual-bound idea of possibility, possibilities determined by the es
sence of the subject for whom (which) they are possibilities. Both
Descartes and Arnauld abide by the principle of whatness closure:
(WC) I f F pertains to what x is, then F is modally neces

sary for X.
Already ( W C ) excludes much of the combinatorialist cre

ation of possibilities cut from whole cloth. What the princess is
is a human being, and what the Loire is is a river. Conse
quently, she cannot be a water pathway leading the liquid from
the Massif Central to the Atlantic. And it—the river—cannot
be a princess or a philosopher or a passions bearer.
O f course, some delicate issues remain. One is this: Do we
have the converse of ( W C ) , that is,
(CWC) I f F is modally necessary for лг, F pertains to what x is.
Equivalently, i f F does not pertain to i ' s whatness, does it fol

low that it is really possible for x to be not-i? This question
will reverberate through the Arnauld-Descartes exchanges as
we move from the possibility argument to the coherent separa
tion argument. But whatever refinements await us, it is fair to
say that Descartes and Arnauld operate through and through
with a nonformalist, genuinely de-subject (what technical phi
losophy would term de re) notion of real possibility. Indeed,
the objection Arnauld is about to raise against Descartes and
Descartes's response will convince us that both assume a most
robust idea of possibility-for-a-subject. 5
5. Sophisticated readers often ask whether the Descartes-Arnauld notion

of possibility corresponds to the modern idea—derived from Saul Kripke—
that is often called metaphysical possibility or, as I prefer to put it, follow-
i6 W H A T A M I?
The possibility argument targets a real possibility for Des-

cartes's mind: Just as it might have had other intellectual acci-
dents—have had different desires and thoughts—it might have
existed without any attachment to this body D B ; indeed, it
might have existed without D B existing at all; stronger yet, it
might have existed without any (human) body in existence.
Possibilities galore, muses Arnauld in the beginning of his
fourth objections. But where from? Ascribed on what grounds?
Descartes's answer is that they are all conceivable of D M .
But "conceivable" how? Remembering our opening quotation
from the letter to Gibieuf; we know that Descartes has some
important qualifications—adverbs of manner—to add here. He
targets a conception that is clear and distinct and, most critical,
complete about the subject. Still, qualifications and all, his de-
fense is this: The foregoing ascriptions of real possibility are
justified once they appear to us as conceivable. I t is this justifi-
cation that Arnauld finds flawed.
Arnauld's Objection
In speaking of Arnauld's objection, we must separate the case

Arnauld presents from the theoretical conclusions (objections)
one may base on it. The case is one; the theoretical conclusions
it may lead to are many. The case runs like this:
ing Kripke's own use, "possibility tout court." Here we encounter one of
the advantages of lecturing about Descartes to immaculate beginners; they
do not ask questions of this kind. More seriously, I flnd it hard to make
such comparisons. The examples used by Descartes and Arnauld concern
mathematics. In contrast, a favorite stock of examples in our age concerns
the natural sciences (e.g., water is made in part of oxygen; gold is an
atomic element; etc.). Nonetheless, I happen to believe that the modern no-
tion of Kripke is more continuous with the Descartes-Arnauld discussion
than the 300-year gap might lead us to think. Still, in what follows I make
an effort not to indulge in time travel from Descartes's period to ours. Our
discussion—inevitably one run by a contemporary mind shaped by contem-
porary texts—aims at understanding Descartes's text in his own terms.
Suppose that someone knows for certain that the angle

in a semi circle is a right angle and hence that the trian
gle formed by this angle and the diameter of the circle is
right angled. I n spite of this he may doubt, or not yet
grasped for certain, that the square on the hypotenuse is
equal to the squares on the two sides; indeed he may
even deny this i f he is misled by some fallacy. (CSM I I ,
141—142)
Since so much concerns Arnauld's right triangle, let us call

it Г, conceived without the Pythagorean property, in short, P.
In analogy to the seeming possibility o f supposing T without
P, Arnauld would like us to consider the seeming possibility of
supposing my mind without my body:
But someone might call this minor premise into doubt

and maintain that the conception you have of yourself
when you conceive of yourself as thinking, non ex
tended, thing is an inadequate one . . . how is my percep
tion of the nature of my mind any clearer than his per
ception of the nature of the triangle? . . . Although I
clearly and distinctly know my nature to be something
that thinks, may I , too, not perhaps be wrong in think
ing that nothing else belongs to my nature apart from
the fact that I am thinking thing? Perhaps the fact that I
am an extended thing may also belong to my nature.
(CSM I I , 141—143)
These are Arnauld's actual words. How are we to understand

the theoretical bite o f his example? What, i f anything, does i t
show?
The First Reaction: The Primacy

of Possibility
What is Arnauld's own reaction? Arnauld prizes above all the

possibility-projection schema—whatever is really conceivable is
i8 W H A T A M I?
really possible. And he is wedded to it not because he views

possibility facts as based on conceivability facts. Quite the op-
posite: Arnauld noticed, very cunningly, indeed, that quite apart
from any conceivability test, it is a brute fact that
(A) I t is not possible for triangle T to exist without P.
This much obtains not because we try to conceive of T with-

out P and fail. Rather, altogether before any use o f conceivabil-
ity, we are given that i f T actually bears P, it necessarily bears
P. Admittedly, not every property o f T allows for such "neces-
sitation"; for example, T has the property of being Arnauld's
favorite figure, in short, Q. But it is not necessary that T bears
Q; Arnauld may have been fonder o f squares. So, some proper-
ties of T—perhaps those expressing mathematical facts—are ne-
cessitation-receptive, some are not, and all o f this does not de-
pend on what we may or may not be able to conceive.
Very well then: I t is impossible for T to exist without P.
Arnauld now argues that it is not really conceivable o f T that
it is without P. Here Arnauld asserts that what is not really
possible is not really conceivable. I n other words, he asserts the
possibility projection scheme—whatever is really conceivable is
really possible.
What Arnauld rejects is rather:
(Real conceivability projection) What we seem to conceive

of x, we really conceive of x.
This, the epistemic transparency of conceivability data, is Des-

cartes's starting point: I t could not turn out that it seemed to me
that I conceived—with the pertinent adverbial qualifications—of
something but I really failed. This much could happen with
more external verbs of activity like "touch" or perhaps even
"see." I may seem to touch and even see the singer Madonna,
but this will not make it a real touching or seeing o f her. But
i f I seem to conceive of Madonna, I , eo ipso, really do conceive
of her. There can be no gap between seeming to conceive and

succeeding i n doing so.
The epistemic transparency made conceivability an attractive
[ starting point for Descartes, and it is exactly what Arnauld de-
nies him. Modern examples—exploiting empirically grounded
necessities—have pushed this point further than Arnauld did.
Thus, I may seem to be able to conceive of water on a planet
where there is no oxygen. I envision a liquid that is raining
from the sky, running in rivers, tasting like water, and so on.
But, many w i l l retort, I have not really conceived of water on
this planet. There is no conceiving of water without the con-
ceiving o f H 0 ; this is simply what water is.
2
So, we must redescribe our activity. What I have done really

is to conceive of some water look-alike, or the-seeming-of-
water. Next, I have misdescribed my act as the real conceiving
of water, as the conceiving of real water. 6
Here, of course, the gap between seeming and being is more

obvious. The essential fact about water—it is H 0 — i s only em-
2
pirically given; men i n Descartes's time did not know it at all

(and were surely already thinking about water). I n Arnauld's
case, we are concerned with mathematical facts that are reput-
edly all knowable a priori; at any rate, the Pythagorean prop-
erty was surely so known by mathematicians for ages. Thus,
the gap seems harder to generate. Nonetheless, Arnauld insists,
we may seem to think about a mathematical object (or kind of
object) and be unaware of its full list of necessary properties;
stronger yet, we may be aware o f the property (we think about
and consider the applicability of the property) and consciously
deny it.
Extended to the case of the mind, Arnauld's point would
come to this: I may very well conceive of my mind—for that
matter, our kind of mind—and apply to this subject a certain
specific property: its need for an underlying body. We suppose
6. This way ofexplaining the gap betweenseeming and real conceivabil-

ity is discussed in some detail in the appendix 1.6, page 42.
20 W H A T A M I?
now that, in fact, it turns out that it is not merely actually true
that my mind is connected to a (this) body but also that of
necessity it is so connected. I f so, I stand corrected in my con-
ceivability claim. I have not really conceived of my mind in
this exercise; I have merely conceived of a mind, or of an epi-
stemically equivalent subject that from the inside is indistin-
guishable from my (our kind of) mind. O f that item, I have
conceived—really conceived—that it exists without bodily con-
nection. This does not show that I can so conceive o f my (our
kind of) mind. A l l in all, concludes Arnauld, D M and D B may
well be necessarily connected. They may or may not be two
items (this much is left open). But there is no real possibility
of D M without D B . I n this sense of "can," there is no real
distinction between D M and D B . Descartes refuted.
The Second Reaction: The Primacy

of Conceivability
Arnauld's reaction to his own case is based on the idea that

possibility precedes conceivability. I n assessing the question of
whether it is conceivable that water exists without oxygen or T
is conceivable without bearing property P, we are already given
the fact that these eventualities are not possible. This much is
not possible independently of any conceivability fact.
A n opposite reaction to Arnauld's case is driven by an oppo-
site doctrine, the primacy of conceivability. Conceivability facts,
what is and is not conceivable, are determined prior to and
independently o f the determination of what is possible.
Now, a strong form of this doctrine proposes that conceiv-
ability facts determine possibility facts. But this strong form—
the metaphysical subordination o f possibility to conceivabil-
ity—is not the motivation o f our presently examined response
to Arnauld's case. The response is rather driven by a concern
about epistemology—the wish to preserve, with Descartes, the
epistemic transparency of conceivability claims. I n this line o f
thought, conceivability is thus intrinsically different from sense
perception; in turn, it is intrinsically different from any notion

of imagination based on such perception. When I claim that I
see Madonna in front of me, I know that I may turn out to be
wrong. Prior facts about the external world may give the lie to
my claim, and twice over: There may be no such item at all ( I
hallucinate) or I am getting the object in front of me wrong (it
is a look-alike and not Madonna that I am seeing). No such
openness to correction by antecedent facts governs conceivabil-
ity claims: I f I seem to conceive of Madonna, I really do. Iri
this respect, "conceive" operates like "seem to see" rather than
tc «
see.
This view sacrifices the reliability of conceivability as a
guide to possibility only to regain the reliability of our own
self-ascriptions of what we conceive. Regarding Arnauld's t r i -
angle, this view proposes that we did manage to conceive T
without P (though it is not possible for T to exist iMess);
furthermore, we do conceive of water even i f we make i t — i n
the story—oxygenless; and we may well conceive of Elizabeth
even i f we misplace her in time, flouting a necessary truth, and
place her in the eighteenth century. I n the mind-body case, the
response asserts that we are successful in conceiving the mind
without any body on hand, even i f it turned out to be impos-
sible for the mind to exist without (this specific) embodiment.
The question of whether we manage to conceive of the mind-
bearing property F is simply prior to the question of whether
that mind could have really been F. I f it could not, fine: We
have already conceived of it as bearing F.
In sum, the present response suggests that we seem to con-
ceive of D M without D B ; thus, we really do. I n our conception
(or " i n the understanding"), D M is distinct from D B . We
would now like to conclude as follows: and so, in reality, D M
is distinct from D B . What is needed for this last step is a certain
kind of conceivability-reflecting-reality projection principle.
Exactly what kind o f principle? A general reality-projection
principle cannot be correct. Just because I conceive of you on
the beach in Rio does not make you be there. But there may
22 W H A T A M I?
be a restricted class of logically privileged relations for which

the passage from mere conception to reality is defensible. This
would be the case not because the conceiving makes reality be
its way. Quite the opposite: I can succeed in conceiving of a
certain—logically privileged kind of—eventuality only because
reality is this way. For example, it is only because in reality
D M is distinct from DB that I can conceive of them apart.
Thus, to get the intended reading of the principle—what fixes
what—we should read it contrapositively: I f in reality the dis
tinctness does not obtain, it can not obtain in the understanding
either. This is why I call the projection schema the reflection (of
reality) principle.
For the reflection principle to work, we need something
more than a restriction to logically privileged relations; we need
restrictions on the manner in which the relata—the objects of
the logical relation—are given. Here we are after special repre
sentations—what Descartes's calls "ideas"—of the purported
objects. Let Madonna and Francesca be in fact one woman.
This woman may conspire to confuse the population by creat
ing the impression of two distinct persons. I may then seem to
conceive o f Madonna singing in a Hollywood club but firmly
deny Francesca's involvement. A n y projection on that basis of
a numerical distinctness in reality would be mistaken.
This would not be so i f the "ideas" in question were essen
tially representative of the worldly objects, so that distinct ideas
would necessarily reflect distinct objects of which they were
ideas. I am not aware of a place where Descartes provides us
with such a theory of reflective ideas (though he says quite a
lot—and offers interesting bits—about ideas in his Meditation
I I I ) . Perhaps, in the end, his notion of a "complete idea" (al
luded to in the quotation from the letter to Gibieuf) can be
made to carry this load; that is, for distinct, complete ideas of
mind and body, reflection would hold. Let us call such complete
ideas С-ideas. I n our present view, when we say, " I t is conceiv
able that D M is not D B , " we only seem to be speaking directly
of the objects D M and D B . What we really are speaking of
directly are the С-ideas. We thus really say, " I n our concep
tion, C ( D M ) Ф C ( D B ) . " But now, by the reflecting reality
principle, we may infer that in reality D M is not D B .
I have at this point two things to say about this line of
argument. First, even granting the foregoing, we would still
have no result of real distinction. We would have the numerical
distinction of D M and D B , very much consistent with the ex
clusion of any sense of "can" according to which the one can
exist without the other. Second, the argument critically depends
on developing Descartes's notion of complete idea, which I do
in the next few pages. Then I return briefly to an evaluation of
the present defense of a conceivability-based yet possibility-free
proof of the numerical distinction of D M and D B . 7
1.5. The Primacy ofWhatness
The Third Way I: "Too Much 99

vs.
e(
Too Little 99
It is not quite clear to me where Descartes comes out in A r

nauld's case. I n his official reply (CSM I I , 158), he cites three
facts that mark a disanalogy between the geometric example
and the mind-body situation:
(i) I n the geometric example, neither the triangle nor the

property P are "things" (substances), whereas both D M
and DB are.
(ii) One may understand a right triangle T without prop
erty P but not think about P without thinking about
right angles; but one can think—symmetrically, as it
were—about mind without body and about body with
out mind.
7. See the notion of "conceptual fix" in I.5 and the analysis of conceiv
ability illusions in ch. 3, section 3.1.
2 4 W H A T A M I?
(iii) One can understand T without thinking about P but

not without any thought about the ratio between the hy-
potenuse and the two other sides; but one can think o f
the mind without reference to any kind of connection
to a body.
Even i f we grant the truth of ( i ) - ( i i i ) , I do not see how they

address the core o f Arnauld's objection, let alone provide a way
out of it. A t best, the points, especially ( i i ) - ( i i i ) , are the begin-
ning of a general theory of what is involved in a complete idea
of a given subject of conception. I t is such a theory that will
provide Descartes's response to Arnauld. I assume that Des-
cartes's full answer to Arnauld must lie beyond these three
short paragraphs.
What is it? O f the two responses discussed so far, Descartes
is certainly not attracted by the primacy of possibility, A r -
nauld's own way out. I t is rather the primacy of conceivability
as a way out that is closer to Descartes's inclinations, in the
fourth replies and in his letter to Gibieuf: De-modalize the real
distinction argument; leave real possibility out of it; insist that
D M and DB have been really conceived of apart; and thus, by
means of the reflection principle, get the result that in reality
D M is not D B .
Such a line of argumentation is not repulsive to Descartes.
But the argument is, as it were, an argument in waiting: I t still
needs to supply a theory of conceptual fix on a subject. A the-
ory of conceptual fix specifies what it takes a given thinker to
have in his or her understanding as a particular target object,
specifically, to manage to really conceive that object and not a
mere aspect of it or some qualitatively similar twin object (the
seeming of the object). We face here a quandary: Might not the
constraints about what it takes to conceive some item x come
from the very—conceivability-free—basic metaphysics of xi
Arnauld proposes one such metaphysical basis, the possibili-
ties involving χ. Let property F be modally impossible for x.
Suppose F has been conceived of some candidate item y. Then,

Amauld concludes, y is not x.
Arnauld may have been wrong about which metaphysical
features—modal possibilities—do the constraining. But even i f
Descartes rejects Arnauld's way out, he must admit that some
facts about x must constraint what it is to successfully conceive
of it. Descartes must supply some such conceptual fix theory.
As we shall see, he does.
So, we are looking for a third response to Arnauld's objec-
tion. Partly, we are looking here on behalf o f Descartes. I can-
not say that such a third response would be one literally enunci-
ated by him in one specific, canonical paragraph. Indeed, I do
not think there is any such single answer from Descartes but
rather fragments o f answers, pushing sometimes in one direc-
tion, sometimes in another. When we have the full third re-
sponse, I hope it w i l l turn out to be very much in tune with
his general line of attack on the real distinction proof.
We are searching for a third way out in the hope of isolating
Descartes's favorite response. But we are also searching because
it is natural to imagine such a third way, given the first two
responses. Both the first two ways out have virtues and vices.
Using Descartes's own terminology, I w i l l , instead of vices,
speak of solutions that give us "too little" or "too much."
As I see i t , Arnauld's response asserts too much in metaphys-
ics and too little in epistemology:
(i) Too much in its possibility involvement: Too much is re-

quired for having a conceptual fix on a target x; that is,
never ran afoul of x's necessary features.
(ii) Too much in the real distinction proof: I f one were to
defend a real distinction proof while adhering to A r -
nauld's theory o f conceivability, one would have to
ground the proof in a real possibility of disembodied ex-
istence for D M . Those interested in the dual key proj-
ect—integrating D M and D B in a single man—would
26 W H A T A M I?
find such a real possibility to be too much, too strong a

premise.
(iii) Too little in its epistemology: We have no reliable ac-
cess to what we conceive. Arnauld's allowance for a
gap between seeming and really conceiving is fine for
some cases (the geometric example). But by claiming
the gap across the board, we are left with too little
about our manner of apprehension of mind and body;
in this case, the concepts (ideas) through which we con-
ceive the target objects are more reliable.
The second way out—the one stressing the primacy of con-

ceivability—presents an opposite profile. I t asserts too little in
metaphysics and too much in epistemology:
(i) Too little constraining of conceivability. Having rejected

Arnauld's possibility constraint, we swing to the other
extreme—there is no constraint from the metaphysics o f
χ on what it would take to have a conceptual fix on x.
(ii) Too little in the real distinction proof We are provided
with merely the numerical distinction o f D M and D B
and no blueprints for how D M can exist without D B .
(iii) Too much in epistemology. The approach asserts an
across-the-board transparency thesis—seeming to con-
ceive suffices for really conceiving. This may be too de-
manding for many ordinary empirical subjects of con-
ception with nontransparent "essences"; it also seems
incorrect for certain mathematical subjects like A r -
nauld's T.
The Third Way II: Two Ground Facts
It would be appealing to submit here a third way out, free o f

the preceding vices and sensitive to the virtues. Before we so
proceed, let me add that such a proposal should respect what
strikes me as two basic facts that have so far been neglected.
Both facts seem to me intuitively right, prior to particular theo-
ries so far offered, and so basic that any theoretical reaction to

Arnauld's case would want to respect them. The first basic
fact—the epistemic—concerns the problem of what it takes to
have a conceptual fix on an object x; the second—the metaphysi-
c<z/-concerns the whatness o f x, what kind of object x is.
The epistemic fact is addressed to Arnauld. Consider his
right triangle Τ. I believe both Arnauld and Descartes feel—at
least at first blush—that we, and even practicing mathemati-
cians, can successfully think of T and deny its necessary feature
P. But we would not feel thus i f the thinker did not respect in
his conception features like triangle, many-sided body, figure
with angles, and even figure with three sides and a right angle.
So these features are more constitutive of having a mental hold
on T than mere necessary accidents of 7"-for example, it satis-
fies property P, has i n Riemann's geometry an angle sum that
is not that of two right angles, has in Euclidean geometry an
angle sum that is twice the atomic number o f uranium minus
four, and so on.
We find the same pattern when we think about empirical
subjects. Consider Descartes's favorite, the river Loire. One
may think of it and flout what may well be necessary of it: I t
originates in the Gerbier de Jonc Massif, it does not fall into
the Indian Ocean and does not flow on the planet Mars, the
liquid stuff flowing in it contains oxygen atoms, and so on. O n
the other hand, i f one were to deny in one's conception that it
is a river, one's claim to be thinking about the Loire would be
endangered.
This suggests that some among x's (necessary) features are
constitutive to our conceptual fix. This much concerns an epis-
temological question about us. But it leads to one o f metaphys-
8. I take it here as necessary about rivers (true in every way, such an

item might have evolved while in existence) that, even as the item changes
its course, it keeps originating in a given mountain range, does not fall into
the sea at the other side of the globe, and carries in it give-and-take impuri-
ties HjO molecules.
28 W H A T A M I?
ics: Might not some among x s necessary features be "more"

9
constitutive to what it is? Reconsider the triangle T. Surely it

has infinitely many mathematical features—all o f them neces
sary—that do not pertain to what it is.
What then is 7? A right triangle; in turn, an extended body.
The last two features articulate what this item is, the very kind
of item it is. I n contrast, many of its necessary features—I use
here, with a twist, the medieval language of "necessary acci
dents"—do not concern what T is. Rather, they are subsequent
to the kind-specifying predication. First we are given the sub
ject: I t is a right triangle. Only now, having articulated its
whatness, we may ponder how this item might have been. We
may find that throughout each and every way (mode) of its
being, it could not have failed to satisfy property P; for that
matter, it could not have failed to be Pythagoras' actually fa
vorite figure (here the addition of "actually" makes this a neces
sary feature of Г, given that it was indeed Pythagoras' favorite
figure). The necessary accidents apply to an already fixed sub
ject with antecedently given whatness—a right triangle.
In the same vein, let us consider another mathematical ob
ject, the number 4. Its being a number—and a positive inte
ger—pertains to what it is. So does the feature o f being the
successor of 3. But 4 has many other necessary features: I t
solves the color map hypothesis in topology; it is twice the
number of hydrogen atoms in a water molecule; it is the num
ber o f dimensions in a Minkowski space-time structure. These
are all accidents 4 could not have failed to have. And yet they
do not articulate what kind of subject 4 is.
The idea of what a given subject is, the kind of thing it is,
is very primordial in Descartes's thinking. I n various contexts,
this—rather than any modal question about necessity—is his
first question about the subject of investigation. He so asks
about the triangle in Meditation V , as well as about mountains
and about God; he so asks about the sun in Meditation I I I (in
separating our different ideas about it from what that body is);
and, of course, Meditation I I is saturated with such questions
about the piece of wax (on which much more later) and about
the mind and the body, and it raises our primal question: What
am I? I n all, Descartes is after the kind of item involved, not
just the listing of its accidents, necessary though they may be.
The theory about whatness-specifying features is part of the
metaphysics of a given subject. But we see that it soon leads to
epistemic corollaries. I n the quest for natural constraints on
what it takes to have a conceptual fix on a given object x, we
look for metaphysically distinguished features of x. Arnauld
proposes the necessary features. But they fail twice over. They
fail epistemically because intuitively this demands too much.
We all feel that arbitrary, hidden necessities about x cannot undo
our mental hold on x. Arnauld's modal answer fails a second time
in not distinguishing, now in the metaphysics of x, the premodal
articulation of just what kind of thing this subject is from a subse-
quent business—how that subject (now already given) might
have been, which ways it could have gone on to be.
I submit that both basic facts must be respected by a theory
of complete ideas or conceivings about x. I t now seems that a
proper treatment of the epistemic basic fact might be a byprod-
uct of respecting the metaphysical basic fact. I n other words, a
theory of genuinely thinking about x may rely on a theory of
what χ is. I believe that Descartes's remarks about complete
ideas supply such a theory, and it is to this theory that I now
turn.9
The Third Way III: Complete Ideas
Arnauld is uneasy about the lack of constraints in Descartes's

theory of conceptual fix: "But someone might call this minor
premiss into doubt and maintain that the conception you have
9. The preceding section introduces the distinction between what a

thing is and how it is. I have discussed the distinction in a more general
metaphysical, Descartes-free context in "The What and the How I , " Jour-
nal of Philosophy (1991); and "The What and the How I I , " Nous (1996).
3o W H A T A M I?
of yourself when you conceive of yourself as thinking, non

extended, thing is an inadequate one . . . " (CSM I I , 141). The
letter to Gibieuf reports similar worries:
You inquire into the principles by which I claim to

know that the idea I have o f something is not an idea
made inadequate by the abstraction o f my intellect. . . i n -
tellectual abstraction would consist in my turning my
thought away from one part of the contents of the richer
idea which I have in myself to another part with greater
attention. . . . O f course, by abstraction we can obtain
the idea of a mountain without a valley or o f an upward
slope without considering that the same slope can be
traveled downhill. (PL, 123-124)
This, then, is the worry. I t arises dramatically when the pri-

macy o f conceivability is assumed. This suggests that from the
distinctness of the ideas associated with D M and D B , we con-
clude that in reality D M is distinct from D B . But what i f D M
and DB were one and the same—rather like Madonna and
Francesca—and the two ideas arose here by intellectual abstrac-
tion?
The worry arises in the context of the primacy of conceiv-
ability solution, but its force is quite general. Descartes rejects
Arnauld's theory of conceptual fix, that is, respect all necessities
about χ. Very well. But he seems to offer nothing instead. This
opens the floodgates to wild claims o f conceivability about x:
N0 matter how off-base the conception of дг, we still conceive
of it. This cannot be Descartes's ultimate defense o f the mecha
nism of conceivability.
It is not. Descartes tells us quite explicitly:
There is a great difference between abstraction and exclu

sion. I f I simply said that the idea which I have of my
soul does not represent it to me as being dependent on a
body and identified with i t , this would merely be an ab

straction, from which I could only form a negative argu
ment, which would be unsound. But I say that this idea
represents i t to me as a substance which can exist
though everything belonging to body be excluded from
it; from which I form a positive argument, and conclude
it can exist without the body. ( P L , 152 ) 10
The second text is the already quoted passage from a letter to

Gibieuf:
The idea which I have o f a thinking substance is com

plete i n this sense and I have i n my mind no other idea
which is prior to it and joined to it i n such a way that I
cannot think o f the two together while denying the one
of the other. ( P L , 124 ) 11
Here, i n a couple o f elegant paragraphs i n later letters, Des

cartes encodes the essential ingredients o f his "complete idea,"
a four-page answer to Arnauld i n the fourth replies (CSM I I ,
156—160) and related remarks to Caterus i n the first replies
(CSM I I , 85—86). What follows presents a way o f understand
ing Descartes's notion o f complete idea.
First, in the term "complete idea," the adjective characterizes
a kind o f idea. But Descartes also uses "complete" to modify
"thing"; for example, i n the reply to Arnauld, he keeps refer
ring to the distinction between complete and incomplete things
(and "entities"): " I took 'complete understanding o f something'
and 'understanding something to be a complete thing' as having
one and the same meaning." A n d i n the next paragraph, he
says: "By a complete thing I simply mean a substance endowed
10. Letter to Mesland, May 2, 1644.

и . Letter to Gibieuf, January 19, 1642.
32 W H A T A M I?
with the forms or attributes which enable me to recognize that

it is a substance."
This is how I intend to understand the notion of complete
ness, that is, as relating primarily to things, to purported sub
jects. I will just read the complete understanding (idea) of
something as an understanding (idea) of a complete thing.
Now, "complete" is a technical adjective. How are we to
understand its intuitive import? When do we have on hand a
complete thing? Arnauld offers one candidate: A story (a sce
nario, a conceiving) operates with a complete idea of дг, that is,
with an idea of the complete thing x i f f the story (etc.) respects
all of *'s necessary traits. Descartes rejects this understanding
as requiring "too much," and he seems to be right.
A t the other extreme lies the worry of a story that is operat
ing with a mere intellectual abstraction on x, a fixation on a
mere "backside" of x. A n example of such an incomplete (idea
of an) entity is given in the letter to Gibieuf. Descartes men
tions a discourse about a shape S without any reference to the
substance whose shape it is. The example is repeated in the first
replies (CSM, 86), and another one is added: T o think of justice
without thinking of the person who is just is to similarly intro
duce an incomplete entity.
Very well: On one end, we are dealing with mere aspects,
modes, or abstractions from the full-blooded subject; at the
other end, our conception of the subject is so full that it is
overloaded with any necessary feature borne by the subject.
We swing between too little and too much. Is there any natural
balancing point here?
I should like it noted that the question posed is not about
the creation point of a subject, as i f we wondered how, by gluing
enough properties together, we can conceive-into-being a real
being. The question is rather about the limits o f subtraction from
a given real being, what Descartes describes as "exclusion" from
a given subject. Alas, many readers, contemporaries of Des
cartes and contemporaries o f us, think that the question is one
of "philosophical (conceptual) creation": When have enough
T H E REAL D I S T I N C T I O N 33
properties been thrown into the conceiver's pot to create a real

subject? T o which the answer should be: never. One never
creates a subject—a real subject like Nixon or a fictional one
like Hamlet—by gluing properties together. Rather the question
is this: O f an already given subject x provided by the history
of the world, what are the limits of subtractions from the list
of its actual features, preserving all the while the sense that the
story (conceiving) is still about x? 12
The Third Way IV: Whatness vs. Necessity
Our question to Descartes is this: Suppose we are conceiving

in the intended subtractive mode and suppose we are focused
on a given subject x. What is the condition for success? Is there
a natural balance point between requiring too much—do not
tinker with the necessities about x—and requiring too little,
whereby x itself is subtracted and lost?
I believe that Descartes submits that there is such a natural
condition. I t is the respect of what the thing is. T o articulate
what an item is is much less than providing its full list of neces-
sary features [ I assume, as did ( W C ) above, that its whatness
is included in the necessary features list.] On the other hand,
to preserve the whatness of the item is much more than to
abstract from the item this or that aspect; in the latter, we
merely focus on a perspective we happen to have on the item.
12. This much applies as well to fictional characters like Hamlet. Shake-
speare did not generate Hamlet by gluing properties together. Rather, the
world (here certain historical acts of the author, Shakespeare) provided
Hamlet. We may now try to conceptually subtract some of Hamlet's fea-
tures, for example, being melancholic. Of course, with fictional items we
have to be careful about what Hamlet is in the story (a man and melan-
cholic) and what he is outside the story (a fictional character). Various com-
plications arise from this duality but they do not take away the main point:
Real subjects—fictional or not—are not created by gluing properties to-
gether.
34 W H A T A M I?
I n assessing the answer I take Descartes to be giving, we

must remind ourselves that we are dealing here with a phenome-
nological question: What captures best our sense that the story
(conceiving) is still about the purported subject x? T o this kind
of question, we should not expect strict "algorithmic" rules.
We test various candidates and find out—by trial and error—
what kind of features get our intuitions satisfied. Descartes's
answer, as I read i t , is this: We may well subtract necessary
features and still be speaking of the purported subject x. But
subtract what-x-is, and gone with it is x itself. We are no
longer subtracting from x; we are subtracting x. I n Descartes's
own language, we are no longer excluding from what x is its
mere how; we are now abstracting from x a new item, a mere
property or mode thereof.
The whatness condition seems right for empirical subjects;
for example, consider Nixon. We may conceptually subtract
from this man features philosophers have taken as necessary,
for example, make him live 20 years earlier than he did or
make him emerge from gametes other than those he actually
emerged from; and though, of necessity, Nixon was the actual
president of the United States in 1970, we may subtract this
feature in the story. But i f we subtract his being human, perhaps
making him a frog or a crocodile, we are destroying the sense
that the story is a coherent story about Nixon.
The condition seems pertinent for mathematical items. Con
sider Arnauld's triangle T. We may conceptually subtract the
property P. Indeed, I believe that Pythagoras, thinking of Г,
may have conjectured to himself, " I bet P is not true o f such
right triangles," only to find to his surprise that 7"bears P after
all. This often occurs in mathematical work: Mathematicians
think about some (kind of) object x. On their way to eventually
proving that x is F, they consider (and sometimes, convince
themselves) that x is not F.
A n interesting example is seen in a branch of mathematics
called set theory. A certain story may be provided—about what
sets are—that from a later perspective turns out to be (i) actu-
ally false about sets and (ii) as such, necessarily false. Nonethe-
less, the story is a coherent story about sets, preserving their
very essence.
Here is a summary o f such a story. The theory of sets is
taken to describe a "real" universe of sets V . By one of set
R
theory's own early results, V itself is not a set; it is a plurality

R
of sets that is not itself collectible into one more set. When set
theoreticians initially thought about the fundamental structural
operations that generate the universe o f sets, they articulated
the following: We generate sets in stages. We start, in stage
zero, by taking as our basis the empty set, the unique set with-
out members. Given this basis, we go on to generate, at later
stages, more sets. We have two such generative operations for
the two kinds of stages we encounter. I n any successor stage—
the one after the first, the one after that, and so on—we take
the power set, the set of all previously generated subsets, and
make that power set the new universe. The second operation
concerns limit stages of construction—the stage beyond the
first, second, and so on—a stage we think of as the first inhnite
stage, dubbed the CO stage. A t such a limit stage, we take the
union o f all previously available sets. Applying these operations
on and on, we generate the universe of sets.
The question now is this: How many stages does it take to
generate the full universe? Suppose we think that there are only
CO many stages o f generation. We get a universe with infinitely
many sets, each of which is merely a finite set. I n this universe
of finite sets, no single set is itself infinite. We may call this the
universe of finite sets, V . This story casts in modern set theo-
F
retic terms (as it were, with hindsight) what was probably the
received view for generations of mathematicians who believed
in the merely "potential infinite," a universe with infinitely
many items but no single infinite item. This common view was
questioned in the nineteenth century by the work o f mathemati-
cians like Bolzano, Dedekind, and Cantor.
Under the impulse of these nineteenth-century pioneers, it
has become a received truth of modern set theory—by the first
36 W H A T A M I?
decade of the twentieth century—that on top of the sets of V , F
there are "larger" sets. Modern set theory asserts not only that
the universe of hitherto generated finite sets is itself infinite but
also that there exists at least one infinite set; for example, ω
itself is such a set. Indeed, it goes on to assert that many such
infinite sets exist. But the existence of at least one such set, set
I , is sufficient for our purposes. I will assume it is true—in the
real universe of sets V —that there is this set I . Since this is a
R
mathematical truth, it is automatically a necessary truth.

We now confront the key question: Consider a story—the
common view for centuries—that provides us with the universe
of finite sets as the full domain of sets. By our hypothesis, this
story is (i) not really true about sets and (ii) not really possible
about them. Nonetheless, it is arguable that this is a coherent
story about sets. The story does not respect how many sets there
are, but it does respect what sets are. This last claim is defensi-
ble because the story—as articulated above—gives us the funda-
mental generative operations that produce sets. We omit some
real sets from the story, but we don't omit what it is to be a
set.13
13. (For the technically minded reader only): Of course, this example is
presented here merely as an illustration. I do not mean to delve into the
philosophy of set theory. What is more, many might think that the opera-
tions under which the universe of finite sets is closed do not exhaust what
sets are. They may think so (i) because V is wrong about how many sets
F
there are or (ii) because, although in principle a model may be wrong—it

may omit some real sets—but still respect the essence of sets, the omitting
of all infinite sets is incoherent. Why? Because it is constitutive of what
sets are (of the "very idea" of set), that there is at least one infinite set.
I cannot address here the objections in detail. The second objection may
seem to have much plausibility. I will only say that even i f it were right,
other examples of large infinite sets 5"-of fancier character—arise, in which
all sides would agree that the existence of S is (i) true in the real universe,
hence necessary, and yet (ii) not pertaining to the very essence of sets. For
example, consider Ernst Zermelo, who stated early in the twentieth century
the axiom that ω is a set, indeed, the least in size-infinite set. He did reveal
The moral so far is that we should not confuse consistency

with possibility. W i t h this in mind, let us go back to Arnauld's
own original geometric example. I t displays nicely the gap be-
tween consistency and possibility that I have been harping on.
To prove property P of T, one needs the axiom o f parallels of
Euclidean geometry. Given that the axiom is true o f Euclidean
space, it is necessarily true, and so is 7"'s bearing P. But we
may consider models of the first four Euclidean axioms, which
do not satisfy the axioms of parallels; that is, right triangles d0
not bear P. Nonetheless, it is arguable that beings living " i n -
side" such a space and conceptualizing (visualizing) it non-
Euclideanly may very well have the idea of a right triangle.
So it goes for the mathematical examples—a gap between
real possibility and the weaker idea of consistency with what is
χ. I n the present reconstruction, we read Descartes as looking
for a similar gap between necessity and whatness in the mind-
body real distinction proof. This leads to Descartes's third re-
sponse to Arnauld.
Let us grant Arnauld that, of necessity, D M and D B coexist
in a union inside the human being whose mind and body they
are. I call this the necessary union hypothesis. We thus grant
Arnauld the analogy he was seeking—just as triangle T is of
a way to number such infinite sets according to size, thus providing for the
second infinite size, the third, and so on. But what about the 00 place in
this sequence, what is standardly referred to as K ? Is it also a set? Zer-
m
melo's own theory "forgot" to state principles that guarantee the existence
of this set. The theory is consistent with there not being such a set. But
given its actual existence in V , it is impossible that this set does not exist.
R
Similar remarks apply to yet larger sets, for example, those called in set the-
ory "the (strong) inaccessible" cardinals, whose theory was articulated by
Zermelo later, in the 1930s. The objector who resists the present line of re-
sponse will have to suggest that each such (existential) truth about sets is
ipso facto pertaining to the essence of sets. I regard this as much too strong
a claim. It does not distinguish arbitrary truths about sets from truths about
what they are.
38 W H A T A M I?
necessity bound to property P, subjects D M and D B are of

necessity bound to each other.
Nonetheless, D M is distinct from D B . H o w so? W h a t - D M -
is is a thinking thing; what-DB-is is an extended thing. Thus
D M is by what it is (nature, essentially) thinking. D B is not.
Hence, by the distinctness of discernibles, D M and D B are two
subjects. We may put this argument in a form analogous to
that displayed by the possibility and conceivability arguments.
That is, D M bears the property consistent-with-what-it-is-to-not-
be-extended; DB does not bear this property. Thus D M and D B
are numerically distinct. Also, the consistency-with-what-DM-
is provides a coherent story, with unextended D M . I t is not
DB in the story. This last may be true because D B is in the
story, but being essentially extended—hence, extended in this
story, too—it is not D M . Stronger yet, D M may be distinct
from D B in the story simply because there are no extended
items in the story and thus, in particular, D B is not in it. As
mentioned earlier, Descartes's Meditation I I insists on the co-
herence of such a story. Assuming it, we have a sense according
to which D M can exist apart from DB. We secure the real
distinction: Even i f it is not really possible, it is logically consis-
tent with what D M is, being a thinking thing, that it exists (i)
without any extended thing or (ii) without the specific extended
thing, DB.
The Third Way V: Evaluation
A l l in all, Descartes's response follows the structure of Arnauld's

own theory without accepting its overly strong premises. I thus
see this third way out as having (i) logical, (ii) metaphysical,
and (iii) epistemic advantages over the two other escape routes.
Logically speaking, we operate throughout with weaker
premises (and sometimes they are not merely weaker; they
seem truer). Whereas Arnauld projects real possibility from
conceivability, Descartes merely projects a coherent story that
might well represent no real possibility. Whereas Arnauld i n -
sists on real possibility of disembodied existence for a real dis-

tinction between D M and D B , Descartes requires merely sepa-
ration in a coherent story (model). Whereas Arnauld demands
respect for all of x's necessary features for a conceptual fix on
χ, Descartes requires merely respect for what x is.
Metaphysically speaking, the third way out allows for an
element that might strike many modern thinkers about the
mind-body problem as more likely than not—the necessary con-
nection of the human mind and body. I , for one, would rather
assume this at the outset and bar claims o f the genuine possibil-
ity o f the disembodied existence of D M . This is all the more
poignant in the context of this book because of the second stage
of our dual key project. I n chapter 2, we are to integrate D M
and D B as the mind and body of a single human being. Such
an integrative theory assumes, at the very least, that the human
mind and body—indeed this specific human mind and body of
this specific human being—are necessarily connected.
Finally, this real distinction proof preserves an attractive
epistemology of conceivability. Arnauld dispenses—across the
board—with the sense that at least sometimes when we claim
to have conceived of something, we are guaranteed to be reli-
able. He turns examples in which our access to the necessary
properties of the subject is flawed into the paradigm for all
cases of conceivability. This strikes me as oversweeping.
Arnauld may provide the right picture for subjects like the
Loire, this table in front of me, the substance water, or the
species tiger—subjects whose essence (what they are) involves
scientific discoveries in chemistry, biology, and geology con-
cerning the material composition of the external world. But the
case of D M and D B seems categorically different, at least from
Descartes's standpoint (which I am here trying to reconstruct).
Let someone assert that "what my mind is is a thinking be-
ing" and "what my body is is an extended thing." I t is not
clear what scientific discoveries—of the kind made about the
nonfishood of whales and the noncompoundness of gold—may
arise and give the lie to these assertions.
4o W H A T A M I?
A t one point in his critique of Descartes's method, Gassendi

(in the fifth objections) says to Descartes:
I f we are asking about wine . . . it will hardly be enough

for you to say "wine is a liquid thing, which is com-
pressed from grapes, white or red, sweet, intoxicating"
and so on. You will have to attempt to investigate and
somehow its internal substance, showing how it can be
manufactured from spirits, tartar, the distillate and other
ingredients in such and such quantities and proportions.
Similarly . . . you must see that it is certainly not enough
for you to announce that you are a thing that thinks and
doubts and understands etc. You should carefully scruti-
nize yourself and conduct a kind of chemical investiga-
tion of yourself, i f you are to succeed in uncovering and
explaining to us your internal substance. (CSM I I , 193)
To this Descartes replies:
You want us, you say "to conduct a kind of chemical i n -

vestigation" of the mind, as we would of wine. This is
indeed worthy o f you, O Flesh, and of all those who
have only a very confused conception of everything and
so do not know the proper questions to ask about each
thing. (CSM I I , 248-249)
The rudeness is gratuitous, but Descartes's point is well

taken. Scientific investigation may reveal to us that what wine
is is not what we took it to be; and though we thought we
were conceiving something about what seemed to be wine, the
subject of the conceiving was not really wine. Thus we were
not really conceiving—whatever it was—of wine. This is not
true of our mind and body. Science may divulge to us various
truths about both mind and body, perhaps necessary truths. But
our basic conception of what each one of us is is not threatened
by future chemical discoveries. I f we seem to conceive of mind
and body—of a thinking and an extended thing—we are really
doing what we seem to be doing. A n d so our two conceivabil-

ity-based results have a bite:
(Ci) We seem to conceive of D M as distinct from D B ;

thus, we really conceive the numerical distinction.
(C2) We seem to conceive of D M as existing without DB;

thus we really conceive the real distinction.
We see how the metaphysics and the epistemology fall into

fheir natural places. Metaphysically, we prove, from conceiv-
ability-free facts about whatness, that
(Wi) What D M is is not what D B is; thus, D M is distinct

from D B .
Also, we prove that
(W2) I t is consistent with what D M is that it exists and D B

does not; thus, D M and D B are really distinct.
So much is metaphysics. But we have a reliable epistemology

with which to come to know i t . Our access—conceptual fix—on
what D M is and what D B is is not as tenuous as Arnauld's
pessimism would have it. I n this domain—if not when we think
about whales, gold, and triangles—our complete ideas are re-
flective of the whatness of the items they are ideas of. By means
of ( C i ) and (C2), we have reliable access to ( W i ) and (W2).
Thus, we conclude chapter 1—focused exclusively on the
duo of D M and D B — i n a rather optimistic vein. I t remains to
be seen whether the optimism can be sustained when we bring
in Descartes's third player, the full-blooded human being;
whether our whatness assignments, to D M and D B , can then
still be the purely generic "a thinking thing" and "an extended
thing"; whether, in turn, we would still be able to claim that it
is consistent with what D M is that it exists without D B ; and
42 W H A T A M I?
finally, whether it would then still be really conceivable of D M

that it exists without D B .
i.6. Appendix: Conceivability—

Does Epistemology Precede Metaphysics?
Descartes, Arnauld, and Kripke
Three arguments for the discernibility of D M and D B were

mentioned—from conceivability, possibility, and whatness. We
experimented with grounding the discernibility in both the pos-
sibility and whatness properties. But we never quite allowed
ourselves to consider the conceivability argument in and o f i t -
self. What could be said in defense of such an autonomy of
conceivability thesis?
As we saw, Arnauld took an unbending negative stand on
this—there can be no "autonomous" conceivability argument;
what is really conceivable is dependent on what is really pos-
sible. I n our own third-way argument, we softened this by
making real conceivability free o f arbitrary possibility facts.
Nonetheless, it did turn out to depend on real whatness facts.
Thus, in both Arnauld's discussion and ours, the metaphysical
precedes the epistemological.
The question before us is whether we cannot do any better
on behalf of "the autonomy o f conceivability" conception; i f
we cannot, we need to know why; that is, is there something
with the force of a result that blocks us here?
In what follows, I would like to consider what may well be
the most powerful attempt to date to explore the question, one
arising from Saul Kripke's Naming and Necessity. Although I
have vowed not to bring into this book modern analyses that
go beyond the context of Descartes and Arnauld, Kripke's dis-
cussion does take us beyond this period. But I believe that this
one reference beyond the seventeenth century provides genuine
illumination over the very issues discussed by Descartes and
Arnauld. My hunch is that, learning of Kripke's analysis, Des-
cartes and Arnauld would have immediately seen the pertinence

of his remarks to the "autonomy of conceivability" thesis they
were arguing over.
Kripke presents a beguiling intermediate position between
Arnauld and Descartes. I n many ways, he is very close to A r -
nauld because of Kripke's fundamental distinction between
seeming to conceive and really conceiving. Also, like Arnauld,
Kripke emphasizes that in the most general case, empirical or
mathematical, we may turn out to be wrong about what we*
really conceive of; it all depends on whether we respected the
essence of the target object, and this essence may not be trans-
parent to us. I n Kripke, as in Arnauld, the epistemological is
reined in by the metaphysical.
A t the same time, Kripke sides with Descartes, against
Gassendi and Arnauld, in conceivings of mental items. Like
Descartes, Kripke takes conceivings of items such as wine in
one way, conceivings of mental items in quite a different way.
Like Descartes, Kripke hopes to avert, in this realm, the gap
between seeming to conceive and really conceiving. I t is my
conviction that because of this nuanced intermediate position,
Kripke's analysis generates a "best possible" result for the au-
tonomy of the conceivability thesis. Indeed, so nunanced is the
result that it does not uniformly treat all things mental. The
result has its surprises.
Logic Matters: De Re vs. De Dicto Arguments
Explorers of a conceivability argument—like Descartes and

Kripke—hope to achieve two aims: (i) a metaphysically simple
discernibility argument, by means of the genuine property con-
ceivably-exists-without-the pertinent-physicalium, and (ii) an epis-
temologically transparent argument, affordable by the direct ac-
cessibility of the conceivability data. I n the process of moving
back and forth between the metaphysics and the epistemology,
we run the danger o f mixing the logical forms of different argu-
ments. T o avert the danger, we need to separate two types of
44 W H A T A M I?
arguments from conceivability. I will first draw the distinction

in Kripke's framework, where these logical differences are ex-
plicitly available, and then extend it to Descartes's conceivabil-
ity argument.
The mind-body arguments of Saul Kripke in his Naming and
Necessity illustrate the two types of argument I have in mind.
Kripke is not directly focused on the distinction of mind and
body (though he does speak o f his arguments from time to time
as showing the distinctness of, e.g., the man Descartes and that
man's body). I n the main, Kripke attends to the distinction
between the mental state (phenomenon) of pain and the brain
state (phenomenon) of the firing of C fibers (henceforth, FCF).
Also, his focus is on imaginability rather than conceivability.
But the distinction we are after applies across these variations;
as mentioned, it is a distinction of logic. I t is between what I
call the de re argument and the de dicto argument.
In Kripke's de re possibility argument, we may use whatever
expression we wish to pick out certain purported objects—say,
a certain pain I had at 6 A.M. and a certain related brain state
occurring in my brain at 6 A.M. We may, for example, pick the
two purported items by speaking of my least favorite sensation
on Monday morning and the last event recorded on the M R I
(magnetic resonant imaging) machine at 6 A.M. Let us use A
for whatever such term we use to pick the pain, B for whatever
term we use to specify the brain state.
Kripke's de re argument calls on the following single, non-
logical premise:
(Ko) A could have existed without B.
Thus, (Ko) is read to have the following structure: A certain

object A—bears a certain modal property—could have existed
without B. Surely B does not have this modal property. A n d so,
by the distinctness of discernibles, Kripke may conclude that A
is numerically distinct from B. Thus, the sensation is not the
brain event. We may, of course, wonder about the truth of the
key premise. But it is clear that the argument is free of any

speculation about our (linguistic) representations o f mind and
body.
In contrast, Kripke's de dicto argument essentially involves
theories about language. Kripke considers the terms "pain" and
"FCF." He asserts about both that they pick out in every count-
erfactual possibility the same state (phenomenon) fhat they
actually pick out. I n this, they are different from such designa
tors as " m y least favorite sensation" and "the state recorded'
last on the M R I o f my brain." These designators may each pick
different phenomena in different possibilities. The phenomena
of pain and FCF are what, respectively, these two designators
actually pick out. But there is a possibility in which "my least
favorite sensation" picks out the sensation of headache, and
"the last event recorded on my M R I " refers to the firing of H
fibers. I n that possibility, the headache sensation is my least
favorite sensation and the firing of H fibers is the last state to
be recorded on my M R I . Kripke calls rigid those designators
that keep picking up through all the possibilities what they ac
tually pick up. He now asserts the first premise of his de dicto
argument:
(Ki) "Pain" and "FCF" are rigid designators.
Next, Kripke takes the English sentence "Pain is the firing

of C fibers" to express the identity "Pain = FCF." We encoun
ter here another linguistic assumption: The form of the English
sentence is that of an identity. I doubt the assumption but act
in what follows merely as an expounder of Kripke's argument.
We are now ready for Kripke's second de dicto premise:
(K2) " I t is possible that pain Ф FCF" expresses a truth.
This is the analog o f his direct de re modal property intuition,

(Ko). But it is merely an analog. W i t h (Ko), we had the straight
predicative claim: This subject, pain, could have existed without
4б W H A T A M I?
that brain state B. The modal locution "possibly" (or "could")

applies to "exists without B" to form a complex modal predi
cate. This predicate is said to hold of the phenomenon of pain.
In the de dicto argument, the modal locution " I t is possible
that" acts as a sentential modifier, applying not to a genuine
property but to a sentential representation—the whole claim (in
latin, dictum), "Pain Ф FCF." This sentence is said to be possi
bly true. Together, ( K i ) and (K2) deliver
(K3) "Pain Ф FCF" is true (in reality).
It is quite clear that the de dicto argument rests critically on

various semantic doctrines about the representations used in our
dicta. One such assumption, ( K i ) , concerns the terms "pain"
and "FCF." For example, i f these two representations meant
something like "Almog's least favorite sensation" and "the
event last recorded on the Almog M R I , " ( K i ) would be false.
The de dicto argument would have failed, even i f the de re
argument remained unaffected.
Many philosophers of language before Kripke did believe
something like this about the meaning of the terms "pain" and
"FCF." Some continue to hold on to such a semantic theory
even after Kripke. I mention this not as a skeptic about Kripke's
theory of language. Quite the contrary: I simply want us to take
notice of the fact that the de dicto argument depends on extra
assumptions about the nature of our linguistic representations.
The distinction between the two types of arguments extends
in a natural way to the conceivability case. The de re conceiv
ability argument assumes this one premise:
(Co) Pain is conceivable-without-FCF.
As before, (Co) is really two-tiered, but both tiers are meta

physical- and language-free. The first tier is that conceivably so
and so is indeed a genuine property of things, borne indepen-

dently o f how the thing is specified. The second is that conceiv-
able-without-FCF is borne by pain. Assume both: Then, surely,
FCF does not bear the property. Thus, pain is not FCF.
Like its modal twin, the de dicto conceivability argument is
more complex. We need first an assumption about our repre-
sentations of pain and the related brain state, assumptions about
what Descartes calls "our ideas" of pain and brain states. We
may want to distinguish here, too, between representations like
"pain" and " m y least favorite sensation" and, respectively, be-
tween "the firing of C fibers" and "the state last recorded on
my M R I . "
One basis for differentiating is this: When I employ the rep-
resentation "my least favorite sensation" in conceivings (Des-
cartes would say, " i n my understanding"), my mind has the
freedom to alter what the representation actually stands for. I t
actually stands for the sensation of pain, but I can employ it in
my understanding to get to the sensation of tasting fish. I can
alter what is being picked out while sticking to the conventional
meanings of the words used; for example, "sensation" still means
sensation, in my conceiving exercise. I n contrast, the word
"pain" is resistant to any such conceivability-induced alteration.
When I deploy i t , thinking to myself, "Let me conceive that
pain . . . " I may have some freedom about which predicates I
apply to pain in the story constructed; but it is to pain—the old
real sensation—that I apply these predicates.
We may summarize this difference by calling representations
like "pain" and "the firing of C fibers" reality-bound: Even in
conceiving exercises, they are employed to pick out the very
items they pick out ordinarily, outside of conceiving exercises.
Representations such as " m y least favorite sensation" and "the
last state recorded on my M R I " are not reality-bound. I shall
call them unbound—my conceivings are free to endow them
with new referents.
We are now ready to state the first assumption of the de
dicto conceivability argument:
48 W H A T A M I?
(Ci) "Pain" and "FCF" are reality-bound designators.
We next need the assumption that
(C2) " I t is conceivable that pain Ф FCF" expresses a truth.
W i t h both ( C i ) and (C2) given to us, we may conclude that
(C3) "Pain Ф FCF" is true (in reality).
Has Transparent Epistemology Triumphed?
We have distinguished the de re argument from the de dicto

variant. A natural question arises in both the possibility and
conceivability cases: W h y would one rely on the much more
complex de dicto argument? After all, given the truth of its
discerning premise, the de re discernibility is logically simpler.
It is also much closer to the intuitive way in which we often
seek a separation of two tightly connected candidates—we work
through and through at the level of objects and properties, lan
guage notwithstanding, to find a discerning property. So, there
remains the mystery: W h y wheel in the much more complex
de dicto argument?
The key lies in my passing remark, "given the truth of its
discerning premise." The question is this: What on earth would
"give" us this truth? On what basis may we assert the first
premise of the de re argument? We seem in urgent need of an
epistemological basis for our argument, of premises that are not
only true but also known for certain (known a priori, known by
sheer reflection, etc.).
This is where the de dicto argument becomes appealing. The
hope is that by not operating directly on objects and properties
but on our representations, it provides media known to us in a
special way. Granted, the argument would get us to the bottom
line only in a roundabout way. But, in the end, the price of
the complexity seems negligible compared with the gain of a
transparent epistemological basis. I believe it is this kind of
reasoning that leads Descartes, and three centuries later,

Kripke—both illustrious de re metaphysicians—to fall back on
the de dicto argument.
Let us trace the epistemological gains promised by the de
dicto argument. Consider Kripke's main de dicto premise,
( K i ) — " I t is possible that pain Ф FCF"—is true. De-representa
tion rather than de-object as the claim is, in itself it still pro
vides no special epistemic basis for assessment. We may well
wonder why "taking" the sentence "Pain Ф FCF" to another
"possible w o r l d " and evaluating it for truth "over there" is any
more epistemically transparent than asking directly, here
in actuality, whether this phenomenon, pain, bears the modal
property possibly exists without FCF} I n both cases, it seems
that we are involved with a primitive modal "fact," that thus
and such is possible.
The de dicto arguer may retort that we think our access
to evaluations of sentences ("dicta") in possible worlds is mark
edly more intimate than to modal properties of objects here in
actuality. A reason to think so is provided by Kripke's bringing
into de dicto evaluations the faculty of the imagination. I n
speaking of the evaluation of sentential representations in possi
ble worlds, he speaks of trying (succeeding or failing) to imag
ine a situation in which the sentence is true or false. I f a repre
sentation "a is F" is imaginable-by-the-mind, Kripke is happy
to pronounce it possible for a to be F; i f not, not.
We seem en route to justifying the dream ticket. We have
on hand a logically sound argument (by means o f possibility
evaluations in "worlds" and the reflection of this onto actuality
by rigid designators) sustained by a sound epistemology—the
imagination providing the access to these evaluation worlds. I n
all, this is a swift victory over a tricky dilemma.
Seeming to Imagine vs. Really Imagining
The foregoing may be a favorite of many modern metaphysi

cians; it may even seem to be Kripke's own account, but it is
5o W H A T A M I?
not his real account. Kripke's notion of possibility (necessity)

is based on many a posteriori (empirical), actual facts. For ex-
ample, it is because water is actually made of oxygen that it is
necessarily so; because gold actually has atomic number 79, it
is necessarily so; because Madonna and Francesca are actually
one and the same woman, they necessarily make one woman
only. I n all such cases, the delimitation o f possibilities succeeds
the facts of actuality. On the other hand, it surely seems to be
easy to imagine, as Kripke himself admits, that water exists
without oxygen, that gold is a compound, and that Madonna
and Francesca make two distinct women. I f imaginability were
to be our guide, Kripke's notion of "metaphysical" possibility
would be all but lost.
And so, in spite o f the initial seeming priority of imaginabil-
ity over possibility, Kripke's real account runs the other way
around—water is not really imaginable without oxygen because
it is not possible for water to exist without oxygen. Real imagi-
nability about an object x is dependent on and posterior to what
is really possible for x. Thus, real imaginability has now be-
come as epistemically nontransparent as real possibility. The
victory over our dilemma was too swift.
Kripke contrasts real imaginability with another notion, seem-
ing imaginability. Seeming imaginability is meant to be closer
to our first-impression intrinsic judgments, prior to subsequent
empirical information about identity, biological origin, and
chemical structure. W i t h seeming imaginability we may regain
a transparent epistemology, but at a price: A gap has now
opened between what is seemingly imaginable about an item x
and what is really imaginable about it. I n sum: Whereas real
imaginability is determined by real possibility, seeming imagi-
nability determines merely seeming possibilities.
Kripke never quite offers a canonical theory of what it is we
relate to when we seem to imagine this table (water, gold, etc.).
A t least two informal suggestions are made. The first is that to
seemingly imagine x, say, water, is to really imagine the seeming
of water, to relate to the sensation (look, feel, etc.), the phenom-
enological presentation, associated with water. This would

make the object of seeming imagination shift from the external,
real-world item—the wooden table or H 0 — t o an internal,
2
phenomenological quality, the how-it-seems. 4
A second suggestion, also developed with great force by

Kripke, runs like this: When we say that we seem to imagine
water without oxygen, we mean that we have really imagined
another substance that is qualitatively indiscernible from real
water. Thus, he who says, that for example, he seems to imag-
ine that this wooden table is made of ice is really asserting that
another object, looking like this table, is really imaginable (and
indeed is really possibly) made of ice.
In either explanation of seemings, we now confront a d i -
lemma. T o get a logically sound argument, one concerning the
target object x (Madonna, the table, water, etc.), we can rely
only on real imaginability (and, in turn, real possibility). How-
ever, we have no irrefragable access to real imaginability facts.
On the other hand, we do know by sheer introspection what
we seem to imagine. But, alas, seeming-imagination facts do
not concern the pertinent target objects; involved are either
mere seemings or worldly objects that are distinct, even i f qual-
itatively indiscernible, from the target objects. And so, in our
search for a discernibility argument for object x (say, from j),
we run the risk of pronouncing a successful discerning on the
basis o f mistakenly identifying x with an indiscernible qualita-
tive twin t(x). Meanwhile, in reality, things run exactly the op-
posite way—χ is necessarily identical withjy; it is the qualitative
twin t(x) that is necessarily distinct from x (=_y). Kripke's ad-
monition is that when it comes to real possibility-imaginability,
nothing is quite what it seems.
14. See Saul Kripke, Naming and Necessity (Cambridge, Mass.: Harvard
University Press, 1980), pp. 144—152.
52 W H A T A M I?
Is Nothing Quite What it Seems?
I f Kripke's bottom line were literally that nothing whatsoever is

quite what it seems, he would have ended up where Arnauld's
objection left us. But Kripke's principle is a touch more re
stricted—nothing external is quite what it seems. This leaves us
with a realm where at least some things are what they seem.
One such item, submits Kripke, is the phenomenon of pain.
The nature ("essence") of real pain is its seeming, the
how-it-feels. Alternatively, something qualitatively indiscern
ible from pain is, for Kripke, pain all over again. Essential to
being this kind of thing, pain, is the having of the qualitative
characteristic.
Assume this much and then consider the hypothesis that I
have merely seemingly imagined pain. I f so, I have imagined
the seeming of pain or really imagined something qualitatively
indiscernible from pain. But then I have really imagined pain.
In this special case, and perhaps with other essentially quali
tative items, we may successfully base our de dicto possibility
on transparent imaginings after all. We seem to imagine "Pain Ф
FCF"; thus, we really imagine "Pain Ф FCF"; and thus "Pain Ф
FCF" is possibly true—hence, actually true. 15
15. I have presented Kripke's argument in a stronger way than he in

tends. His apparatus was made to "positively" assert that it is possible that
pain is distinct from FCF and, thus assert that it is actually distinct from it.
Taken more de jure, Kripke merely makes in his work the following nega
tive point: The exposure of certain scenarios (e.g., that water is distinct
from H 0 ) as mere seeming possibilities does not extend to the case of
2
"Pain is distinct from FCF." I have taken the extra step here of making
Kripke assert that given the failure to expose an illusion, the seeming possi
bility is a real possibility.
I should point out that I stick to the role of an expounder of Kripke's
views on pain. I take Kripke to be proposing that its phenomenological
quality exhausts its list of necessary features. For myself, I have doubts
about this on Kripkean grounds. It is Kripke who inspires us, in Naming
and Necessity, to depart from Hume's famous doctrine that there are no nec
essary connections between distinct existences. To the contrary, everything
real is necessarily connected to, or dependent on, other real things. Notable
Three Morabfrom Kripke's Result
I draw three morals from this discussion of Kripke. The first

concerns the general question of relying on the imagination in
discernibility arguments; the second, the special case o f pain;
the third, any extension to Descartes's mind.
ι. Imaginability. I n the general case of a discernibility argu-
ment over an arbitrary object x, the accessible mechanism of
imaginability offers no solace. We cannot rule out a gap be-
tween seeming to imagine x and really imagining x. What is
more, the question of whether, in some special cases, we are
gap-immune depends on prior facts about the nature (whatness,
essence) of x: I f all there is to what x is is how it seems, we
are gap-immune; i f not, not.
And so, in the general case, we cannot do better fhan
our initial de re discernibility (possibility, conceivability) argu-
ments, with their "x is F" e.g. x is possibly (conceivably) G,
subject-predicate structure. Either we have (i) the real subject
χ but with no transparency of its bearing of the predicate F, or
among such connections is the mndamentality of its dependence on its genera-

tors, the items and processes without which the real item in question would
not have come into history. I take pain—and other mental phenomena—to be
as real as physical constituents of the world. And thus, I don't see my pain at
6 A . M . as just mysteriously "jumping into existence." If, as a matter of causal
fact, it was the firing of FCF in my brain that generated my pain, i f indeed
every pain event turns out to be generated by an FCF firing, then I would
be very inclined to argue that there is no real possibility of pain unless FCF
brings it about. The phenomenological quality of pain would still be distinct
from the FCF that brings it about, and the two would be of different natures
(essence). But once an FCF event occurred, of necessity pain would be felt. I
have discussed the dependence of all reals on other reals in my own work;
see, for instance, "The What and the How I , I I , " op. cit. Among others in
the contemporary literature who have expressed similar criticisms of Kripke's
views (while acknowledging some of his critical intuitions) are Tom Nagel,
"Conceiving the Impossible and the Mind Body Problem," Philosophy (1999)
and Torin Alter, "Essence, Necessity and the Mind Body Problem," manu-
script, University of Alabama, 1999. In my discussion of Kripke here, I
bracket away my own views about pain.
54 W H A T A M I?
else we have (ii) the transparency of the applicability of the

predicate F but not to the real subject x.
2. Pain. What closes the gap for seeming imagination in the
special case of pain? Kripke turns to prior intuitions about es-
sence (whatness, nature): W i t h pain, though not, say, with wa-
ter, the seeming of the phenomenon is what the (essence of
the) very phenomenon is. This assertion about the essence
(whatness, nature) of pain and water is not, in turn, dependent
on what is imaginable. Our explanation has reached rock bot-
tom. Our success in imagining x proper is explained in terms
of what χ is, not the other way round.
3. Extension to Mind. Is Kripke's principle for this one kind
of mental phenomenon, pain, extendable in general to the men-
tal, for example, to discernibility arguments about the mind in
general?
Two remarks are in order at this point. First and to reiterate,
even i f Kripke's pain strategy were extendable to the mental in
general, it would only emphasize the priority of whatness facts
over imagination facts. We would have to argue that our seem-
ing to imagine Descartes's mind, or a certain thinking thing x,
is really imagining x. Here, as elsewhere, it is the essence of the
mind that would determine whether it has really been imagined
(conceived).
This brings us to the second remark. Our question is this:
Kripke emphasizes in the case of pain the essentiality of the
purely internal qualitative character. The principle may extend
to other mental experiences. Is it extendable to everything
mental?
We must separate two kinds of items in the allusion to "ev-
erything mental." On the one hand, we may be concerned with
subjects, what Descartes refers to as "substances." So D M is
such a subject, and the man Rene Descartes is another such
subject (whether, for Descartes, it is indeed a substance will be
discussed in chapter 2). O n the other hand, we may be con-
cerned with mental predicates, for example, "is thinking about
water," "is thinking that water is wet," and "is intending to get
T H E REAL D I S T I N C T I O N 55
some water." I take it that Kripke's discussion of pain was

focused on the kind of predicate it is, that is, what is involved
in applying "is in pain" or "feels pain" to some subject, say,
the man Jones. 16
The extension of Kripke's pain strategy to mentality-involv-

ing subjects—be it the mind D M or the man Descartes—invites
major questions that have to do with according these items the
status of "substances." Chapters 2 and 3 are dedicated to this
issue, and it is within this context that imaginability of such
subjects will be discussed (see, for instance, the sections on con-
ceivable separability and conceivability illusions in chapter 3).
But even before a thorough analysis, it is quite clear that the
extension of the pain thesis to such subjects is very problematic.
One way to realize the problem is this: Surely, i f Kripke's thesis
is not extendable to mental predicates such as "thinking about
water" and "thinking that water is wet," it will not extend to
the mental subjects who entertain such thoughts. I f the mental
states of such subjects are not exhausted by their phenomeno-
logical character, it may be argued that the subjects that bear
those states do not have their identity so exhausted. ( I do not
argue this point here). A n d so, I will concentrate here on the
most likely extension of Kripke's thesis—from experiential
predicates to mental predicates in general. 17
16. Kripke's discussion is rather informal on this matter. It is surely pos-

sible to read his remarks as applying to some sort of (higher order) subject,
a certain species of mental states, in the way we can view, for example, t i -
ger as a species of biological organisms. I , for one, (in "The What and the
How" and "Nothing, Something, Infinity") so view tiger, water, and other
worldly kinds, and I would so view pain, thirst, and so on. But the context
of Kripke's discussion makes it quite likely that his remarks on "states" (of
pain) are directed at a theory of mental properties such as "is in pain." The
problems I mention below about extending Kripke's view to the mental in
general do not depend on whether we read him, with regard to pain, as
predicate- or subject-oriented.
17. As mentioned, (illusions of) conceivability of subjects like D M and
the man Descartes will occupy us in chapter 3.
56 W H A T A M I?
From Descartes's perspective, Kripke's focus (i) on sensa-

tions, as opposed to thinkings, and (ii) on the mechanism of
imagination, as opposed to conceiving, is doubly ironic. Des-
cartes took both Kripke's mental targets and the scanning tech-
nology as exactly those that are not distinctive of the mind
proper. Rather, they show essential dependence on the body
(and the brain in particular) and its perceptual powers. But let us
not give up on the extension of Kripke's strategy just because
Descartes saw a great gulf between perceptual (imaginative)
states and purely intellectual states. Let us ask in a neutral vein
whether the extension from "is in pain" to, say, "is thinking
that water is wet" is feasible.
Consider, then, taking experiences, intentional action states,
and thinking states (to cite three categories of mental states) as
fundamental to a human mind. The critical point here is that
right away it is far from obvious that intentional action and
thinking states are exhausted by their purely qualitative, "phe-
nomenological" character. Contemporary philosophy has stressed
quite the opposite about such cognitive states. Standard exam-
ples in the philosophy of mind allude to two agents, Castor
and Pollux, that are in the same phenomenological state—for
example, enjoy in their heads the same "seeming of water"—
but who nonetheless entertain two different thoughts. I t is con-
cluded that the two are in different thinking (types of) states.
This is so simply because Castor interacted, solely on this
planet, throughout his life with liquid H 0 whereas Pollux,
2
confined to some other planet, interacted only with the qualita-

tively similar but chemically distinct liquid X Y Z . I n a similar
vein, Castor may form the intention to drink some water but
Pollux may form quite a different intention. This much seems
true in spite of their common, inner qualitative feel (say, the
feel of thirst).18
18. (N.B.: This footnote involves "technical" matters in "advanced" phi-

losophy of language.) The intuitions I mention have been emphasized by
Tyler Burge in many writings—see, for example, "Other Bodies," in
The preceding intuitions about the identity of thoughts (in-

tentions) and thinking states have been supported by various
arguments. Those who would like to extend Kripke's pain-
strategy w i l l have to refute the arguments. On the other hand,
i f we accept them, as I am inclined to, the extension of Kripke's
thesis is blocked.
Suppose that two agents—myself, here on earth, and some
twin, Max, on the other planet—engage, respectively, in imag-
inings of Castor and Pollux. I imagine Castor thinking that
water is wet. Max imagines, as he would put it in his dialect,
that Pollux is thinking that water is wet. Asked to enlarge on
our experiences, both o f us may describe an agent, in the pose
of a thinker, and in his head visions of a transparent liquid,
Thought and Object, ed. A. Woodfield (London: Blackburn, 1982). In my

own view, an even more radical thesis about cognition is defensible in
which many thought (and other cognitive) states are essentially relational,
relating agent and worldly item. This is not to be confused with another
Burge claim (e.g., in his "De re Belief," Journal of Philosophy 1977) that
many thoughts are essentially de re. Burge uses "de re" for what are really
some special de dicto thoughts, thoughts involving quasi-demonstrative rep-
resentations. As I read it, Burge's work is part of the classical project of re-
ducing what I call relational thoughts to de dicto thoughts, although those
that involve quasi-demonstrative representations. A locus classicus of this
project is David Kaplan's "Quantifying I n , " Synthese, 1968. Burge's point is
to amend Kaplan's initial reduction by giving, in the de dicto language, rep-
resentations that are not purely qualitative but quasi demonstrative. In con-
trast, here I mean by "relational" or "essentially de re" thoughts essentially
of the very object—thoughts with a direct, representation-free relation, bind-
ing the thinker Castor and his thought object, this table, the substance wa-
ter, and so on. This view denies any reduction to a de dicto form in which
the agent relates to a dictum (thought, proposition). The view I support
may seem natural when we report acts of reference to and visual percep-
tions of the table, as reported, for example, by "Castor sees (refers to) the
table." I extend it to our reports of thoughts of (beliefs about) the
table. I have written, in a technical vein, on such matters in "The Subject
Verb Object Class I , I I , " in Philosophical Perspectives, 1998 and "The Lan-
guage to Describe Our Cognitive Lives" (to appear). Here I do not rely on
my own stronger theses. The point in the text rests on Burge's now famil-
iar view about the individuation of de dicto thoughts.
58 W H A T A M I?
pouring down from the sky, and so on. The phenomenological

description would be very much the same. We seem to imagine
the same of Castor and Pollux. D o we really imagine the same?
I imagine Castor thinking one thought—that H 0 is wet. Max
2
is imagining Pollux thinking a distinct thought—that X Y Z is

wet.
The gap between seeming to imagine and really imagining
has reopened because the nature of the objects—here the
thoughts entertained by Castor and Pollux—is not exhausted
by their phenomenological character. When what we conceive
(imagine) is not purely phenomenological, seeming to imagine
a target is not guaranteed to be a real imagining of it.
In more general terms, the following moral emerges: What
has turned out to be the fundamental fact is not some imagina-
tion exercise but a whatness fact, that is, the nature of thinking
states, as opposed to pain states. A n d so, I conclude the appen-
dix very much in the way I concluded the text before the ap-
pendix. What we need primarily is not so much a dissection of
the workings of imaginability and conceivability. What we need
is more reflection on the nature of the objects about which we
have these imaginings (conceivings). We need more insight
into the nature of the three essential Cartesian factors—the hu-
man mind, the human body, and the human being. This is the
task of chapter 2. 19
19. I have not taken final leave of the question of conceivability. I go

back to it in some detail, after the nature of the human mind, body, and be-
ing has been analyzed. See, for instance, chapter 3.
TWO
Integrative Dualism
2.i. The Prove T o o Much/Prove T o o

Little Dilemma
In chapter ι I argued for the real distinctness o f the human

mind and body, specifically Descartes's D M and D B , under
four assumptions:
(i) Complete subjecthood. D M and D B are complete subjects.

(ii) Modal inseparability. D M and D B necessarily coexist.
(iii) Conceivable existential separability. I t is conceivable of
D M (describable in a scenario consistent with what D M
is) that it exists without D B .
(iv) Whatness separability. D M and D B make different kinds
of things.
However, as really distinct as D M and D B may be, we are only

halfway through. As I have emphasized from the beginning of
this book, the point of the two-stage Cartesian conception of
man is this: The real distinction of the human mind and body
is to be established so that it can next be embedded in a genuine
subject—the real man, the human being whose mind and body
they make.
As I read Descartes, it is only upon the completion of the
full two-stage project that he would feel he had succeeded. But,
alas, here precisely lies our predicament—the separate demands
of the two stages pit them against each other. We seem to be
prisoners of a dilemma I have called, following Descartes's own
59
6o W H A T A M I?
t e r m i n o l o g y i n his reply t o A r n a u l d , the prove too much/prove

too little dilemma.
L e t us recall the w a y i n w h i c h the d i l e m m a is generated. I f
w e make the separability o f D M and D B strong enough t o
count as a real distinction, w e p r o v e " t o o m u c h . " W e assert a
real possibility o f disembodied existence for D M and thus a
strong sense o f self-sufficiency. W h e n w e n o w p u t this one self-
sufficient b e i n g together w i t h another such self-sufficient b e i n g ,
the b o d y , w e no longer get a real natural u n i t y — a f u l l - b l o o d e d
h u m a n subject—but a " l a t e - i n - t h e - d a y " u n i t y o f c o m p o s i t i o n , a
derivative c o m p o u n d . T h i s is the h o r n o f p r o v i n g t o o m u c h .
T o amend things, w e m i g h t start b y carefully p r o t e c t i n g the
p r i m a l i t y o f the full h u m a n being. B u t whatever subsequent
separation b y conceivability experiments w e accord t o the m i n d
and b o d y , i t is n o t enough t o establish t h e m as distinct complete
subjects that can exist w i t h o u t the other. T h i s is the h o r n o f
p r o v i n g t o o little.
A d i l e m m a and, i n its w a k e , a w o r r y — i n d e v e l o p i n g Des-
cartes's conception o f the real m a n , something i n o u r quartet
o f assumptions f r o m chapter i has g o t t o give. O f course, i t
w i l l n o t be the assumption o f m o d a l inseparability. B u t the
three other assumptions seem vulnerable.
First, whatness separability is i n some danger: W i l l w e still
stick with the rather generic essence assignment—merely
" t h i n k i n g " and " e x t e n d e d " — t o the h u m a n m i n d and body?
M i g h t w e n o t have to m e n t i o n that w h a t they are is i n t r i n s i c a l l y
the m i n d and b o d y of a h u m a n being? I n l i k e manner, m i g h t
w e n o t have t o give up—as against ( i i i ) — a n y existential separa-
b i l i t y between m i n d and body? O f course, w e have already
d e n i e d — w i t h m o d a l inseparability—that there is any real possi-
b i l i t y o f D M existing w i t h o u t D B . B u t n o w that w e contemplate
articulating w h a t each is i n terms o f the h u m a n b e i n g whose
m i n d and b o d y they are, w i l l w e n o t have t o deny even the
weaker sense o f conceivable existential discernibility? F i n a l l y ,
the darkest w o r r y o f t h e m all: A s w e are bent o n stressing the
p r i m a l i t y o f the real m a n Rene Descartes ( R D ) , w i l l w e n o t
INTEGRATIVE DUALISM 61
end up w i t h ( i ) refuted? Descartes's m i n d and b o d y w i l l t u r n

out t o be mere modes o f this m a n . G o n e w i l l be the claim o f
D M and D B t o be t w o complete subjects.
2.2. Separatist D u a l i s m : F o u r Basic A n n o t a t i o n s
A d i l e m m a and, i n its w a k e , m a n y w o r r i e s — n o n e o f this gets

o f f the g r o u n d i n the most popular reading o f Descartes, sepa
ratist dualism. T h i s classical dualism contrasts w i t h the t w o -
stage picture that appears b e l o w under the title " I n t e g r a t i v e
D u a l i s m . " I t is t i m e t o lay o u t the m a i n tenets o f classical sepa
ratist dualism so that w e are i n a p o s i t i o n t o contrast i t w i t h
the integrative alternative.
A c c o r d i n g t o the separatist i n t e r p r e t a t i o n , we were i n a p o s i
t i o n t o call i t game, set, and match b y the end o f chapter i .
Just t o p r o v e the real d i s t i n c t i o n is the task o f Descartes's d u a l
i s m , and this is w h a t w e claimed t o have accomplished. H o w
ever, i t n o w turns o u t that quite a few annotations need t o be
added to the p r o o f , each glossing Descartes's understanding o f
the k e y notions i n v o l v e d : ( i ) the idea o f a real, complete subject
(substance)—both the m i n d and the b o d y are meant t o be
such—and ( i i ) the idea o f the k i n d o f distinction i n v o l v e d w h e n
the relata are real subjects, a real distinction, so that D M and
D B can each exist w i t h o u t the other.
Separatist Dualism I: Existential Separability
Separatist dualism demands that w e read "can exist w i t h o u t the

o t h e r " as a f u l l - b l o w n , real p o s s i b i l i t y — t h a t is, i t is really possi
ble for D M t o exist w i t h o u t D B ; i n t u r n , i t is really possible
for D B t o exist w i t h o u t D M . A real d i s t i n c t i o n requires the real
possibility o f separate existence. H o w e v e r , i n chapter ι, I ar
gued that w e m a y grant A r n a u l d his m o d a l o b j e c t i o n and as
sume m o d a l inseparability o f m i n d and b o d y . " C a n exist w i t h
out the o t h e r , " I assumed, was weaker, d e m a n d i n g mere
62 W H A T A M I?
c o n s i s t e n c y - w i t h - w h a t - D M - i s rather t h a n a real possibility for

D M . I n this respect, the f r a m e w o r k o f chapter i is an attenuated
f o r m o f separatist dualism: w e need n o t assume that conceiv-
ability o f D M w i t h o u t D B entails a real possibility o f D M w i t h -
out D B ; w e o n l y assume that the conceivability entails the co-
herence o f a story ( m o d e l ) w i t h D M existing w i t h o u t D B .
Attenuated as o u r assumptions o f chapter i are, the frame-
w o r k is that o f separatist dualism: I n some " l o c u s " (a conceived
story) respectful o f w h a t D M and D B are, D M exists without
D B . I t is i n this last that an essential feature o f separatist d u a l -
ism lies: T h e separation o f D M and D B rests u l t i m a t e l y o n
some such existential separability.
Separate Dualism II: Generic Essences
Existential separability is one fundamental feature o f separate

dualism. A second is the development o f a symmetric account
o f the independence o f m i n d and b o d y .
A s stated i n chapter i , i n M e d i t a t i o n I I (as i n the treatise
that underlies its metaphysics m o r e theoretically, The Principles
of Philosophy), there is a general tendency t o specify v e r y ab-
stract and general whatnesses (essences). T h e h u m a n b o d y ends
up h a v i n g as its nature—just l i k e a stone, the L o i r e , or the
Sun—the p r i n c i p a l attribute o f b e i n g extended. I n a symmetric
manner, the h u m a n m i n d is assigned this one p r i n c i p a l a t t r i -
b u t e — i t is a t h i n k i n g t h i n g . A n d so, even t h o u g h the subtitle
o f M e d i t a t i o n I I is " O n the N a t u r e o f the H u m a n M i n d , " b y
the t i m e Descartes articulates the essence o f this h u m a n m i n d ,
he does n o t m e n t i o n that this m i n d o f his is of a h u m a n b e i n g
rather than o f a M a r t i a n or an angel or a g o d .
T h i s generic approach t o the whatness o f D M and D B calls
for a finer f o r m u l a t i o n o f one o f o u r principles above, ( i v ) ,
whatness separability. W e should say that the separatist frame-
w o r k presupposed b y o u r argument i n chapter i asserted i n
effect t w o related principles concerning w h a t D M and D B are:
(iv) Whatness separability: W h a t D M is is distinct f r o m

w h a t D B is.
(v) Generic whatness: T h e articulation o f w h a t D M and
D B are—respectively, t h i n k i n g and extended—is b y
means o f generic, k i n d - b l i n d attributes.
Separatist Dualism III: No Endurance Theory
T h e genericity o f essence has so far s h o w n i n the absence o f .

interkind distinctions. F o r example, w e find i n M e d i t a t i o n I I
there is n o separation i n the essence o f h u m a n , angelic, and
d i v i n e m i n d s ; n o r is there an intrinsic d i s t i n c t i o n a m o n g h u m a n ,
d o g , and stone bodies. B u t the g e n e r i c i t y o f essence strikes a
second t i m e , this t i m e at the intrakina level. Descartes's m i n d ,
D M , is one h u m a n m i n d ; Plato's m i n d , P M , is another h u m a n
m i n d . A s k e d to say w h e n t h e y existed, w e w o u l d answer that
the t w o existed at different times. O f course, w e m a y (perhaps
Descartes d i d ) have a theory a c c o r d i n g to w h i c h b o t h always
existed and, for that matter, b o t h still d o . B u t I have n o t been
asking about Descartes's t h e o r y , just about the natural answer
to the question asked.
G i v e n the disparate periods o f existence, i t is arguable ( I ,
for one, w o u l d be happy t o argue) that D M c o u l d n o t have
existed at the time P M actually d i d ; i n t u r n , P M c o u l d n o t have
existed i n the first h a l f o f the seventeenth century. W h a t is
m o r e , i t m a y be argued that the manner and time i n w h i c h D M
came i n t o b e i n g — b y the c o m i n g i n t o b e i n g o f the man Rene
Descartes—constrained the w a y i n w h i c h i t was possible f o r
D M t o change while i n existence, d u r i n g the 54 years i t d i d
exist between 1596 and 1650. I t c o u l d thus be argued that D M
c o u l d n o t have gone t h r o u g h its existence, e n d u r i n g t h r o u g h a
lifetime, i n the manner actually gone t h r o u g h b y P M , itself
b o u n d t o the vicissitudes o f that в . с . m a n Plato. I n a nutshell,
different h u m a n m i n d s have essentially different possible lives,
different possible histories. N 0 explanation o f such differences
seems f o r t h c o m i n g f r o m the generic essence t h e o r y , w h i c h as-
6 4 W H A T A M I?
signs the same schematic essence—thinking t h i n g — t o b o t h D M

and P M .
W h a t exactly is the problem here? First, a w o r d about what
problem is not raised b y these endurance-in-time issues: I ' m not
w o r r i e d here about "criteria o f i d e n t i t y " for D M and P M . I stress
this simply i n v i e w o f the dominance i n contemporary discussions
o f what are called "personal i d e n t i t y " theories. T h i s vocabulary
w i l l make many readers surmise that I am lamenting here the
absence o f some such theory, multiplied b y three—a theory o f
m i n d identity, b o d y identity and p e r s o n a l ( m a n ) identity.
But this is n o t so. I am n o t d r i v e n b y a quest for " c r i t e r i a
for i d e n t i t y " — b e they for m i n d s , bodies, or persons—a quest
that dominates the practice o f personal i d e n t i t y theories. Q u i t e
the opposite: I v i e w i t as an essential m a r k o f e v e r y t h i n g real
that i t is governed b y Bishop Butler's m a x i m that e v e r y t h i n g is
w h a t i t is and n o t another t h i n g , surely n o t a "definable t h i n g , "
somehow reducible t o its " c r i t e r i a o f i d e n t i t y , " "necessary and
sufficient conditions o f i d e n t i t y , " and other such i n d i v i d u a t i o n
blueprints for b e i n g the distinct i t e m o n e — m i n d , b o d y , or
person—is.
F o r Descartes, h u m a n minds and bodies (as w e l l , as w e shall
see, h u m a n beings) are real subjects. A s such, I believe that
they escape r e d u c t i o n t o criteria, definitions, and so o n . I also
believe that i f Descartes were a m o n g us and aware o f the d o m i -
nance o f such theories o f criteria and defining conditions for
real things, he w o u l d v e r y m u c h insist o n the mdefinability o f
any real being s i m p l y because i t is a real being.
So, m y w o r r y about the e n d u r i n g D M and P M is n o t d r i v e n
b y the quest for criteria o f i d e n t i t y . Rather m y w o r r y is that
separatist d u a l i s m — w i t h its generic essence assignment—forces
a disjointed treatment o f t w o questions that call for a unified
account. T h e t w o questions concern ( i ) the whatness (essence)
o f a subject and ( i i ) its endurance i n t i m e .
W e l o o k e d at the first question, the whatness articulation
p r o b l e m , i n some detail i n chapter i w h e n w e asked repeatedly
about a variety o f items: " W h a t is this t h i n g ? " T h e second
question, w h i c h has n o t so far been broached, concerns the

endurance i n t i m e o f r e a l subjects. I call i t the lifetime descHption
p r o b l e m . E x i s t i n g i n t i m e — c o m i n g i n t o h i s t o r y , changing i n
i t , and c o m i n g to an end—is the m a r k o f a real subject. Given
such a subject, o u r metaphysics needs to describe w h a t i t
t o o k for this distinct t h i n g to come to be, w h a t sustained its
changes, and w h a t i n the w o r l d had to happen for i t to go o u t
o f existence.
Descartes was v e r y m u c h concerned w i t h lifetime descrip-
tions o f subjects. F o r example, i n M e d i t a t i o n I I he provides a
most abstract and generic discussion o f the essence o f various
items, i n c l u d i n g a certain piece o f wax. B u t even i n this abstract
and generic setting, he goes o n to speculate n o t just about w h a t
this w a x is b u t also about w h a t is i n v o l v e d i n its " r e m a i n i n g the
same w a x , " w h a t is i n v o l v e d i n describing the wax's lifetime.
T h e trouble is that separatist dualism and its generic essence
t h e o r y o f m i n d and b o d y — t h e former is t h i n k i n g ; the latter is
extended—tells us n o t h i n g m u c h about the lifetime description
p r o b l e m . F o r example, the t w o h u m a n minds D M and P M have
different histories and endurance conditions, b u t n o t h i n g i n the
generic essence t h e o r y explains these differences—a silence that
is even more t r o u b l i n g i f w e t h i n k o f another m i n d , one that is
internally similar to D M (e.g., a bundle o f similar intellectual
contents, o r t h o u g h t s , as those had b y D M ) . C a l l the m i n d
D M * and make i t exist i n the b o d y o f a c o n t e m p o r a r y person,
a m a n w e shall call C. O b v i o u s l y , D M * is an intellectual t w i n
o f that 400-years predecessor, D M . H o w are w e to explain the
n o n i n t e r n a l differences i n the lifetimes o f these t w o minds?
Similar worries m a y be raised for a c o n t e m p o r a r y t w i n b o d y
( D B * ) o f t h e 1651 dead-and-gone, o r i g i n a l D B . N o t h i n g i n their
c o m m o n generic essence tells us about their distinct lifetime
profiles.
T h i s , t h e n , is the concern about generic separatist dualism,
a w o r r y w e shall attend to i n some detail. A t the m o m e n t , let
us r o u n d up the exposition o f separatist dualism b y f o r m u l a t i n g
this one last principle:
66 W H A T A M I?
(vi) N o endurance (lifetime description): T h e t h e o r y o f ge

neric essences ( v ) passes i n silence over the endurance
i n t i m e o f D M and D B . 1
Separatist Dualism IV: The Notion

of Substance
T h e p r o v e - t o o - l i t t l e h o r n o f o u r d i l e m m a brings t o the fore a

fundamental feature o f separatist dualism—its understanding o f
the idea o f substance. T h e p r o v e - t o o - l i t t l e w o r r y is this: Sup
pose w e assume the p r i m a l i t y o f the Human b e i n g and regard,
i n t u r n , the u n i o n between his m i n d , D M , and his b o d y , D B ,
as i n the v e r y nature o f such a m i n d and b o d y . T h i s is so
because the v e r y nature o f each is n o w to be the m i n d and
b o d y o f that p r i m a l l y g i v e n h u m a n being. I t seems that w e are
o n the verge o f m a k i n g the three o f t h e m — h u m a n m i n d , b o d y ,
and being—three existentially interdependent substances. B u t t o
those familiar w i t h classical glosses o f the n o t i o n o f substance—
assumed b y separatist dualism—this w i l l seem t o be a contradic
t i o n i n terms: H o w can existentially interdependent items c l a i m
the rarefied title o f a "complete t h i n g , " "substance," o r "ens
per se"? Is i t n o t part o f the v e r y definition o f a substance that
it enjoys existential independence f r o m all other beings (per
haps excepting the creator o f t h e m all)?
T h e r e is no d o u b t that Descartes speaks o f "substance" i n a
separatist v e i n . B u t as prevalent i n his w r i t i n g s is quite another
ι. Let me add here a general remark, going beyond our strict Descartes
agenda. The separation between personal identity problems and lifetime
metaphysics questions strikes me as of general use. Personal (ship, bicycle,
etc.) identity theories fuse in my view (i) atemporal (pseudo) issues about
individuation and defining criteria with (ii) genuine lifetime (endurance)
problems for individuals (with a given, primitive identity). In many of the
genuine questions raised in personal identity puzzles, I flnd a quest for an
explanation of the limits in the change potential of a specific item (Rene
Descartes, the Titanic, etc.) of a specific kind (a person, a ship, etc.). More
on the methodology of personal identity theories arises in chapter 3, in our
discussions of illusions of conceivability.
conception o f substance, and i t is i m p o r t a n t t o understand the

difference. A s t r i k i n g encapsulation o f this d u a l i t y shows itself
i n Descartes's v a c i l l a t i o n i n his response t o A r n a u l d ( C S M I I ,
156—160), a text already discussed i n detail t h r o u g h o u t chapter
ι. 2
I n the space o f four pages ( C S M I I , 156—160), the separatist
idea shows up once and then v e r y crisply: " T h e n o t i o n o f sub
stance is just t h i s — t h a t i t can exist b y itself, that is w i t h o u t the
aid o f any other substance."
T h i s is stark enough. I w o u l d l i k e t o refer t o the separatist
idea o f substance as the existential conception. W e test for real
subjecthood, because this is w h a t real subjecthood consists of,
b y checking the existential profile o f an i t e m : C a n i t o r can i t
n o t exist all b y itself?
T h e potential for solitary existence is qualified w h e n D e s
cartes acknowledges the existential dependence o f pure sub
stances o n G o d , their creator. B u t even w i t h this qualification,
the separatist's gloss o f "substance" remains essentially existen
t i a l — a sine qua n o n for an item's b e i n g a substance is its exis
tential independence f r o m every other t h i n g b u t the d i v i n e . 3
I n the v e r y same pages, w e encounter another, existence-free

n o t i o n o f substance. T h e idea is that o f ( i ) a subject o f predica
t i o n ( i i ) i n w h i c h inhere attributes (modes), b u t ( i i i ) i t itself
does n o t exist in a subject. T h i s approach emphasizes w h a t I
call categorical facts about the candidate, facts r e l a t i n g t o its
profile i n a " l o g i c a l g r a m m a r " o r i n a t h e o r y o f categories
( t y p e s ) — i t is the subject o f predication and itself is n o t one to
2. Concerning terminology: As we saw in the discussion of complete

ideas in chapter 1, Descartes uses the two notions "substance" and "com
plete thing" interchangeably (see CSM I I , 156, quoted in chapter 1). Even
though I quote extensively in the section that follows, the reader may flnd
it useful to have a copy of the fourth replies (CSM I I , 156—160).
3. The only other place in these pages where the existence-involving
notion is evoked is a bit later (CSM I I , 160), when Descartes says in the
arm-body passage already quoted: "And saying that the arm belongs to the
nature of the whole man does not give rise to the suspicion that it cannot
subsist in its own right."
68 W H A T A M I?
be borne b y subjects; i t cannot be said t o exist in subjects. I

refer t o this gloss o f "substance" as the categorical conception.
W h e n he introduces the idea o f "complete t h i n g , " Descartes
says:
B y a "complete t h i n g " I s i m p l y mean a substance en-

d o w e d w i t h the forms o r attributes w h i c h enable to rec-
ognize i t as a substance. W e do n o t have immediate
k n o w l e d g e o f substances, as I have noted elsewhere. W e
k n o w t h e m o n l y b y perceiving certain forms o r a t t r i -
butes w h i c h must inhere i n something i f they are t o ex-
ist; and w e call the t h i n g i n w h i c h they inhere a " s u b -
stance." ( C S M I I , 156)
So far, this n o t i o n o f substance is existence-free. Descartes goes

o n to apply this idea t o the m i n d - b o d y case:
Just as b e i n g extendible and divisible and h a v i n g shape

etc. are forms or attributes b y w h i c h I recognize the sub-
stance called b o d y , so understanding, w i l l i n g , d o u b t i n g ,
etc. are forms b y w h i c h I recognize the substance w h i c h
is called m i n d . A n d I understand a t h i n k i n g substance t o
be just m u c h a complete t h i n g as an extended t h i n g .
( C S M I I , 156)
T h i s is still free o f any allusion t o existential independence.

T h e k e y continues t o be the presence o f a subject i n w h i c h
certain attributes inhere, the subject itself n o t existing i n an-
other subject as w o u l d an attribute or a mode.
A t this p o i n t , Descartes reminds A r n a u l d that he, Descartes,
has already articulated this (categorical) n o t i o n o f substance
elsewhere. O n e such venue is the gloss ( " d e f i n i t i o n " ) i n the
geometrical exposition i n the end o f the second replies (CSM
II, 114):
E v e r y t h i n g i n w h i c h there exists i m m e d i a t e l y , as i n a
subject, or t h r o u g h w h i c h exists a n y t h i n g w e perceive
that is any p r o p e r t y , q u a l i t y , o r attribute o f w h i c h a real

idea is i n us is called substance. N o r do w e have any
other idea o f substance itself, precisely t a k e n , than that i t
is a t h i n g i n w h i c h f o r m a l l y or e m i n e n t l y exists this
something w h i c h w e perceive, that is, w h i c h is objec-
t i v e l y i n some o f o u r ideas, since i t is k n o w n b y the nat-
u r a l l i g h t that n o t h i n g can be a real attribute o f n o t h i n g .
(CSM I I , 156)
Descartes also refers A r n a u l d back t o the gloss i n the first re-

plies ( C S M I I , 86), where each o f the real distincts has t o be
" a n entity i n its o w n r i g h t w h i c h is different f r o m e v e r y t h i n g
else" ( C S M I I , 156).
A l l o f this glosses "substance" and " c o m p l e t e t h i n g " i n exis-
tence-free terms. T h i s , then, is the second idea o f Descartes. A
substance is a subject o f p r e d i c a t i o n , i n w h i c h w e recognize
b o t h the inherence o f a p r i n c i p a l attribute—specifying the v e r y
k i n d o f subject i n v o l v e d — a n d other properties (ways o r modes)
o f the subject that are cast, i n t u r n , as modes o f that p r i n c i p a l
attribute. T h e attributes—principal or modifications—inhere i n
the substance; the substance itself does n o t exist i n o r t h r o u g h
another subject. T h e substance is n o t a way o f b e i n g ; i t is the
being p r o p e r . 4
W e have before us the t w o conceptions, and I w a n t t o e m -

phasize this one difference between them: T o n o t exist i n a
subject does n o t i m p l y fhe potential o f solitary existence. I n the
categorical conception, a substance may be of necessity—
stronger y e t , b y its v e r y whatness—connected i n its existence
t o other substances. A l l w e are t o l d b y the categorical concep-
t i o n is that t h e y — t h e candidate substances—do n o t exist in each
other. T h e k e y t o complete subjecthood is n o t existential i n d e -
pendence b u t existence-as-a-subject and n o t in-a-subject.
4. Also in this vein are Descartes's remarks about "complete thing" to

Caterus in the first replies, CSM I I , 85—86.
70 W H A T A M I?
From Real Subjects to Real Distinctions
I n the categorical conception, we separate an existential real dis

tinction f r o m a real distinction simpliciter. A real distinction sim-
pliciter is a distinction between t w o real subjects. T h i s Descartes
w o u l d like to contrast w i t h a variety o f other distinctions, such
as between the subject and its modes or between t w o modes.
Suppose n o w o u r n o t i o n o f a real subject is existential. B o t h
D M and D B are, b y hypothesis, real subjects. So, inasmuch as
they are distinct at a l l , n u m e r i c a l l y distinct, i t f o l l o w s — o n sepa
ratist lines o f reasoning—that they are existentially separable
f r o m each other, that they c o u l d exist w i t h o u t the other. O f
course, this does n o t settle the question o f whether D M and
D B , i n the first place, make one t h i n g or t w o . B u t w e are
required to a d m i t as a necessary c o n d i t i o n o n their numerical
distinction that they are existentially independent; o f course,
existential separability is assumed t o be a sufficient condition
for numerical distinction. T h u s , the existential n o t i o n o f sub
stance engenders w h a t w e m a y call the idea o f existential real
distinction.
Suppose, o n the other h a n d , that o u r n o t i o n o f real subject
is based o n the categorical idea. Assume again that D M and
D B are real subjects. Assume further that they are n u m e r i c a l l y
distinct. N o w , first, i t is no longer required that they w o u l d be
existentially separable, whether i n a possibility or even merely
i n a conceived story. Second, there is nonetheless a sense o f
"real d i s t i n c t i o n " according to w h i c h the t w o m a y be really
distinct. T h e sense is d i r e c t l y connected t o the categorical sense
o f substance. T h e latter submits that t o be such a substance is
to be a subject o f attributes and n o t to exist i n a subject. T h e
correlate idea o f a categorical real distinction between x and y is
that χ and y are n u m e r i c a l l y distinct and each exists w i t h o u t
existing i n the other as i n a subject. A n d so, for all that was
said so far, i t is quite conceivable that the h u m a n m i n d , b o d y ,
and being are categorically really distinct f r o m each other w i t h
out b e i n g existentially really distinct.
2.3. Integrative Dualism: T h e Quest

for Symmetry
A feature o f separatist dualism emphasized above is its symmet-

ric treatment o f m i n d and b o d y : Just as the essence o f D B is
free o f any allusion t o the h u m a n b e i n g whose b o d y i t is, the
essence o f D M is also free o f reference t o the m a n whose m i n d
it is. I n m o r e general terms, the generic essence assignment is
free o f human involvement.
Integrative dualism casts itself at the outset as an opposite
m e t h o d o l o g y . T h e ur-fact t o be respected is that w e are dealing
w i t h a man's m i n d and b o d y . I refer to this basic fact as the
primality of the man. I believe this one switch vis-a-vis separatist
dualism is b e h i n d m a n y others. B u t fundamental as the p r i m a l -
i t y o f the m a n is t o integrative dualism, i t confronts us w i t h a
textual mystery.
Descartes's w r i t i n g s include an explicit integrative proposal
about the h u m a n b o d y b u t n o t a n y t h i n g as direct about the
h u m a n m i n d . A letter t o Mesland i n 1645, w h i c h w e shall be
l o o k i n g at, offers clear blueprints for an integrative account o f
the h u m a n b o d y . Some w i l l d o u b t its authenticity, w r i t t e n as i t
was i n a theological context. A s m e n t i o n e d at the outset, I am
n o t a historian and w i l l n o t engage i n detective-like psychoanal-
ysis o f Descartes's " r e a l " intentions. W h e t h e r he " r e a l l y " i n -
tended its contents o r n o t , the contents are crystal clear. A n d
this is the p r o b l e m — n o equally explicit, integrative account o f
the h u m a n m i n d is suggested b y Descartes.
W h e r e are w e t o l o o k then? I n d i r e c t indications o f such an
integrative account o f the h u m a n m i n d are sown i n t o D e s -
cartes's late essay Passions of the Soul. W h a t is m o r e , I argue
that some fundamental features o f Descartes's general meta-
physics o f substances—manifest already i n M e d i t a t i o n I I — l e a d
us sooner o r later, even i n the case o f h u m a n m i n d s , t o a t h e o r y
o f whatness and endurance i n t i m e that is i n the integrative
m o l d . A n d finally, since w e speak o f the letter t o Mesland, i t is
even possible that enough is said there about
72 W H A T A M I?
(Mi) the endurance i n t i m e o f objects i n general (e.g.,

rivers)
(M2) the role o f the m i n d i n the endurance o f the h u m a n
body
t o a l l o w us t o infer f r o m ( M i ) and ( M 2 ) an integrative account

o f the h u m a n m i n d . Such an argument f o l l o w s , b u t first let
me reiterate the basic fact here: N 0 matter h o w appealing the
argument, there is nowhere i n Descartes a n y t h i n g e x p l i c i t l y i n -
tegrative about the m i n d , as there is i n the Mesland letter about
the h u m a n b o d y .
T h i s absence sets a task w e m a y formulate as f o l l o w s : H o w
are w e to reinstate, w i t h i n the integrative f r a m e w o r k , the m i n d -
b o d y s y m m e t r y manifested i n separatist dualism? W e m a y p u t
the question this w a y : W h a t w o u l d solve the f o l l o w i n g equation
for Descartes?
Separatist m i n d ?
Separatist b o d y Integrative b o d y (Mesland letter)
Integrative Dualism I: Bodies,

Human and Other
M y proposal is to investigate the letter o f Mesland w i t h an eye 5
to stating Descartes's account o f w h a t a h u m a n b o d y is. W i t h

this m u c h f o r m u l a t e d , w e m a y be clearer about w h a t a symmet-
r i c — " i n t e g r a t i v e " — a c c o u n t o f the h u m a n m i n d w o u l d come t o :
I consider w h a t is the b o d y o f a m a n and I find that this

w o r d " b o d y " is v e r y ambiguous. W h e n w e speak i n gen-
eral o f a b o d y , we mean a determinate part o f matter, a
part o f q u a n t i t y o f w h i c h the universe is composed. I n
this sense, i f the smallest a m o u n t o f that q u a n t i t y were
5. Letter to Mesland, February 9, 1645. I was first made aware of this

letter many years ago by a remark in a lecture of Rogers Albritton.
r e m o v e d w e w o u l d eo ipso judge that the b o d y was

smaller and n o longer complete; and i f any particle o f
the matter were changed w e w o u l d at once t h i n k that
the b o d y was no longer quite the same, n o longer numer-
ically the same. But w h e n w e speak o f the b o d y o f a
m a n , w e d o n o t mean a determinate part o f matter w i t h
a determinate size; w e mean s i m p l y the w h o l e o f the m a t -
ter j o i n e d to the soul o f that m a n . A n d so, even t h o u g h
the matter changes, and its q u a n t i t y increases o r de-
creases, w e still believe its the same b o d y , numerically
the same, p r o v i d e d that i t remains j o i n e d i n substantial
u n i o n w i t h the same soul; and w e t h i n k that this b o d y is
w h o l e and entire p r o v i d e d that i t has i n itself all the dis-
positions required to preserve that u n i o n . N o b o d y denies
that w e have the same bodies as w e had i n o u r infancy,
a l t h o u g h their q u a n t i t y has m u c h increased, and accord-
i n g to the c o m m o n o p i n i o n o f doctors, w h i c h is d o u b t -
less t r u e , there is n o longer i n t h e m any part o f the m a t -
ter w h i c h t h e n belonged to t h e m and even t h o u g h they
do n o t have the same shape any longer; so that they are
o n l y n u m e r i c a l l y the same because they are i n f o r m e d b y
the same soul. . . . Consequently, I do n o t t h i n k that
there is any part o f o u r bodies w h i c h remains n u m e r i -
cally the same for a single m o m e n t , a l t h o u g h o u r b o d y ,
qua h u m a n b o d y , remains always the same p r o v i d e d i t is
u n i t e d w i t h the same soul. I n that sense, i t can even be
called i n d i v i s i b l e ; because i f an a r m o r a leg o f a m a n is
amputated, w e t h i n k that i t is o n l y i n the first sense o f
" b o d y " that his b o d y is d i v i d e d — w e do n o t t h i n k that a
m a n w h o lost an a r m o r leg is less a m a n than any
other. A l t o g e t h e r t h e n , p r o v i d e d that a b o d y is u n i t e d
w i t h the same rational soul, w e always take i t as the
b o d y o f the same m a n whatever matter i t m a y be and
whatever q u a n t i t y o r shape i t m a y have. ( P L , 154—159)
T h i s concerns the human b o d y . T h e account o f its existence

t h r o u g h t i m e , n o t i n terms o f some fixed q u a n t i t y o f matter,
m i g h t be t h o u g h t special to h u m a n bodies. B u t Descartes makes
74 W H A T A M I?
sure to stress a general metaphysical p o i n t earlier i n the letter:

" W e can say that the L o i r e is the same r i v e r as i t was ten years
ago, a l t h o u g h i t is n o t the same water, and perhaps there is no
longer a single part left o f the earth w h i c h surrounded that
water" ( P L , 156).
I isolate i n these passages t w o themes that w i l l play a central
role i n the integrative p i c t u r e . T h e first concerns the metaphys-
ics o f bodies (objects) i n general—rivers, pieces o f wax, and
h u m a n bodies—and their endurance i n t i m e . T h e second p o i n t
concerns specifically the i d e n t i t y o f h u m a n bodies and w h a t i t
takes for one t o endure t h r o u g h change. I consider first the
specific h u m a n b o d y c l a i m . T h e general metaphysics o f endur-
i n g objects w i l l concern us later.
Generic Bodies: Descartes Cuts
I n the first sentence o f the passage, Descartes alludes t o an

a m b i g u i t y i n the concept o f " b o d y " :
W h e n w e speak i n general o f a b o d y , w e mean a deter-

minate part o f matter, a part o f q u a n t i t y o f w h i c h the
universe is composed. I n this sense, i f the smallest
amount o f that q u a n t i t y were r e m o v e d w e w o u l d eo ipso
judge that the b o d y was smaller and n o longer complete.
(PL, 156)
T h i s is also the n o t i o n assumed b y the separatist treatment o f

body-in-general i n M e d i t a t i o n I I . I see i t as m a r k e d b y six
6
fundamental theses:
(SDBi) Ontologicalprimality: D B , s i m p l y qua b o d y , is o n t o -

logically p r i m a l ; the h u m a n b e i n g R D — t h e entity
composed f r o m i t and D M — i s derivative i n its c o n -
6. CSM I I , 17.
ditions o f ( i ) existence, ( i i ) i d e n t i t y , and ( i i i ) a r t i c u -

l a t i o n o f w h a t i t is.
(SDB2) Essence: T h e essence o f D B is generic; w h a t - i t - i s is
an extended t h i n g .
(SDB3) Existence: D B ' s existence, s i m p l y qua b o d y , does
n o t depend o n the existence o f the m a n , R D .
(SDB4) Endurance: T h e instantaneous conception o f b o d y ,
for example, o f D B , does n o t a l l o w for its endur-
ance t h r o u g h change.
(SDB5) Substance: D B is a substance i n the sense o f existen-
tial independence.
(SDB6) Real distinction: D B is really distinct f r o m the m a n
R D (and f r o m the m i n d D M ) i n the existential
sense—it can exist w i t h o u t R D and w i t h o u t D M .
T h e k e y theses are ( S D B i ) and ( S D B 2 ) . B y ( S D B i ) , D B is

g i v e n p r i o r t o and independently o f its eventual c o n t r i b u t i o n t o
the m a k i n g o f that derivative e n t i t y , the h u m a n b e i n g R D . T h e
separatist picture is o f bodies and minds created separately and
leading their autonomous existence. T h e n o t i o n o f a h u m a n
b e i n g — w h e t h e r the result o f a c o m p o u n d i n g , c o m p o s i t i o n , o r
u n i o n o f a m i n d and a b o d y — i s just this: a subsequent b i n d i n g
o f antecedently g i v e n m i n d and b o d y . I n particular, i n itself,
D B is just a material b o d y — a cluster o f matter o r an exten-
s i o n — n o t the b o d y o f an e n t i t y o f a certain k i n d (a h u m a n
b o d y ) , let alone the b o d y o f a specific entity o f a given k i n d
(the b o d y o f this h u m a n b e i n g R D ) .
T h i s conception is further articulated i n ( S D B 2 ) . The es-
sence o f D B is said t o be exhausted b y the generic a t t r i b u t e — i t
is an extended t h i n g . F o r all that its generic essence specifies,
D B c o u l d v e r y w e l l be the b o d y o f another h u m a n b e i n g , for
example, Princess Elizabeth; and w i t h equal ease, i t c o u l d be
the b o d y o f an e n t i t y o f quite a different k i n d , the b o d y o f a
giraffe o r a statue.
T h u s , D B is g i v e n as an instantaneous cut, a d i s t r i b u t i o n o f
matter i n space. T h e slightest alteration i n the material c o m p o -
sition leads t o another such instantaneous b o d y . I call this defi-
7 6 W H A T A M I?
n i t i o n b y abstraction o f a b o d y the postulation o f a Descartes

cut: C o r r e s p o n d i n g t o every three-dimensional, instantaneous
cut i n the material c o m p o s i t i o n o f the universe, w e postulate a
body. 7
The r e m a i n i n g principles ( S D B 3 ) - ( S D B 6 ) fall i n t o place

once w e read t h e m as intended for Descartes cuts. F o r its exis-
tence, the instantaneous cut does n o t depend o n this specific
h u m a n b e i n g R D , " c o n s t r u c t e d " f r o m i t and other such cuts.
S i m i l a r l y , this b o d y , g i v e n strictly as a simple cluster o f matter,
is n o t a real historical subject—one that comes i n t o b e i n g i n
t i m e , changes t h r o u g h t i m e , and eventually comes t o an end.
F i n a l l y , a b o d y m a y be v i e w e d as a substance w h e n w e classify
b y the existential conception assumed b y the separatist. Each
such Descartes cut owes its existence t o the cut definition and
n o t t o any robust metaphysical connections w i t h other real his-
torical objects.
Bodies of a Given Kind
T h e preceding discussion articulates the conception o f b o d y

Descartes relies o n t h r o u g h o u t M e d i t a t i o n I I . W h a t is s t r i k i n g
about the letter t o Mesland is that Descartes sketches a second
conception o f the h u m a n b o d y — i n d e e d , o f other kinds o f b o d -
ies, for example, rivers. T h i s alternative conception is n o t
merely i n t r i g u i n g for its consequences concerning the m i n d -
b o d y question. A s m e n t i o n e d , i t lays o u t a substantially differ-
ent metaphysics o f objects i n general. I see this second concep-
t i o n , w h e n applied t o D B , as characterized b y a corresponding
sextet o f theses:
(IDBi) Ontological primality: T h e h u m a n b e i n g , for exam-

ple, R D , is the e n t i t y i n terms o f w h i c h i t is speci-
7. My terminology is meant to remind us of a mathematical analogy

from a procedure of Dedekind in which for each "cut" in the system of ra-
tional numbers, we postulate the existence of a real number.
fied w h a t D B is, that is, the b o d y o f a m a n . Conse-

q u e n t l y , the conditions o f ( i ) existence and
( i i ) i d e n t i t y o f D B are specified i n terms o f the m a n
whose b o d y i t is.
(IDB2) Essence: D B is assigned a f u l l - b l o o d e d essence—the
body o f a human being.
(IDB3) Existence: D B existentially depends o n R D (and vice
versa); i n t u r n , i t depends o n D M .
(IDB4) Endurance: D B is a historical subject, e n d u r i n g
t h r o u g h change o f material c o m p o s i t i o n .
(IDB5) Substance: D B is a substance i n the categorical
sense—a subject o f a category o f predications; i t
itself is n o t a m o d e or w a y o f b e i n g o f a subject.
(IDB6) Real distinction: D B is really distinct, i n the categori-
c a l — i f n o t the existential—sense, from R D (and, i n
turn, from D M ) .
Y e t again, the k e y lies i n the first t w o theses. T h r o u g h o u t

his letter, Descartes emphasizes the contrast between t h i n k i n g
o f a b o d y abstractly ( " i n general") and t a k i n g i t as the b o d y o f
"a m a n , " indeed, o f a specific historical subject o f this k i n d , a
m a n w i t h a particular h i s t o r y . T a k e n i n the latter w a y , the b o d y
has this "of-ness" b u i l t i n t o w h a t - i t - i s : I t is the b o d y o f a h u -
m a n b e i n g . W e are n o longer g i v e n an array o f "generic b o d -
ies" and "generic m i n d s , " o n l y t o fancy ourselves creative ar-
chitects w h o are to construct f r o m t h e m an array o f h u m a n
beings. W e start w i t h a g i v e n h u m a n b e i n g and ask about its
b o d y a v a r i e t y o f metaphysical questions: W h a t is the nature
o f this b o d y , the k i n d o f i t e m i t is? U n d e r w h a t conditions can
it come i n t o existence? W h a t does i t take for i t t o endure
t h r o u g h change? F i n a l l y , w h a t a m o n g bodies o f its k i n d d i s t i n -
guishes i t as this specific one?
T h e k e y answer is the one about whatness. W h a t D B is is a
h u m a n b o d y ; this is glossed as the b o d y o f a h u m a n being.
T h e r e is a latent a m b i g u i t y i n this last c l a i m because o f the
indefinite article: F o r example, is D B the b o d y o f some h u m a n
7 8 W H A T A M I?
b e i n g or other ( " a n y o l d h u m a n being")? O r is i t essentially

the b o d y o f a specific h u m a n b e i n g , say, RD?
T h e intended answer is based o n the h o m e t r u t h that D e s -
cartes's b o d y , D B , and Elizabeth's b o d y , E B , are the same kind
o f object—a h u m a n b o d y . Suppose w e a l l o w e d i n the whatness-
specifying statement a reference to the particular m a n , Rene
Descartes: W h a t D B is is the b o d y o f this m a n , R D . T h u s , D B
and EB w o u l d no longer share whatness. T h i s seems t o be an
error: D B and E B are the same k i n d o f t h i n g , a h u m a n b o d y .
W h i c h b o d y , a m o n g h u m a n bodies, isJDB? T h e b o d y o f Rene
Descartes. T h i s is true o f D B t h r o u g h o u t its lifetime and i n
any counterfactual possibility. I t is also i n d i v i d u a t i v e o f D B
a m o n g its k i n d — n o other h u m a n b o d y is Rene Descartes's
b o d y . B u t w h e n i t comes to w h a t D B is—the k i n d o f i t e m i t
i s — i t is just w h a t EB is, a h u m a n b o d y . 8
T h e p r i m a l i t y o f m a n encoded i n t o ( I D B i ) and ( I D B 2 ) i m -
pinges o n existence questions broached i n ( I D B 3 ) . T h e exis-
tence o f the b o d y is subordinated t o that o f the m a n : W h e n the
m a n dies, its l i v i n g h u m a n b o d y is dead and gone; the r e m a i n -
i n g alignment o f molecules o n the pathologist's table is—despite
its b e i n g the same material extension as before—another k i n d
o f t h i n g , a corpse. T h u s , as u r g e d i n ( I D B 4 ) , D B has been
transformed f r o m a generic abstraction—an instantaneous dis-
t r i b u t i o n o f matter i n space—into a historical subject, one that
is generated i n h i s t o r y , undergoes change, survives material a l -
terations i n c o m p o s i t i o n , and eventually corrupts w h e n the m a n
whose b o d y i t is has come t o an end. F i n a l l y , ( I D B 5 ) and
8. This view precludes the possibility—in actual or counterfactual his-

tory—that this body, DB, will "become" the body of a man other than Des-
cartes. And given a certain man, say, Rene Descartes, it precludes its com-
ing to have any other body than DB. This pits the present metaphysics
against more modern, neo-Lockean theories of "personal identity." The al-
leged possibilities of man-body swaps allowed by such modern views are
discussed in detail in chapter 3. I also return there to the important ques-
tion of whether what DB is is not just the body of a man but also the body
of this particular man, Descartes.
( I D B 6 ) insist o n an alternative sense i n w h i c h D B is a sub-

stance; D B has n o w been made existentially dependent o n t w o
o t h e r s — R D and D M . B u t i t is the distinct subject o f predica-
tions, i n c l u d i n g the p r e v i o u s l y m e n t i o n e d changes i n predicates
across t i m e . A n d a l t h o u g h i t is the subject o f predications, i t
itself does n o t exist i n others—for example, in D M or i n
R D — a s a predicate inheres i n a subject; i t is, rather, a subject
whose existence is interdependent o n the existence o f these
other subjects. Consequently, D B is really distinct f r o m RD'
( a n d , i n t u r n , D M ) i n the categorical sense—both relata o f the
d i s t i n c t i o n relation are subjects, and neither exists i n the other
as a p r o p e r t y inheres i n a subject.
Integrative Dualism II: The Human Mind
T h e subtitle o f M e d i t a t i o n I I is " O n the N a t u r e o f the H u m a n

M i n d . " B u t soon enough the account turns o u t t o be o f m i n d
i n general, n o t w i t h s t a n d i n g any allusion t o the k i n d o f m i n d i t
is. W e have seen that i n the letter t o Mesland, Descartes p r o -
vides a nongeneric account o f b o t h the m a n and its h u m a n
b o d y , b o t h treated as real, historical subjects. The question
is w h a t account o f the man's m i n d w o u l d fit i n t o this second
approach.
O n e qualification is called for at the outset. I t c o u l d be ar-
gued that for all that Descartes literally says i n the letter, the
account o f m i n d suggested m a y be taken i n the generic way.
Descartes speaks o f the h u m a n b o d y as b e i n g the same as l o n g
as i t remains u n i t e d to the same m i n d . T h i s r e m a r k seems t o
i m p l y that the m i n d itself endures i n time—remains the same:
T h e b o d y remains u n i t e d t o the same m i n d b y r e m a i n i n g u n i t e d
to a m i n d that remained the same. B u t even i n such a reading,
it is still possible t o give quite an abstractionist account o f m i n d .
F o r example, the e n d u r i n g m i n d i n question need n o t be speci-
fied as a m i n d o f a h u m a n being: M i g h t n o t m y h u m a n b o d y
be i n f o r m e d t h r o u g h and t h r o u g h b y an angelic mind?
8o W H A T A M I?
I n a more general w a y , w e m a y frame a consistent reading

o f the Mesland letter b y p r o p o s i n g a t h e o r y o f whatness and
endurance for minds that does n o t account for t h e m as i n t r i n s i -
cally human minds. Here w e are o n l y t o recall the " a m b i g u i t y "
i n the n o t i o n o f b o d y that Descartes points o u t i n the first line
o f the q u o t a t i o n o n page 72 ( P L , 156). T h a t is, w e m a y c o n t e m -
plate a symmetric " a m b i g u i t y " i n the n o t i o n o f m i n d .
Consider the generic idea o f m i n d , m i n d " i n general." T h i s
idea is n o t a mere hypothetical analog o f the b o d y - i n - g e n e r a l
idea o f the Mesland letter. I t is the actual idea o f m i n d proposed
b y Descartes i n M e d i t a t i o n I I ( C S M I I , 18-19). T h e conception
is laid out i n the paragraphs f o l l o w i n g Descartes's decision t o
shift f r o m the p r i m a l question " W h a t am I ? " t o " W h a t is
m i n d ? " A l l w e have t o do n o w is reconsider the separatist b o d y
principles ( S D B 1 ) - ( S D B 6 ) , except one ( w h i c h w e shall discuss
s h o r t l y ) , and make the obvious substitutions. W e have before
us the principles that characterize the separatist conception o f
mind:
(SDMi) Ontologicalprimality. D M , s i m p l y qua m i n d , is o n t o -

l o g i c a l l y p r i m a l ; the h u m a n b e i n g R D — t h e e n t i t y
composed f r o m i t and D B — i s derivative i n its c o n -
ditions o f ( i ) existence, ( i i ) i d e n t i t y , and ( i i i ) a r t i c u -
l a t i o n o f w h a t i t is.
(SDM2) Essence. T h e essence o f D M is generic; w h a t - i t - i s is
a thinking thing.
(SDM3) Existence. D M ' s existence, qua m i n d , does n o t de-
pend o n the existence o f the m a n , R D .
(SDM5) Substance. D M is a substance i n the sense o f existen-
tial independence.
(SDM6) Real distinction. D M is really distinct f r o m R D and
D B i n the existential sense—it can exist w i t h o u t
them.
O n e principle is missing, ( S D M 4 ) , w h i c h concerns endur-

ance i n t i m e . A n account o f m i n d - i n - g e n e r a l that is f u l l y s y m -
metric to the b o d y - i n - g e n e r a l idea o f the Mesland letter w o u l d

insist o n the analog p r i n c i p l e ( S D M 4 ) :
(SDM4) Endurance. T h e instantaneous conception o f m i n d ,

for example, o f D M , does n o t a l l o w for its endur-
ance t h r o u g h change.
A s w e shall see i n a m o m e n t , there is no w a y o f reading

M e d i t a t i o n I I ' s generic account o f m i n d as subscribing t o this
principle. T h e nontransfer o f this one separatist p r i n c i p l e chal-
lenges reflection. W h a t i f Descartes were n o t w h o l l y separatist
about mi nds after all?
2.4. Subjects i n T i m e : T w o
Cartesian Frameworks
T h e materials that make up the instantaneous n o t i o n o f b o d y

i n the Mesland letter are " p h y s i c a l " (a cluster o f particles, an
assortment o f physical qualities o f the extension, etc.). T h e ana-
l o g — a n instantaneous m i n d — w o u l d be yet again a k i n d o f D e s -
cartes cut, b u t this t i m e o f intellectual ingredients: an instanta-
neous cut o f thoughts and w i l l i n g s . T h e materials, o r fabrics,
that make u p instantaneous bodies and m ind s are different; l o g i -
cally, the conception is the same, that is, c o m p o s i t i o n f r o m the
relevant instantaneous " p a r t s . "
T h e slightest change i n c o m p o s i t i o n — i n t e l l e c t u a l o r p h y s i -
c a l — w o u l d give us a different i t e m : m i n d o r b o d y . I n b o t h
cases, w e d o n o t have real, historical subjects that are generated
i n t i m e , endure i n h i s t o r y , and eventually c o r r u p t . Rather, w e
have abstractions, with fixed identities imposed o n logical
grounds b y the cut definitions and e x c l u d i n g b y fiat any sur-
v i v a l t h r o u g h alteration o f qualities. 9
9. I speak here of a bundle of purely intellectual ingredients because in

various contexts Descartes asserts that perceptual, imaginative, and affective
states require essentially the existence of a body (note that in an existen-
82 W H A T A M I?
T h e conception o f m i n d as a b u n d l e o f intellectual qualities

is interesting; variations o n the theme have been alluded t o b y
eminent readers o f Descartes (most famously, H u m e , b u t also
L o c k e and m o d e r n personal i d e n t i t y theorists). B u t n o matter
h o w generic the other principles o f M e d i t a t i o n I I are about
m i n d , Descartes h i m s e l f does n o t endorse this last analogy c o n -
c e r n i n g endurance i n t i m e . F o r Descartes, a g i v e n h u m a n m i n d
is never analyzed as an instantaneous b u n d l e o f intellectual
qualities; a specific h u m a n m i n d is an e n d u r i n g subject. A g i v e n
h u m a n m i n d can change and remain* n u m e r i c a l l y the same
m i n d — d i f f e r e n t thoughts b u t still the same m i n d . W h e n H u m e ,
L o c k e , and K a n t pursue a c u l p r i t that so sustains ("blunders
i n t o " ) the existence o f a mysterious inner subject, i t is Descartes
they are after.
T h i s suggests that i f any b o d y , any " p h y s i c a l i a , " o f M e d i t a -
t i o n I I is t o serve as a m o d e l for Descartes's conception o f a
g i v e n h u m a n m i n d , the i t e m m a y n o t be the abstractly g i v e n
instantaneous b o d y (extension). Rather, i t should be a r o b u s t l y
g i v e n i t e m , constantly changing i n h i s t o r y . Such an i t e m does
come up for dissection i n M e d i t a t i o n I I . I t is the piece o f w a x
(first appearing i n C S M I I , 2 0 - 2 1 ) .
T h e (piece of) w a x is b r o u g h t t o the fore late i n M e d i t a t i o n
I I as a " s i m p l i f i e d case" for an analogy w h e n w e frame o u r
account o f the h u m a n m i n d . T h i s m u c h is n o t i n dispute. W h a t
is more controversial is the " d i m e n s i o n " over w h i c h the piece
o f wax is to serve as a " m o d e l " for the h u m a n m i n d . B u t before
I take a stand here, I must note a simple fact, one h o l d i n g for
various interpretations o f the piece-of-wax-as-a-model: Side b y
tially separatist notion of a substance, this dependence would endanger the

substancehood of the mind). Thus, it is somewhat surprising to find Des-
cartes himself, in the Synopsis to his Meditations, mention "desires" and
"sensations" in a context relevant to the present discussion (see CSM I I ,
10). Perhaps he relates to the purely intellectual aspect of a desire. In this
discussion, I assume that the ingredients, for example, thoughts, that "make
up" an instantaneous mind need nothing but God to exist.
side w i t h the abstractionist, ahistoric account o f bodies, D e s -

cartes's discussion o f the c h a n g i n g w a x i n M e d i t a t i o n I I shows
a keen interest i n a metaphysics o f robust historical subjects
that endure i n t i m e .
T h e side-by-side presence o f t w o opposite o u t l o o k s strikes
me as a general p r o b l e m i n Descartes's metaphysics, even be-
fore w e come t o the particular case o f h u m a n m i n d s . I n differ-
ent contexts and sometimes i n a single, b r o a d context (e.g.,
M e d i t a t i o n I I ) , Descartes operates w i t h t w o c o m p e t i n g concep-
tions o f objects ("substances"), their essence (whatness) and
their endurance i n t i m e . T o reiterate, the a m b i g u i t y a n d / o r
ambivalence cuts across the b o a r d , a p p l y i n g t o all k i n d s o f sub-
jects—pieces o f w a x , rivers, h u m a n bodies, and ( p u r p o r t e d sub-
jects that t h e y are) h u m a n m i n d s .
Consider rivers, for example, Descartes's L o i r e . B y the cut
definition, w e fix o n a certain set o f earth-and-water particles
at a g i v e n t i m e . T h e r e n o w f o l l o w s an abstractionist t h e o r y o f
essence ( w h a t - i t - i s ) , n u m e r i c a l i d e n t i t y , and existence i n t i m e .
F o r essence, w e are m e r e l y t o l d the v e r y abstract and generic:
I t is an extended object. F o r i d e n t i t y , the Descartes cut slices
the object as t h i n l y as possible: A n y v a r i a t i o n i n the q u a l i t y set
gives us another object. F i n a l l y , for endurance, there is n o en-
d u r i n g i n t i m e for this object, as the water flows and the r i v e r
b a n k erodes, and so o n . A l l i n a l l , the whatness, i n d i v i d u a t i o n ,
and endurance conditions abstract f r o m the robust historical
subject, the Loire-through-the-ages. W e are g i v e n a " l o g i c a l
object," an abstraction.
T h i s project o f Descartes involves a f o r m o f subspecies-aeter-
nitate metaphysics. T h e metaphysician is n o t o p e r a t i n g " f r o m
b e l o w , " f r o m inside the m u c k o f things; he is n o t describing
the life o f this b o d y and this m i n d , let alone this man they make
u p . N o r is i t his o r her task t o account for the e v o l u t i o n i n
h i s t o r y o f this r i v e r and this piece o f wax. T h e project is rather
t o classify e v e r y t h i n g whatsoever b y a simple system o f genera
(a " t w o - s o r t e d l o g i c " w o u l d have said the m o d e r n l o g i c i a n ) :
extended vs. t h i n k i n g . T h e metaphysician is operating " f r o m
84 W H A T A M I?
above," really f r o m b e y o n d the cosmological receptacle w i t h i n

w h i c h things are endowed w i t h a specific h i s t o r y and are m e m -
bers o f specific kinds. I call this project o f describing the w h o l e
w o r l d " f r o m above" Descartes's abstractionist metaphysics.
I call the k i n d o f second project Descartes is interested i n
lifetime metaphysics because here o u r focus is o n describing,
f r o m inside the cosmos, the lifetimes o f specific historical items,
w h a t i t takes for t h e m to come to be and the l i m i t s o f change
they can endure. F o r essence, instead o f the generic, t w o - s o r t e d
classification (extended vs. t h i n k i n g ) , objects are n o w subsumed
i n specific k i n d s — r i v e r s , pieces o f w a x , stones, h u m a n m i n d s ,
angelic m i n d s , and d i v i n e m i n d s . W h e n i t comes to i n d i v i d u a -
t i o n , each such object is n o t reduced t o this or that cut o f
qualities; the qualities are s i m p l y borne b y the object.
T h e r e is n o w , at least l o g i c a l l y , the potential for an alteration
o f qualities that preserves the i d e n t i t y o f the pertinent objects
such as the piece o f wax o r the L o i r e . B y logical potential I
mean to indicate this: I t m i g h t still t u r n o u t that, i n the cosmos,
things are so organized that no object ever alters its qualities.
B u t if, indeed, each object is changeless, this w o u l d n o t be due
to a " f r o m a b o v e " logical doctrine about i d e n t i t y that stipulated
the result; rather, i t w o u l d be due t o the fact that, metaphysi-
cally, this is the w a y things stand i n the w o r l d .
T h i s potential for logical change is accorded to all mundane
objects. I t is realized i n different ways b y different objects, de-
p e n d i n g o n the k i n d o f objects they make. T o b r i n g o u t the
metaphysical potential o f a specific object, w e engage i n its life-
time d e s c r i p t i o n — h o w items o f its k i n d actually evolve and
h o w , at different points i n the object's h i s t o r y , i t might have
gone o n to be. W e thus discover different metaphysical p o t e n -
tials—to a specific whatness belongs a specific manner o f p r e -
serving i t as t i m e goes b y ; rivers are rivers, w a x is wax.
A s I read h i m , Descartes is interested i n t w o different p r o j -
ects i n metaphysics. O u r task, therefore, is to keep his distinct
projects separate. W h e n p r o v i d i n g a lifetime metaphysics for
the h u m a n b e i n g and his b o d y , Descartes puts b e h i n d h i m the
abstractionist account o f b o d y and calls o n his robust metaphys-

ics o f e n d u r i n g items. M y suggestion is that a similar m o v e is
needed i n the case o f the h u m a n m i n d for t w o reasons. First,
to make sense o f w h a t Descartes says o f the h u m a n b o d y , i f
the m i n d is t o c o n t i n u a l l y " i n f o r m " the e n d u r i n g b o d y , keep i t
and the h u m a n b e i n g whose b o d y i t is living, w e can n o longer
r e l y o n a generic-abstractionist conception o f m i n d ; w e need t o
treat the m i n d as a historical subject, e v o l v i n g w i t h this h u m a n
b o d y . M o r e o v e r , w e need to understand this c o e v o l u t i o n as
determined b y the k i n d s o f items the e v o l v i n g subjects are—a
human m i n d and b o d y (the c o e v o l u t i o n o f other symbiotic
duos, e.g., b i n a r y stars or pairs o f elementary particles w o u l d
proceed d i f f e r e n t l y ) . Second, quite apart f r o m the Mesland let-
ter, w h i c h is focused o n the " i n f o r m i n g " - t h e - b o d y r o l e — t h a t
is, just t h i n k i n g o f the m i n d b y itself—we need t o account for
its intellectual life: the changes i n content ( " I have just changed
m y m i n d " ) as i t remains one and the same m i n d .
I t is m y suggestion, t h e n , t o start i n M e d i t a t i o n I I w i t h w h a t
Descartes says about a perfectly mundane i t e m , the piece o f
w a x , w h i c h he intends t o be a simplified m o d e l for the case o f
the h u m a n m i n d . W e must understand w h a t i n his metaphysical
dynamics is the basis for the wax's e n d u r i n g t h r o u g h change;
then, w i t h this m o d e l before us, w e must reflect o n w h a t i t
takes another e n d u r i n g subject, D M , t o persist t h r o u g h change.
Subjects in Time I: The Enduring

Piece of Wax
W h a t does Descartes tell us about the piece o f w a x i n M e d i t a -

tion I I ? 1 0
I t is a cluster o f epistemological issues that has d r a w n
attention i n the w a x discussion. H o w do w e track this i t e m as
i t changes and h o w do w e come t o k n o w its essence? B u t o u r
10. In what follows, when I speak of "the wax passage" I refer to CSM
I I , 20—21.
86 W H A T A M I?
focus here is the metaphysical question: W h a t i n the w o r l d

grounds this item's endurance t h r o u g h change?
W e need a name for our piece o f wax. I n t r o d u c i n g i t , D e s -
cartes tells us, " I t has just been taken f r o m the h o n e y c o m b , "
and so I shall refer t o i t as H o n . Descartes tells us that after
b e i n g w a r m e d up and melted d o w n , H o n has changed radically,
losing m a n y o f its o r i g i n a l , sensible qualities. T h e qualities i t
has kept are "extendible, changeable, flexible." The trio o f
qualities is Descartes's official proposal for the essence o f H o n .
Descartes also repeatedly uses the phrase " i t remains the same
w a x . " O f course, he wants to indicate the stable numerical i d e n -
t i t y o f w h a t he calls "one particular b o d y , " that is, H o n . I n so
referring t o H o n , however, he n o t o n l y assures us that i t has
persisted i n b e i n g one and the same e n t i t y b u t he also concedes
that i t has remained t h r o u g h and t h r o u g h o f the same k i n d —
wax. So, i n m y reading, being-a-piece-of-wax is also a persist-
i n g p r o p e r t y o f H o n . I n a l l , Descartes tells us that t h r o u g h o u t
the wax's shedding o f its sensible qualities, there stably remains
one and the same i n d i v i d u a l i t e m ( H o n ) o f the same k i n d o f
stuff ( w a x ) .
A friend o f Descartes's abstractionist metaphysics m i g h t w e l l
object at this p o i n t i n the f o l l o w i n g w a y :
F o r all that was said so far, Descartes need n o t call o n a

lifetime metaphysics o f e n d u r i n g objects to get a c o m -
plete description o f the changes that have occured i n the
pertinent part o f the w o r l d . H e should rather describe
things i n this way: H a v i n g w a r m e d up a certain D e s -
cartes cut, a definite b o d y extension ( H o n ) , w e have af-
fected that item's shape. A s a result, w e d r o v e this exten-
sion out o f existence. T h e resulting m e l t e d - d o w n i t e m
(call i t M e l ) is a n e w Descartes cut, another b o d y v i z . an-
other extension. Sure enough, the changes have been
continuous and have f o l l o w e d a familiar path o f c h e m i -
cal deformation. B u t w h e n w e r e m i n d ourselves that
each b o d y is "a determinate part o f matter, a part o f
q u a n t i t y o f w h i c h the universe is composed," w e see

that the w a r m i n g up o f H o n has g i v e n us another b o d y -
object. T h e fact that the change between H o n and M e l
seems continuous e n o u g h is w h a t lies at o u r i n c l i n a t i o n
t o posit a single e n d u r i n g object. B u t , strictly, all that
has happened t o the w o r l d is that one extension—a
g i v e n cluster o f matter—was replaced b y another. P r o -
ceeding i n this w a y , w e c o u l d give a complete descrip-
t i o n o f the h i s t o r y o f the w o r l d , the changes that c o l l o -
q u i a l l y w e w o u l d describe as i n c u r r i n g i n H o n and i n
the L o i r e and i n m y l i f e l o n g h u m a n b o d y .
T h e objection claims that w e c o u l d p r o c e e d — w i t h pieces o f

w a x , as w i t h rivers and h u m a n b o d i e s — w i t h o u t a dynamics o f
e n d u r i n g objects. C o u l d we?
Possibly—or so at least w e are t o l d b y m a n y metaphysicians,
ancient and m o d e r n , f r o m Heraclitus t o Q u i n e , all w e a r y o f o u r
marketplace descriptions i n terms o f e n d u r i n g objects. N o w , m y
task i n this b o o k is n o t t o evaluate the correctness o f this " a n y -
t h i n g y o u (marketplace man) c o u l d d o , I (abstractionist meta-
physician) c o u l d do ( b e t t e r ) . " M y focus is o n understanding
Descartes. G i v e n the t w o metaphysical schemes he operated
w i t h , he surely had the means t o issue such a theoretical-physi-
cist " c o u l d describe the w o r l d " c l a i m . H e c o u l d so describe the
w a x and the L o i r e and the h u m a n b o d y . B u t he d i d n o t . I n
describing the changes f r o m the w a r m i n g u p o f the w a x , he
used the language o f the marketplace—as changes in the endur-
i n g H o n . A n d here lies o u r question: W h a t facts about H o n
g r o u n d this marketplace description?
M y hypothesis is that i t all turns o n the essence o f H o n , o n
w h a t - i t - i s . M a k e its whatness generic, and persistence t h r o u g h
change is n o t open t o i t . I n contast, accord H o n robust w h a t -
ness, and y o u have e n d o w e d i t w i t h a real p o t e n t i a l for change.
W e m i g h t summarize the basic hypothesis:
(E) N 0 endurance w i t h o u t robust whatness.

88 W H A T A M I?
M y reasoning about Descartes is this: H e describes H o n as en-

d u r i n g ; b y principle ( E ) , he must v i e w i t as h a v i n g a robust
whatness.
Indeed, he does. Descartes emphasizes that the w a x stays
t h r o u g h o u t "extendable, changeable and flexible." B u t he also
stresses—by repeatedly using the phrase "remains the same
w a x " — t h a t t h r o u g h and t h r o u g h H o n remains the same k i n d
o f t h i n g , that is, a piece o f wax. T h e l i m i t s o f change that H o n
can endure rest v e r y m u c h o n this last a t t r i b u t i o n : F o r example,
H o n m a y n o t change i n t o a piece o f g o l d ; n o r c o u l d some
initial piece o f g o l d change i n t o H o n . T h e process o f w a r m i n g
up and expanding i n extension as described b y Descartes per-
tains t o wax: A t the temperatures described, o u r candidate piece
o f g o l d ( G o l d i e ) w o u l d n o t have melted at a l l . A t temperatures
at w h i c h G o l d i e w o u l d have m e l t e d , the waxness o f H o n w o u l d
have l o n g been destroyed and, w i t h i t , H o n itself. T h e m o t i o n
o f the particles inside the w a x molecules w o u l d be h i g h e n o u g h
to dismantle the chemical c o m p o s i t i o n that is a sine qua n o n for
b e i n g b e e s w a x — C H O C O C H . O n the other h a n d , G o l d i e ,
3 0 6 l 1 5 3 I
being made o f a heavy atomic element, is subject t o different

procedures: T o melt it d o w n , let alone t o break u p the g o l d
atoms that i t is made of, quite a different chemical reaction is
necessary.
Such are the differences at the " c o r r u p t i o n " stage. T h e r e are
correlate differences i n their respective " g e n e r a t i o n . " H o n and
Goldie must have f o l l o w e d different processes o f c o m i n g t o be.
H o n was just d r a w n f r o m the h o n e y c o m b . O r d i n a r y earthly
temperatures p e r m i t the b r e e d i n g o f that k i n d o f stuff. B u t
golden stuff—recall that the a t o m is one o f the heavier ele-
ments—could n o t get g o i n g i n the h o n e y c o m b . I t t o o k rather
dramatically h i g h temperatures i n the i n t e r i o r o f stars to engen-
der atoms o f that k i n d .
I n sum, suppose w e describe the t w o pieces o f matter i n
front o f u s — H o n and G o l d i e — i n m e r e l y generic terms as "ex-
tended t h i n g s . " T h e r e is v e r y little w e can do t o describe their
distinct profiles o f persistence i n time i n a natural w a y . F o r all
that is said under the generic description, someone—a p h i l o -

sophical alchemist—can attempt t o convert the one i n t o the
other; there is n o t h i n g t o b l o c k the transformation o f ( w h a t w e ,
f r o m the outside, using k i n d terms w o u l d describe as) a piece
o f gold-at-r i n t o a piece o f w a x - a t - / .
T h i s result is f o r t h c o m i n g for t w o , seemingly opposite meta-
physical theories that are abstractionist i n their approach t o H o n
and G o l d i e . O n e is Descartes's o w n t h e o r y o f i d e n t i f y i n g i n d i -
viduals w i t h their Descartes cuts and m o r e generally, an i t e m
w i t h its " b u n d l e o f qualities." T h e second t h e o r y seems t o be
at the other end o f the metaphysical spectrum; that is, each such
i n d i v i d u a l — l i k e H o n and G o l d i e — i s a "bare p a r t i c u l a r , " w i t h
n o p r o p e r t y whatsoever p e r t a i n i n g t o its essence o r w h a t i t is.
F o r that matter, the particular c o u l d be made "almost bare,"
w i t h " e x t e n d e d " f o r m i n g the sole p r o p e r t y p e r t a i n i n g t o its na-
ture. T h e t w o pictures m i g h t seem radically different, the first
canonizing every p r o p e r t y o f H o n and G o l d i e as a sine qua
n o n for b e i n g - i t , the second d e n y i n g any p r o p e r t y that title.
B u t , i n t r u t h , the t w o o u t l o o k s are variations o n a c o m m o n
theme—abstraction f r o m the robust historical i n d i v i d u a l . T h e
one picture abstracts the bundle o f qualities, the other some
i m a g i n a r y p o i n t l i k e subject. I n so abstracting, b o t h b l u r the
fundamental d i s t i n c t i o n between what the mundane i n d i v i d u a l
is and features p e r t a i n i n g merely t o how things stand w i t h i t at
the instant o f abstraction.
Suppose, t h e n , that w e abstract f r o m the facts as f o l l o w s :
(Kind) T h e k i n d o f t h i n g H o n is is a piece o f wax; the k i n d

o f t h i n g G o l d i e is is a piece o f g o l d .
B y abstracting ( K i n d ) , w e have n o w " h o m o g e n i z e d " H o n and

G o l d i e . A s t w o p o i n t l i k e bare subjects, they can come t o bear
any properties whatsoever; as t w o Descartes cuts, they cannot
change at a l l . E i t h e r w a y , w e do n o t get the natural description
o f t w o lifetimes o f change, for example, the one Descartes h i m -
self provides for H o n .
90 W H A T A M I?
Subjects in Time II: The Enduring

Human Mind
W h a t is g o o d for the w a x is g o o d for the h u m a n m i n d , o r so

says Descartes's the wax-as-a-model principle. M o s t often the
principle is applied i n relation t o the epistemology o f essence.
I n such applications, w e ask: B y w h a t mental faculty and k i n d
o f data do w e get to k n o w the essence o f a g i v e n subject? T h e
w a x is t h e n deployed t o illustrate the proposal that i t is the
pure intellect, n o t the senses, that reveals t o us w h a t the w a x
is. W e are n o w u r g e d b y the wax-as-a-model p r i n c i p l e t o p r o -
ject the result t o the manner i n w h i c h the essence o f the m i n d
is to be d i v u l g a t e d .
I n contrast, our present application is t h r o u g h l y metaphysical.
W i t h H o n , w e reasoned as follows: Descartes allows i t t o endure
t h r o u g h change; there is no endurance w i t h o u t robust whatness;
thus, he must be e n d o w i n g H o n w i t h a robust whatness. I p r o -
pose the same line o f reasoning for the human m i n d . O u r exam-
ple o f a specific m i n d is, as before, Descartes's m i n d , D M .
Descartes tells us at the outset, i n the synopsis o f the M e d i t a -
tions, that D M and its k i n d are subjects o f change: " E v e n i f all
the accidents o f the m i n d change, so that i t has different objects
o f the understanding, and different desires and sensations, i t
does n o t o n that account become a different m i n d " ( C S M I I ,
i o ) . Descartes emphasizes later, i n M e d i t a t i o n I I ( C S M I I , 18—
19) that, as he sleeps and awakes, perceives and daydreams,
walks and t h i n k s , one and the same h u m a n m i n d remains.
M e d i t a t i o n V I and his f o u r t h replies t o A r n a u l d add i m p o r -
tant facts about the e n d u r i n g linkage o f this g i v e n m i n d and its
sustaining b o d y . W e find that D M is n o t l i n k e d t o D B as a
" p i l o t t o a s h i p . " Such a h o o k u p w o u l d have been forged rather
late i n the life o f b o t h relata. T h e linkage is rather t o be t h o u g h t
of—according to Descartes's f o u r t h replies ( C S M I I , 160)—on
the m o d e l o f the h o o k u p o f Descartes's a r m ( D A ) w i t h D B .
T h i s relation is part o f the v e r y c o m i n g i n t o b e i n g o f b o t h
relata and is i n the v e r y — I am u s i n g Descartes's word—nature
o f b o t h . I n M e d i t a t i o n V I , Descartes goes o n t o speculate o n

where, i n the h u m a n b o d y , the relation is based. I t is i n the
pineal gland (conarion) that a l i f e l o n g dynamics o f back-and-
f o r t h " i n f o r m a t i o n transfer" occurs: T h e m i n d gets its i n f o r m a -
t i o n f r o m the senses and the b r a i n and instructs t h e m about
actions t o be t a k e n . 11
I w a n t t o emphasize that w h a t is u p f o r evaluation here is n o t

a matter o f correctness—does the c o n a r i o n hypothesis reveal t o
us w h a t the u n i o n between m i n d and b o d y is really like?
Rather, o u r objective is a better grasp o f Descartes's conception
o f the modalities o f the connection. H e tells us that i t is w r i t t e n
i n t o the nature o f a h u m a n m i n d t o be, t h r o u g h o u t its life, i n
u n i o n w i t h a g i v e n b o d y that i t i n f o r m s . T h e accidents o f the
b o d y t o w h i c h i t is connected w i l l change as i t — t h e mind—goes
t h r o u g h life; so w i l l the intellectual accidents o f the m i n d
proper. B u t t h r o u g h o u t , w e w i l l have o n l y one m i n d and one
b o d y interconnected.
Subjects in Time III:

The Abstractionist's Challenge
So far, Descartes provides, as w i t h the w a x , marketplace de-

scriptions o f a single h u m a n m i n d that is e n d u r i n g t h r o u g h
b o t h b o d i l y and intellectual changes. Is this f o r m o f description
inevitable? T h e description i n terms o f a single subject that is
e n d u r i n g t h r o u g h change has been challenged i n the case o f the
w a x b y the abstractionist metaphysician. Reasoning b y D e s -
cartes's o w n wax-as-a-model p r i n c i p l e , various philosophical
readers o f Descartes—most n o t a b l y L o c k e — t u r n the tables
against h i m : W h a t is g o o d f o r the w a x , is g o o d f o r the m i n d ;
n o endurance metaphysics is called f o r the w a x and thus none
for the h u m a n m i n d . T h e c h a l l e n g e r u n s l i k e this: T o explain
" t h e lifetime o f a m i n d , " there is n o need t o posit the m i n d as
и . CSM I I , 59-60.
92 W H A T A M I?
an e n d u r i n g substance. A succession o f m o m e n t a r y minds,

p r o p e r l y strung t o preserve memories and other " t e l l i n g " expe-
riential facts w o u l d do just as w e l l .
A m o n g philosophers w h o argue i n this v e i n , there are differ-
ent emphases o n the sense i n w h i c h such a situation is " i n d i s t i n -
guishable" f r o m the e n d u r i n g subject assumed b y marketplace
descriptions. T o some, the p o i n t concerns the epistemology o f
the situation v i e w e d from the outside: I , an external observer,
w o u l d n o t be wiser i f the m a n I observe, Rene Descartes, were
sustained i n t i m e n o t b y the single persisting D M b u t b y a
succession o f m o m e n t a r y minds. T o other philosophers, the
p o i n t concerns the epistemology o f the situation v i e w e d from
the inside: Descartes h i m s e l f — r e l y i n g o n his inner feeling o f
w h a t i t is l i k e , as w e l l as i n t r o s p e c t i o n — w o u l d n o t be wiser
i f G o d were t o revolve a succession o f m o m e n t a r y minds inside
his head instead o f the single, l i f e l o n g D M . T o yet a t h i r d
g r o u p o f philosophers, the p o i n t intended here concerns b o t h
outside and inside epistemology: T h e r e is s i m p l y no stand-
p o i n t f r o m w h i c h w e c o u l d tell the operation o f an e n d u r i n g -
m i n d hypothesis f r o m the sequence o f the m o m e n t a r y minds
counterpart.
I suspect that all three groups leap f r o m these epistemic o b -
servations—right o r w r o n g as they m i g h t be about h o w things
w o u l d appear—to metaphysical conclusions about the k i n d o f
facts necessary to describe the life o f Rene Descartes. A s e m -
phasized, the leap is n o t special to the case o f the m i n d . I t
is based o n the m e t h o d o l o g y relied o n w h e n the issue is the
prerequisites o f a complete description o f the piece o f w a x , the
L o i r e , or the h u m a n b o d y . I n m y understanding, w h a t is b e i n g
assailed b y L o c k e is n o t essentially and d i s t i n c t l y the case o f
the h u m a n m i n d as a subject o f change b u t the v e r y m e t h o d o l -
o g y o f p o s i t i n g e n d u r i n g subjects o f change.
A s emphasized i n o u r discussion o f the wax and the L o i r e ,
our task here is n o t t o j o i n this metaphysical dispute b y evaluat-
i n g the abstractionist's challenge. O u r task is to understand
Descartes. There is n o t m u c h o f a question about w h a t l a n -
guage o f description he adopts. Descartes, w h o uses a dynamic

metaphysics for pieces o f w a x , rivers, and h u m a n bodies, e m -
phatically proposes such a lifetime metaphysics for h u m a n
minds. T h e question before us is rather this: W h a t facts accord-
i n g t o Descartes's lifetime metaphysics e n d o w the h u m a n m i n d
w i t h its change potential?
A s before, i t all turns o n w h a t the essence o f o u r subject is.
Make D M ' s essence generic—merely a " t h i n k i n g t h i n g " — a n d
the endurance metaphysics is jeopardized. Make the whatness
robust—the m i n d o f a g i v e n h u m a n b e i n g — a n d change is n o t
o n l y possible for i t b u t also necessary for its o w n l i f e l o n g sur-
v i v a l . O f course, i n the context o f his generic metaphysics, D e s -
cartes is famous for assigning D M the mere " t h i n k i n g t h i n g "
as its generic whatness. B u t i n the setting o f his lifetime meta-
physics, a robust whatness is called o n . I t is t o this robust es-
sence for D M that w e n o w t u r n . 1 2
Subjects in Time IV: Three Lifetime Principles
Descartes submits that t h r o u g h o u t the wax's shedding o f its

sensible qualities, there remains one and the same i n d i v i d u a l
i t e m ( H o n ) ; i t stably remains o f the same k i n d o f stuff i t was
( w a x ) . W i t h D M , w e are t o l d the same: T h r o u g h o u t its changes
o f desires and t h o u g h t s , there stably remains one and the same
i n d i v i d u a l i t e m ( D M ) and one the same k i n d o f i t e m (a h u m a n
m i n d ) . B u t there is m o r e : T h r o u g h o u t the changes, i t stably
remains true that D M is the m i n d o f the same h u m a n b e i n g .
T h i s t h i r d stable f e a t u r e — D M is always the m i n d o f the
same h u m a n b e i n g — I call the prime lifetime fact because i t is
12. Related to the robustness of essence is what the metaphysician

would assert to be conceivable. Many in the generic essence tradition have
argued that they can conceive for D M a variety of body-shedding scenar-
ios. In contrast, a robust essence of D M exposes such claims of conceivabil-
ity as unfounded, i f we read them as de re conceivings, that is, scenarios
for this specific item, D M . I dwell on this difference in chapter 3.
94 W H A T A M I?
the most fundamental fact i n Descartes's account o f the life o f

a m i n d l i k e D M . I see the p r i m e lifetime fact as encapsulating
three principles i n one claim.
T h e first principle proposed here is that the k i n d o f m i n d a
m i n d is is stable i n t i m e . G i v e n that D M is a human mind, it
m a y n o t t u r n i n t o an angelic or d i v i n e m i n d . A n d thus w e have
(Stability) T h e k i n d o f a m i n d is lifetime stable.
Second, w e get a principle about k i n d subsumption. Consider,

for example, the b o d y o f a d o g . I t is natural t o propose that a
b o d y is a d o g b o d y i f f i t is the b o d y o f a d o g b e i n g , that is, o f
a d o g . T h e principle suggested for minds is a natural general-
ization:
(Of-ness) A m i n d is o f a g i v e n k i n d i f f i t is the m i n d o f a
b e i n g o f that k i n d .
T h u s , the k i n d o f a m i n d is the v e r y k i n d o f the b e i n g whose

m i n d i t is. So, to be a h u m a n rather than an angelic o r a d i v i n e
k i n d o f m i n d , D M is t h r o u g h and t h r o u g h the m i n d o f a h u m a n
being. 13
F o r its w h o l e l i f e t i m e , D M is stably a h u m a n m i n d ; and,

moreover, the m i n d o f a h u m a n being. W h i c h one? Here enters
the t h i r d and last principle packed i n t o the p r i m e lifetime fact,
a principle about the c o u n t i n g o f i n d i v i d u a l h u m a n minds:
(One-One) O n e m i n d , one m a n ; one m a n , one m i n d .
T h e one-one principle tells us that g i v e n a certain m i n d — s a y ,

D M — t h e r e is t h r o u g h o u t its lifetime o n l y one h u m a n b e i n g
13. A symmetric of-ness principle was seen to govern, in the letter to

Mesland, human bodies. What makes DB throughout its life a human body
is that it is the body of a human being.
(and b y Mesland's letter, h u m a n b o d y ) i t i n f o r m s . B y this p r i n -

ciple ( i f n o t b y the previous t w o ) , the familiar philosophical
fantasies ( w h i c h , alas, since L o c k e dominate the subject called
"personal i d e n t i t y " ) o f m i n d swapping between t w o humans
are r u l e d o u t . T h a t is, D M m a y n o t t u r n i n t o the m i n d o f
Princess Elizabeth; her m i n d , E M , cannot become Rene D e s -
cartes's m i n d .
I t should also be noted that the one-one p r i n c i p l e is u n i t a r i a n
i n the opposite d i r e c t i o n . G i v e n a certain h u m a n being—say
Rene Descartes—he m a y have t h r o u g h o u t his life o n l y the
m i n d (and b o d y ) he was engendered w i t h ; he m a y get perhaps
another a r m or leg, b u t w h e n i t comes t o his m i n d i t is the
original or n o t h i n g . 14
W e m a y w e l l appreciate that these three f o r e g o i n g p r i n c i -

ples—stability, of-ness, and one-one—are n o t natural i n the ge-
neric whatness conception. L e t D M ' s essence be s i m p l y "a
t h i n k i n g t h i n g . " First, there is n o t h i n g n o w t o b l o c k D M ' s be-
c o m i n g the m i n d o f an angel; i n essence, angel Gabriel's m i n d
and D M are alike. Second, there is n o t h i n g t o force of-ness o n
D M . T h e nature o f the m i n d is n o t determined b y the specific
k i n d o f b e i n g whose m i n d i t is. Rather, the nature o f the m i n d
is determined first, generically, as i t were, as a t h i n k i n g t h i n g .
Its serving as the t h i n k i n g apparatus o f a h u m a n b e i n g is part
o f the historical vicissitudes befalling that already essentially
determined m i n d .
T h i r d , i n the generic conception the one-one p r i n c i p l e is
o b v i o u s l y false. O n e , starting w i t h a g i v e n m i n d , i t is conceiv-
14. Some modern anti-Lockean theorists of personal identity find attrac-

tive an analogous view about the brain. The attractiveness is independent
of whether one also holds the further reductive thesis that the mind is iden-
tical to the brain. I , for one, would hold with Descartes that my mind and
brain are two subjects. But I would also hold with Descartes the integrative
dualist that both abide by the one-one principle. Such intuitions about the
brain suggest, as mentioned above, that Descartes would have been better
off using the brain-body rather than arm-body model in his fourth replies.
9 6 W H A T A M I?
able that this t h i n k i n g t h i n g becomes the t h i n k i n g u n i t o f an-

other m a n - b o d y u n i t . A n d easy t o conceive or n o t , there is
surely n o t h i n g i n the nature o f the m i n d to tie i t to the p a r t i c u -
lar h u m a n b e i n g i n w h i c h i t started its life (assuming i t is o n l y
then that the m i n d ' s life started, an assumption m a n y a separat-
ist dualist denies). T w o , g i v e n a certain h u m a n b e i n g , i t is
readily conceivable—recall again the familiar philosophical fan-
tasies—that this b o d y w o u l d be r u n b y another t h i n k i n g t h i n g ,
a different p i l o t for the same ship. 15
I n closing, w e m a y summarize the difference between the

abstractionist and lifetime theories o f essence i n the f o l l o w i n g
terms. T h e abstractionist begins b y abstracting f r o m the real
m a n — f r o m that real i n d i v i d u a l and f r o m the specific real k i n d
o f b e i n g i t is. T h e basic ingredients w e start w i t h are a m i n d ,
a t h i n k i n g t h i n g , and a b o d y , an extended t h i n g . Inasmuch as
a m a n emerges, i t is the result o f a subsequent h o o k u p o f these
t w o ingredients. T h i s late linkage lends c r e d i b i l i t y t o the idea
that the man's m i n d can be conceived w i t h o u t his b o d y and
vice versa; i t also lends c r e d i b i l i t y t o the assertion that one can
exist w i t h o u t the other.
T h e integrative o u t l o o k begins w i t h the real h u m a n b e i n g .
T h e quest is t o better understand the p r i m a l answer (a m a n ) t o
the p r i m a l question ( W h a t am I ? ) . I t turns o u t that such an
understanding must proceed i n t w o stages. First, w e are b e i n g
s h o w n that inside each such real m a n , a h u m a n m i n d and b o d y
15. This concerns the classical separatist conception. A variation on it

goes like this: The freedom of mind swapping is accorded to a given hu-
man body and being but not to a person. In this variant separatist picture,
the identity of the person is given in terms of his or her mind, whose iden-
tity is specified independently and does not depend on that of the human
being. What is more, in this picture, unlike the integrative picture that
emerges from the letter to Mesland, a person may associate in a lifetime
with many different (human) bodies and beings. Such accounts are sug-
gested by the "personal identity" literature, which I discuss in detail in
chapter 3.
must be discerned. Second, h a v i n g ascertained this d i s t i n c t i o n ,

w e are s h o w n h o w the m i n d and b o d y are integrated inside
that p r i m a l i t e m , the m a n . N o w h e r e along the p r o o f o f the
distinction do w e forget where w e came f r o m ; that is, w e never
abstract so m u c h that w e forget that the m i n d and b o d y w e are
to distinguish are ( i ) a human m i n d and b o d y and, moreover,
( i i ) o f a particular h u m a n b e i n g , for example, RD.
T h e k i n d s o f m i n d and b o d y w e are operating w i t h are es-
sential f r o m the outset: T h e y are those o f a m a n . T h i s sets the
limits for what changes either can endure in two ways.
First, just as w e get restrictions f r o m the fact that the subject
o f change is a r i v e r or a piece o f w a x , the changes open
to a h u m a n m i n d ( b o d y ) are different f r o m those open t o an
angelic o r d i v i n e m i n d o r those for a d o g o r tree b o d y . Second,
further restrictions are induced b y the particular h u m a n be-
i n g whose m i n d ( b o d y ) is under discussion. As i t is w i t h ,
say, rivers—the changes i n geological t i m e open t o the N i l e
are n o t those open to the A m a z o n — i t is for the h u m a n m i n d
and b o d y : different h u m a n minds (bodies), different change
potentials.
W e m a y summarize w h a t the integrative f r a m e w o r k says
about the h u m a n m i n d i n six principles that are symmetric t o
those i t offered for the h u m a n b o d y :
(IDMi) Ontologicalprimality. T h e h u m a n b e i n g , for exam-

ple, R D , is the e n t i t y i n terms o f w h i c h i t is speci-
fied w h a t D M is, that is, the m i n d o f a m a n . Conse-
q u e n t l y , the conditions o f D M ' s ( i ) existence and
( i i ) i d e n t i t y are specified i n terms o f the m a n whose
m i n d i t is.
(IDM2) Essence. D M is assigned a f u l l - b l o o d e d essence: I t is
the m i n d o f a h u m a n b e i n g .
(IDM3) Existence. D M existentially depends o n R D (and
vice versa).
(IDM4) Endurance. D M is a historical subject, e n d u r i n g
t h r o u g h change.
98 W H A T A M I?
(IDM5) Substance. D M is a substance i n the categorical

sense—a subject o f a category o f predications that
does n o t itself inhere as a p r o p e r t y i n a subject.
(IDM6) Real distinction. D M is really distinct, i n the categor-
ical sense i f n o t the existentiaL f r o m R D . 1 6
16. As noted in the preface, integrative dualism excludes the coherence

of the immortality of the soul. This is not a "surprise" but part and parcel
of the view, just as separatist dualism makes room for the "coherence" (if
not possibility) of such survival. I do not wish to engage the question of
how essential was the immortality thesis to Descartes and whether he really
believed in it. I think it is more illuminating to explain what framework of
his allows it and what framework excludes it.
T H R E E
The Real Man
3.1. T h e Real D i s t i n c t i o n R e v i e w e d
By attending t o their different essences, integrative dualism

proves D M and D B t o be n u m e r i c a l l y distinct. T h i s preserves
one o f o u r four m a j o r assumptions—(iv) whatness separabil-
i t y — b u t w i t h a t w i s t because w e keep the letter o f the assump-
t i o n b u t change its spirit. F o r n o w , the essence o f each o f the
t w o subjects is n o t generic. W h a t D M is is the m i n d o f a m a n ;
w h a t D B is is the b o d y o f a m a n . T h u s D M and D B differ i n
their essential properties. Subjects that differ over properties are
t w o distinct subjects. W e conclude that D M and D B are numeri-
cally distinct. H o w e v e r , are they really distinct? T h i s leads to
the r e m a i n i n g three assumptions.
DM and DB: Complete Subjecthood and

Existence Apart
W h e t h e r D M and D B are really distinct depends o n o u r under-

standing o f the n o t i o n o f substance, o r complete subject. Sup-
pose w e fix o n the existential sense o f complete subject: A c o m -
plete subject (substance) is one that can exist all b y itself. T h u s
an existential real d i s t i n c t i o n between p u r p o r t e d subjects A and
B demands that at least one o f t h e m can exist w i t h o u t the other.
In this existential i n t e r p r e t a t i o n , integrative dualism denies
complete subjecthood t o R D , D M , and D B . I n t u r n , the pair
D M and D B (as w e l l as D M - R D and D B - R D ) are n o t existen-
t i a l l y really distinct.
99
ioo W H A T A M I?
O n the other h a n d , let us assume n o w the categorical c o n -

ception o f substance: Genuine subjecthood, i n the d o m a i n o f
h u m a n beings b u t also w i t h e v e r y t h i n g else, lies n o t i n isolated
existence b u t i n the isolation o f subjects o f predication. T h i s
conception proposes that m a n y real subjects are b y their v e r y
nature existentially interdependent. T h i s k i n d o f interdepen-
dence is n o t read merely i n the " w e a k " m o d a l sense: O f neces-
sity, subjects A and B coexist. Rather, a " s t r o n g " sense o f de-
pendence is contemplated here: B y their v e r y nature ( w h a t each
is), subjects A and B coexist.
W e have turned here the existential conception o f real subjects
o n its head. T o t a l existential independence f r o m all others is n o t
a m a r k o f being real b u t rather o f being essentially unlike the
reals o f the w o r l d . A real subject is embedded i n the nexus o f
reals. A n d to be so embedded is to depend i n the strictest w a y
o n others i n the nexus. D e p e n d i n g o n w h a t k i n d o f t h i n g one is,
one existentially depends o n different kinds o f items. So, real
subjects are existentially not immaculate b u t are real subjects
nonetheless: T h e y are not properties or traits that describe sub-
jects or ways o f being o f subjects; they are the subjects—the
beings—proper. I t is o n l y i n this categorical sense o f "complete
subject" and "real distinction" that integrative dualism preserves
our assumption (i)—complete subjecthood o f D M and D B .
T h e r e may be a place i n the objections and replies t o the
Meditations where something v e r y close t o o u r conclusion is
being debated, that is, the feasibility o f nature separability w i t h -
out (conceivable and) existential separability. I have i n m i n d
the comment Father Caterus makes o n Descartes's m i n d - b o d y
p r o o f at the end o f the first objections ( C S M I I , 72—73). Ca-
terus points out that Scotus conceived o f a f o r m o f d i s t i n c t i o n —
and called i t a f o r m a l d i s t i n c t i o n — i n w h i c h w e have " t w o f o r -
mal concepts that are distinct p r i o r to any operation o f the
intellect." God's mercy and justice m a y be a case i n p o i n t .
Nonetheless, says Caterus, " I t does n o t f o l l o w that because jus-
tice and mercy can be conceived apart f r o m one another they
can therefore exist apart" ( C S M I I , 72—73).
THE REAL M A N 101
Descartes replies that this k i n d o f a d i s t i n c t i o n is a m o d a l

one ( C S M I I , 85—86). H e reserves the relation t o nonsub-
stances, specifically, t o " m o d e s , " or w h a t he also calls " i n c o m -
plete entities." F o r this k i n d o f e n t i t y , a m o d a l d i s t i n c t i o n arises
w h e n w e conceive o f the e n t i t y b y "abstraction o f the i n t e l l e c t . "
I t is t h e n possible t o conceive o f G o d ' s justice w i t h o u t his
m e r c y and vice versa; also conceivable apart are the shape and
m o t i o n properties ( " m o d e s " ) o f a t h i n g . B u t , asserts Descartes,
these " e n t i t i e s " are incomplete; they cannot exist i n their o w n
r i g h t because there is n o complete understanding o f the shape
( m o t i o n ) w i t h o u t the t h i n g whose shape ( m o t i o n ) i t is or o f
justice and m e r c y w i t h o u t the person w h o is just and m e r c i f u l .
F r o m this, Descartes argues as f o l l o w s : Because o f this existen-
tial dependence, " I cannot imagine there t o be m o t i o n i n some-
t h i n g w h i c h is incapable o f shape o r shape i n something w h i c h
is incapable o f m o t i o n " ( C S M I I , 86). A n d , concludes Des-
cartes, the m i n d - b o d y conceivable separability is v e r y different
indeed: H e r e w e do have "complete entities" and n o modes.
A n d here the conceivability apart is indicative o f a genuine
existential separability.
T h e Descartes o f this r e p l y is the existential separatist. The
p o i n t Caterus is (almost) m a k i n g supports o u r integrative d u a l -
ist Descartes.
The question is h o w t o connect ( i ) nature separability,
(ii) conceivable separability ( " i n the u n d e r s t a n d i n g " ) , and
( i i i ) existential separability. A c c o r d i n g t o Descartes, w e may
have conceivable separability w i t h o u t existential separability
o n l y for incomplete entities ("abstractions") o r modes. I n the
case o f complete entities, conceivable separability reflects the
availability o f existential separability.
Let us grant Descartes quite a b i t . First, there can be no exis-
tence and understanding o f a shape (or m o t i o n ) w i t h o u t that o f
w h i c h i t is a shape (or m o t i o n ) . Let i t also be granted that w i t h o u t
existential separability, there can be no conceivable separability.
T h i s is still n o t enough to force o n us that there can be no nature
separability w i t h o u t conceivable and existential separability.
i02 W H A T A M I?
I t m a y w e l l be the case that t w o candidate items—each as

complete a subject as c o u l d be—cannot exist apart; each exists
as a subject o f predication and w i t h its o w n nature. I suspect that
Caterus and Scotus may have realized this. W h a t Caterus literally
says ( C S M I I , 7 2 - 7 3 ) could r u n either way. H e may simply say
that we can conceive apart the existentially inseparable. I f so, his
point does not anticipate the thesis o f integrative dualism-nature
separability w i t h o u t existential and conceivable separability.
But something i n his sentences—(i) " T h e f o r m a l concepts
are distinct p r i o r t o any operation o f the i n t e l l e c t " and ( i i ) "Jus-
tice and m e r c y can be conceived apart of one another"—suggests
to me another reading. R e g a r d i n g ( i ) , Descartes's rheotorical
slide n o t w i t h s t a n d i n g , i t is quite clear the " f o r m a l concepts" i n
question are n o t taken b y Caterus t o be incomplete entities or
"intellectual abstractions." R e g a r d i n g ( i i ) , w e must take notice
o f a slippery a m b i g u i t y i n the l o c u t i o n "conceived apart o f one
another." I n the first, existential reading, this means that x and
y can be conceived t o exist w i t h o u t each other. T h i s is the w a y
i n w h i c h Descartes reads the l o c u t i o n . B u t w e m a y understand
i t i n the existence-free numerical r e a d i n g — x and y can be c o n -
ceived as t w o separate entities, each a distinct e n t i t y , w h e t h e r
or n o t they can be conceived t o exist w i t h o u t each other. T o
m y ear, the previous sentence—"Formal concepts are distinct
. . . "—makes the existence-free conceiving-as-an-entity-apart
( t h o u g h dependent o n others) m o r e plausible.
W e also need n o t accept Descartes's v i e w ( o r even its re-
mote A r i s t o t e l i a n ancestry) that the existential dependence of
shape, m o t i o n , justice, and m e r c y makes t h e m less real beings
or mere modes. W e just saw that inside the lifetime metaphysics
f r a m e w o r k , the m a r k o f b e i n g real is existential dependence o n
other reals. W e m a y v e r y w e l l t h i n k ( I d o ) that m i n d and b o d y ,
t o o , taken as genera, o r h u m a n m i n d and h u m a n b o d y , as
k i n d s , are existentially dependent o n the existence o f specific
(human) minds and bodies. I n any event, i f w e t h i n k that exis-
tential dependence makes the candidate a nonsubstance, w e i n -
THE REAL M A N 103
v o k e an independent thesis, one separate f r o m the particular

issue I am after here: C o u l d w e have t w o real items whose
natures are separate yet are neither existentially n o r conceivably
separable?
I n one reading o f Caterus and Scotus, I see r e c o g n i t i o n o f
this possibility: P r i o r t o any intellectual operation, i n r e r u m
natura, w e have f o r m a l nature separation o f x and y. Thus x
and y can be conceived apart f r o m each other, as distinct e n t i
ties. Y e t , i n n o existential locus respectful o f these f o r m a l na
tures m a y χ be present w i t h o u t y and vice versa. I n conse
quence, i f o u r c o n c e i v i n g is t o describe such an existential
separation and i t is t o be respectful o f the natures o f x a n d j y ,
i t w o u l d fail.
DM and DB: Conceivable Separability
O f the four assumptions o f chapter i , three are preserved—

complete subjecthood, m o d a l indiscernability, and whatness
separability. W e come n o w t o the most delicate o f the f o u r , the
assumption that D M can be conceived t o exist w i t h o u t D B .
T h e discussion o f integrative dualism raises three separate
questions about conceivability. T h e r e is, first, the question o f
conceivability of disembodiment: M a y D M be conceived t o exist
w i t h o u t any b o d y — i n particular D B — o n hand? T h e r e are,
next, t w o questions related t o the " o n e m a n , one m i n d " and
"one mind, one m a n " principles. T h e questions are often
l u m p e d i n t o one i n the popular personal i d e n t i t y discussions,
where " m i n d - s w a p p i n g " ( o r " b o d y - s w a p p i n g " ) cases are i m a g
ined as a matter o f course. B u t there are really t w o separate
questions t o be asked. O n e concerns the conceivability of mind
migration: M a y w e conceive o f this single m i n d , D M , that i t
migrates f r o m b e i n g the m i n d o f one m a n , R D , t o b e i n g the
m i n d o f another man? T h e complementary question concerns
the conceivability of mind replacement: M a y w e conceive o f this
i04 W H A T A M I?
single m a n , R D , as h a v i n g i n i t i a l l y one m i n d , D M , then h a v i n g

i t replaced b y another m i n d ? 1
T h e integrative dualism (one-one) principle does n o t settle

the last t w o questions. T h e (one-one) principle is conceivabil-
ity-free. I t regards the existence o f the items proper. I t asserts
that a single m a n has o n l y one m i n d t h r o u g h o u t its lifetime;
f u r t h e r m o r e , a single m i n d is t h r o u g h o u t its lifetime the m i n d
o f one m a n o n l y . T h i s statement concerns w h a t befalls these
objects d u r i n g their lifetime. I t does n o t settle the question o f
w h a t w e may conceive about t h e m .
O f course, w i t h the help o f some extra principles, w e m a y
get a q u i c k answer t o the conceivability queries. F o r example,
suppose i t were n o t just true o r even necessarily true b u t i n the
v e r y nature o f D M t o be the m i n d o f one m a n o n l y and the
v e r y essence o f R D to have one fixed m i n d ; suppose also that
we agreed that whatever is flouting whatness principles is i n
conceivable. W e w o u l d then have answers t o the above c o n
ceivability questions.
B u t w e have n o t yet e x p l i c i t l y asserted all these principles
j o i n t l y . A n d even i f w e i m p l i c i t l y d i d , there remain the p r i m a
facie contrary conceivability i n t u i t i o n s t o deal w i t h , an intricate
matter t o w h i c h w e n o w t u r n .
Conceivability Illusions I: Disembodiment
O n e o f the marks o f the shift f r o m separatist t o integrative

dualism is the reversal o f the relationship between conceivabil
i t y i n t u i t i o n s and whatness facts. Separatist dualism starts w i t h
conceivability intuitions. I t then struggles t o make the relevant
whatness (essence) claims respect the i n i t i a l conceivability i n t u
itions. B u t i n the m i n d and heart o f a separatist dualist, the
ι. In the posing of these questions, the standard literature assumes that

the person in question, Descartes, is not to confused with the man, the hu
man being. When I speak of mind migration and replacement, I allude to
the man Descartes.
THE REAL MAN 105
conceivability i n t u i t i o n is the most basic fact, w h i c h all others

must be made t o respect.
I n t e g r a t i v e dualism reverses this procedure. W e do n o t start
w i t h o u r concepts and conceptual apparatus for conceiving mind
and b o d y . W e start w i t h the objects o u t there and p r i m a r i l y
w i t h the m a n , Rene Descartes; i n t u r n , w e continue w i t h his
m i n d and b o d y , t w o subordinated objects, and no concepts
whatsoever. T h e logic o f investigation has n o w been reversed:
The facts about what each o f the trio—man, mind, and
b o d y — i s are g i v e n d i r e c t l y , p r i o r t o , and independently o f c o n -
ceivability experiments. T h i s , indeed, is the p o i n t o u r discussion
has reached. W e first analyzed, f o l l o w i n g the integrative b l u e -
prints l a i d o u t i n the Mesland letter, the whatness o f the m i n d
and b o d y o f the m a n R D . W e n o w face the subsequent question
o f w h a t m a y be conceived o f each.
N o w that this is the order o f investigation, a n e w possibility
l o o m s , illusions of conceivability. T y p i c a l l y , i n such an i l l u s i o n
case, w e seem t o conceive o f a g i v e n subject that i t is a certain
w a y . B u t u p o n c o n s u l t i n g w h a t this subject is, i t turns o u t that
w e have n o t — w e c o u l d n o t have—really conceived o f this sub-
ject that i t is this w a y because its b e i n g this w a y flouts w h a t
this subject is. So, o u r c o n c e i v i n g has never really latched o n
to this subject. W e had conceived o f some other subject and
were taken b y the i l l u s i o n that w e had successfully conceived
o f this original one. Illusions are to be expected: W i t h the w h a t -
ness o f subjects g i v e n p r i o r to our conceivings, " c o n c e i v i n g " has
become a "success v e r b . " There is n o w an independent basis for
checking whether w e merely seemed to have conceived a certain
specific item or have really succeeded i n d o i n g so.
T h i s standpoint provides an interesting explanation o f at
least one conceivability i l l u s i o n o f chapter i , one had b y m a n y
separatist dualists and thus b y all o f us at first blush: the c o n -
c e i v a b i l i t y o f disembodiment. A l m o s t e v e r y b o d y i n m y classes
assents t o the conceivability o f D M w i t h o u t D B . T h e question
is, h o w does this i l l u s i o n arise?
io6 W H A T A M I?
T h e explanation I suggest is this: W h a t w e really conceive

is the existence o f a t h i n k i n g b e i n g (note the italicized indefinite
article) w i t h o u t any extended object i n the story; o r , w e c o n
ceive o f an extended object w i t h o u t any t h i n k e r a r o u n d . W h a t
w e go o n t o do is t o misdescribe this successful conceiving. W e
declare that w e have managed t o conceive o f a specific t h i n k i n g
subject, D M , that i t exists w i t h o u t extended objects around.
A n d w e feel that this is an adequate description because the
concept under w h i c h w e carry the conceivability experiment is
taken t o be the full essence o f the p u r p o r t e d subject; that is,
" t h i n k i n g b e i n g " is the full essence o f D M . I f o u r conception
respected i n the story the full essence o f D M , w h a t else c o u l d
disqualify the experiment?
N o t h i n g , really, b u t w e must be careful t o latch o n t o the
full essence o f D M , n o t just to a t h i n , abstractionist-generic
segment o f i t . Suppose n o w that the essence o f D M is the robust
p r o p e r t y "is the m i n d o f a h u m a n b e i n g . " W e can still h o l d o n
to the general p r i n c i p l e ; w h a t w e conceive w h i l e respecting
D M ' s full essence shows us something about w h a t can befall i t .
B u t n o w i t is excluded b y D M ' s v e r y nature, that i t exists w i t h
out b e i n g the m i n d o f an existing h u m a n being. G i v e n our
assumptions about the existence conditions o f a h u m a n b e i n g ,
it is s i m p l y no longer conceivable o f D M — w h i l e respecting its
full essence—that i t exists disembodied.
T h e f o r m o f e x p l a n a t i o n — i n c l u d i n g the defusing mechanism
that explains the i l l u s i o n o f c o n c e i v a b i l i t y — f o l l o w e d here m a y
r e m i n d the reader o f the one w e discussed at the end o f chapter
ι, i n the appendix dedicated t o Saul K r i p k e ' s w o r k . O f course,
K r i p k e is focused o n the connection t o real possibility and ne
cessity, whereas w e have focused o n the weaker l i n k t o mere
consistency w i t h w h a t the subject is. Nonetheless, structurally,
the similarity is s t r i k i n g . O n e p o i n t is especially central—usu
ally, w e start, l o o k i n g for m e n t a l l y accessible evidence, with
conceivability intuitions. W e t r y t o find out facts about the o b
ject's essence (possibility). Such is the order o f evidence gather
i n g — f r o m conceivability t o possibility. B u t things r u n i n a re-
THE REAL M A N 107
verse order when i t comes to the metaphysical issue of

u n f o l d i n g what makes the pertinent facts o b t a i n . F o r example,
w h e n w e say that i t is n o t possible for this table t o be made o f
ice, w h a t makes this fact h o l d has n o t h i n g t o do w i t h conceiv-
a b i l i t y about this table. W h a t is the basis for o u r m o d a l fact is
conceivability-free and g i v e n d i r e c t l y i n terms o f w h y this k i n d
o f p r o p e r t y ( w o o d e n ) b u t n o t that one ( b e i n g b r o w n ) is neces-
sary. So here w e see that facts o f possibility ( f o r that matter,
w h a t the t h i n g is) precede facts about w h a t is genuinely c o n -
ceivable. I t is exactly this s w i t c h c o n c e r n i n g the precedence o f
whatness t o conceivability that integrative dualism performs
v i s - ä - v i s the m e t h o d o l o g y o f separatist d u a l i s m . 2
Conceivability Illusions II: Man-Mind Swaps
T h e preceding discussion rules o u t the conceivability o f disem-

b o d i m e n t b y u p h o l d i n g t w o ideas. First, the p r o j e c t i o n p r i n c i -
ple—whatever is conceivable is whatness-consistent—but now
read contrapositively: W h a t e v e r is whatness—inconsistent is i n -
2. Lurking here in the background is a question that goes beyond our

understanding of Descartes and concerns the metaphysics of essence and ne-
cessity per se: What may serve as conceivability-free grounds for projecting
some properties as whatness pertaining and as necessary? I speculate about
this (and about Kripke's ideas) in "The What and the How I I , " Nous,
1996, and "Nothing, Something, Infinity," Journal of Philosophy, September
1999. Further developments regarding what-x-is, e.g. is the whatness in-
duced by the generative process that brought x into being, are pursued in
"The What and The How I I I , " in press. But see also the discussion in the
following pages. In a different framework, a related departure from conceiv-
ability as the basis of (telescope into) necessity and essence is contemplated
by Tom Nagel and Torin Alter op cit. ch. 1. Although our discussion is re-
stricted to Descartes, both Nagel and Alter discuss a variety of modern
claims in the philosophy of mind based on the use of conceivability exer-
cises (e.g. the possibility of zombies and the possibility of sheer electric cir-
cuitry in a "computer" that induces consciousness). I sympathize with their
discomfort with many such alleged possibilities but will not engage here in
these contemporary controversies.
io8 W H A T A M I?
conceivable. Second, w e insist that w h a t D M is is the m i n d o f

a h u m a n being. G i v e n that such a m i n d cannot exist w i t h o u t a
h u m a n whose m i n d i t is and the latter cannot exist w i t h o u t a
b o d y , D M cannot, i n any sense, exist disembodied.
T h i s is still consistent w i t h c o n c e i v i n g this m i n d , D M , as
the m i n d o f another h u m a n b e i n g , say, Elizabeth, and thus as
connected to another b o d y , Elizabeth's b o d y ( E B ) . Such m i n d
m i g r a t i o n stories are familiar to us f r o m the literature o n per-
sonal i d e n t i t y . A l s o consistent so far is the c o n c e i v i n g o f this
m a n , R D , as shedding away his current m i n d , D M , and adopt-
i n g a n e w m i n d , say, Elizabeth's. T o exist, a m a n needs a man's
m i n d , b u t i t does n o t have to be his o r i g i n a l m i n d .
N o w , w h e n I speak o f the "consistency" o f such stories, I do
n o t mean that they represent real possibilities for the subjects
c o n c e r n e d — R D , D M , and D B . I n t e g r a l t o the integrative d u a l -
ism discussed above is the (one-one) p r i n c i p l e . I t guarantees
that t h r o u g h o u t the lifetime o f these subjects and i n whatever
counterfactual possibility open to t h e m , they are b o u n d t o each
other. Nonetheless, i t is consistent w i t h w h a t D M i s — w i t h its
sheer nature—that i t w o u l d be the m i n d o f another h u m a n be-
i n g ( " h u m a n a n i m a l " w o u l d say the supporters o f the v i e w , just
to d r i v e the p o i n t h o m e ) . A n d i t is consistent w i t h the nature
o f this m a n , Descartes, that he w o u l d have another m i n d .
A s m e n t i o n e d , these whatness-consistent stories are the basis
for the conceivability claims o f personal i d e n t i t y theories, w h i c h
actually assert the yet stronger claim o f the real possibility o f
such swaps. B u t this m a y be due to a general lack o f distinction
i n this literature between a real possibility for x and a mere
consistent story about x. W i t h the distinction n o t e d , w e m a y
reread the familiar stories as asserting the weaker claim o f
whatness-consistent scenarios.
W h a t is proposed b y such swap stories varies. F o r some
w r i t e r s , i t is consistent w i t h w h a t the h u m a n b e i n g R D ( n o t t o
be confused, they w o u l d say, w i t h the person RD) is that the
m a n w o u l d have another m i n d . F o r almost everyone w h o is
friendly t o such swap stories, i t is obvious that i t is consistent
THE REAL M A N 109
w i t h w h a t D M that i t w o u l d be l o d g e d i n another h u m a n b o d y
and, for that matter, another h u m a n b e i n g .
T h e v i e w just o u t l i n e d is certainly appealing. A m o n g m a n y
critics o f classical separate dualism, the v i e w just expounded is
a natural balance p o i n t . T h e y a d m i t that t o assert the conceiv-
a b i l i t y o f full disembodiment is a mistake. B u t equally mistaken
is the attempt t o deny o u r a b i l i t y t o conceive o f D M w i t h o u t
this particular b o d y , D B , and this particular h u m a n , R D , or to
deny o u r a b i l i t y to conceive o f this h u m a n b e i n g w i t h o u t this
v e r y specific m i n d , D M . I call this f o r m o f integrative dualism
generic integrative dualism because i t assigns t o D M and D B the
still somewhat generic essence—the m i n d ( b o d y ) o f a h u m a n
b e i n g , o f some h u m a n or other. 3
Generic vs. Tight Integration
Descartes's texts do n o t contain enough i n f o r m a t i o n t o decide

the f o l l o w i n g question: A s s u m i n g that w e read Descartes ac-
c o r d i n g t o integrative lines, w o u l d he have opted for the ge-
neric f o r m o f the doctrine o r a yet t o be expounded, t i g h t e r
variant? Rather than argue that one o f these views is D e s -
cartes's o w n , I w i l l discuss b o t h forms o f i n t e g r a t i o n .
A c c o r d i n g t o the t i g h t e r f o r m o f i n t e g r a t i o n , i t is n o t really
conceivable o f this m i n d , Descartes's, that i t exists and serves
as the m i n d o f any other m a n . N o r is i t conceivable o f this m a n
that he has any other than his actual m i n d and b o d y . A n d this
m u c h is inconceivable because w h a t D M is is n o t just the m i n d
o f a h u m a n b e i n g — a n y o l d o n e — b u t also the m i n d o f this h u -
m a n b e i n g , Rene Descartes. W h a t is asserted here i n effect is
3. I focus here and below on a certain variant of the personal identity

literature, one that insists that DM has to exist in a human being and body.
Other variants would see any such attachment to this specific biological spe-
cies of beings and their bodies as overrestrictive. The problems pointed out
below in the more plausible variant apply of course to the more liberal
stories.
no W H A T A M I?
that the force o f the one-one p r i n c i p l e o f integrative dualism is

n o t that o f mere necessity; rather, i t pertains t o the v e r y nature
o f the h u m a n m i n d and m a n whose m i n d i t is. I call this strict
version o f integrative dualism tight integrative dualism because
o f the m a x i m a l l y specific whatnesses i t assigns. W h a t D M is is
the m i n d o f this h u m a n b e i n g , Rene Descartes.
T w o theses proposed b y this m a x i m a l l y specific i n t e g r a t i o n
call for some explanation. First, w e seem t o be saying that D M
and E M (Elizabeth's m i n d ) do n o t share whatness. T h i s runs
against the strong i n t u i t i o n that b o t h are the same k i n d o f t h i n g ,
a h u m a n m i n d . T h e second p o i n t concerns conceivability issues.
Familiar m a n - m i n d swapping stories (e.g., the prince-cobbler
swap) suggest that m i n d relocation and m i n d replacement are
coherent. I t is m y belief that b o t h denials are n o embarrass-
ments b u t insights offered b y the t i g h t e r f o r m o f integrative
dualism. L e t me support this claim.
First, concerning whatness, I do agree that D M and E M are
the same k i n d o f t h i n g , a h u m a n m i n d . I n familiar cases, this
coheres w i t h the claim that the objects share whatness. T o ask
o f an o r d i n a r y i n d i v i d u a l l i k e F i d o , " W h a t is F i d o ? " leads t o
the same answer as w h e n w e ask, " W h a t k i n d o f t h i n g is
F i d o ? " B u t w h e n i t comes t o the " d e r i v a t i v e " (as contrasted
w i t h the m a n Descartes) of items, for example, the m i n d and
b o d y o/Descartes, whatness and k i n d questions d r i f t apart.
Before I justify the p o i n t for D M and D B , let us consider a
case f r o m outside the m i n d - b o d y realm. Consider t w o sets: sin-
gleton Descartes, {Descartes} and singleton Elizabeth, {Eliza-
b e t h } . N o w , asked t o say w h a t is { E l i z a b e t h } , I w o u l d answer
w i t h the set whose sole member is Princess Elizabeth; and I
w o u l d give the corresponding b u t different answer for " W h a t
is {Descartes}?": the set whose sole member is the Rene D e s -
cartes. But n o w , asked t o say w h a t k i n d o f t h i n g {Elizabeth}
is, I w o u l d say: a set, indeed, a singleton set. I w o u l d g i v e the
v e r y same answer for {Descartes}. W e get sameness o f k i n d
b u t different whatnesses.
T H E R E A L MAN in
T h e reason seems t o me t o be this: {Elizabeth} is n o t as

" p r i m a l " an i n d i v i d u a l as Elizabeth is. T o specify w h a t the set
is, w e depend essentially o n the m o r e p r i m a l i n d i v i d u a l , the
member o f the set. T h e k i n d o f t h i n g that all singleton sets are
is the same, that is, singleton sets. T h i s classifies t h e m all b y
l o o k i n g for a superordinate l e v e l — t h e i r k i n d . A n d this w o u l d
be the w h o l e story were each such singleton a p r i m a l i t e m , n o t
one whose o w n whatness and i d e n t i t y are specified i n terms o f
a m o r e p r i m a l i t e m . T h u s , for Elizabeth o r F i d o o r the L o i r e ,
all there is t o the whatness is the " u p w a r d " - l o o k i n g k i n d classi-
fication. T h e r e s i m p l y is no further " d o w n w a r d " specification
to be made i n terms o f m o r e basic i n d i v i d u a l s . B u t for the
singleton sets o f all these i n d i v i d u a l s — { E l i z a b e t h } and others—
the situation changes. T h e r e is the c o m m o n u p w a r d explanation
i n terms o f the category they b e l o n g t o — t h e i r k i n d — b u t there
is also the d o w n w a r d explanation, different for each, i n terms
o f the i n d i v i d u a l s they are the set of. 4
W h a t is g o o d for {Elizabeth} is g o o d for her m i n d and

b o d y . T h e items i n question are g i v e n their whatness i n d o w n -
w a r d terms that relate t o the m a n of w h o m they are the m i n d
and b o d y . F o r the superordinate k i n d classification, D M and
E M b e l o n g i n the same k i n d . B u t the d o w n w a r d explanation
runs differently because, b e i n g o f items, they are o f distinct
p r i m a l subjects, Descartes and Elizabeth.
I n s u m , the v i e w proposed here preserves the i n t u i t i o n that
D M and E M are the same k i n d o f t h i n g , a h u m a n m i n d , that
is, the m i n d o f a h u m a n b e i n g . B u t asked t o say w h a t D M is
and w h a t E M is, I find that w e understate things i f w e do n o t
b r i n g o u t the fact that D M is the m i n d o f this h u m a n , whereas
E M is the m i n d o f this other h u m a n . G i v e n that this of-ness is
4. Of course, I here assume a certain view of Elizabeth, Fido, and the

Loire. Some may wish to contest the view. For example, suppose we view
the princess as derivative, say, "the union of EM and EB." In such a view,
her whatness and Descartes's whatness would indeed differ.
ii2 W H A T A M I?
essential to the v e r y k i n d o f t h i n g b o t h D M and E M are, i t

should come out i n the specification o f w h a t each is.
Conceiving of this Specific Human Mind
I t u r n n o w t o o u r second w o r r y — t h e seemingly obvious c o n -

ceivability o f m a n - m i n d swaps. I t is c o m m o n currency i n the
literature o n "personal i d e n t i t y " t o start w i t h conceivability ex-
periments as g r o u n d zero o f metaphysical investigations. T h e
m e t h o d o l o g y m a y w e l l be called " n e o - H u m e a n " after the prac-
tice o f D a v i d H u m e , even i f i t is another empiricist, John
L o c k e , w h o inspires the u l t i m a t e views o n the specific matter
o f personal i d e n t i t y . T h e g u i d i n g idea is that any tale that does
not r u n i n t o logical or otherwise a p r i o r i c o n t r a d i c t i o n ( a m o n g
"ideas" o r "concepts") is ascertained as " g e n u i n e l y conceiv-
able." I n this manner, ahy connection between a g i v e n existing
subject, for example, the N o t r e D a m e i n Paris, and any other
real subject is q u i c k l y dismissed as merely contingent; i n t u r n ,
any substantial p r o p e r t y , for example, "is a cathedral," is
excluded f r o m serving as nature p e r t a i n i n g . I n o u r conceiv-
i n g exercises, w e have subtracted any such candidates with
impunity.
I thus see this general m e t h o d o l o g y as evening the score
before w e confront the particular case o f conceivings o f h u m a n
minds, bodies, and beings. B u t even b y the liberal H u m e a n
standards, i t is s t r i k i n g h o w g o t h i c the scenarios concocted i n
this d o m a i n are. M i n d swaps are talked about as a matter o f
course; sometimes they are accompanied b y " b r a i n transplants,"
sometimes b y b r a i n " i n f o r m a t i o n transfers" conducted b y the
latest philosophical gadgets o r machines; persons are easily de-
tached f r o m one h u m a n b e i n g (and b o d y ) and relocated i n
other h u m a n beings (bodies), sometimes across centuries; entire
mental histories and physical makeups are transferred and reas-
signed b y " t e l e t r a n s p o r t i n g " devices dreamed u p b y Holly-
w o o d ; and so o n . W h e n the shadow o f d o u b t is expressed about
the t a n g i b i l i t y o f all this metaphysical abracadabra, one is t o l d
THE REAL M A N 113
that o u r i n a b i l i t y to really carry o u t , i n real hospitals, these

transformations is a mere "technical i m p o s s i b i l i t y . " 5
A s m e n t i o n e d , free and easy conceivability goes hand i n

hand w i t h a generic essence f r a m e w o r k . Inside this f r a m e w o r k ,
the generic essence theorist seems t o c o m m i t the e r r o r o f sepa-
ratist dualism one level higher. W e gave up o n the absolutely
generic essence " t h i n k i n g t h i n g , " b u t w e still assign the rather
generic "a h u m a n m i n d " ; i n neo-Lockean theories, w e further
b u i l d u p a b u n d l e o f mental qualities that identify the m i n d .
W e do so b y a l l u d i n g t o some desired s t r i n g i n g o f memories
and other c o g n i t i v e faculties, m a k i n g sure n o t t o b u i l d i n t o the
b u n d l e o f mental qualities ( i ) the external, specific objects and
k i n d s o f w h i c h these are memories, perceptions, and so o n and
( i i ) the h u m a n b e i n g whose memories, perceptions, and so o n
these are. W h e n asked i f w e can conceive o f this m i n d , D M ,
as existing w i t h o u t this m a n , R D , w e "translate" the question
t o this: C a n w e conceive o f some h u m a n m i n d ( w i t h , e.g., such
q u a l i t a t i v e l y g i v e n memories) that i t exists w i t h o u t this man?
T h e answer is, yes, o f course.
T h e response o f the integrative dualist should be clear b y
n o w . I t is n o t consistent w i t h w h a t D M is that i t exists and the
m a n does n o t . G i v e n this, i t is n o t really conceivable o f D M —
w h i l e respecting its full essence—that i t exists w i t h o u t this m a n .
W h a t is conceivable is that some h u m a n m i n d exists w i t h o u t
this particular m a n . B u t all o f this is n o t c o n c e i v i n g o f this v e r y
mind—Descartes's—as m i g r a t i n g t o another m a n , actual or
hypothetical.
5. In speaking of Hollywood films as the provider of evidence and not-

ing a mere "technical impossibility," I quote directly from a recent series of
lectures of Sidney Shoemaker at UCLA, Spring 2000. As mentioned, this
line of response is common currency in the contemporary literature. I wish
to note that the personal identity literature is replete with other "technol-
ogy" that (i) is connected to the free and easy use of conceivability and
(ii) is equally suspicious. To take a famous example, it is now standard to
talk about "quasi memories," and Derek Parfit speaks of "Venetian memo-
ries," where the brain of a woman called Jane, immaculate of any contact
ii4 W H A T A M I?
I n the present development o f the Cartesian integrative p i c -

ture, all such free and easy conceivability claims are un-
g r o u n d e d . T h e o r i g i n a l sin, the c o n j u r i n g t r i c k , is the H u m e a n
essence-obliviousness o f the basic n o t i o n o f seeming conceiv-
a b i l i t y ( i m a g i n a b i l i t y ) . T o r e m i n d us, the integrative Cartesian
f r a m e w o r k starts w i t h essence, one founded o n a robust endur-
ance metaphysics. T h i s means that w o r l d l y items—the stone
and the sun; the r i v e r and the piece o f wax; the h u m a n b o d y ,
m i n d , and b e i n g — n o t o n l y change i n h i s t o r y b u t also come i n t o
and go o u t o f existence i n h i s t o r y . I t is the process o f a subject's
c o m i n g i n t o h i s t o r y that determines ( i ) what, for example, this
t h i n g taken f r o m the h o n e y c o m b is and ( i i ) o f the m a n y w a x
pieces, which one was thus generated. Descriptions o f the life-
time o f the piece o f w a x rest o n w h a t i t t o o k t o generate this
l i f e - i n - t i m e i n the first place.
So far this holds o f pieces o f w a x , and i f w e abide b y t h e -
wax-is-the-model p r i n c i p l e , the same applies t o the emergence
o f h u m a n minds. W i t h i n the n e o - H u m e a n and Lockean m e t h -
o d o l o g y , i t is easy t o conceive o f D M as existing i n the h u m a n
b e i n g Plato a couple o f m i l l e n n i a before i t was actually gener-
ated i n the m a n Descartes. I t is also easy t o imagine i t just
" l e a p i n g i n t o existence," " j o i n i n g " w i t h some h u m a n b o d y t o
with the city of Venice, is implanted with "memory traces" from to a man
called Paul and gotten by him through direct perception in the piazzas and
by the canals. I see these as sheer logical constructions tagged misleadingly
to suggest kinship with an existing species of phenomena. None of these fa-
miliar constructions has anything—verbal excesses aside—to do with the
real, cognitive process of memory. I am not merely expressing doubt about
the genuineness of the possibility of applying such a supposedly coherent no-
tion of memory to beings like us. I question the very coherence of the idea
as a notion of memory for any historically real, existing species and/ or for
human beings. I do not pursue here these grave reservations. In the text, I
focus on the use of this one technology in the current literature that is (i)
relatively informal and (ii) pertinent to the sustenance of separatist dualist
theses.
THE REAL M A N 115
" f o r m " a subsequent m a n . B u t this is n o t so i n the current

integrative picture. T o replay Descartes's o w n analogy o f the
a r m - b o d y relation: A h u m a n m i n d is engendered as part o f the
natural development o f a h u m a n b e i n g and its h u m a n b o d y .
W e m i g h t w e l l speak o f its generation process as part o f the
overall h u m a n morphogenesis; o n l y w h e n a certain stage i n the
emergence o f the h u m a n b e i n g , specifically o f his o r her b r a i n ,
has been reached, w i l l a h u m a n m i n d come i n t o b e i n g . The
process is specific t o k i n d s o f beings; that is, an angelic m i n d
w o u l d n o t emerge i n the w a y a h u m a n m i n d does. T h e process
is also specific t o the v e r y h u m a n b e i n g that is its breeding
g r o u n d — t h i s m i n d , D M , g r e w as part o f the g r o w t h o f this
v e r y h u m a n b e i n g , R D . T h e genitive f o r m is absolutely essen-
tial here: I t is this being's m i n d , the m i n d o f this b e i n g .
W e are n o w ready to go back to w h a t is conceivable o f D M
as a f u n c t i o n o f w h a t i t is. I t is o n l y w i t h i n this distinct b e i n g
i n cosmic h i s t o r y , Rene Descartes, that this m i n d , D M , had a
w a y o f c o m i n g i n t o real life. I t is thus o n l y as this being's m i n d
that i t has any conceivable life.
3.2. T h e P r i m a l Q u e s t i o n a n d the P r i m a l A n s w e r
F r o m the v e r y start, Descartes's project was threatened b y the

prove too little/prove too much dilemma. I t is t i m e t o evaluate
h o w the integrative i n t e r p r e t a t i o n handles this threat.
The discussion o f existential separability and illusions o f
conceivability points o u t that Descartes has grounds for c l a i m -
i n g that he does n o t " p r o v e t o o m u c h . " Integrative dualism
does n o t separate m i n d and b o d y so radically as t o claim that
they can exist w i t h o u t the other, b l u r r i n g the sense that they
are essentially interdependent parts o f the single person each o f
us is.
Descartes is also i n a p o s i t i o n t o claim that he escapes the
" p r o v e t o o l i t t l e " h o r n . T h e p r i m a l i t y o f the h u m a n b e i n g is
the cornerstone o f the integrative i n t e r p r e t a t i o n ; the m i n d and
n6 W H A T A M I?
b o d y are g i v e n as those o f a h u m a n b e i n g . Nonetheless, the

p r i m a l i t y o f the m a n does n o t denigrate the status o f his m i n d
and b o d y to that o f mere modes. T h e m i n d and b o d y o f the
m a n are genuine, complete subjects, n o t engendered b y the ex-
istential gloss o f "subject." B u t , t h e n , neither is the m a n h i m s e l f
an existentially independent b e i n g n o r is a v a r i e t y o f other
mundane subjects dear t o Descartes—the sun, the L o i r e , M t .
Blanc, the piece o f w a x , or any other w o r l d l y i t e m — f o r all
depend o n other real beings for their existence. A n d so, s t i c k i n g
to o u r categorical n o t i o n o f a subject i n general and i n the
realm o f m a n i n particular, w e count all three, D M , D B , and
R D , as complete subjects. T h e y are existentially interdependent
but categorically separate; each is a b e i n g , n o t merely a w a y -
of-being.
Safely o u t o f the dilemma's claws, w e have left one question
unattended—the p r i m a l question. W e have n o t answered the
opening question, " W h a t am I ? "
H a v i n g been exposed t o integrative dualism about the m a n -
m i n d - b o d y t r i o , m a n y surmise that the integrative answer is
something l i k e this: W h a t I am is the u n i o n o f a h u m a n m i n d
and a h u m a n b o d y . T a k i n g this t o be the answer, m a n y suspect
a " v i c i o u s c i r c u l a r i t y . " W e have been a r t i c u l a t i n g the essence
o f D M and D B i n terms o f the h u m a n b e i n g they are the m i n d
and b o d y of; and n o w , the essence o f the h u m a n b e i n g R D is
g i v e n i n terms o f the u n i o n o f D M and D B . T h e m i n d and
b o d y explained b y w a y o f the m a n , the m a n b y w a y o f the
(his) m i n d and b o d y .
I believe such c i r c u l a r i t y is n o t vicious, p a r t l y because D e s -
cartes's quest here is n o t for a definition (see b e l o w ) . B u t m y
m a i n answer is that the proposal—what I am is the u n i o n o f a
h u m a n m i n d and b o d y — i s n o t the one g i v e n b y the integrative
picture. T h e answer i t gives o n behalf o f Descartes is the one
he gives o r i g i n a l l y i n M e d i t a t i o n I I , i m m e d i a t e l y after the ques-
t i o n m a r k o f " W h a t am I ? " — t h a t is, "a m a n . " I a l l o w m y s e l f
to call i t n o t just his first answer b u t also " t h e p r i m a l answer."
THE REAL M A N 117
I n the o r i g i n a l text o f M e d i t a t i o n I I (quoted i n this book's

preface) and the c o m p a n i o n , " T h e Search for T r u t h , " Descartes
mentions problems i n h o l d i n g t o this answer:
POLYANDER: Y o u d i d n o t w a n t t o ask a n y t h i n g w h i c h
c o u l d n o t be answered v e r y easily. So I shall say I am
a man.
E U D O X U S : Y o u are n o t p a y i n g attention t o m y ques-
t i o n and the r e p l y y o u g i v e , h o w e v e r simple i t m a y
seem t o y o u , w o u l d plunge y o u i n t o v e r y difficult and
complicated problems, were I t o press y o u even a l i t -
tle. I f , for example, I were t o ask even Epistemon
w h a t a m a n is, and he gave the stock r e p l y o f the
scholastics, that a m a n is a " r a t i o n a l a n i m a l " and, if,
i n order to explain these t w o terms ( w h i c h are just as
obscure as the f o r m e r ) , he were t o ask us further,
t h r o u g h all the levels w h i c h are called " m e t a p h y s i c a l " ,
w e should be dragged i n t o a maze f r o m w h i c h i t
w o u l d be impossible t o escape. F o r t w o questions
arise f r o m this one. First, w h a t is an animal} second,
w h a t is rational? i f i n order t o explain w h a t an animal
is, he were t o reply that i t is a " l i v i n g and sentient be-
i n g , " that a l i v i n g b e i n g is an "animate b o d y " and
that a b o d y is a " c o r p o r e a l substance", y o u see i m m e -
diately that the questions, l i k e the branches o f a f a m -
i l y tree, w o u l d r a p i d l y increase and m u l t i p l y . Q u i t e
clearly the result o f all these admirable questions
w o u l d be pure verbiage, w h i c h w o u l d elucidate n o t h -
i n g and leave us i n o u r o r i g i n a l state o f ignorance.
(CSM I I , 410) 7
So runs Descartes's n i g h t m a r e scenario o f w h a t m i g h t ensue i f

w e gave the p r i m a l answer t o the p r i m a l question. T h e passage
seems t o me t o raise three separate issues.
7. Italics are in the original.

n8 W H A T A M I?
O n e p r o b l e m concerns the p u r p o r t e d form, according to

Descartes, o f the critical t r u t h : T o be a m a n is t o be a rational
animal. A second question concerns w h a t , according t o D e s -
cartes, is the truth o f the matter; that is, is " r a t i o n a l a n i m a l " the
r i g h t answer? These t w o questions focus o n Descartes's v i e w .
A t h i r d and final question goes b e y o n d Descartes's scholarship
and enters i n t o p l a i n metaphysics: W h a t are w e d o i n g w h e n w e
say that " w h a t I am is a . . . " and, i n particular, w h e n w e inject
i n the space the c o m m o n n o u n " m a n " ?
I w i l l start w i t h the t w o Descartes-bound questions. Con-
cerning the f o r m o f the p u r p o r t e d " d e f i n i t i o n " — M a n is a ra-
tional animal—Descartes is r e l a t i n g t o a l o n g t r a d i t i o n before
h i m , a t r a d i t i o n that reached a h i g h p o i n t i n late medieval p h i -
losophy. A n d i n spite o f Descartes's discomfort w i t h i t , the
t r a d i t i o n w e n t o n t o dominate m a n y philosophical discussions,
for example, Kant's, all the w a y i n t o o u r century. R o u g h l y (and
no m o r e than a r o u g h idea is ever g i v e n ) , the " l o g i c a l f o r m "
o f the statement is that o f a " d e f i n i t i o n . " T h i s is taken t o assert
something displayed b y " M a n = d f rational a n i m a l . "
A r e the relata here concepts, properties, or names o f species
and genera? Is the i d e n t i t y sign t o be taken seriously? I do n o t
k n o w . W h a t is m o r e , the general p r e s u m p t i o n o f such definition
is that o n the r i g h t - h a n d side w e are t o get a " r e d u c t i o n , " a
relatum "made u p " o f m o r e basic ingredients (concepts, p r o p e r -
ties, etc.) than the one preceding the copula. T h i s is even m o r e
confusing because an i d e n t i t y sign is used. I n a real i d e n t i t y ,
for example, " M . M o n r o e = N o r m a Jean Baker," there is n o
reduction o n the r i g h t - h a n d side, n o r do w e get w i t h genuine
identities complaints o f c i r c u l a r i t y (arising f r o m the expectation
that w e were t o be offered a noncircular r e d u c t i o n ) .
I w i l l n o t d w e l l o n this medieval t h e o r y o f real definition
that haunted Descartes because I feel that he is absolutely r i g h t
to avoid i t . L e t me just r e c o r d , rather t h a n argue for, t w o
observations.
One observation is that even the theory's idealized m o d e l —
whatness statements i n mathematics—is mishandled. Suppose I
THE REAL M A N 119
say, " W h a t 9 is is a n u m b e r " o r " W h a t the d o u b l e t o n {Eliza-

b e t h , Descartes} is is a set," and y o u question me: " B u t w h a t
is a number? W h a t is a set?" W h a t I give y o u i n r e t u r n is n o t
a d e f i n i t i o n — n o t even an i m p l i c i t d e f i n i t i o n — o f " n u m b e r " and
"set." Perhaps some formalist philosophers of mathematics
w o u l d give y o u such a definition. B u t m y explanation o f what
a n u m b e r is and w h a t a set is w o u l d n o t be b y w a y o f a defini-
t i o n . A n d whatever f o r m i t w o u l d take, i t w o u l d use related
arithmetic notions o f order (finitude) and set theoretic notions
o f collection and w o u l d thus seem t o be " c i r c u l a r . " So, even i n
accounting for the whatness o f mathematical entities, the real
definition m o d e l seems t o me misguided.
C o n c e r n i n g the other observation, suppose w e m o v e f r o m
mathematics t o such questions as " W h a t is F i d o ? " — t o w h i c h
the r i g h t answer, i n t u i t i v e l y speaking, must be "a d o g . " You
n o w challenge me: " B u t w h a t is a d o g ? " Most l i k e l y I w o u l d
answer: "a certain k i n d o f m a m m a l i a n species"; perhaps, i n a
genetics l a b o r a t o r y , I w o u l d add: "a m a m m a l i a n species o f
characteristic D N A D " ; o r i n an e v o l u t i o n a r y b i o l o g y class, I
w o u l d add: "a m a m m a l i a n species o r i g i n a t i n g i n u r - g r o u p U " ;
and so o n . T a k e any such answer—none is a definition o f the
species d o g , the p r o p e r t y o f b e i n g a d o g , the p r o p e r t y o f b e i n g
a member o f the species d o g , and so o n . M y statements are all
further whatness statements about any one o f those, say, the
species d o g , just as the previous one was a whatness statement
about F i d o . T h a t i n i t i a l whatness statement d i d n o t define F i d o .
I t p r o v i d e d a p r o p e r t y o f F i d o that was essential to i t , a r t i c u -
lated w h a t k i n d o f t h i n g i t is, classified i t b y w a y o f a funda-
mental b i o l o g i c a l ensemble i t is a member of, and so o n . N o n e
o f this had the f o r m o f an i d e n t i t y o r a r e d u c t i o n o r replacement
o f p o o r F i d o b y a logical construction o u t o f a bundle o f p r o p -
erties or o u t o f a genus and differentia or the l i k e . I first said
w h a t k i n d o f t h i n g F i d o is, then w h a t k i n d o f t h i n g the species
d o g is. T h a t is a l l .
I n answering the question " W h a t am I ? " Descartes does n o t
need t o carry o n his shoulders the excesses o f medieval defini-
i20 W H A T A M I?
tionalism. Perhaps there is something i n t r i n s i c a l l y elusive about

the f o r m o f a statement such as " W h a t F i d o is is a d o g " and
" W h a t d o g is is a m a m m a l i a n species." B u t w e cannot give u p
i n advance o n such statements just because medieval defini-
tionalists mistreated t h e m .
So far this discussion has concerned Descartes's worries about
the f o r m o f the whatness statement. M y second question concerns
the t r u t h o f the matter, at least f r o m Descartes's standpoint. Parts
o f his w o r k — s u r e l y his w o r k o n language as distinguishing us
f r o m animals i n a most fundamental way—suggest strongly that
any classification o f humans as a k i n d o f animal is just w r o n g .
I f so, the problem w i t h the p r i m a l answer is that i t leads to a
statement—Man is rational animal—that is simply false.
Suppose this is Descartes's last w o r d o n the matter. T h e n ,
all that follows is that w h a t i t is t o be a m a n — w h i c h he is and
I am—is n o t essentially t o be a k i n d o f animal. B u t w h a t I am
and he is is still g i v e n b y the answer: "a m a n . " A n d since the
answer " T o be a m a n is to be a rational a n i m a l " has n o w been
disqualified as false, Descartes is free t o pursue his o w n answer
about w h a t i t is to be a m a n . W h a t e v e r this subsequent answer
is, the answer t o his i n i t i a l q u e s t i o n — " W h a t am I"?—is still "a
man."
So far w e assume a reading o f Descartes that denies that
w h a t i t is to be a m a n is t o be a certain k i n d o f animal. N o w ,
for myself, I do n o t believe that i n order to make sense o f
Descartes's answering w i t h "a m a n " the " W h a t am I ? " q u e r y ,
w e need to take a stand o n whether he u l t i m a t e l y saw m a n k i n d
as an animal k i n d . A l l w e need to make sense o f is his p r i m e
a s s e r t i o n — " W h a t I am is a m a n " — a n d t h e n o f related asser-
tions such as " W h a t F i d o is is a d o g " ; " W h a t the L o i r e is is a
r i v e r " ; " w h a t H o n is is (a piece of) w a x . "
I n addressing the t h i r d question, h o w t o understand such
whatness assertions, w e need yet again t o separate matters o f
logical f o r m and matters o f t r u t h . T h e former seem t o me t o
pose problems at a m u c h m o r e sophisticated level. L e t me give
THE REAL M A N 121
a sense o f this i n order t o eliminate m u c h o f the sophistication

and focus o n the t r u t h o f the matter.
First, f o r m : Consider o u r k i n d - s p e c i f y i n g assertions—"Des-
cartes is a m a n , " " F i d o is a d o g , " and " T h e L o i r e is a r i v e r . "
T h e proper name refers here t o a certain i n d i v i d u a l , n o t a p r o p -
erty o r a concept o r a predicate. B u t w h a t are w e t o say o f the
indefinite n o u n phrase that f o l l o w s "is"? Are we attributing a
p r o p e r t y t o the i n d i v i d u a l named b y the proper name, as i f w e
were u s i n g an adjective, as i n "Descartes is wise"? O r m i g h t
w e still be r e f e r r i n g , even o n the r i g h t - h a n d side o f the " i s , "
this t i m e t o a different object, a k i n d — f o r example, m a n k i n d —
and saying overall that Descartes is a member o f this kind?
Such a referent for the c o m m o n n o u n w o u l d be a bona fide
historical object, w h e t h e r i t is a chemical substance ( l i k e w a x ) ,
a geological k i n d ( l i k e a r i v e r ) , o r a b i o l o g i c a l species (the w a y
I take m a n k i n d and d o g k i n d t o be; " b i o l o g i c a l " is n o t t o say
that m a n k i n d is just another animal k i n d ) . Such historical o b -
jects are part o f the cosmological order o f things: T h e y come
i n t o b e i n g at a certain p o i n t i n the h i s t o r y o f the cosmos; for
example, none o f the preceding k i n d s existed three minutes
after the B i g B a n g , a l t h o u g h , say, the k i n d h y d r o g e n d i d . Such
cosmological k i n d s go o n existing i n t i m e and changing t h r o u g h
h i s t o r y , l o s i n g and g a i n i n g members and b e c o m i n g widespread
and endangered. E v e n t u a l l y , t h e y — t h e k i n d s — m i g h t go o u t o f
existence i f the cosmos were t o so evolve t o exclude the exis-
tence o f members o f the k i n d . I n this sense, m a n k i n d is no
eternal platonic b e i n g b u t one that came t o be ( n o t , e.g., 5000
years ago b u t a r o u n d 3 m i l l i o n years ago), a k i n d that c u r r e n t l y
exists b u t is subject to the danger o f g o i n g o u t o f existence.
U n d e r this understanding, i n saying o f F i d o (Descartes) that
w h a t i t is is a d o g ( m a n ) , w e are r e l a t i n g t w o historical beings.
W e are saying w h a t the specific i n d i v i d u a l ( F i d o ) is b y assign-
i n g i t t o certain natural ensemble ( k i n d o f t h i n g s ) , itself existing
i n h i s t o r y and fundamental t o the description o f cosmological
evolution.
i22 W H A T A M I?
H e r e , t h e n , are t w o simple theories o f the f o r m o f w h a t is

said i n whatness-specifying assertions such as " T h e L o i r e is a
r i v e r , " "Descartes is a m a n , " and " F i d o is a d o g . " O n e views
the f o r m as predicative—that is, the assignment o f a p r o p e r t y ,
a k i n d p r o p e r t y , to a real subject, something the m o d e r n l o g i -
cian w o u l d code b y the predicative f o r m " F a . " T h e other sees
the f o r m as a relation between t w o objects, the i n d i v i d u a l sub-
ject b e i n g said t o be a member o f the k i n d referred t o b y the
c o m m o n n o u n ( w h a t the logician w o u l d code b y a f o r m analo-
gous to the set theoretic relation " I t e m ais a member o f the
set S"). B o t h theories are interesting, and the question o f w h i c h
one is r i g h t is none t o o easy t o answer.
T h e preceding discussion concerns sophisticated questions
about the f o r m o f the whatness specifying assertion "Descartes
is a m a n . " These form-related questions are i n t r i g u i n g and
w o r t h p u r s u i n g i n the r i g h t context. B u t m y p o i n t here is that
w e need n o t await the v e r d i c t o f these refined discussions o f
logical f o r m to be i n a p o s i t i o n t o make sense o f Descartes's
p r i m a l answer t o his p r i m a l question. W h a t e v e r t h e o r y o f f o r m
is u l t i m a t e l y correct, w e k n o w this m u c h : Just as the assertion
that w h a t F i d o is is a d o g , the assertion that w h a t I am is a
m a n makes sense and needs no philosophical analysis t o be seen
to make sense. M o r e o v e r , just as the assertion that w h a t F i d o
is is a d o g , the claim that w h a t I am is a m a n has the undeniable
r i n g o f t r u t h . So does Descartes's line to Elizabeth: " E v e r y o n e
feels that he is a single person w i t h b o t h b o d y and t h o u g h t so
related b y nature that the t h o u g h t can m o v e the b o d y and feel
the things w h i c h happen t o i t . " O n e finds w i t h Descartes, as
w i t h the m i n d - b o d y p r o b l e m i n general, that i l l u m i n a t i o n m a y
lie n o t so m u c h i n the m u r k i e s t depths o f scholastic analysis
b u t i n t a k i n g notice o f l i m p i d shallows.
8. In a rather different, technical context, I discussed in detail the form

of such statements as "Descartes is a man" and the role of the indefinite
noun phrase that follows the "is." See "The Subject-Verb-Object Class,"
PhilosophicalPerspectives 12 I , I I (1998).
THE REAL M A N 123
I t w o u l d be e n l i g h t e n i n g t o have at some p o i n t a general

t h e o r y o f the logical f o r m o f such w h a t - I - a m assertions. I t
w o u l d surely help t o have a general metaphysical t h e o r y o f the
whatness o f k i n d s — w h a t are the k i n d s dog, wax, river, o r man}
I n the latter case, i t is v e r y l i k e l y that mere D N A specification
o r the t r a c i n g o f a historical connection t o some u r - m e n i n
E t h i o p i a w o u l d n o t be a f u l l y satisfying answer, at least n o t for
someone w i t h a Cartesian bent. W e are l i k e l y t o call o n the
observation that each m a n has a m i n d and b o d y ; that they are
interdependent (perhaps i n the w a y discussed above); that they
are essentially and d i s t i n c t i v e l y a h u m a n ( m a n k i n d ) m i n d and
b o d y ; that there is n o understanding o f b e i n g o f that k i n d —
b e i n g a m a n — w i t h o u t a better understanding o f h o w each such
b e i n g depends o n the existence o f his m i n d and b o d y .
A l l o f these metaphysical extras w o u l d help. B u t whatever
appeal such theoretical analyses m i g h t have and whatever c o n -
fidence w e m a y come t o have i n t h e m , they are n o t l i k e l y t o
exceed the appeal and confidence w e find i n Descartes's p r i m a l
answer t o the p r i m a l question.
Index
a priori knowledge, conceivabil- atoms, 88

ity and, 19, 21, 48 attributes, i n separatist dualism,
abstraction 62—63, 67—68
in real distinction, 101
of whatness, 30—31, 33 being, o f substance, 69
abstractionist metaphysics bodies
of human mind, 91—93 composition of, 81, 81n.9
lifetime theories versus, 84—86, Descartes's cuts for, 75—76,
96-97 76n.7, 81
of numerical sameness, 82—83, endurance in time of, 74—75,
86 77, 87
of objects, 84, 86, 88-89 of a given kind, 76—79
allusion, i n mentality, 54 human (see human body)
analog, i n pain distinction, 45 body-in-general, separatist con-
angelic mind ception of, 74—75, 80
in integrative dualism, 79 body-swapping, 78n.8, 96n.15,
in separatist dualism, 62—63 103. See also mind-swap-
stability of, 94—95 ping
animal, rational, man as, 117—123 brain, as mind model, 91, 95n.14
arm-body connection, 67nn.2,3, brain state, o f pain, mental state
73, 90, 95n.14, 115 versus, 44—48
Arnauld brain transplants, 112, 113n.5
on conceivability, 16, 18, 20—
2
з> 42-43 C fibers. See firing o f C fibers
can, i n real distinction, 4, 6, 8,
on possibility, 15n.5, 16—20,
12
2
3 can exist without the other, in
on real distinction, 12n.3, 13n.4 real distinction, 4, 6
on whatness, 23—42
125
i26 INDEX
Cartesian concepts, 8, 59, 114 completeness, notion of, 31-33,

of subjects i n time, 81—98 66
categorical conception conarion hypothesis, o f mind-
of real subjects, 70 body connection, 91
of substance, 67-68, 77, conceivability
100 Arnauld's notion of, 16, 18,
categorical real distinction, 70 20-23, 42-43
change. See also transformation autonomy of, 42-43
human mind endurance of, constraining of, 25—26, 29
90-93, 97 de re notions of, 93n.12
robust whatness and, 87—88, Descartes's notion of, 16, 30—
90, 93, 93n.12 31, 42-44
coevolution, ofhumans, 85, 121 determinism of, 20
cognition. See thinking of idea, 22—23
coherent separation, o f possibil- illusions of, 105, 115
ity, 13, 13n.4, 15 imaginability versus, 52n.15,
coherent story 53-54, 56, 58, 114
in real distinction, 8—10, 12— integrative dualism and, 103—
13, 12n.3 105, 107—108
in whatness quest, 8—10, 38— of life, 115
39, 98n.16 logical, 43-48
coherent story projection, 12 of mathematics, 14, 17, 19, 23
combinatorialist view, of possibil- metaphysical, 10-11, 42—43,
ity, 14 107
complete idea of mind migration, 103
framework for understanding, possibility as preceding, 107
31-33 primacy of, 20—23, 30
in whatness, 22-23, 29-33 a priori knowledge i n , 19, 21,
complete subjecthood 48
denial of, 99—100 projection method for, 10—13,
existence apart and, 99—103 11n.2, 13n.4, 18
in integrative dualism, 59, 61, in real distinction, 5—6, 10—11,
66, 70 59, 101, 103-104
complete thing of reality, 21—23
existence-free terms for, 67— seeming versus real, 19, 62,
69, 69n.4 66n.1
notion of, 31—33, 66 conceivability illusions
in separatist dualism, 66—68, disembodiment as, 23n.7,
67n.2 103—107, 109
INDEX 127
extendedness i n , 106 de re argument

man-mind swaps as, 103—104, for cognition, 56, 56n.18
107—109 for mind-body connection,
conceivability intuitions, 57, 44-47
104—106, I I I premise of, 48—49, 53
evidence i n , 106—107, 1 1 1
definition, i n primal answer,
whatness facts and, 104—107 118—120
conceivability-reflecting-reality dependence
projection principle, existential, 102—103
21—22 in real distinction, 52n.15, 56
conceivable separability, 59, 101, derepresentation, i n de dicto
103—104 premise, 49
conceivably-exists-without-the derivatives, i n integrative dual-
pertinent-physicalium, 43 ism, IIO—III
conceivably so and so, 46—47 Descartes, Rene

conceptual fix on conceivability, 16, 30—31,
application of, 25, 41 42-44
constraints i n , 25—26, 29 cuts o f generic bodies, 75—76,
epistemic fact of, 27—28 76n.7, 81
metaphysical fact of, 27—29 on imagination, 56
theoretical basis of, 23n.7, 24 letter to Mesland, 71—72,
consistency 72n.5, 76, 7 9 ~ > 95
8l
in man-mind swaps, 107—108 meditations o f (see specific

possibility versus, 37 meditation)
consistent-with-what-it-is-to-not- on pain and brain states, 44,
be-extended, 38 47
constraints, i n conceptual fix the- on possibility, 14—16, 15n.5,
ory, 25—26, 29 2
3
corpse, as thing, 78
on primal question, 115—123
creation point, o f subject, 32—33,
on whatness, 23—25, 28—29,
33n.12
38-42
desires, i n real distinction, 81n.9
data articulation, o f projection
determinism, o f conceivability,
mold, I I
20
de dicto argument
disembodiment
epistemological gains of, 4 8 ¬ in integrative dualism, 60,
49> 5 2
103—107, 109
for mind-body connection, in real distinction, 19, 23n.7,
45-48
2
5
128 INDEX
distinction epistemic conceivability

of individuals, 5 metaphysical versus, 42—43
in real distinction, 12—13, of pain, 43, 48-49, 5 2
12n.3, 100 preservation of, 39

distinctness o f discernibles transparency of, 18—21, 43,
applications of, 5-8, 13n.4 48-49
Cartesian doubt of, 8 epistemic possibility, i n real dis-
divine mind tinction, 4, 8
in separatist dualism, 62—63 epistemic whatness
stability of, 94 evaluation of, 38—41
dual key project, objectives of, notion of, 24—27
xvii essence
dualism abstractionist theory of, 83
forms of, xviii of bodies
integrative (see integrative du- generic, 4, 75
alism) given king, 77
metaphysical, 62, 65, 74, 77, generic, 62-64, 71, 75,
78n.8 109—112
numerical sameness i n , 70, 73 metaphysics of, 170n.2
separatist (see separatist du- of mind, 80, 90—91, 97
alism) of real pain, 52, 54
robustness of, 87—88, 90, 93,
endurance 93n.12
of mind-body connection, 63— of separability, 62-64, 75
66, 71-81, 85-93 of whatness, 5, 52, 54, 64—65,
of piece-of-wax-as-a model, 104
85-89 essentialism, 38
of subjects i n time, 65—66, Euclidean axioms, 37
85-91 evaluation worlds, in de dicto
endurance i n time premise, 49
abstractionist theory of, 83 evidence
of bodies in conceivability intuitions,
generic, 74~75> 7 8
106—107, I I I
given kind, 77 of specific human mind, 112—
of minds 113, 113n.5
generic, 79-81, 97 exclusion
human, 71—72, 79 necessity versus, 33—34
of objects, 74-76, 86—87 in whatness, 8—10, 30—33,
of subjects, 65-66, 85-91 39
INDEX 129
existence f mind, separatist conception of,

of bodies 80-81
generic, 75 fictional characters, creation of,
given kind, 77—79 33, 33n.12
of mind, 80, 97 finite sets, i n mathematics, 35—
necessary connections and, 36, 36n.13, 119
52n.15 firing of C fibers ( F C F )
in real distinction, 4, 25, 39, mental state versus brain state
60, 99—103, 106 of, 44-48
existence apart, complete subject- pain versus, 44, 52, 52n.15
hood and, 99—103 formal distinction, i n complete
existence-free numerical reading, subjecthood, 100—101
of distinct entities, 102
existence i n time. See endurance generic bodies, Descartes's cuts
in time of, 75-76, 76n.7, 81
existential conception, o f sub- generic essence theory
stance, 67 of integrative dualism, 71,
existential dependence, 102—103 74—76, 109—112
existential independence, 68, 70, of separatist dualism, 62—64,
75, 79, 100 75
existential real distinction, 5—8, geometry
70, 101, 115 disanalogy w i t h mind-body
existential sameness, numerical, connection, 23—24, 26
102 in real distinction, 14, 17, 19
existential separability as whatness example, 27—28,
of mind and body, 61—62, 66, 34, 37
81n.9, 115 God
o f real distinction, 5—8, 70,101 justice of, 100—102
experiences mercy of, 100—102
as phenomenon, 56—58
predication of, 55 histories
extended thing, 38, 88—89, 106 of different minds, 63, 65
extendedness of subjects, 81, 83
in conceivability illusions, 106 human being
of human body, 62, 65, 82 modes of, 60—61, 116
of whatness, 9, 15, 17, 38, 89 as particular, 86, 89, 109,
extension to mind, Kripke's no- 109n.3, 113
tion of, 54—55 primal answer of, 116—123
external, as what it seems, 52, 92 primal question of, 115—116
i o
3 INDEX
human being (continued) prime lifetime fact of, 93—94

primality of, xviii, 60—61, 66, as specific, 112—115, n3n.5
71, 76-80, 96-97, I I I Hume, David, on personal iden-
prove too much/prove too lit- tity, 112—114
tle dilemma of, 59—61,
115-123 idea
as rational animal, 117—123 complete (see complete idea)
human body conceivability of, 22—23
connection to mind (see mind- of substance, 3
body connection) identity
endurance of, 63—66, 74—77, criteria for, 45
85, 87-88, 90-91 in pain distinction, 45
existential separation of, 5, personal, 64, 66n.1, 95,
61—62, 66, 81n.9 96n.15, 108, 109n.3
extendedness of, 62, 65, 82 in primal answer, 118—119
integrative dualism of, 72—79 in real distinction, 12, 12n.3
whatness of, 77—78, 78n.8 specific man-mind swaps and,
human involvement, freedom 112—114, n3n.5
from, 71 illusions of conceivability, 105, 115
human mind imaginability. See also perceptual
abstractionist metaphysics of, states
91-93 conceivability versus, 52n.15,
connection to body (see mind- 53-54, 56, 58, 114
body connection) Kripke's notion of, 52n.15,
endurance of, 63—66, 71—72, 53-54, 56
79-81, 85, 90-93, 97 mental versus sensational,
existential separation of, 5, 55-58
61—62, 66, 81n.9 predication i n , 53—54
generic essence of, 62—64, 7 , 1
seeming versus real, 49—51,
75, 109-112 52n.15, 53, 58
integrative dualism of, 72—74, imagination, i n de dicto premise,
79-81 49
intellectual qualities of, 38, immortality, o f soul, 98n.16
81—82, 81n.9, 91 incomplete idea, notion of, 32
mental categories of, 56—58 independence
momentary, 92 existential, 68, 70, 75, 79, 100
nature of, 81—82, 81n.9, 91, 96 of mind and body, 62—63, 66,
one-one principle of, 94—96, 96n.15
95n.14 symmetrical, 62-63
INDEX 131
individuals, transformation of, intellectualism

89 of human mind, 38, 81—82,
infinite sets, i n mathematics, 35¬ 81n.9, 91
36, 36n.13 perceptual states and, 56
information transfer, 91, 112 intentional action states, as phe-
inseparability, modal, 59—61 nomenon, 56—58
integrative dualism, 71—81 interdependence, o f subjects, 79
ofbodies, 72—74 internal, as what it feels like, 92
generic, 74—76 intuitions
given kind, 76—79 about identity o f thoughts, 57,
complete subjecthood i n , 59, 104
61, 66, 70 conceivability, 57, 104—106,
denial of, 99—100 III
conceivability and, 103—105, items. See object(s); subject(s)

107—108
derivatives i n , n o — i n justice
disembodiment illusion i n , 60, apart from mercy, 102
103—107, 109 o f God, 100—101
generic, 71, 74—76, 109—112
of human mind, 72-74, 79— kind-blind attributes, i n separat-
81 ist dualism, 62—63
Meditation I I on, 71, 74, 76, kind-specifying assertions, i n p r i -
79-81 mal answer, 120—123
Mesland letter on, 71—72, knowledge, prior. See a priori
72n.5, 76, 7 9 ~ > 95
8l
knowledge
of mind-body connection, Kripke, Saul
xvii—xviii, 39, 72—81 on conceivability, 42—44
of-ness principle i n , 111—112 morals of, 53—54
of primality o f man, 111, on pain phenomenon, 44, 52,
115—116 52n.15, 54-55, 55n.16,
principles of, 78n.8, 96—98 V
separatist dualism versus, 61, on real possibility-imaginabil-
71, 96n.15 ity, 49-51
set theory of, 110-111
symmetrical, 71—74, 80, language theory, i n pain distinc-
94n.13, 97 tion, 45—46
tight, 109—112 life, coevolution of, 85, 121
intellectual abstraction, o f what- lifetime metaphysics, o f enduring
ness, 30—31, 33 objects, 84—86
i2
3 INDEX
lifetime principles matter, particles of, 81—82,

abstractionists versus, 83, 82n.9, 88
96-97 Meditation I I
of subjects in time, 65—66, on integrative dualism, 71, 74,
93-98 76, 79-81
limits, i n set theory, 35 on primal question, 116-117
living, coevolution of, 85, 121 on real distinction, 8—10, 28—
Locke, John 29, 38
on personal identity, 112—114 on separatist dualism, 62—63,
on whatness, 9 65
logic, modal. See modal logic on subjects i n time, 82—83,
logical conceivability, metaphysi- 90
cal versus epistemologic Meditation I I I , on real distinc-
argument for, 43—48 tion, 22, 28
logical possibility Meditation V I , on real distinc-
in real distinction, 4, 11—12, tion, 4, 28, 90—91
12n.3 memory implants, 112, 113n.5
in whatness evaluation, 38— mental categories, o f human
39, 3 I 2
mind, 56—58
logical potential, 84 mental items
conceivings of, 43
magnetic resonance imaging imaginability of, 54—55
( M R I ) , o f pain distinc- mental state
tion, 44-47 of pain
man. See human being brain state versus, 44—48
man-mind-body trio, 116 extension to, 54—55, 55n.16
man-mind swaps. See body-swap- predication and, 54—55, 55n,16
ping; mind-swapping mercy
man relocation, integration w i t h , apart from justice, 102
110 of God, 100—101
materiality, o f whatness, 9 Mesland letter, on integrative du-
mathematics alism, 71-72, 72n.5, 76,
conceivability of, 14, 17, 19, 79-81, 95
2
3 metaphysical conceivability
in dualism theories, 70, 73 epistemology versus, 42—43
set theory of, 34—36, 36n.13, in real distinction, 10—11, 107
119 metaphysical dualism
whatness of, 27—28, 34—35, integrative, 74, 77, 78n.8
118—119 separatist, 62, 65
INDEX 133
metaphysical possibility as dilemma, xvii—xviii,

Kripke's notion of, 49—51 xviiin.i
in real distinction, 4, 15n.5 endurance of, 63—66, 71-77,
of whatness, 24—26 85—88, 90—91
metaphysical potential, 84 geometric disanalogy of, 23¬
metaphysical whatness 24, 26
evaluation of, 38—39, 41 integrative, 79—81
factual basis of, 27—29 necessity of, 39
of kinds, 121, 123 pineal location of, 91
possibility of, 24—26 possibility of, 19—20
metaphysics primal answer to, 116—123
abstraction (see abstractionist mind-body independence, 62—63,
metaphysics) 66, 96n.15
of essence, 170n.2 mind-in-general, 80—81
lifetime, o f enduring objects, mind migration, conceivability
84-86 of, 103, 104n.1, 108
o f necessity, 170n.2 mind replacement
of subjects i n time, 83—84, conceivability of, 103—104,
86-87 104n.1
subspecies-aeternitate, 83 integration w i t h , 110
mind mind-swapping
of angels, 62-63, 79, 94-95 as conceivability illusions,
composition of, 81, 81n.9 103—104, 107—109
essence of, 80, 90—91, 97 conceivability of, 107—109,
generic idea of, 80—81, 85 112
of a god, 62—63, 94 integration of, 110
of humans (see human possibility of, 103—104,
mind) 107—109
lifetime principles of, 93—98 in real distinction, 103—104,
possible histories of, 63, 65 107—109, 112, n3n.5
separatist conception of, specific identity and, 112—115,
80-81 n3n.5
mind-body connection. See also types of, 103—104, 104n.1
human body; human whatness-consistency i n ,
mind 107—108
Cartesian concept of, 8, 59 minds, endurance i n time of
conceivability of, 21 generic, 7 9 ~ > 97
8l
de re versus de dicto argu- human, 71—72, 79

ments for, 43—48 modal inseparability, 59—61
i34 INDEX
modal logic object(s)

of complete subjecthood, conceptual fix theory of,
IOO—IOI 23n.7, 24, 26
failure of, 24-25, 29 endurance i n time, 74—76,
in pain phenomenon, 44, 46, 86-87
49 metaphysics of, 84—86, 88—89
of possibility, 4, 12n.3, 15, 25 seeming versus real
of primality o f human being, conceivability of, 19, 62,
60—61 66n.1
modes imaginability of, 49—53,
o f h u m a n being, 60—61, 116 52n.15, 58
in real distinction, 101 transformation of, 88—89
of substances, 67—68 objectal relation
momentary mind, 92 in real distinction, 12, 12n.3
in reality reflection, 22—23
nature, o f human mind, 81-82, in whatness, 24, 33n.12
81n.9, 91, 96 of-ness principle
nature separability, 101 in integrative dualism,
nature thinking, 38 111—112
necessary union hypothesis, as lifetime, 94—95, 94n.13
37 one-one principle
necessity failure of, 103—104
of connections between dis- as lifetime, 94—96, 95n.14
tinct existence, 52n.15
exclusion versus, 33—34 pain
metaphysics of, 170n.2 essence o f real, 52, 54
in whatness, 28—29, 33-38, firing of C fibers versus, 44,
40, 107n.2 52, 52n.15
no endurance theory imagined, 52
of separatist dualism, 63—66 Kripke's notion of, 44, 52,
of subject, 65-66 52n.15, 54-55, 55n.16,
nothing is quite what it seems, 57
51-52 mental state versus brain state
numerical sameness of, 44-48
abstractionist theory of, 82— transparent epistemology argu-
83, 86 ment for, 43, 48—49, 52
in dualism theories, 70, 73 particles, o f matter, 81—82,
existential, 102 82n.9, 88
INDEX 135
particularism, o f human being, metaphysical, 4, 15n.5, 24—26,

86, 89, 109, 109n.3, 113 49-51
perceptual states. See also imagin- as preceding conceivability,
ability 107
intellectual states and, 56 primacy of, 17—20
personal identity projection method for, 10—13,
man-mind swaps and, 112— 12n.3, 13n.4, 17-18, 39
115, 113n.5 in real distinction, 4—7, 10
theory of, 64, 66n.1, 95, possibility tout court, 15n.5
96n.15, 108, 109n.3 potential, 4
phenomenon of subjects i n time, 84
experiences as, 56—58 predicates and predication
mental states as, 56—58 experiential, 55
necessity i n whatness as, 34 in imaginability, 53—54
pain distinction as, 44—45, 49, mental, 54—55, 55n.16
52, 52n.15, 54-55, 55n.16, of substance, 67—68, 79, 98
57 primal answer
physical qualities definition i n , 118—120
ofbodies, 81—82, 82n.9 to primal question, 116—123
transformation of, 88—89 primal question, 115—116
piece-of-wax-as-a model primal answer to, 116—123
endurance of, 85—89 primality
principles of, 85n.10, 90—91 ofbodies, 74—78
as subject in time, 82—85, integrative interpretation of,
114-115 i n , 115—123
pineal gland, 91 of real man, x v i i i , 60—61, 66,
plurality, i n set theory, 35 71, 76-80, 96-97, I I I
possibility prime lifetime fact, of mind,
Arnauld's notion of, 15n.5, 93-94
16—20, 23 projection mold, two-step, 11
combinatorialist view of, 14 projection principles
consistency versus, 37 conceivability-reflecting-
Descartes's notion of, 14—16, reality, 21—22
15n.5, 23 in man-mind swaps, 107—108
logical, 4, 11—12, 12n.3, 38— in real distinction, 10—13,
39> I 2
3 11n.2, 13n.4, 18, 39
o f man-mind swaps, 103—104, projection step, of projection
107—109 mold, I I
136 INDEX
prove too much/prove too little existence i n , 4, 25, 39, 60, 106
dilemma as apart, 99—103
integrative interpretation of, existential separability of, 5-8,
115-123 70, 101, 115
in separatist dualism, 59—61, generic versus tight integration
66 of, 109—112
Pythagorean theorem, 14, 17, 19 man-mind swaps i n , 103—104,
whatness and, 28, 34 107-109, 112, n3n.5
of mind, 80, 98
quasi-demonstrative representa- necessary connections and,
tions, i n de dicto prem- 52n.15 *"
ise, 56n.18 observations of, 52n.15,56, 101
projection method for, 10—13,
rational animal, man as, 117—123 11n.2, 13n.4, 18, 39
real conceivability, seeming con- simpliciter, 70
ceivability versus, 19, 62, of specific human mind, 112—
66n.1 115, 113n.5
real conceivability projection, real imaginability, seeming imagi-
10—13, 11n.2, 13n.4, 18 nability versus, 49—51,
real distinction 52n.15, 53, 58, 114
Arnauld's notion of, 12n.3, real man
13n.4 primal answer of, 116—123
basis for, 3—7, 70 primal question of, 115—116
of bodies primality of, xviii, 60—61, 66,
generic, 75 71, 76-80, 96-97, I I I
given kind, 77 real possibility-imaginability,
categorical, 70 49-51
coherent stories i n , 8—10, 12¬ real possibility projection, 10—13,
13, 12n.3, 98n.16 12n.3, 13n.4, 39
complete subjecthood i n , reality
99-103 conceivability of, 21—23
conceivable separability i n , necessary connections and,
5-6, 10-11, 59, 101, 52n.15
103-104, 107 in real distinction, 12—13,
Descartes's meditations on, 4, 12n.3
8-10, 22, 28-29, 38, reality-bound representations, o f
90-91 pain, 47-48
disembodiment illusion i n , reality-projection principle,
23n.7, 103—107, 109 21—22
INDEX 137
reflection principle, of reality, integrative dualism versus, 61,

22—24 71, 96n.15
representations, i n de dicto prem- Meditation I I on, 62—63, 65
ise, 49, 56n.18 of mind, 79—81
rigid designators, i n pain distinc- divine, 62—63
tion, 45, 49 no endurance theory of,
63-66
scientific discoveries, whatness prove too much/prove too lit-
of, 39-40 tle dilemma of, 59—61, 66
seeming conceivability, real con- subject(s) i n , 61, 65—66
ceivability versus, 19, 62, substance notion of, 59, 61,
66n.1 66—69, 81n.9
seeming imaginability, real imagi- set theory
nability versus, 49—51, of integrative dualism,
52n.15, 53, 58, 114 IIO—III
self-sufficiency, o f human being, of mathematics, 34—36,

60 36n.13, 119
semantic theory, i n real distinc- simpliciter, i n real distinction,
tion, 4 70
sensations, i n real distinction, 56, soul, immortality of, 98n.16
81n.9 species assertions, i n primal an-
separability swer, 120—123
conceivable, 5, 59, 101, stability principle
103—104 failure of, 103
existential (see existential sepa- as lifetime, 94—95
rability) stages, i n set theory, 35
of mind and body, xvii—xviii, subject(s)
5-8 creation point of, 32—33,
modal, 59—60, 107 33n.12
nature, 101 extension to mind, 54—55
in whatness, 5, 7, 59-60, 99 imaginability of, 53—55
separatist dualism lifetime description of, 65—66,
attributes i n , 62—63, 67—68 93-98
ofbody-in-general, 74-75, 80 real, complete, 59, 61, 66, 70,
complete thing i n , 66—68, 99
67n.2
in separatist dualism, 61, 65
existential, 61—62
whatness of, 28—29
generic, 62—64, 75
subject-predicate structure, i n i m -
generic essences of, 62—64, 75
aginability, 53-54
i 8
3 INDEX
subjecthood, complete, 59, 61, symmetry

66, 70 in integrative dualism, 71—74,
existence apart and, 99—103 80, 94n.13, 97
subjects in time, 81-98 of mind and body indepen-
abstractionists challenge of, dence, 62—63
91-93
endurance of, 65—66, 85—91 teletransporting devices, 112,
human mind as, 90—91 n3n.5
lifetime principles of, 93—98 thinking
Meditation I I on, 82—83, 9 ° de re argument for, 56, 56n.18
metaphysics of, 83—84, 86-87 in whatness, 29, 31, 38—39,
piece-of-wax-as-a model of, 58
82-89, 114—115 thinking being, 106
qualities of, 81—85, 82n.9 thinking state, as phenomenon,
separatist dualism of, 65—66 56-58
subspecies-aeternitate metaphys- thinking thing, 38, 62
ics, 83 stability of, 93, 95—96
substance thoughts, intuitions about iden-
bodies as tity of, 57
generic, 75 tight integrative dualism, generic
given kind, 77 versus, 109—112
categorical conception of, 67— transformation. See also change
68, 77, 98, 100 physical qualities of, 88—89
existence-free terms for, 67— triangle. See geometry
69, 69n.4
existential conception of, 67 universe o f finite sets, 35—36
extension to mind, 54—55
idea of, 3, 30—31 water
of mind, 80 conceivability of, 19—21
modes of, 67—68 imagining of, 50-51, 57
predication of, 67—68, 79, 98 as subject i n time, 74, 83
real, complete, 59, 61, 66, thinking of, 57-58
67n.2, 70, 99 whatness of, 27
separatist ideas of, 59, 61, 66— wax-as-a model. See piece-of-
69, 81n.9 wax-as-a model
subtraction principles wax passage, 85n.10
necessity versus, 33—34 What am I?
of whatness, 8—10, 32—33 primal answer to, 116—123
successor, i n set theory, 35 as primal question, 115—116
INDEX 139
whatness primacy of, 23—42

Arnauld's notion of, 23—42 in primal answer, 118—123
ofbodies, 77—78, 78n.8, 87 priority over imagination, 54
coherent stories i n quest for, in real distinction, 5, 7—8
8-10, 38 robust, change and, 87—88,
complete ideas i n , 22—23, 90, 93, 93n.12
9-33
2
subtraction principles of, 8—
Descartes's notion of, 23—25, 10, 32-33
28—29, 3 8 -
4 2
thinking states for, 29, 31,
epistemic, 24—27, 38—41 38-39, 58
essence of, 5, 52, 54, 64—65, too much versus too little i n ,
104 23-26, 33
exclusion i n , 8—10, 30—33, 39 whatness closure, 15
extendedness of, 9, 15, 17, 38, whatness-consistency, i n man-
89 mind swaps, 107—108
ground facts of, 9, 26—29 whatness facts, conceivability i n -
of mathematics, 27—28, 34—35, tuitions and, 104—107
37, 118-119 whatness separability, 59—60,
metaphysical (see metaphysical 99
whatness) whatness-specifying features,
necessity i n , 28—29, 33~~3 ? 8
28—29
40, 107n.2 wine, as item, 40, 43

Almog, J. - What Am I

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

Almog, J. - What Am I

Caricato da

Copyright:

Formati disponibili

W H A T A M I?

Descartes and the Mind-Body Problem

Copyright © 20o2 by Joseph Almog

First published in 2002 by Oxford University Press, Inc.

First issued as an Oxford University Press paperback, 2005

Oxford is a registered trademark of Oxford University Press

A l l rights reserved. N 0 part of this publication may be reproduced,

Library of Congress Cataloging-in-Publication Data

Printed in the United States of America

The title question is vintage Descartes. I t makes its debut in

This book has been reconstructed, with minor alterations

ι. In what follows, I have relied on The Philosophical Writings of Des-

and decided to imitate him, which was not very sage on my

But what is a man? Shall I say 'rational animal'? No; for

He soon substitutes the primal question by surrogates—What

night, bought cadavers in the "street of veal" so that their dis-

I supposed that your Highness still had in mind the argu-

Driven by these remarks, I have redesigned the class in re-

Running this class for more than 15 years led me to accumu-

Descartes's seeming promise to leave room for the immortality

Kripke to be the leading expounder. I t is thus that I have come

Saint Jean Du Gard, France

Synopsis: The Project and the Challenge, xvii

ι. The Real Distinction, 3

3. The Real Man, 99

Descartes's dual key project targets two objectives. The first is

Let me put the dilemma in my own words. We might be so

cartes calls a mere "unity of composition/' an artificial com-

We shall follow Descartes's two-stage defusing of the prove

ι. I should say that in teaching this critical dilemma-setting paragraph, I

The Real Distinction

i . i . The Basis for the Real Distinction

I n a letter written on January 19, 1642, to Father Gibieuf, Des-

The idea of a substance with its extension and shape is a

A similar observation about the completeness o f the idea o f his

i.2. What Is i n a Real Distinction?

When I speak of "the real distinction o f mind and body/' we

1.3. Conceivability, Possibility, Whatness

Existentially Separatist Arguments

There are at least three arguments of Descartes that we shall

(2) I t is not the case that D B bears F.

Now, by the distinctness of discernibles:

It is then added that given the meaning o f the discerning feature

(4) D M can exist without D B .

It is time to verify how this basic mold shows up i n the

The Argumentfrom Conceivability

In the argument from conceivability, the discerning property F

(iC) D M conceivably exists without D B .

(2C) I t is not the case that D B conceivably exists without

This takes care o f the numerical distinction. But where are

(4C) D M can exist without D B .

The Argumentfrom Possibility

In the argument from possibility, the discerning property is

(iP) D M possibly exists without D B .

(2P) I t is not the case that D B possibly exists without D B .

It is now argued that ( i P ) directly gives us

(4P) D M can exist without D B .

The Argumentfrom Whatness

In the argument from whatness, the target property F relies on

(iW) D M is by whatness thinking.

(2W) I t is not the case that D B is by whatness thinking.

The premises of the whatness argument are much more pru­

tomed to witnessing a numerical distinction by existential sepa-

The premises of the whatness argument are much more pru

(WC) I f F pertains to what x is, then F is modally neces

Already ( W C ) excludes much of the combinatorialist cre

Equivalently, i f F does not pertain to i ' s whatness, does it fol

It is not quite clear to me where Descartes comes out in A r