Cognitive Literary Science

 

COGNITION AND POETICS

Cognition  and Poetics (CAP) fosters high quality interdisciplinary
research at the intersection of cognitive science, literature, the arts, and
linguistics. The series seeks to expand the development of theories and
methodologies that integrate research in the relevant disciplines to fur-
ther our understanding of the production and reception of the arts as
one of the most central and complex operations of the human mind. CAP
welcomes submissions of edited volumes and monographs in English
that focus on literatures and cultures from around the world.

Series Editors:
Alexander Bergs, University of Osnabrück
Margaret H. Freeman, Myrifield Institute for Cognition and the Arts
Peter Schneck, University of Osnabrück
Achim Stephan, University of Osnabrück

Advisory Board:
Mark Bruhn, Regis University Denver, CO, USA
Peer Bundgard, Aarhus University, Denmark
Michael Burke, University College Roosevelt Middelburg, The Netherlands
Wallace Chafe, University of California Santa Barbara, USA
Barbara Dancygier, University of British Columbia Vancouver, Canada
Frank Jäkel, Universität Osnabrück, Germany
Winfried Menninghaus, Freie Universität Berlin, Germany
Keith Oatley, University of Toronto, Canada
Jan Slaby, Freie Universität Berlin, Germany
Reuven Tsur, Tel Aviv University, Israel
Mark Turner, Case Western Reserve University Cleveland, OH, USA
Simone Winko, Georg-​August-​Universität Göttingen, Germany
Dahlia Zaidel, University of California Los Angles, USA

Cognitive Approaches to Early Modern Spanish Literature

Isabel Jaén and Julien Jacques Simon

Cognitive Literary Science: Dialogues between Literature and Cognition

Edited by Michael Burke and Emily T. Troscianko
 

Cognitive Literary
Science
Dialogues between Literature and Cognition

Edited by Michael Burke
and
Emily T. Troscianko

1
 

1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.

Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.

© Oxford University Press 2017

All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.

You must not circulate this work in any other form

and you must impose this same condition on any acquirer.

Library of Congress Cataloging-in-Publication Data

Names: Burke, Michael, 1964- editor. | Troscianko, Emily, editor.
Title: Cognitive literary science : dialogues between literature and
cognition / edited by Michael Burke and Emily T. Troscianko.
Description: Oxford ; New York : Oxford University Press, [2017] |
Series: Cognition and poetics | Includes bibliographical references and index.
Identifiers: LCCN 2016018910| ISBN 9780190496869 (hardcover : alk. paper) |
ISBN 9780190643072 (epub) | ISBN 9780190496883 (online)
Subjects: LCSH: Discourse analysis, Literary—Psychological aspects |
Psychology and literature. | Cognition in literature. |
Literature—Psychology—Handbooks, manuals, etc. | Psycholinguistics.
Classification: LCC P302.5 .C64 2016 | DDC 801/.92—dc23
LC record available at https://lccn.loc.gov/2016018910

9 8 7 6 5 4 3 2 1
Printed by Sheridan Books, Inc., United States of America
 

To our partners, for their patience

 
 

CONTENTS

Acknowledgements  ix
Contributors  xi
Introduction: A Window on to the Landscape of Cognitive Literary
Science  1
Emily T. Troscianko and Michael Burke

PART I: Literature through a Cognitive Lens

1. Scientific Concepts in Literary Studies: Towards Criteria for the
Meeting of Literature and Cognitive Science   17
Marcus Hartner
2. Towards a ‘Natural’ Bond of Cognitive and Affective
Narratology  35
Caroline Pirlet and Andreas Wirag
3. ‘Annihilation of Self’: The Cognitive Challenge of the Sublime   55
David S. Miall
4. The Space between Your Ears: Construal Level Theory, Cognitive
Science, and Science Fiction   73
James Carney
5. Patterns of Thought: Narrative and Verse   93
Brian Boyd

PART II: Cognition through a Literary Lens

6. Simulation and the Structure of Emotional Memory: Learning from
Arthur Miller’s After the Fall  113
Patrick Colm Hogan
7. Cognitive Science and the Double Vision of Fiction   135
Merja Polvinen
8. Fantastic Cognition   151
Karin Kukkonen
 

9. Feedback in Reading and Disordered Eating   169

Emily T. Troscianko
10. Animal Minds across Discourse Domains   195
David Herman

PART III: Literature and Cognition in Cognitive Science

11. Embodied Dynamics in Literary Experience   219
Raymond W. Gibbs Jr.
12. How Readers’ Lives Affect Narrative Experiences   239
Richard J. Gerrig and Micah L. Mumper
13. On Truth and Fiction   259
Keith Oatley
14. Under Pressure: Norms, Rules, and Coercion in Linguistic Analyses
and Literary Readings   279
Alexander Bergs
15. Affective and Aesthetic Processes in Literary
Reading: A Neurocognitive Poetics Perspective   303
Arthur M. Jacobs

Index  327

[ viii ] Contents
 

ACKNOWLEDGEMENTS

This book appears in the Oxford University Press series Cognition and
Poetics. We are especially grateful to the series editors, Alexander Bergs,
Margaret H. Freeman, Peter Schneck, and Achim Stephan for seeing poten-
tial in our project and for commissioning it. We are also indebted to the
anonymous reviewers of our book proposal, who helped focus our ideas
in the planning stages. This volume is the third artefact, as it were, of our
ongoing interest in cognitive literary science (CLSci), which also gives the
book its title. The first of the three was a symposium entitled Science and
Literary Criticism, held in April 2012 at St John’s College, Oxford; we are
extremely grateful to the St John’s College Research Centre and to Terence
Cave (via the Balzan Interdisciplinary Seminar ‘Literature as an Object of
Knowledge’) for financial, organizational, and moral support in making
that event happen. The second of our joint ventures was a special issue on
‘Explorations in Cognitive Literary Science’ published in September 2013
in the Journal of Literary Semantics. Several of the authors whose work fea-
tures in this book were involved in those earlier projects.
Editing a volume of scholarly contributions is a task that requires more
than just the editors. We are grateful to a number of anonymous review-
ers who offered insightful and constructive feedback to the authors. We
also appreciate the help of all those at Oxford University Press working
in copyediting, production, and sales to publish and promote this book.
Any errors that may remain are our responsibility. Finally, we are especially
indebted to Hallie Stebbins at Oxford University Press for all her guidance,
expertise, and kindness during the writing and editing process.
While every effort has been made to contact copyright holders, we would
be pleased to hear of any that have inadvertently been omitted.
M.B. and E.T.
Oxford, UK, and Middelburg, NL
February 2016
 
 

CONTRIBUTORS

Alexander Bergs joined the Institute for English and American Studies at
Osnabrück University in 2006, when he became Full Professor and Chair of
English Language and Linguistics. His research interests include language
variation and change, constructional approaches to language, the role
of context in language, and cognitive poetics. His works include several
authored and edited books (Social Networks and Historical Sociolinguistics
[2005], Modern Scots [2005], Constructions and Language Change [2008],
Contexts and Constructions [2009]), one textbook on Synchronic English
Linguistics (2012) and one on Historical Linguistics (co-​authored with Kate
Burridge, 2016), as well as the two-​volume Handbook of English Historical
Linguistics (edited with Laurel Brinton, 2012). He has taught at the uni-
versities of Düsseldorf, Bonn, Santiago de Compostela, Catania, and
Wisconsin-​Milwaukee, and has organized several international workshops
and conferences. Apart from several terms as Director of the Institute of
English and American Studies, as Dean of the Faculty of Linguistics and
Literatures, and as member of the University Senate, he is one of the
founding directors of the Research Cluster for Cognition and Poetics at
Osnabrück University.
Brian Boyd is University Distinguished Professor in English, Drama, and
Writing Studies at the University of Auckland. He is best known for his
work on Vladimir Nabokov—​a biography, critical books, and editions, most
recently of Letters to Véra (Penguin, 2014, and Knopf, 2015) and hundreds of
articles—​and on literature, evolution, and cognition, including On the Origin
of Stories:  Evolution, Cognition, and Fiction and Why Lyrics Last:  Evolution,
Cognition, and Shakespeare’s Sonnets (Harvard, 2009 and 2012) and the co-​
edited Evolution, Literature, and Film:  A  Reader (Columbia, 2010). He has
written on literature from Homer to the present, on comics and film, on lit-
erary theory and translation, and on art, philosophy, and science. His work
has won awards on four continents and has been published in 19 languages.
He is currently writing a biography of philosopher Karl Popper.
 

Michael Burke is Professor of Rhetoric at University College Roosevelt

(Utrecht University). He is the author of Literary Reading Cognition and
Emotion: An Exploration of the Oceanic Mind (2011). He has published numer-
ous chapters and articles on the topic of cognitive literary science. His areas
of interest also include classical rhetoric, stylistics, and pragmatics.
James Carney is Senior Research Associate in Psychology at Lancaster
University. He previously held a Marie Curie Fellowship and a Junior
Research Fellowship (with Linacre College) at the Department of
Experimental Psychology, University of Oxford. Other appointments have
included working as a lecturer in English literature at University College
Cork and the University of Limerick. His research interests centre on the
application of insights from the cognitive and experimental sciences to cul-
ture, broadly conceived. To date, this has resulted in studies of literature,
religion, mythology, popular culture, poetics, and narrative in a wide vari-
ety of scholarly journals.
Richard J. Gerrig is a professor of psychology at Stony Brook University.
Gerrig’s research focuses on cognitive psychological aspects of language
use. One line of work examines the mental processes that underlie effi-
cient communication. A  second research programme considers the cogni-
tive and emotional changes readers experience when they are transported
to narrative worlds. His book Experiencing Narrative Worlds was published
by Yale University Press in 1993. Gerrig is a Fellow of the Society for Text &
Discourse, the American Psychological Association, and the Association for
Psychological Science. He is the editor of the Journal of Memory and Language.
Raymond W. Gibbs Jr. is Distinguished Professor of Psychology at the
University of California, Santa Cruz. He is the author of Poetics of Mind:
Figurative Thought, Language and Understanding (1994), Intentions in the
Experience of Meaning (1999), Embodiment and Cognitive Science (2006),
and, with Herb Colston, Interpreting Figurative Meaning (2012). His newest
book is Metaphor Wars: Conceptual Metaphor in Human Life (2015). He is
also editor of The Cambridge Handbook of Metaphor and Thought (2008) and
the journal Metaphor and Symbol.
Marcus Hartner is lecturer in English Literature at Bielefeld University
in Germany, where he is also part of a research project on Contemporary
Fictions of Migration. His main research interests include literary theory,
cognitive approaches to narrative, and the study of both contemporary
and early modern literature and culture. In the field of cognitive narra-
tology his work has focused primarily on blending theory, the sociopsy-
chological underpinnings of character construction, and the dynamics of

[ xii ] Contributors
 

character interaction. Among his publications are Perspektivische Interaktion

im Roman:  Kognition, Rezeption, Interpretation [The Interaction of
Perspectives in the Novel:  Cognition, Reception, Interpretation] (de
Gruyter, 2012)  and a co-​ edited volume on Blending and the Study of
Narrative: Approaches and Applications (de Gruyter, 2012).
David Herman, who has taught at institutions that include North Carolina
State University, Ohio State University, and, most recently, Durham
University in the UK, is working to bring ideas from narrative stud-
ies into dialogue with scholarship on animals and human‒animal rela-
tionships. His current projects include an edited collection titled Animal
Comics: Multispecies Storyworlds in Graphic Narratives (forthcoming from
Bloomsbury in 2017) and a monograph on Narratology Beyond the Human
(in progress).
Patrick Colm Hogan is a professor in the English Department at the
University of Connecticut, where he is also on the faculty of the Program
in Cognitive Science. He is the author of nineteen books, including How
Authors’ Minds Make Stories (Cambridge University Press, 2013) and Beauty
and Sublimity: A Cognitive Aesthetics of Literature and the Arts (Cambridge
University Press, 2016).
Arthur M. Jacobs is Professor of Experimental and Neurocognitive
Psychology and founding director of the Dahlem Institute for Neuroimaging
of Emotion (D.I.N.E.) at the Freie Universität Berlin (FUB). As part of
the highly interdisciplinary Languages of Emotions project at the FUB,
Professor Jacobs led a team investigating the Affective and Aesthetic
Processes of Reading. He is (co-​)author of more than 250 scientific publica-
tions in the fields of reading research, psycholinguistics, affective neurosci-
ence, and neurocognitive poetics, including the book Gehirn und Gedicht
[Brain and Poetry] (2011).
Karin Kukkonen is Associate Professor in Comparative Literature at
the University of Oslo and Academy of Finland Postdoctoral Research
Fellow. She has published on cognitive approaches to comics and graphic
novels (Contemporary Comics Storytelling, 2013)  and embodied and prob-
abilistic cognitive approaches to literary narrative, as well as on the 18th-​
century novel. Her forthcoming monograph A Prehistory of Cognitive
Poetics: Neoclassicism and the Novel brings the neoclassical criticism of the
17th and 18th centuries (which was informed by the ‘new science’ of the
time) into conversation with today’s cognitive approaches to literature. In
a project funded by the Academy of Finland, Kukkonen is currently pursu-
ing research on how the rise of embodied strategies of style and narration

Contributors  [ xiii ]
 

in the 18th-​century novel contributed to the immersive, gripping nature

of the genre.
David S. Miall received his doctoral degree from the University of Wales
at Cardiff, after which he taught for 10 years at the College of St Paul &
St Mary in Cheltenham. He moved to Canada in 1989 and took up a posi-
tion in the Department of English in 1990, specializing in literature of the
British Romantic period. His research interests include empirical study of
literary reading—​a field in which he has collaborated with Don Kuiken in
Psychology since 1990. The first of several federal grants for this research
was awarded in 1992, the latest in April 2008. In addition to his book
Literary Reading: Empirical and Theoretical Studies (2006), he is the author
of over 140 chapters and scholarly articles. He became Emeritus Professor
on his retirement in 2012. He now resides in France.
Micah L. Mumper is a doctoral candidate in the Cognitive Science program
at Stony Brook University. He is advised by Dr. Richard Gerrig. Using a
combination of behavioural and self-​report methodologies, he studies how
readers’ global and moment-​by-​moment experiences of fictional worlds
influence narrative impact. In particular, his research considers how basic
cognitive processes support comprehension, how reading may benefit
social-​cognitive abilities, how narratives affect readers’ moral judgements,
as well as how fiction influences real-​world attitudes and behaviours.
Keith Oatley read Natural Sciences at Cambridge and did a PhD at University
College London. With Maja Djikic and Raymond Mar, he has been involved
in developing the psychology of fiction, a movement in which literary anal-
yses are combined with empirical and theoretical research in psychology.
His work has included the relation of reading to writing fiction, exploration
of how literary works enable people to transform themselves, and research
on how reading fiction encourages empathy with others. Among his books
on fiction is Such Stuff as Dreams (2011). He has also published three nov-
els, the first of which won the Commonwealth Writers Prize for Best First
Book. His most recent book is a novella combined with psychological analy-
ses, The Passionate Muse (2012).
Caroline Pirlet is now working as a management consultant in Frankfurt/​
Main. During her time as a PhD candidate at the International Graduate
Center for the Study of Culture (GCSC) in Giessen, Germany, she was a
visiting scholar with Project Narrative at Ohio State University (Columbus,
OH). Her research has focused on the affective dimension of understanding
narratives, and reception-​orientated cognitive-​narratological approaches in
particular. She co-​authored the entry Narratology in English and American

[ xiv ] Contributors
 

Studies (with Monika Fludernik, ed. Martin Middeke et al., 2012), contrib-
uted to Unnatural Narratives—​Unnatural Narratology (ed. Jan Alber and
Rüdiger Heinze, 2011), and reviewed Patrick Hogan’s Affective Narratology
(Germanisch-​Romanische Monatsschrift, 2013). She is currently gaining fur-
ther qualifications as a business coach and doing independent research on
emotions and narrative coaching.
Merja Polvinen is a research fellow at the Helsinki Collegium for Advanced
Studies. Her work focuses on interdisciplinary approaches to literature, lit-
erature and the natural sciences (Reading the Texture of Reality: Chaos Theory,
Literature and the Humanist Perspective, 2008), and cognitive approaches
to literary representation. She is co-​editor of Rethinking Mimesis (2012)
and has recently published articles in the Journal of Literary Semantics and
Interdisciplinary Literary Studies. Polvinen is also a member of the network
Narrative and Complex Systems (University of York), board member of the
Finnish Society for Science Fiction and Fantasy Research, and co-​organizer
with Karin Kukkonen of the Cognitive Futures in the Humanities confer-
ence in Helsinki in June 2016.
Emily T. Troscianko (http://​www.troscianko.com) is a Research Associate
in the Faculty of Medieval and Modern Languages at the University of
Oxford, and in 2014‒2015 was a Knowledge Exchange Fellow at the Oxford
Research Centre in the Humanities, collaborating with Beat, the leading
UK eating disorders charity. The book from her doctoral thesis, Kafka’s
Cognitive Realism, came out with Routledge in 2014, and she is now work-
ing at the intersection of the cognitive and medical humanities, while
co-​authoring, with Susan Blackmore, the third edition of the psychology
textbook Consciousness: An Introduction (forthcoming 2018).
Andreas Wirag is a PhD student at the Teaching & Learning Processes
(UpGrade) Graduate School at Koblenz-​Landau University. As a secondary
schoolteacher of foreign languages and a university lecturer, he empirically
investigates the interface of cognitive approaches, literature, and language
education. He is currently working towards his PhD thesis on the employ-
ment of prototype theory in second language vocabulary acquisition.

Contributors  [ xv ]
 
 

Cognitive Literary Science

 
 

Introduction
A Window on to the Landscape of Cognitive
Literary Science
EMILY T. TROSCIANKO AND MICHAEL BURKE

I n 2013, we asked what the prospects were for the field of cognitive liter-
ary studies not only offering tangible benefits for our understanding of
literature (which it has and continues to do) but also starting to think of
itself, and be thought of by others, as able to offer benefits back to the cog-
nitive sciences that inform it. In our special issue of the Journal of Literary
Semantics (Burke and Troscianko, 2013), we included four examples of
work that made this recursive move back to the scientific side: papers on
parallel processing and consciousness, affect and artifice, the imagination
across the disciplinary divide, and the neuroscience of rhetorical style were
followed by a coda from a neuroscientist asking ‘Can literary studies con-
tribute to cognitive neuroscience?’ (and concluding yes).
Over the past few years, it has been gratifying to see a subtle but distinct
shift in the tone of many contributions to the cognitive-​literary field: not
across the board, but more conspicuously now than before, researchers
working with cognitive concepts, findings, and debates seem to be engag-
ing with them more in the spirit of confident give and take. Not that there is
anything wrong with applying a relevant idea from another field judiciously
to a question in another: this kind of work can be exciting and illuminat-
ing. It is probably also the most sensible first step in an encounter between
disciplines: find something (probably something quite solidly documented)
 

from ‘the other side’ that speaks to a question you already had, or some-
thing that opens your eyes to a question you had never quite thought of,
and see where it takes you. This ‘simple’ strategy of cross-​disciplinary
application is in practice often not very simple at all, and if it takes you as
far as a new insight into an issue of text or response that had previously
been opaque, this in itself is already a real achievement. That should not
be forgotten when we tell ourselves that one-​directional ‘borrowing’ isn’t
enough; it is already a lot.
Quite often, though, it happens that along the way, the act of applying
one thing to another actually makes you rethink the thing (the theory or
method) being applied. In the most basic sense, new evidence for something
(like, say, the characteristics of autobiographical memory as evidenced in
a fictional evocation of memory or a reader’s response to it) always tells us
more about that thing—​and when the evidence comes from something as
unlike the standard experimental psychology or neuroscience protocols as
a work of literary fiction, it would be surprising if something qualitatively
new were not learnt about memory or whatever it might be.
In more emergent areas of scientific inquiry, the likelihood of reciprocal
benefit is greater still: if a subfield explicitly acknowledges its own works in
progress, it automatically opens up space for input from other areas. This is
one of the things that makes the cognitive-​literary dialogue so promising
in the first place: there is so patently so much still to be learned in so many
and varied corners of the cognitive-​scientific field, as well as the literary
one, that nearly everything is still up for grabs.
And up for grabs does not mean the literary people are coming in and
grabbing stuff the scientists would rather keep for themselves. It’s easy,
working in an area where the most obvious method has seemed to be the
application/​borrowing one, to come unthinkingly to the conclusion that
no one on ‘the other side’ cares what you do. This impression is bolstered
by the practical facts that departments and journals and funding bodies
tend to adhere to the disciplinary boundaries, so the opportunities for
researchers in different fields simply to come across your work can be lim-
ited. Nevertheless, researchers tend to become researchers because they
are generally curious, open-​minded people, and our experience is that this
applies unequivocally to those trained in the empirical method: for people
who run experiments as an everyday part of life, the point is to have ques-
tions and enjoy figuring out how to pose them in answerable ways and
then trying to answer them, all the while knowing that your knowledge
will never be absolute.
A few weeks ago, one of us (Troscianko) spoke to someone at a cognitive
classics conference in Oxford who had been involved in an event bringing

[ 2 ] Introduction
 

together psychologists with humanities scholars, and who reported that

one of the scientists had said to her at some point during the event, ‘It’s
obvious what we get out of it, but why would literary people want to col-
laborate with us?’ Ironically enough, this is what humanities researchers
seem to think most of the time too. It seems that for whatever combina-
tion of perfectly explicable reasons—​institutional habit, intellectual inse-
curity, the allure of the greener grass everywhere else—​both ‘sides’ have
concluded that, well, they would quite like to collaborate with the other,
but the other would never be interested in reciprocating.
It’s very easy (for us) to enumerate all the reasons why the humanities
end up thinking this: the apparent status imbalance, the consequent feeling
of being under-​appreciated, the consequent feelings of defensiveness …
But it’s a shame, because all this conspires against giving it a go, whatever
‘it’ may mean in any given context: emailing that person whose paper you
liked but didn’t quite understand, setting up lunch to talk about your very
hazy ideas for an experiment, inviting someone from slightly academically
further afield to speak at your seminar series. This is especially sad if the
scientists do in fact really value the qualitative depth or conceptual sub-
tlety apparent in our work—​but we never get to find out.
However, if you’re reading this book, you are probably one of the people
who does do these things and continues to do them because you see that
they are worthwhile—​if only in making your working day more stimulat-
ing. We know there are a lot of you out there, and we are not going to pre-
tend that this volume is in any sense representative of cognitive literary
research as a whole, except insofar as it showcases the sheer variety and
creativity of our field.
Most of our contributions are single-​authored chapters, and the two
exceptions are co-​authored by researchers from the same field, but we
imagine (and in many cases know) that they are all based on energetic and
careful conversation with people from that ‘other side’: at conferences and
seminars, in common rooms and over lunches, by email, and even through
periods working closely with people trained very differently, in open-​
minded lab groups or interdisciplinary institutes. In the rest of this intro-
duction, we try to draw out some of the commonalities and differences
between the topics tackled and the angles adopted by our contributors;
there are thematic threads to be traced and recurrent patterns of perspec-
tive and method. But our guide in conceiving this volume was not thematic
or method-​specific; it was structural in a broad disciplinary sense.
Many of our contributors took part in a symposium on Science and
Literary Criticism (Burke and Troscianko, 2012) which we held at St John’s
College, Oxford, in the spring of 2012. The talks given there were as diverse

Introduction  [ 3 ]
 

as the title suggests, and the small size of the event combined with the vari-
ety of topics and backgrounds meant we were able to have intimate conver-
sations about the promise and problems of the field. We talked about the
‘laboratory liability’ and what experiments can really be expected to teach
us; about how systems of theoretical knowledge interact; about all the tim-
escales from the evolutionary to the neural; about how much interpersonal
variability there really is; about expertise and the blank-​slate reader, nor-
mality and averaging, introspection and the unconscious, rigour and fidel-
ity. Questions about disciplinary balance and reciprocity have been with us
since, and the idea for this book was to try to instantiate both.
In this spirit, the three parts of the book present the three main itera-
tions on ways of working in the cognitive-​literary field. In the first part,
which would often be thought of as cognitive literary studies proper, lit-
erary scholars draw on some aspect of cognitive science to offer a new
viewpoint on literature or literary reading. In the second, literary scholars
use literary materials or conceptual frameworks to contribute to cognitive-​
scientific debates. In the third, cognitive scientists engage with literature
and literary-​critical methods to shed light on questions in their home dis-
ciplines and/​or those in literary studies. Arguably for total symmetry there
should have been four parts, but in practice we found that the contribu-
tions from cognitive scientists tended in any case to have a dual focus: cast-
ing light on the literary phenomena and on the cognitive. So separating
them out would have felt a little artificial.
In 2013, we suggested the term ‘cognitive literary science’ for a form of
cognitive literary studies that takes its place assertively beneath the capa-
cious cognitive-​science umbrella, giving and receiving in equal measure—​
maybe so it stops even feeling like exchange, and starts feeling simply like
what we do. Originally our thought was that Part I here might not quite
count as part of cognitive literary science thus defined, but as should
become clear in the following survey, it now seems right and important to
see all three variations on cognitive literary research as integral to what a
grown-​up ‘CLSci’ will look like.
Of course, the argument could be made that this model makes the inher-
ently limiting assumption that everyone will be working on their own and
that every individual researcher has only one ‘home’ discipline. Clearly
neither of these things need or should be true. Collaborative work that
in its everyday practices crosses the divide or even forgets that the divide
exists is one of the best ways of making interdisciplinarity meaningful. And
many people have eclectic and active enough academic backgrounds that
pigeonholing them by department makes little sense. But even where these
things are the case, perhaps there is still something to be said for the rough

[ 4 ] Introduction
 

outlines of our structures; perhaps, especially while the field is still rela-
tively young, the directional currents can still on the whole be discerned,
and can tell us interesting things when we stop to look at them.
In the spirit of learning through careful observation, the remain-
der of this introduction will be devoted to an overview of the follow-
ing chapters that asks broad questions about some of the similarities
and differences between our contributions. We will not give a blow-​by-​
blow summary of what each chapter. Instead, it has been interesting
to reflect, at the end of a long editorial process, on the composition
of the book and what it might tell us about the present and future of
the field. Again, we make no claims to representativeness, but 15 chap-
ters in 3 parts offer a decent-​sized window on to the state of CLSci in
2016:  where, right now, are our colleagues applying cognitive-​literary
approaches, to what purpose, with what methods and assumptions? Is
it even possible to generalize at all?
We will start with a few simple questions.
—​ Who? Our contributors range from established to mid-​and early-​
career scholars, working in the United Kingdom, the United States,
Canada, New Zealand, Germany, Finland, and Norway, originally trained
in literary studies, linguistics, and experimental and neurocognitive psy-
chology, and now practising at the interesting intersections of those fields
and others.
—​ What? Broadly speaking, the topics being addressed in our chapters
are the ones we would expect to see. The current Big Six cognitive-​literary
topics—​embodiment, emotion, immersion, mental imagery, simulation,
and social cognition—​are salient in the majority of the chapters. Even
when the focus of inquiry is something less ubiquitous and more specific,
like the sublime in David Miall’s chapter or the fantastic in Karin
Kukkonen’s, the conceptual underlay is still shaped by those major themes.
The three exceptions are perhaps the two chapters on different aspects of
readerly pattern extraction (by Alexander Bergs and Brian Boyd), and James
Carney’s chapter on construal level theory and science fiction. The latter
proposes hypotheses about psychological distance and abstraction that
certainly touch on empathy and emotion more broadly, but subordinates
them to questions about the foregrounding of human agency through spe-
cific stylistic means. The role of feedback and predictive processing, which
Kukkonen describes as ‘still vastly under-​represented in cognitive literary
studies’, makes an appearance in two other chapters—​briefly in Caroline
Pirlet and Andreas Wirag’s, and centrally in Emily Troscianko’s—​and feels
like one that could play an important connective role in the future of the
field, with clear relevance to topics like the interplay between memory and

Introduction  [ 5 ]
 

emotion discussed by Patrick Hogan or between immersion and reflection

as explored by Merja Polvinen. If, as Andy Clark predicts, predictive cod-
ing turns out to kick off one of ‘the major intellectual events of the early
twenty-​first century’ (2013, p. 21), then it seems likely that CLSci will get in
on the act. But then, we are currently keen on this, so our predictions may
be revealing primarily of our ‘priors’.
—​ Where? If we look at the cognitive-​scientific disciplines on which our
contributors are drawing, we again find the usual suspects of experimental
psychology and neuroscience, with a little bit of evolutionary psychology
and some philosophy of mind, notably an adapted form of ‘heterophe-
nomenology’ (David Herman). Herman also brings in ethology and some
anthropology, which seems like an obvious area for expansion in CLSci,
as do developmental psychology and questions about life-​course changes
(broached by Richard Gerrig and Micah Mumper, and by Keith Oatley) and
the medical/​psychiatric realm dealt with by Troscianko. Alexander Bergs
and Arthur Jacobs make cognitive (neuro)linguistics central to their chap-
ters, though it does not feature much in the contributions of those outside
that field, with the exception of Miall, who explains how EEG findings on
functional shift speak to the style of the sublime. There is, though, a dis-
connect across the field between researchers who adopt a linguistics model
and those who do not which continues to feel surprising—​it would be nice
to see more integration on this front in future. A possible facilitator here
could be the field of cognitive stylistics: the linguistic analysis of literary
texts conducted through the lens of either cognitive psychology or cogni-
tive linguistics. Another common absence also found here is that of social
psychology: like anthropological methods, it tends to be under-​represented
in CLSci, as it is here (though there is a little discursive psychology in Pirlet
and Wirag’s chapter).
Is this because when turning to ‘science’, the inclination is to turn to
the ‘harder’ rather than the ‘softer’ versions first, because they promise
the most solid foundation of empirical validation? Marcus Hartner would
warn us that the principle of autonomy should make us hesitate before leap-
ing over too many explanatory levels on the path between our home dis-
ciplines and those we make connections with. We would also add that a
cogent link from the humanities to social psychology can be found in the
precepts and principles of classical rhetoric and its modern guises of persua-
sion and communication studies. Meanwhile, it’s clear that for the majority
of our contributors, the behavioural and self-​report methods of experimen-
tal psychology are the natural stepping stone: not too near and not too far.
Generally speaking, though, there seems quite a contrast between those
contributors who (to oversimplify somewhat) jump straight to the science

[ 6 ] Introduction
 

and those who ground their arguments in theoretical or empirical work that
has already taken place within the cognitive-​literary field. That preference
will depend on all kinds of factors including subject matter and probably
personality, but tracking whether the relative proportions change over time
may tell us something about the likely future size, shape, and constitution
of CLSci.
Another ‘where?’ question we might ask, of course, is a cultural-​
geographic one: where do our contributors’ primary texts come from? In
this we are, for the obvious pragmatic reasons, fairly Anglocentric, but
Kukkonen introduces us to an 18th-​century French novella and Jacobs
guides us through the word valleys, sentence slopes, verse lifts, and stanza
rises of German linguistic beauty, idiom, and poetry. Where time and
expertise permit, it would be great to see more cultural-​linguistic diver-
sity in future CLSci studies out beyond the main Germanic and Romance
languages.
—​ When? The primary texts our contributors discuss range from
Longinus, reproducing in the 1st century c.e. a poem by Sappho from
6 centuries earlier (Miall), to three North American novels from 2013
(Gerrig and Mumper). Shakespeare’s sonnets win the prize for the most
attention, with three contributors considering them. Otherwise, the 20th
century is the best represented, as might be expected—​but with much less
of a focus on high Modernism than has often been the case.
—​Why and how? These two questions meld somewhat into one, since
it’s hard to neatly separate out the question being asked from the method
used to answering it. The methods adopted by our contributors take in
the full range from meta-​theoretical overview (Hartner) to theories that
encompass facets of the overarching distinction between lyric and narra-
tive (Boyd) or fiction and non-​fiction (David Herman); from accounts of
genre characteristics (Carney) and rethinkings of disciplinary structures
and boundaries (Jacobs, Pirlet and Wirag) to inquiries into literary phe-
nomena like the sublime (Miall) and the fantastic (Kukkonen) or linguistic
phenomena like coercion (Bergs); from a question about how a particular
cognitive context or individual history changes the reading experience
(Gerrig and Mumper, Troscianko) to higher-​level ones about why readers
(critical and recreational) vary and resemble each other in their responses
(Raymond Gibbs) and how reading changes people (Oatley); from a chal-
lenge to received ways of thinking about readerly engagement (Polvinen) to
a knotty puzzle posed by a specific text (Hogan). It will become clear to you
once you read them, though, that these encapsulations are only one way of
conveying what the chapters do: we could just as well describe Herman’s as
a critical survey of the problem of non-​human other minds, or Kukkonen’s

Introduction  [ 7 ]
 

as a case study on the probabilistic models of the Bayesian reader. But the
variety of scales and scopes of questions asked and evidence presented,
approached with deductive and/​or inductive methods, all with their own
rationales and priorities, makes clear that there really is no single template
for a standard CLSci publication: we could hardly be any further from, say,
a field in which all anyone does is apply a scientific finding to the reading of
a single text to generate a new interpretation. This can be and is done bril-
liantly, but there are a myriad other options for researchers in the field, and
it is heartening to see the inventiveness keep growing. Sceptics may say
that this heterogeneity is the field’s fatal flaw, but it must also be its forte.
When it comes to the use of primary literary texts, too, there is a huge
range of strategies, from more or less close readings of just one or a very
few texts to high-​level surveys of general characteristics of a large num-
ber of texts or analysis of numerous small text fragments, to chapters that
do not discuss specific texts at all. Interestingly, the closest reading and
the very broadest argument go hand in hand in Boyd’s chapter on the con-
trast between narrative and lyric; here the specifics of textual patterning
are analysed at the lowest level to provide evidence for the ultra-​high-​level
hypothesis about the levels of effort required for cognitive pattern extrac-
tion. And while for the most part the texts considered are literary prose
fiction, poetry, and drama, Kukkonen brings in discussion of the links
between literature and visual art, and Herman compares and contrasts fic-
tional and non-​fictional accounts of non-​human minds.
Having exhausted the ‘Five Ws and an H’, our next set of questions
relates to the currents and tensions of interdisciplinarity: in the shifts or
mergings between disciplines, is consensus emerging or not, what happens
to terminology, to what extent are attitudes critical or embracing or both
at once, and are people worrying about the interdisciplinary or just getting
on with it?
There are some striking points of convergence in our contributors’
conclusions—​Bergs and Boyd on the centrality of pattern recognition in
(literary) reading, as already noted, or Oatley and Polvinen on the nature
of literature as cognitive training. There are some areas of divergence too,
whether in differing attitudes to things like measures of transportation
(compare Gerrig and Mumper with Polvinen), or in thinking about whether
contrasting attitudes to texts manifest through simultaneity or vacillation
(compare Polvinen and Kukkonen). We see these differences not as incom-
patibilities, but as excellent starting points for future exchange.
In many of the chapters, there seems to be an easy interplay between
concepts and terms deriving from the cognitive and the literary side—​cog-
nitive frames and natural narratology, construal level and characterization,

[ 8 ] Introduction
 

the P600 response and the sublime—​with established terms of literary ref-
erence clearly still serving useful purposes when put in dialogue with oth-
ers that have quite different histories and conventions. The use of certain
cognitive terms indicates that there is still a lot of fluidity in the concep-
tual systems in use: Oatley, for instance, uses inferencing, theory of mind,
and simulation in an inclusive way that the more terminologically hard-​
line might say one shouldn’t. Who knows where the scientific and memetic
competition will take us in the end. Perhaps surprisingly, though, no one
suggests that we need to replace existing concepts with new ones designed
specifically for cognitive-​literary purposes: although there is plenty of crit-
ical engagement with the definitions and/​or implications of well-​known
concepts—​like heterophenomenology in Herman, or aesthetic illusion in
Polvinen—​the tendency here seems to be to work with the terms we have
inherited rather than offering up new ones.
On the matter of critique, we might expect the contributions in our sec-
ond part—​literary scholars offering something back to the sciences—​to
be the most overtly critical of scientific practices and frameworks, and this
turns out to be the case:  Hogan remarks on the limitations of lab-​based
experiments, for example, Polvinen on the problems with thinking com-
putationally about the imagination, and Herman not only on the need to
rethink narratology with the help of philosophy and anthropology but
also on how elements of that philosophy can and should be rethought
with the help of literary insights. By contrast, though, both Kukkonen and
Troscianko apply feedback or prediction principles quite uncomplicatedly
to the study of literature, but both with the aim of advancing the study
of the cognitive phenomena under discussion:  predictive processing and
disordered eating, respectively. That said, the contributors to our third
part are happy to acknowledge the limitations of current scientific practice
too: Gibbs in relation to typical literary reading studies investigating ‘naïve
readers’ first-​time pass through, and quick comprehension of, brief seg-
ments of text, usually artificially constructed for experimental purposes’,
for instance, or Bergs on the ‘substantial drawbacks’ of fMRI. (Although as
a linguist working at an Institute for English and American Studies using
historical and solidly empirical methods, Bergs is an excellent example
of where the opposition of ‘scientist’ versus ‘humanities scholar’ breaks
down.) A bit of healthy scepticism about traditional literary-​critical meth-
ods might also be anticipated from the scientists writing in Part III, but
this is not really in evidence at all, with the possible exception of Gibbs’s
comments on the tendency of critics to think of their acts of reading as
quite unlike those of ‘ordinary’ readers, and so to feel legitimized in reject-
ing findings about the latter as inapplicable to critical reading. A brief note

Introduction  [ 9 ]
 

of warning is, however, sounded by Carney when he considers what hap-

pens when prescriptive notions of the literary collide with ordinary read-
ers’ experiences—​and advocates siding with the latter.
As for whether interdisciplinarity itself is the object of questioning,
doubt, or other kinds of meta-​reflection, on the whole it seems not to be.
Assessing the status quo and offering suggestions for how to strengthen
the field is the point of Hartner’s opening chapter, but otherwise, though
most of our authors give brief scene-​setting remarks about the disciplin-
ary encounters they will be drawing on, these are more explanation than
defence, and the usual procedure seems to be:  set out why a cognitive-​
literary approach is meaningful, and then put it into practice. On the
meta-​level, Hartner makes some concluding suggestions about the aims of
interdisciplinary research that contrast with Herman’s position on ‘trans-
disciplinarity’, suggesting that although conducting research that demon-
strates the benefits of the humanities in broader contexts is an excellent
aim, it needn’t be one we always have in mind: ‘Literature is worth studying
for a vast variety of reasons; not all of them will necessarily be of scientific
or transdisciplinary value.’
By now it will have become clear that, like any categorical structure
imposed on complex works of individual scholarship, our ordering schema
is far from watertight. It’s easy to make the case, in particular, that the
contributions in Part I  offer ‘transdisciplinary’ benefits back to the cogni-
tive sciences just as those in Part II do. By offering rich evidence of cogni-
tive phenomena that are manifested in salient and complex ways in literary
encounters, they arguably do what Kukkonen says of literature and ‘fantas-
tic cognition’:  throwing each of their cognitive subjects into sharp relief,
cognitive literary study ‘helps make … more or less automatic features of
cognition noticeable and thus subject to analysis’. One of the most subtly and
unexpectedly encouraging trends in the whole book, actually, is that many of
our authors do not seem to feel the need to specify, in disciplinary terms,
where the projected benefits of their contributions lie: when investigating
what distinctive processes might be involved in the reading of full-​length
novels, or how the linguistic phenomenon of coercion behaves in aesthetic
contexts, or what exactly the sublime is, these questions are of intrinsic
interest, and working out which ‘side’ ‘gains’ more from any given increase
in understanding may be beside the point. Of course, articulating which
disciplinary stockpiles we want to contribute to is often important, but
sometimes we can allow our questions and answers to speak for themselves.
Finally, we might ask how the classic flashpoints of the cognitive literary
field are dealt with by our contributors. We might name three in particu-
lar: How do findings about averaged-​out experimental participants relate

[ 10 ] Introduction
 

to insights into individual experiences? How do the theoretical and the

empirical relate to each other? And how do findings about 20th-​or 21st-​
century experimental participants relate to questions about texts many
centuries or even millennia old?
The matter of the individual versus the general is broached in many
chapters, and takes centre stage in two—​interestingly, both by scientists
(Gibbs, and Gerrig and Mumper). This concern from the scientific side with
the specificity more usually thought of as the domain of the humanities is
echoed in Oatley’s chapter too, though Pirlet and Wirag also engage with it,
and it makes brief appearances in lots of other chapters. The problem and
a solution are expressed concisely by Carney, who notes that literary texts
inherit the variability of the human mind, and so can suffer from shoehorn-
ing typologies, but that both also have regularities which emerge at the
statistical level, so that generalizations are not meaningless. Many of our
chapters are beautiful demonstrations of the simple reality that although
empirical methods can be used to iron out the differences between people,
they can also be used to highlight those differences—indeed, empirical
work that investigates responses other than one’s own is the only way of
doing that. This is part of perhaps the most immediately satisfying justifi-
cation for the entire field of CLSci (should one be needed): that instead of
basing conclusions about textual effects on the singular experience of the
critic-​as-​reader disguised as the generic reader, or accumulating new inter-
pretations of texts without acknowledgement of the cognitive factors on
which they depend, we can understand interpretations as cognitive effects,
and investigate their natural variations in others as well as ourselves. This
logic is put into practice not just in the kinds of research questions and
empirical evidence manifest in our contributions, but in some cases also in
their approaches as expressed through choice of writerly tone: the chapters
by Gibbs and Jacobs, both scientists, make particularly clear that personal
experience is a touchstone for how research is conducted and/​or conveyed.
Another question bound to be asked about work in CLSci as conducted
by researchers trained in literary studies is what combination of theory
and empirical work it draws on or contributes to. David Miall is well known
as one of the pioneers of empirical literary studies, and Gibbs, Gerrig and
Mumper, Oatley, and Jacobs all present findings from experimental work
they have carried out. Amongst our humanities contributors, several give
clear outlines of how their suggestions could be tested empirically: a hypo-
thetical study using the three extant versions of Cazotte’s 18th-​century
novella to see how readers tread the interpretive line between the uncanny,
the fantastic, and the marvellous (Kukkonen), or a prediction of changes
to readers’ approval of Joyce’s eponymous heroine Eveline as a function

Introduction  [ 11 ]
 

of their varying tendencies towards empathic engagement with other

people (Pirlet and Wirag). Troscianko presents some pilot data from the
start of a ‘knowledge exchange’ collaboration with a mental health char-
ity, and Herman also makes clear the real-​world ethical implications of the
research project he sketches out: mental-​state attribution in narrative can
have effects back on the discourse domains in which they are located, and
thus help change how we think about other species’ minds and so treat
other animals.
When it comes to testing hypotheses, or even making the hypotheses
in the first place, about writers or readers of texts written centuries or mil-
lennia ago, there are obvious complications—​indeed, they were raised by
one of the reviewers of our book proposal as needing more attention—​
but our contributors do not seem fazed by them. Miall’s chapter deals with
the historically furthest removed textual examples, but he aligns Sappho’s
poem with travel accounts from the 18th and 19th centuries, and his analy-
sis makes clear that there is no principled obstacle to creating this kind
of line of connection between periods. Boyd’s argument relates to a fun-
damental enough feature of human cognition that illustrating it through
Shakespeare’s sonnets poses no problems. Kukkonen’s argument links the
history of aesthetic trends with the predictive models they engage, but in
general from our contributions one can infer that an adapted version of
Carney’s response to the ‘problem’ of individual variation applies to that of
historical variation: there are variations, but there are also commonalities.
The difference between the two cases is, of course, that if we have hypoth-
eses about reader response (or indeed authorial creation) that we want
to test, this simply cannot be done with historical readers or writers as it
can with 21st-​century individuals; the most we can do are observational
studies along the lines of corpus analysis. But maybe this does not hugely
matter. We can do experiments with readers now and interpret our results
in the light of wide-​ranging evidence of what is known about historical-​
evolutionary trends in human cognition. This requires more interdisciplin-
ary collaboration, but perhaps that is no bad thing. It is certainly one more
tempting territory staked out for future exploration.
What we take from this survey of the territory of CLSci, at least as it
is inhabited by the 17 contributors to this volume, is the sense of a field
growing confidently into maturity. We imposed the tripartite structure,
but probably we needn’t have: people are doing all kinds of creative borrow-
ing and lending, from different starting points and with varying aims. You
must judge for yourself whether Hartner’s three principles for a respon-
sible CLSci are being adhered to, or whether you agree with them in the
first place, and we are sure you will have your own set of criteria by which

[ 12 ] Introduction
 

to assess what follows. But we hope that your reading experience will have
something of the quality of eavesdropping enjoyably on a mixture of ani-
mated conversations.
As for the near future of CLSci, well, we predict that scholars and sci-
entists from across the disciplines will work together more frequently on
closely collaborative projects, and that these projects will develop new
ways of doing mixed-​methods research combining theory with qualita-
tive and quantitative measures. We also predict that 4E cognition—​the
embodied, the embedded, the enactive, and the extended—​will stay big,
but grow more differentiated as debates on what strength of claim can be
made about the contributions of context to cognition continue to mature.
Investigations of contextual effects, priming, and framing will, we imagine,
connect the linguistic and the rhetorical more closely with the other aspects
of the cognitive. There will be more work on how important dimensions
of reader variation affect the processes and the outcomes of literary read-
ing, and how these interactions may have implications for today’s social
and psychiatric challenges. The ever-​seductive question of whether reading
literature makes us better (cleverer, more empathic, more moral) people
will be tackled from new angles, especially by developing ways of track-
ing longer-​term changes in readers’ mental states and behaviours. In this
regard, we anticipate an increasing concern with more ecologically valid
methods for studying literary reading empirically, via more dialogue with
social anthropology and mobile tech innovation. Lab-​based experiments
will continue to ask detailed questions about readers’ responses, with the
4E paradigm bringing the haptics, kinaesthetics, and ergonomics of liter-
ary text processing under scrutiny. Neuroscientific methods will, we hope,
grow more nuanced too, as conceptual developments like ‘second-​person
neuroscience’ accompany technological advances. All this should keep the
cognitive literary scientists of the near future agreeably busy. Of course, we
could be wrong about any or all of this, but in a field as young and as vibrant
as ours, there is nothing particularly disquieting about that.

REFERENCES

Burke, Michael, and Emily Troscianko. (2012). Science and literary criticism. St John’s
College Oxford. Programme and abstracts at http://​www.sjc.ox.ac.uk/​4073/​
Science%20and%20Literary%20Criticism%202012_​v4.pdf.download
Burke, Michael, and Emily T. Troscianko. (2013). Explorations in cognitive literary sci-
ence. Journal of Literary Semantics, 42(2).
Clark, Andy. (2013). Whatever next? Predictive brains, situated agents, and the future
of cognitive science. Behavioral and Brain Sciences, 36(3), 181–​204.

Introduction  [ 13 ]
 
 

PART I

Literature through a Cognitive Lens

 
 

CHAPTER 1

Scientific Concepts in Literary Studies

Towards Criteria for the Meeting of Literature
and Cognitive Science
MARCUS HARTNER

INTRODUCTION

Cognitive approaches to the study of literature are not a new phenom-

enon. More than 25 years ago, inspired by the cognitive turn in linguis-
tics, Eldrud Ibsch published an article on the ‘The Cognitive Turn in
Narratology’ (1990), and at the turn of the millennium Tony Jackson
(2000, p.  340) suggested that the influence of cognitive science might
soon become irresistible. ‘How can empirically established, scientific
claims about the biology and psychology of reading, writing, and respond-
ing’, he asked, ‘have no bearing on the discipline whose meat and pota-
toes is reading, writing, and responding?’ Yet, despite such predictions,
the overall situation of cognitive approaches within the academic study
of literature has changed only partially over the last two decades. General
interest in the field has undoubtedly increased over the past years, but
many students of literature still complete their studies without ever being
introduced to cognitive approaches, and the majority of scholars remain
sceptical, or at least indifferent, towards this field of research. Cognitive
approaches, in other words, have remained ‘on the verge of affecting the
mainstream’ (Burke and Troscianko, 2012) but have never quite managed
to do so. As a result, the field’s breakthrough continues to be projected
 

into the future. It is still hailed as ‘the next big thing’ in literary stud-
ies (Cohen, 2010), despite the fact that cognitive approaches have been
around for more than 20 years.1
The reasons for the lack of enthusiasm many scholars in the humani-
ties display towards cognitive approaches are diverse. They range from
the field’s lack of a common theoretical framework to an old-​fashioned
concern about the ‘infiltration and contamination [of literary criticism]
by other disciplines’ (Waugh, 2006, p. 24). Furthermore, scholars such
as Paul Sheenan consider cognitive literary studies to be fundamentally
incompatible with the entire body of approaches traditionally subsumed
under the shorthand of ‘Theory’ (2014, p.  31). Cognitive approaches,
in his view, represent ‘a deliberate turn away from the historical, social
and political conditions that shape the literary, toward the universal
structures of cognition’ (p.  53). The concerns about cognitive literary
studies also include epistemological and methodological uncertainties
surrounding the intersection of empirical (cognitive) science and (non-​
empirical) literary studies, which have formed part of an ongoing debate
on the explanatory potential, the scope, and the problems of cognitive
approaches.2
In the context of this discussion, my chapter engages in a reflection on
methodologically sound ways of conceptualizing the meeting of science
and the humanities. Taking up some of the methodological and theoretical
issues affecting cognitive approaches, I try to identify general criteria by
drawing on a standard model of the structural relationship between dif-
ferent levels of scientific investigation, outlined, for example, by Patrick
Hogan (2003, pp. 202‒210). From this model, I derive a set of basic heu-
ristic guidelines for cognitive literary studies: the principles of coherence,
moderation, and autonomy. By elaborating on the conceptual underpinnings
and the practical consequences of those guidelines, I hope to provide some
practical orientation for research situated at the intersection of literature
and science. In this way, I  aim both to further the field’s development
and to counter some of the methodological criticism to which it has been
subjected.3

1.  On the origins of cognitive approaches to literature, see Richardson (2004, p. 1)
and Vandaele and Brône (2009, p. 1).
2.  See, for example, Jackson (2000, 2005), Adler and Gross (2002), Hogan (2003),
Gottschall (2008), Slingerland (2008), Ryan (2010), and Sheenan (2014).
3.  For a more extended discussion of the project of cognitive literary studies and the
idea of heuristic guidelines, see also Hartner (2012, pp. 13–​56).

[ 18 ]  Literature through a Cognitive Lens

 

MAPPING THE FIELD: CRITICAL RESPONSES

TO COGNITIVE APPROACHES

Although general interest in the meeting of literature and cognitive sci-

ence has increased over the past years, there remains considerable resis-
tance towards cognitive approaches within literary studies—​a resistance
that takes the form sometimes of open hostility, sometimes of wariness,
and sometimes of indifference. On one level, such reactions are not sur-
prising. As we have learned from Thomas Kuhn (1962), new theoretical
paradigms typically have to struggle against established academic struc-
tures, or what Kuhn calls ‘normal science’, before they eventually take
hold and occasionally even lead to a paradigm shift in research. Moreover,
as Patricia Waugh (2006) reminds us, there has always been a somewhat
irrational anxiety in literary criticism about being infiltrated and con-
taminated by other disciplines. This concern is ‘as old as criticism itself’
(p. 24), but it is also somewhat paradoxical, since most critical approaches
within the last hundred years, from structuralism to gender studies and
deconstruction, have taken conceptual inspiration from outside the dis-
cipline. On another level, the sheer ferocity and breadth of this resistance
can seem baffling, with even scholars genuinely interested in cognitive
approaches, such as Marie-​L aure Ryan (2010), remaining overtly critical
towards the field.
One of the factors that may have kept cognitive literary studies from
going mainstream is its heterogeneous conceptual and methodological
foundations. Although most scholars working with cognitive approaches
share an interest in reading as a cognitive act, the field does not yet offer a
common theoretical framework or methodology; it creates the impression
rather of a kind of ‘interdisciplinary bricolage’ (Ryan, 2010, p. 476) and has
famously been defined in a rather loose way as ‘the work of literary critics
and theorists vitally interested in cognitive sciences and neuroscience, and
therefore with a good deal to say to one another’ (Richardson, 2004, p. 2).4
Over the past two decades, the work of those scholars has not solidified
into a shared set of concepts and/​or methods. Instead, a variety of equally
loosely defined subfields such as cognitive poetics, cognitive narratology,
and neuroaesthetics have emerged, each offering a diverse set of topics and
approaches. This diversity can be seen as an asset, but the lack of an identi-
fiable conceptual core has its downside, making it difficult, for instance, to

4.  For an overview of current trends in the field, see Jaén and Simon (2012).

S c i e n t i f i c C o n c e p t s i n L i t e r a r y S t u di e s  [ 19 ]
 

design introductory courses and textbooks covering the field as such, and
inviting others to view cognitive literary studies as methodologically vague
and theoretically confusing or inconsistent.
The latter criticism in particular strikes at the heart of the cognitive
field because notions of vagueness and inconsistency run counter to the
promise seemingly inherent in cognitive approaches to bring a touch of sci-
entific clarity into the notoriously vague study of literature. Indeed, both
the special appeal and the most poignant criticism of cognitive approaches
seem to be firmly tied to their interdisciplinary nature. Although Frederick
L. Aldama (2010) has claimed that C. P. Snow’s (1965) famous separation
between ‘the two cultures’ is artificial, ‘a line drawn in sand’ (2010, p. 1),
both critics and proponents of cognitive literary studies have repeatedly
highlighted the particular promises and challenges of attempting to bridge
the gap between the two spheres. While some critics have pointed out the
dangers of the scientific method (Sheenan, 2014, pp.  49‒52), others see
the move towards science as a way of establishing ‘a new humanities on
surer foundations’ (Gottschall, 2008, p. 176) and an opportunity ‘to make
the discipline and the institution of literature more accessible and more
connected with the world outside university and college life’ (Stockwell,
2002, p. 11). In any case, literary scholars who attempt to connect the two
cultures and venture into the realm of science seem to face a particular and
substantial set of challenges.
One of those challenges is connected to the perceived differences in
the nature of research in the two spheres of academia. The humanities
have long embraced the idea that research in literature and art does not
subscribe to the same methodology as the empirical sciences. While
Jonathan Gottschall (2008, p. 176) sees this as a disadvantage and com-
plains that the humanities seem to have ‘dismissed the possibility of gen-
erating reliable knowledge’, other scholars stress what they believe to be
an essential difference between scientific objects of inquiry and litera-
ture. The latter, in the words of Wolfgang Iser, ‘can be assessed, but not
predicted’ (2006, p. 5), as its study is neither centred on the analysis of
experimental data nor proceeds by solving explicitly spelled out problems
(pp. 5‒7). His position, which is still shared by many traditional literary
scholars, is based on the premise that empirical scientists and literary
scholars have fundamentally different research interests. According to
this somewhat simplistic and polarizing point of view, science attempts
to find evidence for solving problems and conclusively answering research
questions, while a successful or productive reading of a literary work is
based ‘on such parameters as originality, appropriateness, inventiveness,
or “insight value” ’ (Adler and Gross, 2002, p. 214). For traditional literary

[ 20 ]  Literature through a Cognitive Lens

 

scholars, textual analysis can accordingly not be measured by empirical

data but rather depends on

our degree of satisfaction with what is revealed or illuminated about a text.

Most of all, literary interpretation generally does not aspire to the once-​and-​
for-​allness implied by the term solution. On the contrary, it is often unabashedly
nonfinal, inviting supplementation or revision, in other words: conscious of its
own historicity. (p. 214)

Despite such claims there is, of course, no univocal agreement on the aims
of literary analysis and the practice of interpretation within ‘traditional’ lit-
erary scholarship. Nevertheless, many researchers working with cognitive
approaches still seem to be interested in creative readings and interpreta-
tions, and few have proposed that we should subscribe exclusively to sci-
entific methodologies. For this reason, scholars such as Marie-​Laure Ryan
see the field as being ‘uncomfortably sandwiched between the speculative
and interpretive disciplines of the humanities and the experimental disci-
plines of the hard sciences’ (2010, p. 474). This situation creates a unique
challenge. By drawing on data, concepts, and methods from the sciences,
cognitive literary studies appear to promise insight into the ‘reality’ of
reading and writing. As a consequence, even those aspects of the literary
scholar’s work which remain ‘speculative’ and/​or non-​empirical run the
danger of becoming invested with an air of scientific and empirical author-
ity. The project of cognitive literary studies generally conveys an impres-
sion of offering positivist clarity, empirical evidence, and the possibility of
definite answers. At the same time, few studies really live up to this image
and go ‘as far as adopting the rigours of experimentation’ (p.  476). Lisa
Zunshine’s (2007) investigations of ‘theory of mind’ and ‘metarepresenta-
tion’, Alan Palmer’s work on ‘social minds’ (2010), or literary applications
of Gilles Fauconnier and Mark Turner’s blending theory (e.g. Schneider and
Hartner, 2012) are examples of studies that design hermeneutic analytical
tools of their own by adapting concepts from the ‘hard’ sciences. As Ryan
points out, the majority of works in the field do not subscribe to the scien-
tific standards of the disciplines they borrow from, but usually remain ‘in
spirit strictly speculative’ (2010, p. 476).
Moreover, the strategy to borrow and adapt concepts comes with several
methodological pitfalls. As scholars seldom have thorough scientific train-
ing, they often lack detailed knowledge about the precise scope and the
potential shortcomings of the scientific theories they are using. A frequent
result, for example, is an exaggerated trust in specific scientific concepts or
perspectives, as work from areas in which one is an amateur often somehow

S c i e n t i f i c C o n c e p t s i n L i t e r a r y S t u di e s  [ 21 ]
 

feels more reliable than ‘the more familiar, but embattled assertions’ of one’s
own field of expertise (Spolsky, 1993, p. 41). As a consequence, literary schol-
ars, according to Hans Adler and Sabine Gross, frequently ‘succumb to the
seductiveness of scientific terms and import them into literary analysis with
little consideration for their actual scientific use, treating them in effect with
poetic licence and happily engaging in creative analogies’ (2002, p. 211).
There is, of course, nothing wrong with creative analogies per se. In
fact, instances of what Gregory Bateson calls ‘loose thinking’ (1972, p. 84)
often constitute the initial stages of creative and successful research.
Bateson recommends that researchers ‘look … for wild analogies to their
own material’ (p. 87) in other fields of inquiry and then move from ideas
with yet ‘unsound foundations’ to stricter scientific thinking (p. 86). In my
opinion, however, ‘loose thinking’ should be made recognizable as such.
When poetic licence comes in the disguise of scientific terminology, when
what claims to be neuroscience turns out to be ‘neurospeculation’ (Tallis,
2008), cognitive literary studies run the danger of undermining their own
credibility. Combining a rhetoric of scientificity with epistemological care-
lessness is a sure way of providing ammunition to the critics of cognitive
literary studies. Vicious attacks such as Raymond Tallis’s article on ‘The
Neuroscience Delusion’ bear witness to the fact that epistemological short-
comings do not go unnoticed; they trigger critical responses which accuse
cognitive approaches of promoting reductionist views that fail to do justice
to the complexity of both scientific theory and literary texts.
As a scholar deeply interested in cognitive approaches, I  believe that
we need to take seriously the concerns that have been voiced concerning
the special interdisciplinary nature of our field. The incorporation of sci-
ence into the work of literary scholarship arguably demands a heightened
degree of epistemological awareness and conceptual deliberation. In order
to forestall criticism, cognitive literary studies therefore need to engage in
a continued reflection on their concepts, aims, and methods. A helpful heu-
ristic tool for this purpose can be found in the model of explanatory levels.5

LEVELS OF EXPLANATION: A HEURISTIC MATRIX

Complex phenomena often require complex methods of analysis. The bio-

logical study of animals or plants, for example, is so multifaceted that scien-
tists cannot rely on a single approach in order to arrive at a comprehensive
understanding. A variety of perspectives and methods need to be enlisted

5.  See Hogan (2003, pp. 202–​210) and Paulson (1991).

[ 22 ]  Literature through a Cognitive Lens

 

in order to deal with distinct yet interdependent aspects such as physiol-

ogy, genetics, behaviour, and ecology. Only their combined efforts, each
focusing on a different aspect, can lead to a thorough grasp of the object
of study. In other words, a productive and established heuristic strategy
in the sciences is to differentiate between different levels of analysis. The
study of termites, for instance, draws on biochemical theory to investigate
how the animals communicate with the help of pheromones. At the same
time, the study of termites cannot be reduced to the level of biochemistry,
since questions about termite sociology cannot be adequately addressed
from the vantage point of biochemical theory alone; input from behavioural
studies and biosemiotics, for example, is required too. Simply accumulating
knowledge from different perspectives tends not to be enough, however.
Generally speaking, it is not enough to analyse the individual components
of a composite phenomenon; at some point active synthesis is required. In
Douglas Hofstadter’s famous example: ‘A traffic jam is just not on the level
of an individual car. It is a pattern composed of cars, a pattern that more-
over has a deep repercussion on the cars it is composed of’ (1987, p. 787).
The malfunctioning engine of an individual vehicle may well be responsible
for a particular traffic jam, but the phenomenon as such requires us to go
beyond the level of individual cars. Hofstadter’s example thus highlights the
need to differentiate between different levels of analysis while simultane-
ously keeping their mutual interdependence in mind.
The difference between cars and traffic jams also illustrates the fact
that the relationship between explanatory levels generally takes a hierar-
chical form, in which ‘each level depends on the existence of the level from
which it emerges’ (Paulson, 1991, p.  44). The laws of biochemistry, for
example, shape and delimit the chemical means by which termites com-
municate via pheromones. Those laws do not cease to exist when the sci-
entist moves to a different level of analysis. According to Patrick Hogan,
they ‘are conserved in the conceptual or explanatory movement from lower or
more basic levels to higher levels’ (2003, p. 202, emphasis in original). Put
differently, ‘the laws that govern elements on level one constrain the pos-
sibility of patterns that can emerge on level two’ (p. 203). Simultaneously,
level two transcends level one and thus requires a different set of research
strategies and theories. The study of chemical compounds, for instance,
cannot be comprehensively conducted on the subatomic level of particle
physics, as this level does not account for the emergent characteristics of
chemistry.

Emergent qualities, which are found in every corner of complexity in the uni-
verse, are qualities not included in, and generally not predictable from knowledge

S c i e n t i f i c C o n c e p t s i n L i t e r a r y S t u di e s  [ 23 ]
 

of, the qualities of the system in which they arise. The emergent qualities of the
simple chemical combination H2O, for example, are not found in the two gases
taken separately or mixed together. Similarly, complete physical and chemical
knowledge of the DNA molecule would not predict its function in reproduction.
Wilden, 1987, p. 1706

The phenomenon of emergent qualities is central to the interplay of explan-

atory levels, for ‘each level is defined by the emergence of some structure that is
not accounted for by laws at the lower level’ (Hogan, 2003, p. 203, emphasis
in original).
Although the examples so far have been primarily taken from physics,
chemistry, and biology, the phenomenon of emergence is not restricted to
those areas of research. In the context of this chapter, emergence is partic-
ularly interesting with regard to mental functions and social phenomena.
In the view of Nobel laureate Roger Sperry, mental experiences need to
be understood as emergent properties of brain processes which cannot be
fully captured by lower-​level theories:

The events of inner experience, as emergent properties of brain processes,

become themselves explanatory causal constructs in their own right, interacting
at their own level with their own laws and dynamics. … The qualitative, holistic
properties at all different levels become causally real in their own form and have
to be included in the causal account.
Sperry, 1982, p. 1226

In this context, Joseph Carroll’s provocative statement that ‘literature

is … a biological phenomenon’ (1995, p. 1) is, on the one hand, fully justified
in reminding us of the biological grounding of cognition and culture. The
physical, chemical, and neurobiological laws governing our brains and bod-
ies do in fact both enable and delimit the phenomenon of literature, includ-
ing its production, reception, and communication, in several respects. On
the other hand, Carroll’s statement is arguably misleading in implying that
the different facets of literature can be adequately approached from the
vantage point of biology alone, that they lack emergent properties which
require us to look beyond cells and nerves.7

6.  See also Paulson (1991, pp.  44–​50). For a critical introduction to the notion of
emergence, see also O’Connor and Wong (2012).
7.  On the notion and discussion of eliminative materialism which underlies Carroll’s
position, see Dietrich (2007, pp. 52–​59) and Ramsey (2008).

[ 24 ]  Literature through a Cognitive Lens

 

IMPLICATIONS FOR COGNITIVE LITERARY STUDIES: THE

PRINCIPLES OF COHERENCE, MODERATION, AND AUTONOMY

The hierarchy of explanatory levels outlined in the previous section is

obviously a conceptually simplified and heuristic model. Nevertheless, if
applied to cognitive literary studies, its inherent dialectics of hierarchical
constraints and emergent structure can be used to develop a number of
general guidelines for the intersection of literary study and cognitive sci-
ence. To begin with, the matrix suggests that cognitive research should be
conceptually divided into different explanatory levels. In this context, I pro-
pose to distinguish between at least three relevant hierarchical domains of
analysis: (1) brain structures and neurological processes, (2) mental states
and mental processes, and (3) literature as a social and cultural phenom-
enon. Again, this distinction is heuristic in nature; it is certainly possible
to make further differentiations or delineate the levels in a different way.
Nevertheless, a number of observations can be deduced from their hierar-
chical interplay.
As we have seen, explanatory levels are generally constrained by their
subordinate levels, which means that they should not postulate ideas in vio-
lation of these levels. Put differently, theories and concepts should satisfy
what could, loosely based on Gregory Currie (1997), be called the principle
of coherence. This means they should cohere with the tenets of established
scientific theorizing on those levels and ‘should not postulate any mecha-
nisms or causal pathways not sanctioned by that theorizing’ (pp. 63‒64).8
Consequently, scholars attempting to build bridges between the two cul-
tures need to possess a degree of expertise that allows them to assess
potential conceptual conflicts arising from the intersection of the levels of
brain, mind, and culture. This guideline sounds uncontroversial, but it has
tremendous implications in practice. The principle of coherence implicitly
requires the literary scholar working with scientific concepts to master the
theories that are being adapted to such an extent that he or she is famil-
iar not merely with their conception and application but also with their
criticism. Only in this way is it possible to differentiate between (a) empiri-
cally corroborated findings and speculative hypotheses, and (b) sufficient
and necessary conditions for a certain phenomenon—​distinctions of the

8.  In this context, see Gregory Currie’s discussion of the paradox of caring, in which
he argues that solutions to this paradox need to observe the constraint of coherence,
i.e. that they ‘should cohere with the best psychological theorizing’ and refrain from
postulating psychological mechanisms ‘not sanctioned by that theorizing’ (1997,
pp. 63–​64).

S c i e n t i f i c C o n c e p t s i n L i t e r a r y S t u di e s  [ 25 ]
 

utmost importance for any sensible cognitive approach to literature and

its processing.
Unfortunately, the need for such expertise constitutes one of the
basic challenges of interdisciplinary research in general.9 Work across
disciplinary borders ideally requires solid expertise in both fields. Yet
mastering expert knowledge in an additional discipline beside one’s own
is indeed a daunting task. Most traditionally trained literary scholars
therefore may find it difficult to acquire a solid footing in the cognitive
sciences.

Adequate scientific knowledge is simply outside the expertise of all but a

minority of literary scholars. The literary in literary study is simply outside
the expertise of all but a minority of cognitive scientists and evolutionary psy-
chologists. Few scholars will have time or inclination to learn the other field
sufficiently to challenge what is most scientific about the one or most literary
about the other.
Jackson, 2000, p. 340

The principle of coherence nevertheless demands that literary scholars

working with the sciences face up to this challenge of interdisciplinary
expertise. As adequate scientific knowledge nevertheless often remains
unrealistic, one possible solution for conducting interdisciplinary research
lies in the collaboration of scholars and scientists. Yet, ‘even collaboration
requires that both parties know quite a bit about both areas’ (Hogan, 2003,
p. 3). Collaborative work, in other words, does not entirely solve the prob-
lem of expertise. For this reason, I would like to propose a second heuristic
guideline: the principle of moderation.
According to this guideline, the transfer of theories and concepts from
lower to higher levels should be moderate in a twofold way. First, it should
be firmly restricted to established, well-​corroborated theories and con-
cepts. This means that conceptual transfer from, for example, neurology
or psychology should be limited to relatively reliable insights from those
disciplines and refrain from taking up highly controversial, speculative,
or experimental ideas that have yet to be tried and tested. Observing this
seemingly trivial suggestion may at first appear as a means of stifling cre-
ativity. But it enables the scholar to adhere to the principle of coherence
without having to acquire expert-​like scientific knowledge. It prevents us
from unintentionally misinterpreting or misapplying ideas from highly

9.  On the theory, practice, and history of interdisciplinarity in general, see Klein
(1990, 1996).

[ 26 ]  Literature through a Cognitive Lens

 

specialized scientific discourses which we in literary studies, even if we

do not realize it, may lack the training to understand, contextualize, and
evaluate. Secondly, scholars should ideally limit the use of scientific theo-
ries to cases in which new insight into the literary objects and phenomena
in question can really be gained (Mansour, 2009, p. 156). Adhering to this
suggestion helps to counter accusations that cognitive literary scholars
uncritically succumb to the lure of scientific theory and give preference to
scientific concepts much more because of their popularity and success in
their own disciplines than ‘from any likelihood that they will help us here’
(Midgley, 2006, p. 3). The principle of moderation can thus help to weaken
the frequently voiced criticism that ‘despite regular, enthusiastic claims for
radically new insights, the actual application of theories to texts has much
too often produced interpretations that are painfully obvious’ (Jackson,
2005, p. 528).10
Beyond the principles of coherence and moderation, the phenomenon of
emergent structure also draws attention to the conceptual and explanatory
gaps that can be found between different levels of investigation. Perhaps
the most important gap with respect to cognitive literary studies is the one
discussed in philosophy as the mind‒body problem. Although the mind is
widely seen today as being somehow ‘based’ on neuronal activity in the
brain, science is still unable to explain the emergence of consciousness or
mental functioning (Ryan, 2010, p.  473). While some see this merely as
an epistemological problem that may be solved one day, others maintain
that there is an unbridgeable ontological gulf separating consciousness and
matter:

Either way you look at it, there is an apparent gap that separates the mental
phenomena from solid matter. Consciousness does not seem to fit naturally into
the scientific framework that explains the physical universe. The difficulty of

10.  See also Vandaele and Brône (2009, p. 3). Ryan (2010) makes a related point in
her discussion of cognitive narratology. Taking the example of a brain-​imaging study by
Speer and colleagues (2009), she suggests that neurological findings often seem self-​evi-
dent to the narratologist: ‘Do we need an MRI to tell us that reading isolated words does
not require the same mental activity as reading a story? Are brain scans necessary to
make us realize that there is something in common between apprehending an image of
something—​be it verbal or visual representation—​and apprehending its referent? …
Instead of opening new perspectives on narrative cognition the experiments of Zack
and Speer confirm what common sense tells us’ (pp. 471–​472). For Ryan the study’s lack
of narratological relevance is symptomatic of a deeper problem: ‘current techniques of
brain imaging have not yet reached the necessary precision to tell narratologists some-
thing truly new and interesting concerning the cognitive foundations of narratives’
(p. 472).

S c i e n t i f i c C o n c e p t s i n L i t e r a r y S t u di e s  [ 27 ]
 

relating the phenomenal world to the very different world described in physics
is known as the explanatory gap.11
Dietrich, 2007, p. 13, emphasis in original

Even if bridging this gap is theoretically possible, the solution will most
likely have to go beyond developing brain scans with higher resolution.
It will probably require a new understanding of the relationship between
mind and matter, since mental phenomena, according to Benjamin Libet,
‘cannot a priori be described by any knowledge of physical events and
structures; and, conversely, physical events (including the neuronal ones in
the brain) cannot be described by knowledge of the accompanying mental
subjective events’ (2004, pp. 181‒182).12
The mind‒body problem is certainly the most widely discussed explana-
tory gap in cognitive science, but an equally important gulf opens up spe-
cifically in cognitive literary studies between the analysis of general mental
processes and the subjective reading of specific literary texts—​not to men-
tion the question of how individual acts of reception shape literature as
a cultural or political phenomenon. Again we find emergent structure on
those levels that scientific theory, for example in neuroscience or social
psychology, is currently incapable of adequately dealing with, partly due
to the staggering complexities involved in the reading of literature. In my
opinion, the existence of those gaps and the emerging idiosyncrasies found
on all explanatory levels suggest a third criterion for cognitive literary
studies: the principle of autonomy.
The idea of methodological and conceptual autonomy in literary stud-
ies derives from the notion of emergence and the related gap between the
explanatory potential of scientific approaches and the phenomena inves-
tigated by the humanities. While Thomas Gottschall, for example, wants
literary studies to ‘move closer to the sciences in theory, method, and gov-
erning ethos’ (2008, p. 3), I believe that the notion of a hierarchy of explana-
tory levels suggests that the most illuminating way of addressing literature
and culture is by working with conceptual means developed specifically
for these phenomena. During the development of adequate approaches,
scholars should by all means look beyond the borders of their discipline for
inspiration. Yet the incorporation of scientific concepts into literary anal-
ysis should ideally be free from false scientific pretence and avoid broad-​
sweeping reductionist claims that naïvely equate lower-​level phenomena
such as mirror neurons with higher-​level phenomena like empathy.

11.  The term ‘explanatory gap’ was originally coined by Joseph Levine (1983).
12.  Again, this is a hotly contested issue in the study of consciousness. For a debate,
see Dietrich (2007, pp. 63–​82) and Blackmore (2010).

[ 28 ]  Literature through a Cognitive Lens

 

Given that a literary text is not a naturally occurring phenomenon but

a cultural artefact and a work of art, the research methods developed for
its study can also differ from scientific procedures because literary schol-
arship may be interested in what could be called non-​scientific aspects
of reading. While cognitive analysis in the sciences by and large tends to
focus on illuminating ‘general patterns and models of comprehension’,
cognitive approaches to literature may, for example, remain interested in
individual works and their interpretation (Adler and Gross, 2002, p. 215).
Correspondingly, many scholars still try ‘to develop highly individualis-
tic, sophisticated readings’—​an interest not necessarily shared by experi-
mental psychologists who focus on automatic mental operations which
readers perform unconsciously (Ryan, 2010, p. 475). Whereas scientific
approaches thus generally aim at gaining conclusive insight into the
workings of the mind, the prime concern of many scholars in literary
studies still lies within the reading of individual texts. The study of litera-
ture is usually seen not as a ‘problem-​solving undertaking’ but rather as a
means of mapping the potential of art and literature and a way ‘to achieve
understanding, to assess context-​relatedness, to investigate meaning and
function, and to evaluate art and literature’ (Iser, 2006, p. 7).
The principle of autonomy reminds us that the study of literature
may differ significantly from many fields of research in the cognitive sci-
ences in terms of methodologies and the aims of research. Of course,
some voices strongly advocate a genuinely scientific approach to litera-
ture.13 But given the methodological idiosyncrasies of each explanatory
level, a literary scholar aiming at a thoroughly scientific approach would
necessarily have to meet the strict standards of science, which would
include solving ‘explicitly spelled-​out problems via explicit problem-​
solving strategies or methods’ (Schmidt, 2000, p.  621). However, as
Tony Jackson emphasizes, the moment we do not want to stick exclu-
sively to arguments and methods from the cognitive sciences, ‘we will
necessarily [have to] come back around to culture—​that is, language,
ideology, politics, history—​if we are to account for difference’ (2000,
p. 341). In any case, the principle of autonomy draws our attention to
the fact that various methods and research interests may exist and inter-
act on different levels. It thus invites us to reflect upon the aims of our
research and their compatibility with the methods and theories we are
recruiting from the sciences. Together with the principles of coherence
and moderation it can serve as a helpful guideline and an invitation for
meta-​theoretical reflection.

13.  For example, Gottschall (2008) and Slingerland (2008).

S c i e n t i f i c C o n c e p t s i n L i t e r a r y S t u di e s  [ 29 ]
 

CONCLUSION: THE MEETING OF LITERATURE AND SCIENCE

There are many good reasons for a literary scholar to be interested in the
cognitive sciences. After all, ‘the mind as such, and perception in particu-
lar’, as Monika Fludernik and Greta Olson point out, ‘have … been stock
features of all narrative enquiry since the days of Henry James’ (2011,
p. 8). There is no a priori objection to taking science as a source of infor-
mation and inspiration. In fact, I  would go one step further and hold
with Tony Jackson, quoted earlier, that our discipline can hardly afford to
ignore empirical investigations of mind and brain and the resulting claims
‘about the biology and psychology of reading, writing, and responding’
(2000, p. 340) if we want to keep up with the pulse of time. Whenever new
information about mental processes becomes available and sufficiently
persuasive, it is irrational to refrain from exploring its explanatory value
for literature and culture. Engaging with science can in my opinion also be
a valuable asset with regard to the standing of the humanities in interdis-
ciplinary dialogue—​if only because it enables the humanities scholar to
critically examine and question the scientist’s assertions.14
The contributions to the present volume testify to the exciting possibili-
ties of an illuminating and fruitful dialogue between science and literature.
The volume demonstrates that the particular appeal of cognitive literary
studies lies specifically in its interdisciplinary nature, which invites us to
think outside the boxes of our established disciplines. The meeting of the
two cultures not only has the potential to draw our attention to interest-
ing questions we have failed to consider so far but also asks us to review
basic theoretical tenets and to re-​examine familiar phenomena from new
methodological angles. But although the appeal of cognitive approaches
lies in the field’s interdisciplinary nature, its potential problems do too. As
this chapter has argued, attempts at bridging the two cultures always come
with conceptual, methodological, and practical challenges that must not be
ignored.
Drawing on the idea of a hierarchical matrix of explanatory levels, I have
tried to outline some of the most common pitfalls and faux pas in the adap-
tation of scientific theory to literary studies. Among other things, I have
drawn attention to the awkward blending of scientific rhetoric and poetic

14. An important function of a scientifically informed humanities, in my opin-

ion, is to critically examine seemingly scientific claims regarding such general topics
as society, culture, art, and human nature. The reason is that scientists, for example
evolutionary psychologists, ‘all too often … try to explain a higher-​level structure
(e.g., literature) in terms of a lower-​level structure (biology) without first acquiring a
detailed understanding of what that higher-​level structure is’ (Hogan, 2003, p. 213).

[ 30 ]  Literature through a Cognitive Lens

 

licence, as well as the danger of naïvely subscribing to reductionist explana-

tions that confuse necessary with sufficient conditions and ignore the real-
ity of explanatory gaps and the emergence of new meaningful structures
on different levels of inquiry. Subsequently, I  have proposed three basic
guidelines for the study of cognitive approaches to literature: the principle
of coherence, the principle of moderation, and the principle of autonomy.
The idea informing these heuristic guidelines was to provide a general
sense of methodological and conceptual orientation for cognitive literary
studies, as a response to some of the more valid criticism levelled at this
exciting field of study. I believe that in order to further advance cognitive
approaches within our discipline we need to take this criticism seriously.
In this spirit, the principles are not designed to restrict interdisciplinary
exchange but to render it more methodologically sound and thus more suc-
cessful. We need to continue engaging in meta-​theoretical reflections on
the methods, scope, and aims of cognitive approaches which form a neces-
sary prerequisite for the successful intersection of two such fundamentally
different research cultures as cognitive science and literary studies.
Personally, I  believe that the work of literary scholars should gener-
ally not ‘aspir[e]‌to the condition of the definitive, of a scientific theory’
(Waugh, 2006, p. 16) if it does not want to lose its intimate link to the lit-
erature it is interested in. Thus, the field of cognitive literary studies should
engage in dialogue with, and seek inspiration from, the cognitive sciences
without kowtowing to them on a conceptual level. Others have voiced
different opinions about the field. David Herman (2011), for example,
eloquently advocates forms of transdisciplinary research that go beyond
unilateral adaptations of ideas from other disciplines and instead aim to
establish new areas of inquiry together and on an equal footing with the
sciences. ‘In these increasingly difficult budgetary times’, he argues, schol-
ars can contribute to protect the humanities

by demonstrating the benefits of their research in broader contexts of inquiry.

And they can do this … by helping to identify transdisciplinary objects of inves-
tigation such as narrative, defining the role that humanities research can play
in articulating questions about those objects, and then developing strategies for
crossing disciplinary boundaries to address the questions that they themselves
have helped formulate.
2011, emphasis in original

Herman makes a convincing case for transdisciplinary research, and I can-

not think of any objection to the idea of advancing our field by means of
developing new questions and areas of investigation that demonstrate the

S c i e n t i f i c C o n c e p t s i n L i t e r a r y S t u di e s  [ 31 ]
 

benefits of humanities research in a broader context. Nevertheless, I won-

der whether this is all that literary scholarship in general and cognitive liter-
ary studies in particular should endeavour to do.
The notion of emergent structure and the related principle of auton-
omy indicate that each explanatory level poses questions that may not
necessarily bear on other levels of inquiry. Literature is worth studying
for a vast variety of reasons; not all of them will necessarily be of sci-
entific or transdisciplinary value. However, if unidirectional but meth-
odologically sound borrowing from the sciences enables us to improve
our ‘reflective understanding’ of a literary work and helps us ‘to increase
the associative complexity of our response to that work—​thereby, one
hopes, enriching our aesthetic experience’ (Hogan, 2008, p. 193)—​then
cognitive literary studies, in my opinion, does not require any further
justification.

REFERENCES

Adler, H., and Gross, S. (2002). Adjusting the frame: Comments on cognitivism and
literature. Poetics Today, 23(2), 195–​220.
Aldama, F. L. (2010). Introduction: The sciences and the humanities matter as one. In
F. L. Aldama (Ed.), Toward a cognitive theory of narrative acts (pp. 1–​9). Austin,
TX: University of Texas Press.
Bateson, G. (1972). Steps to an ecology of mind. Chicago, IL: University of Chicago Press.
Blackmore, S. (2010). Consciousness:  An introduction (2nd ed.). London:  Hodder
Education.
Burke, M., and Troscianko, E. (2012). Science and literary criticism. St John’s College
Oxford. Retrieved from http://​www.sjc.ox.ac.uk/​4073/​Science%20and%20
Literary%20Criticism%202012_​v4.pdf.download
Carroll, J. (1995). Evolution and literary theory. Columbia, MO:  University of
Missouri Press.
Cohen, P. (2010, 1 April). The next big thing in English: Knowing they know that you
know. New York Times. Retrieved from http://​/​www.nytimes.com/​2010/​04/​01/​
books/​01lit.html
Currie, G. (1997). The paradox of caring: Fiction and the philosophy of mind. In M.
Hjort and S. Laver (Eds.), Emotions and the arts (pp. 63–​92). New York: Oxford
University Press.
Dietrich, A. (2007). Introduction to consciousness: Neuroscience, cognitive science, and phi-
losophy. New York: Palgrave Macmillan.
Fludernik, M., and Olson, G. (2011). Introduction. In G. Olson (Ed.), Current trends in
narratology (pp. 1–​33). Berlin: de Gruyter.
Gottschall, J. (2008). Literature, science, and a new humanities. New  York:  Palgrave
Macmillan.
Hartner, M. (2012). Perspektivische Interaktion im Roman:  Kognition, Rezeption,
Interpretation [The interaction of perspectives in the novel:  Cognition, recep-
tion, interpretation]. Berlin: de Gruyter.

[ 32 ]  Literature through a Cognitive Lens

 

Herman, D. (2011). Narrative worldmaking across media and disciplines. American

Council of Learned Societies. Focus on Research Series. Retrieved from http://​
www.acls.org/​news/​5-​24-​11
Hofstadter, D. P. (1987). Metamagical themas: Questing for the essence of mind and pat-
tern. Harmondsworth: Penguin.
Hogan, P. C. (2003). Cognitive science, literature, and the arts. New York: Routledge.
Hogan, P. C. (2008). On interpretation: Meaning and inference in law, psychoanalysis, and
literature (2nd ed.). Athens, GA: University of Georgia Press.
Ibsch, E. (1990). The cognitive turn in narratology. Poetics Today, 11(2), 411–​418.
Iser, W. (2006). How to do theory. Malden, MA: Blackwell.
Jackson, T. E. (2000). Questioning interdisciplinarity: Cognitive science, evolutionary
psychology, and literary criticism. Poetics Today, 21(2), 319–​347.
Jackson, T. E. (2005). Explanation, interpretation, and close reading: The progress of
cognitive poetics. Poetics Today, 26(3), 519–​533.
Jaén, I., and Simon, J. J. (2012). An overview of recent developments in cognitive liter-
ary studies. In I. Jaén and J. J. Simon (Eds.), Cognitive literary studies: Current
themes and new directions (pp. 13–​32). Austin, TX: University of Texas Press.
Klein, J. T. (1990). Interdisciplinarity:  History, theory & practice. Detroit, MI:  Wayne
State University Press.
Klein, J. T. (1996). Crossing boundaries: Knowledge, disciplinarities, and interdisciplinari-
ties. Charlottesville, VA: University Press of Virginia.
Kuhn, T. S. (1962). The structure of scientific revolutions. Chicago, IL:  University of
Chicago Press.
Levine, J. (1983). Materialism and qualia:  The explanatory gap. Pacific Philosophical
Quarterly, 64, 354–​361.
Libet, B. (2004). Mind time: The temporal factor in consciousness. Cambridge, MA: Harvard
University Press.
Mansour, J. (2009). Chancen und Grenzen des Transfers kognitionspsychologischer
Annahmen und Konzepte in die Literaturwissenschaft: Das Beispiel der ‘Theory of
Mind’ [Potential and limits of the transfer of cognitive-​psychological assumptions
and concepts into literary studies: The example of ‘Theory of Mind’]. In Martin
Huber and Simone Winko (Eds.), Literatur und Kognition: Bestandsaufnahmen und
Perspektiven eines Arbeitsfeldes [Literature and cognition:  Surveys and perspec-
tives on a field of study] (pp. 155–​164). Paderborn: Mentis.
Midgley, M. (2006). Science and poetry. London: Routledge.
O’Connor, T., and Wong, H. Y. (2012). Emergent properties. In E. N. Zalta (Ed.), The
Stanford encyclopedia of philosophy (Spring 2012 ed.). Retrieved from http://​
plato.stanford.edu/​archives/​sum2013/​entries/​properties-​emergent/​
Palmer, A. (2010). Social minds in the novel. Columbus, OH: Ohio State University Press.
Paulson, W. (1991). Literature, complexity, interdisciplinarity. In N. K. Hayles (Ed.),
Chaos and order: Complex dynamics in literature and science (pp. 37–​53). Chicago,
IL: University of Chicago Press.
Ramsey, W. (2008). Eliminative materialism. In E. N. Zalta (Ed.), The Stanford ency-
clopedia of philosophy (Summer 2013 ed.). Retrieved from http://​plato.stanford.
edu/​archives/​sum2013/​entries/​materialism-​eliminative/​
Richardson, A. (2004). Studies in literature and cognition: A field map. In A. Richardson
and E. Spolsky (Eds.), The work of fiction: Cognition, culture, and complexity (pp.
1–​29). Aldershot: Ashgate.
Ryan, M.-​L . (2010). Narratology and cognitive science: A problematic relation. Style,
44(4), 469–​495.

S c i e n t i f i c C o n c e p t s i n L i t e r a r y S t u di e s  [ 33 ]
 

Schmidt, S. J. (2000). Interpretation:  The story does have an ending. Poetics Today,
21(4), 621–​632.
Schneider, R., and Hartner, M. (Eds.). (2012). Blending and the study of narra-
tive: Approaches and applications. Berlin: de Gruyter.
Sheenan, P. (2014). Continental drift: The clash between literary studies and cognitive
literary studies. In C. Danta and H. Groth (Eds.), Mindful aesthetics: Literature
and the science of mind (pp. 47–​58). New York: Bloomsbury.
Slingerland, E. (2008). What science offers the humanities: Integrating body and culture.
Cambridge: Cambridge University Press.
Snow, C. P. (1965). The two cultures and a second look: An expanded version of the two cul-
tures and the scientific revolution. Cambridge: Cambridge University Press.
Speer, N. K., Reynolds, J. R., Swallow, K. M., and Zacks, J. M. (2009). Reading sto-
ries activates neural representations of perceptual and motor experiences.
Psychological Science, 20(8), 989–​999.
Sperry, R. (1982). Some effects of disconnecting the cerebral hemispheres. Science,
217(4566), 1223–​1226.
Spolsky, E. (1993). Gaps in nature: Literary interpretation and the modular mind. Albany,
NY: State University of New York Press.
Stockwell, P. (2002). Cognitive poetics: An introduction. London: Routledge.
Tallis, R. (2008, 9 April). The neuroscience delusion. The Times Literary Supplement.
Retrieved from http://​www.the-​tls.co.uk/​tls/​reviews/​other_​categories/​arti-
cle758025.ec
Vandaele, J., and Brône, G. (2009). Cognitive poetics:  A  critical introduction. In
G. Brône and J. Vandaele (Eds.), Cognitive poetics: Goals, gains and gaps (pp. 1–​
29). Berlin: de Gruyter.
Waugh, P. (2006). Introduction:  Criticism, theory, and anti-​ theory. In P. Waugh
(Ed.), Literary theory and criticism: An Oxford guide (pp. 1–​33). Oxford:  Oxford
University Press.
Wilden, A. (1987). The rules are no game: The strategy of communication. London: Routledge
and Kegan Paul.
Zunshine, L. (2007). Why we read fiction. Columbus, OH: Ohio State University Press.

[ 34 ]  Literature through a Cognitive Lens

 

CHAPTER 2

Towards a ‘Natural’ Bond of Cognitive

and Affective Narratology
C AROLINE PIRLET AND ANDRE A S WIR AG

C urrent research on emotion has had only limited impact on narrative

theory. By and large, cognitive approaches to narrative have treated
emotion as a subsidiary effect of the mental processes governing story-
telling practices and have relegated the affective component of cognition
to a subordinate position vis-​à-​vis the supposedly rational, deliberate or
analytic operations of cognition—​whether temporally, causally, and/​or
investigatively (Sternberg, 2003, pp. 313, 382; Miall, 2006; Hogan, 2011).
Understanding narrative, however, just like understanding in general, is
never a purely rational act. A  cognitive account of story comprehension
that omits the affective dimension of narrativization, i.e. the act of impos-
ing a narrative structure on experience or discourse (Fludernik, 1996,
p. 14), is consequently an inadequate representation of how we understand
stories and what constitutes them.
As Peter Stockwell (2002, p.  171) notes with a view to literary texts,
emotions—​regarded as prompted by textual cues that are constitutive to
any narrative—​are easily amenable to the study of literature through the
field of research known as cognitive poetics. This, we argue, is also true
for cognitive narratology, which incorporates the study of both factual and
fictional texts. In cognitive narratology and cognitive poetics, respectively,
two recent outstanding contributions to this research strand are Patrick
Colm Hogan’s Affective Narratology (2011) and Michael Burke’s Literary
Reading, Cognition and Emotion (2011). While Hogan persuasively links
 

emotion with cross-​culturally recurring story patterns and our under-

standing of the temporal and spatial dimension of story structure, Burke
systematically charts what happens in the embodied mind of readers as
a dynamic, bi-​directional process of bottom-​up and top-​down affective-​
cognitive inputs.
The present chapter hopes to advance this line of research on the affective
dimension of reader response and move towards creating synergetic effects
between cognitive poetics and narratology to arrive at a more holistic under-
standing of how we understand stories. The argument is therefore divided
into three larger sections. First, echoing Stockwell’s general conviction that
affect is integral to fictional narrative, a theory of ‘literary emotions’ (under-
stood as those emotions typically evoked in the act of reading works of fiction)
is proposed in order to explain the prevalence of affect in literary reading per
se, as well as to account for similarities and idiosyncrasies in readers’ affec-
tive responses to fiction. Second, following this theoretical exposition, a sub-
sequent close reading of James Joyce’s short story ‘Eveline’ (1914) illustrates
the purported salience of emotion in decision-​making processes of read-
ers and fictional characters alike. Third, broadening out the narratological
perspective, these findings and their implications for the larger agenda of
developing an affect-​based narratology are explored within the framework
of a ‘natural’ narratology (Fludernik, 1996).
For the purposes of this discussion, the term ‘broad’ cognition will des-
ignate ‘any kind of mental operation or structure that can be studied in
precise terms’ (Lakoff and Johnson, 1999, p. 11), a cognitive umbrella that
includes a putatively rationalistic component (henceforth ‘narrow’ cogni-
tion) accountable for deliberate reasoning tasks, such as memory, atten-
tion, problem-​solving, and planning (see also the notion of ‘system 1’ in
Kahneman, 2011). Broad cognition also includes affect-​based operations,
which are argued to emerge from a small number of basic emotions, each
associated with distinct physiological expressions and evolutionarily adap-
tive values (Ekman, 1992)  that dynamically shape and complement the
purposes of all ‘narrow’ cognition (Damasio, 1994, 2000; Prinz, 2004).

A THEORY OF LITERARY EMOTIONS

Generally speaking, ‘emotions are part of a solution to problems of orga-

nizing knowledge and action in a world that is imperfectly known and in
which we have limited resources’ (Oatley, 1992, p. 3). To that end, emotions
serve a twofold function, both within and between organisms (see also
Hogan, 2003, pp. 141‒144). Within the organism, one basic way of thinking

[ 36 ]  Literature through a Cognitive Lens

 

about emotions is as information-​selection devices. As such, they structure

and coordinate our perceptual input by arranging and prioritizing relevant
information. As selective constraints on input, emotions initially focus
our attention on subjects important to our goals, wants, and interests (de
Soussa, 1987; Oatley, 1992, pp. 19, 98). Then, reinforced by physiological
changes, they move us towards action (Elster, 1999, pp. 60‒61, 281‒283;
Frijda, 2007). Despite the habitual discursive distinction between intellect
and affect, therefore, emotions can be considered rational in a narrower
sense, since they constitute an effective response in a given set of circum-
stances by supplying information about reasonable action. Claiming that
emotions are vital for intelligent action means adopting what Dylan Evans
calls ‘the positive view of emotion’ (2002, p. 33). Findings from evolution-
ary theory further substantiate the idea that ‘the benefits of having emo-
tions outweigh the drawbacks’ (p. 35; see also Damasio, 2000, pp. 41‒42).
The neuroscientist Antonio Damasio (1994), who has studied the role of
emotion in decision-​making and other tasks commonly considered ‘purely
rational’, goes so far as to conclude that, in matters of social and personal
importance, good decisions require an emotional component (p.  192).
Consequently, in addition to adopting rational strategies such as cost-​
benefit analyses, people appear to deliberate, narrow down, and choose
from a range of behavioural options through an initial emotional assess-
ment of envisioned outcomes as either beneficial or harmful. In simpler
terms, in order to know what to do and what to think about, we habitually
resort to how we feel about an issue rather than what we know about it
(Kahneman, 2011, pp. 19‒30, 137‒145).
Apart from their internal directive function, emotions serve to commu-
nicate intentions to others. As social beings, we often depend on collabo-
ration and mutual rapport to achieve our goals, a task greatly facilitated
by the communicative function of emotions. In this manner, people will
generally respond affectively to manifestations of emotions in others in
either a ‘parallel’ or a ‘complementary’ manner (Hogan, 2003, p. 143). An
empathic response (i.e. I feel what you feel) would be an example of the
former, whereas a sympathetic response (i.e. I  see your pain and conse-
quently pity you) is an instance of the latter. Crucially, because emotions
are ways to express our intentions and attitudes, people frequently assess
and morally judge one another by their emotional responses. In a nutshell,
emotions are tools for defining and understanding ourselves and others.
But how do these functions of emotion relate to our understanding of fic-
tional narrative?
Emotions, thus described, can be paraphrased as a form of engagement
with our internal and external environment—​real or fictional. The common

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 37 ]
 

phenomena of readers feeling for (and suffering with) fictional characters,

becoming invested in their fate (i.e. the story outcome), or developing read-
ing preferences for a particular genre based on the specific mood or emotion
it typically evokes—​these all attest to the affective dimension of narrative
engagement. But how can these characteristic forms of engagement with
narrative be accommodated within a narratological framework? What func-
tional or operative contribution do they make to the reading practice?
First, by way of their directive function, affective responses draw our
attention to certain facets of the narrative that strike us as relevant, yet
might have escaped us in a more intellectual engagement with the text.
Emotions, in other words, ‘alert us to important aspects of the story such
as plot, character, setting, and point of view’ (Robinson, 2005, p. 107; see
also Miall, 2006, p. 53). By selecting and establishing a hierarchy among
the given pieces of information, emotions ipso facto shape readers’ under-
standing of what the text is about. Secondly, regarding the external com-
municative function of emotion, readers will assess fictional characters and
their actions based both on a ‘narrow’ cognitive (i.e. analytic and dispas-
sionate) evaluation and on affective criteria. In this context, empathic and
sympathetic responses towards story characters represent means of relat-
ing to the emotional experiences of fictional entities. By exploiting their
understanding of how emotion and individual motivation and agency are
interlinked in real life (knowledge gained via introspection or observations
of others), readers are equipped to apply emotion-​based heuristics to ‘fill
in the gaps’ (Iser, 1978, pp. 170‒179) in characters’ portrayals or assumed
mindsets; they are, in other words, able to make informed (although idio-
syncratic) inferences about what motivates fictional characters to act in
the way they do in particular contexts and circumstances. In short, we
understand characters and the situations in which they find themselves
not merely rationally but also emotionally. To reiterate this in narratologi-
cal terms, emotional engagement with narrative is thus characterized as a
means of directing readers’ attention to story elements that correspond to
a field of current affective interest (internal function) as well as providing
an affect-​derived scaffold that facilitates reader inferences about the fic-
tional characters’ motivation or agency (external function).
Beyond these general observations, acquiring a more differentiated
understanding of how readers’ affective responses to literature are grounded
in the way emotions function for readers in real life (i.e. outside the realm
of reading) requires that we delineate the nature and structure of emotions
in more detail. Emotion, for this purpose, can be more narrowly described
as a ‘process that unfolds as a situation is appraised and reappraised, and
as continuous feedback occurs’ between affective and cognitive-​rational

[ 38 ]  Literature through a Cognitive Lens

 

assessments of a given state of affairs (Robinson, 2005, p.  76; see also
Ekman, 2003, p. 13). In this manner, the unfolding of an emotion consists
of (1) an initial heuristic of affective appraisal of the situation that focuses
attention on its significance for the organism’s immediate wants, goals,
and interests; which gives rise to (2) physiological and behavioural changes;
which are then subject to (3) a further, more discriminating, ‘narrow’ cogni-
tive evaluation or monitoring of the situation (Robinson, 2005, pp. 3, 59).
The different components of the emotion process are interconnected in
various ways: pragmatic cognitive monitoring may (dis)confirm the initial
affective appraisal; physiological and behavioural changes may alter the
environment, thereby requiring an emotional reassessment of the situ-
ation; and so on (pp.  93, 145‒146). Most importantly therein, the emo-
tion process is set in motion by a fast and automatic (or ‘quick and dirty’;
LeDoux, 1996, p. 164) affective appraisal of the baseline scenario, a routine
that ‘evaluates in a rough and ready way the personal significance of some-
thing in the internal or external environment’ (Robinson, 2005, p. 62). The
subsequent ‘narrow’ cognitive monitoring of the initial affective appraisal,
on the other hand, modifies the responses, changes the focus, and moder-
ates the behaviour. While deliberate cognition is thus a fixed component
of the emotion-​regulation process, pragmatic reflection is habitually post-
poned to follow an initial spontaneous, affective evaluation (Kahneman,
2011, pp.  103‒104, 364‒367). Humans (and other organisms) therefore
possess a highly efficient instinctive ‘affect heuristic’ that precedes the
rational component of cognition in a regulatory routine that appears to
have been beneficial to the survival of the species.
Foregrounding the role of affective appraisal in the constitution of emo-
tions1 entails a change in focus when it comes to story comprehension that
offers a number of advantages for a cognitive approach to narrative. First,
and significantly, automatic, non-​rationalistic assessments are thought to
be subject to individual and cultural socialization, i.e. they can be con-
ditioned or learned (Damasio, 1994; Ekman, 2003). An appraisal-​based
psychological foundation for responses to literature thus accommodates
the fact that works of art and their recipients are always embedded in cul-
tural contexts which shape the patterns of significance among a work’s
properties as well as the patterns of readers’ responses. Additionally, tak-
ing into account that spontaneous affective appraisal typically precedes
‘narrow’ cognitive judgement, the ‘fictionality problem’ (Hogan, 2003,
p.  167) will prove to be less problematic. Seeing that ‘narrow’ cognitive

1.  See also Solomon (1976/1993), and, for a critical review of judgement theories of
emotion, Robinson (2005, pp. 8–​16).

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 39 ]
 

monitoring feeds back on affective components only in the subsequent

stages of emotion-​regulation, the post-​hoc recognition that ‘it is a story
we are engaged with and that there is no appropriate action to take’ will
neither undo nor reverse the initial affective appraisal. Emotional stimuli,
moreover, do not need to derive from the factual realm (or extra-​text),
but can equally be hypothetical or imagined. As Jenefer Robinson sug-
gests, ‘more complex cases of emotion in human beings might involve
affective responses not to a perception but to a thought or belief’ (2005,
p.  59). Presupposing, furthermore, what Jerrold Levinson (1997, p.  24)
calls an ‘anti-​judgementalist stance’—​that is, challenging the proposition
that ‘emotions for objects logically presuppose beliefs in the existence and
features of those objects’ (p. 23)—​effectively provides a potent (although
partial) answer to the paradox of fiction.
Henceforth, granting emotional responses to literature the onto-
logical status of ‘real’ emotions is no longer a problematic proposition
(contrast this with the notion of ‘quasi-​emotions’, i.e. emotions expe-
rienced through second-​order, make-​believe knowledge; Walton, 1990,
pp.  195‒204). As regards their intensity, however, (literary) emotions
depend on the power of the stimulus, which, in this case, is linked to
the cognitive activity of reading. But how do we account for the numer-
ous idiosyncrasies in readers’ affective responses to the same literary
work (i.e. to an identical matrix of plot design, character depiction, nar-
rative perspective, and so on)? In the case of literary reading—​and in
blatant contrast to real-​life events—​the provision of information in any
story (including the pre-​selected framing of events through narratorial
perspective or comment) is exactly equivalent for any reader; patently,
however, readers’ emotional evaluation differs not only in degree but
also in kind (joy, anger, disgust, etc.) for any given literary text. In sum,
therefore, how do we approach the ‘differing emotions problem’ (Hogan,
2003, p.  185) for an identical literary artefact? To start with, because
interpretations are the result of ‘narrow’ cognitive monitoring of our
non-​cognitive appraisals, and these in turn depend on differing goals,
desires, and interests, there is likely to be disagreement in the resulting
‘broad’ cognitive overview of plot, character, and theme. In line with this
argument, experimental research by Dolf Zillmann (1995) has demon-
strated that (deliberate) moral considerations play a significant role in
justifying, conceding, and motivating discordant affective responses to
the emotional experience of others. In addition, differences in readers’
emotional susceptibility to a literary work can be partially explained by
the closeness or distance of the events and the situations of the charac-
ters to readers’ own lives (Myyry and Helkama, 2007).

[ 40 ]  Literature through a Cognitive Lens

 

Furthermore, what exactly is being evaluated and how is not only con-
tingent on the personal repertoire and horizons of readers but also hinges
on their ‘embedding’ in particular cultures. Along the lines of social con-
structionism, contexts determine emotional life and these evidently vary
significantly: cultures differ in the value they attach to specific emotions,
in the elicitors and language of emotion, in the rules of emotional display
and the complexity of emotional experience (Oatley, Keltner, and Jenkins,
2006, pp.  57‒80). Discursive psychologists have introduced the concept
of ‘emotionology’ as an umbrella term spanning the collective emotional
standards of a given culture (Stearns and Stearns, 1985). As David Herman
(2007b) notes, narrative discourse and story provide insights into the
emotional standards from which they emanate. At the same time, stories
can shape an emotionology itself, and ‘constitute a primary instrument
for adjusting systems of emotion terms and concepts to lived experience’
(pp. 324‒325). Like authors during production, readers will draw on their
respective culture-​specific emotionologies for the construction of narrative
during reception. Divergence in idiosyncratic as well as cultural emotional
standards can consequently complicate or even prevent readers’ access to
a literary text.
Notwithstanding intercultural or individual variation, there are a sig-
nificant number of recurring cross-​cultural patterns which render the (fic-
tional) other’s emotions intelligible to observers (Hogan, 2011). Although
anthropological research documents that every culture has its own emo-
tional climate, we largely share the same basic emotional repertoire; our
common emotional heritage, in evolutionary terms, is deeper than the cul-
tural differences that set us apart (Evans, 2002). Accordingly, there are a
number of basic (as opposed to complex) emotions that anyone from any
culture is capable of experiencing. Usually identified by means of uniquely
distinguishable physiological symptoms (including, e.g. facial expressions;
Ekman, 1992, 2003), these basic emotions are essential to our everyday
responses to fundamental experiences and life tasks (LeDoux, 1996, p. 40).
Although the specific inventories of basic emotions vary among emotion
theorists (Elster, 1999, p. 61), identifying a conclusive set of basic emotions
is an empirical, not a theoretical issue, and will not affect their usefulness in
cognitive narratology. What is important to note, therefore, is that a high
degree of ‘similarity … makes cross-​cultural relations possible and that
allows for art and literature, music and film, to cross frontiers’ (Damasio,
2000, p.  53) between readers who live in different cultures, whether
defined in synchronic (concurrent) or diachronic (historical) terms. For the
sake of illustration, think of how some contemporary readers still empathi-
cally re-​enact the emotions of 18th-​century heroines or shake their fists at

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 41 ]
 

Renaissance villains. Pace constructionism, there seems to be something

‘real’ that can be shared across cultures, historical divides, and the realm
of readerly subjectivity alike. While individual life experience and moral
predispositions may therefore account for differing emotional assessments
among readers, basic emotion theory effectively provides an answer to the
‘shared emotions problem’ (Hogan, 2003, p. 167), or the question of why
readers habitually experience an overlap in their affective responses to the
same literary text.

AFFECTIVELY READING (IN)ACTION

IN JAMES JOYCE’S ‘EVELINE’

We have argued that emotions in response to a literary text represent a

form of access to a story in their own right. Emotional responses shape
all cognitive activity, including that involved in narrativization; emotions,
moreover, underlie our understanding of self and (fictional) others, as
well as our understanding and evaluation of real-​life and fictional envi-
ronments. The following close reading of James Joyce’s ‘Eveline’ focuses
on the eponymous heroine and shows her to be represented at the nexus
of action, motivation, and emotion. We will illustrate how literary emo-
tions experienced by readers vis-​à-​vis the text (in the manner outlined
earlier) can complement an overly narrow intellectual response to the
story in varying narrativizations that may be carried out by readers. In
this fashion, we will illustrate both a sympathetic (complementary) and an
empathic (parallel) affective response to Eveline’s struggle as they extend
and alter a ‘narrow’ cognitive assessment of the heroine’s failure to pursue
individual freedom.
The story itself is set in early 20th-​century Ireland. The reader initially
finds the protagonist gazing out the window of her room, deeply absorbed
in contemplation. Although still adolescent (just ‘over nineteen’; Joyce,
1914/​2000, p. 30), the young woman appears physically and mentally ‘tired’
(p. 29) beyond her age. The reader learns that Eveline’s childhood ended ‘a
long time ago’ (p. 29) and that she leads a current life of hard work ‘both
in the house and at the business’ (p. 30). While Eveline is reflecting on her
past and present situation, her body language (sitting with ‘her head …
against the window curtains’, p. 29) bespeaks her weary and pensive state
of mind. The window, pointing to the street below, marks the threshold
between a sheltering domestic space and an erratic, unstable outside world;
while the room is filled with symbols of stasis and recollection (‘dust’, p. 29;
a ‘yellowing photograph’, a ‘broken harmonium’, p. 30), the outside world

[ 42 ]  Literature through a Cognitive Lens

 

appears in an unsteady, continual flux (‘Everything changes. Now she was

going to go away like the others’, p. 29).
Stylistically, Eveline’s anxiety about change and her attachment to the
past are emphasized by the exclamatory ‘Home!’ (p. 29), which introduces
the description of the family’s living room in free indirect discourse, and
by the close repetition of ‘familiar objects’ (pp.  29, 30)  that appear dear
to her. The heroine’s characteristic passivity and defensiveness are fore-
grounded linguistically through the use of a future-​directed verb phrase (to
be + infinitive) that conveys a sense of obligation on the subject: ‘she was
to go away … , to be his wife and to live with him’ (p. 31). Her arguments
for maintaining the status quo in private life revolve around the finality
of a departure (‘Perhaps she would never see again those familiar objects’,
p.  30), the gravity of imminent change (‘She had never dreamt of being
divided [from those objects]’, p. 30), and her discomfort at being judged by
others (‘What would they say of her in the Stores’, ‘Say she was a fool, per-
haps’, p. 30). The reader learns that, since the death of her mother, she has
taken on stereotypical roles of both a maternal/​domestic and a paternal/​
financial type to keep the family together. Eveline attends to everyone but
herself while her reflections are tinged by a nagging sense of familial duty.
Eveline, the reader surmises, has come to define herself in terms of
paternal expectations imposed on her, an image she has embraced and
internalized as part of an identity-​sustaining sense of self. While con-
templating her options, she remains obedient to and considerate of her
parents’ desires:  ‘Her father was becoming old lately, she noticed; he
would miss her’; ‘the promise to her mother, her promise to keep the
home together as long as she could’ (p. 32). Her habitual prioritization of
others seems to have thwarted her ability to act for herself; when faced
with the prospect of freeing herself from the constraints of a burden-
some life, internalized obligations weigh heavily. In her mind, the tire-
some daily drudges, although unwanted, have come to signify security.
Although she rightly assesses her dead mother’s life as a ‘commonplace
sacrifice closing in final craziness’ (p. 33), and momentarily grows deter-
mined to avoid her mother’s fate (‘she wanted to live’, ‘She had the right
to happiness’, p. 33), Eveline is eventually immobilized by indecision and
inertia: ‘Her time was running out but she continued to sit by the win-
dow, leaning her head against the window curtain’ (p. 32).
In a final climactic scene, the heroine is found at the docks of Dublin,
standing ‘among the swaying crowd in the station’ (p. 33), on the point of
eloping with her lover. Eveline’s conflict of interests comes to a head when,
now deprived of her familiar surroundings, she is overcome by the scope
and finality of her decision, which she consequently questions: ‘If she went,

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 43 ]
 

to-​morrow she would be on the sea with Frank, steaming towards Buenos
Ayres [sic]. The passage had been booked. Could she still draw back after all
he had done for her?’ (pp. 33‒34, emphases added). By now, Eveline is in
the grip of her anxiety, which manifests itself in her physical and emotional
distress: ‘All of the seas of the world tumbled about her heart’, ‘nausea in
her body’ (p. 34). When Frank urges her to ‘Come!’, the entreaty is affec-
tively assessed as a threat, not as tenderness. Eveline responds with an
instinct to flight. In a ‘frenzy’ (p.  34), her strenuous opposition consists
in a—​now literal—​clinging to the present (‘She gripped with both hands
at the iron railing’, p. 34), a resolution reflected in a palilogical exclama-
tion: ‘No! No! No! It was impossible’ (p. 34). Eventually, in the final stage
of her distress, the accustomed routine of emotional stasis sets in: ‘She set
her white face to him, passive, like a helpless animal. Her eyes gave him no
sign of love or farewell or recognition’ (p. 34). Eveline has finally slipped
back into a state of numbness and inaction, fittingly signalling her immi-
nent return to paralysis in life.
Eveline appears to epitomize the plight of many early 20th-​century
women in Irish society, and ad-​hoc rationalizations for her predicament
abound:  the reader may attribute her inability to liberate herself to the
restrictiveness of social prospects designated to women like Eveline in a
patriarchal Irish-​Catholic culture (i.e. Eveline as a victim of confining soci-
etal roles); he or she may relate it to an individual character flaw (i.e. Eveline
as a narrow-​minded coward). Or else, in a synthesis of these ideas, her dire
living conditions might have prevented the heroine from being receptive
to new experiences and paths to happiness. Whichever intuitive rationale
is favoured, Eveline longs to escape, yet her reliance on routine and famil-
iarity overrides these impulses. In facing and rejecting a life-​altering deci-
sion, and in her inability to seize the chance of finding happiness, Eveline
emerges as a tragic heroine.
Cognitively speaking, retracing the pathway of reasoning that underlies
the reading of ‘Eveline as a coward’, readers may attempt to understand her
plight intellectually, which will result in a ‘narrow’ cognitive judgement of
the basic situation she finds herself in as constricting and painful. A further
pragmatic comparison of Eveline’s present hardship with the possibility of
a blissful life abroad with Frank, who she knows is willing to provide for
her, protect her, and love her (p. 33), will lead readers to rationally assess
Eveline’s character as fearful and weak. Her indecision and eventual refusal
to leave might, upon dispassionate reflection, reveal her to be nothing
short of foolish. Without considering the heroine’s emotions, therefore, the
reader cannot even comprehend why there should be an internal conflict in

[ 44 ]  Literature through a Cognitive Lens

 

the first place. Yet, as Damasio has shown, when we anticipate, weigh, and
assess our arguments regarding the decisions we are about to take, we do so
not least emotionally. This principle applies equally to characters (Eveline’s
choice) and, in the present context, to readers in the process of narrativiza-
tion. By emotionally engaging with the story, readers respond in a manner
at once complementary and analogous to Eveline’s predicament. Therefore,
in contrast to a purely pragmatic rendering, if the story induces the literary
emotions of sadness, pity, and fear for Eveline through the intermediary
of sympathy, this affective response might provoke or reinforce a reader’s
evaluation of the heroine as an undeserving target of trauma or psychologi-
cal abuse (i.e. Eveline as victim of societal expectations). By the same token,
readers may share empathically in the protagonist’s fears, which represent
an equally integral part of their general understanding of the difficulty she
faces. As a result, empathic readers will criticize her refusal to leave less
harshly, since these readers’ subjective re-​experiencing of Eveline’s plight
might suggest a virtual incapacity to break free from such paralysis—​even
if, for the heroine’s sake, we would pragmatically wish for her to escape. If
the reader empathizes with Eveline’s fear, her anxiety becomes intelligible.
A more empathic or sympathetic reader will consequently place greater
weight on this aspect than a reader who engages with her situation in a
more cerebral fashion and thus remains at a distance.2 Literary emotions, in
brief, constitute an important dimension of readers’ understanding of the
conflict at hand. As Eveline is immobilized and held prisoner by her emo-
tions, readers affectively, by way of the two described mechanisms (parallel
and complementary), relate to her experience of what it is like to lose con-
trol at a decisive moment of crisis.
As we have suggested, the reader brings his or her real-​world emotional
knowledge and experience to bear in interpreting Eveline and her situation.
The analysis of the story illustrates one of the ways in which readers’ affec-
tive engagement with a character may plausibly change their assessment
of the fictional other, and thus the way in which the story is narrativized.
In this manner, the transition in analytic focus from a cerebral-​cognitive
to an emotion-​based story reading (exploiting sympathy and empathy
alike) reveals how literary interpretations may not only be enriched, but

2.  This is evidently not to suggest that emotional involvement is a purely optional
or volitional activity in any given reader; like the vast majority of character traits in
personality psychology, emotional sensitivity can be modelled to vary across the read-
ership according to a Gaussian ‘bell curve’ or normal distribution (Howe, 2013, p. 65;
on the Gaussian distribution, see e.g. Reis and Judd, 2010, pp. 395–​397).

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 45 ]
 

be significantly altered by the incorporation of affective judgement. As

Herman suggests, to

take into account the cognitive and emotional states and processes of the charac-
ters as they act and interact in the storyworld … must be construed as integral
to the core events or ‘gist’ of the narrative, not as optional or peripheral ele-
ments that can be safely omitted.
2007a, p. 247, emphasis added

The final ‘broad’ cognitive assessment of Eveline’s inability to act is

therefore argued to shift from a rationalistic critique of the heroine as
apathetic to an affect-​derived condemnation of societal context in grad-
ual proportion to the sympathetic and empathic capacities of a given
reader.
The suggested claim about (actual) readers’ responses to the anal-
ysed story—​that is, the emergence of a positive correlation of read-
ers’ empathic aptitude and appreciation towards Eveline—​represents a
thesis that, in a stricter scientific sense, could plausibly be investigated
through adequate empirical procedures. In this manner, the psychomet-
ric test scores of readers on, for example, the Mayer-​Salovey-​Caruso
emotional intelligence scale (Mayer, Salovey, Caruso, and Sitarenios,
2003) might fruitfully be compared to a Likert-​type reader approval rat-
ing of Eveline’s conduct after reading (in psychological parlance, assess-
ing positive and negative valence of readers’ emotions). In the ideal case,
this set-​up might reveal a significant positive correlation between the
readers’ empathic capacities and their support for the heroine’s ultimate
(in)decision.
Moving beyond the current analysis, the suggested impact of affect-​
based cognitive operations on readers’ literary perception and evaluation
of story characters appears to entail implications for the larger enterprise
of narratological theory formation. The final section therefore discusses
the locus of emotions within an embodied reading process—​ that is,
within the dynamics of the mind‒body continuum on the level of reader
reception—​and the prospect of jointly theorizing ‘narrow’ cognition and
emotion within a unified narratological framework. More specifically, it
appears that the interplay of cognition and emotion in reader narrativiza-
tion supports Monika Fludernik’s (1996, 2003) conception of an (embod-
ied) reader consciousness at work in the reception of the analysed tale. It
is this notion of embodied consciousness, which we regard as central to
how emotional contours are ascribed to narrative throughout the reading
process, to which we will now turn.

[ 46 ]  Literature through a Cognitive Lens

 

EMOTION AND ‘NATURAL’ NARRATOLOGY

As a persistent trend, cognitive literary scholars faced with the chal-

lenge of theorizing the ‘mind-​relevant dimensions of storytelling prac-
tices’ (Herman, 2009, p. 30) are increasingly prepared to incorporate
into their own methodologies the insights emerging from cognitive sci-
ence that reject Cartesian dualism, i.e. the strict separation of (mate-
rial) body and (immaterial) mind (e.g. Turner, 1991; McConachie and
Hart, 2006; Caracciolo, 2012, 2014; Kukkonen, 2013; Kukkonen and
Caracciolo, 2014; Troscianko, 2014). As Margaret Boden puts it in her
monumental study of the history of ideas of cognition, cognitive sci-
ence now ‘covers all aspects of mind and behaviour’ (2006, p. 9). The
increasing stress on the inseparability of cognition and body in cog-
nitive science has brought forth ‘situated approaches to cognition’
(Cantwell Smith, 1999, p. 769) which consider the subjective reality of
readers a manifold mental, embodied, and environmentally embedded
phenomenon; the life of mind is seen as a continuous blend of men-
tal and physical elements, arising from an individual mind located in a
physical body, situated within a real physical environment. These situ-
ated approaches to cognition call for ‘“thick descriptions” of real people
acting in real-​life situations’ (p. 770), a task that is arguably a strong-
hold of narrative fiction. Literature—​and narrative as a cultural tech-
nique more generally—​with its potential for conveying richly embodied
as well as embedded experience is therefore able to represent situated
cognition in action.3
Reception-​orientated cognitive approaches to narrative, in turn, have
taken heed of readers’ minds as embodied and of their narrativization of
stories as an interplay of cognition, body, and emotion. One step towards
such a development is to incorporate emotions into existing narrato-
logical approaches (for one such attempt, see Troscianko, 2014, ch. 4).
Although emotions are marginalized in its original design, Fludernik’s
‘natural’ narratology (1996) appears to provide a further framework well
suited to accommodate readers’ affective responses to literature. By no
means exhaustive, the arguments put forward in what follows in favour
of a ‘natural’ bond between cognition and emotion in narratology will
centre, first, on the locus of emotions within the current conception
of ‘natural’ narratology and, secondly, on how the framework might

3.  This claim is reminiscent of Käte Hamburger’s (1968, p.  67) assertion that lit-
erature is the sole discourse with the potential for conveying the ‘I-​Origo’ of personal
experience.

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 47 ]
 

fruitfully be extended by ascribing emotions to the intrinsic core of nar-

rativity, namely ‘experientiality’ (as defined later).
The general framework of natural narratology is a cognitive constructiv-
ist one (Fludernik, 1996, pp. xii, 12). Narrativity, consequently, ‘is not a
quality adhering to a text, but rather an attribute imposed on the text by
the reader who interprets the text as narrative, thus narrativizing the text’
(Fludernik, 2003, p. 244). By stressing the constructivist aspects of nar-
rativity, natural narratology essentially foregrounds the reader and focuses
on the cognitive mechanisms underlying readers’ construction and inter-
pretation of narrative. Drawing on insights from cognitive linguistics and
schema (or frame) theory, natural narratology is predicated on the reader’s
pre-​existing cognitive parameters, and based on the assumption that read-
ers rely on the same frames for the construction of meanings to interpret
(fictional) texts and (factual) real-​life experience alike (1996, p. 12; see also
Gerrig, 1993). The term ‘natural’ in Towards a ‘Natural’ Narratology there-
fore corresponds to the human, i.e. it relates to narratology’s ‘anchorings
in human everyday experience’ (1996, p.  19) and to the ‘cognitive frames
[constitutive of prototypical human experience] by means of which texts
are interpreted’ (p. 12, emphasis in original).
The appeal of a reader’s recourse to established frames of knowledge
lies in the fact that this way of organizing knowledge is psychologically
credible, highly dynamic, and draws on readers’ individual real-​life expe-
rience. Yet the experience of narrative is more than a mechanical imple-
mentation of pre-​existing cognitive and affective mechanisms. Stories
feed back into how we perceive these contexts; narrative, by providing
aesthetically modelled experience, will engage readers with the lives
of (fictional) characters in a process that drives the formation of new
behavioural and cognitive patterns (such as perspective-​taking, imagina-
tive powers, social learning, etc.; see Carroll, 2004; Boyd, 2009; Dutton,
2009)  that might, in turn, enrich and alter readers’ responses to other
stories (Herman, 2007b).
Incorporating individual cognitive parameters into the framework (of
frames, scripts, and schemas) also offers crucial advantages for explaining
idiosyncrasies and similarities in readers’ cognitive and affective engage-
ment with narrative (see also Stockwell, 2002, p. 87). In this context, criti-
cism of ‘natural’ narratology has targeted the universality of the cognitive
set-​up in particular. As a response, and particularly in recognition of the
subjective dimension of narrativization, Fludernik acknowledges that

the aesthetics of the narrative text changes drastically over time and that these
developments will tend to influence narrativization. Moreover, individual

[ 48 ]  Literature through a Cognitive Lens

 

readers’ personal background, familiarity with literature, and aesthetic likes and
dislikes will also have a bearing on how texts are narrativized.
2003, p. 262

In line with Fludernik, we argue that this is precisely where the heuris-
tic value of literary emotions can be located within the present ‘natural’
framework. Within natural narratology, emotional responses to literary
artefacts are seen to be particularly pertinent to the subjective dimension
of reader experience. Natural narratology, consequently, is able to expand
its purview and become even ‘more natural’ by exploring the affective com-
ponent of narrativization. By integrating emotions as a central subjectiv-
izing element of readers’ construction of narrativity, natural narratology
is able to strike a balance between the universal and the particular dimen-
sions of narrativization. The universal aspects of emotion (as represented
by the basic emotion approach) can, at least partially, account for why read-
ers can agree on certain interpretations in the first place, and why they
are able to recuperate narrativity from texts beyond their own restricted
historical period or culture (the ‘shared emotions problem’). On the other
hand, the cultural and individual contingencies of (literary) emotions form
part of the subjective dimension of narrativization and are therefore partly
responsible for differences in readers’ interpretation of a single literary
text (the ‘differing emotions problem’).
Additionally, and consistently with ‘natural’ narratology’s focus on
cognitive reader parameters, narrativity is reconceived as constituted by
experientiality, defined as the reader’s ‘quasi-​mimetic evocation of “real-​
life experience” ’ (Fludernik, 1996, p. 12) as established during the reading
process (p. 36).4 Experientiality is hence mediated by means of the reader’s
consciousness, or, in other words, all narrative is argued to ‘fundamentally
represent another’s consciousness’ (p. 374, emphasis in original) as stimu-
lated through narrative and temporarily construed by the reader. With
experientiality as the universal topic of narration (p. 50), every narrative
conveys the special subjective quality of events as they are experienced by
the individual reader.5 A natural narratology conception of narrative conse-
quently not only foregrounds the reader’s consciousness as locus of simu-
lated (fiction-​derived) experience but posits experientiality as the defining
core of narrative—​narrative, as a minimum definition, cannot exist without
an anthropomorphic experiencer (i.e. a human consciousness) at the level of

4.  See Abbott (2009) for more traditional conceptions of ‘narrativity’.

5.  Note that Alan Palmer (2010) argues that the experiencing consciousness is not
necessarily an individual’s, but can belong to a collective instead.

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 49 ]
 

narrative reception. Such a conception, incidentally, appears to diminish the

role of plot as an essential constituent of narrativity, a controversial posi-
tion contested, for instance, by Jonas Grethlein. As he argues for ancient
texts and modern paralittérature, ‘what entices the reader are not the rather
schematic consciousness processes, but the drive of the action’ (2015, p.
267) characteristic of numerous literary genres. This apparent contradic-
tion, however, can fruitfully be reconciled within the present approach if the
dynamics of plot (as presented in the story) are equally modelled to derive
from general parameters of cognitive reader construal, i.e. as based on those
common mechanisms of sense-​making that enable individuals to project
time, action (and ultimately suspense) onto everyday memories, planning
for future events, or, in the present case, fictional literary incidents.
In much the same manner, and finally turning to affect, in The Feeling of
What Happens: Body and Emotion in the Making of Consciousness, Damasio
elaborates that ‘we continually have emotional feelings … sometimes low
grade, sometimes quite intense, and we do sense the general physical tone
of our being’ (2000, pp. 285‒286). Reader experientiality (and ultimately
reader consciousness) consequently includes emotions as an experien-
tial facet of the reading experience that accompanies story construals at
all times and is aptly described as ‘continuity of the melodic line of back-
ground emotions’ (p. 93) throughout the reading process. In line with this
idea, Fludernik has emphasized the specific emotional quality of the nar-
rativizing experience as follows:

For the narrator [i.e. reader] the experientiality of the story resides not merely
in the events themselves but in their emotional significance and exemplary
nature. The events become tellable precisely because they have started to mean
something to the narrator on an emotional level. It is this conjunction of experi-
ence reviewed, reorganized, and evaluated (‘point’) that constitutes narrativity.
2003, p. 245, emphases added

Like Damasio’s work and other situated approaches to cognition in con-

temporary cognitive science, natural narratology accordingly highlights
the close connection between narrativity (as projected in the act of nar-
rativization), experientiality (as the phenomenological quality of narra-
tivization), consciousness (as mediator of experientiality), and ultimately
emotions, which arguably constitute an indispensable experiential facet of
reader consciousness in story construals. To conclude, therefore, by main-
taining natural narratology’s cognitive constructivist design and by fore-
grounding experientiality at the heart of what constitutes a narrative, the
approach presented in this chapter can be argued to advocate a ‘natural’

[ 50 ]  Literature through a Cognitive Lens

 

bond between affective and cognitive narratology. We therefore propose

that Hogan’s and Fludernik’s approaches join forces in a concerted effort to
further integrate emotion and cognition in future narrative study.

ACKNOWLEDGEMENT

The authors would expressly like to thank Monika Fludernik for her invalu-
able advice and suggestions on all subjects pertaining to ‘natural’ narratology.

REFERENCES

Abbott, H. P. (2009). Narrativity. In P. Hühn, J. Pier, W. Schmid, and J. Schönert (Eds.),

Handbook of narratology (pp. 309–​328). Berlin: de Gruyter.
Boden, M. A. (2006). Mind as machine: A history of cognitive sciences (2 vols.). Oxford:
Clarendon Press.
Boyd, B. (2009). On the origin of stories:  Evolution, cognition, and fiction. Cambridge,
MA: Belknap.
Burke, M. (2011). Literary reading, cognition and emotion: An exploration of the oceanic
mind. London: Routledge.
Cantwell Smith, B. (1999). Situatedness/​embeddedness. In R. A. Wilson and F. C.
Keil (Eds.), The MIT encyclopedia of cognitive sciences (pp. 769–​771). Cambridge,
MA: MIT Press.
Caracciolo, M. (2012). J. M. Coetzee’s Foe and the embodiment of meaning. Journal of
Modern Literature, 36(1), 90–​103.
Caracciolo, M. (2014). The experientiality of narrative: An enactivist approach. Berlin: de
Gruyter.
Carroll, J. (2004). Literary Darwinism:  Evolution, human nature, and literature.
New York: Routledge.
Damasio, A. R. (1994). Descartes’ error: Emotion, reason, and the human brain. New York:
G. P. Putnam.
Damasio, A. R. (2000). The feeling of what happens: Body and emotion in the making of
consciousness. London: Routledge.
De Soussa, R. (1987). The rationality of emotions. Cambridge, MA: MIT Press.
Dutton, D. (2009). The art instinct: Beauty, pleasure, and human evolution. New York:
Bloomsbury.
Ekman, P. (1992). An argument for basic emotions. Cognition & Emotion, 6(3–​4),
169–​200.
Ekman, P. (2003). Emotions revealed: Recognizing faces and feelings to improve communi-
cation and emotional life. New York: Henry Holt.
Elster, J. (1999). Alchemies of the mind:  Rationality and the emotions. Cambridge:
Cambridge University Press.
Evans, D. (2002). Emotion: The science of sentiment. Oxford: Oxford University Press.
Fludernik, M. (1996). Towards a ‘natural’ narratology. London: Routledge.
Fludernik, M. (2003). Natural narratology and cognitive parameters. In D. Herman
(Ed.), Narrative theory and the cognitive sciences (pp. 243–​ 267). Stanford,
CA: Center for the Study of Language and Information.

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 51 ]
 

Frijda, N. H. (2007). The laws of emotion. Mahwah, NJ: Lawrence Erlbaum.

Gerrig, R. J. (1993). Experiencing narrative worlds: On the psychological activities of read-
ing. New Haven, CT: Yale University Press.
Grethlein, J. (2015). Is narrative ‘the description of fictional mental functioning’?
Heliodorus against Palmer, Zunshine & co. Style, 49(3), 257–​284.
Hamburger, K. (1968). Die Logik der Dichtung [The logic of literature]. Stuttgart: Klett.
Herman, D. (2007a). Cognition, emotion, and consciousness. In D. Herman (Ed.),
The Cambridge companion to narrative (pp. 245–​259). Cambridge:  Cambridge
University Press.
Herman, D. (2007b). Storytelling and the science of mind: Cognitive narratology, dis-
cursive psychology, and narratives in face-​to-​face interaction. Narrative, 15(3),
306–​334.
Herman, D. (2009). Cognitive narratology. In P. Hühn, J. Pier, W. Schmid, and
J. Schönert (Eds.), Handbook of narratology (pp. 30–​43). Berlin: de Gruyter.
Hogan, P. C. (2003). Cognitive science, literature, and the arts:  A  guide for humanists.
New York: Routledge.
Hogan, P. C. (2011). Affective narratology:  The emotional structure of stories. Lincoln,
NE: University of Nebraska Press.
Howe, D. (2013). Empathy:  What it is and why it matters. Basingstoke:  Palgrave
Macmillan.
Iser, W. (1978). The act of reading. London: Routledge.
Joyce, J. (1914/​2000). Eveline. In Dubliners (pp. 29–​34). London: Penguin Classics.
Kahneman, D. (2011). Thinking, fast and slow. London: Penguin.
Kukkonen, K. (2013). Space, time and causality in graphic narratives:  An embod-
ied approach. In D. Stein and J.-​N. Thon (Eds.), From comic strips to graphic
novels:  Contributions on the theory and history of graphic narrative (pp. 49–​66).
Berlin: de Gruyter.
Kukkonen, K., and Caracciolo, M. (2014). Introduction:  What is the ‘Second
Generation?’ Style, 48(3), 261–​271.
Lakoff, G., and Johnson, M. (1999). Philosophy in the flesh: The embodied mind and its
challenge to Western thought. New York: Basic Books.
LeDoux, J. (1996). The emotional brain: The mysterious underpinnings of emotional life.
New York: Simon and Schuster.
Levinson, J. (1997). Emotions in response to art: A survey of the terrain. In M. Hjort
and S. Laver (Eds.), Emotions and the arts (pp. 20–​36). New  York:  Oxford
University Press.
Mayer, J. D., Salovey, P., Caruso, D. R., and Sitarenios, G. (2003). Measuring emotional
intelligence with the MSCEIT V2.0. Emotion, 3, 97–​105.
McConachie, B., and Hart, F. E. (2006). Performance and cognition: Theatre studies and
the cognitive turn. London: Routledge.
Miall, D. S. (2006). Literary reading: Empirical & theoretical studies. New York: Peter Lang.
Myyry, L., and Helkama, K. (2007). Socio-​cognitive conflict, emotions, and complex-
ity of thought in real-​life morality. Scandinavian Journal of Psychology, 48(3),
247–​259.
Oatley, K. (1992). Best laid schemes: The psychology of emotions. New York: Cambridge
University Press.
Oatley, K., Keltner, D., and Jenkins, J. M. (2006). Understanding emotions (2nd ed.).
Malden, MA: Blackwell.
Palmer, A. (2010). Social minds in the novel. Columbus, OH: Ohio State University.

[ 52 ]  Literature through a Cognitive Lens

 

Prinz, J. J. (2004). Gut reactions:  A  perceptual theory of emotion. New  York:  Oxford
University Press.
Reis, H., and Judd, C. (2010). Handbook of research methods in social and personality psy-
chology. Cambridge: Cambridge University Press.
Robinson, J. (2005). Deeper than reason: Emotion and its role in literature, music, and art.
Oxford: Clarendon Press.
Solomon, R. C. (1993). The passions:  Emotions and the meaning of life. Indianapolis,
IN: Hackett.
Stearns, P., and Stearns, C. (1985). Emotionology: Clarifying the history of emotions
and emotional standards. American History Review, 90(4), 813–​836.
Sternberg, M. (2003). Universals of narrative and their cognitivist fortunes (I). Poetics
Today, 24(2), 297–​395.
Stockwell, P. (2002). Cognitive poetics: An introduction. London: Routledge.
Troscianko, E. T. (2014). Kafka’s cognitive realism. New York: Routledge.
Turner, M. (1991). Reading minds:  The study of English in the age of cognitive science.
Princeton, NJ: Princeton University Press.
Walton, K. (1990). Mimesis as make-​believe: On the foundations of the representational
arts. Cambridge, MA: Harvard University Press.
Zillmann, D. (1995). Mechanisms of emotional involvement with drama. Poetics,
23(1–​2), 33–​51.

C o g n i t i v e a n d A f f e c t i v e N a r r at ol o g y   [ 53 ]
 
 

CHAPTER 3

‘Annihilation of Self’
The Cognitive Challenge of the Sublime
DAVID S. MIALL

INTRODUCTION

The immense, beyond human grasp, infinite and timeless, a grandeur

of emotions, annihilation of self—​whether encountered in nature or in
poetry, this is the sublime, or one version of it at least, and one that par-
ticularly interested the aesthetic theorists and poets of the 18th and early
19th centuries. Induced in particular by the immensities encountered in the
Alps of France or Switzerland, one early traveller to Mont Blanc struggled
to express his response to the mountain:  ‘the mind is almost lost in the
sublimity of its own idea, and no tongue whatever is capable of describing,
and conveying justly to others, the humiliating, elevated, awful feelings of
the soul upon the sight of such an object’ (Bourrit, 1776, p. 8). This differs
from the account by Kant (1790/​1987), which would later predominate in
aesthetic theory: for him the sublime exemplifies the human mind’s asser-
tion of its pre-​eminence over nature in possessing the powers of reason. In
his explanation of the sublime, found for instance in our response to ‘shape-
less mountain masses’, the mind ‘feels elevated in its own judgment of itself
when it contemplates these without concern for their form … and finds all
the might of the imagination still inadequate to reason’s ideas’ (p. 113).
Such accounts lay claim to an unusual power or state of mind revealed by
an encounter with the sublime in nature, although nature is seen in oppos-
ing roles, being rendered either superior in some accounts or inferior to the
 

human mind in others. In this chapter I investigate what processes of mind

are at issue in possessing or invoking such powers. What do we know from
studies in cognition or neuropsychology that may help to validate or illumi-
nate these accounts of sublime experience? In fact, despite numerous theo-
retical accounts, especially those originating in the 18th century, it appears
that almost no empirical studies of the sublime experience have been car-
ried out. This chapter will not undertake another theoretical analysis of the
sublime: my aim, rather, is to map several locations where empirical work
might be productive by showing that existing studies of the mind in other,
often distant, fields can cast some light on experiences of the sublime.
In this way I hope to show that the sublime is of more than antiquar-
ian interest, allowing us a vision of states of consciousness that continue
to promise new insights, beyond the ascendancy of the term two to three
hundred years ago. Nowadays the term is little used, except perhaps by
specialists discussing the art and literature of that period. Have we, along
with the word, also lost our feeling for the sublime? Do we no longer have
the capacity to be impressed by the immensity of a high mountain, the
power of a great river, or the magnificence of a cathedral such as Chartres?
I  think not, although we may now be impressed by a different range of
sublime experiences: the most prominent of these are illustrated in films
set in space, from 2001 (1968) to Gravity (2013), although the sublime also
appears notably in advertisements for vacations, and it surely must, among
other things, motivate mountain climbers, such as those who now line up
to achieve their moment on the summit of Everest.
But not everyone would agree. Here is Thomas Weiskel writing in 1976
in his well-​known book The Romantic Sublime:

To please us, the sublime must now be abridged, reduced, and parodied as the
grotesque, somehow hedged with irony to assure us we are not imaginative ado-
lescents. The infinite spaces are no longer astonishing; still less do they terrify.
They pique our curiosity, but we have lost the obsession, so fundamental to the
Romantic sublime, with natural infinitude. We live once again in a finite natural
world. (p. 6)

In this view, we have lost that vision of the powers of reason that accord-
ing to Kant enable us to rise superior to nature in response to the sub-
lime, while recognizing the ‘unattainability’ of both the object in nature
and the ideas of reason. Put another way, as Wordsworth (1974) writes,
around 1811‒1812, in his fragmentary essay on the sublime: ‘it rouses us to
a sympathetic energy & calls upon the mind to grasp at something towards
which it can make approaches but which it is incapable of attaining—​yet so

[ 56 ]  Literature through a Cognitive Lens

 

that it participates [in an engagement with the] force which is acting upon
it’ (p. 354).
But Kant’s idealism is a problem. In Weiskel’s view, Kant’s definition of
the sublime ‘must be purged of its idealist metaphysics. Can the sublime
be construed at all’, he asks, ‘outside the presuppositions of idealism—​
whether Platonic or Kantian, theological or simply egotistical? It is pos-
sible, I believe,’ he says,

to preserve the dichotomous structure of Kant’s formulation in a ‘realist’ or

psychological account. … The ‘unattainability’ of the object with respect to the
mind would be duplicated as an inner structure, so that in the sublime moment
the mind would discover or posit an undefinable (ungraspable) domain within.
1976, p. 23

Here, Weiskel seems to be saying, is where we need to engage with the find-
ings of neuropsychology, and that what we may then find is that a series of
features and processes map our cloudy mental suppositions on to the real-​
time operations of a working brain.
Weiskel in fact takes his psychological intuitions in the direction of
psychoanalysis, where I do not propose to follow him. What he suggests,
however, is that we may find within ourselves a cognitive structure cor-
responding to the sublime. Perhaps this can be identified through EEG or
fMRI, or some similar brain-​mapping technique. We cannot, of course,
expect to identify a specific set of neurons where the sublime response
takes place; we might, rather, find a constellation of features in the brain
that contribute to sublime experiences, each of which may differ from
the next. There may be no ‘rules’, as Edmund Burke (1757/​1998, p. 49)
puts it; ‘art can never give the rules that make an art’ (see also Attridge,
2004, p. 12). In what follows I outline an approach that may give us access
to some of the cognitive and neuropsychological features that help cre-
ate the sublime experience. I begin with an examination of one written
account of a sublime experience dating from the early 19th century by
Lady Morgan (1821), and go on to introduce descriptions by other authors
later in the chapter.

PREDICAMENTS OF THE SUBLIME

Since the early 18th century, the crossing of the Alps has been a sought-​
after experience of the sublime. One traveller, Lady Morgan, crossing the
Alps in 1820, had read such accounts but felt compelled to disagree with

C o g n i t i v e C h a ll e n g e of t h e S u b li m e   [ 57 ]
 

them. Her account, published in 1821 in her book on Italy, forms a particu-
larly interesting moment, since it suggests that for her the discourse on the
sublime is radically deficient. She is on the usual route for travellers from
France to Italy (via Lyon and Turin), about to cross Mont Cenis. Since the
time of Gray and Walpole who made the crossing in 1739, this mountain
has had a fearsome reputation for travellers, who engage themselves to be
carried by porters across its heights in baskets.

Whoever has wandered far and seen much, has learned to distrust the prom-
ises of books; and (in respect of the most splendid efforts of human labour)
must have often felt how far the unworn expectation starts beyond its pos-
sible accomplishment. But nature never disappoints. Neither the memory
nor the imagination of authorship can go beyond the fact she dictates, or the
image she presents. … An aspect of the material world then presents itself,
which genius, even in its highest exaltation, must leave to original creation,
as unimitated and inimitable. The sensation it produces is too strong for plea-
sure, too intense for enjoyment. There, where all is so new, novelty loses its
charm; where all is so safe, conscious security is no proof against ‘horrible
imaginings;’ and those splendid evidences of the science and industry of man,
which rise at every step, recede before the terrible possibilities with which
they mingle, and which may render the utmost precaution of talent and phi-
lanthropy unavailable. … Here experience teaches the falsity of the trite
maxim, that the mind becomes elevated by the contemplation of nature in the
midst of her grandest works, and engenders thoughts ‘that wander through
eternity.’ The mind in such scenes is not raised. It is stricken back upon its
own insignificance. Masses like these sublime deformities, starting out of the
ordinary proportions of nature, in their contemplation reduce man to what he
is—​an atom.  … Well may the countless races of successive ages have left the
mysteries of the Alps unexplored, their snows untracked:  but … Gratitude
as eternal as the snows of Mount Blanc to them or him, who grappled with
obstacles coeval with creation, levelled the pinnacle and blew up the rock,
pierced the granite, and spanned the torrent, disputing with nature in all her
potency her right to separate man from man, and ‘made straight in the desert
an highway’ for progressive civilization!
1821, pp. 38‒40

From Kant’s perspective, Morgan here fails to rise above her own fearful
sensory imaginings; contemplating the destructive forces of nature, she
refuses to see that human reason is superior to anything in nature; on
the contrary, the scene shows man to be ‘an atom’. Yet, Morgan’s account
suddenly changes direction: perhaps she swings behind Kant after all in

[ 58 ]  Literature through a Cognitive Lens

 

her last remark. If human reason is superior to anything in nature, then

perhaps reason’s cherished ideal of human solidarity is also an adequate
motive to blast away rocks and pinnacles in order to make a road across
Mont Cenis. Napoleon, after all, had recently ordered a series of such
engineering feats in the Alps, notably at Mont Cenis itself and in the
Simplon Pass.
But Morgan’s account is also conflicted in another way. While dispar-
aging all books that purport to describe the sublime experience of the
Alpine traveller, Morgan’s need to describe her own response to the land-
scape obliges her to set at defiance the ‘unimitated and inimitable’. In
addition to her own language of destruction and catastrophe, she calls
in aid the phrases of Milton and Shakespeare. It is Macbeth who says
that ‘Present fears /​Are less than horrible imaginings’ while anticipating
the murder of Duncan; the thoughts ‘that wander through eternity’ are
those of the fallen angels, cherishing their intellectual survival in the pit
of hell, in Book II of Paradise Lost. In this way Morgan reaches for mean-
ings that, in her own prose, lie beyond her. She hollows out what Virginia
Woolf (1959, p. 60) was to call ‘caves’ of consciousness that scaffold the
further limits of expression. What is striking about her choice of phrases
is that, in each case, what is at issue is a monstrous violation of the order
of nature (Macbeth about to murder the king; the devils whose very being
represents disorder, casuistry, and ruin). While declaring invalid the
standard conceptions of the sublime, Morgan seems to put in their place
a sense of trauma, a nature seared by horror and woe, where man is out
of place: ‘In such regions’, she says, ‘nothing is in conformity with him,
all is at variance with his end and being, all is commemorative of those
elementary convulsions which sweep away whatever lives and breathes,
in the general wreck of inanimate matter’ (1821, p. 39)—​just as, for those
immersed in them, all is threatened in the worlds of Macbeth and the
fiends of Paradise Lost.
As Morgan puts it, ‘nature never disappoints’ (1821, p. 38). It is not that
there is no alternative to the false sublime of which she complains; on the
contrary, Morgan is still able to endow the forbidding precipices of Mont
Cenis with an imagined, humanly conceived meaning, thanks to Milton
and Shakespeare (and the Bible, in her last quotation), although a meaning
that reduces human significance close to zero. Perhaps it is not so strange,
then, that in the later part of this paragraph Morgan should seek to restore
human values by envisaging the makers of roads, those who ‘grappled with
obstacles coeval with creation’, those who disregarded the convulsions of
nature, its ravines and pinnacles, those who ‘ “made straight in the desert
an highway” for progressive civilization!’ (1821, p. 40)

C o g n i t i v e C h a ll e n g e of t h e S u b li m e   [ 59 ]
 

CONFLICTS OF THE SUBLIME

In Morgan’s account, as we will note, admiration is overcome by terror: anal-

ysis of sublime moments shows that two or more emotions confront one
another, of which the predominant one (terror) places the moment beyond
sensory capacity (as Kant proposed in using the term ‘imagination’). Yet
this is not a conflict that is static or frozen, but one between forces amongst
which one is tending to supplant another. Thus genius, a reigning term for
insight and power in contemporary aesthetics, is in Morgan’s account inad-
equate to portray the natural scene, which must be left ‘to original cre-
ation, as unimitated and inimitable’. In other words, our admiration for the
powers of human creation (‘evidences of the science and industry of man’)
recedes in the face of the terror (‘the terrible possibilities’) that can beset
us amidst the powers of nature. As her commentary continues she sets the
‘trite maxim’ of the mind’s elevation against its actual felt ‘insignificance’.
Stated even more forcefully (so that we trace a negative elevation building
across the whole passage), Morgan observes that ‘these sublime deformi-
ties … reduce man to what he is—​an atom’.
More than most landscape descriptions in the high mountains, Morgan’s
account emphasizes the conflict between the traveller’s physical predica-
ment, imperilled by storms, avalanches, torrents, cliffs, and the mere
scale of an Alpine crossing too demanding for human effort, against the
novel appearances of a landscape that is not only ‘unimitated and inimi-
table’ by the artist but amidst which the sensory powers of the mind fail to
encompass the scene before it. The mind is shown experiencing an altered
state of consciousness:  this is evident, for example, in the awareness of
time. For Morgan this awareness emerges from her sense of the enormous
convulsions that created this landscape over time, but which then ‘settle
into awful stillness and permanent desolation’ (1821, pp. 39‒40). For the
Alpine traveller this sense of the timeless invests the landscape, overcom-
ing the episodic, time-​limited nature of ordinary consciousness and mem-
ory, and extending indefinitely the ‘now’ usually considered to last just a
few seconds (Dainton, 2014). As the anonymous author of Picturesque Tour
through the Oberland (1823, p. 55) observes: ‘the ideas of eternal duration,
unlimited power, an inviolable asylum, take possession of the soul’. This
author also notes the challenge to our sensory powers on first catching
sight of the Jungfrau: ‘The eye is dazzled; it looks around for a point of sup-
port and of comparison, but none is to be found.’ And this in turn imposes
changes on our being: ‘One world ends, another begins, and that a world
governed by laws of a totally different existence.’ The sight of the Jungfrau
confronts us with our insignificance:  ‘Before this mass mankind appear

[ 60 ]  Literature through a Cognitive Lens

 

a race of pygmies.’ Numerous passages expressing a similar response to

Alpine landscape can be found not only in this book but in many other
contemporary publications. The structure of the sublime can also be seen
in other locations far from the Alps.
One of the most celebrated examples of the sublime is reproduced by
Longinus (2000) in his 1st-​century c.e. treatise: a verse by the Greek poet
Sappho (which would be lost to us if Longinus had not quoted it). Sappho
appears to be speaking to her female lover. This situation is radically dif-
ferent from mountain scenes of the sublime, but a comparable structure
can be recognized. The poem reads, in part (translated by Murray and
Dorsch, 2000):

For if I gaze on you but for a little while, I am no longer master of my voice, and
my tongue lies useless, and a delicate flame runs over my skin. No more do I see
with my eyes, and my ears hum. The sweat pours down me, I am all seized with
trembling, and I  grow paler than the grass. My strength fails me, and I seem
little short of dying. (p. 127)

Longinus draws our attention to one key feature of the sublime: that ‘how,
uniting opposites, she freezes while she burns, is both out of her senses
and in her right mind’; and that this shows us not one emotion but ‘a con-
course of emotions’ that is beyond control or comprehension. Longinus
emphasizes the fusion of the emotions into a single whole as the feature
that gives the poem its distinction (p.  127). Yet, looking more carefully,
we find paradoxically that each sense of the poet is confronted by some
greater power that brings it close to failure: eyes that cannot see, hearing
obstructed by a hum. Her capacities for speaking or making sense of her
sensations and emotions are threatened. Again, experience of the sublime
creates an altered state of consciousness. The sublime, paradoxically, can
involve the effect of quotidian feelings and emotions being subjected to
a diminishment or humiliation in the face of an incalculably superior and
powerful emotional agency. As Longinus notes, speaking of the power of
the great writers: ‘they all rise above the human level. All other attributes
prove their possessors to be men, but sublimity carries one up to where
one is close to the majestic mind of God’ (p. 156). Thus in literature of the
sublime ‘we look for something transcending the human’.
Although focused on quite different contexts, both Morgan and Sappho
describe the impact of the transcendent aspect of the sublime as a total-
izing experience, pre-​empting all else. Morgan dismisses the ‘trite maxim’
that the sublime elevates the mind; on the contrary it reduces ‘man to
what he is—​an atom’ (1821, p. 39). For Sappho it involves a failure of the

C o g n i t i v e C h a ll e n g e of t h e S u b li m e   [ 61 ]
 

senses that seems ‘little short of dying’ (Murray and Dorsch, 2000, p. 127).
Despite the fact that such remarkable literary and dramatic implications
are underpinned by profound cognitive phenomena, the aesthetic fea-
tures of the sublime have received little attention from cognitive science,
although there has been some study of related phenomena. In the next
section I outline briefly the work of two scholars in this field, Nico Frijda
and Jonathan Haidt.
Frijda (1986), who makes no specific mention of the sublime, provides
a brief discussion of similar concepts, amazement, surprise, and wonder (we
might also consider adding awe, astonishment, exaltation, and the verb to
astound). Frijda’s primary focus is on the physical correlates of these states,
beginning with the response to a sudden appearance. The emotion, whether
of surprise or wonder, initiates a passive, receptive state, an immobilizing of
the body. Amazement, he adds, involves ‘widening of the eyes, brief suspen-
sion of breathing, and general loss of muscle tone. The loss of muscle tone
causes the mouth to fall open, and may make the subject stagger or force him
to sit down’; the raising of the eyebrows and opened eyes set the attention
‘for peripheral stimuli wherever these may come from’ (p. 18). The system, in
other words, is set to respond primarily and immediately to danger. In Frijda’s
scheme amazement can be thought of as the most negative and the swift-
est version of several closely related emotions (amazement, astonishment, sur-
prise). In addition, they displace whatever emotions were taking place when
the interruption occurred. In contrast, wonder and awe encroach on the mind
and take it over more slowly. Examples might be approaching a great cathe-
dral such as Chartres, or attaining the summit of a mountain, or reading pas-
sages of poetry that describe such experiences. However, we may find in these
or other instances that emotions such as surprise or astonishment prepare us
for an ensuing experience of the sublime: within a brief interval (perhaps a
few seconds), a sense of the predicament of the self has developed in an array
of secondary emotions that confront the primary one.
While Frijda’s account, based as it is on evolutionary considerations, is
mainly concerned with negative construals of emotions such as surprise,
Haidt examines a positive emotion that is close to the sublime, that is,
elevation and its cognates. Elevation is regarded as a positive emotion, and
Haidt classifies it as one of the moral emotions in his contribution to the
Handbook of Affective Sciences (2003). Elevation, says Haidt, ‘seems to make
people stop, admire, and open their hearts and minds in a striking experi-
ence of liminality’, so that, as Haidt puts it, ‘elevation is caused by seeing
people blur the upper boundary between humans and God’ (p. 864).
In order to understand elevation, Haidt (2000) offers sketches of the
three components that he attributes to it: the circumstances in which the

[ 62 ]  Literature through a Cognitive Lens

 

elevating event occurred; what changes in thought and feeling occurred,

and what their outcomes are. In a systematic study of the contexts in which
people recalled experiences of elevation, the most frequent type of circum-
stances were seeing somebody give help to a person who was poor, sick, or
in difficulties. Asked to write in detail about ‘a specific time when you saw a
manifestation of humanity’s “higher” or “better” nature’, participants

commonly described themselves as being surprised, stunned, and emotionally

moved. Their descriptions imply that cognitive structures were changing under
the surface—​changing their views about humanity in a more optimistic way and
triggering more prosocial goals for themselves.

The results of the study showed that ‘love and a desire for affiliation’ appear
to be common consequences of witnessing such events. Feeling ‘more lov-
ing and inspired’, participants in a subsequent experimental study in which
experiences of elevation were induced in the laboratory also reported being
more likely to subsequently volunteer for a charitable organization.
While Haidt designates elevation, wonder, and awe as moral emo-
tions, however, the sublime—​or at least, some versions of it—​appears
to be amoral, beyond the grasp of ethical systems. In Morgan’s account,
for instance, the sublime object, Mont Cenis, is figured as indifferent to
human interests. The verse by Sappho shows her speaker as entirely pos-
sessed by her response to the beloved; no ethical perspective is offered or
relevant (although we know virtually nothing about the circumstances in
which the poem was written). While it should be pointed out that many
18th-​century accounts of the sublime appeal to the work of God in cre-
ating such landscapes, these references appear primarily to put in place
what we described earlier as the predominating emotion of the sublime
(such as astonishment), and no moral implications are present. Over time,
towards the end of the century, even such references to the divine tend to
disappear, replaced by a more immediate sense of the powers of nature. The
spectator of these scenes, comments one observer, ‘is struck with the com-
parative littleness of fleeting man … contrasted with the view of nature in
all her vast, eternal, uncontrolable [sic] grandeur’ (Williams, 1798, p. 63).

IMAGERY AND EMBODIMENT

Sublime experiences may be mediated by several of the senses, such as

sound or touch. But it seems likely that most experiences occur primarily
in the visual realm, perhaps supplemented by sound: we may, for instance,

C o g n i t i v e C h a ll e n g e of t h e S u b li m e   [ 63 ]
 

contemplate a picture of a river (in silence), or stand beside the river itself
where we can see a similar scene in reality and hear the sound of the water.
But how similar are these experiences? One depends on inner resources of
imagery and memory (the schema for a river), with an image presented in
two dimensions; the other depends on these too, but also on visual pow-
ers that allow for motion and the experience of force in three dimensions.
A number of studies in neuroscience have found that, far from being sep-
arate resources, imagery and perception activate the same brain regions.
I will outline the findings of one study by O’Craven and Kanwisher (2000).
Their study was based in part on the contrast of two conditions: visual
perception (as in viewing photographs of well-​known faces or buildings
on the campus where the study was carried out) and imagery (hearing the
name for the same faces or buildings and being asked to form a mental
image of it). It was found that regions of the brain that were more active, as
shown by fMRI data, were specific to the stimulus type, being specialized
either for face perception or for specified places. Greater activation occurred
for vision than for mental imagery. The main finding of the study, however,
was that vision and imagery drew upon basically the same processing activ-
ities. Inspection of fMRI data showed that it was detailed enough to iden-
tify the stimulus from a single response: each type of response, whether
perception or image, had its neural signature. In summary, say the authors,
the areas in ‘the ventral pathway that were activated during imagery for a
particular stimulus type fell within the region activated during perception
of the same stimulus class (on average, 92% for places and 84% for faces)’
(O’Craven and Kanwisher 2000, p.  1016). Thus, ‘the neural instantiation
of a mental image resembles the neural instantiation of the corresponding
perceptual image’ (p. 1019). In addition, they point out that ‘our data are
the first to show that the content of a single thought can be inferred from
its fMRI signature alone’ (p. 1019).
Given the importance of the visual processing system in the human
brain, the main finding of this study, that perception and imagery involve
the same brain regions, suggests some far-​reaching implications. It points
to another feature of sublime experience: the enactment in the brain of the
structures and stresses of the sublime object. We may hypothesize, then,
that the structures of the sublime are too powerful to be experienced in
their totality; they can be understood as exceeding and hence disturbing
the individual’s capacity for neurally representing and comprehending such
visions. As we have seen, this fact is anticipated by Morgan, in her com-
ments about our grasp of the powers of nature, which genius ‘must leave to
original creation, as unimitated and inimitable’. In this context, following
the insights of O’Craven and Kanwisher (2000), we may conclude that the

[ 64 ]  Literature through a Cognitive Lens

 

powers represented by outward forms of perception (whether representing

pictures or actual landscapes), are in some elementary forms unmanage-
able, even though they deploy the same neural resources as perception’s
inward replica, the image. In some experiences of the sublime, indeed,
these powers are disturbing, perhaps even traumatic—​as suggested by
Morgan’s account with her references to Shakespeare’s Macbeth and the
fiends of Milton.

FOREGROUNDING AND THE SUBLIME

A sublime experience at its most powerful induces feelings of astonish-

ment, elevation, or awe, as we have seen; in its verbal forms, therefore it
can be regarded as a type of defamiliarization evoked by the devices of fore-
grounding (unfamiliar and striking linguistic features). This experience,
which in the words of Longinus ‘scatters everything before it like a thun-
derbolt’ (p.  114), is dependent on the resources of language drawn upon
by the writer in order to arouse powerful feelings in a hearer or reader.
As commentators have often noted, critical discussion of sublime texts
may develop the powers of the sublime experience in the language being
deployed. In the words of Philip Shaw (2006, p. 6), ‘words have a power
… to raise the idea of the sublime, such that the distinction between the
sublime object and its description no longer applies’.
In this section I  analyse examples of foregrounding, showing how
they bring the sublime into being. My main examples will be drawn from
a letter by the poet Shelley written in 1816. The letter describes Shelley’s
first encounter with the sight of Mont Blanc as he approaches it along
the Chamonix valley. The letter was written to his friend Peacock back in
England. (For a fuller account, see Miall, 2007.)

Mont Blanc was before us but was covered with cloud, & its base furrowed with
dreadful gaps was seen alone. Pinnacles of snow, intolerably bright, part of the
chain connected with Mont Blanc shone thro the clouds at intervals on high.
I never knew I never imagined what mountains were before. The immensity of
these aeriel summits excited, when they suddenly burst upon the sight, a senti-
ment of extatic [sic] wonder, not unallied to madness—​And remember this was
all one scene. It all pressed home to our regard & our imagination.—​Though
it embraced a great number of miles the snowy pyramids which shot into the
bright blue sky seemed to overhang our path—​the ravine, clothed with gigan-
tic pines and black with its depth below.—​so deep that the very roaring of the
untameable Arve which rolled through it could not be heard above—​was close

C o g n i t i v e C h a ll e n g e of t h e S u b li m e   [ 65 ]
 

to our very footsteps. All was as much our own as if we had been the creators
of such impressions in the minds of others, as now occupied our own.—​Nature
was the poet whose harmony held our spirits more breathless than that of the
divinest.
1964, p. 497

First, most obvious, is the sense of defamiliarization, of being over-

whelmed:  ‘I never knew I  never imagined what mountains were before.’
Second, the pressure of the experience seems to cause his syntax to break
down at least once (‘I never knew I never imagined …’), while the effort
to relate the different parts of what was ‘all one scene’ prompts an unusual
number of dashes. Third, Shelley’s language shows the experience press-
ing on, and threatening to overwhelm the senses, not only vision (‘intoler-
ably bright … burst upon the sight’) but also the kinaesthetic response
suggested by ‘seemed to overhang our path’, and ‘held our spirits more
breathless’. Fourth, and perhaps most significant, a merging of mind and
nature: Shelley represents the scene as though he had participated in pro-
ducing it. In effect, this abruptly shifts the deictic centre in the last two
sentences, blurring the boundary between self and landscape.
This analysis (necessarily brief) helps confirm the importance of fore-
grounding in developing a response to the sublime. To examine its effec-
tiveness in a neuropsychological perspective requires identifying a feature
that can be repeatedly tested. Whether this gives us access to a sublime
feature is a question I will raise later.
I mention a neuropsychological study in this context by Philip Davis
and his colleagues (Thierry, Martin, Gonzalez-​Diaz, Rezaie, Roberts, and
Davis, 2008). They studied responses to a single component of foreground-
ing, the functional shift, with all examples culled from Shakespeare’s plays.
Functional shift is the conversion of one part of speech to another. For
instance, a pronoun becomes a noun—​‘the cruellest she alive’ (Twelfth
Night); or a noun is turned into a verb:  ‘He childed as I  fathered’ (King
Lear), ‘he godded me indeed’ (Coriolanus). In a study using EEG, they were
able to show that the original version of such lines (‘he godded me indeed’)
uniquely evoked a P600 response in the participants. P600 is the signal of
a syntactic anomaly: a positive wave peaking at 600 ms following exposure
to the unusual word. In contrast, manipulating the lines to replace them
with literal (‘he deified me’) or anomalous versions (‘he charcoaled me’ or
‘he poured me’) evoked either N400 and P600 waves together or no signifi-
cant wave at all. This effect, where N400 and P600 reliably distinguish the
function shift from related verbal forms, suggests that we may find other
foregrounded effects that reflect measurable neural structures.

[ 66 ]  Literature through a Cognitive Lens

 

The study also suggests that the function shift (e.g. ‘the cruelest she
alive’, Twelfth Night) may contribute to a sublime experience despite its
lack of context. As Davis (2006) remarks, ‘the P600 surge means that
the brain was thereby primed to look out for more difficulty, to work at
a higher level, whilst still accepting that, fundamentally, sense was being
made’. Thus each figure, standing out from the literal or anomalous ver-
sions, places the brain in readiness for the next figure. One important theo-
retical implication of this study is that we do not habituate to the function
shift. The participants in the study must have soon realized that they were
constrained to listen to a series of similar figures, whether literal, function
shifted, or anomalous, yet P600 showed that their responses to this figure
remained constant, at a high level.
The validity of this claim is supported by several other studies in neu-
ropsychology that show a consistent lack of habituation in response to
emotional types of verbal expression (Kissler, Herbert, Peyk, and Kissler,
2007)  or to visual stimuli (Schupp, Stockburger, Codispoti, Junghofer,
Weike, and Hamm, 2006). Schupp and colleagues’ study, for example,
reported that ERP patterns showed a virtual absence of habituation to
emotional pictures: initial response, in the 150‒300 ms window, remained
almost as strong after repeated exposures as at the first. For the partici-
pants in the study by Kissler and colleagues (2007), early ERP responses
at around 250 ms distinguished repeated pleasant and unpleasant words
from neutral words being read. No habituation of emotional response
occurred, as shown by the absence of any repetition effect.
The responses in these three studies are not, of course, sublime: they are
likely to disappear within a second or so, whereas response to a sublime
passage may last a lifetime. However, they do help confirm the phenom-
enon of dehabituation: with repeated affective response to foregrounded
features (that contrast with background figures:  anomalies, etc.) literary
expressions of the sublime remain permanently fresh and influential. As
Longinus notes, a sublime passage survives numerous hearings: ‘If it can
stand up to repeated examination’, and if it is universal in its appeal, it
‘pleases all men at all times’ (2000, p. 120).

AT THE RHINE FALLS

Lastly, one further example of sublime writing is taken from another

writer of the Romantic period, Helen Maria Williams (1798). This passage
will demonstrate the power of the sublime to change or suspend the self-​
concept of the reader.

C o g n i t i v e C h a ll e n g e of t h e S u b li m e   [ 67 ]
 

Williams was an English woman who settled in France early during the
Revolution. She sent back reports on what she saw and learned about the
Revolution. She was briefly imprisoned in 1793, then obtained a passport
and left for travels in Switzerland during 1794. The book she published in
1798 includes both reports on the political situations she found in the can-
tons and descriptions of the scenery.
Williams’s report of a visit to view the Rhine Falls near Schaffhausen in
northern Switzerland is written in sublime style, as follows:

Our path … concealed for some time the river from our view; till we reached a
wooden balcony, projecting on the edge of the water, and whence, just sheltered
from the torrent, it bursts in all its overwhelming wonders on the astonished
sight. That stupendous cataract, rushing with wild impetuosity over those bro-
ken, unequal rocks, which, lifting up their sharp points amidst its sea of foam,
disturb its headlong course, multiply its falls, and make the afflicted waters
roar—​that cadence of tumultuous sound, which had never till now struck upon
my ear—​those long feathery surges, giving the element a new aspect—​that
spray rising into clouds of vapour, and reflecting the prismatic colours, while
it disperses itself over the hills—​never, never can I can forget the sensations of
that moment! when with a sort of annihilation of self, with every past impres-
sion erased from my memory, I felt as if my heart were bursting with emotions
too strong to be sustained.—​Oh, majestic torrent! which hast conveyed a new
image of nature to my soul, the moments I have passed in contemplating thy
sublimity will form an epocha in my short span!—​thy course is coeval with time,
and thou wilt rush down thy rocky walls when this bosom, which throbs with
admiration of thy greatness, shall beat no longer! (pp. 59‒61)

For Williams, the Falls are both dangerous and inspiring. What may distin-
guish this account from the earlier ones I have presented in this chapter is
the sense of a process in thought. As she stands within a few metres of the
water, she seems to come to an understanding of something about the Falls
and about herself.
Her response to the Falls is articulated in three phases. As can be seen
from the tenses she employs, it moves from the present (‘it bursts in all
its overwhelming wonders’), through the past (‘I felt as if my heart were
bursting’), to the future (‘thou wilt rush down thy rocky walls’). In addition
to the distinctions due to time, Williams also traces three phases in the
process of her feelings:  first she is overwhelmed (‘the astonished sight’),
then she senses herself as transformed (‘every past impression erased’),
and lastly she claims some novel historical insights (‘thy course is coeval
with time’). Also notable is the transition from detailed description of the

[ 68 ]  Literature through a Cognitive Lens

 

Falls, hinting at its anthropic presence in such terms as ‘impetuosity’, ‘lift-

ing’, and ‘afflicted’; then a turn to realizing the cognitive impact this has on
her, so powerful that it ‘annihilates’ the self. Finally she realizes the brief-
ness of her life in comparison with the time course of the Falls.
Thus, Williams’s reflection of her experience is richly animated by the
relations of time, sensation, and understanding of the self. These drama-
tize her position on the balcony in close proximity to the water, yet in a
position of personal safety—​a classic stance for the 18th-​century sublime.
Referring to these aspects of Williams’s account, however, points up
that we have said little elsewhere about the most far-​reaching experiences
of the positive sublime, those that evoke transforming experiences. Almost
no empirical studies on this topic have been carried out (though see Eskine,
Kacinik, and Prinz [2012] on the visual sublime, and Silvia, Fayn, Nusbaum,
and Beaty [2015] on personality and the sublime), yet suggestive parallels
exist. Although we cannot readily elicit a sublime experience in the MRI
machine, studies of analogous phenomena indicate what we might expect
to find if we could. I will briefly mention studies in three areas: mysticism,
sexuality, and music.

ANNIHILATION OF SELF

Studies of neuropsychological correlates of mystic experience were pursued

by d’Aquili and Newberg (1999) over a number of years. They noted paral-
lels between ‘aesthetic, spiritual, and mystical states’ (2000), but their pri-
mary claim was that mystic experiences involve deactivation of the sense of
space, together with loss of the bounded self and the self/​other distinction.
Deactivation, or ‘deafferentation’ as they call it, ‘can occur through
inhibitory fibers from other nervous system structures’ (1999, p.  112).
Taking effect in the frontal cortex (the executive domain) and parietal lobe
(the orientation area), it can be willed, as in meditation, by dismissing all
thought. They explain the summit of the mystical experience as follows: ‘the
total deafferentation of the left orientation association area results in the
obliteration of the self-​other dichotomy at precisely the same moment that
the deafferentation of the right orientation association area is associated
with a sense of absolute transcendent wholeness’ (p. 112). Although their
work has been dismissed (for a critique, see Kelly and Grosso, 2007), the
similarities to the sublime experience, as recorded by Williams and many
others, are evident—​above all in what Williams called ‘annihilation of self’.
Sexual experience at its height has often been regarded as transcen-
dental, offering insights into a spiritual world different from this one. For

C o g n i t i v e C h a ll e n g e of t h e S u b li m e   [ 69 ]
 

example, Tantric sexual practices, drawing on ancient Indian religion, have

been highly regarded as a gateway to spiritual wisdom. Their sublime emo-
tions offer an interesting parallel to those of the mystic. My comments here
refer mainly to the work of Georgiadis and his colleagues (2006), who were
able to study sexual experience in the MRI scanner. Perhaps the most sig-
nificant finding they report is that sexual climax in both sexes ‘was mainly
associated with profound rCBF [regional cerebral blood flow] decreases in
the neocortex’; this provides ‘strong evidence’ (if only correlational) ‘that
the frontal and temporal lobes have inhibitory control over sexual behav-
iour. Decreased perfusion [i.e. deactivation] of these regions would release
the inhibition and enable sexual behaviour’ (p. 3313). This effect was stron-
ger in women (Holstege and Huynh, 2011). They also found ‘additional
significant deactivation in the dorsomedial prefrontal cortex (PFC), a
region best known for its involvement in moral reasoning and social judge-
ment… . The significant deactivation of this region during orgasm implies
absence of moral judgement and self-​referential thought’ (Georgiadis et al.,
2006, p. 3314). Again it is as if the boundaries of self, and self and other,
are effaced at that moment. This study, then, provides another example of
annihilation of self. (In an analysis of a poem by Baudelaire, Tsur [2003,
pp. 258‒261] arrives at a similar conclusion.)
Blood and Zatorre (2001) conducted a PET (positron emission tomog-
raphy) study of moments of musical experience, chosen by participants,
at which they reliably felt ‘chills’ (emotional and bodily responses). They
found, as stated in the title of their paper, that ‘intensely pleasurable
responses to music correlate with activity in brain regions implicated in
reward and emotion’. However, correlations were negative for rCBF to
‘chills’ from music with the median prefrontal cortex and posterior neo-
cortical regions. In other words, the more intense the chills, the greater the
deactivation of the frontal cortex.
These three domains—​mysticism, sexuality, and music—​are quite dif-
ferent, yet it appears that the main effect in each is to blur or erase the
neural representation of the self. There are many ways in which this insight
can be expressed. In contrast to the annihilation of the self, another way
of representing it is to refer to the sense of unity of self and universe. Here
is one of Wordsworth’s statements, for example (from the ‘Essay on the
Sublime and Beautiful’), which suggests the dissolution of the self that
thinks and reasons: ‘For whatever suspends the comparing power of the
mind & possesses it with a feeling or image of intense unity, without a con-
scious contemplation of parts, has produced that state of the mind which
is the consummation of the sublime’ (p. 354). Another context in which the
sublime self can occur is that of imminent death: Longinus, as we noted,

[ 70 ]  Literature through a Cognitive Lens

 

draws attention to this in Sappho’s verse, which concludes with this state-
ment: ‘My strength fails me, and I seem little short of dying’ (2000, p. 127).

CONCLUSION

It seems likely that deactivation of the frontal cortex can occur under many
other conditions, as Carter (1999) suggests. A heavy input of information to
the frontal cortex from below, she says, is why ‘a sudden flood of emotion may
occlude thought’ or ‘why terror can (momentarily at least) wipe a brain clean
of thought’ (p. 183). But Williams’s account of the Rhine Falls, and Shelley’s
of Mont Blanc, go well beyond these examples. They seem to propose that the
most powerful experiences of the sublime available to us require the borders
of the self to be dissolved, allowing it to become a part of infinite space. What
remains to be studied with the empirical means at our disposal—​behavioural
and neuropsychological—​are the aesthetic implications of the moment of
poetic sublime: its ‘annihilation of self’, its temporary erasure of the memory,
its powerful feelings. These seem to be the constituents of the altered states of
consciousness that help develop our experience of the sublime.

REFERENCES

Anon. (1823). Picturesque tour through the Oberland. London: Ackermann’s Repository.

Attridge, D. (2004). The singularity of literature. Abingdon: Routledge.
Blood, A. J., and Zatorre, R. J. (2001). Intensely pleasurable responses to music correlate
with activity in brain regions implicated in reward and emotion. Proceedings of
the National Academy of Sciences of the United States of America, 98, 11818–​11823.
Bourrit, M. T. (1776). A relation of a journey to the glaciers in the Dutchy of Savoy (2nd ed.,
C. Davy and F. Davy, Trans.). Norwich: Richard Beatniffe.
Burke, E. (1998). A philosophical enquiry into the origin of our ideas of the sublime and the
beautiful (A. Phillips, Ed.). Oxford: Oxford University Press.
Carter, R. (1999). Mapping the mind. London: Seven Dials.
d’Aquili, E. G., and Newberg, A. B. (1999). The mystical mind: Probing the biology of reli-
gious experience. Minneapolis, MN: Fortress Press.
d’Aquili, E. G., and Newberg, A. B. (2000). The neuropsychology of aesthetic, spiritual,
and mystical states. Zygon, 351, 39–​51.
Dainton, B. (2014). Temporal consciousness. In E. N. Zalta (Ed.), The Stanford encyclo-
pedia of philosophy (Spring 2014 ed.). Retrieved from http://​plato.stanford.edu/​
archives/​spr2014/​entries/​consciousness-​temporal/​
Davis, P. M. (2006). The Shakespeared brain. The reader, 23, 39–​43. Retrieved from
http://​thereaderonline.co.uk/​features/​the-​shakespeared-​brain/​
Eskine, K. J., Kacinik, N. A., and Prinz, J. J. (2012). Stirring images: Fear, not happiness
or arousal, makes art more sublime. Emotion, 125, 1071–​1074.
Frijda, N. H. (1986). The emotions. Cambridge: Cambridge University Press.

C o g n i t i v e C h a ll e n g e of t h e S u b li m e   [ 71 ]
 

Georgiadis, J. R., Kortekaas, R., Kuipers, R., Nieuwenburg, A., Pruim, J., Reinders, et al.
(2006). Regional cerebral blood flow changes associated with clitorally induced
orgasm in healthy women. European Journal of Neuroscience, 24, 3305–​3316.
Haidt, J. (2000). The positive emotion of elevation. Prevention and Treatment (Vol. 3).
Retrieved from http://​journals.apa.org/​prevention/​volume3/​pre003003c.html
Haidt, J. (2003). The moral emotions. In R. J. Davidson, K. R. Scherer, and H. H.
Goldsmith (Eds.), Handbook of affective sciences (pp. 852–​870). Oxford: Oxford
University Press.
Holstege, G., and Huynh, H. K. (2011). Brain circuits for mating behavior in cats and
brain activations and de-​activations during sexual stimulation and ejaculation
and orgasm in humans. Hormones and Behavior, 59, 702–​707.
Kant, I. (1987). The critique of judgment (W. S. Pluhar, Trans.). Indianapolis, IN: Hackett.
Kelly, E. F., and Grosso, M. (2007). Mystical experience. In E. F. Kelly and E. W. Kelly,
Irreducible mind: Toward a psychology for the 21st century (pp. 495–​575). Lanham,
MD: Rowman and Littlefield.
Kissler, J., Herbert, C., Peyk, P., and Kissler, M. J. (2007). Buzzwords:  Early cortical
responses to emotional words during reading. Psychological Science, 18, 475–​480.
Longinus. (2000). On the sublime. In P. Murray and T. S. Dorsch (Eds. and Trans.),
Classical literary criticism (pp. 113–​166). Harmondsworth: Penguin.
Miall, D. S. (2007). Foregrounding and the sublime: Shelley in Chamonix. Language and
Literature, 16, 155–​168.
Morgan, Lady (Sydney Owenson). (1821). Italy (Vol. 1, 3rd ed.). London:  Henry
Colburn. Also online at https://​archive.org/​details/​italymorgan01morgiala
O’Craven, K. M., and Kanwisher, N. (2000). Mental imagery of faces and places
activates corresponding stimulus-​specific brain regions. Journal of Cognitive
Neuroscience, 12, 1013–​1023.
Schupp, H. T., Stockburger, J., Codispoti, M., Junghofer, M., Weike, A. I., and Hamm,
A. O. (2006). Stimulus novelty and emotion perception:  The near absence of
habituation in the visual cortex. NeuroReport, 17, 365–​369.
Shaw, P. (2006). The sublime. London: Routledge
Shelley, P. B. (1964). The letters of Percy Bysshe Shelley (Vol. 1, F. L. Jones, Ed.).
Oxford: Oxford University Press.
Silvia, P. J., Fayn, K., Nusbaum, E. C., and Beaty, R. E. (2015, 17 August). Openness to
experience and awe in response to nature and music: Personality and profound
aesthetic experiences. Psychology of Aesthetics, Creativity, and the Arts, 9(4),
376–​384.
Thierry, G., Martin, C. D., Gonzalez-​Diaz, V., Rezaie, R., Roberts, N., and Davis, P. M.
(2008). Event-​related potential characterisation of the Shakespearean func-
tional shift in narrative sentence structure. NeuroImage, 40, 923–​931.
Tsur, R. (2003). On the shore of nothingness: A study in cognitive poetics. Exeter: Academic.
Weiskel, T. (1976). The Romantic sublime. Baltimore, MD:  Johns Hopkins
University Press.
Williams, H. M. (1798). A tour in Switzerland; or, a view of the present state of the govern-
ments and manners of those cantons (Vol. 1). London: G. G. and J. Robinson.
Woolf, V. (1959). A writer’s diary. Being extracts from the diary of Virginia Woolf (L. Woolf,
Ed.). London: Hogarth Press.
Wordsworth, W. (1974). The prose works of William Wordsworth (Vol. 2, W. J. B. Owen
and J. W. Smyser, Eds.). Oxford: Oxford University Press.

[ 72 ]  Literature through a Cognitive Lens

 

CHAPTER 4

The Space between Your Ears

Construal Level Theory, Cognitive Science,
and Science Fiction
JAMES C ARNEY

INTRODUCTION

The critical study of science fiction (SF) has always had a cognitive dimen-
sion. From the outset, researchers in the field highlight SF as a genre of
‘cognitive estrangement’ (Suvin, 1972, 1979)  that alienates readers from
their habitual environment—​a claim that orientates SF scholarship to the
present day (Hollinger, 1999; Milner, 2012). Nevertheless, for all that the
critical literature centres on the term ‘cognitive’, it makes relatively little
engagement with empirical research on how human beings actually think,
feel, or behave. In this, SF studies runs counter to recent trends in liter-
ary studies, which use insights from cognitive science, neuroscience, and
experimental psychology to explain how texts impact on readers (Zunshine,
2006; Hogan, 2013; Carney, 2014; Troscianko, 2014). Although none of
this mandates that cognitive approaches should be used in the analysis of
SF, the historical preoccupation with cognition on the part of SF scholars
does point to a potentially useful collaboration between the two fields.
My goal in the present chapter will be to pursue this collaboration.
Specifically, I  hope to develop an account of SF that is attentive to how
the characteristic features of the genre trigger empirically attested ten-
dencies in human cognition. My main tool in doing this will be construal
level theory (CLT), a framework developed in social psychology for dealing
 

with the effects of distance upon cognition. (‘Distance’, in this connec-

tion, includes spatial displacement but also comprehends figurative forms
like temporal, social, and probabilistic distance.) The essential discovery
of CLT is to show that there is a reciprocal relationship between psycho-
logical distance and construal level (that is, level of abstraction in how
something is cognitively assessed), to the extent that extremes of distance
in the spatial, temporal, social, or probabilistic dimensions cause individu-
als to construe objects as being abstract, and extremes of abstraction cause
individuals to construe objects as distant (Liberman and Trope, 2009;
Trope and Liberman, 2010). It is evident, for instance, that we are inclined
to imagine probabilistically remote objects or events in very general terms
(‘if I win the lottery I’ll be generous to my friends’), whereas we envision
more likely objects or events quite concretely (‘if John pays me the £100
he owes me I’ll treat myself to dinner in that Italian place on Friday’).
Equivalently, abstract objects and descriptions are felt to be consonant
with distance (‘the archaeological record for H. sapiens evidences symbolic
play as far back as the Pleistocene’), where concrete objects and catego-
ries are consonant with nearness (‘Frank’s children played hide-​and-​seek
yesterday’). Starting from these modest insights, the experimental pro-
gramme of CLT has gone on to generate a rich body of results that link
manipulations of distance and construal level to changes in moral orienta-
tion, self-​perception, emotionality, tolerance, conformity, creativity, and a
range of other attitudes and behaviours.
The value of CLT for the study of SF is that it provides an intellec-
tual resource for understanding key features of SF writing. For the fact
is, almost all SF relies on the maximization of psychological distance to
achieve its effects. Most obviously, this manifests in the cosmic setting of
much SF, but thematizations of time, the alien, and the improbable are also
staples of the genre. Given that CLT predicts very specific cognitive effects
to follow from this type of distance manipulation, it is thus follows that the
theory may well provide a powerful resource for understanding the psy-
chological impact of SF upon its audience. In the first instance, this will
offer valuable psychological insights to reception theory—​a field that has
recently come to recognize the benefits of empirical results for understand-
ing how readers respond to texts (Hamilton and Schneider, 2002; Emmott,
Sanford, and Alexander, 2013). In the second, any successful account of
how a genre achieves its effects should be able to demonstrate its value by
accounting for features of that genre that are problematic or puzzling. My
CLT-​informed account will do exactly this by resolving problems concern-
ing characterization, purpose, and linguistic register that have persistently
absorbed the energies of SF critics. Thus, using CLT to understand how SF

[ 74 ]  Literature through a Cognitive Lens

 

manipulates psychological distance will make an important contribution to

our understanding of a culturally significant genre.
The following discussion will comprise four stages. The first will involve
a short exposition on the nature of CLT; after this, I shall conduct a CLT-​
informed survey of how psychological distance is maximized in SF; then,
I will develop a new understanding of the genre on the basis of CLT; and
finally, by way of conclusion, I  will offer some thoughts on how the CLT
approach can be extended to other genres and aesthetics. The result, ide-
ally, will be to advance the conversation between literary studies and cogni-
tive science in a manner that is valuable to both partners.

CONSTRUAL LEVEL THEORY

It is a truism to say that human cognition is influenced by context; what is

less easy is to specify how, exactly, influence is transmitted. Different theo-
ries make different commitments in this regard, variously assigning the
mechanism to learned association (Watson, 1913; Skinner, 1976; Staddon,
2014), innate biological programmes (Darwin, 1871; Wilson, 2000) or some
combination of these (Kahneman, 2012; Evans and Stanovich, 2013). CLT
is probably the most recent attempt to theorize the effect of context on
thinking, and proceeds by emphasizing how two variables—​distance and
construal level—​shape judgements and preferences. Distance is a psycho-
logical measure; it concerns the remoteness (or closeness) of an event or
object in terms of space, time, social affinity, and likelihood. Although it is
certainly true that all of these measures also pick out real-​world quantities,
their use in CLT is defined in relation to the perceptions of the perceiving
subject—​thus, the experience of distance in a dream (say) will be predicted
to cue the same effects as this experience in reality. Construal level, on the
other hand, is a measure of abstraction, and picks out how concretely (or
abstractly) we perceive an object or event. CLT proposes that both variables
interact, insofar as changes in distance affect perception of construal level,
and vice versa. Evidence suggests that these interactions can have pro-
found impacts on an unexpectedly wide range of thoughts and behaviours.
To start with the first variable, CLT predicts that greater distance will
result in a stronger focus on the general, primary features of an object
or event over its incidental secondary features. For instance, thinking
about a lecture happening tomorrow might prompt concern with con-
crete issues like its venue (‘that lecture hall has terrible sound’), whereas
thinking about a lecture in the distant future is likely to suppress inci-
dental details in favour of essential ones (‘that lecture will be worth

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 75 ]
 

attending’)—​even if the detail in question (the venue) is held constant

(Trope and Liberman, 2000). More generally, studies have found that
linguistic, affective, taxonomical, and procedural judgements all tend
to focus on essential rather than contextual or incidental features when
paired with maximizations of spatial, temporal, social, or probabilistic
distance (Todorov, Goren, and Trope, 2007; Liberman and Trope, 2009;
Bruehlman-​Senecal and Ayduk, 2015).
Conversely, manipulating the second variable, construal level, affects
perceptions of distance. For instance, social descriptions performed
using high-​level, abstract designations such as ‘civilization’ are likely
to cue expectations of temporal or spatial distance (‘Roman civiliza-
tion’, ‘Japanese civilization’). Against this, low-​level, concrete construals
evoked by the local condition of local-​global processing tasks have been
shown by Woltin and colleagues (2011) and Luguri, Napier, and Dovidio
(2012) to decrease perceptions of social distance by inducing increased
levels of empathy towards minorities and out-​groups. Experimental
manipulations of other measures of abstraction have also been shown
to affect judgements concerning self-​control, psychological coherence,
creativity, and affect. These results confirm the CLT axiom that ‘differ-
ent levels of construal serve to expand and contract one’s mental hori-
zons and thus mentally traverse psychological distances’ (Trope and
Liberman, 2010, p. 442).
From all of this, it follows that if the claims of CLT theorists are correct,
CLT describes a fundamental aspect of human cognition. Judgements of
distance and/​or abstraction inform almost every aspect of human behav-
iour, so the discovery of a link between these two variables represents an
important result. This is particularly so when it comes to researching coun-
terfactual thinking, or how humans conceive of what is not the case. Unlike
factual thinking, which is constrained by real-​world causal structures,
counterfactual thinking can readily engage with extremes of distance or
abstraction that may only rarely be encountered otherwise. For this reason,
fiction—​and in particular, speculative or experimental fiction—​represents
a natural target for CLT, given that it is explicitly designed to evoke the
extremes discussed in CLT research. Nevertheless, though proponents
of CLT have been quick to apply it in areas like marketing, architecture,
and product design (Dhar and Kim, 2007; Dębek and Bożena, 2013; Luca,
2015), it has yet to be used in the analysis of literature. Thus, a strong prima
facie case can be made for the utility of CLT as a tool for explaining how lit-
erary works achieve their effects. While I shall demonstrate this here only
with respect to SF, I volunteer this exposition as an illustration of how CLT
might be operationalized in the wider field of literary studies.

[ 76 ]  Literature through a Cognitive Lens

 

PSYCHOLOGICAL DISTANCE IN SCIENCE FICTION

My fundamental claim is that SF manipulates psychological distance to

affect construal level. To justify this claim requires a survey of the genre
that shows the precise ways in which it does this. Inevitably, reviews of this
type must be summary in character; nevertheless, they retain a value as
map-​building exercises that can be subsequently finessed in greater detail.
Here, I will begin by showing that many of the major texts and subgenres
of SF work by foregrounding one or more of the four dimensions of psycho-
logical distance. Once I have done this, I shall comment on the likely effects
of doing so by way of what we know from the CLT literature.

Spatial Distance

Edmund Burke long ago noted that ‘as the great extreme of dimension
is sublime, so the last extreme of littleness is in some measure sublime
likewise’ (1757/​1909, p. 100). Given the role played by the sublime in SF
(Robu, 1988; Nicholls, 2000), it is unsurprising that extremities of space
should represent a core feature of the genre. This is visible on a terrestrial
scale in early SF like Jules Verne’s Twenty Thousand Leagues Under the Seas
(1870/​2001) and Journey to the Centre of the Earth (1864/​2008); the cosmic
analogue to this is the premise of almost all popular articulations of SF
up to the present day. However, extremes of displacement are not merely
background features of the genre: a number of core texts foreground the
‘spatial sublime’ as a topic of cognitive interest in its own right. One crucial
example is Poul Anderson’s Tau Zero (1970), where a malfunction with a
colonization ship’s drive means that it is unable to decelerate, so it tra-
verses intergalactic distances and, due to relativistic time dilation, outlasts
the universe itself. Other instances are Stephen Baxter’s Manifold series
(2000‒2003) and Gregory Benford’s Galactic Center saga (1979‒1999),
which both feature traversals of space and time that defy integration into
any conventionally human scale; striking cinematic evocations of the same
process can be found in Stanley Kubrick’s 2001:  A  Space Odyssey (1968)
and Christopher Nolan’s Interstellar (2014). Other manipulations of scale
occur with respect to what Peter Nicholls terms ‘Big Dumb Objects’ (2000,
p.  13)—​namely, cosmic artefacts that dwarf human technologies and
explanatory frameworks. Prototypical examples can be found in Arthur
C.  Clarke’s Rama novels (1973‒1993) and Larry Niven’s Ringworld series
(1970‒2012), which both focus on the encounter with mysterious alien
creations that are equal to (or vastly greater than) planetary dimensions.

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 77 ]
 

At the opposite extreme, the literature of the extremely small can be

found in those SF texts that explore the implications of nanotechnology.
Richard Fleischer’s film Fantastic Voyage (1966) offers an early (if unreflec-
tive) exploration of this; more recently, texts like Greg Egan’s Blood Music
(1985) and Neal Stephenson’s The Diamond Age (1996) meditate in detail on
the implications of manipulating objects on the nanoscale. Finally, extre-
mal topologies of space are dealt with in narratives like Robert Heinlein’s
‘He Built a Crooked House’ (1941/​2010), Joe Haldeman’s The Forever War
(1974), and Christopher Priest’s The Inverted World (1974); in these works,
the emphasis falls on manipulations of the geometry of space by way of
extra dimensions, wormhole technology, and perceptual alteration. In all
cases, it is evident that SF accents psychological distance by foregrounding
extremes of physical space with respect to both extension and topology.

Temporal Distance

Although Frederic Jameson maintains that ‘SF as a genre has less to do

with time (history, past, future) than with space’ (2005, p.  313), this
statement is true only in relation to the most rarefied interpretation of
‘time’. Starting with works like H.G. Wells’s The Time Machine (1895) and
Tourmalin’s Time Cheques by F. Anstey (1891), extreme temporal displace-
ments have been a feature of SF writing since its inception. Moreover,
the profound cultural impact of special and general relativity in the sub-
sequent two decades went on to make the ‘fourth dimension’ particu-
larly prominent as a topic for literary speculation (Bohn, 2007). As with
space, much of this merely involved exploiting time travel as a useful plot
device; nevertheless, a number of important works interrogate the topic
in a more thoughtful way. Central among these are Heinlein’s stories ‘By
His Bootstraps’ (1941/​2010b) and ‘All You Zombies’ (1959/​2005), which
initiate the trope of having all the principal characters be the same per-
son at different points in their timeline. The psychological potential of this
perspective is developed in narratives like Kurt Vonnegut’s Slaughterhouse
Five (1969), Audrey Niffeneger’s The Time Traveller’s Wife (2004), and Ted
Chiang’s ‘Story of Your Life’ (2013), where the space-​like treatment of time
facilitates a synoptic overview of a human life.
Other reflections on time come by way of the ‘deep future’ subgenre,
which deals with scenarios that are removed from the present in cosmo-
logical timescales. An early example of this is William Hope Hodgson’s
The Night Land (1912), which is set in the last city on Earth after the sun
has gone dark—​a trope that is subsequently developed in narratives like

[ 78 ]  Literature through a Cognitive Lens

 

Arthur C. Clarke’s The City and the Stars (1956) and Greg Bear’s The City at
the End of Time (2008). In the reverse direction, travel to the distant past is
a staple of pulp SF, but more interrogative treatments of humanity’s pro-
genitors and coevals come in William Golding’s The Inheritors (1955) and
Wolfgang Jeschke’s The Last Day of Creation (1982). Finally, perhaps the
most profound meditation on time is to be found on Olaf Stapledon’s Star
Maker (1937), where the narrator enters into a type of visionary ‘super-​
time’ that allows him to survey the creations of the eponymous Star Maker,
ranging from the earliest primitive universes to the extraordinarily com-
plex ‘ultimate cosmos’. As with space, these meditations on time all serve
to maximize the experience of temporal distance on the part of the reader.

Social Distance

SF is, for Adam Roberts, ‘a genre devoted to the encounter with difference’
(2000, p. 118). Inevitably, the most effective vehicle for representing this
difference is the alien—​a being that, in virtue of its different planetary and
evolutionary history, can only with difficulty be reconciled with human
values and behavioural norms. As might be expected, this difference is to
an overwhelming degree aligned with hostility, especially in popular SF;
this is readily seen in H.  G. Wells’s The War of the Worlds (1898), one of
the first convincing attempts to depict extraterrestrials. (Indeed, Wells’s
Martian tripods initiated an enduring association between the alien and
the insectile—​subsequently visible in narratives like Heinlein’s Starship
Troopers [1959], Orson Scott Card’s Ender’s Game [1985], and Ridley Scott’s
Alien [1979]; see Carney [2012] for a discussion of this trope in modernist
literature.) A  more nuanced representation of the alien can be found in
explorations like Ursula K. Le Guin’s Hainish cycle (1966‒2000), where she
explores the outcomes of changing human sexual biology (in The Left Hand
of Darkness [1969], for instance, characters can be both male and female and
are only periodically sexually potent). Similarly, the Lilith’s Brood trilogy by
Octavia E. Butler (2000) focuses on the sexual dynamics of cross-​breeding
humans with morphologically different aliens possessing three sexes.
On a larger scale, Stapledon’s Star Maker delivers an expansive survey
of the different forms of life across several possible universes. This tradi-
tion is continued in Stephen Baxter’s Xeelee series (1994‒2003), which
centres on a multi-​billion-​year war between species composed of baryonic
and non-​baryonic matter. It is, however, in the work of Stanisław Lem that
the encounter with the alien is given its most philosophical treatment. In
novels like Solaris (1961/​2012), His Master’s Voice (1968/​1999), and Fiasco

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 79 ]
 

(1986/​1987), Lem explores what it might mean for humanity to encoun-

ter an intelligence so alien that meaningful communication is impossible
(in Solaris this takes the form of a sentient, planet-​sized ocean that defies
generations of human theorizing). A more recent iteration on the theme
of the radically alien can be found in Peter Watts’s Blindsight (2006), a
first-​contact novel in which technologically sophisticated, non-​humanoid
aliens are discovered to be entirely without consciousness (i.e. ‘zombies’
in the philosophical sense). Finally, in Philip K. Dick’s The Three Stigmata of
Palmer Eldritch (1965), VALIS (1981b), and The Divine Invasion (1981a), the
alien is conflated with the divine by way of a gnostic cosmology in which
reality is revealed to be a corrupted simulacrum. As will by now have been
anticipated, these examples are given to illustrate the extent to which SF
amplifies social distance by engaging with forms of agency that escape easy
anthropomorphic categories.

Hypothetical Distance

For Umberto Eco, the narrative text is ‘a machine for producing possible
worlds’ (1984, p. 246)—​a claim that is particularly true of SF, which explic-
itly thematizes the counterfactual. The most obvious way in which this
occurs is through the alternative history genre. As shown by psychological
research on the ‘availability heuristic’, familiar events are assessed as being
more likely to occur than unfamiliar ones (Tversky and Kahneman, 1973;
Wänke, Schwarz, and Bless, 1995); thus, any exploration of alternative his-
torical timelines is at the same time a traversal of probabilistic distance.
A typical early example of such an exploration is Castello Holford’s Aristopia
(1895), which imagines a settlement of North America that proceeds along
socialistic lines. Subsequent developments of the genre generally retain this
pattern of focusing on counterfactual outcomes to world-​historical events,
with World War II inevitably attracting a large number of treatments. Most
of these are merely diverting, but narratives like Philip K. Dick’s The Man
in the High Castle (1962), Christopher Priest’s The Separation (2002), and
Michael Chabon’s The Yiddish Policeman’s Union (2007) offer more sus-
taining meditations on the role of chance in human affairs. Equally, Kim
Stanley Robinson’s The Years of Rice and Salt (2002) delivers a thoughtful
counter-​history of a world in which European civilization has been aborted
by a more virulent version of the Black Death.
More speculative use of the alternative history idea is to be found in
the ‘competing timeline’ theme, where narrative conflict is predicated on
the need to preserve or alter a given timeline in the face of counter-​action

[ 80 ]  Literature through a Cognitive Lens

 

by opponents. Isaac Asimov is one of the first to use this trope, in The End
of Eternity (1955), which features a corps of ‘guardians’ who intervene
in history to minimize human suffering. Other explorations of this idea
include H. Beam Piper’s Paratime (2001) series and, more recently, Charles
Stross’s Palimpsest (2011). Harry Harrison’s West of Eden (1984) and Harry
Turtledove’s A Different Flesh (1988), for their part, are representative
of generic variations wherein the extinctions of prehistory never took
place; on a bigger scale, Greg Egan’s Orthogonal series (2011‒2013) and
Bob Shaw’s Land and Overland trilogy (1986‒1989) imagine universes in
which physical and dimensional constants are different from their present
values. Thus, the recurring exploration of counterfactuals in SF serves to
foreground remote possibilities to a degree that is rarely visible in non-​
speculative fiction.

THREE HYPOTHESES CONCERNING SCIENCE FICTION’S

MANIPULATION OF CONSTRUAL LEVEL

By now, it should be clear that SF systematically maximizes all four dimen-

sions of psychological distance. It is difficult, for instance, to think of a text
that maximizes none of spatial, temporal, social, or probabilistic distance,
and the most representative texts (like Stapledon’s Star Maker) often
maximize all of them. (In fact, the CLT literature predicts that distance
maximization on one dimension leads to greater expectations of distance
on the other three—​see Bar-​Anan, Liberman, Trope, and Algom [2007]
and Liberman and Förster [2011].) If this is so, the question becomes what
impact these manipulations have upon readers’ reception of SF literature.
Here, I  will engage with this query by formulating three CLT-​informed
hypotheses about how SF is likely to affect readers and showing that three
corresponding problems in the critical literature on SF are resolved by
viewing them through this lens. This, certainly, does not exhaust what
one could say about SF using insights from CLT; nevertheless, I  submit
that part of the value of a new approach lies in its being able to resolve or
overturn long-​standing problems in the study of a topic.
My first hypothesis is that distance maximization should lead to greater
expectations of purposiveness in SF texts. Specifically, CLT suggests that
distant actions will cue greater preoccupation with motive over implemen-
tation relative to closer actions. This is shown to be the case for tempo-
ral distance in Liberman and Trope (1998), where events like ‘cleaning the
house’ were described by participants as indicating general character traits
like ‘cleanliness’ when projected into the distant future, and described in

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 81 ]
 

terms of practical tasks (‘vacuuming the floor’) when located in the near
future. Similar results are obtained for spatial distance by Fujita and col-
leagues (2006), for probabilistic distance by Wakslak, Trope, Liberman,
and Alony (2006), and for social distance by Liviatan, Trope, and Liberman
(2008). In all cases, the point to retain is that experimental amplifications
of distance had the effect of dampening speculation about the contextual
determinants of an action and emphasizing its goal-​directedness.
This observation is important for SF because it shows how the genre
enacts its heroic character. The development of ‘literary’ SF left behind the
naïve adventurism of the early pulp magazines, but the genre remains, in
Ursula K. Le Guin’s words, ‘a modern descendent of the epic’ (1980, p. 92).
As such, it is concerned with a cognitive stance that is proactive rather than
reactive, and which subordinates proximal actions to remote, even tran-
scendent, goals (Kreuziger, 1986; Cowan, 2010); in Elana Gomel’s words,
it ‘defamiliarizes humanism and points the way to transcendence’ (2014,
p. 32). Necessarily, this programme of representation will be facilitated by
using manipulations of distance that enhance the expectation and recog-
nition of self-​motivated agents who are not imprisoned in the quotidian.
Indeed, viewing SF through this lens explains the puzzling lack of
well-​drawn characters in the genre. As noted by Gwyneth Jones, SF has
‘little space for deep and studied characterisation’, preferring instead to
treat characters as ‘pieces of equipment’ (1999, p. 5). For critics like Brian
Attebery, this is explained as a reflexive attempt to foreground ‘the rela-
tionship between character as imitated person and character as story func-
tion’ (1992, p. 73). Critics like Scott Sanders, on the other hand, interpret it
as a deliberate critique of the inheritance of ‘complex, autonomous, unique
individuals—​the idea at the heart of the Continental and Anglo-​American
novel’ (2014, p. 133). However, though both explanations touch off impor-
tant points, they fail to explain why readers should be so forgiving of the
dearth of compelling characterization—​elsewhere, the very motor of audi-
ence engagement with fiction (Keen, 2007). Here, CLT suggests that read-
ers tolerate formulaic characterization because that is exactly what they
have been primed to expect. Specifically, CLT shows that maximizing dis-
tance de-​emphasizes the concrete, singular details that make the evoca-
tion of a character lifelike; instead, it produces a preoccupation with the
abstract idea of agency—​and from this follows the formal interchangeabil-
ity of many SF protagonists.
My second hypothesis is that maximizing distance in SF narratives
should produce an exaggerated expectation of moral engagement on the
part of SF readers. Results from Eyal, Liberman, and Trope (2008) and Eyal
and Liberman (2012) show that moral aversion and moral approbation are

[ 82 ]  Literature through a Cognitive Lens

 

more intense when applied to distant objects. This is theorized to be the

case because high-​level construals leave out concrete details that confuse
or mitigate the application of abstract moral principles. For instance, in
Eyal and colleagues (2008), vignettes about cleaning the house with the
national flag, sibling incest, and eating the family dog were collectively
judged as being more morally offensive when performed a year as opposed
to a day in the future. These results are particularly interesting when com-
pared with CLT discussions of empathy. Although empathy and morality
usually align, the formal character of morality means that it can sometimes
deliver prescriptions that run counter to empathic inclination—​such as
when the moral demands of fairness run against our emotional disposition
to favour family and friends. CLT highlights this mismatch by showing how
empathy, in contrast with morality, is cued by concrete, nearby constru-
als (Woltin et al., 2011; Luguri, Napier, and Dovidio, 2012). Thus, different
construals can place different accents on how we relate to the needs of oth-
ers, and these accents can sometimes be in conflict.
For these reasons, CLT leads us to expect that distance maximization
in SF narratives should lead to more concern with morality and less con-
cern with empathy. And this is precisely what we see—​most visibly in the
SF preoccupation with utopian (or dystopian) alternatives to present-​day
social arrangements (Suvin, 1979; Parrinder, 2000; Jameson, 2005; Reeve-​
Tucker and Waddell, 2013). As Tom Moylan notes, ‘the axis of the utopian-​
dystopian imaginary’ (2000, p. 9) is a core theme of SF and is expressed
through ‘the conflict between the originary world and the utopian society
opposed to it so that the process of social change is more directly articu-
lated’ (2000, p. 36). The key emphasis here is on the ideological project of
utopianism, which blends moral commitment with the praxis that seeks to
make the utopia a reality. By maximizing psychological distance, SF thus
makes this moral impulse more salient to readers by negating the emo-
tional immediacy of empathic identification in favour of long-​term, large-​
scale moral and ethical planning—​leading, ultimately, to what Darko Suvin
terms ‘estrangement arising out of an alternative historical hypothesis’
(1979, p. 49).
The final hypothesis I wish to discuss concerns linguistic register in SF.
Specifically, CLT experiments show that formal speech registers are acti-
vated by psychological distance. For instance, experiments on politeness
show that participants assessed temporally distant requests as more polite
than identical temporally proximal ones, and created more polite requests
when asked to write them for a far-​future recipient (Stephan, Liberman,
and Trope, 2010). These results are partially explained by the effect of spa-
tial distance upon emotionality, insofar as CLT shows that events moving

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 83 ]
 

away from the subject are imagined as having less emotional valence than
events moving towards (Davis, Gross, and Ochsner, 2011; Hart, Stedman,
and McComas, 2015). Given that a core function of politeness is to drain
social encounters of arousing emotions by way of ritual forms of deference
(Brown and Levinson, 1987; Keltner, Young, and Buswell, 1997), it thus fol-
lows that low levels of emotionality will be signalled by a polite register. In
fact, this is likely to be true of all registers, sociolects, and jargons that are
concerned with the communication of information over emotional stance,
even if CLT proponents have yet to test this experimentally.
In the context of SF, these results explain the notoriously formulaic style
of most SF writing. As Peter Stockwell notes, ‘the vast majority of science
fiction published over the last forty years has retained the conventional
pattern of fantastic content with prosaic stylistic delivery’ (2014, p. 102).
What is less clear is why this might be the case, given that innovators in SF
are, presumably, no less sensitive to the aesthetic opportunities of styliza-
tion than writers in other genres. Adopting a CLT perspective resolves this
issue by showing that, by eschewing the somatic and affective affordances
implicit in literary style (M. Burke, 2010; Bolens, 2012), SF satisfies the
refusal of emotionality associated with extremes of psychological distance.
Equally, the preponderance of technical neologism in the genre evinces a
move away from an affective register and into a pseudo-​objective scien-
tific one. This is not to discount the claim that the use of scientific neolo-
gism in SF world-​building forms ‘part of the establishment of plausibility
and verisimilitude’ (Stockwell, 2014, p. 117); instead, it is to suggest that
this world-​building is itself informed by the psychological consequences of
maximizing psychological distance. Doubtless, for critics who identify the
literary quality of a text with its capacity to evoke the qualitative aspects
of experience (Pilkington, 1996; Fludernik, 2004; Herman, 2011; Lodge,
2012), this flattening of stylistically mediated affective impact may well
count against SF’s claim to ‘literary’ status. My view is that when a theory
devalues the concrete experiences of one group of readers on the basis of
an a priori prescription, it is the theory that should be rejected and not the
readers.
If one wished, one could discuss the effects of distance maximization
in SF at far greater length: the CLT literature, for instance, offers findings
on topics like categorization, self-​control, psychological coherence, and
creativity that can be directly integrated into a critical discussion on SF.
However, even apart from limitations of space, the result would only be
to confirm the results established so far. Specifically, the three explored
effects—​ purposiveness, moral engagement, and affect-​ flattening—​ all
unite in a view of SF that sees it as consistent with inflated depictions

[ 84 ]  Literature through a Cognitive Lens

 

of the scope of human action (in texts) and exaggerated expectations of

the efficacy of human action (in readers). That is, SF cues a set of cogni-
tive dispositions that are instrumental in establishing collective inten-
tionality on scales that transcend individual agency. In this, the genre
could not contrast more with traditional literary forms, where the world
is treated as an immutable framework in which human beings—​usually
tragically—​pursue their destiny. Instead, my CLT-​motivated claim is that
SF overturns the fatalism implicit in such a position by prompting highly
abstract modes of cognition in authors and readers; these transform
metaphysical absolutes into formal variables that can be manipulated by
human beings and human technologies. As such, the genre is continu-
ous with the fundaments of modernity, which match an unprecedented
increase in human productive capacity with processes of secular disen-
chantment. Indeed, this is probably why SF has consistently proved so
attractive a target for Marxist criticism (Fekete, 1988; Jameson, 2005;
Bould and Miéville, 2009), whose protagonists place a similar emphasis
on the capacity of collective human praxis to overturn the wreckage of
history. However, rather than collapse SF into any given political faction-
alism, it is more useful to see it as a cultural amplification of the principle
of formal planning that underwrites all forms of politics. As such, SF is
the natural literature of those supra-​individual entities like the state, the
corporation, and the confederacy that subordinate the human lifeworld
to the forces of rational administration.

CONCLUSIONS

This chapter has attempted to characterize the defining features of the SF

genre using CLT. Its core results are (1) that SF maximizes psychological
distance in the areas of space, time, social relations, and hypotheticality;
and (2)  that the effect of doing this is to foreground, on a cosmic scale,
the transformative power of human agency and human technologies. As
such, the outcome of SF is to diminish readerly preoccupation with the per-
sonal and cultivate instead a disposition towards the remote, the abstract,
and the principled. SF is, in this sense, the genre that most explicitly
infuses popular representations with intellectual (as opposed to affective or
somatic) content—​hence its affinity with rationalist or progressive ideolo-
gies. By way of conclusion, I will now touch off two important issues that
are raised by the preceding discussion.
The first issue is methodological and concerns the representativeness
of my conclusions. A recurring objection to the use of psychological and

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 85 ]
 

cognitive approaches in literary studies is that, as ‘reductionist material-

isms’, they are insufficiently nuanced to ‘explain the textual particulars
that make works memorable’ (Gilmore, 2012, p. 313). Even if this objec-
tion were valid for cognitive readings of individual texts (it isn’t), large-​
scale surveys of the type conducted here would remain unaffected by it.
As products of the human mind, literary texts inherit human variability;
thus, they do not usually reward being shoehorned into rigid typologies. At
the same time, like the human mind, they exhibit regularities that emerge
at the statistical level—​and it is these that are of interest to genre theo-
rists (Moretti, 2004). Counterexamples can no doubt be cited against any
generic definition, and this is certainly true with respect to the SF defini-
tions offered here. Nevertheless, as John Holloway notes, to ‘promulgate
a law that literary illumination is always promoted by being sensitive to
shades and nuances, and never by assimilating respects of sameness which
they overlie and conceal, would be the most peremptory and sweeping
of all acts of standardizing assimilation’ (1979, p.  108). Ultimately, the
account of SF developed in the preceding pages is meant to be representa-
tive rather than prescriptive, and for this ambition to be successful, it need
only explain the aggregate features of the genre.
My second point concerns the value of CLT in highlighting how SF relates
to its contemporary genres and aesthetics. In this regard, CLT is useful in
showing how SF relates to modernism—​a topic that has repeatedly come
up in critical discussions (McHale, 1991; Wegner, 2007; March-​Russell,
2015). This becomes possible when we remember that CLT is concerned not
solely with the effects of maximizing distance on construal level but also
with the effects of maximizing construal level on distance. On this issue,
existing research suggests that ‘as psychological distance increases, con-
struals would become more abstract, and as level of abstraction increases,
so too would the psychological distances people envisage’ (Trope and
Liberman, 2010, p. 440). It is striking to note the extent to which these
inverse operations characterize the difference between SF and modernism.
Where SF maximizes distance by way of fantastic content, the modernist
emphasis on formal, mathematical, fragmentary, and kinetic modes of rep-
resentation (Marinetti, 1909; Lewis et al., 1914; Malevich, 1916; Gropius,
1919) often serves to maximize abstraction. One outcome of this is that
many of the affordances present in the content of SF are yielded by the
form of modernist art. Thus, where SF makes everyday social and psy-
chological arrangements abstract by foregrounding distance, modernism
makes these arrangements seem distant by foregrounding abstraction. The
result, in both cases, is to alienate audiences from the quotidian by subject-
ing the latter to exaggerated forms of logico-​conceptual rearrangement. As

[ 86 ]  Literature through a Cognitive Lens

 

I can do little more than register this claim here, I will only say that I intend
to pursue it in a more developed way in future.
To close, my hope is that this chapter has made evident the value of CLT
as a tool in literary studies. Although many of the predictions of CLT will
have been intuitively obvious to literary scholars, others will not—​and in
any event, the fact that these predictions have been experimentally tested
means that they help locate critical speculation on solid ground. To be
sure, this does not mean that the results of CLT (and their application in
literary studies) are beyond dispute. Instead, it is merely to recognize that,
if critics wish to make claims about how readers psychologically respond
to textual manipulations, empirically derived models of cognition are by
a long measure a better place to start than uninformed intuition or dis-
credited paradigms (Currie, 1995; Freeland, 2004). Moreover, considering
that the two central concepts of CLT—​distance and construal level—​are
multimodal in character, its applications are far from exhausted by the
discussion here. Given this, the next step will come with exploiting CLT to
make new discoveries about other areas of literature and culture.

REFERENCES

Anderson, P. (1970). Tau zero. New York: Doubleday.

Anstey, F. (1891). Tourmalin’s time cheques. Bristol: J. W. Arrowsmith.
Asimov, I. (1955). The end of eternity. New York: Doubleday.
Attebery, B. (1992). Strategies of fantasy. Bloomington, IN: Indiana University Press.
Bar-​Anan, Y., Liberman, N., Trope, Y., and Algom, D. (2007). Automatic processing of
psychological distance:  Evidence from a Stroop task. Journal of Experimental
Psychology, 136(4), 610–​622.
Bear, G. (2008). City at the end of time. London: Orion.
Bohn, W. (2007). Writing the fourth dimension. Comparative Critical Studies, 4(1),
121–​138.
Bolens, G. (2012). The style of gestures: Embodiment and cognition in literary narrative.
Baltimore, MD: Johns Hopkins University Press.
Bould, M., and Miéville, C. (2009). Red planets: Marxism and science fiction. Middletown,
CT: Wesleyan University Press.
Brown, P., and Levinson, S. C. (1987). Politeness:  Some universals in language usage.
Cambridge: Cambridge University Press.
Bruehlman-​Senecal, E., and Ayduk, O. (2015). This too shall pass: Temporal distance
and the regulation of emotional distress. Journal of Personality and Social
Psychology, 108(2), 356–​375.
Burke, E. (1909). A philosophical inquiry into the origin of our ideas of the sublime. In
The works of Edmund Burke (pp. 49–​178). London: George Bell and Sons.
Burke, M. (2010). Literary reading, cognition and emotion: An exploration of the oceanic
mind. London: Routledge.
Butler, O. E. (2000). Lilith’s brood. New York: Grand Central.
Card, O. S. (1985). Ender’s game. New York: Tor.

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 87 ]
 

Carney, J. (2012). The buzzing of B: The subject as insect in Beckett’s Molloy. In J. Carney,
L. Madden, M. O’Sullivan, and K. White (Eds.), Beckett re-​membered: After the
centenary (pp. 228–​241). Newcastle-​upon-​Tyne: Cambridge Scholars.
Carney, J. (2014). Supernatural intuitions and classic detective fiction:  A  cognitivist
appraisal. Style, 48(3), 203–​218.
Chabon, M. (2007). The Yiddish policeman’s union. New York: Harper Collins.
Chiang, T. (2013). Story of your life. In Story of your life and others (pp. 91–​146).
New York: Tor.
Clarke, A. C. (1956). The city and the stars. London: Frederick Muller.
Cowan, D. E. (2010). Sacred space: The quest for transcendence in science fiction film and
television. Waco, TX: Baylor University Press.
Currie, G. (1995). Image and mind:  Film, philosophy, and cognitive science. Cambridge:
Cambridge University Press.
Darwin, C. (1871). The descent of man. London: D. Appleton and Co.
Davis, J. I., Gross, J. J., and Ochsner, K. N. (2011). Psychological distance and emo-
tional experience: What you see is what you get. Emotion, 11(2), 438–​444.
Dębek, M., and Bożena, J.-​D. (2013). Attitude towards architectural objects and the
Construal Level Theory (CLT) by Liberman and Trope. Polish Journal of Applied
Psychology, 11(4), 81–​96.
Dhar, R., and Kim, E. Y. (2007). Seeing the forest or the trees: Implications of con-
strual level theory for consumer choice. Journal of Consumer Psychology, 17(2),
96–​100.
Dick, P. K. (1962). The man in the high castle. New York: Putnam.
Dick, P. K. (1965). The three stigmata of Palmer Eldritch. New York: Doubleday.
Dick, P. K. (1981a). The divine invasion. New York: Simon and Schuster.
Dick, P. K. (1981b). VALIS. New York: Bantam.
Eco, U. (1984). The role of the reader: Explorations in the semiotics of texts. Bloomington,
IN: Indiana University Press.
Egan, G. (1985). Blood music. Gettysburg, PA: Arbor House.
Emmott, C., Sanford, A. J., and Alexander, M. (2013). Rhetorical control of readers’
attention: Psychological and stylistic perspectives on foreground and background
in narrative. In L. Bernaerts, D. De Geest, L. Herman, and B. Vervaeck (Eds.),
Stories and minds: Cognitive approaches to literary narrative (pp. 39–​58). Lincoln,
NE: University of Nebraska Press.
Evans, J., and Stanovich, K. E. (2013). Dual-​process theories of higher cognition:
Advancing the debate. Perspectives on Psychological Science, 8(3), 223–​241.
Eyal, T., and Liberman, N. (2012). Morality and psychological distance: A construal level
theory perspective. In M. Mikulincer and P. R. Shaver (Eds.), The social psychol-
ogy of morality: Exploring the causes of good and evil (pp. 185–​202). Washington,
DC: American Psychological Association.
Eyal, T., Liberman, N., and Trope, Y. (2008). Judging near and distant virtue and vice.
Journal of Experimental Social Psychology, 44(4), 1204–​1209.
Fekete, J. (1988). The stimulations of simulations: Five theses on science fiction and
Marxism. Science Fiction Studies, 15(3), 312–​323.
Fleischer, R. (1966). Fantastic voyage [Motion picture]. United States:  Twentieth
Century Fox.
Fludernik, M. (2004). Towards a ‘natural’ narratology. New York: Routledge.
Freeland, C. (2004). Explaining the uncanny in The double life of Véronique. In S. J.
Schnieder (Ed.), Horror film and psychoanalysis: Freud’s worst nightmare (pp. 87–​
105). Cambridge: Cambridge University Press.

[ 88 ]  Literature through a Cognitive Lens

 

Fujita, K., Henderson, M. D., Eng, J., Trope, Y., and Liberman, N. (2006). Spatial dis-
tance and mental construal of social events. Psychological Science, 17(4), 278–​280.
Gilmore, P. (2012). Reading minds in the nineteenth century. In R. Castronovo (Ed.),
The Oxford handbook of nineteenth-​century American literature (pp. 327–​344).
Oxford: Oxford University Press.
Golding, W. (1955). The inheritors. London: Faber and Faber.
Gomel, E. (2014). Science fiction, alien encounters, and the ethics of posthumanism: Beyond
the golden rule. Palgrave Macmillan: Basingstoke.
Gropius, W. (1919). Bauhaus manifesto and programme. Weimar. Reproduced as
Gropius, W.  (1998). Manifesto of the Bauhaus, April 1919. In V. Kollocotroni,
J. Goldman, and O. Taxidou (Eds.), Modernism: An anthology of sources and docu-
ments (p. 301). Edinburgh: Edinburgh University Press.
Haldeman, J. (1974). The forever war. New York: St. Martin’s.
Hamilton, C. A., and Schneider, R. (2002). From Iser to Turner and beyond: Reception
theory meets cognitive criticism. Style, 36(4), 640–​658.
Harrison, H. (1984). West of Eden. New York: Bantam.
Hart, P. S., Stedman, R. C., and McComas, K. A. (2015). How the physical proximity of
climate mitigation projects influences the relationship between affect and public
support. Journal of Environmental Psychology, 43, 196–​202.
Heinlein, R. A. (1959). Starship troopers. New York: G. P. Putnam’s Sons.
Heinlein, R. A. (2005). All you zombies. In A. Wheeler (Ed.), Off the main sequence: The
other science fiction stories of Robert A. Heinlein (pp. 725–​738). New York: Science
Fiction Book Club.
Heinlein, R. A. (2010a). And he built a crooked house. In R. James and W. H. Patterson
(Eds.), Creating a genre: The short fiction of Robert A. Heinlein 1939‒41 (pp. 281–​
304). Houston, TX: Virginia Edition.
Heinlein, R. A. (2010b). By his bootstraps. In R. James and W. H. Patterson (Eds.),
Creating a genre:  The short fiction of Robert A.  Heinlein 1939‒41 (pp. 333–​386).
Houston, TX: Virginia.
Herman, D. (2011). Basic elements of narrative. Hoboken, NJ: John Wiley and Sons.
Hogan, P. C. (2013). Ulysses and the poetics of cognition. London: Taylor and Francis.
Holford, C. (1895). Aristopia: A romance-​history of the new world. Boston: Arena.
Hollinger, V. (1999). Contemporary trends in science fiction criticism, 1980‒1999.
Science Fiction Studies, 26(2), 232–​262.
Holloway, J. (1979). Narrative and structure: Exploratory essays. Cambridge: Cambridge
University Press.
Hope Hodgson, W. (1912). The night land. London: Eveleigh Nash.
Jameson, F. (2005). Archaeologies of the future: The desire called utopia and other science
fictions. London: Verso.
Jeschke, W. (1982). The last day of creation (G. Mander, Trans.). New York: St. Martin’s.
Jones, G. A. (1999). Deconstructing the starships:  Science, fiction and reality.
Liverpool: Liverpool University Press.
Kahneman, D. (2012). Thinking, fast and slow. London: Penguin.
Keen, S. (2007). Empathy and the novel. Oxford: Oxford University Press.
Keltner, D., Young, R., and Buswell, B. N. (1997). Appeasement in human emotion,
personality, and social practice. Aggressive Behavior, 23, 359–​374.
Kreuziger, F. A. (1986). The religion of science fiction. Bowling Green, OH: Bowling Green
State University Popular Press.
Kubrick, S. (1968). 2001: A space odyssey [Motion picture]. United Kingdom: MGM.
Le Guin, U. K. (1969). The left hand of darkness. New York: Ace.

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 89 ]
 

Le Guin, U. K. (1980). The language of the night:  Essays on fantasy and science fiction.
New York: Ultramarine.
Lem, S. (1987). Fiasco (M. Kandel, Trans.). New York: Harcourt Brace.
Lem, S. (1999). His master’s voice (M. Kandel, Trans.). Evanston, IL:  Northwestern
University Press.
Lem, S. (2012). Solaris (B. Jonston, Trans.). Los Angeles, CA: Premier Digital.
Lewis, W., Aldington, R., Arbuthnot, M., Atkinson, L., Dissmore, J., Gaudier-​Brzeska,
H., et al. (1914). Vorticist manifesto. BLAST, 1. No pagination.
Liberman, N., and Förster, J. (2011). Estimates of social distance: A construal level the-
ory perspective. In T. Schubert and A. Maass (Eds.), Spatial dimensions of social
thought (pp. 109–​128). Berlin: Mouton de Gruyter.
Liberman, N., and Trope, Y. (1998). The role of feasibility and desirability consider-
ations in near and distant future decisions: A test of temporal construal theory.
Journal of Personality and Social Psychology, 75(1), 5–​18.
Liberman, N., and Trope, Y. (2009). The psychology of transcending the here and how.
Science, 322(11), 1201–​1206.
Liviatan, I., Trope, Y., and Liberman, N. (2008). Interpersonal similarity as a social
distance dimension:  Implications for perception of others’ actions. Journal of
Experimental Social Psychology, 44(5), 1256–​1269.
Lodge, D. (2012). Consciousness and the novel. New York: Random House.
Luca, C. (2015). The conceptual effects of verticality in design. In R. Batra, C. Seifert,
and D. Brei (Eds.), The psychology of design: Creating consumer appeal (pp. 40–​52).
London: Routledge.
Luguri, J. B., Napier, J. L., and Dovidio, J. F. (2012). Reconstruing intolerance: Abstract
thinking reduces conservatives’ prejudice against nonnormative groups.
Psychological Science, 23(7), 756–​763.
Malevich, K. (1916). From cubism and futurism to suprematism. Moscow.
March-​Russell, P. (2015). Modernism and science fiction. London: Palgrave Macmillan.
Marinetti, F. T. (1909, 20 February). The founding and manifesto of futurism. Le figaro.
Reproduced as Marinetti, F.  T. (1998). The founding and manifesto of futur-
ism 1909 /​The variety theatre 1913. In V. Kollocotroni, J. Goldman, and O.
Taxidou (Eds.), Modernism: An anthology of sources and documents (pp. 249–​256).
Edinburgh: Edinburgh University Press.
McHale, B. (1991). POSTcyberMODERNpunkISM. In L. McCaffery (Ed.), Storming the
reality studio (pp. 308–​323). Durham, NC: Duke University Press.
Milner, A. (2012). Locating science fiction. Liverpool: Liverpool University Press.
Moretti, F. (2004). Graphs, maps, trees: Abstract models for literary history—​3. New
Left Review, 28, 43–​63.
Moylan, T. (2000). Scraps of the untainted sky:  Science fiction, utopia, dystopia.
Oxford: Westview.
Nicholls, P. (2000). Big dumb objects and cosmic enigmas: The love affair between sci-
ence fiction and the transcendental. In G. Westfahl (Ed.), Space and beyond: The
frontier theme in science fiction (pp. 11–​24). Westport, CT: Greenwood.
Niffeneger, A. (2004). The time traveller’s wife. Orlando, FL: Harcourt.
Nolan, C. (2014). Interstellar [Motion picture]. United States: Paramount Pictures.
Parrinder, P. (2000). Introduction. In P. Parrinder (Ed.), Learning from other worlds (pp.
1–​18). Liverpool: Liverpool University Press.
Pilkington, A. (1996). Introduction: Relevance theory and literary style. Language and
Literature, 5(3), 157–​162.
Piper, H. B. (2001). The complete paratime. New York: Ace.

[ 90 ]  Literature through a Cognitive Lens

 

Priest, C. (1974). The inverted world. London: Faber and Faber.

Priest, C. (2002). The separation. New York: Scribners.
Reeve-​Tucker, A., and Waddell, N. (2013). Utopianism, modernism, and literature in the
twentieth century. London: Palgrave Macmillan.
Roberts, A. (2000). Science fiction. London: Routledge.
Robinson, K. S. (2002). The years of rice and salt. London: Harper Collins.
Robu, C. (1988). A key to science fiction: The sublime. Foundation, 42, 21–​37.
Sanders, S. (2014). The disappearance of character. In P. Parrinder (Ed.), Science fic-
tion: A critical guide (pp. 131–​147). London: Routledge.
Scott, R. (1979). Alien [Motion picture]. London: Twentieth Century Fox.
Skinner, B. F. (1976). Beyond freedom and dignity. Harmondsworth: Penguin.
Staddon, J. (2014). The new behaviorism (2nd ed.). New York: Psychology Press.
Stapledon, O. (1937). Star maker. London: Methuen.
Stephan, E., Liberman, N., and Trope, Y. (2010). Politeness and psychological dis-
tance: A construal level perspective. Journal of Personality and Social Psychology,
98(2), 268–​280.
Stephenson, N. (1996). The diamond age. London: Penguin.
Stockwell, P. (2014). The poetics of science fiction. Oxford: Routledge.
Stross, C. (2011). Palimpsest. Burton, MI: Subterranean.
Suvin, D. (1972). On the poetics of the science fiction genre. College English, 34(3),
372–​382.
Suvin, D. (1979). Metamorphoses of science fiction: On the poetics and history of a literary
genre. New Haven, CT: Yale University Press.
Todorov, A., Goren, A., and Trope, Y. (2007). Probability as a psychological dis-
tance:  Construal and preferences. Journal of Experimental Social Psychology,
43(3), 473–​482.
Trope, Y., and Liberman, N. (2000). Temporal construal and time-​dependent changes
in preference. Journal of Personality and Social Psychology, 79(6), 876–​889.
Trope, Y., and Liberman, N. (2010). Construal-​level theory of psychological distance.
Psychological Review, 117(2), 440–​463.
Troscianko, E. T. (2014). Kafka’s cognitive realism. New York: Routledge.
Turtledove, H. (1988). A different flesh. New York: Congdon and Weed.
Tversky, A., and Kahneman, D. (1973). Availability: A heuristic for judging frequency
and probability. Cognitive Psychology, 5(2), 207–​232.
Verne, J. (2001). Twenty thousand leagues under the seas (W. Butcher, Trans.).
Oxford: Oxford University Press.
Verne, J. (2008). Journey to the centre of the earth (W. Butcher, Trans.). Oxford: Oxford
University Press.
Vonnegut, K. (1969). Slaughterhouse Five. New York: Delacorte.
Wakslak, C. J., Trope, Y., Liberman, N., and Alony, R. (2006). Seeing the forest when
entry is unlikely: Probability and the mental representation of events. Journal of
Experimental Psychology, 135(4), 641–​653.
Wänke, M., Schwarz, N., and Bless, H. (1995). The availability heuristic revisited:
Experienced ease of retrieval in mundane frequency estimates. Acta Psychologica,
89, 83–​90.
Watson, J. B. (1913). Psychology as the behaviorist views it. Psychological Review, 20,
158–​177.
Watts, P. (2006). Blindsight. New York: Tor.
Wegner, P. E. (2007). Jameson’s modernisms; Or, the desire called utopia. Diacritics,
37(4), 2–​20.

C o n s t r ua l L e v e l a n d S c i e n c e Fi c t i o n   [ 91 ]
 

Wells, H. G. (1895). The time machine. London: Heinemann.

Wells, H. G. (1898). War of the worlds. London: William Heinemann.
Wilson, E. O. (2000). Sociobiology:  The new synthesis. Cambridge, MA:  Harvard
University Press.
Woltin, K. A., Corneille, O., Yzerbyt, V. Y., Förster, J., Luguri, J. B., Napier, J. L., et al.
(2011). Narrowing down to open up for other people’s concerns: Empathic con-
cern can be enhanced by inducing detailed processing. Journal of Experimental
Social Psychology, 23(7), 418–​424.
Zunshine, L. (2006). Why we read fiction:  Theory of mind and the novel. Columbus,
OH: Ohio State University Press.

[ 92 ]  Literature through a Cognitive Lens

 

CHAPTER 5

Patterns of Thought
Narrative and Verse
BRIAN BOYD

C ognitive neuroscientists have recently emphasized that minds work

as pattern extractors—​that, in Gerald Edelman’s words, ‘the “pri-
mary mode” of thought is “pattern recognition” ’ (2006, p. 103)—​and that
human minds in particular are hierarchical pattern extractors (Kurzweil,
2012)  with a special appetite for novel pattern (Bor, 2012). In my own
recent work, I suggest that we need to see art as cognitive play with pat-
tern. In On the Origin of Stories (2009), I focus especially on narrative and
show how minds have evolved to understand events in terms of hierarchies
of pattern, from micro-​actions and micro-​responses to goals and intentions
and all the way to stories, genres, and history. In its companion piece, Why
Lyrics Last (2012), I focus on verse, especially lyric in its strictest sense, as
verse without narrative. Lyric plays with pattern in very different ways,
both more conventional (less natural, in for instance rhyme and stanza
structure) and more exploratory (more effortful, in for instance imagery
and word choice).
Here I will develop the difference between these two books to contrast
narrative and verse, literature’s two longest and richest strands. Often,
especially in the past, they intertwine. Sometimes, though, verse seeks an
appeal without narrative, in lyrics at their purest. What unites and what
distinguishes narrative and verse? What can we learn about them and
about the mind by disentangling them?
 

Shakespeare offers an especially fruitful test case. In much of his work,

he weaves together narrative and verse more memorably than any other
writer, while his Sonnets form the most successful collection of Western
literary lyrics. The Scottish poet Don Paterson, in his ebullient recent book
on the Sonnets, asserts that they ‘have to be read as a narrative of the prog-
ress of love’ (Paterson, 2010, p. 14). I argue, on the contrary, that we need
to read Shakespeare’s sonnets as lyrics, as verse without narrative, and to
appreciate the deep differences between, say, the love lyricism in the dra-
matic narrative of Twelfth Night and the love lyrics in his Sonnets.
I do not intend to highlight the differences between narrative and lyric
by contrasting texts with almost nothing in common. Were I to choose such
a contrast, even within high literature, even roughly contemporaneous—​
Pride and Prejudice versus ‘To Autumn’, for instance—​some might object
that I had begged the question by not considering cases where narrative and
lyric might seem harder to distinguish. I therefore choose two Shakespeare
works where the differences may at first appear slight but where the
kinds of attention and response invited are, as the writer himself clearly
knew, utterly distinct. I  will contrast the almost automatic convergence
of patterns in fiction, or narrative more generally, and the compounding
of patterns upon patterns—​patterns athwart or concealed behind other
patterns, patterns we have to attend very closely to appreciate—​in verse,
especially in lyric.
The world swarms with information, which animals and even plants can
interpret in order to respond appropriately to threats and opportunities in
their environment. But analysing information is costly in time and effort.
Slowly evolved modes of pattern recognition—​like, in vertebrate visual
systems, recognizing outlines, shapes, movement, and direction—​reduce
the cost of interpreting information and allow animals to respond to it
quickly, in real time. Brains therefore operate, observes Edelman, ‘not by
logic but rather by pattern recognition’ (2006, p. 58).
We humans depend for our survival on our superior handling of infor-
mation, and hence on our capacity to recognize—​and to find new ways of
recognizing—​pattern. We therefore have a natural appetite for pattern.
Neuroscientist Daniel Bor has recently stressed that because of the differ-
ence pattern detection makes to information processing, we humans ‘are
alone in the animal kingdom in just how aggressively we constantly search
for patterns, and even in how they may be a source of so much pleasure’
(2012, p. 149).
Cognition evolved to guide action. At its everyday richest in event com-
prehension, cognition integrates the most salient information patterns
around us, allowing us to understand where we are and where we have come

[ 94 ]  Literature through a Cognitive Lens

 

from so that we can predict where to move next. Bor notes that we have a
uniquely ‘ravenous desire to find structure in the information we pick up in
the world… . We … develop strategies to further help us—​strategies that
themselves are forms of patterns that assist us in spotting other patterns… .
Some of our greatest insights can be gleaned from moving up another level
and noticing that certain patterns relate to others, which on first blush appear
entirely unconnected—​spotting patterns of patterns’ (2012, pp. 147, 150).
Because we are active and social animals, narrative channels our pat-
terns of patterns and shapes much of our thought and much of our
literature. Indeed, narrative seems highly likely to be the default task ori-
entation of the human mind. By that I mean that if our minds can process
information in narrative terms, if they can interpret what they experience
as events, they will—​and they will do so through a more or less automatic
convergence of pattern-​recognition processes.
The neuroscientists Antonio Damasio and Joseph LeDoux propose
convergence zones in the brain, where, for example, the distinct what and
where pathways in our visual processing converge to allow us to understand
both the nature of objects and their location (Damasio, 2000; LeDoux,
2002). These convergent information pathways feed in turn into higher-​
level convergence zones, where information from still more different kinds
of pathways, like visual and aural and emotional, meet. The pressure to
understand events, especially events involving one’s own species, has pro-
duced superconvergence zones in the brain.
In everyday event comprehension and in narratives like gossip, history,
fiction, and drama, information naturally salient to members of our ultraso-
cial, highly cooperative, and highly competitive species naturally converges
in many kinds of pattern, as soon as clear information or inference becomes
available. Think of a taxonomic chart, where lower hierarchies like subspe-
cies, species, and genera converge into higher classifications like families,
orders, classes, and phyla; or a family tree, where distinct family lineages con-
verge in the past towards a common stock. But these convergences stretch
back in time, and slowly. In narrative comprehension, on the other hand,
neural data feed forward into convergence and superconvergence zones,
almost automatically, and on a scale of milliseconds. Data and inferences
about identity, age, sex, appearance, speech, role, status, kinship, friendship,
romantic attachment or interest, alliance, the various factors in personal-
ity, and the attitudes of others all converge in patterns of character. These
in turn converge with other input patterns, like settings, situations (situ-
ational schemas, strategic social information), plot (beliefs and expectations,
desires, goals, intentions, actions, reactions, and outcomes), and genre.

Pat t e r n s of T h o u g h t   [ 95 ]
 

Now to our narrative example, Twelfth Night (1600). From the play’s first
speech, Shakespeare foregrounds Orsino’s prime aim—​to woo Olivia—​and
the special personality of this man that would make him ready to love at
a distance, able to feel confident of ultimately winning a woman despite
her firm rebuffs, and sufficiently sanguine and in control to keep sending
envoys to her on his behalf. The dramatist establishes character as the cause
of the ensuing effects, and the relevance of almost everything that follows
to Orsino’s initial aim of winning Olivia. Shakespeare then launches what
seems to be another plot line, and in another mood. After introducing
Orsino indulgently languid in the first scene, he now presents Viola, ship-
wrecked in a strange land but resolute and decisive. As soon as she hears
Duke Orsino named as the ruler of the region, she comments that she has
heard her father mention him, and adds, ‘He was a bachelor then.’ ‘And so
is now’, she is told (1.2.25‒26). Aha: a possible line of complication looms
already.
She hears that Orsino woos Countess Olivia and wishes she could serve
the countess, but when she hears that Olivia ‘will admit no kind of suit, /​
No, not the duke’s’, she decides to travel to the duke in disguise as a male,
to be ‘an eunuch to him’ (1.2.41‒52). The problem she sees ahead of her,
merely surviving and finding support in foreign parts, we suspect may soon
be no longer the problem she actually faces. Indeed, in her next scene, we
find that Orsino not only has taken her on, but feels an unusual emotional
closeness towards this sensitive young ‘man’, and wants to share his inti-
mate feelings with ‘Cesario’—​who in return, being in fact a woman, feels
even more unreservedly for him.
Romantic comedy thrives on the distance between frustration and ful-
filment in love. Shakespeare rapidly sets up a classic comedic pattern of
frustrated love: Orsino loves Olivia (1.1), who loves Cesario (i.e. Viola) (1.5),
who loves Orsino (1.4).
One of the most famous, most ‘lyrical’, as we say, most poetic, imagi-
native, and emotional speeches in Shakespeare is Twelfth Night’s ‘willow
cabin’ speech, purportedly characterizing Orsino’s love for Olivia, obliquely
expressing Viola’s love for Orsino, inadvertently awakening Olivia’s love
for ‘Cesario’. If it were a lyric, it would have a more open emotional reso-
nance, relevant to writer and reader, to anyone at all. But here, although it
has explosive eloquence, it first fits tightly into patterns of character and
cause and effect within the forward movement of the narrative, which is
what allows us to understand it so immediately.
Shakespeare needs to have Olivia fall in love promptly with Viola-​as-​
Cesario, and that despite the facts—​hurdles he has raised himself—​that
Olivia is in mourning, and that she has said she will foreswear the whole

[ 96 ]  Literature through a Cognitive Lens

 

world for seven years. The quick-​wittedness of Viola-​Cesario as envoy for

Orsino’s love, as she is about to be turned away from Olivia’s gate, piques
the Countess’s interest and earns ‘Cesario’ an audience. Olivia insists to
the envoy, all the same, that despite her knowing Orsino’s noble qualities,
‘I cannot love him. /​He might have took his answer long ago.’ Viola replies
to this that ‘If I did love you in my master’s flame, … In your denial I would
find no sense.’ Olivia asks ‘Cesario’: ‘Why, what would you?’ (what would
you do?) (1.5.246‒256). Viola answers:

Make me a willow cabin at your gate

And call upon my soul within the house,
Write loyal cantons of contemnèd love,
And sing them loud even in the dead of night;
Hallow your name to the reverberate hills,
And make the babbling gossip of the air
Cry out ‘Olivia!’ O, you should not rest
Between the elements of air and earth
But you should pity me. (1.5.257‒265)

To which Olivia replies:  ‘You might do much.’ Because of the context of

romantic comedy, because of the criteria of relevance to the details of
the narrative, because of the promise of complications implicit in cross-​
dressing, we can infer from Olivia’s four plain words—​even before quick-​
witted Viola can—​that the Countess has already started to fall for the envoy.
But looking just at Viola’s speech, we can see also, from our knowledge
of (1) the patterns of fiction in general, and (2) the patterns of romantic
comedy in particular, and (3) the emotional pattern of Viola’s intense and
frustrated love for Orsino, and (4)  the patterns of Orsino’s and Viola’s
responsiveness of character, and (5) the pattern of Olivia’s responsiveness
too (first to Feste, then to reports of the envoy, then to ‘Cesario’ in person),
little though we’ve seen her—​we can see from all these converging pat-
terns that that startling little counterfactual narrative inset expresses the
intensity of Viola’s love for Orsino, both in the vignette she invents and the
yearning she expresses, and that its emotional heat will fire Olivia. Viola
loves Orsino so purely that she is prepared to use her secret love for him as
the imaginative impetus for her utmost eloquence on his behalf. She will
give everything she has, all her considerable heart and mind, to serve the
man she loves, even if her success in winning Olivia’s attention to Orsino’s
suit would ruin forever her own faint chance of winning his love for herself.
And the imaginative sensitivity she shows instantly and comically wins
Olivia not for Orsino but for her poor self, his envoy.

Pat t e r n s of T h o u g h t   [ 97 ]
 

Because the narrative patterns converge in this way, we understand the

dramatic import of this speech instantly, even if we might be hard pressed
to say exactly what some of these lines mean. We understand their ‘lyri-
cal’ eloquence better than their exact meaning; but we understand their
role within the story effortlessly, because Shakespeare has prefocused
our understanding and expectations1 through so many different kinds of
pattern—​of genre, conventions, characters, beliefs, desires, aims, inten-
tions, causes, and effects. For all this set speech’s detachability and ‘lyri-
cal’ intensity, its resonance is not expansively open as in lyric, but tightly
channelled into its narrative context. Even when we do come across the
passage detached from the scene and the play, we either recall its narrative
situation or, if new to the passage, scramble to infer something of its con-
text. Without more than that, it can seem mere grandiloquence. Within its
context, everything falls effortlessly and poignantly into place.
Now let’s switch from narrative to lyric verse, verse characterized by not
depending on narrative and, at its purest, excluding it altogether.
In Why Lyrics Last, I note that the only common feature of verse across
languages is that in verse, poets determine where lines end: they control
our attention by making us focus on particular strings of words in one
mental moment (Boyd, 2012, p. 16). Poets may not know it, or barely intuit
it, but their verse lines shape language to fit a human cognitive constraint,
the capacity of working memory. Focusing audience attention on a line at a
time, poets invite close scrutiny of each line and repay it by satisfying our
appetite for pattern (Hogan, 2003; Boyd, 2012; Fabb, 2014, 2015).
Patterns such as rhythm, rhyme, and syntactic or sonic parallelism,
independently or together, serve to demarcate and integrate lines of verse
in diverse traditions. So too does the patterning of imagery, juxtaposing
not just words but one domain of life and another (‘patterns of patterns’,
Bor comments, ‘which is essentially what analogies are’; Bor, 2012, p. 150).
And so does the very different kind of pattern—​also common in lyric but
less recognized there—​of information compression, compacting multiple
observations into a single statement, an abstraction or generalization.
Narrative organizes experience as automatically as a magnet organizes
iron filings, and it still does this even when recounted in verse. But pre-
cisely because lyric at its purest eschews narrative, it can turn the absence
of story to advantage.
After On the Origin of Stories I  wanted to call my follow-​up On the
Absence of Stories, but my publishers thought it sounded negative. It wasn’t

1.  For narrative prefocusing (in film), see Carroll, 1998.

[ 98 ]  Literature through a Cognitive Lens

 

meant to. The absence of story in pure lyric ‘allows us the illusion of access
to another’s thought at its least constrained by circumstance, in the very
act of appealing to others regardless of their circumstances’ (Boyd, 2012,
p. 29). Liberation from narrative allows lyric thought to shape its own con-
texts and prompts—​and indeed to shape itself to the hilt, even when it
chooses set forms, which at their best can set off new-​found freedoms.
Lyrics also invite an expansively resonating response. Since they ‘for-
feit the supplied circumstances of a story, they need to appeal to our cir-
cumstances whatever these might be—​and this usually means appealing
to concerns connected with any reader’s life’ (Boyd, 2012, pp.  29‒30) or
with human experience in general. Where narrative automatically chan-
nels implications to flow down the gradients of the story—​towards Viola’s
predicament at this precise moment, for instance—​lyric allows them to
radiate out from some vivid but generalized human state. The emotional
resonance is transmitted not via the characters as in narrative but directly
from writer to reader, through what we all share.
In Why Lyrics Last I  show why it is a mistake to regard Shakespeare’s
Sonnets as narratives, and why it seems much more probable that they are
a deliberate attempt by the poet—​after having demonstrated by 1594 his
prowess in all the main dramatic and verse forms of narrative of his time
(comedy, tragedy, history; comic and tragic narrative verse)—​to show what
he can do without narrative.
Although he creates an intense emotional engagement with the Mistress
and especially the Youth, Shakespeare does not tell a story though the
Sonnets:

He maximises the openness of lyric, its freedom from the linearity of story. He
offers each new poem as an unpredictable challenge, not least in the unpredict-
ability of its relation to the poems before and after. He exploits the tension
between the autonomy of each sonnet and the variety of its potential relations
to its neighbors—​emotional, thematic, verbal, rhyming, imagistic, structural;
continuation, contrast, echo, variation, reversal.
Boyd, 2012, p. 74

Rather than trying to read the sonnets as a narrative, as a kind of verse

journal, I  suggest in Why Lyrics Last that we should see Shakespeare’s
art in the Sonnets as in some ways like a kaleidoscope. He tries out, as
it were, ‘small taps or larger jolts from poem to poem, so that themes,
moods, or patterns can persist for a time in slight reconfigurations or
suddenly drop out of sight behind other themes, moods, or patterns’
(2012, p. 101).

Pat t e r n s of T h o u g h t   [ 99 ]
 

For all his well-​known challenges to Petrarchan convention, Shakespeare

still focuses on love, but complicates its psychology and attention-​earning
power by adding elements of hate, or by mingling desire with dislike in the
Mistress sonnets, and idealization in the Youth sonnets with undertones
of disenchantment. Meanwhile, he also ramps up the appeal of patterns
other than the amatory. Helen Vendler (1997) has argued eloquently for
the centrality of pattern within individual sonnets; I echo and extend her
emphasis, but I also stress how Shakespeare amplifies the appeal and sur-
prise of the patterns between sonnet and sonnet.
Some sonnets stand by themselves, except as they form part of the
opening sub-​sequence of 126 focused on the Youth if on anyone, or
the closing sub-​sequence of 28 focused on the Mistress. Others form
sub-​sub-​sequences, up to seventeen at a time. Let’s focus on one such sub-​
sub-​sequence, starting with sonnet 33.

Full many a glorious morning have I seen

Flatter the mountain tops with sovereign eye,
Kissing with golden face the meadows green,
Gilding pale streams with heavenly alchemy,
Anon permit the basest clouds to ride
With ugly rack on his celestial face,
And from the forlorn world his visage hide,
Stealing unseen to west with this disgrace:
Even so my sun one early morn did shine,
With all triumphant splendour on my brow;
But out alack, he was but one hour mine,
The region cloud hath masked him from me now.
Yet him for this my love no whit disdaineth:
Suns of the world may stain, when heaven’s sun staineth.
Shakespeare, 1609/​2002:447

Unusually, Shakespeare structures this sonnet around a single image.

Images themselves are patterns: patterns of experience, at their simplest—​
here, a day that begins in glorious morning but turns to dark cloud, a pat-
tern familiar to us all, and one that effortlessly evokes memories by means
of automatic mental pattern-​matching. Although a fine morning turning
foul is an event, this image is no narrative. In Why Lyrics Last I define a nar-
rative as ‘a representation of a coherent particular sequence of events, usually
involving some agency and purpose’ and consider various cases that do not
quite cross the border, including this: ‘A report of the weather for the day
in a particular place would be a representation of a coherent sequence of

[ 100 ]  Literature through a Cognitive Lens

 

events, but without agency and purpose we would not normally consider
it a narrative, although of course it could become the basis for a narrative
by providing setting or circumstances for events about to unfold’ (2012,
pp. 24‒25). In sonnet 33, the image is not even a report of a particular day,
but a reference to a general phenomenon. It cannot by itself constitute a
narrative.
In quatrain 3 the poet’s report of what has happened in his love for
‘my sun’, that something has darkened it, involves agents and could eas-
ily become narrative, were Shakespeare to add specific detail within this
sonnet or its successors, or build on something in its predecessors, but as
it stands he pointedly thwarts the particularity of narrative. Nothing here
indicates whether the darkening was a waning of love, or an instance of
infidelity, anger, suspicion, or anything else. Readers may therefore feel an
emotional resonance with any unwelcome change whatever that they have
experienced in love. Nothing in the ensuing sonnets crystallizes any nar-
rative, and indeed, as we will see, sonnet 36 completely reverses what son-
nets 34 and 35 might seem very vaguely to suggest. Shakespeare knew how
to tell a story. The evidence here shows he also knew how to avoid telling
one, even while implying powerful emotional changes and inviting power-
ful emotional resonances.
Rather than focusing on the patterns that converge automatically as
we process narrative, Shakespeare explores other kinds of patterns that
he invites us, caught by the emotional hook, to explore with him, slowly,
lingeringly, line by line, image by image, quatrain by quatrain. He invites a
very different kind of attention than the attention that we focus on Viola’s
predicament and that enables us instantly to feel her passion, anticipate
Olivia’s, and predict Viola’s imminent plight.
Repeatedly throughout the Sonnets, Shakespeare draws on the most
familiar natural patterns by which we make sense of experience:  the life
cycle, the seasonal cycle, the cycle of day and night. Here, he also draws
on the perennial association of changing emotions and changing weather.
He freshens the overall image through diction and imagery. In the
first two lines, the implicit sun becomes ‘a sovereign eye’, flattering the
mountaintops with the prolonged attention of the monarch of the sky.
Shakespeare lays the pattern of human hierarchy in general, and royalty in
particular, over the pattern of a brilliant early morning.
In line 3, ‘kissing with golden face the meadows green’, the metaphor
of kissing extends into ‘with golden face’, at this point the sun’s. But the
phrase acquires a double-​sidedness characteristic of the especially close
attention Shakespeare invites in the sonnets, as the green meadow, at
first in shadow in our minds, looks back at us ‘with a golden face’ in the

Pat t e r n s of T h o u g h t   [ 101 ]
 

sunlight. An additional pattern of alliteration links the two colour adjec-

tives, golden and green, one before and one after its noun. In line 4 gilding
continues the g-​alliteration, and links tightly in sound, sense, etymology,
and evocation with golden, but the imagery, still focused on the same scene
and sun, moves all the way from kissing to alchemy.
As so often, Shakespeare uses patterns of structure as units of sense
and strategy, in this case in the setup in this first quatrain: the pattern of
the brilliant sun persists, but in three consecutively different images, the
royal look, the kissing face, the alchemical transformation, on mountain
and meadow and streams. As so often, he also varies the patterns as much
as he can on the syntactical level: in lines 2‒4, three verbs with with, one
late in the line, another early, the third late again; one line with a finite
verb, then two with participles; two lines with face imagery, then one with-
out. As throughout the Sonnets, he plays with patterns of variations, like
Beethoven varying the tone, tempo, idiom, and harmonies of the Diabelli
theme he transforms thirty-​two times.
Shakespeare then follows the shift to the second quatrain with a shift
to clouds covering the sun and the bright day, and a sense of corruption or
shame (basest, ugly, hide, steal, disgrace) besetting the radiant morning, but
still linking with the language of the first quatrain (face, l. 3, l. 6; heavenly,
l. 4; celestial, l. 6). He extends the glorious morning-​beclouded day image
through the first two quatrains, as if it were the octave of a Petrarchan
sonnet. He often complicates the pattern of the Elizabethan sonnet, with
its three quatrains plus couplet, by overlaying, as here, the pattern of the
Italian sonnet:  an octave, then the volta, the shift of thought into what
would be the sestet of the Italian sonnet.
All Shakespeare’s sonnets are love poems, but exceptionally here he
has offered for the entire octave only a purely visual image that keeps love
unmentioned for more than half the sonnet. He has conjured up, evoca-
tively, the visually vivid vehicle of the image, but only right where the volta
comes, with the ‘Even so’ of line 9, does he start to introduce its tenor.
He pays us the compliment of trusting us to infer from ‘my sun’ that he
means the Youth who has been the focus of all the other thirty-​two sonnets
so far (and has often been associated with the splendour of the sun, as in
‘Shall I compare thee to a summer’s day?’)—​an implication strengthened
but still far from explicit in the he and him of lines 11 and 12. He uses the
structural patterns of the sonnet to perfection: just as the first two qua-
trains of the image’s vehicle are compressed into the one quatrain of the
tenor, so the four lines describing the glorious morning compact into two
on the initial radiance of his relationship with the Youth, and the onset of
gloom likewise contracts from four lines to two.

[ 102 ]  Literature through a Cognitive Lens

 

For the first time in the sequence, Shakespeare suggests that some-
thing has troubled the love between Poet and Youth. Something has come
between them; the Youth’s radiance no longer shines on the Poet. This
could become narrative, if it became specific enough; and in the context of
the previous radiance, the shift certainly has emotional overtones, match-
ing those that we know ourselves from the exhilaration of a perfect morn-
ing turning to a dark and threatening day, or a cloudless love turning to
foul emotional weather. But Shakespeare is not aiming for narrative speci-
ficity: we have no idea what the change involves, except disappointment in
a love that had seemed ideal. But the patterns of day and disappointed love
resonate with us precisely because they lack narrative specificity, because
we can instantly link them to a whole range of our own experience.
Then comes the couplet. A Shakespearean sonnet sets up expectations
in its three quatrains, three units of four lines of alternating rhymes, abab,
which it often overturns with the shift to the couplet, a single unit of adja-
cent rhymes, gg. That shift allows the poet

to condense the rest of the sonnet, as it were, from twelve lines into two, into
an emphatic closing epigram; or to advance just one stage further, to a clinching
argument; or to take the thought to a new plane; or to turn the tables suddenly
on the rest. On a first encounter we can never be sure whether the couplet will
repeat, condense, advance, divert or reverse.
Boyd, 2012, p. 33

The couplet of sonnet 33 takes us by a double surprise. Whatever has

come between the Poet and the beloved, it has not affected the Poet’s love.
Whatever gloom has been cast over the Poet’s emotional sky, the radiance of
the Youth persists in his eyes: he is still a sun of the world, still lights up the
Poet’s life. Or so the couplet professes. But stain and staineth pull another
way, echoing the notes of moral corruption in the second quatrain—​basest,
ugly, stealing, disgrace—​as if the Poet wants to be heard to excuse the Youth
but cannot help feeling that the young man’s behaviour deserves reproach.
Other Elizabethan sonnet sequences do not include this degree of psycho-
logical complexity in the love relationship:  the sudden darkening of the
relationship in something shameful, or the love persisting despite the rec-
ognition of some ‘stain’. The Poet is disappointed but besotted, covertly
reproachful yet still overtly idealizing. Despite his cause for complaint, he
wants to affirm his love as emphatically as ever.
A few words more on sonnet 33’s patterns. The whole sonnet is about
love, but love, as the tenor of the glorious morning image, does not
enter explicitly until past half way. The whole sonnet uses the vehicle of

Pat t e r n s of T h o u g h t   [ 103 ]
 

the sunny morning beclouded, but although the image of the sun quite
dominates the octave, the word sun never in fact occurs there. It enters
only with the tenor, the Poet’s love, in the sestet:  ‘Even so my sun’. The
poem is about the disappearance of the emotional sun that lights up the
Poet’s life, yet in this tour de force the literal sun itself remains concealed
by the sonneteer. And after that single appearance of the word sun at the
start of the sestet, the word recurs only in the last line, twice, linking tenor
sun and vehicle sun explicitly within the one line. But if the word sun hides
throughout the octave, the word seen occurs in its first line (meaning ‘I have
seen such glorious sunny mornings’), and unseen in its last (the sun unseen
behind cloud), a deliberate formal bracketing, and with the three letters
of sun scrambled or obscured in ‘unseen’. And the s-​n consonance of sun
and seen recurs in a different key in the last line, suns … stain … sun …
staineth. The sun also reappears in a different mode in the linking of heav-
enly and celestial in the first two quatrains with heaven’s in the couplet.
These are the kinds of things Helen Vendler, MacDonald P. Jackson, and
I  have in mind when we stress the pleasure of pattern, the invitation to
discovery, and the rewards for controlled attention in the Sonnets (Vendler,
1997; Jackson, 2000; Boyd, 2012, 2016).
Like storytellers, poets are instinctive psychologists, but they apply
their instincts to different facets and conditions of human attention.
They instinctively parcel our attention a little at a time: first, to the line,
which when spoken takes about as long as we can hold sound in working
memory—​and as we have seen, this is what explains the rough equivalence
of verse line lengths around the world. In short lyrics, such as sonnets,
poets also invite our concentrated attention to a whole work, enticing us to
linger until we have extracted as much pattern and point as we can for the
moment before we move on.
Poets appear to have recognized instinctively not only the duration
of our working memory but also the rough limitations of the capacity
of working memory: we can hold only about four chunks of information
there at a time (Bor, 2012, pp. 150‒153). In narrative, we can easily chunk
information into higher-​level units (Viola’s situation, Olivia’s responsive-
ness, Viola’s willow cabin speech as a plea on behalf of Orsino and as an
expression of her own sense of steadfast unrequitedness). We can also
chunk together the rough sense of a sonnet, or of its sub-​units, octave and
sestet, or quatrains and couplet, but other patterns, like rhymes, images,
syntax, words, sounds, letters (like the [implied but unspoken sun]-​unseen-​
sun pattern), are not automatically convergent, and not easily chunked.
We discover them slowly, returning to a poem and focusing brief beams of
attention on one or two new features at a time.

[ 104 ]  Literature through a Cognitive Lens

 

The poet invites us to feel the immediate emotional resonance of the

sonnet’s open and accommodating situation, to feel with him as a per-
son or persona, but he also invites us to return, to linger, to discover and
therefore feel with him also as poet, as craftsman, as pattern-​maker and
pattern-​concealer. That bears little relation to our emotional responsive-
ness to a character in an ongoing plot, like Viola as cross-​dressing proxy
wooer, about to inflame Olivia’s unsought passion, even if that too finds
expression in verse.
Committed readers of poetry want to linger over the patterned plea-
sures of lyrics. But in narrative, the onward pull of the story, what philoso-
pher of art Noël Carroll calls its erotectic (‘what next?’) force, especially
in performance but even in solitary reading, keeps us moving ahead. We
know there’s a lot more of the story to come, we’ll exhaust ourselves, and
we won’t ever reach the end if we read this passage, however striking, too
intently. In Shakespeare’s sonnets, on the other hand, no one unless in the
grip of a theory about the sonnets as a concealed story moves on from one
sonnet to its successor to find out what happens next. Many readers might
have hoped for a story before they began, although the static repetitions
of the first fourteen sonnets usually put an end to that expectation. Some
have even tried to construct a story by attempting to rearrange the son-
nets’ sequence, and others, like Paterson, have laboriously and grimly tried
to infer a narrative from the sonnets as they stand. But very few readers
without a point to prove will be impelled from one sonnet to its successor
thinking, what’s going to happen next?
When we move from sonnet 33 to sonnet 34, we find something more
like repetition with variation than progression. Sonnet 34 appears to link
closely with its predecessor in occasion and imagery, but it also compresses,
alters, and complicates the mood, yet still ends by expressing the forgive-
ness of devoted love.

Why didst thou promise such a beauteous day,

And make me travel forth without my cloak,
To let base clouds o’ertake me in my way,
Hiding thy brav’ry in their rotten smoke?
’Tis not enough that through the cloud thou break,
To dry the rain on my storm-​beaten face,
For no man well of such a salve can speak,
That heals the wound, and cures not the disgrace;
Nor can thy shame give physic to my grief;
Though thou repent, yet I have still the loss.
Th’offender’s sorrow lends but weak relief

Pat t e r n s of T h o u g h t   [ 105 ]
 

To him that bears the strong offence’s cross.

Ah, but those tears are pearl which thy love sheds,
And they are rich, and ransom all ill deeds.

Here Shakespeare seems to give his kaleidoscope the tiniest tap, while
showing how much even that slightest touch can reconfigure. He retains
the imagery of a promising day dismally dulled by cloud, but this time
makes it instantly personal. Whereas sonnet 33’s octave describes a per-
fect morning marred by a stormy sequel in concrete but universal terms,
and introduces the beloved only obliquely, only in the third person and not
until the start of the sestet, sonnet 34 turns immediately to the beloved,
in the second person and the first line, and to the Poet’s own experience of
this initially flawless morning: ‘Why didst thou promise such a beauteous
day, /​And make me travel forth without my cloak … ?’ The ‘cloak’ adds
a note of almost novelistic realism, yet this sonnet remains a metaphor
about the relationship, not a narrative.2 The two opening quatrains of son-
net 33, fine weather and foul, condense again into the opening quatrain
here, as they had condensed in the third quatrain of sonnet 33, and the
‘basest clouds’ of 33 fly back as ‘base clouds’.
The personalization applies a new force to the cloud-​sun imagery in the
opening of the second quatrain, ‘’Tis not enough that through the cloud
thou break /​To dry the rain on my storm-​beaten face’: the lines imply the
Poet’s tears, dried by the beloved’s showing again a sunny face. Where the
Poet forgave the Youth at the end of the previous poem, here he seems
unready to forgive, and after this ‘’Tis not enough’ he shifts from meteo-
rological metaphor to direct if unspecific reproach: a barrage of negatives
(not, no, not, nor), and then wound, disgrace, shame, grief, repent, loss, offend-
er’s sorrow, weak relief, the strong offence’s cross. Again Shakespeare evokes
the emotional shift from blithe love to wounded feelings, without specify-
ing at all what has happened.
But once again the sonnet pivots, in a different way, in the couplet. The
reproaches of the last six lines seem to have caused the Youth to show his

2.  In her editorial comments, Emily Troscianko objected that this ‘feels like a forced
distinction: why can’t a metaphor have a narrative structure? (cf. Lakoff and Johnson
on e.g. the path metaphor).’ But the image of, say, life or a project as a path is not a
narrative but a template or schema, or rather a warehouse row of templates, that could
become narrative were specific detail selected and elaborated. To call the metaphor
schema life is a path a narrative is itself a misleading metaphor. Our minds respond
to abstractions or even concrete nouns in isolation, but respond in a different way to
the specifically advancing situations of narrative (Viola’s speech) than to the kind of
vague change of circumstance expressed in the central metaphor of sonnets 33 and 34.

[ 106 ]  Literature through a Cognitive Lens

 

sorrow vividly enough and fetchingly enough to soften the Poet’s appar-
ently firm anger: ‘Ah, but those tears are pearl which thy love sheds, /​And
they are rich, and ransom all ill deeds.’ The imagery of the Poet’s tears
implied in line 6, ‘To dry the rain on my storm-​beaten face’, recurs as tears
on the Youth’s face, which immediately win back the Poet who had deter-
mined to hold out in anger. As in sonnet 33, but with a very different emo-
tional contour, even if with the same sense of besottedness, sonnet 34 ends
by forgiving the Youth for some strong personal offence against their love.
Sonnet 35 draws on the two previous poems but compresses still further
and shifts in a new direction. Where for much of sonnet 34 the emotion
had felt raw, here Shakespeare begins with a generalization and four swift,
almost glibly proverbial, images. He expands and distances the phenom-
enologically felt sun-​cloud image of the earlier sonnets into the cool imper-
sonal assurance of ‘Clouds and eclipses stain both sun and moon’ (notice
‘stain’ here echoing the repeated stain in the last line of sonnet 33).
But the note of calm, resigned, wise forgiveness in the first line of the
second quatrain, ‘All men make faults’, itself turns into another sting,
another rebuke: ‘and even I in this, /​Authorizing thy trespass with com-
pare’. In earlier sonnets (18, 21), the Poet has spoken of comparisons only
in terms that extol the Youth; now he acknowledges that the comparisons
he introduces here to allay the Youth’s offence actually corrupt him in
excusing a foul fault in the one he loves. The notes of easy besotted for-
giveness in sonnet 33 and besotted forgiveness after rankling resentment
in sonnet 34 change here to a sense of inextricable ambivalence, reproach
of the other and of himself, that lingers through the last ten lines of the
sonnet. This time the couplet offers no switch in the complex emotion but
simply compresses the complicity and the irony: ‘That I an accessary needs
must be /​To that sweet thief which sourly robs from me.’
Sonnets 33‒35 offer a succession of shifts of highly personalized and far
from idealized or conventionalized emotions that seem to suggest a vividly
personal sequence of feelings. But Shakespeare takes great care to specify
no events to dampen down the resonance that any of these sonnets might
have with our own experiences of emotional disenchantment and ambiva-
lence. As always in his sonnets, he evokes the immediacy of emotion in
intense personal relationships that he had also mastered in drama, and at
least the illusion of his particular personal feelings—​the sense that here
in the tight focus of lyric we eavesdrop on him at his most intimate—​but
without the specificity of narrative.
Sonnet 33 introduces an emotional cloud in a hitherto sunny relation-
ship, which becomes in sonnet 34 not just an emotional darkening but
some emotional wound or offence. Yet in both sonnets the Poet dismisses

Pat t e r n s of T h o u g h t   [ 107 ]
 

the felt shadow or offence, easily in sonnet 33, with difficulty in sonnet 34.
Sonnet 35 tries another angle, showing the Poet complicit in the beloved’s
offence by his very wish to excuse it more than it deserves. Sonnet 36 then
offers one further tap on the emotional and relational kaleidoscope: a new
thought, a new imaginative opportunity. Shakespeare flips the situation
over, in a way utterly different from, and indifferent to, the causal sequences
of narrative: What if the fault is not the beloved’s but the Poet’s, what if
the Poet’s mere complicity in the beloved’s guilt in 35 were expanded into
the Poet as the sole source of guilt? What if not the beloved other but the
speaking self is the guilty one, while the postulate of intense love remains
the same? In that case, given his love and his guilt, the Poet urges the
beloved not to associate with him, lest he become tainted by association.
In all of these sonnets Shakespeare takes the postulate of an absolute,
idealized love, but showing its first flaws: an emotional dimming (33), an
actual offence by the other (34), an offence shared by the poet in seeking
to excuse it (35), an offence or shame on the Poet’s part that he wants the
other not to have to share (36). Not only does this gloom or guilt start up
suddenly in sonnet 33, and vary itself with different intensities within the
sub-​sub-​sequence, then transfer itself to the opposite party, but it has no
consequences in the sonnets that follow, which easily return to the primary
postulate of the beloved’s perfections (sonnet 53 ends: ‘But you like none,
none you, for constant heart’).
How different this experience of reading is from that of Shakespeare’s
narrative and dramatic verse. Narrative and its constraining and instantly
apprehended patterns do not shape the Sonnets: lyric and its plethora of
open but delayed patterns do.
While superficially sonnets 33‒36 might seem to describe the same
trouble that has arisen in Shakespeare’s relationship to the youth, in fact it
becomes clear that the poet is testing out different kinds of possible trou-
ble. He is experimenting with vague scenarios or templates, offering new
ones each time, but not developing a story. Rather than moving from 0
to, say, 14 to 28 to 42 to 56 along a single axis, he moves from 0,0 to 0,14
in the first sonnet (to use Cartesian coordinates anachronistically), from
0,0 to 14,0 in the next, from 0,0 to 0,−​14 in the next, from 0,0 to −​14,0
in the next. He experiments in possibility space rather than advancing in
narrative time.
Narrative, especially literary narrative, rarely ignores the opportunities
of enriched attention to language and pattern that are central to verse.
Verse rarely ignores the additional interest, the supplied relevance, the
forward impetus, and the ease of processing that are central to story. But
when verse unwinds itself from narrative, in lyric, the mind’s processing of

[ 108 ]  Literature through a Cognitive Lens

 

pattern changes from effortless to effortful. No wonder so many more read-

ers prefer narrative to the more intense demands and the more deferred
delights of lyric.

REFERENCES

Bor, D. (2012). The ravenous brain: How the new science of consciousness explains our insa-
tiable search for meaning. New York: Basic Books.
Boyd, B. (2009). On the origin of stories:  Evolution, cognition, and fiction. Cambridge,
MA: Belknap Press of Harvard University Press.
Boyd, B. (2012). Why lyrics last:  Evolution, cognition, and Shakespeare’s sonnets.
Cambridge, MA: Harvard University Press.
Boyd, B. (2016). Experiments with experience: Consilient multilevel explanations of
art and literature. In J. Carroll, D. P. McAdams, and E. O. Wilson (Eds.), Darwin’s
bridge:  Uniting the humanities and sciences (pp. 223‒244). Oxford:  Oxford
University Press.
Carroll, N. (1998). A philosophy of mass art. Oxford: Clarendon.
Damasio, A. (2000). The feeling of what happens: Body, emotion, and the making of con-
sciousness. London: Vintage.
Edelman, G. (2006). Second nature:  Brain science and human knowledge. New Haven,
CT: Yale University Press.
Fabb, N. (2014). The verse-​line as a whole unit in working memory, ease of process-
ing, and the aesthetic effects of form. Royal Institute of Philosophy Supplement,
75, 29–​50.
Fabb, N. (2015). What is poetry? Language and memory in the poems of the world.
Cambridge: Cambridge University Press.
Hogan, P. C. (2003). The mind and its stories: Narrative universals and human emotion.
Cambridge: Cambridge University Press.
Jackson, M. P. (2000). Aspects of organisation in Shakespeare’s Sonnets (1609).
Parergon, 17, 109–​134.
Kurzweil, R. (2012). How to create a mind:  The secret of human thought revealed.
London: Viking.
LeDoux, J. (2002). The synaptic self: How our brains become who we are. New York: Viking.
Paterson, D. (2010). Reading Shakespeare’s sonnets: A new commentary. London: Faber.
Shakespeare, W. (2002). Complete sonnets and poems. C. Burrow (Ed.). Oxford: Oxford
University Press.
Vendler, H. (1997). The art of Shakespeare’s sonnets. Cambridge, MA: Belknap Press of
Harvard University Press.

Pat t e r n s of T h o u g h t   [ 109 ]
 
 

PART II

Cognition through a Literary Lens

 
 

CHAPTER 6

Simulation and the Structure

of Emotional Memory
Learning from Arthur Miller’s After the Fall
PATRICK COLM HOGAN

I n recent years, researchers have come increasingly to recognize that

the study of literature may converge with research in cognitive science,
even in some cases contributing to the latter. This is most obvious when it
comes to emotion, and perhaps most famously associated with emotion-
ally tinged memory (as in Lehrer’s popular treatment of Proust [2008]).
In keeping with this, What Literature Teaches Us About Emotion (Hogan,
2011) argues for a wide range of ways in which the study of literature may
advance affective science. The potential value of literature seems equally
clear for understanding simulation.
Simulation is the imaginative generation of particular conditions and
trajectories beyond direct perceptual experience and conceptual inference.1
Cases of simulation include a reader’s visualization of a character’s actions
in a story or a recipient’s subvocalization (inner speaking) of a correspon-
dent’s tone of voice in a letter. The most obvious cases of simulation occur
with hypothetical and counterfactual scenarios, when one does not simply
infer possible trajectories by deduction or induction, but assumes a per-
spective on events and actions, often someone else’s perspective, concretely

1.  For a fuller discussion of simulation and literature, treating the principles and
parameters guiding simulation, see Hogan, 2013a.
 

imagining particulars from that perspective. However, as the mention of

character visualization and correspondent subvocalization suggests, simu-
lation is not confined to hypotheticals and counterfactuals. Indeed, though
it is not always fully recognized as such, simulation is a dynamic, ongoing,
interactive process that is part of every social encounter, including encoun-
ters with novels and letters. For instance, it occurs anytime one engages in
a conversation. Indeed, Theory of Mind (our understanding of other peo-
ple’s thoughts and feelings) centrally involves not only inference but also
simulation (on the involvement of both processes in Theory of Mind, see
e.g. Doherty, 2009, p. 48).
In all cases, simulation is a rule-​governed procedure that integrates
both dispositional and situational factors, both enduring and ephem-
eral features of personality and circumstance. It is generally not a self-​
conscious process and is open to introspection in only very limited ways.
Specifically, the rules governing simulation are no more available to intro-
spection than are grammatical rules. Other aspects of simulation are open
to introspection in roughly the manner that aspects of grammar are open
to introspection. For example, I  just typed and subvocalized ‘aspects of
grammar’. In doing that, I was not self-​consciously thinking of a rule for
forming regular plurals. Moreover, I could not introspect that rule (which
is more complex than the apparent ‘add “s” ’ procedure). Even introspecting
the mere fact that I formed a plural is complex and mediated by knowledge
that there is a word stem and a suffix. The situation is much the same with
simulation. For example, when I initially read the comments of the editors
on the first draft of this chapter, I  found them rather disagreeable. This
was in part due to the way I unselfconsciously simulated the tone of voice
in which the text of the cover letter was (subvocally) delivered. I only real-
ized this when I self-​consciously thought about the tone of that text and
re-​simulated it differently.
Simulation is particularly important in literature. Indeed, the ‘imagina-
tive’ in ‘imaginative writing’ may plausibly be understood as a particular
use of ordinary processes of simulation. As the influential emotion psy-
chologist and award-​winning novelist Keith Oatley put the idea, ‘the way
fiction works in the mind and brain is that when we read, hear, or watch a
story, we create and run a simulation of selves in the social world. This kind
of simulation was invented long before computers. It could not run on a
computer. It can only run on a mind constructed to understand itself and
others’ (2012, p. 172). The particular use of simulation in literature is argu-
ably a matter of intensification. Novels differ from ordinary counterfactual
imaginations in being more sustained, more elaborated, and more nuanced
in both their cognitive particularity and their emotional consequence. In

[ 114 ]  Cognition through a Literary Lens

 

this way, authors often have particular expertise in simulation—​not only in

simulating, but in representing and successfully sharing those simulations.
The representation and sharing are enabled by the author’s tacit simulation
of readers’ reception and his or her adjustment of the text to avoid possible
misunderstandings or to enrich the emotional response of readers. As dis-
cussed in How Authors’ Minds Make Stories (Hogan, 2013a), all this makes
literary works prime candidates for contributing to our understanding of
simulation. That is perhaps particularly true in the relation between simu-
lation and emotion.
The chapter begins by outlining some ways in which literature may be
valuable for the study of emotion and simulation. It then briefly sketches
some points in our current understanding of those topics. The bulk of the
chapter concerns Arthur Miller’s 1964 play After the Fall, based on (but
deviating from) Miller’s biography (and thus his personal memories),
including his relation with Marilyn Monroe.2
The analysis of the play begins by considering two moments of anom-
alous rage on the part of Quentin, the protagonist (based in part on
Miller himself). The question posed by these moments is the same as
one we often encounter in ordinary life:  Why do people—​indeed, why
do we ourselves—​sometimes respond to events with emotion that is
not merely inappropriate, but bafflingly disproportionate in the circum-
stances? In part, cognitive science helps us to isolate aspects of the play
that explain Quentin’s rage. More significantly, the play suggests that
emotional memories and simulation have structural features which have
not been widely recognized in cognitive science. Once we have isolated
the reasons for Quentin’s response, we begin to see that the play sug-
gests something about the nature of emotional memory and its relation
to simulation.3

THE VALUE OF LITERATURE FOR UNDERSTANDING

EMOTION AND SIMULATION

In What Literature Teaches Us About Emotion (Hogan, 2011), I  have argued

that literary works may provide one important source of evidence for the
scientific understanding of human emotion. In brief, the argument is that

2.  On Miller’s use of autobiographical elements in the play, see Moss, 1966.

3.  There has been little cognitive work on Miller. Analyses of the play are somewhat
limited and often focus on more straightforward themes, such as McCarthyism (Smit,
2010) and the Holocaust (Forsyth, 2008; Balakian, 1997 treats both topics), as well as
literary sources for the play (Royal, 2000).

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 115 ]
 

successful works of literature are in effect thought experiments. They develop

scenarios of human action and interaction that depict emotions at work in
complex situations. As such, they are closer to achieving ecological validity
than most laboratory experiments, which tend to present participants with
highly artificial tasks removed from any ordinary human context. Of course,
they lack the full reality of studies that focus on factual occurrences—​for
example, through some parallel to ethnographic observation or when par-
ticipants are asked to record their feelings and experiences at certain points
in the day. However, literary works have two advantages over such obser-
vational and prompted self-​report studies. First, they do not involve the
distorting effects of outside scrutiny. The intervention of a researcher in a
social situation necessarily changes that social situation. Second, at least in
many cases, they are more fully and more expertly depicted. Field research
necessarily relies on the reports from the field, either those of the scientist
or those of the self-​observing participant. It is rare that such reports have the
richness and precision of George Eliot or Balzac.
Of course, the mere fact that an author has envisioned a particular sce-
nario is hardly a guarantee that it is accurate to the way human cognition
or emotion operates. This is where the importance of ‘successful’ artistic
representation enters. The representational value of a work of art is sug-
gested by the systematically correlated response of a broad range of read-
ers. A broad range of readers have been moved by Othello’s jealousy, Lear’s
remorse, and Hamlet’s grief. Moreover, we may infer that they have been
moved systematically, not at random. If a reader is affected by the play, it is
likely that Othello’s jealousy led him or her to share those feelings empathi-
cally (even while seeing that they are based on errors) or to feel anger and
resentment—​thus to respond with parallel or complementary emotions.
In either case, the response is systematically related to the depiction of
Othello, suggesting that there is something in that depiction that bears on
our own experiences of jealousy and our responses to jealousy in ourselves
and others. We might contrast the case of a work in which the representa-
tion of jealousy was not in keeping with our experiences of its origins and
trajectory. In that case, we would presumably not respond with parallel or
complementary emotions, but with indifference or disorientation.
This is not to say that we should simply assume the representational
validity of any work by Shakespeare—​or by other successful writers, such
as Eliot, Balzac, or the case we will be considering, Arthur Miller. In other
words, we should not assume that the processes or structures a work sug-
gests are psychologically or socially accurate. But this is true for all research.
A  given laboratory experiment or a particular field study suggests some
conclusions about the nature and operation of emotion. But it necessarily

[ 116 ]  Cognition through a Literary Lens

 

has limits. Moreover, there are invariably complications in our interpreta-

tion of the findings from any form of research. In the end, we have to rely
on the convergence of multiple forms of study and multiple results within
each form of study.
As to literature and the arts, there are numerous complications in mov-
ing from the artwork to conclusions about emotion. Indeed, the very suc-
cess of a work suggests its representational validity in some areas, but its
representational invalidity in other areas. There are two obvious ways in
which successful works are likely to be biased, thus inaccurate, in their
simulation of emotions in actions and situations. (There are undoubtedly
other biases as well; these two appear particularly important and, in any
case, illustrate the general point.) The first is that they are likely to devi-
ate from accuracy in a direction that intensifies the emotional response of
readers. For example, a reader is likely to find the conclusion of a roman-
tic comedy more satisfying if the united couple has no ambivalence about
their union. This does not mean that romantic love is unambivalent in real-
ity. The second obvious form of deviation is ideological. Works are likely
to be socially successful to the extent that they represent characters and
their emotions in keeping with social norms. For example, in a patriarchal
society, it is likely that a successful work will depict women as more content
with their subordinated position than they in fact are. (Of course, ‘likely’
does not mean ‘inevitable’, and works are often ambiguous or inconsistent.
But the caution holds in any case.)
On the other hand, these biases are not wholly deleterious. Distortions
of both sorts may contribute to our understanding of human emotion.
However, they contribute by their depictions of, for example, ideals or
social norms of emotion, rather than actual conditions. In each case, we
must, once again, rely on convergent evidence from different forms of
study. In some instances, one form of study (e.g. literary analysis) may pri-
marily suggest hypotheses that may be fruitfully examined in other forms
(e.g. laboratory experiments).
Emotion and simulation seem to be particularly apt topics for under-
standing through literature. Emotion is arguably the central concern of lit-
erature. We read literature that engages and moves us, and authors write,
in part, to engage and move us. Moreover, literature does this through
exploring the motivations and actions of characters—​often their sexual
desires and attachment bonds, their pride and shame (both individual and
collective), and their guilt and anger.4 Emotion is in many ways the stuff

4.  See Hogan (2003) on these emotions and their relation to cross-​culturally occur-
ring narrative genres.

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 117 ]
 

of literature. As to simulation, literature is (again) the product of our sys-

tematic, causally particular, perspective-​guided imagination of situations
and events, most often people acting in social conditions. Indeed, litera-
ture and the arts are probably our most intense and fully elaborated forms
of simulation.
Simulation is especially germane here due to its importance for the
understanding of emotion. Focusing in particular on our comprehension
of other people’s minds, cognitive scientists distinguish two ways in which
we understand motivations and ideas:  simulation and inference. In this
context, simulation is a (largely unselfconscious) process by which we envi-
sion someone’s feelings, beliefs, and intentions, integrating circumstances
and dispositions, as noted above. We do this based on our own implicitly
remembered experiences and actions, with their feelings, beliefs, and
intentions—​also integrating circumstances and dispositions. Inference is a
more self-​conscious or reflective process whereby we rely on more general,
‘theoretical’ premises, from which we draw conclusions. Some researchers
have suggested that we may be particularly prone to understand emotion
through simulation (Doherty, 2009, p.  49). This is perhaps unsurprising
in that the relative speed of spontaneous simulation (in contrast with
theoretical inference) would appear to have evolutionary advantages for our
response to other people’s (or even animals’) motivational attitudes in criti-
cal situations. As perhaps our most elaborate form of simulation, literature
seems particularly well suited to draw on emotional sensitivities—​through
an author’s creative simulations and through the related, responsive simu-
lations of readers.
These points suggest that emotion is not the only psychological phe-
nomenon that might be illuminated by the cognitive study of literature.
Simulation is a crucial process in ordinary life. We engage in it all the
time: when we try to understand other people’s intentions, when we imag-
ine how we might have behaved differently in the past to produce more
desirable outcomes, when we envision future courses of action, and so on.
However, as just noted, there is perhaps no type of simulation that is so
fully and intensively developed as that of literary works. This is true not
only with regard to the characters and their actions in the storyworld. It is
true also of the author’s ongoing sense of audience, his or her tacit simula-
tion of the reader’s (simulative) response to the conditions and events of
the literary work, and thus the author’s sensitivity to possible misunder-
standings or divergent emotions on the part of a reader.
Put differently, our most accomplished authors are likely to be our most
accomplished simulators, at least in certain respects. As such, we would
expect their simulations—​including their simulations of the process of

[ 118 ]  Cognition through a Literary Lens

 

simulation—​to be particularly revealing. Of course, knowing how to do

something does not entail having explicit knowledge of just what one is
doing. But even in the simulation of simulation (e.g. an author’s imagina-
tion of a character’s imagination), that is not what is at issue. Rather, as
with emotion, the author is not articulating a theory. He or she is repre-
senting a process that is intimately familiar and that he or she evaluates
implicitly, working and reworking the depiction until it feels right to his or
her own experience.

A NOTE ON THE NATURE OF EMOTION AND SIMULATION

Before turning to Miller’s play, it is important to sketch briefly some-

thing of what we currently know about emotion and simulation. There has
been enormous advancement in the study of emotion in recent decades.
Nonetheless, even fundamental principles of the field remain controver-
sial. For example, there is a broad difference between theorists who see
emotion as a function of appraisals (see e.g. Oatley, Keltner, and Jenkins,
2007, ch. 7) and those who conceive of emotion as a more automatic, per-
haps sense-​based process. In the account of appraisal theorists, we have
goals and we evaluate events or conditions in relation to those goals. Those
appraisals (which may be unconscious) yield emotions. For example, one of
Smith’s goals is to avoid pain. When the dentist tells him that she needs to
do root canal surgery, Smith becomes afraid because he appraises the situ-
ation as increasing the likelihood that he will experience pain.
In What Literature Teaches Us About Emotion (2011), I  have argued
in favour of a more concrete and automatic account of emotion. In this
account, we have emotional responses to some sensory properties (e.g.
to the sight of bared fangs). However, these sensory properties are not
confined to actual perception. They occur also with memory and with
simulation. Smith’s fear is not the result of a calculation or appraisal. Put
differently, it is not the result of theory-​like inference, even one that is
unconscious. It is, rather, the result of memories and simulated images.
Inference is not irrelevant to this account. Smith engages in inference and
without that process he would not feel fear. However, the emotion does not
result from the inference per se. Rather, it results from the sensory images
(including auditory, kinaesthetic, or other non-​visual memories and simu-
lations) that are activated in connection with that inference.
In this second model of emotion, there are again three sources of emo-
tion elicitation:  perception, memory, and simulation. But here the ques-
tion arises as to just what connects a given image with an emotion system.

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 119 ]
 

The case of memory is simplest. Emotional memories are a specific type

of memory (LeDoux, 1996, p. 182). When activated, they serve to activate
the initial emotion system (or, we might add, an appropriately correlated
emotion system, especially in the cases of pleasure and pain). For exam-
ple, a sorrowful or angry emotional memory, once triggered, will revive
a feeling of sorrow or anger (though a memory of pain is likely to elicit
fear). However, perceptions and simulations of certain targets have emo-
tional effects independent of emotional memories. In some cases, these
propensities are innate. Among the innate elicitors of emotion are almost
certainly the emotion expressions of other people. In other words, we do
not have to acquire a happy response to smiling and laughter or a fear-
ful response to screams of fear or facial gestures of terror. Indeed, innate
emotional responsiveness to emotion expressions seems to be almost all
that is logically needed to build up most of our adult emotional repertoire.
Specifically, early sensitivities to caregivers’ emotional expressions help
to develop our emotion systems through critical period experiences.5 One
famous case of this concerns monkeys and snakes. It was thought for some
time that monkeys were innately afraid of snakes. However, subsequent
research indicated that monkeys routinely developed a fear of snakes by
observing their mothers’ fear of snakes (Damasio, 2003, p. 47). This illus-
trates the great importance of early experiences of empathy and emotion
contagion for the development of later emotional propensities. (Emotion
contagion involves egocentric emotion, whereas empathy is centred on the
other person. For example, in empathic fear, I  am afraid for the person
screaming in terror; in fear contagion, I am afraid for myself.)
Simulation is much less well understood than emotion. Again, it is a
cognitive capacity for imagining particular conditions or causally entrained
sequences of situations and events that are not being and have not been
directly experienced. In How Authors’ Minds Make Stories (2013a), I argued
that simulation may be understood in terms of complexes of principles
with variables that allow for small changes in imagined scenarios. I am now
inclined to expand the account, viewing simulation as akin to some other
cognitive processes that are governed by broad and variable principles,
standard cases, and/​or significant instances.6 The idea is consistent with

5.  The phrase ‘critical period’ is used somewhat differently by different authors (for
discussion in relation to the theoretically and historically important area of language
acquisition, see e.g. Hyltenstam, 2011). I use the phrase rather broadly to refer to a
developmental period in which a particular system, such as an emotion system, is
uniquely sensitive to formative experiences.
6.  For example, aesthetic response seems to involve all three (see Hogan, 2016,
ch. 4).

[ 120 ]  Cognition through a Literary Lens

 

the view that there are three distinct ways in which categorization oper-
ates: through rules, through prototypes, and through instances or exem-
plars (Murphy and Hoffmann, 2012). For present purposes, the last may be
particularly important. In other words, it seems likely that simulation may
be orientated by general semantic or empirical structures—​schematic or
prototypical—​but also and perhaps most significantly by episodic memo-
ries. The idea of episodic memory being involved with the imagination of
particulars has obvious intuitive appeal in modelling particulars on partic-
ulars. It is also compatible with what little scientific evidence there seems
to be that might bear on the issue. For instance, in some cases, when imag-
ining what Smith will do, I may rely on an unselfconscious use of general
principles, such as a tacit assumption that people pursue cherished goals.
In many cases, however, I may assume a trajectory parallel with a specific
act of his in the past. The point is in keeping with Nisbett and Ross’s obser-
vations on the cognitive importance of single, consequential experiences
(1980, p. 15).
There has been some valuable research on organizational procedures
governing simulation.7 However, exemplar-​ based simulation does not
appear to have received much attention. (There appears to have been little
attention of any sort to degrees of generality in the organizational sources
of simulation.) This seems particularly unfortunate, given the relation of
simulation to particularity and point of view, as well as its close connection
with emotion. Indeed, given these points, we might expect that episodic
memory and emotional memory—​thus exemplars—​would be key factors
in developing simulation. However, these likely relations have hardly been
touched on in the research. In this case, too, literature seems to be a field
particularly well suited to help us think about the issue.

AFTER THE FALL (I): EMOTIONAL OVERREACTION

We are all familiar with the general scenario. Something happens that is
annoying, but one of the participants responds with anger; or something
happens that could give rise to anger, but one of the participants responds
with rage. Most of us have seen this happen both to other people and to
ourselves. There are two points where we find such overreactions in After
the Fall. Both concern the main character, Quentin, and one of his wives.

7.  See e.g. Carlson and Kenny (2005) on spatial language comprehension and Decety
and Stevens (2009) on motor simulation.

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 121 ]
 

Specifically, they are instances of surprising rage that initiates violent

action, fortunately stopped by Quentin before it leads to harm.
The first instance is the culmination of an argument between Quentin
and his first wife, Louise. Quentin has come home late. In consequence,
he missed a meeting at his daughter’s school (which he had forgotten).
Earlier in the day, he had told Louise that he had to stay on after normal
work hours. In fact, he did have to stay on. Indeed, the executive commit-
tee of his law firm was meeting with him to decide whether he should leave
the firm. However, he somehow forgot about the meeting and went to the
park. When one of the partners in the law firm calls his home, Louise learns
that Quentin was not at work and appears to assume he is having an affair.
She treats him with cold contempt when he returns, though he tells her he
did not stay at work before she has a chance to confront him with the fact.
She does not change her attitude when he explains his problems at work,
but tells him that he has to sleep on the sofa. He protests that it will upset
their daughter when she sees this in the morning. Louise does not relent.
Quentin receives a telephone call telling him that a friend of his has com-
mitted suicide. Even after this, Louise remains adamant in her attitude.
She criticizes him further, with particular heartlessness given the preced-
ing events. He certainly has reason for irritation, even anger. But it is still
surprising, even somewhat shocking when ‘He starts a clench-​fisted move
toward her and she backs away, terrified’. The ‘violence’ is ‘aborted’ (p. 608),
but remains distressing and unexplained, particularly given the fact that
Quentin is not an especially mercurial character. Although their divorce
does not actually take place for several years, this is the point when it is
first suggested. Indeed, it is the culmination and, in effect, the end of this
marriage for purposes of the play; Louise hardly appears after this point in
the narrative.
The second overreaction involves Quentin and his second wife, Maggie.
Quentin was delayed in returning home after a trip. Maggie questions
where he has been. It is not developed, but there is some suggestion that
Maggie suspects deceit. Although there is tension between them, it is still
surprising that Quentin announces that he will ‘sleep in the living room’
(p.  102). Not long after this, we learn that Quentin has been fired from
his job.
Quentin’s decision to sleep in the living room—​subsequently extended
to sleeping at an inn—​gives us a hint that the emotional memory of his fight
with Louise has been activated. Experiencing anxiety over his professional

8.  This and subsequent citations of After the Fall refer to Miller, 1964.

[ 122 ]  Cognition through a Literary Lens

 

life and finding himself in a similar situation to that night with Louise,
Quentin seems to have taken up the action previously forced on him (sleep-
ing alone). This is already interesting and consequential. This sequence of
events suggests potentially important points about both emotional mem-
ory and simulation. First, it makes it particularly clear that simulation is
not confined to hypotheticals and counterfactuals. One’s response to ongo-
ing situations and events involves considerable simulation as well. Here,
Quentin is extensively filling in aspects of his current situation by (explicit
or implicit) reference to his past.
Second, this sequence suggests that emotional memories are not simple,
isolated impressions (e.g. Quentin’s feeling of ‘clench-​fisted’ rage). They
are structured into particular causal sequences, which is to say narratives.
Those causal sequences need not have a very high degree of probability. For
example, it is hardly a law of nature that a quarrel between a husband and
wife will lead to the man’s exile from the marital bed (though it presum-
ably is highly probable that a quarrel will lead to some form of alienation).
Nonetheless, it seems clear that this particular causal sequence is highly
consequential for Quentin’s understanding and even behaviour.
This leads us to a third implication of the scene: that simulation is con-
nected with narrative memories in two significant ways. First, the previ-
ous experience serves as a model for filling in anticipated outcomes of the
current situation and for providing options for action. In itself, nothing
in the dialogue of Quentin and Maggie thus far suggests the likelihood of
Maggie evicting Quentin from the master bedroom. Moreover, nothing
thus far suggests that this would be a situationally relevant response on
Quentin’s part. Rather, the action, and the anticipation to which it appears
to respond, are supplied by the particular causal sequence of the memory.
Perhaps even more important, this model is directly motivational. The
memory of the fight with Louise does not simply lead Quentin to envision
the possibility of sleeping elsewhere; it leads him to decide on that as the
best course of action (in effect engaging in rejection, rather than experi-
encing rejection). In short, the scene indicates that simulation and actions
based on simulation derive in part from particular causal sequences of
memory associated with strong emotions. Crucially, the emotions are con-
gruent in these two cases, as both situations involve employment anxiety
and marital alienation. That congruence makes the memory more likely to
be active and consequential in Quentin’s current thought and action. It is
also probably necessary in the motivation of his current and anticipated
actions.
Not entirely unlike Louise, Maggie does not respond with sympa-
thy when Quentin says that he has been ‘fired’ (p. 102). Indeed, Maggie

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 123 ]
 

actually denies that the loss of employment has occurred. Quentin

responds, ‘I didn’t expect you to take it seriously’ (p. 103). Expectation
here is a form of simulation, probably implicit. At this point, we can
infer that Quentin’s anticipation of Maggie’s lack of compassion is prob-
ably the result of his implicit simulation being guided by his emotional
memory of the fight with Louise—​an emotional memory that is, again,
narratively structured. In that memory, Louise was unsympathetic with
Quentin’s job crisis.
On the other hand, Maggie is clearly inebriated on alcohol and pills. She
is not quarrelsome, but pleading. Her terrible unhappiness is clear, though
so is Quentin’s understandable exasperation with her self-​destructiveness.
What is less comprehensible is the fact that he berates her, calling her
‘inexcusably vicious’ (p. 107). Indeed, he attacks her the way that Louise
attacked him. The one concrete accusation he makes, the one that shows
she is ‘inexcusably vicious’, is that she ‘called [her] husband idiot in public’
(p. 107). If true, this could be viewed as an instance of humiliation. But we
are presented with no evidence that Maggie ever called Quentin an idiot.
Indeed, given her respect for his intelligence (evident repeatedly in the
course of the play), and the general social view of their mental capabilities
(as we can infer this from the play), this seems extremely unlikely. Here, we
may have an instance of the model actually adding elements to the simula-
tion, or in this case to the simulative reconstruction of explicit memories.9
Louise and Quentin had more than one argument. In the argument that
preceded the one we considered above, Quentin was asking Louise to admit
her faults. That is precisely what he is doing now with Maggie. At the end
of that earlier argument, Louise had called Quentin ‘an idiot’ and left, end-
ing the disagreement (p. 42). It seems at least (simulatively) possible, then,
that Quentin has filled in this element of his simulation of Maggie from an
emotionally salient memory of Louise.
This accusation does precipitate something of a quarrel, but it quickly
peters out and Maggie returns to her pleas for attachment-​based care, or
mothering, from Quentin. Quentin’s indifference and rigidity in effect
keep him in the Louise role until, in a rash act, Maggie tries to swallow
a handful of pills. Understandably both frightened and angered, Quentin
knocks her hand, then begins to struggle with her for the bottle. For a cer-
tain time, there is a physical fight centred on trying to wrest the pills from
her. But, ‘suddenly … he lunges for her throat and lifts her with his grip’
(p. 111). Here, we have the same unexplained surge of violence that we saw
with Louise. It was undoubtedly reasonable for him struggle with her for

9.  On memory as a reconstructive process, see Schacter, 1996.

[ 124 ]  Cognition through a Literary Lens

 

possession of the pills, a struggle that necessarily involved physical force

(particularly as she was hitting him). But this sudden, blind rage—​an abor-
tive attempt at murder, as both of them acknowledge—​is shocking. As in
the case of the argument with Louise, the play presents this as the turning
point that leads to divorce.

AFTER THE FALL (II): CRITICAL PERIOD EMPATHY

The preceding analysis indicates that some of the peculiarities of Quentin’s

conflict with Maggie are comprehensible. The particular causal sequence
of his past conflict with Louise has partially structured his simulation of
the ongoing conflict with Maggie. Moreover, the emotional memories of
those past events have guided his emotional response to the—​partially
simulated and partially experienced—​current situation. But this is, at
best, a very partial explanation. It gives us a fair understanding of some
minor peculiarities in Quentin’s behaviour, such as his statement that he
will sleep in the living room. But it leaves largely unexplained his murder-
ous rage at the end of the sequence. The mere parallelism of the two scenes
does not in itself account for anything unless we have an adequate expla-
nation of the events in the conflict with Louise. In effect, we have thus far
accomplished little more than recognizing that the two anomalies are ver-
sions of the same problem.
It seems clear that generally innate propensities are inadequate to
explain Quentin’s rage. For example, he does not appear to have a broad
proneness to violence, resulting from some genetically based neurological
disorder. Nor does he seem to be responding to the usual sorts of event that
we might consider common triggers for rage. Indeed, if this were the case,
the rage would not be anomalous. Given the account of emotion sketched
above, the obvious alternative is some critical period experience, either an
egocentric experience or one based on empathy, especially an empathic
response to a caregiver. The latter is just what we find. Moreover, that
critical period experience presents us with a particular causal sequence, a
narrative that serves to structure simulations of subsequent experiences,
suggesting the great significance of early emotional experiences not only
for emotion systems proper, but for simulation as well.
Specifically, one of Quentin’s most important memories concerns an
uncle’s wedding. Quentin’s father is late coming home. Earlier, Quentin’s
father had forgotten another uncle’s wedding, and Quentin’s mother
worries that this is the case again. Thus we already have a possible link
with the Louise-​Quentin scenario: a delayed homecoming connected with

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 125 ]
 

forgetfulness. The connection is strengthened when we learn the reason

for the delay. Quentin’s father has suffered terrible financial losses in the
Depression and now learns that he is entirely ruined. He turns to his wife,
Quentin’s mother, devastated. Quentin’s mother has no sympathy. Rather,
she calls him an ‘idiot’ and says that she ‘ought to get a divorce!’ (p. 20).
The links with both anomalous scenes are striking. In each case, Quentin
himself is experiencing a crisis at his job, an insecurity that must inevi-
tably activate the emotional memory of his own father’s disaster. Indeed,
that disaster shaped Quentin’s own emotional relation to employment and
financial security. Louise is genuinely unsympathetic, at least in Quentin’s
memory (itself perhaps biased by the critical period experience). As such,
she does in many ways recall the reaction of his mother. Moreover, the
connection of both incidents with divorce is straightforward, even if the
divorce did not in fact occur with Quentin’s parents.
The connections with the Maggie scenario are looser, but perhaps more
striking for that reason. Again, after Maggie expresses some limited scepti-
cism about his losing his job, Quentin says ‘I didn’t expect you to take it
seriously’ (p. 103), indicating that he anticipated a lack of sympathy on her
part. We noted earlier that this might have had its source in the emotional
memory of his conflict with Louise. We can now see that there is an earlier,
critical period source in the story of his parents’ conflict over his father’s
professional failure. It is this critical period experience, enhanced by the
emotional memory of the Louise story, that guides Quentin’s spontane-
ous simulation of Maggie’s reaction beforehand and that leads him to take
her brief—​and inebriated—​response as definitive. We also now see why
Quentin might (simulatively) reconstruct his memory of Maggie to include
the accusation that she ‘called [her] husband idiot’ (p.  107). Empirical
research shows that people falsely remember the presentation of a par-
ticular word when they are given many items associated with that word.
For example, presented with ‘chalkboard’, ‘desk’, ‘podium’, and ‘classroom’,
participants may believe that they have heard the word ‘school’ (Eysenck,
2009, p. 278). Quentin’s recollection of Maggie’s insult could be of just this
sort. Having experienced all the concomitants of that derogation, Quentin
feels quite certain that he heard the derogation as well, though it seems
implausible to the reader familiar with Maggie’s behaviour and attitudes.
If so, this suggests that such simulatively created memories are a function,
not only of semantic networks, but of specifically emotional narratives
as well.
But we still have not quite explained the anomalous violence of the
two scenes. We have a clue in Quentin’s accusation against Maggie. Again,
Quentin’s criticisms of Maggie are vague generalities, for example that she

[ 126 ]  Cognition through a Literary Lens

 

has been ‘cruel’ (p. 107). He manages only the single concrete accusation
about calling him an idiot. Again, it is his only piece of evidence that she
is ‘inexcusably vicious’ (p. 107), implicitly comparable to the Nazis treated
elsewhere in the play. Considered on its own, the idea is ludicrous. But it
indicates that Quentin’s critical period relation to his mother’s use of that
derogation is of great emotional consequence.
The repetition of this word recalls the psychoanalytic ideas of Jacques
Lacan with his emphasis on the ‘signifier’.10 However, the crucial point here
is that it is not the ‘signifier’ as such that is at issue. It is, rather, the par-
ticular utterance of the word (‘idiot’) by a particular person in a particu-
lar context. It is, in other words, an emotional memory, not an abstract
semantic object that is key. That memory recurs in the course of the play,
haunting Quentin’s later experiences. It first appears in the recollection
of his parents’ quarrel, as just noted. It returns twice in brief images as
Quentin’s mother stretches menacingly over his father, who in the second
instance sits ‘dejectedly’ (p. 30; see also p. 26). The final recurrence is just
before Quentin’s recollection of his murderous conflict with Maggie (p. 100;
the scene with Maggie begins on p. 101). In each case, it is a very physical
fact:  the mother’s and father’s bodies with their emotionally expressive
postures and facial expressions, or the mother’s voice, with its anger and
disdain. Moreover, the context is crucial. Quentin was a successful lawyer.
It is hardly devastating for him to be called an idiot. Indeed, Maggie is the
one who is likely to be thought of (unfairly) as stupid. If she did call him
an idiot, it would be ludicrous. In contrast, Quentin’s mother was educated
and intellectually ambitious, whereas his father was illiterate (p. 17). Her
criticism of him as an idiot is wounding in a way that such a statement
from Maggie could not be.
We begin to see here just what the accusation of idiocy provokes emo-
tionally: a sense of shame. But, again, the shame at issue is the shame of
Quentin’s father, not of Quentin himself. This leads us to the precise phras-
ing of the criticism. It is not simply that Maggie putatively called Quentin
an idiot, but that she did so ‘in public’ (p. 107). In other words, it was not
simply private shame, but public humiliation. Shame and humiliation are
bound up with rage. Specifically, both are common sources of rage and asso-
ciated violence (e.g. Gilligan, 1997; Scheff, 2011; and Walker and Knauer,
2011)—​including spousal abuse (on the importance of childhood experi-
ences of shame, see Mills, 2008, p. 634; see also Lawrence and Taft, 2013).

10.  Lacan treats the signifier in a number of his Écrits (1966). A psychoanalytic con-
nection is not surprising, given the frame of the play, which recalls a psychoanalytic
session (Murphy, 2002, p. 314).

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 127 ]
 

But, again, it does not seem that Maggie did humiliate Quentin in this
way. Even if she did call him idiot in public (which seems unlikely), that
would have been more absurd than humiliating. Rather, the shame and
humiliation were Quentin’s father’s feelings. But where does the public
enter in Quentin’s father’s case? Immediately after his wife calls him ‘idiot’,
Quentin’s father comments that he hears ‘Somebody crying’ (p. 20). It is
Quentin. Quentin has heard the whole exchange. However unintention-
ally, Quentin’s mother has exposed her husband to shame in front of his
son—​an act even more humiliating than such an insult made before a large
group of strangers.
This critical period experience had simulative and motivational conse-
quences. For example, it sensitized Quentin to his daughter’s feelings as he
simulated the results of his conflict with Louise. More important, it led to
his accusation about Maggie insulting him in public. In both cases, we see
again that Quentin’s simulation and emotional response were organized
by reference to this particular causal sequence, this narrative, in emotional
memory. Indeed, we begin to see further connections as well. Shame is to
a great extent a response to the disgust of other people. Quentin’s mother
clearly exhibits disgust at her husband. Later, Quentin’s memory of his
conflict with Louise highlights Louise’s exclamation, ‘You are disgusting!’
(p. 57). Indeed, it is in response to this insult that Quentin first says that
‘Betty will see’ (p.  57). In other words, his concern about his daughter,
though explicitly related to his sleeping on the sofa, is actually prompted
by his wife’s expression of disgust—​just what he had witnessed as a child.
The most important feature of this critical period experience, however,
is the culminating rage of the later events. It is easy to infer that young
Quentin’s response to his mother’s disgust is empathic shame and humili-
ation for his father. Indeed, it is striking that Quentin’s father notices that
Quentin is there, but does not go to Quentin himself. Rather, he tells his
wife, ‘You better talk to him’ (p. 20). It is possible that this is simply the
gender-​based division of labour that apportions comfort-​giving to the
woman, thus stressing the uncomforting role the mother has just been
playing. However, it is also possible to understand the father’s reticence as
the result of shame and a feeling of inadequacy. In any case, it seems clear
that young Quentin felt this shame and humiliation empathically. These
empathic feelings may have combined with his own sense of shame and
humiliation at merely crying over the incident and, as a child, being inca-
pable of defending his father against the joint devastations of professional
and domestic life. Moreover, these feelings may have been enhanced by his
own response of disgust at his father’s apparently passive attitude, thus
his father’s inability to respond adequately to either the professional or

[ 128 ]  Cognition through a Literary Lens

 

the familial crisis. Indeed, such a feeling (of disgust at passivity when faced
with humiliation) could readily motivate the sort of proactive rage that we
find in the conflicts with Louise and Maggie. The point is particularly likely
in the case of Maggie where Quentin shames and humiliates Maggie before
their assistant, rather than being shamed and humiliated himself.
Whatever the precise details, it seems clear that this early experience is
the crucial factor in explaining Quentin’s excessive rage in his arguments
with Louise and Maggie. In each case, then, a narratively structured emo-
tional memory from a critical early period in his emotional development
has served to organize and orientate his ongoing simulation of the interac-
tion, his reconstruction of relevant episodic memories, and his emotional
responses to both. What seemed incomprehensible when considered in
isolation becomes comprehensible when located in a cognitive and specifi-
cally affective history, a history that crucially includes simulation guided by
exemplars from emotional memory.11

CONCLUSION

In short, After the Fall presents us with two instances of anomalous rage—​a
phenomenon all too familiar from ordinary life. It implicitly explains those
events by reference to a series of structurally parallel narrative memories
rooted in Quentin’s childhood experiences. (See Table 6.1 for a schematic
representation of these parallels.)
Perhaps more significantly, the play has implications for our under-
standing of the human mind. Specifically, it indicates that emotional mem-
ories are organized into stories, which is to say, particular causal sequences.
These causal sequences are not necessary or law-​like, nor even probabilis-
tic. Nonetheless, they serve as models for construing and simulating later

11.  Towards the end of the play, Miller directly indicates that there are childhood
precedents for Quentin’s rage. Specifically, Quentin stops strangling Maggie and
begins strangling a phantasm of his mother (p. 111). Miller links this with a different
childhood memory—​when Quentin’s parents had left him behind to take a trip. But
this explains nothing. The incident with the parents leaving Quentin behind has no
distinctive relation to his moments of rage. Thus, it is not clear why Quentin’s mind
would link the two at all. That distinctive relation bears on his father’s humiliation.
Moreover, the preceding analysis does not indicate that his anger at Louise and Maggie
is simply displaced anger at his mother. Rather, that rage results more indirectly from
cognitive processes in which emotional memories of his father’s humiliation guide his
simulation of later, complex interactions in a way that produces rage. It may be that,
in part misled by a certain version of psychoanalysis, Miller failed to understand the
psychological subtlety of his own play—​an ordinary phenomenon (on authors’ misun-
derstanding of their own works, see Hogan, 2013b, ch. 3).

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 129 ]
  
Table 6.1   A SCHEMATIC OUTLINE OF THE STRUCTURAL PARALLELS AMONG QUENTIN’S EXPERIENCES OF MARITAL CONFLICT.

Parents Father not back, Job crisis Unsympathetic Expression of Mention of Hurt child Rage for/​at father’s Paternal
late for an wife (‘idiot’) disgust; shame divorce (Quentin) humiliation passivity
important event
Quentin and Father not back, Job crisis Unsympathetic Expression of Divorce Hurt child Rage/​humiliation Near violence
Louise late for an wife (‘idiot’) disgust; sleeps (Betty)
important event alone
Quentin and Father not back Job crisis Unsympathetic Chooses to sleep Divorce ‘in public’ Rage/​humiliation Violence
Maggie wife (‘idiot’) alone
 

events: defining their causal configurations, filling in intentions or unob-

served actions, reconstructing relevant memories, and so on. In conse-
quence, the play suggests that one’s emotional responses are not responses
to the current situation alone. They are, rather, responses to the current
situation as organized and partially re-​simulated by tacit reference to nar-
ratively structured emotional memories. The play provides further sup-
port for the view that empathic response is central to the development of
emotional propensities through childhood, critical period experiences. It
also indicates that such early, empathic experiences, perhaps particularly
those involving an attachment object, may have important consequences,
not only for emotion, but for simulation as well. Finally, After the Fall hints
that there may, in certain cases, be a perhaps surprisingly important role
for the emotional memory of words—​or, rather, not words alone (thus not
language as a pseudo-​objective structure), but the emotionally powerful
event of a particular utterance at a particular time in a particular voice.
One significant feature of this analysis is that it suggests connections with
psychoanalytic accounts of transference. It re-​understands the psychoana-
lytic idea by reference to critical period experiences in the development of
emotion systems, emotional memories organized through particular causal
sequences or narratives, and simulation processes. This fundamentally
changes the descriptive characterization and explanation of the phenom-
ena. However, it does seem to capture the relevant insights of psychoanaly-
sis. The same point holds for the relations among the different narratives
isolated here and Lévi-​Strauss’s idea of ‘transformation sets’ (1969). Here,
too, a cognitive and affective account both re-​construes and re-​explains the
phenomena, while preserving relevant insights. Finally, there is some link
with Jacques Lacan’s psychoanalytic emphasis on the ‘signifier’. However,
the present account shifts away from the linguistic autonomism of Lacanian
theory to the particularity and, indeed, corporeality (or embodiment) of the
emotional memory—​once again, to some degree preserving the insights,
but with a more descriptively and explanatorily adequate characterization
and development. Each of these links, suggested by Miller’s play—​as well as
the exact processes governing the organization of emotional memory and
simulation—​points to avenues for further cognitive and affective study,
both within and outside literature and the arts.

ACKNOWLEDGEMENTS

An earlier version of part of this essay was delivered at ‘The Science of Story
and Imagination’, Stanford University (2014), and at the Arthur Miller

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 131 ]
 

Centennial conference, Saint Francis College (2015). I am grateful to the

organizers and participants for their comments. The essay has also ben-
efited from the comments of Michael Burke and Emily Troscianko.

REFERENCES

Balakian, J. N. (1997). The Holocaust, the Depression, and McCarthyism: Miller in the

sixties. In C. Bigsby (Ed.), The Cambridge companion to Arthur Miller (pp. 115–​138).
Cambridge: Cambridge University Press.
Carlson, L., and Kenny, R. (2005). Constraints on spatial language comprehen-
sion:  Function and geometry. In D. Pecher and R. Zwaan (Eds.), Grounding
cognition: The role of perception and action in memory, language, and thinking (pp.
35–​64). Cambridge: Cambridge University Press.
Damasio, A. (2003). Looking for Spinoza: Joy, sorrow, and the feeling brain. New  York:
Harvest.
Decety, J., and Stevens, J. (2009). Action representation and its role in social interac-
tion. In K. Markman, W. Klein, and J. Suhr (Eds.), Handbook of Imagination and
Mental Simulation (pp. 3–​20). New York: Psychology Press.
Doherty, M. (2009). Theory of mind: How children understand others’ thoughts and feel-
ings. New York: Psychology Press.
Eysenck, M. (2009). Memory in childhood. In A. Baddeley, M. Eysenck, and M.
Anderson (Eds.), Memory (pp. 267–​291). New York: Psychology Press.
Forsyth, A. (2008). The trauma of articulation: Holocaust representation in After the
fall and Broken glass. Arthur Miller Journal, 3, 41–​60.
Gilligan, J. (1997). Violence: Reflections on a national epidemic. New York: Vintage.
Hogan, P. (2003). The mind and its stories:  Narrative universals and human emotion.
Cambridge: Cambridge University Press.
Hogan, P. (2011). What literature teaches us about emotion. Cambridge:  Cambridge
University Press.
Hogan, P. (2013a). How authors’ minds make stories. Cambridge: Cambridge University
Press.
Hogan, P. (2013b). Narrative discourse: Authors and narrators in literature, film, and art.
Columbus, OH: Ohio State University Press.
Hogan, P. (2016). Beauty and sublimity: A cognitive aesthetics of literature and the arts.
Cambridge: Cambridge University Press.
Hyltenstam, K. (2011). Critical periods. In P. Hogan (Ed.), The Cambridge encyclopedia
of the language sciences (pp. 238–​240). Cambridge: Cambridge University Press.
Lacan, J. (1966). Écrits. Paris: Éditions du Seuil.
Lawrence, A., and Taft, C. (2013). Shame, posttraumatic stress disorder, and intimate
partner violence perpetration, Aggression and Violent Behavior, 18, 191–​194.
LeDoux, J. (1996). The emotional brain: The mysterious underpinnings of emotional life.
New York: Touchstone.
Lehrer, J. (2008). Proust was a neuroscientist. New York: Mariner.
Lévi-​Strauss, C. (1969). The raw and the cooked (J.  and D. Weightman, Trans.).
New York: Harper.
Miller, A. (1964). After the fall. New York: Penguin.
Mills, L. (2008). Shame and intimate abuse: The critical missing link between cause
and cure. Children and Youth Services Review, 30, 631–​638.

[ 132 ]  Cognition through a Literary Lens

 

Moss, L. (1966). Biographical and literary allusion in After the fall. Educational Theatre
Journal, 18, 34–​40.
Murphy, B. (2002). Arendt, Kristeva, and Arthur Miller: Forgiveness and promise in
After the fall. PMLA, 117, 314–​316.
Murphy, G., and Hoffman, A. (2012). Concepts. In K. Frankish and W. Ramsey (Eds.),
The Cambridge handbook of cognitive science (151–​170). Cambridge:  Cambridge
University Press.
Nisbett, R., and Ross, L. (1980) Human inference: Strategies and shortcomings of social
judgment. Englewood Cliffs, NJ: Prentice-​Hall.
Oatley, K. (2012). The passionate muse:  Exploring emotion in stories. Oxford:  Oxford
University Press.
Oatley, K., Keltner, D., and Jenkins, J. (2007) Understanding emotions (2nd ed.).
Malden, MA: Blackwell.
Royal, D. (2000). Camusian existentialism in Arthur Miller’s After the fall. Modern
Drama, 43, 192–​203.
Schacter, D. (1996). Searching for memory:  The brain, the mind, and the past.
New York: Basic.
Scheff, T. (2011). Social-​emotional origins of violence:  A  theory of multiple killing.
Aggression and Violent Behavior, 16, 453–​460.
Smit, D. (2010). An epitaph for the long fifties:  The new American sensibility and
Arthur Miller’s After the fall. CEA Critic, 72, 76–​93.
Walker, J. and Knauer, V. (2011). Humiliation, self-​esteem, and violence. Journal of
Forensic Psychiatry and Psychology, 22, 724–​741.

S i m u l at i o n a n d E m o t i o n a l M e m or y   [ 133 ]
 
 

CHAPTER 7

Cognitive Science and the Double

Vision of Fiction
MER JA POLVINEN

INTRODUCTION

What is the difference between readers’ experience of fictional and non-​

fictional narratives? How might the cognitive sciences be able to capture
that difference? In order to work towards answers to these questions, this
chapter focuses on the cognitive processes engaged in readers of fictional
narratives. It brings together some recent literary theories of the imagina-
tion and fictionality, on the one hand, and views of perceptual experience
presented within the enactive paradigm of cognitive science, on the other.
What these approaches share is a view of the human imagination as a pro-
cess that is both fundamental to our ability to negotiate our physical and
social world, and complex and multilayered enough to resist easy analysis.
Fictionality studies and enactive cognition have also both presented valu-
able alternatives to some intuitive assumptions about the imagination,
assumptions which have to a large extent been reproduced within the com-
putational paradigm of cognitive science.
Within computationally inflected neuroscience, the imagination has
mainly been discussed in the context of visual perception, memory, and
future projection (the production of mental imagery on the basis of recol-
lected details, for example, or the mental replication of actions through
the mirror neuron system; Denis, 2001). While many of these earlier stud-
ies understood the imagination specifically as a process of forming mental
images, today’s research explores more widely the ways in which human
 

beings imagine alternate realities—​whether in everyday situations or as

the audiences of works of fiction.1 However, because the computational
model of the mind follows centuries of tradition in assuming that human
cognition is primarily rational, it presupposes that the fundamental
operations of the imagination must also follow rules of logical informa-
tion processing. While studies of the imagination in the humanities might
be quick to deny such an assumption, the computational drive inherited
by many cognitive studies of literature leads, I suggest, to the neglect of
those effects of fictionality that depend on contradictions between what is
known and what is experienced. Cognitive literary studies are thus at risk
of pushing the imagination to fit within a model which erases many of the
capacities that actually make it interesting to a humanities scholar. Alan
Richardson (2011, 2015), from his position as specialist in the Romantic
period, points out how little attention the mainstream cognitive discussion
has paid to those humanities debates that emphasize the irrational, emo-
tional, and experiential aspects of the phenomenon. The imagination as
conceptualized within the cognitive sciences is ‘strangely attenuated—​one
might even say tamed’, Richardson suggests, because the computational
perspective of the scientists aims to ‘render the imagination rule bound
and quotidian’ (Richardson, 2011, pp. 664–​665; see also Danta and Groth,
2014, p. 7).
But it should also be noted that the cognitive sciences, in moving away
from the computational paradigm, are themselves becoming more inter-
ested in the wilder and weirder sides of the imagination, and in this chapter
I would like to focus on the ways in which the combination of the enactive
paradigm within the cognitive sciences and theories of fictionality within
literary studies can help each other in forming a more nuanced view of the
imagination for both fields of study.

FICTIONALITY IN EMPIRICAL STUDIES OF LITERATURE

Taking up Richardson’s call for more complex views of the imagination in

cognitive literary studies, I  argue that approaching fiction as fiction is a
central element in literary imagining. The underlying idea is that readers of
fiction engage in a seemingly contradictory interplay of the immersive and

1.  Other terms in use when discussing the imagination include ‘make-​believe’ (e.g.
Walton, 1990), which is in use in developmental psychology and the anthropology of
play, and ‘simulation’, which is seen by many cognitive scientists as an integral part of
social cognition (e.g. Gallese, Keysers, and Rizzolatti, 2004).

[ 136 ]  Cognition through a Literary Lens

 

the reflective aspects of fiction. Such a view, while amply theorized within
literary studies, has been largely absent from the empirical studies on read-
ers’ experiences of fiction. This is true, for example, of Melanie Green and
Timothy Brock’s (2000, p. 704) much-​referenced ‘Transportation Scale’, a
questionnaire formulated for studying the depth of readers’ immersion in a
fictional world.2 The scale measures aspects such as imagery, involvement,
emotional effects, and distance from reality, and respondents are asked to
rate these effects on a scale of 0–​60. However, some items on the ques-
tionnaire reveal the scholars’ assumption that involvement with narrative
can only mean engagement with the events and characters represented,
not engagement with the artefact itself. Thus, even though questions 4 (‘I
was mentally involved in the narrative while reading it’) and 8 (‘I found
myself thinking of ways the narrative could have turned out differently’)
can be answered positively when readers’ involvement is with the level of
storytelling rather than (or as well as) the events, the scale interprets all
positive answers as counting towards the respondents being, in Green and
Brock’s (2000, p. 701) terms, ‘transported’ into the world of the narrative
and [becoming] involved with its protagonists’. Another example comes
from one of the most publicized recent experiments concerning the effects
of literature, David Comer Kidd’s and Emanuele Castano’s 2013 article in
Science. Their results seem to show that literary fiction improves our under-
standing of other minds more than do either non-​fiction or popular fiction.
The argument is based on the statistical analysis of readers’ self-​reports
in five separate experiments, of which the first includes the fiction/​non-​
fiction condition. In this study, however, the comparison is undermined by
the fact that the authors deliberately chose non-​fiction texts that ‘primar-
ily focused on a nonhuman subject’, while the fictional texts included ‘at
least two characters’ (Kidd and Castano, 2013, supplementary material), as
if the difference between the two modes consisted simply in the presence
or absence of people to empathize with. With such a choice of material, the
more active priming of interpersonal skills in readers of the fiction texts
can hardly be a surprise, and the findings, such as they are, speak only to
the effects of characterization, not those of fiction or non-​fiction.
These studies, like most empirical work on fictionality, used self-​report
questionnaires. One of the few neurophysiological studies of the difference
between fiction and non-​fiction was conducted by Altmann and colleagues
(2014) to examine how paratextual information shapes the reading process.

2. The Transportation Scale is cited by Appel and Richter (2007); Sanford and
Emmott (2012); Bae, Lee, and Bae (2014); Phillips (2015); amongst others.

D o u b l e Visi o n of Fi c t i o n   [ 137 ]
 

Participants read dozens of c. 50-​word narratives arbitrarily labelled as

either fiction or non-​fiction, while their brains were scanned using fMRI.
The main findings here indicate that the texts flagged as non-​fiction engaged
systems associated with the simulation of physical actions, whereas with
texts flagged as fictions the activation patterns were more like those asso-
ciated with ‘mind-​wandering’ and ‘relational inferences’ (p.  26). Such
results, creating a tenuous connection between fiction-​reading and the
open-​minded readiness of hypothetical thinking, are attractive (see also
Richardson, 2011, pp. 685–​687), and indicate that it is worthwhile to pay
attention to relational as well as action-​related cognition in the processing
of narratives. However, since Altmann and colleagues wanted to focus pri-
marily on the intersubjective aspects of reading, the participants were also
asked to fill out the Interpersonal Reactivity Index (IRI) devised by Mark
H. Davis (1980). These results were then compared with the strengths of
the activation patterns indicating mentalizing—​the act of imagining other
minds. Altmann and colleagues are careful as to the extent of the conclu-
sions they draw from these results, but the suggestion is that because brain
patterns indicating mentalizing are more strongly activated in individuals
who self-​report a strong identification with fictional characters in general,
and because the same patterns of activation are more strongly manifest
in reading fiction than non-​fiction, fiction can be said to differ from non-​
fiction because it engages our interpersonal cognition more fully.
While the fMRI results themselves are intriguing, I would like to draw
attention to one detail in this study that is arguably problematic from the
point of view of the literary view of the imagination. Altmann et al. (2014,
p. 24) base the correlation between the fMRI results and the respondents’
general tendency to engage with characters on one of the four factors iden-
tified in the IRI—​the ‘Fantasy Scale’ (Davis, 1980, table 3). This section
of the IRI questionnaire focuses on reactions to fictional characters, and
even though it is designed for the measurement of empathic skill rather
than the reading of fiction more generally, in it empathic identification is
conceptualized in roughly similar ways to immersion or transport in liter-
ary studies, and equated with successful engagement with fiction. Thus,
some Fantasy Scale items adopt the common shorthand of assuming that
‘good’ or ‘interesting’ stories or films are by definition those that promote
an empathic identification, and positive answers to questions like ‘When I
am reading an interesting story or novel, I imagine how I would feel if the
events of the story were happening to me’ (item 26) and ‘When I watch
a good movie, I can very easily put myself in the place of a leading char-
acter’ (item 23) are assessed as indicating a high tendency to empathize.
However, responses indicating whether or not participants tend to get

[ 138 ]  Cognition through a Literary Lens

 

‘caught up in’ or ‘involved in’ a story (items 7 and 12) are judged the same
way, revealing an underlying assumption that such generally engaged reac-
tions to fictions also designate empathy towards fictional characters. Since
no option is given for respondents to indicate that they may be caught up
in a novel’s way of using language, in the intricacies of its narrative struc-
tures, or, indeed, in its fictionality, all responses to these questions end up
counting towards an individual respondent’s tendency to empathize. This
automatic equation of an empathic tendency with the tendency to become
engaged with fiction causes obvious problems once we start to inquire
more openly into the possible components of fictional engagement.
What I suggest, therefore, is that we look critically at results from stud-
ies which start with the assumption that engagement with fiction functions
only through empathic identification with fictional characters, or that such
engagement requires a form of illusion or loss of a sense of the fictionality
of the story. In the following I will argue that fiction is perceived through a
double vision that is unlike the kind of ‘aesthetic illusion’ (Wolf, 2013) that
sets such awareness and engagement to work against each other.

ENACTIVE PERCEPTION OF FICTIONS

Despite the fact that many psychological and neurological studies of fic-
tionality have suffered from the conceptual problems described above, the
cognitive sciences can offer literary scholars valuable insights into how
our general cognitive skills are activated not only by the environments or
people represented in a text but also by the fact of their fictionality. These
insights are offered by the critique of the computational model that has
appeared in the form of the ‘4E’ paradigm. This paradigm takes the mind to
be embodied, embedded, enactive, and extended (Menary, 2009; Stewart,
Gapenne, and Di Paolo, 2010), and it presents a view of cognition that
replaces the computer-​like input-​processing-​output model with a system
incorporating more complex—​and more intractable—​feedback between
an embodied being and a dynamic environment. As such, 4E approaches
draw not only on neuropsychology but also on phenomenology—​a combi-
nation that the computational paradigm has resisted (e.g. Noë, 2004, 2012;
Gallagher and Zahavi, 2007; Thompson, 2007).
My focus is on the consequences of enaction and embodiment for the
imagining that readers undertake in experiencing fiction. Enactive cog-
nition broadly takes thought to be ‘the exercise of skillful know-​how in
situated and embodied action’, in the sense that all our thinking—​however
abstract and introspective—​ is in constant feedback with ‘recurrent

D o u b l e Visi o n of Fi c t i o n   [ 139 ]
 

sensorimotor patterns of perception and action’ (Thompson, 2007, p. 13).

Consequently, an enactive understanding of literary reading conceives it
also as a skill-​orientated interaction between a reader’s embodied mind and
the literary object. Enactive cognition has already been shown to affect our
understanding of fictionality in interesting ways, and new scholarship has
presented analyses of embodied reactions to texts, including topics such
as sensing fictional spaces (Caracciolo, 2011), experiencing movement and
body boundaries (Esrock, 2001; Kuzmičová, 2012), or identifying embod-
ied feelings coded into the rhythms of narrative (Caracciolo, 2014). In her
examination of perceptual experiences of narrative, for instance, Emily
Troscianko (2013) emphasizes the difference between the enactive view of
imagining and the ‘picture in the head’ variety presented in older forms
of cognitive neuropsychology: ‘I don’t have a mental image of the cat I’m
imagining’, she writes, ‘but I perform the same kinds of exploratory behav-
iours as when I see one, with weaker forms of sensory feedback provided
from memory’ (Troscianko, 2013, p. 185). What is crucial in such literary
scholarship is its interaction with the enactive paradigm to produce a view
of the literary imagination as a set of complex processes that engage the
mind-​body with the fictional environment offered by the text.
Troscianko (2013) has also drawn attention to the way in which our folk-​
psychological assumptions about the imagination often lump together sev-
eral experiential aspects. Her example (drawing on Jajdelska et al., 2010) is
the way in which the concept of ‘vividness’, much used in questionnaire
studies of literary imagination, actually conflates two aspects: actual visual
detail and emotional intensity. The unacknowledged presence of this con-
ceptual amalgam, Troscianko argues, results in flawed experimental data
about the exact processes involved. In a similar fashion, I wish to unpack
another conceptual conflation: that what is being encountered during read-
ing is in some senses like a world, but is a fiction. I am particularly inter-
ested in the role of the clearly signalled fictionality of the literary work in
the reading process, and in how readers’ minds are able to assume a per-
spective that is simultaneously aware of the fictionality of the events it fol-
lows and yet fully cognitively and emotionally engaged with them.
For the purpose of unpicking this conflation, I draw on the theory of
enactive perception as presented in Alva Noë’s Action in perception (2004)
and Varieties of presence (2012). Perception, Noë argues, ‘is constituted not
only by the perceiver’s mastery of patterns of sensorimotor dependence,
but by the fact that the perceiver knows that his or her relation to the envi-
ronment is mediated by such knowledge’ (Noë, 2004, p. 65). According
to this idea of the ‘full-​blooded duality of perceptual experience’, seeing
a silver dollar from an angle includes an experience of the elliptical shape

[ 140 ]  Cognition through a Literary Lens

 

presented from that angle as well as the circularity we know the object to
have, since our learned patterns of sensorimotor perception tell us that
from an angle a circular object presents itself to us as elliptical (Noë,
2012, p. 52). Such experience includes not only how an object looks, but
also how it is—​two aspects of experience that arise from our knowledge
of how objects look from various spatial positions we can have in relation
to them.3 I suggest that the fictionality of a narrative could be seen as an
analogy of our angled stance towards the silver dollar. The representation,
by drawing on readers’ everyday cognitive patterns, creates a sense of the
verisimilitude of the storyworld, while at the same time that effect is itself
dependent on readers’ ability to negotiate narrative’s fictional mode of rep-
resentation. In some sense, then, in perceiving fictions we perceive them as
both elliptical (verisimilar) and circular (fictional). It should be noted that
for Noë the ‘full-​blooded duality’ of perception implies that experiencing
how things are and how they seem is possible without shifting attention.
Consequently, the view of the perception of fiction presented here differs
from the plot-​oriented prediction and hesitation effects examined by Karin
Kukkonen in her contribution to this volume, in that fictionality and the
fictional world are not like Wittgenstein’s duck-​rabbit—​two aspects impos-
sible to see at the same time—​but instead like seeing either one because we
see the lines of the famous drawing. That is, we see the lines as either a duck
or a rabbit, but in both cases we see both the lines and the animal (see Noë,
2012, pp. 21−22 and Polvinen, forthcoming).
The full ramifications of this analogy should not be seen only in the
context of specifically visual imagining, nor is this conceptualization likely
to support clear-​cut ontological categories of fiction and non-​fiction into
which narratives could be divided (see also Polvinen, in press). If applied
to the entirety of our complex engagements with fictions, a theory of enac-
tive perception of fictions as fictions demands much more work in terms
of teasing out the various emphases in different texts. For this kind of
work we can draw on literary scholarship which relies on ideas of praxis
recognizable to those working within the enactive paradigm. For example,
Joshua Landy’s volume How to do things with fictions (2012) combines lit-
erary analysis, hermeneutics and the philosophy of narrative fiction, and
argues that one of the major functions of fictionality is to train the mind in
assuming particular mental states. ‘Rather than providing knowledge per
se,’ Landy writes, ‘whether propositional knowledge, sensory knowledge,

3.  For a challenge of Noë’s argument concerning sensorimotor knowledge, see e.g.
Hutto and Myin (2013), who argue for a more radical version of contentless cognition.

D o u b l e Visi o n of Fi c t i o n   [ 141 ]
 

knowledge by acquaintance, or knowledge by revelation—​what [fictions

can] give us is know-​how; rather than transmitting beliefs, what they equip
us with are skills; rather than teaching, what they do is train. They are not
informative, that is, but formative’ (Landy, 2012, p. 10). Of particular
interest is the way Landy bases the power of these ‘formative’ fictions on
the fact that in order to do the training they do, they must generate in
their readers a state of ‘conviction and distrust’ in the enchantment being
offered (ibid., p. 76). In his analyses of texts that range from the Gospel
of Mark to Mallarmé and Beckett, Landy thus lays open a form of writing
that connects with its readers most acutely on a level of ‘lucid self-​delusion’
(p. 12). The combination of engagement with what seems to be the case,
and awareness of the fact that we are being presented with an illusion, is
a seemingly paradoxical mental state that is nevertheless required of us
when experiencing such fictions. At the same time, the fictions themselves
hone our skill in entering that state to an enduring and easy habit (see also
Landy, 2015, p. 572).
Landy’s ‘lucid self-​delusion’ follows a tradition running from Aristotle’s
mimesis (see Halliwell, 2002)  to Coleridge’s willing suspension of
disbelief—​of seeing fiction as something that calls not for a loss of a sense
of reality, but for the maintenance of a dual attitude.4 It might be argued
that these two aspects of engagement with fiction should be seen as dis-
tinct processes, one a low-​level and intuitive perceptual process, the other
a conscious and culturally organized process of interpretation (e.g. Hutto
and Myin, 2013, p. xviii). However, it is crucial to this view of fictionality
to recognize that perceptual and interpretive processes are always inter-
twined in reading, and that there are qualities in fictions which are avail-
able to the audience only when they use specific fiction-​related cognitive
skills. Such is the argument made by Richard Walsh in Rhetoric of fiction-
ality (2007), where fictionality is presented as a communicative strategy
built into works of fiction by authors, and recognized as such by readers.
Fictionality is therefore a rhetorical mode that changes the way readers
comprehend the thing being represented: ‘awareness of its artifice is innate
in any response whatsoever to fiction as such’ (Walsh, 2007, p. 172). Thus,
losing sight of fictionality as a quality of the text would mean readers are no
longer experiencing fiction but have, instead, slipped into a non-​fictional
mode of reading.

4.  Further work in this tradition includes Wolfgang Iser’s The fictive and the imagi-
nary (1993) and Paul Ricoeur’s three-​level mimesis in Time and narrative (1983‒1985/​
1984‒1988).

[ 142 ]  Cognition through a Literary Lens

 

By combining these 4E-cognitive and literary-​theoretical perspectives,

I suggest, we can understand the experience of fiction as a form of enac-
tive perception where fictionality, instead of being a quality that fades to
invisibility in order to be effective, is rather the perspective that makes per-
ception and comprehension of fiction as fiction possible. Furthermore, our
experience of fictionality need not clash with our perception of and engage-
ment with the fictional world, but is conjoined with it. The enactive actual-
izing of the cognitive process presented by a fictional narrative should not
be understood only as a re-​enactment of a character’s experience (as the
verisimilar content of the fictional representation), but must instead be
seen to involve the discourse patterning of the narrative—​everything from
individual linguistic details to its communicative status as a work of fiction.

FICTIONALITY AND THE PRODUCTS OF MAGIC: THE PRESTIGE

In the space remaining, I will extend this theoretical discussion to the anal-
ysis of a novel which I  believe cashes in on those imaginative processes
which the enactive approach to fictionality brings to light. Christopher
Priest’s The Prestige (1995/​2004) is a meditation on the 19th century and
its tensions between spiritual and materialist sensibilities. This is a cultural
moment that many other writers have approached through the spiritualist
séances popular at the time (e.g. A. S. Byatt in her 1990 novel Possession),
but Priest chooses as his entry point a feud between two stage magicians.
In the novel, this conflict draws on two conceptualizations of magic, either
as naturalized craft or as actual supernatural power, and the novel itself is
similarly built on a conflict between naturalized narrative puzzles and fan-
tastical story events. As a result, The Prestige has resisted easy categoriza-
tions, and has been cited as science fiction or fantasy, and as Neo-​Victorian
metafiction, having won both the mainstream James Tait Black Memorial
Prize and the World Fantasy Award.
The novel opens in the present day with a young journalist receiving
a copy of the diary of his Victorian ancestor, Alfred Borden.5 Borden was
a hard-​working tradesman’s son who taught himself conjuring tricks and
eventually made his way to the stage under the name ‘Le Professeur de
Magie’. In addition to the story of Borden’s life, the diary includes sections

5.  For an analysis of Christopher Nolan’s film version from 2006, where some sub-
stantial changes were made to the way the battle between the magicians is presented to
the audience, see Heilmann (2009/​2010). The modern-​day frame of the novel is hon-
estly less interesting than the Victorian magicians, and it was left out of Nolan’s film.

D o u b l e Visi o n of Fi c t i o n   [ 143 ]
 

where he explains his own attitude towards the secrets of his art. ‘Magic
has no mystery to magicians’, Borden believes:

We work variations of standard methods. … Every illusion can be explained,

be it by the use of a concealed compartment, by an adroitly placed mirror, by an
assistant planted in the audience to act as ‘volunteer’, or by simple misdirection
of the audience’s attention. (Priest, 1995/​2004, p. 66)

Borden’s most famous stage trick is exactly such a naturalizable mystery: in

‘The Transported Man’ he steps into a cupboard on one side of the stage,
seems to cross the entire width of that space instantaneously, and steps
out of another cupboard on the opposite side. The pleasure and thrill of the
trick, as Borden emphasizes in his writings, is that everyone knows that no
magic is actually involved, and that instead the mystery is created by skilled
misdirection. The preservation of the mystery itself, however, is crucial,
and Borden is willing to go to insane lengths to maintain it. For what the
diary slowly reveals is that ‘The Transported Man’ is made possible by the
fact that Borden is actually a pair of twins who, in order to protect the secret
of their trick, take turns to live the life of a single individual—​sharing their
wife and children, as well as their mistress, and never letting on that they
are, in fact, not one man but two.
Misdirection is also present in the form of Priest’s novel, as the diary
performs its own narrative trick on readers. Written entirely in the first-​
person singular—​even though the two brothers take turns narrating their
story—​the diary simultaneously presents and conceals the solution to its
own mystery:

I write in the year 1901.

My name, my real name, is Alfred Borden. The story of my life is the story of
the secrets by which I have lived my life. …
First let me in a manner of speaking show you my hands, palms forward,
fingers splayed, and I will say to you (and mark this well): ‘Every word in this
notebook that describes my life and work is true, honestly meant and accurate
in detail.’ (Priest, 1995/​2004, pp. 31‒34)

Just like the audience of stage magic, the readers of Priest’s novel receive
this announcement of honesty from Borden, and both audiences acquiesce
to experiencing the mystery that follows. Even the very first word of the
diary, ‘I’, is simultaneously a truth and a lie, one that readers are designed
to accept at face value at first, but whose duplicity is made explicit later on.
Borden’s secret remains a secret to his stage audiences, but in the diary it is

[ 144 ]  Cognition through a Literary Lens

 

eventually naturalized by the revelation of the twins’ unorthodox life and

life-​writing. This process takes up most of the first half of the book, mak-
ing it read like a realist historical novel, with the twist of a psychological
thriller.
This effect, however, is offset by the fate of Priest’s second diary-​writing
magician, Rupert Angier. Angier is in some ways Borden’s opposite—​the
younger son of an Earl, he is wealthy and educated—​but he does share
Borden’s obsessive drive to succeed as a magician. The fates of the two
men are entangled by an accident that sets them in a spiral of compe-
tition and revenge, sabotaging each other’s performances with tragic
consequences. Against Borden’s methodical ‘Professeur’, Angier is the
showman—​enamoured with the magic act as spectacle. He finds Borden’s
performances to be impossibly skilful, yet banal in their use of ‘standard
stage trickery’. But after trying and failing to produce the uncanny effect
of Borden’s ‘Transported Man’ by engaging a stage double, Angier decides
to match the same level of illusion by producing the truly impossible. He
obtains from Nicola Tesla an electrical machine which transports him
from the stage to the back of the theatre instantaneously. The trick is a
huge success and it establishes Angier’s career as one of the greatest magi-
cians of his age. But because his audience expects to be engaging with
an act of conjuring rather than with reality, Angier now needs to hide an
actual scientific sensation: although his trick ‘by scientific method, in fact
achieves the hitherto impossible’, he ‘cannot allow this ever to be known,
for science has in this case replaced magic’ (Priest, 1995/​2004, p.  282).
Even though the truth of his trick is a scientific miracle, Angier strives
to maintain the traditional, ‘magical’ audience relation. ‘By careful art’,
he has to ‘make [his] miracle less miraculous’ in order to have it accepted
as magic.
Tesla’s machine is, of course, an impossibility in our reality, and its pres-
ence transforms the historical realism of Priest’s novel into science fiction
or steampunk. It also ushers in the Gothic and the grotesque. For, strictly
speaking, Tesla’s machine does not transport anyone anywhere; instead it
places a living copy of the person at any location chosen. The Angier who
steps into the machine is thus copied every night he performs the trick,
and the inert but living original is each time secreted away from the theatre
and placed in Angier’s family vault. This goes on until the night Borden
interferes with the trick, resulting in the ‘original’ Angier remaining mobile
(though weak) and the new copy gaining only a ghostlike existence. Angier,
like Borden, becomes two halves of a single man, but in this case neither
half is physically viable. The original Angier is eventually riddled by cancer
and dies, while the copy remains mostly insubstantial.

D o u b l e Visi o n of Fi c t i o n   [ 145 ]
 

Angier’s diary knowingly repeats Borden’s diary illusion, as both the

original but physically damaged version and the final surviving but incorpo-
real version of Angier use the first-​person singular in writing it. The switch
between the two occurs initially without express signal, but the narrative
situation of the diary eventually alerts readers by becoming seemingly
impossible, with Angier describing his own unconsciousness and paraly-
sis and finally even his own death: ‘At a quarter to three this morning my
life was brought to its end by a sudden seizure of the heart’ (Priest, 1995/​
2004, pp. 323‒325). But unlike Borden’s diary, this time the narrative has
dropped enough hints about the doubling to make it reasonably easy for
readers to understand that Angier is writing about the death of his other
half. And even those readers who are caught by this first-​person-​for-​two-​
men trick for the second time are quickly let off the hook by Angier making
explicit reference to Borden’s doubled voice: ‘I have borrowed a technique
from Borden, so that I am I as well as myself’ (p. 325).
Yet, despite the fact that Angier’s narrative situation is naturalized to
an extent—​he turns out not to be an undead man speaking from beyond
the grave—​the fact of his doubling into corporeal and incorporeal versions
is itself a deviation from the rules of our reality. Thus with Angier, we are
no longer able to explain the doubled man as a psychologically twisted but
ultimately possible set of twins, and Priest no longer continues to operate
within the naturalist or realist literary tradition. The genre of science fic-
tion has been for a long time theorized mainly through its presentation
of scientifically believable speculation (e.g. Spiegel, 2008)—​an approach
which relies on the assumption that invented technology inherently offers
more cognitive grounding than the supernatural phenomena typical to
fantasy. In The Prestige, Priest clearly plays with that assumption by hav-
ing Angier’s impossible magic act be made plausible by a machine invented
by a historical person. But even the science-​fictional naturalization is only
the first step in the process dominating Angier’s story. During his career,
he was able to fool his theatre audiences into continuing to take the fic-
tional attitude towards his performances, but the full, grotesque conse-
quences of actual impossibility are represented to the reading audience.
These are manifested, first, by the description of the frame-​tale narrator’s
final descent into the crypt among dozens of immobile but still conscious,
rubbery Angier copies—​undead interstitial beings (Csicsery-​Ronay, 2008,
pp. 195‒198) that are all the more horrifying for appearing in a series of
absurd poses:

The corpses all lay in different positions. Some were straight, others were twisted
or bent over. None of the bodies was arranged as if lying down; most of them

[ 146 ]  Cognition through a Literary Lens

 

had one foot placed in front of the other, so that in being laid on the rack this leg
was now raised above the other.

Every corpse had one foot in the air. …

I eased myself backwards, not looking. As I reached the main aisle and turned
slowly around, [I]‌brushed against the raised foot of the nearest corpse. A patent-​
leather shoe swung slowly to and fro. (Priest, 1995/​2004, pp. 354‒357)

Furthermore, intertextual allusions at the end of the novel usher in the

ghosts of both Mary Shelley’s Frankenstein and Robert Luis Stevenson’s
Dr. Jekyll and Mr. Hyde, as the final phantom-​Angier, re-​embodied in one of
his own earlier copies, disappears off into a blizzard. Thus this novel pres-
ents its readers with multiple perspectives on the phenomenon of magic—​
as stage performance, as supernatural power, and as fiction.
The motivation for using stage magic as an analogy for fiction in this
chapter—​and, I believe, in Priest’s novel—​is to emphasize the roles of the
two aspects of fictionality I wanted to focus on: the audience’s expectation
of unreality and their engagement with the entire act of conjuring, rather
than just with the world or the actions it seems to represent. It is through
the contrast between the two magicians and their attitudes, as well as
the genre conventions of realism and the fantastic adopted in the novel,
that Priest expressly signals to his readers the kind of enaction they are
engaged in. That signalling gesture, like the empty hands of the magician,
is designed to remind readers to engage with the fiction as fiction, with ‘a
verbal performance in which the events depicted never happened, and in
which everyone knows they didn’t’ (Landy, 2012, p. 3). But the central pur-
pose of underlining this role of artificiality in fiction is not to undermine
the readers’ enactment—​on the contrary. For even while we are enacting
the experiences represented in the fiction, what is also being enacted is
the work of fiction itself. And what follows from that enaction is not what
would follow from engagement with reality—​a point underlined by the title
of Priest’s novel. ‘The prestige’ refers, in the vocabulary of stage magic, to
the ‘effect’ or the ‘product of magic’, to that which ‘apparently did not exist
before the trick was performed’ (Priest, 1995/​2004, p. 65). Angier’s trick
produces copies of himself, whereas Borden talks about his whole bizarre
life as the prestige of ‘The Transported Man’. But fiction, I suggest, can also
produce in readers its own prestige, something that emerges as if by magic
from the process of readerly enactment. The prestige of fiction is not a rab-
bit pulled out of a hat, nor something as grotesque as Angier’s copied bod-
ies; it is a cognitive state of lucid self-​delusion in readers, which would not
exist without having been performed through active complicity in a fiction

D o u b l e Visi o n of Fi c t i o n   [ 147 ]
 

as fiction, and with full knowledge of the meaning-​making actions required

in that performance.

CONCLUSION

Fictions are interactive cognitive environments that require from read-

ers a combination of skills that is much more complex and seemingly
self-​contradictory than the traditionally computational cognitive sciences
assume. In recent years, however, it has become clear within the cognitive
sciences too that the use of residual common-​sense concepts may lead schol-
ars into making oversimplifying claims. As Howard Casey Cromwell and Jaak
Panksepp (2011) have noted, the cognitive and behavioural neurosciences
sometimes fall prey to a form of circularity in their attempts to accurately
describe the workings of the human brain. Citing the NYU neuroscientist
Gyorgy Buzsaki, they note how the conceptual structures created through
slow processes of tradition can lead empirical research to merely reproduce
those structures, to take ‘a man-​created word or concept … and search
for brain mechanisms that may be responsible for the generation of this
conceived behaviour’ (Cromwell and Panksepp, 2011, p. 2034). In a similar
fashion, the risk I see in some of the recent cognitive and empirical studies
of fiction is that they uncritically adopt apparently common-​sense concepts
that in fact derive from the rational/​computational tradition. Such studies
may end up just confirming the preconceptions of the researchers because
of the way the conceptualizations guide the set-​up of the experiments, in
for example the verbal instructions or choices provided in questionnaires,
or even because they limit the vocabulary available for the participants to
describe their experiences (see Troscianko, 2013, pp. 190‒191).
My aim here has therefore been to show how, by focusing on fictional-
ity, cognitive literary studies is better able to analyse those functions and
effects that arise from the artefactual nature of the text, instead of focus-
ing on just the effects that are thought to exist in spite of that artefactu-
ality. Secondly, I have tried to suggest that the cognitive sciences might
benefit from the ideas and analyses presented by literary studies concern-
ing the complexities involved in the processing of fictional narratives, and
the ways in which that processing is unlike the forms of immersive illu-
sion often used as the paradigm cases of literary imagining. While there
is already much intriguing data relating to our imaginary abilities coming
from the empirical neurosciences, the results of such studies will be dif-
ficult to interpret as long as the underlying conceptualizations are still
rooted in oversimplifying models. By combining the theory of fictionality

[ 148 ]  Cognition through a Literary Lens

 

and the enactive paradigm of perception and imagination, the cognitive

sciences might be able to further sharpen their own conceptual and termi-
nological apparatus and continue to generate fruitful hermeneutic circula-
tion between different approaches to the mind and the imagination.

REFERENCES
Altmann, U., Bohrn, I. C., Lubrich, O., Menninghaus, W., and Jacobs, A. M. (2014). Fact
vs fiction:  How paratextual information shapes our reading processes. Social
Cognitive and Affective Neuroscience, 9, 22–​29. doi:10.1093/​scan/​nss098
Appel, M., and Richter, T. (2007). Persuasive effects of fictional narratives over time.
Media Psychology, 10, 113–​134. doi: 10.108/​15213260701301194
Bae, H.-​S., Lee, D., and Bae, R. E.  G. (2014). Emotional engagement with the plot
and characters:  A  narrative film on hearing-​impaired sexual assault victims.
Narrative Inquiry, 24(2), 309–​327.
Caracciolo, M. (2011). The reader’s virtual body: Narrative space and its reconstruction.
Storyworlds, 3, 117–​138.
Caracciolo, M. (2014). Tell-​ tale rhythms:  Embodiment and narrative discourse.
Storyworlds, 6(2), 49–​73.
Cromwell, H. C., and Panksepp, J. (2011). Rethinking the cognitive revolution from a
neural perspective. Neuroscience and Biobehavioral Reviews, 35, 2026–​2035.
Csicsery-​Ronay, I., Jr. (2008). The seven beauties of science fiction. Middletown, CT:
Wesleyan University Press.
Danta, C., and Groth, H. (2014). Introduction:  Between minds. In Mindful aesthet-
ics: Literature and the science of mind (pp. 1–​14). New York: Bloomsbury.
Davis, M. H. (1980). A multidimensional approach to individual differences in empa-
thy. JSAS Catalog of Selected Documents in Psychology, 10, 85. Retrieved from
http://​www.eckerd.edu/​academics/​psychology/​files/​Davis_​1980.pdf.
Denis, M. (2001). Mental imagery, Psychology of. In N. J. Smelser and P. B. Baltes
(Eds.), International encyclopedia of the social and behavioral sciences (pp. 9675–​
9681). Amsterdam: Pergamon.
Esrock, E. J. (2001). Touching art: Intimacy, embodiment, and the somatosensory sys-
tem. Consciousness and Emotion, 2(2), 233–​253.
Gallagher, S., and Zahavi, D. (2007). The phenomenological mind. London: Routledge.
Gallese, V., Keysers, C., and Rizzolatti, G. (2004). A unifying view of the basis of social
cognition. Trends in Cognitive Sciences, 8(9), 396–​403.
Green, M. C., and Brock, T. C. (2000). The role of transportation in the persuasiveness
of public narratives. Journal of Personality and Social Psychology, 79(5), 701–​721.
doi:10.1037//​0022-​3514.79.5.701
Halliwell, S. (2002). The aesthetics of mimesis:  Ancient texts and modern problems.
Princeton, NJ: Princeton University Press.
Heilmann, A. (2009/​2010). Doing it with mirrors: Neo-​Victorian metatextual magic in
Affinity, The Prestige and The Illusionist. Neo-​Victorian Studies, 2(2), 18–​42.
Hutto, D. D., and Myin, E. (2013). Radicalizing enactivism: Basic minds without content.
Cambridge, MA: MIT Press.
Iser, W. (1993). The fictive and the imaginary: Charting literary anthropology. Baltimore,
MD: Johns Hopkins University Press.
Jajdelska, E., Butler, C., Kelly, S., McNeill, A., and Overy, K. (2010). Crying, moving,
and keeping it whole: What makes literary description vivid? Poetics Today, 31,
433–​463.

D o u b l e Visi o n of Fi c t i o n   [ 149 ]
 

Kidd, D. C., and Castano, E. (2013, 18 October). Reading literary fiction improves
Theory of Mind. Science, 342, 377–​380, with online supplementary material.
doi: 10.1126/​science.1239918
Kuzmičová, A. (2012). Presence in the reading of literary narrative: A case for motor
enactment. Semiotica, 189(1), 23–​48. doi: 10.1515/​semi.2011.071
Landy, J. (2012). How to do things with fictions. Oxford: Oxford University Press.
Landy, J. (2015). Mental calisthenics and self-​reflexive fiction. In L. Zunshine (Ed.),
The Oxford handbook of cognitive literary theory (pp. 559–​580). New York: Oxford
University Press.
Menary, R. (2009). Introduction to the special issue on 4E cognition. Phenomenology
and the Cognitive Sciences, 9, 459–​463. doi: 10.1007/​s11097-​010-​9187-​6
Noë, A. (2004). Action in perception. Cambridge, MA: MIT Press.
Noë, A. (2012). Varieties of presence. Cambridge, MA: Harvard University Press.
Phillips, N. M. (2015). Literary neuroscience and history of mind: An interdisciplinary
fMRI study of attention and Jane Austen. In L. Zunshine (Ed.), The Oxford hand-
book of cognitive literary theory (pp. 55−81). New York: Oxford University Press.
Polvinen, M. (in press). Enactive perception and fictional worlds. In P. Garratt (Ed.), The
cognitive humanities: Embodied mind in literature and culture. London: Palgrave.
Priest, C. (1995/​2004). The prestige. London: Gollanz.
Richardson, A. (2011). Defaulting to fiction:  Neuroscience rediscovers the Romantic
imagination. Poetics Today, 32(4), 663–​692. doi: 10.1215/​03335372-​1459845
Richardson, A. (2015). Imagination:  Literary and cognitive intersections. In L.
Zunshine (Ed.), The Oxford handbook of cognitive literary theory (pp. 225–​245).
New York: Oxford University Press.
Ricoeur, Paul. (1984–​1988). Time and narrative, Vols. 1–​3 (K. McLaughlin and D.
Pellauer, Trans.). Chicago, IL: University of Chicago Press.
Sanford, A. J., and Emmott, C. (2012). Mind, brain and narrative. Cambridge: Cambridge
University Press.
Spiegel, S. (2008). Things made strange: On the concept of ‘estrangement’ in science
fiction theory. Science Fiction Studies, 35(3), 369–​385.
Stewart, J., Di Paolo, E. A., and Gapenne, O. (2010). Introduction. In J. Stewart, O.
Gapenne, and E. A. Di Paolo (Eds.), Enaction: Toward a new paradigm for cognitive
science (pp. vii–​xvii). Cambridge, MA: MIT Press.
Thompson, Evan. (2007). Mind in life: Biology, phenomenology, and the sciences of mind.
Cambridge, MA: Belknap.
Troscianko, E. T. (2013). Reading imaginatively: The imagination in cognitive science
and cognitive literary studies. Journal of Literary Semantics, 42(2), 181–​198.
Walsh, R. (2007). The rhetoric of fictionality. Columbus, OH: Ohio State University Press.
Walton, K. L. (1990). Mimesis as make-​believe: On the foundations of the representational
arts. Cambridge, MA: Harvard University Press.
Wolf, W. (2013). ‘Unnatural’ metalepsis and immersion: Necessarily incompatible? In
J. Alber, H. S. Nielsen, and B. Richardson (Eds.), A poetics of unnatural narra-
tive (pp. 113–​141). Theory and Interpretation of Narrative. Columbus, OH: Ohio
State University Press.

[ 150 ]  Cognition through a Literary Lens

 

CHAPTER 8

Fantastic Cognition
KARIN KUKKONEN

Our brains build models of the world and continuously modify these models on the basis
of the signals that reach our senses. So, what we actually perceive are our brain’s models
of the world. They are not the world itself, but, for us, they are as good as. You could say
that our perceptions are fantasies that coincide with reality.
Frith, 2009, pp. 134‒135

Put simply, the brain is—​literally—​a fantastic organ (fantastic: from Greek phantastikos,
able to create mental images, from phantazesthai).
Friston, 2013, p. 1328

Reality or dream? truth or illusion?

Which brings us to the very heart of the fantastic.
Todorov, 1970/​1975, p. 25

A ccording to recent developments in predictive processing and Bayesian

cognition, our thinking is fantastic: we grasp the world through predic-
tive, probabilistic models that we compare against the feedback from the
actual environment, which really only plays a significant role if it proves
our predictions to be wrong and forces us to revise the probabilistic model.
Rather than our eyes and brains registering every detail of the environment
and configuring it into the larger whole of the percept, perception works the
other way around. We already know what we are likely to perceive on the
 

basis of our predictive probabilistic models (or ‘fantasies’) and correct these
only if discrepancies with the environment create prediction errors. The
predictive processing model of cognition, with its feedback loops of virtual
models and prediction errors, has in recent years been extended into many
other areas of the cognitive sciences, from perception to motor control,
from emotions to our sense of self (for comprehensive overviews, see Clark,
2013, and Hohwy, 2013). Not only perception but cognition more generally
might turn out to be nothing short of fantastic.
The predictive, probabilistic models of perception and cognition which
Frith and Friston gloss as ‘fantasies’ are, as we shall see, usually not noticed
in everyday life. Tzvetan Todorov’s definition of the literary genre of ‘the
fantastic’, as a genre that makes us hesitate between ‘truth’ and ‘illusion’,
however, suggests that literary texts might often test boundaries and high-
light the workings of such cognitive models. As I will go on to show, the
fantastic in literature throws ‘fantastic cognition’ into relief.

Karl Friston, one of the leading proponents of the predictive processing

model in the cognitive sciences, makes his statement about the brain as a
‘fantastic organ’ in a place that is perhaps unexpected: a review of a book
on art. Eric Kandel’s (2012) The Age of Insight: The Quest to Understand the
Unconscious in Art, Mind and Brain, from Vienna 1900 to the Present begins
by painting a lively picture of the intellectual scene in Vienna around the
turn of the 20th century, where psychologists exchanged views with the
most progressive artists, such as Gustav Klimt, Egon Schiele, and Oskar
Kokoschka. From this, Kandel moves into a more general overview of
current neurological investigations into the perception of art, in particu-
lar its visual properties (as in the work of Semir Zeki) and its empathy-​
evoking capacities (as in the work of Chris and Uta Frith). He makes a
double claim about the interaction between visual arts and the cogni-
tive unconscious. On the one hand, the formal features of artistic style
exaggerate what we would see in the real world (think of Klimt’s gold-​
studded portraits), and thereby guide the perceiver’s attention in particu-
lar ways. On the other hand, the creative work of artists can call attention
to the usually unconscious processes of perception, in particular to what
Hermann von Helmholtz called ‘unconscious inference’ (through which
we match predictive models with the visual input before actually perceiv-
ing something).
In a way, Kandel’s book recreates the Kaffeehaus exchanges between sci-
ence and art by bringing together a rich catalogue of the work of Klimt,
Schiele, and Kokoschka with a very detailed account of the latest research
on the cognitive processes at play when perceiving such paintings. Friston

[ 152 ]  Cognition through a Literary Lens

 

chooses to home in on one particular aspect of this exchange: the role which

expectations play in the perception of art. This is not surprising, given that
he is the champion of predictive-​processing approaches in psychology.
Kandel’s book reminds us also, however, that predictions and expectations
have played a central role in art theory for a long time. The work of Ernst
Gombrich, for instance, who himself has Viennese roots, brings together
art perception with the psychological research of J. J. Gibson and William
James in his seminal Art and Illusion: A Study in the Psychology of Pictorial
Representation (1960/​2002). Gombrich foregrounds ‘the beholder’s share’
in the co-​construction of artistic images. The paintings and statues that
most successfully engage viewers are usually not those with the smoothest
finish and the most precise detail, but rather those that leave the beholder
space to complete details and thereby engage with the work in more cog-
nitively profound ways. Artists provide, for example, blurry pictorial ele-
ments that lead perceivers to reconstruct more vivid details (Gombrich
mentions here the late work of Rembrandt and the Impressionists) or sim-
plified outlines of facial features that let perceivers focus on the emotional
import of the expression. Because perceivers know from the context (of the
man and the coat in, say, Rembrandt’s Portrait of Jan Six) that they should
perceive a set of gold buttons where there are blurry lines and knobs of gold
and black on the canvas, they do—​and, according to Gombrich, they imag-
ine a richer play of lights reflecting off these buttons to boot.
To some extent, Gombrich reimagines the history of (mostly Western)
art as a history of artists devising ever more refined tools to engage the
beholder and to put her ‘share’ in the image to ever more sophisticated
uses. They do not strive towards an exact representation of reality, though.
Instead, artists (often wilfully) engage in visual fraud and use shortcuts of
representation that distort what perceivers would actually see in the real
world, but at the same time engage our schemata of perception (or rather,
our expectations) more efficiently. Not just in the real world, as work on
inattentional blindness and change blindness suggests (O’Regan and Noë,
2001), but also in the perception of art, we fall prey to surprising lapses
in perception. Perceivers do not notice impossible mirror angles, contra-
dictory foreground-​ background compositions or inconsistent shadow-
ing (Cavanagh, 2005). These typical mistakes in the perception of art are
exploited by artists to create engaging images (Gombrich argues). At the
same time, these techniques, usually devised by the artist after long experi-
mentation with the effect in interplay with intuition, give scientists access
to the ‘simplified physics’ which the brain uses to recognize scenes. In a
Nature article of 2005, Patrick Cavanagh details how the typical shortcuts
in perception that I listed above depend upon the ‘physiology of the visual

Fa n ta s t i c C o g n i t i o n   [ 153 ]
 

brain’ (p. 301). He makes the claim that the artistic process constitutes ‘a
type of found science’ about perceptual shortcuts and the simplifications
which the brain employs when recognizing objects and scenes. Similarly,
in Art and Illusion, Gombrich discusses Constable’s cloud studies, showing
how the artist draws sketches and changes their set-​up systematically, as
he gets a grip on his visual possibilities as an artist. This practice arguably
constitutes the experimental set-​up of ‘found science’. Gombrich writes, ‘I
think [Constable] felt that the history of science presented a story of con-
tinuous advance in which the achievements of one observer were used and
extended by the next’ (p. 175).
It seems that this discussion of Gombrich’s ‘beholder’s share’ and the
‘simple physics’ of artistic shortcuts have taken us a long way away from
predictive processing (which indeed does not play quite as central a role in
Kandel’s book as Friston makes out, and which Cavanagh does not mention
at all). Friston, however, is right to connect Gombrich’s ‘psychology of art’
with predictive processing. The mimesis of the work of art in Gombrich
is based not on its truthful representation of the real world, but on the
degree to which it engages the expectations of the spectator, thereby cre-
ating the illusion of mimesis that is both more artificial and more power-
ful. The viewer needs to be given both the opportunity and the means to
‘project what is not there’, through under-​defined elements on the canvas
(a ‘screen’), and also clues for what inference to draw (Gombrich, 1959/​
2002, p. 232). The viewer’s predictive models take centre stage, as artists
devise depth compositions and colour constellations which are empirically
incorrect but do not create immediate prediction errors. These instances
can give us insights into the make-​up of the predictive model, not only for
the perception of art but also for the perception of the rest of the world.
Remember Chris Frith’s suggestion, quoted at the beginning of this arti-
cle, that ‘perceptions are fantasies that coincide with reality’. For him, the
brain literally ‘makes up’ the mind in that it constitutes the physical basis
of cognitive operations. The brain ‘makes up’ the mind in the extended
sense as well, because predictive processing in the brain gives rise to cog-
nitive illusions, such as the assumption that our minds are isolated and
private. It would take up too much space to go into the details of Frith’s
delightful account here, but basically, he suggests that cognition generally
relies on Helmholtz’s ‘unconscious inferences’ in visual, proprioceptive,
interoceptive, and other kinds of perception. We perceive, feel, intend, and
think after these unconscious inferences have related the sensory stimulus
to our predictive, probabilistic models. Like Gombrich’s notion of artistic
mimesis, Frith’s concept of cognition depends on expectations, understood
as virtual models. Predictive models, for example, readjust our perception

[ 154 ]  Cognition through a Literary Lens

 

of the source of sound when watching TV from the speakers of the TV

set (where it actually comes from) to the mouth of the speaking actor on
screen (Hohwy, 2013, p. 131). As Frith puts it, ‘For us to act upon the world
it doesn’t matter whether or not our brain’s model is true. All that matters
is that the model works’ (2009, p.  136)—​and, he goes on to add, that it
matches the models of others.
Artistic styles and strategies, as Cavanagh and Kandel suggest, might
offer one way of getting to the bottom of these predictive models (or fan-
tasies) that are so important for our perception. Indeed, art historians like
Gombrich who carefully trace the emergence and development of these
stylistic devices (or artistic shortcuts) in light of the cognitive schemata
and predictive models they engage can provide the sciences with hypoth-
eses to test in the Bayesian paradigm. As Ladislav Kesner (2014) points
out, the same might be true for Erwin Panofsky’s notion of ‘iconography’
(conventional features of images that lead to the identification of the char-
acters and the scene depicted), for example, and Michael Baxandall’s (1958)
notion of the ‘period eye’ (historical ideas about visual perception that find
their way into the presentation of reality in the image). Indeed, accord-
ing to Kesner, pre-​modern works of art (such as Chardin’s Lady taking tea,
which Baxandall discusses in detail, pp. 74‒104) contain their own ‘script
for action’ which art historians can trace (Kesner, 2014, p. 10). These pic-
tures provide viewers both with a particular artistic vision and with inbuilt
instructions (through compositional arrangements or references to con-
temporary debates around perception) for how the images should be per-
ceived. In other words, works of art not only engage the predictive models
on which (in the Bayesian paradigm) our cognition is based, but they also
offer viewers little prediction errors that cue them to refine their predictive
model in a particular way.

From our discussion of visual perception and artistic styles emerges a rela-
tively well-​defined research programme for science and visual art within
a predictive processing framework. Can we posit a similar kind of ‘found
science’ for literature? Within the growing field of the cognitive study of
literature, the historical sciences have been taken into account for the dis-
cussion of cognitive phenomena in the texts of earlier periods (see e.g.
Crane’s Shakespeare’s Brain [2001], Richardson’s The Neural Sublime [2010],
and Anderson’s The Renaissance Extended Mind [2015]), and there are note-
worthy attempts to make sense of neurological evidence through the inter-
pretive paradigms of the humanities. In Feeling Beauty (2013), G. Gabrielle
Starr discusses neurological studies of the sister arts (literature, painting,
and music) through Baumgarten’s notion of the aesthetic as ‘a blend of

Fa n ta s t i c C o g n i t i o n   [ 155 ]
 

sensation and knowledge such that we almost feel thought itself’ (p. xiv).
Paul B. Armstrong, in How Literature Plays with the Brain (2013), combines
the seemingly contradictory neurological indications of our predilection
for familiarity and regularity, on the one hand, and novelty, on the other
hand, through the principle of the hermeneutic circle that develops out
of the interplay between what is known and what needs to be explained.
Both Starr and Armstrong give rather general statements about the inter-
pretive models they propose, rather than tracing a specific set of cognitive
shortcuts through stylistic devices (as Cavanagh proposes) or unfolding a
historical narrative of the development of these devices through an archive
of artistic experimentation (as Gombrich does). With the predictive pro-
cessing and Bayesian approaches to literature still vastly underrepresented
in cognitive literary studies, the value of literary stylistic and narrative
features that create particularly instructive and insightful instances of
what I called ‘fantastic cognition’ in the title of this article still needs to be
asserted for literature.
Fantasy is traditionally the domain of literary study and is the subject of
many literary texts. So, in the interests of developing a ‘found science’ of
fantastic cognition from a literary point of view, let us have a look at what
is generally considered the foundational text of the literary fantastic, used
to exemplify Todorov’s initial discussion of the term: Jacques Cazotte’s Le
Diable Amoureux (The Devil in Love, 1772/​1776/​1965).1 In Cazotte’s story,
young Alvare chooses to dabble in the dark arts, and he summons a crea-
ture that takes the shape of first a camel’s head, then a dog, and later the
page Biondetto. Sometimes, Alvare finds it really difficult to make out who
is in front of him:


Le feu de ses regards perçait à travers le voile, il était d’un pénétrant, d’une dou-
ceur inconcevables: ces yeux ne m’étaient pas inconnus. Enfin, en assemblant les
traits tels que le voile me les laissait apercevoir, je reconnus dans Fiorentina le
fripon de Biondetto; mais l’élégance, l’avantage de la taille, se faisaient beaucoup
plus remarquer sous l’ajustement de femme, que sous l’habit de page.

1.  Most of the works which we would intuitively classify as ‘fantasy’, such as Tolkien’s
Lord of the Rings or Martin’s Song of Ice and Fire, are in Todorov’s classification instances
of the ‘marvellous’, because here wizards and dragons actually exist in the fictional
world. The marvellous (le merveilleux) is a long-​standing term in literary criticism that
pertains to the divine interventions, supernatural passions, and larger-​than-​life nar-
ratives that are not uncommon, for example, in the romances of the 16th and 17th
centuries. The realist 18th-​century novel rejects it (generally), but with the rise of the
Gothic towards the end of the 18th century (of which Cazotte’s Diable Amoureux can be
considered a forerunner), the marvellous turns into a supernatural challenge for newly
won empirical epistemic certainties.

[ 156 ]  Cognition through a Literary Lens

 

(The fire of her gaze broke through the veil, it was of an inconceivable pen-
etration and sweetness: these eyes were not unknown to me. Finally, combin-
ing the features which the veil let me perceive, I recognized in Fiorentina the
mischief of Biondetto; but the elegance, the shapeliness of her waist were much
more remarkable in the guise of a woman than when dressed as a page) (p. 323)2

In the singer Fiorentina, he perceives the features of Biondetto, but in

turn, he considers that the female form fits Biondetto much better than
his outfit as a page. (Indeed, soon afterwards Biondetto is revealed to be a
seductive young woman: Biondetta).
Alvare’s perception seems to be akin to perceptual illusions such as the
duck-​rabbit and ‘binocular rivalry’, some of the most discussed issues in
relation to predictive processing in visual perception. In the experimental
paradigm of ‘binocular rivalry’, an image of a house, say, is presented to one
eye and a face to the other, and most participants’ perception switches back
and forth between perceiving a house and perceiving a face, rather than
combining the two into a single image. Similarly, Alvare looks at Fiorentina,
but perceives the eyes and the mischief of Biondetto. At the same time, the
female form seems to be more fitting for Biondetto than his original gender
suggests. Alvare is unsettled by the dual percept of Fiorentina the woman,
or Biondetto the young man, because these two options correspond to dif-
ferent predictive, probabilistic models he has which should not coincide.
In the predictive coding account of binocular rivalry, the reason why we
have trouble perceiving both the house and the face lies in the fact that our
expectations of the world (deeply ingrained in our bodily experience) do
not allow us to posit two such conflicting percepts in the same place—​no
matter what the actual visual stimulus is and no matter that participants
know very well that they are in a laboratory setting, where the expecta-
tions derived from the natural environment do not necessarily pertain (for
a more detailed discussion, see Hohwy, 2013, pp. 19‒23). Similarly, Alvare
struggles to see both Fiorentina and Biondetto in the same person, even
though he knows that the supernatural capacities of the creature he has
summoned would make such shape-​changing quite possible.
Clearly, the similarities between visual illusions (which depend on
predictive models working as they should) used in scientific experiments
and in artistic devices and styles across the history of visual art are more
straightforward, because they both deal with visual perception. In literary
texts, written language supplies a complex web of cues for reconstructing

2.  This and the following translations are mine.

Fa n ta s t i c C o g n i t i o n   [ 157 ]
 

perception, proprioception, actions, thought processes, and direct speech

cognitively. The linguistic cues draw on a plethora of cognitive modes
and, for the most part, they add a screen of representation to the events,
actions, and thoughts in the narrative, and this needs particular attention
when matching the cognitive process to the textual example. Nonetheless,
we can say that literary texts work through a similar strategy of mimesis as
the one that Gombrich identified for visual art. In the linguistic mode, too,
literary texts offer readers space to respond and specify their inferences (to
‘project’, in Gombrich’s words), and they also provide the necessary cues
for doing so. Both verbal and visual mimesis work through the interplay
between cognition and the crafted exploitation of our cognitive predictive
models through the text.3
Literary texts like Cazotte’s differ from pictures not only in the verbal
mode of representation but also in their explicitly narrative construction.4
Elsewhere (2014), I have discussed narratives as containing a probability
design, a feedback loop between the probabilities of the fictional world
and the events of the plot, which create prediction errors and force readers
to revise their predictive model of the fictional world. Narrative, in other
words, shapes our Bayesian inferences. What does this have to do with bin-
ocular rivalry and duck-​rabbits? It turns out that there is an interesting con-
nection between work on visual illusion and narrative. Jakob Hohwy notes
that when we are presented with pairs of ambiguous visual stimuli (such
as two duck-​rabbits), a change of belief in the situation can determine how
we perceive them. A weakly narrative contextualization, such as ‘the duck
is about to eat the rabbit’ (2013, p. 131) creates what Hohwy calls ‘cogni-
tive penetrability’ from higher-​level propositional expectation to cognitive
percept. The predictions of our conscious knowledge actually have an effect

3.  Indeed, in the years before cognitive approaches to literature, such an understand-
ing of literary art was developed in great detail through hermeneutic approaches to
literature. Critics speak of ‘Unbestimmtheitsstellen’ (‘spots of indeterminacy’; Roman
Ingarden, 1968) or ‘Leerstellen’ (‘empty places, gaps’; Iser, 1976) which the text leaves
for readers to fill in, thus contributing to the meaning-​making process. The artistic
reconfiguration of reality in the literary text leads readers to revisit their predictive
models and to refine their inferences in the ‘hermeneutic circle’ (Gadamer, 1960) or
in the engagement with ‘mimesis II’ (Ricoeur, 1984‒1988). This is a wide and complex
field in the history of literary criticism, which I cannot discuss in any detail here, but
these brief references hopefully go to show that a long tradition of pre-​cognitive liter-
ary theory suggests that something akin to the predictive processing account of predic-
tive, probabilistic models that work in interplay with (designed) prediction errors can
also be considered when discussing the reading of the literary work of art. But this is a
research programme which will need a separate treatment.
4.  Of course, images and in particular image sequences as found in comics can be
narrative, too.

[ 158 ]  Cognition through a Literary Lens

 

on our perception. Note that in the usual cases of visual illusion, there is
no such ‘penetrability’:  we know that this is a duck-​rabbit, but we only
ever perceive a duck or a rabbit, because other—​preconscious—​predictive
models apply. Hohwy cites only one study in support of this claim (Jensen
and Mathewson, 2011),5 but if he is right, this finding offers many excit-
ing avenues for cognitive literary study and the cognitive sciences, because
narrative then directly affects the interplay between conscious predictive
models and Helmholtz’s ‘unconscious inference’, and thereby shapes how
we employ predictive processing (Anderson, 2017). In what follows, I will
outline some possible case studies from Cazotte and the fantastic.

The ambiguity of Biondetto/​Fiorentina is indicative of the epistemic desta-

bilization that runs through Cazotte’s entire novella and which led Tzvetan
Todorov, in The Fantastic (1970/​1975), to posit Le Diable Amoureux as the
paradigmatic example of the genre of the fantastic. According to Todorov,
the key sign of the fantastic is the moment of hesitation over whether the
events in the narrative can be explained through supernatural forces or
through the machinations of a trickster. If the events are supernatural
within the fictional world, the text tilts into the marvellous (le merveilleux).
If they can be realistically explained, we get the uncanny (l’étrange). If the
text continues to hesitate until its very end, as is the case with Cazotte’s
novella, it is an example of the truly fantastic (le fantastique). Todorov
has defined the fantastic as ‘that hesitation experienced by a person who
knows only the laws of nature, confronting an apparently supernatural
event’ (p. 25). In other words, the fantastic is a literary genre which creates
tension between competing predictive models.
In Cazotte’s Diable Amoureux, readers are invited to hesitate between
two general predictive models that also describe the hesitation of the
fantastic: Alvare is seduced by the devil and Alvare is duped by the charlatan
Soberano versus All the events are hallucinated by Alvare. While the suspicion
that all is staged by Soberano can be dismissed relatively quickly, the jux-
taposition between the devil’s seduction (which would make the narrative
marvellous) and Alvare’s dream or hallucination (which would make the
narrative uncanny) remains. Each of these predictive models privileges dif-
ferent ‘unconscious inferences’ for readers. If the devil actually takes the
shape of Biondetta, then readers need to understand everything she says
as quite likely being part of a strategy to seduce Alvare. Her confession that

5.  Other stimuli used in earlier studies of binocular rivalry, such as grates with dif-
ferent orientations (Hohwy, 2013, p. 21, Figure 3), can perhaps also not as easily be
turned into narratives as the duck and the rabbit.

Fa n ta s t i c C o g n i t i o n   [ 159 ]
 

she is a sylph turned human, first to teach the magicians around Soberano
a lesson and second to be fit for Alvare’s love, will have to be read as a ruse.
The assassination attempt on Biondetta is then also probably set up by her
herself. If, on the other hand, Alvare dreams most of this, then we need
to start questioning the reliability of his narrative. Maybe he just imag-
ines the similarities between Fiorentina and Biondetto? Does he spin his
own narrative around the young woman Biondetta, perhaps in order to
mask the fact that he cannot commit to her? Does he wish to become a
charmer of the devil with such petulance that he simply imagines it rather
than practise actual witchcraft? After all, he admits himself that ‘jamais
rendez-​vous galant ne fut attendu avec tant d’impatience’ (‘Never was an
amorous rendezvous awaited with such impatience’, p. 318) as his meeting
with Soberano to conjure the devil.
The two general predictive models (or ‘hyper-​priors’, in the parlance of
predictive processing) reconfigure the textual evidence in conflicting ways.
These predictive, probabilistic models cue readers to pick up on different
elements of the text in the hope that they will confirm one hypothesis and
help to disambiguate the situation. As Friston puts it, ‘the raison d’être for
inference is to disambiguate among plausible and competing hypotheses’
(2013, p.  1329). The events in the narrative, the statements of the nar-
rators (and their slips of the tongue), serve readers as evidence to decide
which predictive model to apply to their reading of the text. Such infer-
ences are usually not part of the conscious experience in reading, mostly
because there is typically only one general predictive model for the fictional
world. In the feedback loop of what I call the ‘probability design’ of the liter-
ary narrative (Kukkonen, 2014), the plot events (and the new information
about the fictional world which they carry) leads to a revision of the pre-
dictive, probabilistic model of the fictional world. New observations hence
usher in a modification of the existing model that can be surprising and
unusual. The fantastic (in Todorov’s sense), however, systematically brings
the ‘unconscious inferences’ in literary reading to the fore because it forces
readers to hesitate between contradictory models, and thereby makes the
inferencing process in many instances more conscious.
Of course, not all the inferences in Diable Amoureux are foregrounded
explicitly through Alvare’s narration. Consider the following example, in
which Fiorentina appears at the dinner where Alvare is entertaining his
cabalist friends, right after he mentions to Biondetto that Fiorentina has
promised to attend:

‘Biondetto, dis-​je au page, la Signora Fiorentina m’a promis de me donner un

instant, voyez si elle ne serait point arrivée.’ Biondetto sort de l’appartement.

[ 160 ]  Cognition through a Literary Lens

 

Mes hôtes n’avaient point encore eu le temps de s’étonner de la bizarrerie

du message, qu’une porte du salon s’ouvre, et Fiorentina entre, tenant sa harpe,
elle était dans un déshabillé étoffé et modeste, un chapeau de voyage et un crêpe
très clair sur les yeux, elle pose sa harpe à côté d’elle, salue avec aisance, avec
grâce:  ‘Seigneur Don Alvare, dit-​elle, je n’étais pas prévenue que vous eussiez
compagnie, je ne me serais point présentée vêtue comme je suis, ces messieurs
voudront bien excuser une voyageuse.’
(‘Biondetto, I say to the page, Signora Fiorentina has promised me a moment
with her, see if she has not arrived.’ Biondetto leaves the apartment.
My guests had not had the time to be surprised by the strangeness of the
message when a door opens and Fiorentina enters, holding her harp. She was in
a sweeping and modest dishabille, with a travelling hat and a sheer veil in front
of her eyes. She puts her harp down beside her, gives a poised and graceful greet-
ing: ‘Signor Don Alvare, says she, I was not warned that you would have com-
pany, or I would not have presented myself dressed like this. Messieurs, please
excuse the appearance of a traveller.’) (p. 322)

Fiorentina appears with the same promptness with which Soberano’s

pipe had been refilled earlier, and it seems to go without saying that she
comes from the same supernatural source as the ‘promptitude merveil-
leuse’ (‘marvellous promptness’, p. 322) of the servants. Alvare’s offhand
comment to Biondetto serves as the summons to the singer—​an inference
which readers can make if they adopt the ‘supernatural’ super-​prior to the
situation. Then, the message has no ‘bizarrerie’ for readers and the infer-
ence remains preconscious (or ‘unconscious’ in Helmholtz’s sense).
Fiorentina’s ‘deshabillé’ and her excuses about it, on the other hand,
suggest something unexpected. If she is indeed summoned by the devil to
appear before Alvare, why does she not know (or why did he not tell her)
that Alvare has company? If we think further about this, however, we can
explain this instance either through Alvare’s delusion and vanity (because
he thinks highly enough of himself to expect that a fêted opera singer will
appear just for his own pleasure) or through the devil’s well-​judged play
on Alvare’s vanity (making it appear as if she just came for Alvare). The
situation seems strange, but it does not call for immediate disambiguation.
Instead, it can serve readers as a reference point later on. For example,
when Alvare describes his enchantment at Fiorentina’s song, he says:

‘J’étais ému jusqu’au fond du cœur et j’oubliais presque que j’étais le créateur du
charme qui me ravissait.’
(‘I was moved to the bottom of my heart and I almost forgot that I was the
creator of the charms that delighted me.’) (p. 323)

Fa n ta s t i c C o g n i t i o n   [ 161 ]
 

In the earlier instance, too, he might just as well have forgotten that his
wish had been the creator of Fiorentina’s appearance. As the narratorial
discourse reminds readers of the supernatural powers of the character
(ironically by stating that he had forgotten about them), they can revisit
the earlier instance and inscribe it (more or less consciously) into the gen-
eral predictive model which they are currently developing.
In Le Diable Amoureux, Cazotte sets in motion an intricate machinery
that draws on ‘unconscious inference’ (of the kind that pervades our every-
day cognition), the conscious inferences of Alvare as narrator, and the
(more or less conscious) revisiting of previous inferences on the part of the
reader in light of new information. At times, the narrative contextualiza-
tion does not serve to create the cognitive ‘penetrability’ through which
expectations shape percepts, but instead, creates the narrative, verbal
equivalent of a duck-​rabbit. In the dinner scene, both Alvare’s act of van-
ity and the devil’s manipulation of him remain possibilities when consid-
ered generally, but as soon as we start thinking through the implications of
each of these options, we lose our grasp of the other option. Is Fiorentina
in ‘deshabillé’ due to the volition of Alvare (and his need to show off in
front of his cabalist friends)? Is it due to the devil’s psychological skill of
granting Alvare his wish in such a way as pleases his vanity best and gives
him the impression of being in control? These instances of ambiguity build
up throughout the narrative. They constantly remind readers of the basic
conflict between the realist hyper-​ prior (Alvare’s delusion) and the
supernatural hyper-​prior (the devil’s trickery), and they make it difficult to
disambiguate between the competing hyper-​priors, because the chains of
inferences that these enable can be pursued to such a degree that we lose
the other option from view.

What, then, in such a wilfully ambiguous set-​up, would be feasible disam-

biguation devices? Do they constitute something like a ‘found science’ in
verbal narrative? And what are their larger implications for the study of
cognition?
Cazotte’s novella starts with a general statement from Alvare (as the
first-​person narrator) which orientates readers in the time and space
of the narrative, much in the tradition of the popular genre of the
‘mémoire’: ‘J’étais à vingt-​cinq ans capitaine aux gardes du roi de Naples’
(‘At the age of twenty-​five I  was a captain in the guard of the king of
Naples’, p.  315). The communicative situation that the narrator evokes
takes readers back to an earlier stage of his life. It implies (1) that Alvare
as the narrator can shape his narration of the events to whatever degree
he likes, because he is our only source of its authenticity, and (2) that the

[ 162 ]  Cognition through a Literary Lens

 

narration takes place after the events, so that the narrating I of Alvare
already knows how the story of the experiencing I of Alvare is going to
end. Cazotte, however, decides to mix the features of the narrating I of
Alvare with those of the experiencing I of Alvare. Throughout the novella,
we get passages narrated in the present tense. For example, after Alvare
has heard from the peasant Marcos that Biondetta has left, the narration
turns to the present tense: ‘Marcos sort. Machinalement je me frotte les
yeux’ (‘Marcos leaves. Mechanically, I rub my eyes’, p. 372). He continues
to wonder whether the seduction of Biondetta has actually taken place,
and is interrupted in these thoughts when his carriage is announced. ‘Je
descends du lit; à peine puis-​je me soutenir, mes jarrets plient sous moi’
(‘I get out of bed; I can barely support myself, my knees buckle’, p. 372). In
these instances (as also in the description of Fiorentina’s entrance quoted
above), the narration becomes immediate, and it seems as if the narrating
I of Alvare loses the distancing mode of his narration, just as the experi-
encing I of Alvare loses control over his limbs. Are these instances due to
the devil’s machinations that threaten to control the narrative (and can
they hence be related to the supernatural hyper-​prior)? Or does Alvare
(temporarily and without acknowledging it explicitly) get shaken out of
the flow of his delusions (and can this instance hence be related to the
realist hyper-​prior)?
The narrative of the novella itself ends with the words of Quebracuernos,
a doctor of Salamanca. Considering, he says, that none of the strategies
of the devil that we find in Le Diable Amoureux can be traced back to ear-
lier demonological literature (such as Bodin’s Déonomanie and Bekker’s Le
Monde Enchanté), it seems likely that the devil has devised new strategies
and is more dangerous than ever. The devil could come once more for Alvare,
and the only thing that will keep him from falling into seduction would be
a wife who has celestial qualities, such that ‘vous ne serez jamais tenté de
la prendre pour le Diable’ (‘you would never be tempted to take her for
the Devil’, p. 376). The doctor’s own assessment of the situation keeps the
ambiguity alive. Is it likely that Alvare will find a wife whom he would not
mistake for the devil? This is presented, explicitly, as a strategy for warding
off the devil’s influence, but we wonder whether it is not Alvare’s imagina-
tion (rather than the devil’s machinations) that led him to take Biondetta
for the devil. Indeed, the entire statement of Quebracuernos is reported
by Alvare, so that readers cannot even be sure how reliable the words of
the man of science are. In terms of a ‘found science’ of the literary text,
these instances of Quebracuernos’s argument from authority and Alvare’s
potentially unreliable report of it keep the ambiguity between hyper-​priors
alive on the level of interpersonal cognition, where one might come to ask

Fa n ta s t i c C o g n i t i o n   [ 163 ]
 

whether models that others teach or endorse are reliable or prone to preju-
dice (Frith, 2009, pp. 167‒183).
Even if the fictional text remains ambiguous to the very end, perhaps
Cazotte himself can be prevailed upon to tell his readers how to read his
novella? In the epilogue, he tells us of the different incarnations which his
novella went through. In the version of 1772, Alvare sees through the tricks
of Biondetta and escapes the snares of the devil. Cazotte then reports a
second version (presented only to ‘personnes de sa connaissance’, ‘acquain-
tances’, p. 377), in which Alvare falls prey to the devil and suffers the well-​
known consequences of eternal damnation. Finally, in the version of 1776,
Cazotte seeks to combine both options, creating a narrative in which
‘Alvare y est dupe jusqu’à un certain point, mais sans être victime’ (‘Alvare
is the dupe to a certain point, but he does not become the victim’, p. 377).6
Only this final version of the novella provides readers consistently with the
kind of ambiguity that leads to the sustained hesitation of Todorov’s fan-
tastic. Strictly speaking, all three versions of Cazotte’s novella would tend
towards the marvellous, because the devil plays a role in the fictional world,
but they do so to different degrees, because the powers of the devil change
across versions. As Todorov acknowledges, ‘the fantastic in its pure state’
is best ‘represented by a median line’ between fantastic texts that tend
towards the uncanny and those that tend towards the marvellous (1970/​
1975, p. 44). Nevertheless, one can read the versions of Cazotte’s novella as
a process of experimentation, comparable to the artists’ sketchbooks which
Gombrich analyses in Art and Illusion. Smaller changes between the 1772
and the 1776 versions, and the long alternative ending of the 1776 ver-
sion, presumably lead to vastly different effects on readers. Here, Cazotte’s
novella might offer a ready-​made experimental design for the empirical
study of reading and hence of predictive processing more generally.
A final duck-​rabbit that enhances the ambiguity of Le Diable Amoureux
is the very title of the novella. Who is the ‘devil in love’? At first glance, it
seems most likely that this amorous demon is Biondetta. After the seem-
ingly successful seduction, she reveals herself to be the devil (which—​at this
point—​corresponds to readers’ expectations), and yet at the same time,
rather than triumph over the hapless soul she has snared, she confesses her
love for Alvare once more (‘ce cœur qui t’adore’, ‘this heart that adores you’,
p.  370). Even if characters make definite statements, confirming either
the natural or the supernatural, Cazotte immediately supplies clues that
make a conclusive inference problematic. Is the devil actually in love with

6.  Cazotte goes on to add (p. 377) that the devil might have duped Alvare, but if he
had, Alvare would still have retained his virtue and hence triumphed over the devil.

[ 164 ]  Cognition through a Literary Lens

 

Alvare? That assumption would work against any standard expectations as

to the devil’s character and his actions, and in turn might suggest that the
entire narrative is created by Alvare’s delusion. Franc Schuerewegen (1985)
suggests an alternative interpretation of the title: none other than Alvare
himself could be called the ‘devil in love’. Schuerewegen traces through-
out the narrative instances of Alvare copying Biondetta’s behaviour, as well
as making promises and declarations that serve to set her up (p. 65): he
begins to imitate the strategies of the devil (p. 69). It seems to me an open
question whether readers wonder about the title, which they encounter on
every even page in the running head of the text (in a traditional edition) as
they read the novella. Similarly, it seems to me an open question whether
readers pick up on the similarities between Alvare’s promises and ruses
and those of Biondetta. What, in terms of ‘found science’, is then the role
of what we might call ‘super hyper-​priors’, such as titles or general tags of
situation (like ‘doctor’s visit’ or ‘dissertation viva’)? Surely, they provide
predictive set-​ups for our cognition in certain moments, but in how far are
we (or do we need to be) aware of these tags for these predictions to take
effect? Or in how far do we simply forget about them in order to facilitate
cognition in the situation itself?

One of the earliest and most powerful claims of cognitive approaches to lit-
erature is that the human mind works through devices that are commonly
considered ‘literary’, such as metaphor (Lakoff and Johnson, 1980/​2003;
Turner, 1996) and narrative (Bruner, 1986, 1990). Fiction and the imagina-
tion, as scenarios we entertain and as a ‘default mode’ of thinking, have
more recently entered the fray (Richardson, 2011). Arguably, the study of
literary texts is as important as the cognitive sciences in the endeavour of
working out the elements of the ‘literary mind’ because it helps make these
more or less automatic features of cognition noticeable and thus subject to
analysis.
In the predictive processing model, we have virtual predictive models, or
‘fantasies’, which guide our perception and our cognition more generally.
The literary genre of the fantastic highlights hesitation in the use of these
‘fantasies’ in cognition, and hence it can serve as a repository of ‘found sci-
ence’, of the ways in which predictive processing operates between differ-
ent modes of cognitive ‘penetrability’. Cognitive literary study, informed by
Bayesian models of cognition and the literature of the fantastic, can con-
tribute to studies of how we refine our predictive, probabilistic models and
of whether awareness contributes to or detracts from these recalibrations.
In the cognitive sciences, the literary strategies of the fantastic can then be
employed for the design of experiments to study the cognitive penetrability

Fa n ta s t i c C o g n i t i o n   [ 165 ]
 

entailed by more general, propositional knowledge of the situation, as well

as the functions of awareness of such knowledge and its interpersonal reli-
ability. In cognitive literary studies, we can pursue ‘found science’ through
the ways in which authors experiment with different effects in manuscript
drafts or editions, how they make conscious the unconscious inferences
that predictive processing depends on, and how literary history more gener-
ally provides us with a body of evidence for our ‘fantastic cognition’.

ACKNOWLEDGEMENT

This work was supported by a postdoctoral research grant from the

Academy of Finland (no. 267599).

REFERENCES

Anderson, M. (2015). The Renaissance extended mind. Edinburgh: Palgrave Macmillan.

Anderson, M. (2017). The Extended and Predictive Mind in Shakespeare. In C. Bourne
and E. Bourne (Eds.), The Routledge companion to Shakespeare and philosophy.
London: Routledge.
Armstrong, P. B. (2013). How literature plays with the brain: The neuroscience of reading
and art. Baltimore, MD: Johns Hopkins University Press.
Baxandall, M. (1958). Patterns of intention: On the historical explanation of pictures. New
Haven, CT: Yale University Press.
Bruner, J. (1986). Actual minds, possible worlds. Cambridge, MA:  Harvard University
Press.
Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard University Press.
Cavanagh, P. (2005, 17 March). The artist as neuroscientist. Nature, 434, 301–​307.
Cazotte, J. (1965). Le diable amoureux. In M. du Cheyron (Ed.), Romanciers du XVIIIe
siècle II (pp. 303–​378). Paris: Gallimard.
Clark, A. (2013). Whatever next? Predictive brains, situated agents and the future of
cognitive science. Behavioral and Brain Sciences 36(3), 181–​204.
Crane, M. T. (2001). Shakespeare’s brain: Reading with cognitive theory. Princeton, NJ:
Princeton University Press.
Friston, K. (2013). The fantastic organ. Book review. Brain: A Journal of Neurology, 136,
1328–​1332.
Frith, C. (2009). Making up the mind: How the brain creates our mental world. Hoboken,
NJ: Wiley Blackwell.
Gadamer, H.-​ G. (1960). Wahrheit und Methode:  Grundzüge einer philosophischen
Hermeneutik [Truth and method: Foundations of a philosophical hermeneutics].
Tübingen: Mohr.
Gombrich, E. H. (2002). Art and illusion: A study in the psychology of pictorial representa-
tion (6th ed.). London: Phaidon.
Hohwy, J. (2013). The predictive mind. Oxford: Oxford University Press.
Ingarden, R. (1968). Vom Erkennen des literarischen Kunstwerks [The cognition of the
literary work of art]. Tübingen: Niemeyer.

[ 166 ]  Cognition through a Literary Lens

 

Iser, W. (1976). Der Akt des Lesens:  Theorie ästhetischer Wirkung [The act of read-
ing: A theory of aesthetic response]. Munich: Fink.
Jensen, M. S., and Mathewson, K. E. (2011). Simultaneous perception of both interpre-
tations of ambiguous figures. Perception, 40, 1009–​1011.
Kandel, E. R. (2012). The age of insight: The quest to understand the unconscious in art,
mind and brain, from Vienna 1900 to the present. New York: Random House.
Kesner, L. (2014). The predictive mind and the experience of visual art work. Frontiers
in Psychology, 5(1417), 1–​12.
Kukkonen, K. (2014). Bayesian narrative:  Probability, plot and the shape of the fic-
tional world. Anglia 132(4), 720–​739.
Lakoff, G., and Johnson, M. (2003). Metaphors we live by (2nd ed.). Chicago,
IL: University of Chicago Press.
O’Regan, J. K. and Noë, A. (2001). A sensorimotor account of vision and visual con-
sciousness. Behavioral and Brain Sciences, 24(5), 939–​1031.
Richardson, A. (2010). The neural sublime:  Cognitive theories and Romantic texts.
Baltimore, MD: Johns Hopkins University Press.
Richardson, A. (2011). Defaulting to fiction:  Neuroscience rediscovers the Romantic
imagination. Poetics Today, 32(4), 663–​694.
Ricoeur, P. (1984‒1988). Time and narrative (3 vols., Kathleen McLoughlin and David
Pellauer, Trans.). Chicago, IL: University of Chicago Press.
Schuerewegen, F. (1985). Pragmatique et fantastique dans le Diable amoureux de
Cazotte. Littérature, 60(4), 56–​72.
Starr, G. G. (2013). Feeling beauty: The neuroscience of aesthetic experience. Cambridge,
MA: MIT Press.
Todorov, T. (1975). The fantastic: A structural approach to a literary genre (R. Howard,
Trans.). Ithaca, NY: Cornell Paperbacks.
Turner, M. (1996). The literary mind. Oxford: Oxford University Press.

Fa n ta s t i c C o g n i t i o n   [ 167 ]
 
 

CHAPTER 9

Feedback in Reading
and Disordered Eating
EMILY T. TROSCIANKO

Reading supplements my feelings rather than creates my feelings.

anonymous survey respondent

INTRODUCTION

Nowhere is the impossibility of separating mind from body clearer than

in an eating disorder (ED): both sicken in a reciprocal back-​and-​forth, and
both recover that way too. Typically classified as ‘mental illnesses’, EDs are
an excellent example of how the psychological and the physical have to be
understood as interacting parts of the same system. This might seem an
unremarkable statement, especially if you have already read plenty about
‘4E’ (embodied, embedded, enactive, and extended) cognition or second-​
generation cognitive science or anti-​computationalism in theories of the
mind (e.g. Wallace, Ross, Davies, and Anderson, 2007; Menary, 2010;
Kukkonen and Caracciolo, 2014). But here I want to draw attention less to
the inseparability than to the constituent interactions.
In much cognitive literary work that draws on and contributes to
research on 4E cognition, including most of my own (e.g. Troscianko,
2014), the emphasis has been on thinking through the ways in which our
situated bodies in action are fundamental to how our minds work. What
has therefore been teased out in such research is the many ways in which
embodiment functions as an underlying substrate, a long-​ignored but now
 

incontrovertible bedrock determining what we used to think of, in our more

Cartesian moments, as pure reason or the intellectual act of interpreta-
tion. So, for example, we have gained insights into how embodiment struc-
tures poetic texts and readers’ responses to them (Miall, 2011), or informs
readers’ mental imagery (Kuzmičová, 2014) or intersubjective experience
(Chesters, 2014). These sorts of approaches go a long way towards mapping
out the extent of the mind‒body dialogue in fiction and fiction-​reading,
but typically they say less about its dynamics.
So in this chapter I will try to attend to the dynamics rather than the
state of interconnection between the psychology and the body—​though
the latter will necessarily be a given. In this endeavour, I follow in the foot-
steps of some noteworthy work that has already moved from the founda-
tional to the interactional stance on embodiment in reading, in contexts
like the feedback between more emotional and more rational appraisal in
readers’ responses (Pirlet and Wirag, this volume), the feedback between
brain, body, and world in literary writing (Bernini, 2014), and the feedback
between constraints and affordances (Caracciolo, 2014) or between predic-
tive models and real-​world data (Kukkonen, this volume) in readers’ inter-
pretive engagements with texts. My approach will differ from these others
by focusing on a particular embodied cognitive reading situation—​that of
having an ED—​rather than dealing primarily with the kinds of dynamics
that characterize cognition more generally. My focus on disordered eat-
ing is informed by my own past experience of anorexia; you can read more
about that experience, in dialogue with scientific and clinical work, on
my blog (Troscianko, 2009). Anorexia will feature with particular promi-
nence in this discussion, partly because of my own experience, but mainly
because as an ED that by definition involves severe malnutrition, it makes
certain mind‒body interactions especially salient. However, many of the
same principles apply across the ED spectrum. My aim in this chapter is
to understand both disordered eating and reading better, by tracing some
of the recursive relationships that may mediate between the two. And the
single most fundamental concept I will be relying on is feedback.

FEEDBACK SYSTEMS AND DISORDERED EATING

Feedback occurs when the output of a system is routed back to become an

input to the system. There are two basic kinds of feedback: positive and neg-
ative. Although these terms are sometimes used qualitatively to indicate
appraisal or valence (‘she tactfully gave him negative feedback on his essay’,
‘the feedback was positive: over time it made them gradually happier’), I will

[ 170 ]  Cognition through a Literary Lens

 

be using them here in their technical, quantitative sense: positive feedback

is usually self-​reinforcing and unstable, leading to an ever-​greater discrep-
ancy between the starting and the end states; negative feedback is typi-
cally self-​cancelling and stable, tending towards equilibrium (Mitrophanov
and Groisman, 2008; Åström and Murray, 2014, p. 22). A familiar example
of positive feedback is the deafening screech of a microphone picking up
the sound of the speaker used to amplify it:  the microphone feeds the
speaker sound back to the speaker for amplification, and the speaker ampli-
fies that sound and feeds it into the microphone, and so on until your ears
hurt. The crescendo stops only when someone unplugs one or both bits of
kit. A simple example of negative feedback is a central heating thermostat,
which measures the actual temperature, compares it with the reference
temperature it is set to, and sends a signal to the heating system to increase
or reduce its output, resulting in a new actual temperature to measure. If
the thermostat is working well, the fluctuations should get smaller and
smaller until a balmy 21°C is reached—​and then adjust quickly if someone
opens an outside door.
These examples may seem a long way from human cognition, but the
structures and mechanisms of feedback are the same in mechanical, elec-
trical, and biological systems. Feedback is present in all biological systems.
In the human body, homeostatic stability is the ideal state for everything
from body temperature to blood pressure and metabolic rate, and it is
achieved via negative feedback. Failure to maintain homeostasis is seen in
heatstroke or hypothermia, in hypertension or hyperthyroidism. Although
positive feedback is inherently unstable, it is not always bad:  during
childbirth, for example, pressure of the foetus on the cervix causes nerve
impulses to be transmitted to the brain, causing release of the hormone
oxytocin into the bloodstream, causing the smooth muscle of the uterus to
increase the rate and force of the contractions, pushing the foetus harder
against the cervix, producing more oxytocin causing more contractions.
Although in experiential terms the dynamics of the positive feedback here
are associated with pain and any number of other negative responses, ulti-
mately (when all goes well) the feedback loop is broken by the birth of the
baby, which is thus both the desired outcome and a self-​limiting event. In
most biological cases, though, positive feedback that is not embedded in a
larger negative feedback loop is a sign of something having gone awry.
This observation applies equally when we take the biological to include
the cognitive. No one understands precisely how the experience of being me
now relates to the genetic and cellular composition and mechanisms of my
body. This is the hard problem of consciousness, and even those who claim
it does not exist (e.g. Churchland, 1996)—​that the receptor and neuron

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 171 ]
 

activity just is the mysteriously predictable thrill you get from that song
you currently have on repeat—​still have to explain to everyone else why
an explanation that does not feel at all adequate actually is. In the absence
of a solution to the hard problem, my working assumption will be that it
makes sense to talk about experiential things—​like thoughts, sensations,
and emotions—​as different from but in direct interaction with physical
things like muscular contractions, hormone secretion, or nerve signalling.
Of course, there cannot be any completely neat separation: all the terms on
the experiential side also involve physical activity on the part of the neu-
rons, the muscles, the receptors, and so on. And some of this activity, like
the contractions of some muscles, can be directly experienced; some, like
the pH regulation of the blood, cannot. The relationship between the unex-
perienceable physical elements, the experienceable physical elements, and
the experienceable apparently nonphysical elements remains bafflingly
opaque. This opacity means that the simple feedback loops I will be discuss-
ing between ‘mind’ and ‘body’ pertain only at the highest level, where they
remain partially separable. Go down a little further, and the loops instantly
multiply and entwine with each other so thoroughly that the mind‒body
distinction becomes rapidly meaningless; go down far enough, and no one
knows what the loops might really look like, because no one has solved
the mystery of consciousness yet. A quarter of a century ago, Dan Dennett
wrote that ‘human consciousness is just about the last surviving mystery’
(1991, p. 21)—​defining a mystery as something we don’t even know how to
think about yet. The same can still be said today.
So, let’s return to the more easily comprehensible top level. In a healthy
person, physiological factors like bodyweight and energy metabolism,
behavioural factors like diet, exercise, and other routines, and psycho-
logical factors like attitudes to food and the body exist in an equilibrium
enabled by multiple forms of negative feedback. Just as small changes in
energy intake are adjusted for metabolically (Molé, 1990)  to maintain a
stable bodyweight, so fluctuations in body-​directed self-​confidence may be
absorbed by small changes to dietary or social habits. If we then imagine
that a small but significant amount of weight is gained, thanks say to ill-
ness or Christmas, it is easy to see how a minor reduction in contentment
with body size and shape might result in a plan to make a small increase
in exercise or change in diet, or indeed how these might happen without
an active decision even being required, and how habits will then revert to
normal once the previous equilibrium is returned to (see Figure 9.1).
In someone with an ED, or vulnerable to developing one, things hap-
pen very differently. Figure 9.2 shows one possible way of modelling the
primary high-​level feedback loop in anorexia nervosa. As you can see, the

[ 172 ]  Cognition through a Literary Lens

 

Experience of bodyweight or shape

Influence of body shape and weight on self-evaluation

Change in energy intake and/or exercise

Bodyweight

Feelings of fullness
Thoughts about food
Social habits
Mood
Self-esteem

Figure 9.1  Healthy mind‒body feedback.

Disturbance in how body weight or shape is experienced

Undue influence of body shape and weight on self-evaluation
Persistent lack of recognition of the seriousness of low bodyweight

Persistent restriction of energy intake

Compensatory weight control behaviour

Significantly low bodyweight

Heightened feelings of fullness

Preoccupation with food
Obsessive thought patterns
Social withdrawal
Depression
Lowered self-esteem

Figure  9.2 Positive mind‒body feedback in anorexia nervosa. This model is a specific

instantiation of the general model presented in Figure 9.1, in which the feedback could oper-
ate in either direction. (See also Fairburn, 2008, p. 21.)
 

feedback here is positive rather than negative:  the system moves away
from equilibrium rather than maintaining it. Wherever you enter the loop
(whether with unintended weight loss, or preoccupation with body shape
and weight for other reasons, or a temporarily low mood), each factor exac-
erbates the next, and a spiral deeper into illness is initiated. The reasons
why negative feedback may fail to maintain equilibrium in one particular
input‒output relationship in one person and not another may be genetic,
biological, and/​or socioculturally informed: one person might lose a large
amount of weight due to a viral infection but then recover mentally and
physically as soon as the infection is fought off, whereas for another this
episode might be the start of a prolonged struggle with eating and their
body. But the crucial point to retain is that as soon as negative feedback
fails to correct a movement away from stability—​that is, whenever the sys-
tem is insufficiently robust to perturbations—​the cycle of positive feed-
back kicks in.
This kind of feedback model is used to understand ED psychopathol-
ogy in cognitive behavioural therapy, which emphasizes the interactions
between thought, emotion, behaviour, and physical state (e.g. Fairburn,
2008). Similar models have also been developed in the context of catas-
trophe theory (e.g. Zeeman, 1976), where the progressive abnormality
of attitudes towards food combined with ever-​increasing hunger con-
stitute positive feedback leading towards instability between the two
extremes of bingeing and fasting (which depending on bodyweight
might be classified as bulimia nervosa, or as anorexia binge-​purge sub-
type). The lack of stability assumed by these models is also supported
anecdotally by the tendency of ED sufferers to characterize their condi-
tions using words and phrases like ‘reinforcing’, ‘vicious circle’, or ‘spi-
ralling out of control’. (And a quick Google search for combinations like
‘bulimia spiral’ or ‘anorexia vicious circle’ can help start to turn these
anecdotes into data.)

INCLUDING SOCIETY AND CULTURE IN EATING

DISORDER FEEDBACK SYSTEMS

Like any model at this level of generality, the model shown in Figure 9.1 is
incomplete when it comes to the lower-​level mechanisms of change, but
it is significantly incomplete at this high level too. Given that physiology,
behaviour, thought, and emotion are represented at least cursorily in the
model, the main factors that are obviously missing are the contributions of
social and cultural factors.

[ 174 ]  Cognition through a Literary Lens

 

The importance of social, and particularly familial, factors in the devel-

opment and maintenance of anorexia has been analysed from a systems
theory perspective by Salvador Minuchin and colleagues (Minuchin,
Rosman, and Baker, 1978). Their argument for understanding anorexia not
as the product of the system but as one of its parts has since been further
validated by the increasing popularity of and evidence supporting various
kinds of family therapy for EDs (Fisher, Hetrick, and Rushford, 2010). The
aim of Minuchin and colleagues’ ‘structural family therapy’ (also used for
many conditions other than anorexia) is to ameliorate the feedback dynam-
ics of the family system, often through positive feedback that initiates a
change, followed by negative feedback that sustains the new state. Their
characterization of the ‘psychosomatic family’ in terms of ‘enmeshment’
and ‘rigidity’ (e.g. 1978, p. 30), amongst other qualities, reinforces what we
have already observed about feedback systems in cognitive contexts: the
basic structural features are closely bound to their experiential counter-
parts, and both operate at once as cause and as effect.
A wide range of cultural factors can also be understood as contributing,
through feedback, to the psychopathology of anorexia. Steps in this direc-
tion have been made in research on the role of the media in body image
problems (for a general review, see Grabe, Ward, and Shibley Hyde, 2008).
Steven Kirsh (2010) describes a chicken-​and-​egg situation in which

Youth with disordered eating seek out thin-​ideal media while at the same time
being influenced by the thin-​ideal media that they consume. In turn, a feedback
loop develops (a downward spiral), in which thin-​ideal media reinforces and
exacerbates eating disordered symptomatology, and disordered eating increases
interest in thin-​ideal media. (p. 146)

A recent study (Cohen and Blaszczynski, 2015) investigated whether female

undergraduates (with no clinical diagnosis related to EDs) respond differ-
ently to images promoting a thin ideal presented via Facebook as compared
with other online media, based on the hypothesis, from social comparison
theory, that people are more likely to compare themselves to similar oth-
ers. As expected, they found that body dissatisfaction increased as a result
of appearance comparison on Facebook but not conventional media. They
also found that Facebook use in life beyond the experimental intervention
was higher among those at high risk of EDs than those at low risk (see
also Latzer, Spivak-​Lavi, and Katz, 2015). This correlation of course tells
us nothing about causation (and it also neglects the potential for bene-
fits to self-​esteem through Facebook use suggested by other studies, e.g.
Gonzalez and Hancock, 2011), but the two findings taken together provide

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 175 ]
 

further support for the ‘feedback loop whereby exposure to body-​related

stimuli activates and reinforces an over-​concern with one’s own body,
which in turn reactivates attentional biases toward body-​related stimuli’
(Cohen and Blaszczynski, 2015, p. 9).
When feedback is present, causal relationships can be difficult to disen-
tangle. Cohen and Blaszczynski (2015, p. 9) suggest two possible ways of
interpreting the observed correlation between Facebook use and ED risk.
On the one hand, frequent exposure to thin-​ideal content on Facebook may
reinforce body-​related concerns, eliciting cognitive biases that prioritize
attention to thin-​ideal content on Facebook. On the other hand, people
with a higher risk of EDs may be more likely to use Facebook, and given
the association of EDs with selective attention for appearance-​related
cues, Facebook use may further reinforce ED risk via this particular vulner-
ability. In an ideal world, one would be able to establish which came first,
but given real-​world complexities, this may never be possible. But taking
a feedback perspective means that deciding between hypotheses becomes
less important. For any individual within a given sample, the starting point
for increased body dissatisfaction may be either Facebook or a pre-​existing
vulnerability, or the two may be temporally and causally indistinguishable.
The point is that a positive feedback loop is initiated, and once it is in place,
its result is the predictable instability of a cyclical movement away from
the starting state. The system dynamics rather than the initial trigger are
of primary importance in understanding what is going on.
The field is thus opened up for taking a similar approach to studying the
effects of other forms of cultural activity on those with disordered eating—​
and here, of course, I want to talk about fiction-​reading.

FICTION-​R EADING AND THE EATING

DISORDER FEEDBACK SYSTEM

As an easy point of departure, we could make the same basic claims about
fiction-​reading as Cohen and Blaszczynski do about Facebook use: exposure
to materials that deal with body-​related phenomena activates or reinforces
an over-​concern with one’s own body, which in turn reactivates attentional
biases towards body-​related stimuli, which makes people seek out or notice
such materials preferentially to others, which worsens the over-​emphasis
on the body. There are bound to be important differences here between the
picture-​dominated social media and the wholly linguistic nature of most
adult fiction; in particular we might expect imaginative responses, includ-
ing mental imagery, to be elicited in very different ways (Troscianko, 2013,

[ 176 ]  Cognition through a Literary Lens

 

esp. pp. 187‒188). Nonetheless, the same structural principles are likely to

apply: in someone who already has an active ED, the pre-​existing tendency
to pay more attention to appearance-​related stimuli may make this positive
feedback all the more likely to be initiated. This is one obvious way in which
reading texts of certain types, with a particular interpretive bias, may end
up contributing to the positive feedback already characteristic of EDs. In
Figure 9.3 you can see how the same basic structure as shown in Figures 9.1
and 9.2 can be expanded by adding in several aspects of a possible reading
process.
But reading is a complex business, and the fiction people read (let alone
all the non-​fiction) is complex and varied too, not least in the cognitive
demands it makes on readers. So we should expect that fiction-​reading
might have other contributions to make to the feedback loops of disordered
eating. I recently collaborated with the ED charity Beat to gather data on
perceptions of how reading habits and preferences may be connected with
mental health, and specifically EDs. Eight hundred and eighty-​five respon-
dents took part in our online survey (773 of whom had a personal history
of disordered eating), providing us with a great wealth of both quantitative

Engaging emotionally with

fictional characters

Emotional state

Engaging with fictional

characters’ thought processes

Thought patterns

Reading about physical ideals

Bodily sensations

Reading about dietary and/or

exercise strategies

Diet and exercise

Bodyweight

Figure 9.3  Mind-body feedback incorporating cultural factors.

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 177 ]
 

(forced-​choice) and qualitative (free-​response) data. Other findings from

the data will be set out elsewhere (Troscianko, in preparation), but here
I want to concentrate on the free-​response data and the evidence they pro-
vide for the role of feedback in the relations between disordered eating and
fiction-​reading.
The data I will be presenting come from responses to a number of sur-
vey questions in which respondents were invited to elaborate on previous
forced-​choice answers. The questions divide into three types.
One sequence of questions concerned the potential helpful or harm-
ful effects of reading in general. Having answered a series of forced-​choice
questions on this topic (for example, indicating which of a list of possi-
bilities fitted their experiences of finding books helpful or harmful, like
‘Letting you see your eating disorder through someone else’s eyes’ or
‘Causing you to reflect obsessively on your eating and/​or exercise habits’),
respondents were asked to ‘Please list any authors and/​or titles of books
that have affected you in the ways described in the previous question, and
please briefly specify which book had which effect.’
Another sequence concerned the effects of reading on general mood, self-​
esteem, feelings about your body, and diet and exercise habits. The questions
were asked once with respect to fiction about EDs and then repeated with
respect to ‘your preferred type of other fiction’ (the type which respondents
usually read, chosen from a list including genres like fantasy fiction, roman-
tic fiction, suspense/​thriller, etc.). After each sub-​section respondents were
given the chance to elaborate on the forced-​choice answers with a general
prompt: ‘If you wish, please provide more details about [the change to your
eating and/​or exercise habits] after reading [fiction about eating disorders]
here (including authors and/​or titles if applicable)’.
Lastly, at the end of the survey we asked respondents, ‘Finally, is there
anything else you would like to share about your reading habits or how they
relate to your mood, eating, exercise habits, or similar? If so, please feel
free to use the space below.’
Responses to these three question types—​which constitute all the open-​
ended questions we asked—​are included in the analysis that follows, which
takes the form of a search of all responses for the presence of descriptions
of feedback of any kind. Respondents were primed to employ particular
words and phrases (‘helpful’, ‘harmful’, ‘how you feel about your body’,
etc.), and were encouraged to think about possible causal relationships
between mental health and reading (‘affect’, ‘effect’, ‘change’, ‘improve’,
‘worsen’, etc.). However, at no point were they prompted specifically to
think about the more complex kind of causality manifest in feedback loops
either positive or negative. All indications that such feedback might be in

[ 178 ]  Cognition through a Literary Lens

 

play are to this extent the respondents’ spontaneous reflections on their

personal experience.
My own analysis, through close reading, identified 19 types of positive
feedback loop in the survey responses. These loops were identified in 97
responses (of which 9 were included in more than one category, and 1 was
classified in three categories) from 71 respondents, including 4 with no
personal history of disordered eating; all the rest reported current or past
ED experience. I have grouped them into broad categories to help give an
overview, but the distinctions are by no means absolute. After each type
I  specify the number of times it recurred in our respondents’ testimony
(and note when this included anyone without a personal ED history), and
give a short example. All example responses are reproduced in full in their
original form, with typos and other idiosyncrasies unaltered.

GENERAL

1. Feel worse → reading → feel worse and/​or feel better → reading

→ feel better [26, of whom 11 mention both directions, 15 only the
worsening direction] My mood can improve or worsen depending on my
starting frame of mind. If i am hopeful they can motivate and increase will-
power and self esteem. If i am in a bad place, I come away feeking worse and
more likely to engage in eating disorders behaviour.

MOOD, SELF-​E STEEM, AND BODY IMAGE

2. Reading → worsened ED habits and/​or worsened mood → preoccu-

pation with ED → worsened mood [1]‌After reading Monkey Taming by
Judith Fathallah it left me feeling low. I had been doing well in recovery but it
sort of triggered a bit of a relapse and left me feeling very down and conscious
of my eating disorder, and I found myself comparing my own experience with
the character’s. Thinking too much about my eating disorder puts me in a
more melancholy state of mind.
3. Reading → worsened ED habits (e.g. desire for weight loss) → wors-
ened mood → worsened ED habits (e.g. behaviours promoting weight
loss) in attempt to ameliorate /​distract from low mood [1]‌I read the
book a long time ago however found it made me more determined to lose
weight. My mood tended to become more low and this cemented the desire to
lose weight in order to feel better about myself.
4. Reading → preoccupation with ED → negative self-​assessment for
being thus preoccupied → exacerbation of ED → preoccupation with
ED [1]‌My mindset shifts toward obsessive ED thoughts, and I feel guilty,
ashamed, and lonely when I have these thoughts.

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 179 ]
 

 5. Reading → exacerbation of ED → negative self-​assessment of self

for letting this happen → exacerbation of ED [1]‌Makes me obsessive
and competitive and pushes me further into my eating disorder making me
feel inferior.
 6. Reading → lower self-​esteem and/​or increased shame/​embarrass-
ment → exacerbation of ED → more reading about EDs [1]‌Can’t
really answer this. Most reading on eating disorders, factual or othrwise,
lowers my self-​esteem. I’m so embarrassed to be reading about it that I don’t
let anyone see me with a book about eating disorders—​i hide them.
 7. Reading → worsened mood → worsened self-​esteem/​wellbeing
→ exacerbation of ED → worsened mood [1]‌I tend not to read any-
thing that I know will negatively affect my mood, as this could then have a
knock-​on effect on my eating habits or general self-​esteem and wellbeing. So
I usually pick thoughtful but uplifting fiction. I also read a lot of non-​fiction
on things such as animals, travel, anything that I  can focus on. Since my
negative experience of reading fiction about an eating disorder (in Monkey
Taming by Judith Fathallah) I have avoided anything like this again for fear
of it triggering anything or just making me feel low.
 8. Feel worse → read more → disengage from rest of life → feel worse
[1]‌When I am having difficulty with my eating disorder I am more likely to
seek out stories about eating disorders, particularly short stories (published
or posted online) that glamourise eating disorders. I also spend more time
reading any fiction, often to the detriment of other aspects of my life, and
end up spending less time completing my academic readings or fulfilling my
role as an editor.

THOUGHT AND KNOWLEDGE

 9. Reading → preoccupation with ED → more self-​analysis and self-​

directed thought → exacerbation of ED → heightened preoccu-
pation with ED [2—​incl. 1 non-​ED, implying other mental health
problems] As it is on my mind I am more aware of habits and more likely to
fall into old patterns which can set me back.
10. Reading → positive assessment of ED behaviours as coping mecha-
nisms → exacerbation of ED habits → increased tendency to assess
ED habits as coping mechanisms [1]‌For a few days after being exposed
to ED ideas or stories I  will restrict food or increase exercise. Just being
reminded of them as a coping mechanism makes me want to engage in
them again.
11. Recovery from ED → reading → increased knowledge about how
to recover alone → progress in recovery [1]‌I don’t feel that reading

[ 180 ]  Cognition through a Literary Lens

 

unrelated fiction affects and eating disorder much other than providing an
escape and a sanctuary away from the every day struggles of it. Reading
about eating disorders in specific and sometimes even food can be quite trig-
gering and effect you so I only read it during recovery to help gain advice on
how to get through it alone.

SELF-​T RIGGERING

12. Desire to exacerbate ED by being triggered → reading (especially in

a narrowly filtered mode) → triggered response → exacerbation of
ED → increased likelihood of desire to exacerbate ED [17—​including
1 non-​ED, describing her daughter who has an ED] When I am feeling
‘anti-​recovery’ feeling fit or trying to lose weight, I tend to search for the
kind of books that will trigger me. I feel for me the pathway would have to
be something like … . Weight gain -​> relapse-​> seeking triggering books,
magazines or documentaries out-​> change in behaviour. For me, the starting
point is most definitely that ED voice in my head, not the books.
13. Reading while exercising → (intentionally) longer /​more intense
exercise → exacerbation of ED → more exercise /​perceived need
to exercise [1]‌I tend to read ED fiction while exercising, which is royally
fucked up. It encourages me to keep going.

DISTRACTION

14. Reading as a distraction from ED → action towards recovery is not

taken → exacerbation of ED → greater need for distraction through
reading [5—​incl. 1 non-​ED, referring to other life stresses] At the
worst stage of my eating disorder, reading was my way of passing the time
between meals and used as a distraction from hunger pangs. I  could get
lost in a book and not notice the time pass like I would otherwise. In that
sense, reading was having an adverse affect. But I still love reading just as
much now, and I would not blame it as any sort of accomplice to my physi-
cal deterioration. I’d much rather read than do exercise! That’s how I get my
endorphins.

ENGAGEMENT WITH TEXTUALLY EVOKED CHARACTERS

15. Reading focused on competitive comparisons of self with charac-

ters → preoccupation with ED, changed points of reference → vali-
dation/​exacerbation of ED → reading more likely to be focused on
competitive comparison [5]‌I feel more anxious and more obsessed with
my weight and eating behaviour. I compare myself unfavourably to the ED

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 181 ]
 

sufferer in the book. I feel inadequate and worried that I’ve been complacent
and not previously realised quite how lazy, fat, etc i was being and that I
need to do more to meet the eating disorder’s required standards because the
book just changed the goalposts.

READING AS PHYSICAL/​P SYCHOLOGICAL CHALLENGE

16. ED compromises ability to sit still and/​ or concentrate and/​ or

empathise needed for reading → prevention of reading → prevention
of positive effects of reading → exacerbation of ED and inability to
read [25] I find reading can offer me an escape from my eating disorder. I do
find if I read for too long I can become depressed and the eating disorder voice
usually starts saying I’m lazy and I should be exercising not reading.
17. Recovery increases ability to sit still and/​or concentrate and/​or
empathise needed for reading → increased reading → positive
effects of reading → progress in recovery [5]‌During my lowest points
I  really struggled with concentration so I  couldn’t actually sit and read a
book. I found it a massive boost when I started to recover that I could re-​read
books and eventually read new books. Read books of my preferred genre
gave me massive amounts of comfort and helped me feel detached from my
problems-​if only for a little while.
18. ED compromises ability to concentrate needed to read → worsened
mood and/​or self-​esteem due to difficulty of reading → prevention
of positive effects of reading attained → exacerbation of ED [and/​
or direct exacerbation from worsened mood/​self-​esteem] [1]‌I find
it more difficult to read when I’m in the grips of my eating disorder, i’ve
just come out of a month i hospital where the OT only had John Grisham
books, and it was difficult to read due to concentration levels. While my self
esteem and view of my body may not change following these works of fic-
tion, I usually do feel worse as I’m upset at my inability to read like I used to.
Regardless of content.
19. Finding/​making time to read improves mood → increased likeli-
hood of enjoying reading → increased likelihood of reading more
[1]‌I feel good finding the time to read. I love reading about different charac-
ters and their inner and outer world.

It is important to note that alongside quite some variation in degree of

specificity—​particularly between the first (very general) category and the
others—​not all these loops were described in their entirety; in many cases
two or more of the steps were described, and the completion of the loop
is trivially inferred. The illustration given for type 7 is fully articulated,

[ 182 ]  Cognition through a Literary Lens

 

whereas an example of partial articulation is given under type 15. Here, the
respondent does not explicitly say that the exacerbation of her ED makes it
more likely that she will want to further exacerbate it in future, but this fol-
lows predictably from what we know about the psychopathology of EDs; in
the absence of any indication that the exacerbation led, for the respondent,
to heightened awareness or other change towards greater health through a
specific self-​limiting event (for instance, reaching such a point of physical
weakness or psychological debility that change is sought out or enforced by
others), it is unproblematic to infer that the exacerbation continues.
Another point to note is that not every loop is a full loop back to the
starting state or event; in some, the latter acts as a feedforward element
of the loop. In types 2, 3, 4, 5, 7, 9, 10, and 13 reading is an initiating event
for a positive feedback loop which then becomes self-​perpetuating whether
or not more reading is undertaken. In such cases, the potential for reading
to have significant effects may be quite independent of reading frequency,
which underlines the importance of not jumping to conclusions when
assessing the likelihood of certain populations being affected (whether
towards health or illness) by reading encounters. As indicated by the exam-
ple given for type 2, one book can be enough to make a lasting difference.
Interestingly, one respondent also described how reading led, quite sim-
ply, to the desire to enter into a positive feedback loop, which she expressed
through the verb ‘to spiral’:  I feel like a failure for never reaching that low
weight or find myself comparing habits. I get angry for what I’ve eaten that day
and obsess over it. I become more determined to spiral. The same respondent
said something very similar of one of the characters in a particular TV pro-
gramme (Skins), saying here that she found herself desperate to spiral.
It is also worth stressing that positive feedback need by no means be
associated exclusively with detrimental effects. In three of the feedback
types listed above (types 1, 11, 17, and 19), positive feedback has beneficial
effects by amplifying benefits that alone would be much less powerful. The
prevalence of positive feedback structures in these data also does not at
all mean that negative feedback, tending towards stability, never features.
Although it is found in significantly fewer responses (19 as opposed to 97,
from 18 different respondents, 1 with no personal history of disordered
eating), four basic types are identifiable, the first (and most general) by far
the most frequently occurring:

1. Feel bad → read → feel better [13—​including 1 non-​ED] If I feel very

low I like to read my favourite childhood or teenage books and this always
makes me feel happier.

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 183 ]
 

2. Low mood → thoughts about body → reading → distraction from

thinking about body → improved mood [1]‌It’s a distraction so I’m not
thinking about my body. Also thinking about my body is connected to low
mood—​I usually only do it when depressed. So cheering up means I  think
less/​less negatively about my body.
3. Temptation to restrict eating → reading → reduced anxiety, increased
likelihood of eating [1]‌I am more likely to prepare myself a snack or a meal
after reading. Sometimes when I  am tempted to restrict I  read for half an
hour before a meal to get myself into the mood. Often I read at table as a way
to counter anxiety and distract myself from the fact that I’m eating.
4. In recovery → reading → compromised recovery [4]‌Eating disorder fic-
tion always negativly effects my recovery. I should not read them but I always
do. I have this sick fascination that draws me into them. I like to live vacarisly
through the characters because I cannot have my eating disorder.

The first three of these forms of feedback offer concrete avenues for devel-
oping therapeutic strategies involving fiction-​reading. The positive poten-
tial suggested by the most common negative feedback structure, the simple
progression from feeling bad to reading to feeling better—​centring on
mood as the initiating and the altered variable—​is supported by quanti-
tative data from this survey. An improvement to mood is reported by a
majority of respondents as a result of reading their preferred type of ‘other
fiction’ (i.e. fiction not about EDs): of the 465 respondents with a history
of disordered eating who reported having read other fiction (68 said they
had not, and 240 did not answer this question), 336 (72%) reported a posi-
tive effect on general mood resulting from it. This result combined with
the beneficial negative feedback loops identified here together testify to
the possibility of displacing the often dangerous positive feedback in the
relationship between EDs and reading by stability-​promoting negative
feedback, perhaps particularly with mood as a mediator. This means that
reading may be seen not just to offer benefits where none would otherwise
be available, but even more significantly to offer the potential to replace
structurally dangerous effects of reading with structurally stabilizing
effects.
Further work along these lines might also start to identify more spe-
cific features of the observed feedback structures which are likely to have
a bearing on stability. Two such features are the sensitivity with which
changes in input are detected by the feedback system, and the aggressive-
ness with which input changes are adjusted for by the system ‘controller(s)’
through outputs from the system. There is always a trade-​off between
the two:  the more aggressive the controller (attempting to control for

[ 184 ]  Cognition through a Literary Lens

 

even small perturbations very quickly through excessive compensation),

the greater its sensitivity to errors in input measurement and the greater
the likelihood of overshooting the required response. This balancing act
has been identified as a distinctive feature of ED pathology (Smith, 2002,
p. 96). In anorexia, for example, over-​frequent self-​weighing often leads to
unnecessary adjustments in food intake (reduced) and exercise (increased)
when a small increase in bodyweight is observed. On the other hand, when
weight loss is observed, it usually leads either to no behavioural change
at all (more of the same), or to further adjustments in the same direction
(more exercise, less food), now spurred on by evidence of ‘success’. In both
cases, rather than behaviour being appropriately geared to real trends in
bodyweight change, the ‘noise’ of transient fluctuations is amplified by
obsessive self-​weighing and the feedback control that is too aggressively
calibrated to the input data from the scales. This results in a bodyweight
that has natural day-​to-​day fluctuations (i.e. is stable within given bounds),
but is unstable (i.e. constantly reducing) on a timescale of weeks and
months. In some of the survey data presented here, we can see that read-
ing fiction, especially ED fiction, has the potential to further heighten the
oversensitivity of the system, for example: Often, it can trigger that loop of
obsessive thinking, remind you to keep a food diary and weigh in. More spe-
cifically, this reciprocal ratcheting-​up of sensitivity and aggressiveness may
occur through comparison of one’s own body and/​or dietary intake with
the character’s:

If I read about someone who is very anorexic (either states weight or graphic descrip-
tion) I feel fat and ugly. When i was anorexic and the author weghed more than me
I felt good, but inevitably they would describe losing weight and weiging less than me,
and then I felt fat and ugly. Sometimes I feel inspired to lose weight. (see also the
example under loop type 15 above)

This last example leads on to a second feature of interest: the possibility of

feedback loops being short-​circuited. In the ED context, this often happens
by way of cognitive biases, of which the best-​known is body dysmorphia
(in the popular—​and exaggerated—​cliché, the emaciated woman looks in
the mirror and sees a fat person). Here feedback is operating, but is act-
ing on incorrect input signals due to perceptual distortion. A specific vari-
ant of this is a phenomenon known as ‘thought-​shape fusion’ (TSF), in
which just thinking about a ‘forbidden’ food increases someone’s estimate
of their body size, shape, and/​or weight (Shafran, Teachman, Kerry, and
Rachman, 1999). In some of the survey responses we see an interesting
variation on TSF which we might call ‘inverse TSF’: here it is not thoughts

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 185 ]
 

about eating, but thoughts about someone else not eating that have the
distorting effects. This is potentially all the more dangerous, because the
sufferer need not think directly about food, but may think about it at two
removes:  about (1)  someone else (2)  not eating it. Of course, this might
happen just as easily with other real people as with fictional characters,
but as an activity where the immediate task demands of normal life are
suspended, reading does offer a very good opportunity for uninterrupted
comparisons, as well as for comparisons made on the basis of even more
inadequate evidence than in direct social encounters. ED fiction, of course,
also offers easy access specifically to protagonists who eat unhealthily, but
while the not-​eating is in these texts pathological, not-​eating can figure
more or less innocuously in all kinds of ‘literary’ and ‘genre’ fiction, where
eating gets in the way of plot, conflicts with the evocation of a particular
type of character or situation, or is otherwise not the kind of subject mat-
ter deemed quite worthy of inclusion. When as a reader one has cognitive
biases resulting from an ED, even the harmless absence of mentions of
mealtimes could create a reliable supply of false feedback about characters
and their relation to one’s own body and eating habits.

FEEDBACK SYSTEMS IN THE READING PROCESS

I hope now to have shown how important feedback is for understand-

ing the relationship between reading and disordered eating. The question
now is how this relationship can most effectively be studied. In the more
basic examples of feedback structures reported by our survey respondents,
one could imagine that many other cultural artefacts (like music or film,
say) might have similar effects:  cheering you up, distracting you, forcing
you to sit still, making you realize you can’t concentrate. Another set of
effects could easily be attributed to ED-​specific non-​fiction material online,
whether letting yourself be almost deliberately ‘triggered’ into emulating
new ED behaviours, or conversely gathering information to help with recov-
ery. (One respondent gestures towards this medial equivalence: I wanted
books on the in’s and out’s of a ‘successful’ eating disorder. Especially before
I discovered the pro-​ana community online.) Many of these factors could be
meaningfully investigated using the methods of psychiatry, sociology, and/​
or media studies, as in the research on the mass media I discussed above.
Even in cases like these, however, incorporating insights from cognitive
literary studies would arguably help give appropriate weight to the relation-
ships between textual and psychological factors, especially where the texts
include metaphor, conspicuous rhetorical devices, the use of fictionalized

[ 186 ]  Cognition through a Literary Lens

 

perspectives, and so on. But in many of the examples above, the case for
bringing a cognitive-​literary perspective to bear can be made cogently and
with precision. One way to think about the usefulness of cognitive literary
studies to research on the possible detrimental and therapeutic effects of
reading in an ED context is to identify one salient element of the feedback
structure and then try to clarify its role with reference to existing cognitive-​
literary findings and debates.
For example, respondents’ testimony about their engagement with fic-
tional characters may be elucidated with reference to Don Kuiken and col-
leagues’ (2004) investigation of different kinds of resonance established
between readers and fictional situations or characters. They make a dis-
tinction between identificatory acts based on simile and on metaphor. In
our data here, both can be found: I find this a reminder that I should not be
eating as a woman who is stressed with uni work etc (simile); Books in which
other people face struggles, not necessarily eating disorders are also helpful as
they highlight that you are not the only one who suffers/​struggles and allows you
to gain some clarity/​perspective (metaphor). The authors’ suggestion that
the metaphorical form of connection-​making might have more potential
for changing readers’ sense of self ties in closely to our concern here with
the readerly changes, both beneficial and detrimental, that can be effected
through reading.
The mental health context bears some similarity to that of bereavement,
which has also been investigated in connection with reading by Kuiken and
his colleagues: dissociative experiences, for example, in which people feel
distanced from their own feelings and actions, are common in both. Links
have been suggested between loss and dissociation, ‘sublime disquietude’
(depth of self-​perception combined with feelings of discord and release),
and insight (including inclination to change) in reading; and connections
have been observed between physical and psychological well-​being, iden-
tification, and self-​efficacy in the modulation of sadness when watching
films (for a summary, see Kuiken and Oliver, 2013).
In a broader sense, however, one might also identify important struc-
tural parallels between the feedback loops found to operate in reading in
the ED context and feedback loops identified as relevant to the reading pro-
cess more generally. The wide range of processes that come under the head-
ing of ‘interpretation’ all involve important forms of feedback. At the level
of text comprehension, the decoding of letters and words, which is driven
substantially by the physiology of eye movements and fixations, drives
semantic comprehension, which in turn has effects back on the way the text
is perceptually processed. For instance, a sentence-​completion study and a
self-​paced reading study by Rohde and colleagues (Rohde, Levy, and Kehler,

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 187 ]
 

2011) indicated that, contrary to what had previously been believed, prag-

matic expectations and inferences about discourse coherence can influence
low-​level syntactic processing, contributing to disambiguation as quickly
as do lexical, morphological, or syntactic cues. At a higher level, aspects
of cultural knowledge as well as individual differences in goals, expertise,
and experience drive interpretive choices that may also affect strategies at
the level of comprehension if not decoding. In these senses, reading can
be seen as one manifestation of the predictive processing that has been
argued (by e.g. Clark, 2013) to structure cognition as a whole: very broadly
speaking, we make predictions about what we are reading, which are either
confirmed or disconfirmed by the text itself, causing the prior model to be
either strengthened or updated. We might think of ourselves as ‘Bayesian
readers’ who make optimal decisions based on the available information
(Norris, 2006).
This basic structure is obviously subject to variation depending on the
type of text. For example, Richard Walsh (2006; and personal commu-
nication, 19 January 2016) has argued that fictional and non-​fictional
narrative can be understood as different forms of semiotic feedback loop
(see also Carney, 2008, on lyric and catastrophe theory, and Rinaldi,
2008, on dynamical systems in Petrarch’s love poetry). In what began as
a productive misreading of Walsh’s argument, I would suggest that we
can think of fiction as tending to create a positive feedback loop between
textual content and interpreted meaning, whereas non-​fiction creates
negative feedback because it refers more straightforwardly to things
outside the text. In non-​fiction, broadly speaking, the role of the lan-
guage is to point towards a real-​world referent, such that readers’ inter-
pretive expectations are progressively narrowed down towards identity
with the textual references and their real-​world referents. In fiction, by
contrast, where the essence of the textual communication is as much
significance as referentiality (the ratio, insofar as it can be categorically
established at all, will depend on genre), more noise is present in the
system. Here, the reader is more likely to include a wider set of pos-
sible interpretations for every linguistic element, and the text is more
likely to encourage such interpretive openness; so convergence between
readerly expectations and textual reference does not necessarily occur
(see Figures 9.4a and 9.4b). The fictional structure would leave much
more space for the kind of highly filtered readerly engagement reported
by some survey respondents, driven as much by attitude as by the text
itself: Its hard to blame the books or the authord. I feel it’s more than your
ED screens out that information about pain and suffering and focuses on the
success, the control and power.

[ 188 ]  Cognition through a Literary Lens

 

(a)

“m” “a” “t” [t = 2]

Textual input “c” “a” “t” [t = 1]

Semantic interpretation “cat”

Recognition

Meow
Associations Furry
Mat

(b)

“t” “r” “e” “e” [t = 2]

Textual input “c” “a” “t” [t = 1]

Semantic interpretation “cat”

Goddess
Recognition Good/bad luck
Cheshire

Jabberwocky, grin
Associations Witches, broomsticks
Puss in Boots

Figure 9.4  (a) Text-​interpretation feedback in non-​fiction. The interpretive process and the
textual inputs are typically constrained and convergent. (b) Text-​interpretation feedback in
fiction. Here both the interpretation and the textual inputs are more likely to be heteroge-
neous and therefore also divergent.
 

Alongside the specifics of text type, the nature of the reading situation
may also be expected to affect the probability of unstable positive feedback
developing in the reading process. If we consider the distinction between
solitary and group reading, for example, it is clear that the former allows
much more scope for interpretations that diverge from what is given in the
text to multiply unchecked, whereas the latter provides an inbuilt mecha-
nism for correction, or control, from the rest of the group; group discus-
sion, including perhaps criticism of a particular line of interpretation as
implausible or not supported by the textual evidence, can thus fulfil the
function of a feedback control mechanism. Where there is no such mecha-
nism, self-​regulation is less likely, and ‘solipsistic interpretation and error’
more likely (Majkut, 2014). Similar considerations may apply to profes-
sional (traditional literary-​critical, exegetical) reading versus reading for
pleasure. For scholarly readers, the pressures are primarily towards finding
readings that diverge from existing ones and are superior to them in detail
and ingenuity, skewing the selection pressure in favour of divergence from
textual reference. For recreational readers, the general aims of distraction
or escapism, or reading for plot, may encourage convergence between the
primary semantic associations of the words on the page and interpretive
possibilities that readers entertain. These kinds of factors, of course, make
designing experiments difficult, since even small changes in setting (test-
ing room versus living room, during the working day or in the evening) are
likely to be significant.
When it comes to the structures of fictional plot and fictional worlds,
Karin Kukkonen (2014) has suggested that a feedback model is needed
to account for their relationship with readers’ expectations, or predic-
tive models; readers may well find themselves with competing probabi-
listic models for a given fictional world, and ‘as readers move through
the narrative, they revise their beliefs about the shape of the fictional
world and (usually) get a progressively better grasp of its probabilities’
(p. 725), using any and all textual evidence at the fine grain of words and
phrases to the coarser grain of plot structure to contribute to the loop
between prior hypotheses and new observations. Fiction here differs
from the real world, and from non-​fiction, in being designed expressly
‘to enable constantly new, unpredicted observations and thus reconfig-
ure the probabilities of the fictional world’ (p. 725).
Also importantly for our purposes here, this kind of model makes space
not just for narrowly ‘intellectual’ inference but also for embodied emo-
tional responses and the wider patterns of appraisal they contribute to,
whether directly plot-​related emotions like suspense and surprise, or the

[ 190 ]  Cognition through a Literary Lens

 

broader range of emotional responses we may have to other elements of

a text, including the feeling of being ‘immersed’ or ‘transported’ into its
world (e.g. Gerrig and Mumper, Jacobs, and Polvinen, all this volume).
The appraisal theory of emotion (e.g. Frijda, 2007)  posits the idea that
emotional responses result from the appraisal of what a given object or
situation means to me now, with respect to my current goals and expec-
tations, and that this initial appraisal can be subject to further elabora-
tive appraisals, as part of a feedback loop also involving physiological and
behavioural changes (Pirlet and Wirag, this volume). This adds another
layer to the structures through which a small anomaly in evaluative priors
can result in a wide-​ranging distortion in the cognitive landscape during
the act of reading.
It is easy to see how the contextual specifics of the ED psychopathol-
ogy could slot into this concatenation of recursive loops by virtue of the
particular pressures and probabilities it creates. Andy Clark reports on
evidence that offers a twist on his global model of predictive processing
to account for deviations that occur in mental illness. He focuses spe-
cifically on conditions that involve alterations in the dopamine system;
these are suggested to lead to false generation and high weighting of pre-
diction error signals that then drive maladaptive learning. In the case
of schizophrenia, the ‘false errors’ propagate throughout the hierarchy
of perceptions and beliefs, creating a self-​entrenching process in which
the influence of new beliefs ‘flows back down so that incoming data is
sculpted by the new (but now badly misinformed) priors so as to “con-
form to expectancies”. … False perceptions and bizarre beliefs thus form
an epistemically insulated self-​confirming cycle. This, then, is the dark
side of the seamless story … about perception and cognition’ (2013,
p. 17; see also Brisch et al., 2014).
In the ED case, the ‘false perceptions and bizarre beliefs’ may be less
surreal than in schizophrenia, but they are no less self-​perpetuating: my
tummy is still too fat, I ate too much today, if I eat less it makes me a bet-
ter (stronger, purer, more moral) person. Disturbances to the dopaminer-
gic reward system have been implicated in ED psychopathology (Kaye,
Frank, and McConaha, 1999; Méquinion et  al., 2013), and although it is
hard to establish whether these abnormalities are cause or effect (or both),
the increased dopamine release during fasting (Bergh and Södersten,
1996) means that dopamine acts in anorexia and other EDs involving peri-
ods of dietary restriction as a ‘learning signal’ in a way comparable to its
role in substance addiction. Steven Hyman discusses the short-​circuiting
that occurs when the signal that dopamine release sends—​‘better than

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 191 ]
 

expected’—​is sent repeatedly by direct pharmacologic action, bypassing

the usual controls that compare the current circumstance with prior expe-
rience (Hyman, 2007). The short-​circuiting is not as extreme in restrictive
EDs as in most drug use, but the overlearning and overvaluation that are
lastingly inscribed through misweighted prediction errors have similar
implications for how we think about the relationship between interpreta-
tion and mental health.
We can understand the excessive significance that may be given to par-
ticular aspects of a text—​like the body size or shape of the protagonist, for
example—​as part of a complex series of feedback loops that structure all
elements of the reader’s interaction with the text, from the decoding and
semantic processing of the words on the page or screen, to the engage-
ment with plot, character, and genre, to the contextual effects of physical
and psychological state, motivations and intentions, mood, and setting.
Acknowledging the power of feedback for both good and ill at the many
contact points between reading and mental health can help us understand
and perhaps ultimately prevent or treat EDs more effectively. I hope to
have shown here how significant a contribution fiction-​reading can make
to these feedback loops as a mediator of cultural causes and effects—​some
common to other cultural forms, some specific to fiction. Unravelling the
details of this contribution, and developing new therapeutically valuable
models in which reading may act as a control mechanism to modulate feed-
back in beneficial ways, requires an ambitious cognitive literary science
able to talk and listen to literary studies, the medical humanities, psychol-
ogy and psychiatry, as well as disciplines more apparently distant like sys-
tems and control theory.

ACKNOWLEDGEMENTS

This research was made possible by a Knowledge Exchange Fellowship at

The Oxford Research Centre in the Humanities (TORCH), and by the gener-
ous and creative collaboration of Jonathan Kelly at Beat. Ethical approval
for the online survey was provided by the Central University Research
Ethics Committee at the University of Oxford (MSD-​IDREC-​C1-​2014-​219).
I am grateful to Michael Burke, James Carney, and Dhruva Raman for help-
ful comments on earlier drafts of this chapter, and to my partner (and con-
trol theorist) James Anderson for all his tireless assistance and advice on
feedbacks. Above all, I thank everyone who gave their time to complete the
survey; your thoughtful responses have already taught me a lot, and will
continue to.

[ 192 ]  Cognition through a Literary Lens

 

REFERENCES

Åström, C. J., and Murray, R. M. (2014). Feedback systems: An introduction for scientists
and engineers (2nd ed.). Princeton, NJ: Princeton University Press.
Bergh, C., and Södersten, P. (1996). Anorexia nervosa, self-​starvation and the reward
of stress. Nature Medicine, 2(1), 21–​22.
Bernini, M. (2014). Supersizing narrative theory: On intention, material agency and
extended mind-​workers. Style, 48(3), 349–​366.
Brisch, R., Saniotis, A., Wolf, R., Bielau, H., Bernstein, H.-​G., Steiner, J., et al. (2014).
The role of dopamine in schizophrenia from a neurobiological and evolutionary
perspective: Old fashioned, but still in vogue. Frontiers in Psychiatry, 5, article
47, see http://​www.ncbi.nlm.nih.gov/​pmc/​articles/​PMC4032934/​pdf/​fpsyt-​
05-​00047.pdf
Caracciolo, M. (2014). Interpretation for the bodies:  Bridging the gap. Style, 48(3),
385–​403.
Carney, J. (2008). ‘Unweaving the rainbow’:  The semantic organization of the lyric.
Journal of Literary Semantics, 37, 33–​53.
Chesters, T. (2014). Social cognition: A literary perspective. Paragraph, 37(1), 62–​78.
Churchland, P. S. (1996). The hornswoggle problem. Journal of Consciousness Studies,
3(5‒6), 402–​408.
Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of
cognitive science. Behavioral and Brain Sciences, 36(3), 1–​73.
Cohen, R., and Blaszczynski, A. (2015). Comparative effects of Facebook and con-
ventional media on body image dissatisfaction. Journal of Eating Disorders,
3(23), 1–​11.
Dennett, D. (1991). Consciousness explained. Boston, MA: Little, Brown.
Fairburn, C. G. (2008). Cognitive behavior therapy and eating disorders. New  York:
Guildford.
Fisher, C. A., Hetrick, S. E., and Rushford, N. (2010). Family therapy for anorexia ner-
vosa (Review). Cochrane Database of Systematic Reviews, 4, CD004780.
Frijda, N. H. (2007). The laws of emotion. Mahwah, NJ: Lawrence Erlbaum.
Gonzalez, A. L., and Hancock, J. T. (2011). Mirror, mirror on my Facebook Wall: Effects
of exposure to Facebook on self-​esteem. Cyberpsychology, Behavior, and Social
Networking, 14(1‒2), 79–​83.
Grabe, S., Ward, L. M., and Shibley Hyde, J. (2008). The role of the media in body
image concerns among women:  A  meta-​analysis of experimental and correla-
tional studies. Psychological Bulletin, 134(3), 460–​476.
Hyman, S. E. (2007). The neurobiology of addiction: Implications for voluntary control
of behavior. American Journal of Bioethics, 7(1), 8–​11.
Kaye, W. H., Frank, G. K. W., and McConaha, C. (1999). Altered dopamine activity after
recovery from restricting-​ type anorexia nervosa. Neuropsychopharmacology,
21(4), 503–​506.
Kirsh, S. J. (2010). Media and youth:  A  developmental perspective. Chichester, UK:
Wiley-​Blackwell.
Kuiken, D., Miall, D., and Sikora, S. (2004). Forms of self-​implication in literary read-
ing. Poetics Today, 25(2), 171–​203.
Kuiken, D., and Oliver, M. B. (2013). Aesthetic engagement during moments of suffer-
ing. Scientific Study of Literature, 3(2), 294–​321.
Kukkonen, K. (2014). Bayesian narrative:  Probability, plot and the shape of the fic-
tional world. Anglia, 132(4), 720–​739.

F e e d b a c k i n R e a di n g a n d Dis or de r e d E at i n g   [ 193 ]
 

Kukkonen, K., and Caracciolo, M. (2014). Introduction:  What is the ‘second genera-
tion’? Style, 48(3), 261–​274.
Kuzmičová, A. (2014). Literary narrative and mental imagery: A view from embodied
cognition. Style, 48(3), 275–​293.
Latzer, Y., Spivak-​Lavi, Z., and Katz, R. (2015). Disordered eating and media exposure
among adolescent girls: The role of parental involvement and sense of empower-
ment. International Journal of Adolescence and Youth, 20(3), 375–​391.
Majkut, P. (2014). Smallest mimes:  Defaced representation and media epistemology.
Bucharest: Zeta Books.
Méquinion, M., Langlet, F., Zgheib, S., Dickson, S., Dehouck, B., Chauveau, C., et al.
(2013). Ghrelin:  Central and peripheral implications in anorexia nervosa.
Frontiers in Endocrinology, 4, article 15, see http://​www.ncbi.nlm.nih.gov/​pmc/​
articles/​PMC3581855/​pdf/​fendo-​04-​00015.pdf
Menary, R. (2010). Introduction to the Special Issue on 4E Cognition. Phenomenology
and the Cognitive Sciences, 9(4), 459–​463.
Miall, D. (2011). Wordsworth’s ‘first-​born affinities’: Intimations of embodied cogni-
tion. Poetics Today, 32(4), 693–​715.
Minuchin, S., Rosman, B. L., and Baker, L. (1978). Psychosomatic families: Anorexia ner-
vosa in context. Cambridge, MA: Harvard University Press.
Mitrophanov, A. Y., and Groisman, E. A. (2008). Positive feedback in cellular control
systems. Bioessays, 30(6), 542–​555.
Molé, P. A. (1990). Impact of energy intake and exercise on resting metabolic rate.
Sports Medicine, 10(2), 72–​87.
Norris, D. (2006). The Bayesian reader:  Explaining word recognition as an optimal
Bayesian decision process. Psychological Review, 113(2), 327–​357.
Rinaldi, S. (2008). Laura and Petrarch:  An intriguing case of cyclical love dynamics.
SIAM Journal of Applied Mathematics, 58(4), 1205–​1221.
Rohde, H., Levy, R., and Kehler, A. (2011). Anticipating explanations in relative clause
processing. Cognition, 118(3), 339–​358.
Shafran, R., Teachman, B., Kerry, S., and Rachman, S. (1999). A cognitive distor-
tion associated with eating disorders: Thought-​shape fusion. British Journal of
Clinical Psychology, 38(2), 167–​179.
Smith, J. L. (2002). The psychology of food and eating:  A  fresh approach to theory and
method. Basingstoke, UK: Palgrave.
Troscianko, E. T. (2009). A hunger artist. Retrieved from https://​www.psychologyto-
day.com/​blog/​hunger-​artist
Troscianko, E. T. (2013). Reading imaginatively: The imagination in cognitive science
and cognitive literary studies. Journal of Literary Semantics, 42(2), 181–​198.
Troscianko, E. T. (2014). Kafka’s cognitive realism. New York: Routledge.
Troscianko, E. T. (In preparation). Reading and eating disorders: An online survey.
Wallace, B., Ross, A., Davies, J., and Anderson, T. (Eds.). (2007). The mind, the body and
the world: Psychology after cognitivism? Exeter, UK: Imprint Academic.
Walsh, R. (2006). The narrative imagination across media. Modern Fiction Studies,
52(4), 855–​868.
Zeeman, E. C. (1976). Catastrophe theory. Scientific American, April, 65–​70, 75–​83.

[ 194 ]  Cognition through a Literary Lens

 

CHAPTER 10

Animal Minds across

Discourse Domains
DAVID HERMAN

I n this chapter, I seek to demonstrate the advantages of the ‘transdisci-

plinary’ approach to research on the mind-​narrative nexus outlined in a
previous study (Herman, 2013). The goal of transdisciplinary work, in gen-
eral, is to promote genuine dialogue and exchange among multiple fields
of research around a shared focus of inquiry, rather than engaging in uni-
directional borrowing from a particular field that thereby becomes domi-
nant (see also Hartner’s chapter in this volume). Thus, my earlier study
discussed how mind-​orientated frameworks for narrative scholarship can
not only be informed by but also inform research on human intelligence.
In this chapter, I shift the emphasis to fictional and non-​fictional accounts
of the experiences of non-​human animals, and consider strategies for
fostering fuller, more open dialogue between narratological approaches to
stories that engage with animal subjectivity, on the one hand, and phenom-
enological, ethological, anthropological, and other studies of animals and
human‒animal relationships, on the other hand. Indeed, as is the case with
questions about storytelling vis-​à-​vis human minds, because of their com-
plexity and many-​sidedness questions about the narrative projection of
animal experiences can arguably only be addressed by a cross-​or transdis-
ciplinary approach that brings together insights from the natural sciences,
the social sciences, and the arts and humanities. Reciprocally, inquiry into
narratively organized discourse about animal worlds and human‒animal
interactions across a variety of contexts can foster transdisciplinarity,
 

providing new opportunities to connect the aforementioned areas of

inquiry or, as Kagan (2009) calls them in his update of C. P. Snow (1998),
cultures.
I begin by situating my analysis vis-​à-​vis a representative debate about
animal subjectivity:  namely, the debate between Nagel (1974, 1986)  and
Dennett (1991) concerning the opacity or accessibility of animal minds. I seek
to reframe this debate by working towards a transdisciplinary synthesis of
narratology and ‘heterophenomenology’, a term that I adapt from Dennett
to refer to the ascription of subjective experiences to others. Focusing on
interspecies rather than intraspecies ascriptive practices, i.e. heterophe-
nomenological engagements that cross the species boundary, I  suggest
that Nagel-​Dennett-​like debates organized around a polarity between leg-
ible and illegible animal minds should give way to a new, transdisciplinary
project: developing techniques for documenting and analysing the attested
range of mind-​ascribing practices in a given culture or subculture, as they
manifest themselves in non-​fictional as well as fictional narratives anchored
in a variety of ‘discourse domains’. I use this expression as a technical term to
refer to frameworks for activity that, operative in the full range of cultural,
subcultural, and interpersonal settings, determine what sorts of ascriptive
practices will be deemed appropriate and warranted in a given context; I also
argue that the norms for ascription associated with such discourse domains
cut across the fiction‒non-​fiction divide. Thus, making a claim that has
implications both for narratological research on thought presentation and
for scholarship on animal subjectivity, I  argue that domain, not genre, is
the key determinant of how prolific and detailed the heterophenomenol-
ogy projected by a given narrative will be. It is not the fictional versus non-​
fictional status of a narrative that sets the upper (or lower) limit on how
many mental-​state attributions can be made and the degree to which those
attributions will be fine-​grained and particularized rather than coarse and
general. Rather, the relative richness and granularity of accounts of animal
subjectivity will be determined by how a given narrative bears on the norma-
tive assumptions about species of minds that structure discourse domains—​
my term for the more or less distinctive arenas of conduct in which ways of
orientating to self‒other relationships take shape.

ANIMAL MINDS BETWEEN NARRATOLOGY

AND HETEROPHENOMENOLOGY

A key goal of this chapter is to recontextualize—​and suggest strategies

for moving beyond—​a debate that has grown up around the question

[ 196 ]  Cognition through a Literary Lens

 

of animal minds in the field of philosophy. Nagel stands on one side of

this debate; he posits that animal minds are radically inaccessible—​but
arguably without doing justice to observable modes of relatedness and
rapport between humans and other animals. Dennett stands on the
other side; he posits that human and non-​human minds are equally
accessible—​but arguably without doing justice to potential heteroge-
neities in the structure of experience across species lines. My aim here
is to steer a course between the Scylla of the radical inaccessibility of
non-​human minds and the Charybdis of experiential homogenization
or flattening, by arguing that mind-​ascribing acts, rather than occurring
in decontextualized, one-​off acts of attribution, always unfold within
particular arenas of practice, or discourse domains. Such domains deter-
mine when, to what extent, and in what manner it is appropriate and
warranted to impute subjective experiences to others, non-​human as
well as human. Thus, in lieu of any top-​down dichotomization of leg-
ible and illegible animal minds, I propose working inductively towards
an understanding of the spectrum of attested mind-​ascribing practices
as they take shape in a given culture or subculture, with this spectrum
ranging from minimal to maximal projections of mind across the species
boundary.
Both to constrain my analysis and to open up possibilities for transdis-
ciplinary exchange, I focus on the way such ascriptive practices unfold in
narratively organized discourse about animals, non-​fictional as well as fic-
tional. Here it should be noted that Nagel’s thesis of radical opacity carries
the corollary that only fictional accounts of animals can support abundant,
detailed ascriptions of mental experiences to non-​human agents. Dennett’s
antithesis carries the opposite corollary: namely, that the construction of a
non-​ fictional, or falsifiable, account is required to build up a profile
of another being that can reliably capture what it is like to be that sort
of being. To push past these interlinked polarities in discourse on animal
minds—​polarities because of which a division between legibility and illeg-
ibility leads in turn to a dichotomized approach to narrative genres—​it is
necessary to rethink the core assumptions on which the Nagel‒Dennett
debate is grounded.

Rethinking Heterophenomenology

Nagel concludes his 1974 article ‘What Is It Like to Be a bat?’ with specula-
tions about the possibility of establishing an ‘objective phenomenology’,
or a theory of what it is like to be a particular kind of creature based on

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 197 ]
 

the creature’s physiological structure, perceptual capabilities, behavioural

dispositions, and so forth. As Nagel puts it, ‘though presumably it would
not capture everything, its [objective phenomenology’s] goal would be to
describe, at least in part, the subjective character of experiences in a form
comprehensible to beings incapable of having those experiences’ (p. 449).
Nagel further develops these ideas in his 1986 book The View from Nowhere.
In this study, Nagel seeks to make space for ineliminably subjective experi-
ences within a broadly naturalistic and scientific worldview. This project
leads to the following memorable formulation: ‘We will not know exactly
how scrambled eggs taste to a cockroach even if we develop a detailed
objective phenomenology of the cockroach’s sense of taste. When it comes
to values, goals, and forms of life, the gulf may be even more profound’
(p. 25).
By contrast, Dennett for his part suggests that there is no funda-
mental distinction between the way heterophenomenology plays out
in human‒human interactions, on the one hand, and in human‒non-​
human interactions, on the other. As Dennett puts it in Consciousness
Explained, taking up Nagel’s central example, we can ‘rank order hetero-
phenomenological narratives for realism,’ rejecting those that assert or
presuppose

discriminatory talents, or reactive dispositions, demonstrably not provided for

in the ecology and neurophysiology of the bat. … When we arrive at hetero-
phenomenological narratives that no critic can find any positive grounds for
rejecting, we should accept them … as accurate accounts of what it is like to be
the creature in question. (pp. 443‒444)

In stark opposition to Nagel, Dennett wishes to expurge ineliminably

subjective aspects of experience; his project is to dispute the utility and
even the coherence of the concept of ‘qualia’, or raw sensory feels such as
those putatively associated with seeing the colour red. The result is that
for Dennett non-​human experiences are as accessible as human experi-
ences, but at the cost of a flattening out of qualitative differences in how
differently structured beings might encounter the world—​differences that
the philosopher-​biologist Jakob von Uexküll sought to capture with his
concept of Umwelt, or the phenomenal, subjectively experienced world to
which a creature’s organismic structure gives rise.1

1.  Uexküll’s idea of the Umwelt has been glossed by Evan Thompson (2007) as ‘an
animal’s environment in the sense of its lived, phenomenal world, the world as it pres-
ents itself to that animal thanks to its sensorimotor repertoire’ (p. 59).

[ 198 ]  Cognition through a Literary Lens

 

I return to issues raised by Nagel’s emphasis on the unbridgeability

of the divide between humans and other animals later; for the moment
I  focus on how the premises of the Nagel‒Dennett debate can be
undercut from another direction—​namely, via Dennett’s account of het-
erophenomenology. One line of criticism against Dennett targets his
mentalizing approach to ascriptions of subjective states to others. Thus,
Gallagher and Hutto (2008) critique Dennett’s notion of ‘the intentional
stance’, or humans’ evolved tendency to construe others’ behaviours in
terms of interconnected intentional and volitional states, as an over-​
intellectualizing approach that is also misleadingly spectatorial in orienta-
tion. For Gallagher and Hutto, what Dennett describes as the intentional
stance should be thought of not as the default orientation towards inter-
subjective encounters, but rather as a specialized attitude or interpretive
strategy that will be adopted only when it becomes pragmatically expedi-
ent to do so. From this perspective, when I interact with another person
I  will resort to the intentional stance only when the default, embodied,
pre-​or nonconceptual modes of sense-​making that are ontogenetically
prior and cognitively more basic than deliberative reasoning (Trevarthen,
1993) do not suffice to make clear exactly what someone is doing or why
he or she may be doing that. If I see you standing near a broken window
frame with a hammer in your hand, I will not need to compute your inten-
tions but will be able to grasp them even as (or when) I grasp the situation
at hand. The same goes for a dog who barks and pushes eagerly against the
front door: time to go out for a walk!
Likewise James (2009) and Shapiro (1997) emphasize how non-​
mentalizing, embodied modes of interaction allow for coordinated
interplay—​and mutual understanding—​between humans and companion
animals. For example, in his account of the play behaviours in which he and
his dog Lucy engage, James writes that

talk of a meeting of minds is, to speak loosely, too ‘mentalistic’ to capture my

interactions with Lucy. My being-​with Lucy, if it may be so described, [involves]
an intertwining of bodily intentions, a shared response of two lived bodies to
a common situation. It certainly cannot be understood on the basis of a model
that is merely cognitive. (p. 39)

Not only does this line of thinking further the argument that ascriptions
of subjectivity (both within and across the species boundary) should be
viewed as embedded in and shaped by particular kinds of contexts rather
than as singular, one-​off attributions; what is more, James’s and Shapiro’s
work, like Gallagher and Hutto’s, suggests that despite Dennett’s critique of

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 199 ]
 

what he calls the ‘Cartesian Theater’ model of consciousness,2 his approach

reveals a residual mind‒body dualism, which involves a detached figuring
out of rather than a cognizant co-​involvement with the postures and move-
ments of bodies, non-​human as well as human.
As these considerations suggest, in any culture a range of contexts—​or
what I will go on to describe in more detail as discourse domains—​shape
acts of mental-​state ascription across as well as within the species bound-
ary. Heterophenomenology, in this sense, is a cover term for a whole ecology
of ascriptive practices; this wider ecology needs to be taken into account in
characterizing human‒animal interactions and attributions of intentions,
emotions, volitions, and other mental states and dispositions across species
lines. Thus, again using the example of companion animals such as dogs,
Noë (2009) notes that although it is possible to treat a dog in biophysical
terms as a merely mechanistic locus of conditioned response, ‘if one is to
enter into the kind of relationship of cooperation and companionship that
characterizes our actual relations with dogs, one must leave the standpoint
of mechanism behind and instead view the dog as … a thinking being’
(p. 37). Stating the point in even starker terms, Noë writes, ‘There are two
fundamentally different ways of thinking about things. … From within one
perspective, it is impossible to doubt the mind of others. From within the
other, it is impossible to acknowledge it’ (p. 39). Here, rather than opting for
one or the other polarities of legible versus illegible animal minds, Noë sug-
gests that this dichotomy itself emerges from a larger ecology of ascriptive
practices, ranging from those in which animal subjectivity is blocked out as
a non-​factor to those in which particularized, prolific ascriptions of mind to
non-​human others are not only possible but mandated.
Crist (1999) reveals an equally diverse ecology of mind-​ascribing prac-
tices in her investigation of changing patterns in scientific discourse about
animal behaviour from the 19th to the 20th centuries. Crist focuses on the
contrast between the vernacular language of action used by analysts such
as Charles Darwin and the French naturalist Jean-​Henri Faber, on the one
hand, and the technical terms used by ethologists such as Konrad Lorenz
and Nikolaas Tinbergen, on the other hand. As Crist puts it,

grasped in the ordinary language of action [where animals are the subject of
verbs such as see, feel, pursue, etc.], knowledge of animal behavior is oriented

2.  For Dennett (1991) the Cartesian Theatre model assumes that specific subsystems
of the mind/​brain (e.g. those bound up with perception, long-​term memory, and plan-
ning) come together in some ‘central thinking area’, ‘a Cartesian Theater, a place where
“it all comes together” and consciousness happens’ (p. 39).

[ 200 ]  Cognition through a Literary Lens

 

toward the behavior’s intrinsic meaning, including the subjective experience in

(and of) the world that behavior expresses and embodies. (p. 4)

However, when animal behaviour is mediated through technical terms,

such as Lorenz’s ‘specialized escape-​ and-​
defense reaction’, ‘innate
releasing mechanism’, and ‘stimulus-​emitting object’, understanding
of animal behaviour becomes equivalent to something like the physi-
cist’s concept of gravity, which rather than constituting the behaviour
of a falling object belongs to a particular explanatory scheme designed
to account for that behaviour (p.  4). In parallel with Noë’s argument
about the two ways of orientating to dogs, Crist’s key point here is that
whereas use of the vernacular language of action casts animals as acting
subjects, use of the technical language of, for example, classical ethol-
ogy casts animals as natural objects. These two modes of discourse
about animals, the one bringing ‘humans’ and nonhumans’ phenom-
enal worlds into alignment’ and the other ‘alien to any possible expe-
rience or perspective of animals’ (p.  3), thus motivate very different
kinds of ascriptive practices.
In the present subsection, I  have drawn on the work of analysts in
fields ranging from the philosophy of mind and phenomenology to the
sociology of science in order to question the premises on which the
Nagel‒Dennett debate has been set up. In turn, in disputing the premises
of this debate, I  have argued for the need to replace a binarized model
of animal minds—​transparent accessibility versus radical otherness—​
with a scalar or gradient model involving different degrees of projected
relatedness, mutuality, and rapport across species lines. From this
perspective, heterophenomenology is no monolithic affair; it encom-
passes, rather, a complex ecology of mind-​ascribing practices, of which
Nagel’s and Dennett’s accounts capture only specific, localized sectors.
Leveraging ideas from narratology, my next subsection continues the
process of mapping out the ecology in question, even as it highlights the
transdisciplinary nature of this mapping project.

From Heterophenomenology
to Narratology (and Back Again)

In narrative contexts, mental-​state ascriptions entail forms of embed-

ded world-​building. Narrators ascribe subjective experiences to charac­
ters by portraying them as perceiving, remembering, imagin-
ing, or explicitly recounting (as intradiegetic narrators) further

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 201 ]
 

Figure 10.1  A continuum of modes of speech and thought presentation (based on Leech

and Short, 2007; Toolan, 2001)

worlds-​within-​the-​storyworld. To account for the discourse environments

in which such nested worlds take shape, analysts have identified an array of
positions or increments along a continuum or scale; this scale stretches from
characters’ verbal or mental performances that are more overtly mediated
by a narrator to performances that are less overtly mediated in this sense.
One version of the scalar model is shown in Figure 10.1, where ‘discourse’
serves as a cover term for presentations of speech as well as thought.
Significantly, research in narratology and stylistics has not yet fully
investigated issues that come into play when this model is brought to
bear on methods used to present the experiences of non-​human animals
in stories.3 Extending the model across species lines highlights the need to
rethink its conceptual underpinnings, in a way that bears directly on the
questions about heterophenomenology broached in my previous subsec-
tion. At issue is the range of techniques, from among those registered in
Figure 10.1, that are deemed available (or appropriate) for presentations of
animal subjectivity—​within as well as across the fiction‒non-​fiction divide.
Compare passages (1)  and (2); the first excerpt is taken from William
Horwood’s 1982 novel The Stonor Eagles (1982), and the second from Esther
Woolfson’s 2008 memoir Corvus, about her and her family’s experiences
while living with several birds. Despite their shared focus on birds, the
two passages are marked by different modes and degrees of mental-​state

3.  See Herman (2016) for a cognate discussion of narratological and stylistic research
on speech and thought presentation vis-​à-​vis questions of ‘mind style’ (Fowler, 1977)
and related phenomena in animal narratives.

[ 202 ]  Cognition through a Literary Lens

 

attribution, in ways that might suggest the primacy of genre-​based differ-

ences in the norms governing ascriptive practices in narrative contexts.

(1) For days Cuillin [the last surviving white-​ tailed eagle in Britain] had
remained at a stance on the high cliffs at the easternmost part of the Shetland
Islands, facing the dark sea. She had groomed, she had stared, she had hunted;
now another dawn had come, and she knew there could be no further delay or
excuse. … How vast and grey the sea looked, how treacherous its swells and
dark places, how fearful the day! … She flew at 350 feet, and within an hour the
coast-​bound fulmar were behind her and she was alone over the sea. It stretched
ahead, frighteningly vast, and she could only close her mind to what lay behind
her, and commit herself to its care. … when the first bout of real tiredness hit
her … she found her altitude sinking down to less than 200 feet. … A spar of
driftwood. A dead cormorant … too far out!
Horwood, 1982, pp. 45‒46

(2) As we began to look at all corvids with new interest, we saw Chicken [a rook
rescued by Woolfson’s daughter] do as the corvids around us did. In time, we
could recognise the complex series of movements of body, wings and feathers
that told of mood and inclination. … We began to discern her state of mind from
her stance, her walk, her feathers, to know that, when going about her day-​to-​day
business, untroubled and busy, the head feathers would be smoothed to her skull,
her auricular feathers (the panels of feathers by the sides of her head that cover
the openings that are her ears) flattened, with no ‘eyebrows’ or ‘ears’ visible—​
the raised head feathers that indicate alternations of mood—​no raised, irritated
crown of Dennis the Menace feathers around the top of her head, a posture that
indicates surprise, alarm, anger. Annoyance or some other stimulus, we saw,
could bring this about instantly; when teased, or crossed in any way, she’d fluff
her feathers, lower her head, adopt an aggressive stance, her leg feathers bagged
out and full.
Woolfson, 2008, pp. 72‒73

Passage (1), participating in the genre of historical fiction, features a wide

range of ascriptive techniques, spread out across the continuum shown
in Figure 10.1.4 The first part of the passage morphs from scene-​setting

4.  Horwood’s text can be categorized as historical fiction because it draws on the
resources of fiction to explore the significance, for the last-​remaining member of the
species, of the historically documented extinction in 1918 of the white-​tailed eagle
in Britain, prior to its reintroduction in 1975 (see Royal Society for the Protection of
Birds, 2014).

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 203 ]
 

narration into narrative report of thought act (‘she knew there could be no
further delay or excuse’), as Cuillin prepares to make her long, difficult flight
from Scotland to Norway. The exclamation mark that concludes the third
sentence flags this material as free indirect thought; modelling the frightful
aspect of the sea as it appears to the eagle, the succession of ‘how’ clauses
(‘How vast and grey’, etc.) evoke the animal’s landscape of consciousness
(Bruner, 1990), even as they simultaneously project the landscape of action
through which the bird must make its way. After a resumption of scene-​
setting narration (‘She flew at 350 feet …’), the use of the adverbial phrase
frighteningly vast reintroduces the eagle’s emotional state and frames the
following further narrative reports of thought acts by the bird. Then, in the
concluding lines of the excerpt, Horwood uses the ascriptive method posi-
tioned rightmost on the scale in Figure 10.1: the italicized phrase too far out
with its attendant exclamation mark can be glossed as an instance of free
direct discourse. Here the narrative creates the sense that Cuillin’s surprise
at the cormorant’s atypical location, instead of merely colouring the narra-
tor’s presentation of events as it would in free indirect thought, manifests
itself directly on the page, maximally free of narratorial mediation.
By contrast, passage (2), from Woolfson’s non-​fictional account, lim-
its itself mainly to recounting the ‘complex series of movements of body,
wings and feathers that’—​as Woolfson and her daughters learn to infer—​
‘told of [Chicken’s] mood and inclination’. Here the narrative focuses not
so much on the rook’s subjectivity per se as on the process by which her
human observers construct inferential pathways leading from the bird’s
bodily performances to hypotheses about her mental states and disposi-
tions. The passage recounts how Woolfson and her family initially used
a comparison set of corvids to identify salient behavioural patterns in
Chicken’s comportment and then derived, on the basis of repeated obser-
vations, translations of those patterns into subjective states—​and vice
versa. Thus, rather than projecting Chicken as experiencing equanimity,
surprise, anger, annoyance, and so on, Woolfson’s account centres on the
process of familiarization through which such projections may become
possible over time. Passage (2) therefore remains positioned at or near the
leftmost end of the scale in Figure 10.1; it recounts how inferences concern-
ing avian thought acts or emotional states might be arrived at, rather than
directly ascribing those subjective states via techniques situated further to
the right on the scale.
Do passages (1)  and (2)  therefore support that argument that
genre—​the categorization of a text as fictional or non-​fictional—​is the
main determinant of the relative richness and detail of mental-​state
ascriptions to animals in narratives? Here I seek to push back against

[ 204 ]  Cognition through a Literary Lens

 

what can be construed as a pervasive assumption about animal narra-

tives: namely, that only fictional accounts of animals support prolific,
particularized ascriptions of mental experiences to non-​human agents.
As mentioned previously, this dichotomization of fictional and non-​
fictional animal minds follows as a corollary from Nagelian assumptions
about the radical opacity of animal minds. Yet the variety of ascriptive
practices at work within as well as across the fiction‒non-​fiction divide
belies any such dichotomy. Studying this variety, in turn, forms part of
the project of redefining heterophenomenology as an ecology of mind-​
ascribing practices, rather than a series of one-​off, decontextualized
acts of ascription.
Along these lines, consider passages (3)  and (4)  against the backdrop
afforded by passages (1) and (2), respectively. Passage (3) is excerpted from
Daphne du Maurier’s 1952 novella ‘The Birds’, Alfred Hitchcock’s film adap-
tation of which appeared in 1963. Passage (4) derives from The Peregrine,
J.  A. Baker’s 1967 non-​fictional account of the ten-​year period he spent
pursuing and observing peregrine falcons, condensed into a diary format
covering one year.

(3) Nat [du Maurier’s protagonist] hurried on. … As he jumped the stile [lead-
ing to his family’s cottage], he heard the whir of wings. A black-​backed gull dived
down at him from the sky. It missed, swerved in flight, and rose to dive again. In
a moment it was joined by others, six, seven, a dozen, black-​backed and herring
mixed. … Covering his head with his arms he ran towards the cottage. They kept
coming at him from the air, silent save for the beating wings, the terrible flutter-
ing wings. He could feel the blood on his hands, his wrists, his neck. Each stab
of a swooping beak tore his flesh. If only he could keep them from his eyes. …
They had not learnt yet how to cling to a shoulder, how to rip clothing, how to
dive in mass upon the head, upon the body. But with each dive, with each attack,
they became bolder.
du Maurier, 1952/​2004, pp. 19‒20

(4) He [the male peregrine falcon] flew fast, banking narrow turns, winding in
steep spirals, wings lashing and quivering. Soon he was high above me. He could
see the hills sinking down into the shadowed valleys and the far woods rising all
around, the towns and villages still in sunlight, the broad estuary flowing into
blue, the grey dimness of the sea. All that was hidden from me was shining clear
to his encircling eye. … He was desperate with the rage of the hungry hawk. …
Searing through the sky, the hawk in torment saw the land beneath him work
and seethe with birds and come alive. Golden plover broke their wild cries

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 205 ]
 

along the green surface of the lower air. The peregrine hissed among them like a
burning brand.
Baker, 1967, p. 150

Here, relative to passages (1)  and (2), the ascriptive patterning is

reversed:  the excerpt from du Maurier’s fictional text limits itself to
reportage-​like narration of the birds’ atypical swarming and attacking
behaviour, refraining from imputing to the birds subjective states that
might account for their actions, whereas the passage from Baker’s non-​
fictional text engages in relatively prolific, wide-​ranging ascriptions of sub-
jective experiences to the male falcon he has been observing—​with Baker
exploiting techniques further to the right on the scale in Figure 10.1 than
the ones used by du Maurier. Indeed, part of the disturbing effect created
by passage (3) (and by du Maurier’s text as a whole) stems from the way
it refrains from building any coherent subjective profile for the massing,
hyper-​aggressive birds. Thus, over the course of the novella, restlessness
caused by the change of seasons (p.  2), unusually cold weather originat-
ing from the Arctic Circle (p. 8), intense hunger (p. 12), poisoning by the
Russians (p. 19), and a collective ‘urge for battle’ (p. 5) all feature as poten-
tial explanations for the birds’ attacks. The heterogeneity of these accounts
creates the effect of a desperate attempt to identify reasons for the ani-
mals’ actions, which in consequence take on, more and more, the charac-
ter of brute events. In turn, the indiscernibility of the birds’ motives helps
anchor the narrative in the subgenre of horror fiction.
Reticence concerning animal subjectivity can also be found in parts
of Baker’s The Peregrine—​for example, in material drawing on studies of
falcon physiology, behavioural routines, and geographical distribution
(Baker, 1967, pp. 21‒22, 34, 116).5 But passage (4) parallels the trajectory
followed by Horwood in passage (1):  scene-​setting narration gives way
to ascriptions of perceptual and emotional acts performed by the bird.
Thus, based on the falcon’s elevated position in the sky, Baker ascribes
to the bird perceptions of particular features of the environment hidden

5.  Passages such as the following mirror Woolfson’s method in passage (2), where
the emphasis is less on the bird’s subjective experiences than on how bodily structures
and performances can be interpreted to generate hypotheses about what it might be
like to be a peregrine: ‘The whole retina of a hawk’s eye records a resolution of distant
objects that is twice as acute as that of the human retina. Where the lateral and bin-
ocular visions focus, there are deep-​pitted foveal areas; their numerous cells record a
resolution eight times as great as ours. This means that a hawk, endlessly scanning the
landscape with a small abrupt turn of his head, will pick up any point of movement; by
focussing upon it he can immediately make it flare up into larger, clearer view’ (Baker,
1967, p. 35).

[ 206 ]  Cognition through a Literary Lens

 

from him. Extrapolating, Baker imputes to the falcon a general acuity of

visual perception that, embodied in the birds ‘encircling eye’, has been
denied to humans such as him. After attributing emotional states to the
bird—​desperation, rage, and torment—​the passage concludes with a pro-
jection of a nested structure of perceptual acts, with Baker perceiving the
falcon perceiving the plovers perceiving the predator bird. In this con-
text, the allusion to the biblical figure of the burning brand foregrounds
the falcon as the point of triangulation linking human and non-​human
subjectivities.
To underscore my larger claim here:  excerpts (1)–​(4) reveal the co-​
presence of different norms for mental-​state attribution within as well as
across the fiction‒non-​fiction divide. Fictional accounts can be more or
less prolific in their ascriptions of mental experiences to animals, as can
non-​fictional accounts—​with the result that some non-​fictional narra-
tives about animals may make more detailed and abundant attributions
of mind than fictional accounts. In addition, norms for mental-​state
ascription can vary in one and the same narrative, as suggested by the
variable degrees of detail attaching to Baker’s projections of falcon sub-
jectivity over the course of his account. In short, the case studies I have
considered point again to a complex ecology of ascriptive practices, gov-
erned by norms that cut across the fiction‒non-​fiction distinction. To
account more fully for the variability and plurality of the relevant norms,
I turn now to a fuller discussion of the concept of discourse domains, or
arenas of practice that are governed by more or less distinctive interpre-
tive paradigms and protocols for behaviour.

ANIMALS MINDS ACROSS DISCOURSE DOMAINS

In this section, I move from a discussion of differences in the treatment

of subjectivity in animal narratives to a sketch of the discourse domains
in terms of which these differences can be explained. In a first charac-
terization, discourse domains can be described along the lines of what
Wittgenstein (2009) called ‘language games’ and Levinson (1979) labelled
‘activity types’:  they are frameworks for conduct that organize partici-
pants’ verbal as well as nonverbal comportment around recognized kinds
or modes of activity, which are grounded in more or less fully shared sets
of norms, purposes, and goals. Relevant activities include engaging in pal-
aeontological research, debating the status of animal minds, or going on a
walk with a dog—​in short, activities that involve interacting with one or
more human or non-​human others in a particular setting and for specific

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 207 ]
 

kinds of reasons. Clearly, different sorts of ascriptive practices will be

deemed appropriate and warranted across these different domains: there
is a marked contrast between attributing specific intentional and volitional
states to a companion animal in the context of a familiar play ritual, on
the hand, and ascribing particular subjective experiences to now-​extinct
animals based on the fossil record, on the other.
This last example allows me to home in on the concept of discourse
domains using other descriptive terms. Discourse domains codify or at
least organize more or less distinctive sets of assumptions concerning
what sorts of experiential worlds are available to the various kinds of
beings taken to populate the world.6 Such domains correspond, in other
words, to sectors within a larger ecology of mind-​ascribing practices,
with each sector being distinguished by its own constellation of ascrip-
tive norms. In turn, these norms bear, in a top-​down manner, on the
strategies used to present—​and interpret—​species of mind in narrative
contexts, whether fictional or non-​fictional. For example, in a discourse
domain marked by an emphasis on the biophysical bases for human and
non-​human behaviour, ascriptions of subjectivity will remain severely
curtailed, within as well as across the species boundary and in both fic-
tional and non-​fictional accounts. Hence the ready traffic between behav-
iourist paradigms in psychology and foundational work in ethology in the
mid-​20th century. Hence, too, the way both fictional and non-​fictional
narratives can make use of the technique that Genette (1980) originally
termed external focalization. In this mode, exemplified in texts ranging
from Evelyn Waugh’s Vile Bodies and Ernest Hemingway’s ‘The Killers’ to
bedside shift reports by nurses and sailors’ log books, the narrative dis-
course is largely stripped of explicit references to agents’ subjective states
and experiences.
If domains of this sort profile human and non-​human behaviour as rela-
tively event-​like, and thus as subject to language games centring on concepts
such as ‘cause, law, fact, explanation’, other domains profile behaviour in
terms of actions more than events, and hence as subject to language games
centring on ‘projects, intentions, motives, reasons for acting, agents, and
so forth’ (Ricoeur, 1991, pp. 132‒33; for a fuller discussion of this con-
trast vis-​à-​vis animal narratives, see also Herman, in press). Compared
with discourse domains foregrounding what might be called the register

6.  I write ‘more or less distinctive sets of assumptions’ here to leave room for possi-
bilities (discussed later) for overlapping, interacting, and emergent domains, and also
for domains governed by competing, sometimes contradictory norms for mental-​state
attribution.

[ 208 ]  Cognition through a Literary Lens

 

of events, domains foregrounding the register of actions—​domains that

range from courtship practices to psychological profiling in forensic work
concerned with violent crime—​are marked by more prolific, far-​reach-
ing, and detailed ascriptions of mental states. As exemplified by accounts
of human‒dog interactions in police canine units or search-​and-​rescue
teams, these sorts of domains also cross species lines, and they likewise
encompass different narrative genres, non-​fictional as well as fictional;
hence, the similarly abundant and particularized heterophenomenologies
of texts that are otherwise miles apart, such as Jane Austen’s novels, con-
temporary romance fiction, textbooks on forensic psychology, and manu-
als on the training of rescue dogs.
Figure 10.2 presents a visualization of the idea of discourse domains
vis-​à-​vis ascriptions of subjective experiences to others—​with particu-
lar reference to ascriptions that cross the species boundary. Each dot (or
data point) in the background grid constitutes a mind-​ascribing act, with
such acts being organized into domains governed by more or less distinc-
tive norms. The size of the circles corresponds to the relative salience or
pervasiveness of discourse domains in a given culture, allowing for cross-​
cultural comparisons as well as tracking of the diachronic development and
transformation of domains within a particular culture. Further, the norms
associated with each domain bear in a top-​down fashion on all the repre-
sentational practices that fall within its purview, including non-​narrative
as well as narrative modes, narratives that feature animals as well as nar-
ratives that do not, and, for the narratives featuring animals, non-​fictional
as well as fictional accounts.
In line with my foregoing remarks, Figure 10.2 suggests that discourse
domain trumps genre when it comes to modelling animal minds in nar-
rative contexts, meaning that in a given instance a non-​fictional account
may feature more prolific and more fine-​grained ascriptions of subjec-
tive experiences to animals than would a fictional account—​depending
on the domain in which the narrative is anchored. This hypothesis
is borne out by passages (5)  and (6), both of which centre on human‒
canine interactions. Passage (5)  is taken from Eric Knight’s 1940 novel
Lassie Come-​Home (Knight, 1981); passage (6)  is excerpted from Luis
Carlos Montalván’s 2011 memoir Until Tuesday. A veteran of the Iraq war
with physical disabilities as well as PTSD, Montalván was paired with
Tuesday, a golden retriever, by the ECAD Service Dogs initiative—​with
ECAD standing for East Coast Assistance Dogs. Passage (6) centres on the
impressions about Tuesday that Montalván formed, and vice versa, when
they first meet. (I discuss below the annotation system used to mark up
these two excerpts.)

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 209 ]
 

A culture’s ascriptive practices Narratives

All representational
modes in domain

Discourse domains organizing

these practices (size indicates
relative pervasiveness within the
culture)

Nonfictional
genres
Fictional
genres

Narratives featuring
nonhuman animals

Figure 10.2  Discourse domains and mind-​ascribing practices

(5) That first night Lassie travelled steadily. Never before in her five years of
life had she been out alone at night. So there was no training to help her, only
instinct.
But the instinct within her was keen and alert. Steadily she followed
a path over the heather-​clad land. The path filled her with a warm satis-
faction, for it was going south. She trotted along it confidently and
surely.
At last she reached a rise and then, in a hollow below, she saw the dim shapes
of farm buildings. She halted, abruptly, with her ears thrown forward and
her nose trembling. Her magnificently acute senses read the story of the habita-
tion below as clearly as a human being might read a book. … She started down
the slope warily. The smell of food was pleasant, and she had gone a long time
without eating.
Knight, 1940/​1981, p. 96

(6) When Tuesday [focused] on my face, I saw a sincerity in his dark brown eyes
I hadn’t suspected. This dog was handsome. He was intelligent. But he was also
deep and emotional and hurting at the core. …
We stared at each other for a few seconds, and I could tell Tuesday was check-
ing me out, assessing the situation. He wasn’t timid. And he wasn’t selfish.
Something about the softness in his eyes told me Tuesday craved a relation-
ship, but he was too smart to fawn just because somebody handed me his leash.
I didn’t know why he was wary. I didn’t know he was sensitive. And needy.

[ 210 ]  Cognition through a Literary Lens

 

And that he had lost so much confidence in himself, because of his multiple
abandonments, that I would have to slowly build back the intelligent, caring dog
I glimpsed in those pleading eyes.
Montalván, 2011, pp. 129‒130

Because of the intricate interinvolvement of humans’ and canids’ evo-

lutionary histories, in general the threshold for permissible ascriptions
of mental states to dogs, in fictional as well as non-​f ictional discourse,
is higher than it is for ascriptions in accounts of many other sorts
of human‒animal interactions. But what is noteworthy here is that
Montalván’s non-​fictional narrative, grounded in assumptions about
the strong rapport and mutual regard of humans and dogs also evident
Knight’s text, projects a richer experiential world than does the fic-
tional example. To facilitate a more precise comparison of the degree
of detail associated with the texts’ mind-​ascribing practices, I  have
marked up the two excerpts using the annotation system for which
Figure 10.3 provides a key. The system is ­designed to measure degrees
of what might be called ‘heterophenomenological density’—​that is,
the frequency and scope of ascriptions of subjective experiences to ani-
mal others—​across narrative genres as well as the discourse domains
with which those genres intersect. In essence, the more marked-​up a
text that engages with non-​human beings, the denser or more prolific
the ascriptions of mental states and experiences to the animal agents
involved.
As presented, excerpts (5)  and (6)  are almost exactly the same
length:  142 and 140 words, respectively. The passages thus allow for
an indicative comparison of the frequency and range of mental-​state
ascriptions across an equivalent span of text in the two narratives. In
turn, the annotations suggest that the raw number of mental-​state
ascriptions, especially direct ascriptions of subjective states, is greater
in Montalván’s non-​fictional account of Tuesday than in Knight’s novel.
Excerpt (5)  does feature several direct references to Lassie’s mental
states, as well as locutions that imply the dog’s perceptual activity and
also intentional actions on Lassie’s part. Further, Knight uses the term
south to suggest how Lassie orients to this cardinal direction, again imply-
ing goal-​driven behaviour. Per line of text, however, excerpt (6) is more
marked up, with the overall number of direct references to Tuesday’s
mental attributes and dispositions, as well as his emotional states, being
particularly striking. Thus the passage ascribes to the dog, without any
hedging or qualification, dispositions and states that include sincerity,

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 211 ]
 

Figure 10.3  Key to the annotation system used for excerpts (5) and (6).

intelligence, emotional hurt, timidity, lack of selfishness, craving (for a

relationship), wariness, neediness, lack of confidence, and caringness.
Intermixed with these ascriptions are locutions suggesting Tuesday’s
perceptual activity, references to arrangements and situations to which
the dog orientates intentionally (‘relationships’, ‘multiple abandon-
ments’), and a translation of the dog’s manner of looking at Montalván
himself into a volitional state (‘pleading’). The net result of the greater
density of mental-​state ascriptions in (6) as compared with (5) is a text
that builds a richer profile of non-​human subjectivity than the profile
that emerges from Knight’s novel.
To account for what might seem like counter-​intuitive patterning here,
whereby the non-​ fictional narrative makes more prolific mental-​ state
ascriptions than the fictional text, the idea of discourse domains can be
brought to bear, with the annotation system being designed to capture
how contrasting norms organize the domains to which excerpts (5)  and
(6)  are anchored. The system confirms that the non-​fictional account of
Montalván’s first encounter with Tuesday is more dense with mental lan-
guage, or assumes greater licence in projecting animal subjectivity, than
Knight’s fictional account of Lassie’s experiences during her epic journey
back from Scotland to the Carraclough family in Yorkshire. Indeed, it can
be argued that in Lassie Come-​Home Knight followed the naturalist John
Burroughs (1903) in dichotomizing instinct and reason along species lines,
possibly in an effort to widen his audience by avoiding anthropomorphism
of a sort that some readers may have associated with animal stories for
children. As Knight repeatedly indicates, Lassie only has instinct, and lacks
the ability to reason about the situations and events she experiences (con-
trast London, 1909). Given this set-​up, there is considerably less scope
for detailed modelling of animal subjectivity in Lassie Come-​Home than in
Montalván’s non-​fictional account, where he articulates his understanding

[ 212 ]  Cognition through a Literary Lens

 

of how Tuesday and he orientate to one another within their shared world
of encounter.

CONCLUDING REMARKS

In this chapter, I  have pursued a transdisciplinary triangulation of ideas

from philosophy, narratology, and other fields around questions concern-
ing animal minds. By focusing on the variability of mind-​ascribing practices
in non-​fictional as well as fictional narratives, and arguing that discourse
domain rather than genre determines the prolificness and degree of detail
attaching to projections of animal subjectivity in narrative contexts, I have
sought to reframe debates based on a polarity between legible and illegible
animal minds. In the account outlined here, maximal as well as minimal
projections of non-​human subjectivity must be situated within the larger
ecology of ascriptive practices to which animal narratives afford access.
Reciprocally, debates about the scope and limits of heterophenomenology
provide new contexts for investigating methods of thought presentation
used in narrative, particularly when those methods are extended across the
species boundary.
More broadly, my analysis highlights the need for further study of an
underexplored aspect of the mind‒narrative nexus: namely, how textual
patterns associated with the presentation of animal experiences in narra-
tives are interwoven with cultures’ ontologies, in the sense of that term as
it is used in contemporary anthropological research (Viveiros de Castro,
1998; Descola, 2013; Kohn, 2013; see also Herman, 2014). Such ontologies
specify, in the form of common knowledge, what sorts of beings populate
the world and how those beings’ attributes relate to the attributes imputed
to humans. Discourse domains, as I have described them here, both are
grounded in and also help constitute these ontologies, which entail more
or less parsimonious or prolific allocations of possibilities for subjectivity
beyond the realm of the human. From this perspective, a cultural ontology
can be described as a constellation of discrete as well as overlapping dis-
course domains, in which animal behaviours become normatively profiled
(e.g. via storytelling practices) as relatively action-​like or event-​like, and
hence as more or less appropriately targeted for mental-​state attributions.
Contrast a bare account of an animal’s trajectory of movement in space
with a narrative about the reasons for acting that motivate the animal in
question. But this way of putting the matter suggests why Figure 10.2 tells
only part of the story. If discourse domains shape patterns of mental-​state

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 213 ]
 

attribution in narratives, the patterns of ascription used in individual sto-

ries can reciprocally impinge on discourse domains, and potentially reca-
librate normative assumptions about species of minds—​for instance, by
promoting a shift from the register of events to the register of actions to
account for humans’ relationships with particular (kinds of) animals. Thus
Gabriela Cowperthwaite’s 2013 documentary film Blackfish, which links the
deaths of several animal trainers to the treatment of killer whales kept in
captivity at marine mammal parks maintained by the SeaWorld corpora-
tion, has contributed to the call for legislation to free the whales, on the
grounds that current practices violate the US Constitution’s prohibition
of slavery. This example underscores the need to develop a multi-​scale,
and multi-​directional, approach to the issues under discussion—​one that
explores not only the top-​down normative effects flowing from cultural
ontologies to discourse domains to ascriptive acts found in particular texts,
but also the way storytelling practices can themselves ‘reset’ default norms
for understanding animals and human‒animal relationships, incremen-
tally reshaping cultural ontologies in the process (Herman, in press).
Even more broadly, my analysis raises a number of the wider-​scope
questions that will need to be addressed in future work:  What forms of
relatedness are made possible by ontologies in which an expanded com-
munity of selves extends beyond the species boundary? And how are these
transhuman networks of affiliation figured in fictional texts, non-​fictional
discourse on animals, the storyworlds of cinema, narratives for children,
and other storytelling modes? How do the attested characteristics of par-
ticular species, and the relative (in)frequency of humans’ interactions with
the members of those species, affect allocations of possibilities for trans-
human subjectivity in narrative contexts? To what extent can existing
paradigms for narratological analysis capture forms of cross-​species rela-
tionality, as they manifest themselves in the structures of narrative dis-
course, and to what extent will new, transdisciplinary modes of inquiry be
required to develop what might be characterized as a narratology beyond
the human? How, in turn, might the concepts and methods that emerge
from such a narratology bear on ways of understanding humans’ place in a
more-​than-​human world?

REFERENCES

Baker, J. A. (1967). The peregrine. London: Collins.

Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard University Press.

[ 214 ]  Cognition through a Literary Lens

 

Burroughs, J. (1903). Real and sham natural history. Atlantic Monthly, 91(545),
298–​310.
Crist, E. (1999). Images of animals: Anthropomorphism and animal mind. Philadelphia,
PA: Temple University Press.
Dennett, D. C. (1991). Consciousness explained. Boston, MA: Little, Brown.
Descola, P. (2013). The ecology of others (G. Godbout and B. P. Luley, Trans.).
Chicago: Prickly Paradigm.
du Maurier, D. (2004). The birds. In D. du Maurier, The birds and other stories (pp. 1–​39).
London: Virago.
Fowler, R. (1977). Linguistics and the novel. London: Methuen.
Gallagher, S., and Hutto, D. D. (2008). Understanding others through primary
interaction and narrative practice. In J. Zlatev, T. P. Racine, C. Sinha, and E.
Itkonen (Eds.), The shared mind:  Perspectives on intersubjectivity (pp. 18–​38).
Amsterdam: John Benjamins.
Genette, G. (1980). Narrative discourse: An essay on method (J. E. Lewin, Trans.). Ithaca,
NY: Cornell University Press.
Herman, D. (2013). Storytelling and the sciences of mind. Cambridge, MA: MIT Press.
Herman, D. (2014). Narratology beyond the human. DIEGESIS:  Interdisciplinary
E-​Journal for Narrative Research, 3(2), 131–​143.
Herman, D. (2016). Building more-​than-​human worlds: Umwelt modelling in animal
narratives. In J. Gavins and E. Lahey (Eds.), World building: Discourse in the mind
(pp. 53–70). London: Bloomsbury.
Herman, D. (in press). Hermeneutics beyond the species boundary: Explanation and
understanding in animal narratives. Storyworlds.
James, S. P. (2009). Phenomenology and the problem of animal minds. Environmental
Values, 18(1), 33–​49.
Kagan, J. (2009). The three cultures: Natural sciences, social sciences, and the humanities in
the twenty-​first century. Cambridge, MA: Harvard University Press.
Knight, E. (1981). Lassie come-​home. London: Puffin.
Kohn, E. (2013). How forests think: Toward an anthropology beyond the human. Berkeley,
CA: University of California Press.
Leech, G., and Short, M. (2007). Style in fiction: A linguistic introduction to English fic-
tional prose (2nd ed.). London: Longman.
Levinson, S. C. (1979). Activity types and language. In P. Drew and J. Heritage (Eds.), Talk
at work: Interaction in institutional settings (pp. 66–​100). Cambridge: Cambridge
University Press.
London, J. (1909). The other animals. In J. London, Revolution and other essays (pp.
235–​ 267). New  York:  Macmillan Company. Retrieved from http://​london.
sonoma.edu/​Writings/​Revolution/​animals.html
Montalván, L. C., with Witter, B. (2011). Until Tuesday: A wounded warrior and the dog
who saved him. London: Headline Review.
Nagel, T. (1974). What is it like to be a bat? Philosophical Review, 83(4), 435–​450.
Nagel, T. (1986). The view from nowhere. Oxford: Oxford University Press.
Noë, A. (2009). Out of our heads: Why you are not your brain, and other lessons from the
biology of consciousness. New York: Hill and Wang.
Prince, G. (2003). A dictionary of narratology (2nd ed.). Lincoln, NE:  University of
Nebraska Press.
Ricoeur, P. (1991). Explanation and understanding. In P. Ricoeur, From text to
action:  Essays in hermeneutics, II (K. Blamey and J. B. Thompson, Trans.) (pp.
125–​143). Evanston, IL: Northwestern University Press.

A n i m a l Mi n d s a c r o ss Dis c o u r s e D o m a i n s  [ 215 ]
 

Royal Society for the Protection of Birds. (2014). White-​tailed eagle:  Population
trends. Retrieved from http://​www.rspb.org.uk/​discoverandenjoynature/​dis-
coverandlearn/​birdguide/​name/​w/​whitetailedeagle/​population.aspx
Shapiro, K. J. (1997). A phenomenological approach to the study of nonhuman animals.
In R. W. Mitchell, N. S. Thompson, and H. L. Miles (Eds.), Anthropomorphism,
anecdotes, and animals (pp. 277–​ 295). Albany, NY:  State University of
New York Press.
Snow, C. P. (1998). The two cultures. Cambridge: Cambridge University Press.
Thompson, E. (2007). Mind in life:  Biology, phenomenology, and the sciences of mind.
Cambridge, MA: Harvard University Press.
Toolan, M. (2001). Narrative:  A  critical linguistic introduction (2nd ed.). London:
Routledge.
Trevarthen, C. (1993). The self born in intersubjectivity: The psychology of an infant
communicating. In U. Neisser (Ed.), The perceived self: Ecological and interpersonal
sources of self-​knowledge (pp. 121–​173). Cambridge: Cambridge University Press.
Uexküll, J. von (2010). A foray into the worlds of animals and humans (J.D. O’Neil, Trans.).
Minneapolis, MN: University of Minnesota Press.
Viveiros de Castro, E. (1998). Cosmological deixis and Amerindian perspectivism.
Journal of the Royal Anthropological Institute, 4(3), 469–​488.
Wittgenstein, L. (2009). Philosophical investigations (G. E.  M. Anscombe, P. M.  S.
Hacker, and J. Schulte, Trans.). Chichester, UK: Wiley-​Blackwell.
Woolfson, E. (2008). Corvus: A life with birds. London: Granta.

[ 216 ]  Cognition through a Literary Lens

 

PART III

Literature and Cognition

in Cognitive Science
 
 

CHAPTER 11

Embodied Dynamics
in Literary Experience
R AYMOND W. GIBBS JR .

INTRODUCTION

One of the enduring wonders of literature is that it can touch us in deeply

personal ways which are also evocative of larger shared, symbolic mean-
ings. Let me tell you about one memorable instance when a small segment
of fiction reached out and grabbed my heart. I was reading the novel titled
The Anthologist by Nicolson Baker (2009), which is about a poet who is edit-
ing a large volume of recent American poetry for a publisher. Although the
volume is complete, the poet/​editor still has to write the introductory essay
about the book’s contents and offer some historical context for the poems.
Unfortunately, though, he is simply unable to write the essay because he
suffers from a terrible case of ‘writer’s block’, some of which is due to career
and marital problems he has recently experienced. The novel is quite funny,
and full of insights, as well as gossip, about literature and the lives of poets.
Towards the end of the novel (pp. 196–​197), the poet/​editor summarizes
his dilemma in the following manner:

I wish I  could spill forth the wisdom of twenty years of reading and writing
poetry. But I am not sure I can. …
And now it’s like I’m on some infinitely tall ladder. You know the way that old
aluminum ladders have that texture, that kind of not-​too-​appealing roughness
of texture, and that kind of cold gray color? I’m clinging to this telescoping lad-
der that leads up into the blinding blue. The world is somewhere very far below.
 

I don’t know how I got here. It’s a mystery. When I look up I see people climbing
rung by rung. I see Jorie Graham. I see Billy Collins. I see Ted Kooser. They’re all
clinging to the ladder, too. And above them, I see Auden, Kunitz. Whoa, way up
there. Samuel Daniel, Sara Teasdale, Herrick. Tiny figures clambering, clinging.
The wind comes over, whssssew, and it’s cold, and the ladder vibrates, and I feel
very exposed and high up. Off to one side there’s Helen Vendler, in her trusty
dirigible, filming our ascent. And I look down and there are many people behind
me. They’re hurrying up to where I  am. They’re twenty-​three-​old energetic
climbing creatures in their anoraks and goggles, and I’m trying to keep climbing.
But my hands are cold and going numb. My arms are tired to tremblement. It’s
freezing, and it’s lonely, and there’s nobody to talk to. And what if I just let go?
What if I just loosened my grip, and fell to one side, and just—​fffshhhooooow.
Let go.
Would that be such a bad thing?

As my eyes darted along these lines of text, my mind and heart seem-
ingly exploded with ideas, memories, feelings, and even new understand-
ings, all related to my past, present, and future life. I recalled my time as a
house painter, many decades ago, when I knew all too well the sensations
of being high up on an old, grey, aluminium ladder, feeling the wind blow
around me, sensing the cold, being gripped by loneliness, and wondering
how I got there and what I should do both in that immediate present and
in my life overall. Yet I also knew something of the poets mentioned in this
passage, having read some of their poems. I  laughed aloud when seeing
that the famous American literary critic Helen Vendler, whose books I have
read, was sitting ‘in her trusty dirigible’ while ‘filming our ascent’. This brief
image perfectly captured Vendler’s role as a prominent observer, and ana-
lyst, of the ever-​changing world of poetry.
But I also immediately recognized that this passage spoke to me about
my own present preoccupations as a busy academic. Too often I also feel
overwhelmed with the sheer amount of work I have committed myself to
doing:  articles and books to write, journals to edit, professional talks to
give, teaching, burdensome administrative tasks, and so forth. To be com-
pletely honest, there are moments when I am paralysed by all that I have to
do and feel as if I too were stuck alone on the path of my life’s journey. And
the thought sometimes occurs to me: what if I too simply ‘just loosened my
grip, and fell to one side, and just—​fffshhhooooow. Let go’? Indeed, ‘would
that be such a bad thing?’ My understanding of this possibility, an option
I both fear and welcome, was not an abstract realization, but something
that I, while in the immediate act of reading, imagined in an embodied
manner. I felt myself letting go of the ladder, again similar to ones I have

[ 220 ]  Literature and Cognition in Cognitive Science

 

climbed numerous times before in life, and sensed my body falling, the sud-
den loss of control, the fear of what would happen next, but also a welcom-
ing relief as work and life pressures fell away from me. Why can’t I make
this happen for real?
In short, my brief literary experience reading this passage from Baker’s
novel was rich in personal, bodily imagination and afforded me a chance, as
good literature often does, to think more about how I want to live my life.
Of course, each of us will have a different reaction to the literature we read.
Not everyone has previously climbed high ladders and felt insecure about
one’s place at that moment and about life more generally. Many people
probably have not struggled, sometimes with pleasure, reading the works
of Auden and Kunitz, or understanding the complex analyses that Vendler
provides in many of her writings. Yet  almost all readers will likely come
away from reading the above passage with some sense of what it means,
and how it may, or may not, relate to enduring themes in their own lives.
My aim in this chapter is to advance a specific argument about the
embodied dynamics of literary experience. Specifically, I argue that read-
ing, of all sorts, involves our imaginative, embodied engagement with
texts, and that this engagement does not just constitute our reactions to
literature, but shapes the very process by which linguistic meanings are
interpreted. More broadly, my hope is to suggest a way out of the long-​
standing conflict that exists between science and literary criticism through
acknowledgement of both the stabilities and the variations in how people,
including critics, read and interpret literary texts.

SCIENCE AND LITERARY CRITICISM

The scientific study of literary reading has traditionally focused on naïve read-
ers’ first-​time pass through, and quick comprehension of, brief segments of
text, usually artificially constructed for experimental purposes. For the most
part, these studies try to capture something about what people ordinarily
do when encountering literary works, especially in regard to the fast-​acting
unconscious processes which give rise to more conscious meaning products.
Yet the empirical turn in literary studies over the last few decades has brought
forth different findings related to a diverse array of reading experiences, rang-
ing from speeded interpretations of smaller text passages to slower, more
reflective, analyses of fiction and poetry. For example, one model of emo-
tion in literary reading, based on research and theory from cognitive science,
describes how different stages in reading, ranging from pre-​reading to read-
ing, post-​reading, and even non-​reading, are influenced by various affective,

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 221 ]
 

cognitive, and embodied factors (Burke, 2011). This ‘literary reading loop’
model, as Michael Burke calls it, has been applied to characterizing people’s
experiences of ‘disportation’, such as felt tension, felt motion, and release,
at varying moments during emotional acts of reading, especially at literary
closure. The text that participants read in a key study for this research pro-
gramme was F. Scott Fitzgerald’s The Great Gatsby (1925/​2000).
Not surprisingly, many literary scholars voice strong reactions against
these developments. According to many scholars, studying naïve readers,
such as college students who participate in experimental studies, has lit-
tle bearing on the expertise, and even connoisseurship, brought to bear
when literary critics explicate what texts mean and aesthetically provoke.
After all, ordinary readers do not possess the tremendous knowledge and
experience that trained literary scholars bring to any interpretive proj-
ect. Literary scholars do not aim to find ‘normative’ or ‘correct’ or ‘typical’
interpretations of texts, but create unique readings that are informed by
emotional, aesthetic, historical, cultural, and political concerns. As is seen
in debates within the art world over the rise of behavioural and neuroscien-
tific studies of visual art works (Massey, 2009), literary scholars often pre-
sume that an interpretive gap exists between what ordinary readers do and
the beginnings of their own personal literary analyses. Literary critics are
human beings too, stuffed with the same cognitive structures possessed by
naïve readers. Still, literary scholars have special skills that make their own
readings special and divorced from what ‘ordinary’ people do when they
encounter literature.
Is it possible to reconcile the scientific study of reading with the practice
of literary criticism? My primary claim is that almost all people imaginatively
project themselves into texts as a fundamental part of any act of linguis-
tic understanding. At the same time, the precise nature of these ‘embodied
simulations’ may differ, sometimes in very subtle ways, depending on the
background and experiences of the reader. This perspective asserts that there
are important commonalities between recreational and critical understand-
ings of literature, but that the ultimate products of embodied simulation
processes can vary considerably given variations in the texts read, the people
doing the reading, and their interpretive goals and motivations.

THE SCIENTIFIC CASE FOR EMBODIED SIMULATION

IN LINGUISTIC UNDERSTANDING

The varieties of literary enjoyment are related to how people imagina-

tively project themselves into text worlds via embodied simulations. This

[ 222 ]  Literature and Cognition in Cognitive Science

 

hypothesis asserts that people ordinarily attempt to construct imaginative,

embodied re-​enactments of what the situation described by some specific
discourse must be like to participate in given their own bodily capacities
and experiences (Gibbs, 2006; Bergen, 2012; Gibbs and Colston, 2012). The
embodied simulation hypothesis has been widely debated within the cogni-
tive science community in recent years, in its possible applications to various
cognitive, affective, and linguistic phenomena, including conceptual repre-
sentations, memory, problem-​solving, learning, emotion, and conscious-
ness. One major focus of discussion is whether people can ever create an
embodied simulation for linguistic materials that are abstract or metaphori-
cal (Bergen, 2012; Gibbs and Colston, 2012). First, we will consider some of
the experimental tests of the embodied simulation hypothesis. I will later
extend these research findings to explain people’s experiences of literature.
Imagine first that you are a participant in the following psycholinguistic
experiment. You are seated in front of a computer terminal and shown the
sentence ‘The carpenter hammered the nail into the wall’. After reading
the sentence, you are shown a picture of an object, such as a nail or an
elephant, and asked to quickly judge whether that object was mentioned
in the sentence. Of course, you would quickly say ‘yes’ to the picture of a
nail and ‘no’ to that of the elephant. The primary interest, however, was
with your speeded response to the nail picture, depending on whether it
was shown in a horizontal or vertical orientation. Research indicates that
on average, people are faster to make their ‘yes’ decisions when the pic-
ture is in the same spatial orientation implied by the sentence just read
(Zwaan, Stanfield, and Yarley, 2002). Thus, people are faster to say ‘yes’
when the picture showed the nail in the horizontal orientation than when
it was shown upright, or in the vertical position. However, those who have
first read the sentence ‘The carpenter hammered the nail into the floor’ are
faster, on average, to say ‘yes’ to a nail picture that presents it in a vertical
position than to one in the horizontal orientation.
One interpretation of these findings is that people automatically con-
struct a mental image of an object in its appropriate spatial orientation
in light of what the sentence implies. Even if the nail’s position was not
explicitly noted in the sentence, our immediate understanding of the sen-
tence’s meaning enabled us to create an image of the situation in which the
nail was hammered in a horizontal or vertical position. Readers draw these
inferences not simply because of their abstract knowledge of the world,
but because they imagined themselves hammering a nail into a wall in a
horizontal direction.
Embodied simulations enable people to project themselves into the
minds and actions of others, including the objects and events referred to

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 223 ]
 

in texts. For example, studies on the ‘action-​compatibility effect’ indicate

that people are faster to make comprehension responses for sentences like
‘John opened the drawer’ when they have to move their hands towards
their bodies to push a comprehension button than when they have to
move their hands away from their bodies (Glenberg and Kaschak, 2002).
The reverse pattern of results was observed when people heard sentences
implying movement away from the body, such as ‘John closed the drawer’.
Once again, people interpret sentences by imagining themselves engaging
in the very actions specified in the language, which in turn directly influ-
ences their embodied comprehension task (e.g. moving their hand to push
a comprehension button).
Experimental studies also indicate that embodied simulations are con-
structed incrementally during speeded sentence comprehension. Consider
the following statement:

Before/​the/​big race/​the driver/​took out/​his key/​and/​started/​the/​car.

The implied direction of the key turn in this case is clockwise. In one exper-
iment, participants read through each sentence by rotating a knob after
each chunk of words, indicated by the slashes in the above example (Zwaan
and Taylor, 2006). Participants were instructed to turn the knob in either
a clockwise or a counter-​clockwise manner. The result of interest here was
that people were faster to read and comprehend the verb ‘started’ when
they made their knob turns in a clockwise direction than when making
counter-​clockwise rotations. People essentially understand the key verb
‘started’ by constructing an embodied simulation of the implied movement
the car driver had to perform in order to turn this key and start the engine.
This illustrates that people do not wait till the end of the sentence to ini-
tiate their simulation processes. Embodied simulation processes are not
optional, after-​the-​fact operations that emerge only after a sentence has
been read and understood, but are an immediate part of people’s moment-​
by-​moment processing of linguistic meaning.
Not surprisingly, people with different experiences and expertise may
vary in the ways in which they construct embodied simulations. For exam-
ple, languages like Arabic or Hebrew are written right to left; others, such as
English and Italian, are written left to right; while still others, such as tradi-
tional Chinese, are written top to bottom. When Italian or Arabic speakers
heard sentences like ‘The girl pushes the boy’, and then judged whether
a picture properly captured the event, these participants responded dif-
ferently (Maas and Russo, 2003). Arabic speakers, for instance, took less
time to judge a picture as correctly depicting the event when the girl was

[ 224 ]  Literature and Cognition in Cognitive Science

 

on the right side of the frame, while Italians took less time to make the
same judgement when the girl was shown on the left side of the picture.
Thus, people’s experiences with the spatial direction of written language
affect the embodied simulations created when they interpret simple action
sentences.
These different experimental results highlight the degree to which
people use their perceptual and motor systems for simulation purposes.
Embodied simulations emerge from complex interactions between brains,
bodies, and world. For example, part of the neural basis for embodied simu-
lations is seen in the research on ‘mirror neurons’. Studies have shown,
with both humans and non-​human primates, that motor areas of the brain
are activated when individuals see other actors performing different bodily
motions (Rizzolatti and Craighero, 2004). These findings imply that people
tacitly imagine themselves performing the actions they perceive, which
enables them to understand through simulations what other individu-
als are doing (for a critical discussion of this research, see Hickok, 2014).
Moreover, simply hearing action-​related language, such as the ‘kick’ in
‘kick the bucket’, activates relevant sensorimotor areas of the brain (i.e. the
somatosensory cortex area related to leg actions), once again as if listeners
were partially performing the action implied by the verb (Boulenger, Hauk,
and Pulvermüller, 2009).
One criticism of the embodied simulation hypothesis is that it may not
explain how people understand abstract or metaphorical language (Mahon
and Caramazza, 2008). Consider the simple statement ‘John couldn’t grasp
the concept of infinity’. A ‘concept’ is an abstract entity and it seems odd to
think that one could physically ‘grasp’ something that does not physically
exist. But scientific research has convincingly revealed that people perform
embodied simulations when interpreting verbal metaphors (Gibbs and
Colston, 2012). People’s understanding of abstract events, such as ‘grasp-
ing the concept’, is constrained by aspects of their embodied experience as
if they were immersed in the discourse situation, even when the described
situations can only be metaphorically realized.
For example, studies show that having people first make a hand move-
ment, such as reaching out to grasp something, subsequently facilitates the
speed with which they comprehend a metaphorical phrase such as ‘grasp the
concept’ (Wilson and Gibbs, 2007). Even if people are unable to physically
grasp a concept, engaging in relevant body actions primes the construction
of an embodied simulation to infer the metaphorical meaning of ‘grasp the
concept’. Thus, people interpret the word ‘concept’ as a metaphorical object
which, when grasped, can be examined and understood. When people were
asked to form mental images for metaphorical action phrases, for phrases

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 225 ]
 

like ‘tear apart the argument’ their reported imagery exhibited significant
embodied qualities of the actions referred to by these phrases (e.g. people
conceive of the ‘argument’ as a physical object that when torn apart no lon-
ger persists; Gibbs, Gould, and Andric, 2006). Having people perform, or
even imagine performing, a relevant bodily motion, such as tearing some-
thing apart, enhances the mental images constructed when understand-
ing metaphorical phrases. Most theories of verbal metaphor use would
never predict that people can experience these phrases in such detailed,
embodied ways.
One unique study examined whether hearing an embodied metaphor
influenced subsequent bodily behaviour (Gibbs, 2013). People stood and
looked at a tennis ball 40 feet away as they listened to a short story con-
taining the metaphorical statement ‘Your relationship was moving along
in a good direction’. One version of the story ended up with the relation-
ship continuing to be successful, but another told of the relationship dete-
riorating. After hearing one of these two stories, people were blindfolded
and told to walk out to the tennis ball, stopping when they imagined arriv-
ing at it. Analysis of people’s walking behaviours showed that they walked
significantly beyond the tennis ball when the context suggested a positive
relationship, but, on average, did not get to the tennis ball when hearing
about the unsuccessful relationship. This same difference in walking behav-
iours, however, was not obtained when people read the non-​metaphorical
statement ‘Your relationship was very important’ in the same two sce-
narios. None of the walking behaviours observed were due simply to peo-
ple’s mood after hearing about either the successful or the unsuccessful
relationship. It appears, then, that people understand the metaphorical
statement by building an embodied simulation relevant to the conceptual
metaphor LOVE RELATIONSHIPS ARE JOURNEYS, such that they bodily
imagine taking a longer journey with the successful relationship than with
the unsuccessful one. On the other hand, it is more difficult to create a
detailed embodied simulation for a non-​metaphorical statement such as
‘Your relationship was very important’.
Finally, neuroscientific work has also showed activation in the motor
system of participants’ brains when they read literal (e.g. ‘grasped the
stick’) or metaphorical (e.g. ‘grasped the idea’) statements (Desai, Binder,
Conant, Mano, and Seidenberg, 2011), which offers additional evidence
that embodied simulations may underlie our understanding of metaphori-
cal meanings.
These experimental findings are clearly contrary to traditional accounts
of metaphor understanding, which generally assume that the ultimate
aim is to transcend physical, body-​based meaning to arrive at abstract,

[ 226 ]  Literature and Cognition in Cognitive Science

 

non-​physical metaphorical messages (e.g. pragmatic theories such as ‘rel-

evance theory’, or most psycholinguistic approaches, such as the ‘career
of metaphor’ theory; for discussions of these and other approaches to
metaphor, see Gibbs and Colston, 2012). Metaphorical meaning appears
to be significantly grounded in embodied action and experience, as long
claimed in cognitive linguistics research (Lakoff and Johnson, 1999; Gibbs,
2006). Furthermore, people do not just access passively encoded concep-
tual metaphors from long-​term memory during online metaphor under-
standing. Instead, people may spontaneously create a particular construal
of metaphors via embodied simulation processes operating during think-
ing, speaking, and understanding.

EMBODIED SIMULATIONS IN UNDERSTANDING

ALLEGORICAL LITERATURE

The scientific studies showing the psychological reality of embodied simu-

lation processes may help explain how people interpret literature (Oatley,
2011). For example, Robert Frost’s famous poem ‘The Road Not Taken’
(1916/​1969) describes a man’s journey walking through the woods, and
ends with him considering the dilemma of which of two paths to follow.
The poem concludes with the following lines:

Two roads diverged in a wood, and I—​

I took the one less traveled by,
And that has made all the difference.

Frost uses concrete language to refer to broader symbolic themes regard-

ing the choices people make in life and the consequences that follow from
these decisions. Frost may have intended this work as an allegorical com-
ment on the difficulties one encounters along life’s many journeys (meta-
phorical and otherwise).
One possibility is that people make sense of Frost’s poem, and draw
relevant allegorical inferences, through embodied simulation processes.
Readers may actively imagine themselves physically walking through the
woods, which enables them to understand how their own journey expe-
riences have broader symbolic implications. This hypothesis was tested
by asking college students to read and write interpretations of Frost’s
poem (Gibbs and Boers, 2005). The poem was presented in three-​line
segments with each segment printed on a separate page. The partici-
pants were encouraged to write as much as they could about the poem’s

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 227 ]
 

meanings, including what the author was trying to communicate and

what they thought was the poem’s broader meaning. Finally, partici-
pants were asked to read the poem one more time, and write down what
they now thought the poem was about and what the poet was trying to
communicate.
Participants’ written interpretations for each three-​line segment, as
well as the poem’s overall interpretations, were analysed for their con-
tent. There are several points worth making about these data. First, the
participants provided extremely few personal associations to the dif-
ferent segments, which clearly suggests that people primarily focused
on the poem’s, and the poet’s, messages. This shows that interpreting
poetry, at least in this experimental setting, is not a matter of radical
deconstruction, but is significantly constrained by what readers assumed
may be the poet’s message. Second, although readers mentioned mun-
dane events about the simple topics in the poem (e.g. they noted stand-
ing in one place deciding what to do next), they mostly offered both
metaphorical and allegorical interpretations (72% of all participants’
verbal protocols).
For example, evidence of a general metaphorical theme is seen in one
reader’s thoughts about the first three lines of Frost’s poem:

Two roads diverged in a yellow wood,

And sorry I could not travel both
And be one traveler, long I stood

The participant interpreted these lines in the following manner:

The traveler, in making this tough decision, contemplates for a long time which
path to follow. This could be interpreted as a struggle or challenge in one’s life,
where one must decide which is the better path or way to go.

Another reader gave a specific metaphorical reading of the same three lines,
while also articulating a broader understanding of the allegorical theme
implicit in the poem. As this reader wrote,

The two roads represent different pathways in life that one may or may not
choose to take. Frost is saying that as a singular entity, you may only have a
singular history which is comprised of the choices you have made. Different
choices, or trying to clear a new road between the two existing ones (indecision)
would result in a new person. The last line deals with the hesitancy to make a life
changing decision. Options must be weighed carefully.

[ 228 ]  Literature and Cognition in Cognitive Science

 

In general, university students (studying psychology, not literature) were

quite adept at inferring specific metaphorical readings of both the different
three-​line segments and the poem’s overall metaphorical meanings. More
specifically, the talk-out-loud protocols clearly demonstrated how various
concrete terms in the poems served as source domains for people to think
about, and talk about, metaphorical target domains. Students could, there-
fore, immediately engage in metaphor processing on language that was not
explicitly marked by metaphor.
There are several ways to interpret these empirical findings. One may
simply argue that people interpret the Frost poem by activating a pre-​
existing, embodied conceptual metaphor LIFE IS A JOURNEY from long-​
term memory to make sense of what Frost may have meant (Lakoff and
Turner, 1989). However, as the psycholinguistic research reviewed above
suggests, people are likely creating coherent metaphorical or allegorical
messages by imagining themselves performing the actions mentioned in
the poem. Readers imagine themselves walking through the woods, deal-
ing with obstacles, facing difficult choices, as part of their in-​the-​moment
comprehension of the poem. As stated earlier, people do not interpret
linguistic meaning and only then reflect on its embodied consequences or
aesthetic impact. Running embodied simulations is a critical part of the
online processing of language and helps structure people’s interpretations
of allegorical messages.
A different empirical study further examined the role that embodied
simulations may play in literary interpretation, this time in relation to
prose fiction. Recall the ‘climbing ladder’ excerpt discussed at the begin-
ning of this chapter. This passage metaphorically describes the poet/​edi-
tor’s plight as he struggles to write the introductory essay, establish his
place within the world of poetry, and reclaim some harmony within his
personal life. Fully understanding the text requires readers to engage in an
extended metaphorical comparison in which the poet imagines his present
obstacle ‘like’ being in the unworldly situation of clinging to an ‘infinitely
tall ladder’ which telescopes ‘up into the blinding blue’. In this way, the
passage unobtrusively shifts from a single simile into a full-​blown allegory,
much like that seen in many instances of allegorical fiction. Indeed, it is the
moral character of allegories that makes them something more than mere
extended metaphors.
Can people draw allegorical inferences when reading the ‘climbing
ladder’ passage? If so, might these inferences emerge from embodied
simulation processes? One psychological study indicated that ordinary
university students provided consistent evidence of understanding the
‘climbing ladder’ story as generally referring to the conceptual metaphor

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 229 ]
 

LIFE IS A JOURNEY, an understanding achieved through embodied simu-

lation processes (Gibbs and Blackwell, 2012). People read the passage and
immediately wrote out their responses to a series of questions, including,
among others, the following:
Please describe what the ‘infinitely tall ladder’ refers to or represents. Ninety-​
five per cent of the participants noted that the ‘infinitely tall ladder’ was
symbolic of the LIFE IS A JOURNEY conceptual metaphor. People did not
explicitly mention the conceptual metaphor per se, but remarked on spe-
cific aspects of the mapping of journeys onto life, as when they noted how
a journey consisted of a source (i.e. the beginning or lower parts of the lad-
der), a path (i.e. the ladder itself), and a goal (i.e. success, fame, the solution
to the problem of writing the introductory essay). These references were
often linked to specific bodily actions related to the various poets climbing
the ladder.
What does the poet mean when he says he is ‘clinging to the telescoping lad-
der’ and he does not know how he got there? Almost all of the participants’
responses alluded to the poet’s clinging to the ladder as evidence of inse-
curity over what he was doing as a poet/​editor and his potential for ever
achieving fame as a poet. Thus, the poet being stuck on the ladder referred
to something larger than just his being unable to make progress in physi-
cally climbing the ladder.
Who is a more famous poet—​Graham or Auden? Who is a more famous
poet—​Kooser or Kunitz? Not a single person reported that they knew any of
the poets referred to in the passage. However, almost 80% of the partici-
pants recognized that the ladder represented the journey towards success
such that poets higher up on the ladder were older and more famous, with
those on the lower parts being younger and less famous. Most other people
gave answers that were consistent with this journey metaphor (i.e. poets
higher up were more famous), but were unable to articulate the reasons for
their decisions.
Who is Helen Vendler and why is she ‘filming our ascent’? None of the par-
ticipants reported having ever heard of Vendler. Still, 75% of participants
observed that Vendler was a critic or documentarian whose work it was
to observe and comment on the poets’ activities as they climbed the lad-
der. A few participants even wondered whether Vendler’s presence had any
influence on the different poets’ ladder climbings!
What would happen if the author ‘loosened his grip’ while on the ladder and
‘fell to one side’? People said that the poet wanted to give up trying to be
successful and stop trying to write the introductory essay. Some partic-
ipants felt that the poet’s thoughts about loosening his grip and falling
aside reflected his desire to get a different perspective on trying to write his

[ 230 ]  Literature and Cognition in Cognitive Science

 

introductory essay, or that he even wanted someone else to take over that
project. Here too, participants sometimes gave responses that drew specific
links to the poet’s bodily action and its broader symbolic connotations.
Describe the bodily sensations you felt while reading the story. All par-
ticipants reported having various bodily experiences while reading the
passage. Many referred to concrete physical sensations (e.g. feeling cold,
tired, stomach sinking), while others noted more general psychological
ills (e.g. anxiety, a sense of failure, a fear of competition). These sensa-
tions emerged during the actual reading of the story, as noted by some
participants, and were not after-​the-​fact reflections on some preliminary,
purely linguistic understanding of the text. Such reports are consistent
with the claim that embodied simulations give rise to various bodily
effects, many of which are constrained by the objects, actions, and people
referred to in the text.
Finally, not only did the students interpret specific phrases as having
metaphorical meaning, they also exhibited significant allegorical coherence
in their responses to the different questions, as shown by the consistent
embodied, metaphorical understandings manifest in their answers to the
specific questions described above (86% of the time). At the same time,
students did not simply recruit the general conceptual metaphor of LIFE
IS A  JOURNEY and only report its typical entailments (i.e. the ladder is
the path, the poets are travellers, the top of the ladder is the goal, etc.).
Instead, students offered idiosyncratic interpretations of the story parts,
giving their meaning products a unique character that speaks to everyone’s
individual, embodied, allegorical experience. For example, with the ques-
tion asking about the meaning of the ‘infinitely tall ladder’, participants
responded as follows:

• The function of the ladder is to lead people into the imagination of

poetry.
• The ladder provides the link between the realistic, mechanical world and
the artistic world.
• The ladder represents the world of people who are totally immersed in
literature/​poetry.
• The poet on the ladder has his head in the clouds and he’s out of touch
with reality.

Overall, people frequently provided interpretations that reflected general

allegorical themes, yet also showed very specific, idiosyncratic understand-
ings of different parts of the story. This finding is also consistent with the
idea that people run embodied simulations to interpret the text, but these

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 231 ]
 

all differ depending on individuals’ unique bodily experiences, past and

present, as shaped by their knowledge and expertise.
Interpreting some poems, such as ‘The Road Not Taken’, or certain fic-
tion, such as the excerpt from The Anthologist, may be relatively easy to
do given that these are motivated by the single, recurring embodied met-
aphor LIFE IS A  JOURNEY. But are people capable of interpreting liter-
ary excerpts that prompt a mix of simulation processes? Let me report on
another of my own embodied experiences of literary interpretation that
speaks to this question.
When in high school, I had the pleasure of reading D. H. Lawrence’s
novel Lady Chatterley’s Lover (1928/​2013). This novel describes a young
married woman, Constance (Lady Chatterley), who is married to a
handsome man named Clifford. Due to a war injury, Clifford has signifi-
cant physical (sexual) limitations, but is also emotionally distant from
Constance. Eventually, Constance’s frustration leads her to have a sexual
relationship with the family’s gamekeeper, Oliver. Much of the novel con-
cerns the class differences, and struggles, between Constance and Oliver,
and also Constance’s realization that she must experience sexual love to
be fully alive.
One notable passage in the novel describes a time when Constance and
Oliver make love, which is special because she finally feels tremendous
emotional and physical engagement.

And it seemed she was like the sea, nothing but dark waves rising and heaving,
heaving with a great swell, so that slowly her whole darkness was in motion,
and she was Ocean rolling its dark, dumb mass. Oh, and far down inside her
the deeps parted and rolled asunder, in long, fair-​travelling billows, and ever, at
the quick of her, the depths parted and rolled asunder, from the centre of soft
plunging, as the plunger went deeper and deeper, touching lower, and she was
deeper and deeper and deeper disclosed, the heavier the billows of her rolled
away to some shore, uncovering her, and closer and closer plunged the palpable
unknown, and further and further rolled the waves of herself away from herself
leaving her, till suddenly, in a soft, shuddering convulsion … (p. 210)

Wow! This passage, which is replete with metaphors, was quite memorable
to me when I read it at age sixteen. I clearly recall my imaginative engage-
ment with the text, as I envisioned myself being ‘like the sea’, with all of its
‘rising and heaving’, and what it must be like to be Constance ‘as the plunger
went deeper and deeper, touching lower’ and more. I am not a woman, yet
even so, my teenage imaginative abilities permitted me to ‘be there’ as if
I too was an active participant in the characters’ sexual adventure.

[ 232 ]  Literature and Cognition in Cognitive Science

 

I mention this memorable reading experience to highlight the flexible

and individual manner in which embodied simulations are created dur-
ing literary understanding. Simply recognizing that HAVING SEX IS LIKE
BEING THE SEA fails to capture the depth of my embodied literary expe-
rience that was shaped precisely by Lawrence’s guiding metaphors. My
understanding of the text was constructed by in-​the-​moment embodied
simulations of ‘what it must be like’ to participate in some manner, from
some perspective, in the events depicted in the text.
Embodied simulations are, of course, based on the instructions that lan-
guage provides us. Yet it is also quite possible to create a detailed simula-
tion of a scenario when specific actions are not even mentioned, but only
alluded to. Consider the following sequence from Deirdre Madden’s (2008/​
2010) novel Molly Fox’s Birthday. In this scene, the narrator, who is a play-
wright, is talking on the phone with her friend Molly Fox, who is a stage
actor with a voice ‘infused with a slight ache’. Molly has just received birth-
day wishes from a mutual friend and she asks the playwright regarding this
friend:

‘How did he know that today was my birthday? Did you tell him?’
‘It was in the paper.’
‘What! How old did they say I was?’
‘Forty.’
She swore when I  said this, a sudden, crude outburst. It was all the more
shocking because Molly almost never swears. There was the incongruity of hear-
ing such a thing uttered in that particular voice, and I realised that she was as
capable of drawing forth all the ugly power an oath might contain as she could
the beauty and tenderness of other words. ‘I never heard such nonsense in my
life. I’m only thirty-​eight.’ (p. 219)

This scene is remarkable because it demonstrates how indirection, by not

actually stating what Molly swore in ‘a sudden, crude outburst’, can still
give rise to embodied simulations about what may have occurred. When
reading about Molly’s outburst, I immediately heard the echo of the very
obscenities she likely uttered. Still, the passage would not have been as suc-
cessful if the real words spoken had been presented. But the author’s fram-
ing of the scene, and the mention of Molly’s being ‘as capable of drawing
forth all the ugly power an oath might contain as she could the beauty and
tenderness of other words’, shapes our own individual simulations of what
Molly may have shouted. Even if we are not given explicit words to nar-
rowly define the embodied simulation automatically constructed to under-
stand the passage, we still rely on what the author provides to create our

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 233 ]
 

own sensory experiences of ‘what it must have been like’ to hear Molly’s
outburst and feel her displeasure.

EMBODIED SIMULATIONS IN LITERARY CRITICISM

Literary critics do not simply interpret texts in an abstract manner, look-

ing for deep transcendent meanings. Instead, similar to recreational read-
ers, they understand texts, and write about texts and writers, in embodied
ways. The critic Helen Vendler, who navigates her ‘trusty dirigible’ to
observe literary poetics, reminds us of this when she voices her own view
of a critic’s task when engaged in ‘close reading’ (Vendler, 1975):

It’s a view from the inside, not from the outside. The phrase ‘close reading’
sounds as if you’re looking at the text with a microscope from outside, but
I would rather think of a close reader as someone who goes inside a room and
describes the architecture. You speak from inside the poem as someone looking
to see how the roof articulates with the walls and the wall articulates with the
floor. And where are the crossbeams that hold it up, and where are the windows
that let light through.

Doing literary criticism ‘as someone who goes inside a room and describes
the architecture’ is, we might now say, creating embodied simulations
to discover what it is like to ‘speak from inside the poem’. Vendler then
describes her preferred method for doing literary criticism:

I don’t believe that poems are written to be heard, or as Mill said, to be over-
heard; nor are poems addressed to their reader. I believe that poems are a score
for performance by the reader, and that you become the speaking voice. You
don’t read or overhear the voice in the poem, you are the voice in the poem.
You stand behind the words and speak them as your own—​so that it is a very
different form of reading from what you might do in a novel where a character
is telling the story, where the speaking voice is usurped by a fictional person to
whom you listen as the novel unfolds.

Whenever literary critics ‘are the voice in the poem’ and ‘stand behind the
words and speak them as [their] own’, they are engaging in exactly the
same general process of embodied simulation that scientific studies have
shown recreational readers construct during their readings of both non-​
literary and literary language.

[ 234 ]  Literature and Cognition in Cognitive Science

 

Not surprisingly, the language of literary criticism offers excellent

examples of this sort of embodied engagement with texts. Consider what
one critic, James Wood, wrote about Virginia Woolf as critic (Wood, 1999/​
2010):

In her criticism, the language of metaphor becomes a way of speaking to fiction

in its own accent, the only way of respecting fiction’s indescribability. Metaphor
… is a language of forceful hesitation. Its force lies in the vigor and the original-
ity of Woolf’s metaphors: its hesitation lies in its admission that, in criticism,
the language of pure summation does not exist. Criticism can never offer a suc-
cessful summation, because it shares its subject’s language. …
All criticism is metaphorical in movement, because it deals in likeness. It
asks: what is art like? What does it resemble? How can it best be described, or
redescribed? If the artwork describes itself, then criticism’s purpose is to rede-
scribe the artwork in its own, different language. (pp. 106–​107)

Wood suggests that criticism interprets artworks by experiencing their

movement in metaphors and by communicating with them, in Woolf’s
case, through ‘a language of forceful hesitation’. Literary criticism may
‘redescribe the artwork’, through critics’ personal, embodied engagement
as they share their subjects’ language.

CONCLUSION: MERGING SCIENCE WITH LITERARY CRITICISM

Various empirical studies now support the claim that reading and inter-
preting literature involves embodied simulation processes. These processes
operate at different levels of granularity depending on a host of factors
having to do with the particular language being interpreted, the goals and
motivations of the reader, and that person’s individual personality and his-
tory. Some embodied simulations may be fragmentary, or incomplete, as
when a person is casually reading a news report, while others may be more
complex and ‘fleshed out’, as when literary critics intensely study a particu-
lar text over a long period of time. But all embodied simulations are critical
to readers’ experience of the interior of text worlds, and to feeling trans-
ported into the thoughts and actions of others. Research on the embod-
ied dynamics of literary experience is still in its infancy, and there is much
to learn about the depth and quality of simulation processes in different
situational and personal contexts. Still, research shows that the embod-
ied simulations people typically, unconsciously create when engaging with

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 235 ]
 

literary texts can alter readers’ personalities (Djikic and Oatley, 2014),
and their mood, feelings of empathy, and aesthetic liking (Lüdtke, Meyer-​
Sickendieck, and Jacobs, 2014). These behavioural studies have all been
conducted with naïve readers, yet their results may still offer insights into
the practices of literary interpretation.
Reading and criticizing literature are not separate activities. Of course, the
motivations, goals, and expertise that anyone brings to reading literature may
differ. My argument, however, is that literary experiences of all types are fun-
damentally grounded in embodied simulation processes. Empirical support
for the embodied simulation hypothesis adds weight to the relevance of this
idea for understanding how literary critics often arrive at their sometimes
more idiosyncratic readings of what texts mean and implicate. It may be a
natural response, when hearing of scientific studies of literature, for literary
critics to reject these findings because they characterize ordinary, average, or
normative reading processes. But the scientific research essentially offers a
foundation from which unique interpretations of literature may arise.
Everyone may automatically engage in embodied simulation processes,
where people imagine themselves participating in the acts specified by
language. These simulation activities are not mere neural actions that are
not part of individuals’ meaningful experiences. Even fast-​acting, uncon-
scious processing shapes people’s thoughts, understandings, and actions.
Nonetheless, the meaning products of those simulations clearly differ
according to a wide variety of personal and contextual factors. Recognition
of this fact—​that similar psychological processes may create different
interpretive products—​is critical to closing the gap between the scientific
study of literature and the scholarly practice of literary criticism.

REFERENCES

Baker, N. (2009). The anthologist. New York: Simon and Schuster.

Bergen, B. (2012). Louder than words: The new science of how the mind makes meaning.
New York: Basic.
Boulenger, V., Hauk, O., and Pulvermüller, F. (2009). Grasping ideas with the motor
system:  Semantic somatotopy in idiom comprehension. Cerebral Cortex, 19,
1905–​1914.
Burke, M. (2011). Literary reading, cognition, and emotion: An exploration of the oceanic
mind. New York: Routledge.
Desai, R., Binder, J., Conant, L., Mano, Q., and Seidenberg, M. (2011). The neural career
of sensory-​motor metaphors. Journal of Cognitive Neuroscience, 23, 2376–​2386.
Djikic, M., and Oatley, K. (2014). The art in fiction: From indirect communication to
changes of the self. Psychology of Aesthetics, Creativity, and the Arts, 8, 498–​505.
Fitzgerald, F. S. (2000). The great Gatsby. London: Penguin.

[ 236 ]  Literature and Cognition in Cognitive Science

 

Frost, R. (1969). The road not taken. In The poetry of Robert Frost: The collected poems,
complete and unabridged (p. 105). New York: Holt.
Gibbs, R. (2006). Embodiment and cognitive science. New  York:  Cambridge
University Press.
Gibbs, R. (2013). Walking the walk while thinking about the talk:  Embodied inter-
pretation of metaphorical narratives. Journal of Psycholinguistic Research, 42,
363–​378.
Gibbs, R., and Blackwell, N. (2012). Climbing the stairs to literary heaven: A case study
of allegorical interpretation of fiction. Scientific Study of Literature, 2, 197–​217.
Gibbs, R., and Boers, E. (2005). Metaphoric processing of allegorical poetry. In Z.
Maalej (Ed.), Metaphor and culture. Tunis: University of Manouba Press.
Gibbs, R., and Colston, H. (2012). Interpreting figurative meaning. New York: Cambridge
University Press.
Gibbs, R., Gould, J., and Andric, M. (2005–​ 2006). Imagining metaphorical
actions:  Embodied simulations make the impossible plausible. Imagination,
Cognition, and Personality, 25, 221–​238.
Glenberg, A., and, Kaschak, M. (2002) Grounding language in action. Psychonomic
Bulletin and Review, 9, 558–​565.
Hickok, G. (2014). The myth of mirror neurons: The real neuroscience of communication and
cognition. New York: Norton.
Lakoff, G., and Johnson, M. (1999). Philosophy in the flesh: The embodied mind and its
challenge to Western thought. New York: Basic.
Lakoff, G., and Turner, M. (1989). More than cool reason: A field guide to poetic metaphor.
Chicago, IL: University of Chicago Press.
Lawrence, D. H. (2013). Lady Chatterley’s lover. London: Harper.
Lüdtke, J., Meyer-​Sickendieck, B., and Jacobs, A. (2014). Immersing in the stillness of
an early morning: Testing the mood empathy hypothesis of poetry reception.
Psychology of Aesthetics, Creativity, and the Arts, 8, 363–​377.
Maas, A., and Russo, A. (2003). Directional bias in the mental representation of spatial
events: Nature or culture? Psychological Science, 14, 296–​301.
Madden. Deirdre. (2010). Molly Fox’s birthday. New York: Picador.
Mahon, B., and Caramazza, A. (2008). A critical look at the embodied cognition
hypothesis and a new proposal for grounding conceptual content. Journal of
Physiology (Paris), 102, 59–​70.
Massey, I. (2009). The neural imagination: Aesthetic and neuroscientific approaches to the
arts. Austin, TX: University of Texas Press.
Oatley, K. (2011). Such stuff as dreams: The psychology of fiction. New York: Wiley-​Blackwell.
Rizzolatti, G., and Craighero, L. (2004). The mirror neuron system. Annual Review of
Neuroscience, 27, 169–​192.
Vendler, H. (1975). The art of criticism III. The Paris Review, 190. Retrieved from http://​www.
theparisreview.org/​interviews/​1324/​the-​art-​of-​criticism-​no-​3-​helen-​vendler
Wilson, N., and Gibbs, R. (2007). Real and imagined body movement primes metaphor
comprehension. Cognitive Science, 31, 721–​731.
Wood, J. (2010). The broken estate: Essays on literature and belief. New York: Picador.
Zwaan, R., Stanfield, R., and Yaxley, R. (2002). Language comprehenders mentally rep-
resent the shapes of objects. Psychological Science, 13, 168–​171.
Zwaan, R., and Taylor, L. (2006). Seeing, acting, understanding: Motor resonance in
language comprehension. Journal of Experimental Psychology: General, 135, 1–​11.

E m b o di e d D y n a m i c s i n L i t e r a r y E x p e r i e n c e   [ 237 ]
 
 

CHAPTER 12

How Readers’ Lives Affect

Narrative Experiences
RICHARD J. GERRIG AND MIC AH L . MUMPER

I n the novel The Slippage (Greenman, 2013), William has taken his friend’s
young son, Christopher, to play in a park. Christopher attempts to fly
a kite, with indifferent results: ‘Every few minutes it went into irons and
came crashing back to the ground’ (p. 120). Still, on Christopher’s account,
the crashes don’t spoil his enjoyment:  ‘ “I don’t mind,” Christopher said.
“It’s fun to get it going again” ’ (p. 120). While most readers will find this
scene charming, some will likely have specific recollections of their youth-
ful successes and failures with kites. Some, in fact, might feel inclined to
offer Christopher mental council about how kites might be best kept aloft.
In fact, moments later, William’s neighbour appears and gives some solid
advice, but William isn’t entirely grateful:  ‘He was right, William knew,
but Christopher was having fun running back and forth’ (p. 121). In this
context, some readers will likely feel scolded had they offered comparable
mental advice.
We use this brief scene to support a claim that we expect not to be con-
troversial:  readers’ accumulated memories have a substantial impact on
their narrative experiences. The aim of this chapter is to make that claim
concrete, by drawing upon theories and empirical research from cognitive
psychology. We wish, in particular, to characterize readers’ thoughts and
emotional responses as they engage with a text. We suggest that each read-
er’s experience is unique and that cognitive-​psychological analysis can help
explain how those unique experiences emerge. Note that when we speak
 

of accumulated memories, we include readers’ own life events as well as

the knowledge they have acquired through their interactions with narra-
tive worlds. Also, we use the term readers for convenience. The processes
we outline affect people’s experiences of narratives across media and types
of telling.
In this chapter, we outline basic cognitive processes that make con-
tact with readers’ memory representations as their narrative experiences
unfold. We describe how these basic processes influence readers’ infer-
ences and judgements about narrative events. We suggest how these pro-
cesses shape readers’ participation in narratives, with a particular focus
on readers’ assessments of their similarity to characters with implications
for their experiences of empathy. Finally, we consider how these basic pro-
cesses affect readers’ individual reports of their transportation to narrative
worlds.

MEMORY-​B ASED ASPECTS OF NARRATIVE EXPERIENCES

As readers experience narratives, basic cognitive processes connect the

unfolding narrative to representations in long-​term memory. For example,
theories of text comprehension (beginning with Kintsch, 1988, 1998) gen-
erally make a distinction between processes of construction and integration
(for a review, see McNamara and Magliano, 2009). Construction refers to
the activation of information from the text as well as from readers’ related
knowledge. As readers make their way through successive parts of a narra-
tive, information in the text primes (i.e. makes more accessible) memory
traces from both earlier parts of a narrative and the readers’ general back-
ground knowledge.1 The processes that make these memory traces acces-
sible are not goal-​directed (Myers and O’Brien, 1998; O’Brien and Myers,
1999). Rather, textual information increases the accessibility of a range
of memory traces, only some of which will ultimately be relevant to the
readers’ subsequent narrative experience. Thus, as readers experience The
Slippage’s account of Christopher’s afternoon, personal memories related
to kites and personal memories related to parks will become accessible,

1.  In their classic demonstrations, Meyer and Schvaneveldt (1971) showed that par-
ticipants made faster judgements about related words (e.g. bread, butter) than about
unrelated words (e.g. nurse, butter). The results suggested that exposure to the initial
word of the related pair made the second word more accessible in memory. For narra-
tive experiences, basic priming processes similarly change the accessibility of informa-
tion in memory (for a review, see Cook and O’Brien, 2015).

[ 240 ]  Literature and Cognition in Cognitive Science

 

as well as memories at the intersection of kites and parks. Some of those

memories will help with readers’ comprehension of specific aspects of the
scene including, for example, the low efficacy of Christopher’s strategy of
running back and forth. Other memories—​perhaps a particular reader’s
vivid recollection of a favourite childhood kite—​will be less critical because
they are not relevant to the unfolding narrative.
In fact, as Walter Kintsch (1988) noted, construction processes yield
memory representations that include elements ‘without regard to the dis-
course context, and many of them are inappropriate’ (p. 168). Integration
processes create a ‘coherent whole’ (p.  164) from the products of con-
struction processes. The straightforward point here is that comprehen-
sion processes make contact with readers’ long-​term memories, and those
storehouses of memories will be different for each reader. Those different
memories that become accessible through the construction process will
differentiate readers’ models of narrative worlds as they emerge from the
integration process. We now consider further consequences for the role of
diverse memory representations in narrative experiences.
To begin, we note that readers often differ in high-​level expertise that
may be relevant to particular narrative worlds. Research has indicated that
readers’ expertise affects their understanding (e.g. Fincher-​Kiefer, Post,
Greene, and Voss, 1988; Griffin, Jee, and Wiley, 2009). For example, in one
classic study, participants listened to an excerpt from a fictional baseball
game (Spilich, Vesonder, Chiesi, and Voss, 1979). Participants who were
high in baseball knowledge were better able to recall important features of
the game.
Most extended narratives draw upon several domains of expertise. For
example, The Golem and the Jinni (Wecker, 2013) takes place in New York
City in 1899. Parts of the novel are set in the Jewish community on the
Lower East Side; other parts are set within a Syrian community in an adja-
cent part of the city. One important character, Chava, works as a baker;
another, Ahmed, works as a tinsmith. Each of these aspects of the narrative
world makes expertise relevant. One scene, in which Ahmed finds his way
to the interior of Central Park (p. 104), makes this claim concrete:

Eventually the broad walk descended into a tunnel that cut beneath a carriage
road. On the other side of the tunnel, a broad plaza of red brick curved along the
shore of a pond. In the middle of the plaza he saw what he took at first for an
enormous winged woman, floating above a foaming cascade of water. No, not a
woman—​a sculpture of a woman, perched atop a pedestal. The water flowed into
a wide, shallow basin at her feet, and then into a pool that stretched almost the
width of the plaza.

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 241 ]
 

In the context of this passage, it is possible to invent a great variety of

readers who differ in their Central Park expertise. Some may treat the foun-
tain as pure invention, whereas others will have anywhere from sparse to
extensive personal memories of Bethesda Fountain. In addition, some of
those memories may reflect real-​life encounters with the fountain, while
others may reflect indirect knowledge (e.g. from other narrative works). All
of this diversity will likely have an impact on how readers experience this
moment.
Readers’ knowledge also has an impact on the inferences they encode.2
We use the term inference to refer to information that readers repre-
sent and that was not explicitly stated in a text. Expertise changes the
inferences that readers encode. For example, in the baseball study, high-​
knowledge individuals were more likely to produce accounts of the game
that included elaborations of the original text (Spilich et  al., 1979). We
would expect that as they experience Ahmed’s encounter with the foun-
tain, readers with extensive Central Park knowledge would encode dif-
ferent inferences than would less knowledgeable peers. They might, for
example, supplement their mental model of the text by filling in visual
details of the fountain.
In fact, theories of discourse comprehension have focused extensive
attention on the particular inferences that readers encode as they experi-
ence a narrative. Such analyses are anchored by the claim that any text
permits an infinite number of inferences (Rieger, 1975). In that context,
theorists have been particularly interested in characterizing the inferences
that readers regularly encode through the operation of automatic pro-
cesses (i.e. without engaging strategic effort) (for a review, see McNamara
and Magliano, 2009). Gail McKoon and Roger Ratcliff (1992) originated an
influential theory now known as memory-​based processing. This position,
which has obtained extensive empirical support, asserts that there are no
automatic processes unique to circumstances of text processing. Rather,
‘the only automatic processes readers bring to bear on text processing
are ordinary memory processes’ (Gerrig and O’Brien, 2005, p.  228). The
memory-​based processing approach asserts that readers encode inferences
which are supported by information easily available from memory (McKoon
and Ratcliff, 1992). Information from earlier in the text will often be easily
available, ensuring that some of the inferences readers encode will be quite

2.  We use the term encode throughout the chapter to mean ‘encoded into memory’.
This usage is standard within text processing research (e.g. ‘If a reader can be said to
have understood an inference, then the required information must have been available
during reading and the inference must be encoded into memory,’ McKoon and Ratcliff,
2015, p. 52).

[ 242 ]  Literature and Cognition in Cognitive Science

 

similar. However, the availability of information from readers’ background

knowledge will differ as a product of their own accumulations of memories.
As such, readers’ inferences will potentially show great diversity.
Memory processes also contribute to readers’ experience of narrative
events as normal or abnormal. Daniel Kahneman and Dale Miller (1986)
originated norm theory to provide an account of the processes determin-
ing the judgements of normality that emerge without conscious effort.
In norm theory, new pieces of information serve as probes that resonate
through memory, activating representations of related information. This
memory process yields a norm constructed in the moment. When the cur-
rent event is consistent with that norm, people will experience the event as
normal; a mismatch provides a judgement of abnormality. Kahneman and
Miller also suggested that judgements of abnormality prompt people to
ask ‘Why?’ questions, as a means to understand why current circumstances
depart from the norm. In that sense, readers’ tacit judgements of normality
provide a prompt for the types of active engagement in narrative experi-
ences we detail later. Specifically, we suggest that tacit judgements of abnor-
mality cause readers to experience mysteries that they may then choose to
contemplate (Gerrig and Wenzel, 2015; Wenzel and Gerrig, 2015).
We can illustrate these claims by considering the outset of The Slippage
(Greenman, 2013). As the novel begins, William and Louisa Day are cheer-
fully making preparations for a party that Louisa had suggested they host.
But moments later, as guests begin to arrive, Louisa has gone missing.
William eventually concludes that she has closed herself in their junk room,
but she does not respond to his entreaties: ‘After another trip inside, and
another session spent thumping on the junk room door—​lightly enough,
so as not to draw the attention of the guests—​William went back outside
and collected shards of conversation’ (p.  15). Again, we could invent a
variety of readers. However, we expect that automatic processes operat-
ing over most readers’ memories will indicate that Louisa’s behaviour is, to
some extent, ‘not normal’. As a function of the intensity of their particular
response, individual readers will likely expend different amounts of effort
to try to address the mystery of Louisa’s behaviour. Their solutions to the
mystery will also draw upon their personal memories. Have they experi-
enced parties at which a host has gone missing? Thus, the initial product
of automatic processes (i.e. ‘something here is unusual’) will become more
differentiated as readers engage strategic effort.
As The Slippage unfolds, readers continue to gather knowledge of Louisa.
Specifically, they learn why she hid herself from the party. More generally,
they experience a range of Louisa’s behaviour in other circumstances. In that
way, they begin to encode memories that are specific to Louisa. Researchers

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 243 ]
 

have demonstrated that readers are able to learn quickly what counts as
normal in particular narrative worlds. For example, Mante Nieuwland
and Jos van Berkum (2006) had participants read a story in which a yacht
was undergoing psychotherapy. At first, participants responded to that
information with surprise (as indexed by their brain responses). However,
five sentences later participants were no longer surprised when the yacht
engaged in activities that typically require the subject to be animate. More
generally, research suggests that readers readily use their accumulated
knowledge of particular narrative worlds to adjust their expectations of
what counts as normal (Filik, 2008; Filik and Leuthold, 2008, 2013; Foy
and Gerrig, 2014). We suggest that, in this particular case, readers’ knowl-
edge specific to Louisa changes the sample of memories that will be most
accessible to automatic judgements of normality. Thus, as readers cast their
thoughts back to Louisa’s disappearance, they are likely to experience her
behaviour as more normal (for her) the deeper they get into the novel. In
fact, we speculate that responses among readers will generally show more
variability towards the beginnings of extended narratives than towards the
ends.
The concepts we have reviewed suggest why readers’ narrative expe-
riences will change as they revisit narrative worlds at different points in
their lives. To begin, readers can acquire new expertise. If, for example,
a reader makes a thorough visit to Central Park between readings of The
Golem and the Jinni (Wecker, 2013), the subsequent experience of relevant
passages (e.g. ‘a sculpture of a woman, perched atop a pedestal’, p. 104) will
likely be quite different. More generally, readers will accumulate abundant
new memory traces that will serve as the source of background knowledge
for comprehension. Depending on readers’ local life experiences, memory
traces will differ in accessibility from reading to reading. For example, a
reader who has undertaken a spate of kite flying just before re-​reading The
Slippage will experience that scene with memories in revised resting lev-
els of accessibility. Thus, each reader’s automatic inferences are likely to
change over time as a function of the information that is easily available
in his or her memory. Finally, to the extent that judgements of normal-
ity rely once again on information that emerges from automatic memory
processes, each reader’s sense of what is normal is likely to change as life
events accrue.
Note here that the processes we have described operate almost entirely
outside of readers’ conscious awareness. That is, readers do not need to
expend strategic effort to encode inferences or make judgements of nor-
mality. As such, they will often have little awareness of how and why their
narrative experiences diverge from those of other readers.

[ 244 ]  Literature and Cognition in Cognitive Science

 

We turn now to types of reader responses that typically operate within

conscious awareness to influence narrative experiences.

READERS’ PARTICIPATORY RESPONSES

Consider events in The Golem and the Jinni (Wecker, 2013)  in which the
golem, Chava, has regained control after a fugue state in which she severely
battered a rogue for assaulting a friend. When Chava recalls her inability
to control her aggressive instincts, she decides that she must destroy her-
self:  ‘It was a simple decision, quickly made. She couldn’t be allowed to
hurt anyone again’ (p. 312). In response to this moment, we suspect that
most readers will expend mental effort to dissuade Chava from carrying
out her plan. They might, for example, hear a mental voice crying out,
‘Don’t do it!’
These types of mental contents, which are an important component of
readers’ narrative experiences, are called participatory responses (Allbritton
and Gerrig, 1992). The concept of participatory responses follows from the
participatory perspective on narrative experiences. This perspective suggests
that readers encode the same types of mental contents they would encode
were they actual participants in the narrative events (Gerrig, 1993; Gerrig
and Jacovina, 2009). To create a taxonomy of participatory responses,
Matthew Bezdek and colleagues (2013) asked experimental participants to
speak aloud while they watched brief scenes from Hollywood films. They
counted participants’ productions as participatory responses only if those
productions included content that was not a repetition of plot details or
inferences based on those details. Participants often provided content with
great emotional intensity. For example, participants watched one scene in
which they believed that a girl would trigger an explosion if she answered a
ringing phone. Here is one participant’s verbal response:

Please don’t pick up the phone little girl

Oh my goodness
(Gasps) Noo, no, Oh my god please, no no no don’t get that. Don’t get
it don’t get it don’t get it

Bezdek and colleagues identified several categories of participatory

responses. In this one example, we hear the viewer expressing an outcome
preference (i.e. the participant does not want the girl to answer the phone)
as well as a problem-​solving instruction (i.e. the participant addresses the girl
with the instruction not to answer the phone). Based on their analyses, the

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 245 ]
 

authors asserted that viewers and, by analogy, readers encode participa-

tory responses with reasonable frequency.
Other research suggests that such responses affect how readers assimi-
late narrative events. For example, narratives often provide a clash between
what outcomes appear likely versus what readers would prefer to happen.
We see such circumstances in The Slippage (Greenman, 2013). William has
been put on leave because he has punched his boss. We suspect that most
readers would prefer that William not be fired, although the firing seems
highly likely. In experiments that captured these types of clashes, par-
ticipants found it difficult to assimilate outcomes that mismatched their
preferences (Rapp and Gerrig, 2002, 2006). For example, participants
took longer to indicate that they understood an outcome sentence when
that outcome was unwanted. Of course, most narratives allow readers to
develop diverse preferences. Matthew Jacovina and Richard Gerrig (2010)
demonstrated that readers’ particular responses helped predict the time
course with which they assimilated narrative outcomes. They asked par-
ticipants to indicate their preferences in advance of characters’ decisions.
Imagine, for example, that readers had officially weighed in before William
had punched his boss. Some readers would likely have endorsed the action
(because the boss is an insufferable fool); others would likely have advised
William to forgo the momentary pleasure of decking his boss. Jacovina
and Gerrig’s data indicated that participants read statements of outcomes
more slowly if those outcomes clashed with their particular preferences.
As we have revealed, William does in fact punch his boss. Ultimately, he is
fired. Jacovina and Gerrig’s results suggest that readers who were in favour
of William’s actions would find it more difficult to assimilate the informa-
tion that he had been fired.
These projects indicate that readers encode participatory responses, and
that those responses have lingering consequences. For example, when read-
ers have different outcome preferences, those individual differences affect
their reception of the subsequent narrative events (Jacovina and Gerrig,
2010). Some of those outcome preferences will arise, no doubt, from read-
ers’ own collections of memories. To expand on that point, we narrow our
focus to readers’ responses to characters. Fundamentally, readers may pre-
fer that William punch his boss or not as a consequence of how much they
empathize with him. To move towards a discussion of empathy, we begin
with a consideration of similarity.
In everyday social interaction, people’s judgements of similarity have
important consequences. For example, people often determine which
other individuals count as members of their in-​groups and out-​groups
based on similarity (Turner and Reynolds, 2003; Shkurko, 2015). People

[ 246 ]  Literature and Cognition in Cognitive Science

 

respond to others quite differently as a function of those assessments

(Brock, 1965; Mitchell, Macrae, and Banaji, 2006; Alabastro, Rast, Lac,
Hogg, and Crano, 2013). For example, Tessa West and colleagues (2014)
created circumstances in which college students participated in interac-
tions with strangers of other races. Such interactions often provoke anxi-
ety among participants (for a review, see MacInnis and Page-​Gould, 2015).
West and colleagues reduced the students’ anxiety by manipulating their
perceptions of similarity. Specifically, participants in a high self-​revelation
condition believed that out-​group members had provided a number of rev-
elations similar to their own on questions such as ‘Would you rather be
extremely lucky or extremely smart?’ Similarity for questions that involved
lower levels of self-​revelation (e.g. ‘Would you rather go to Burger King or
McDonald’s?’) failed to have an impact on cross-​racial interactions. Thus,
the students’ assessments of similarity changed only when the informa-
tion focused on ‘something important about the self’ (p.  94). Literary
theorists have drawn upon research on social interaction to highlight the
importance of similarity. They have suggested, for example, that readers’
judgements of their similarity to characters affect the extent to which they
experience empathy (e.g. Keen, 2006, 2007; Hogan, 2011). We agree with
those claims, and return to them shortly. However, we also suggest that
theories of empathy should recognize the complexities of how readers’
judgements of similarity emerge.
Consider Amos Tversky’s (1977) classic contrast model of similarity.
Tversky’s model argued that people assess similarity based on weighted
combinations of the common and distinctive features of a pair of items.
Features can be any characteristic of a stimulus and may be concrete (i.e.
big) or abstract (i.e. beautiful). The model is specified as (p. 332)

S ( A, B ) = θ f ( A ∩ B ) − α f ( A − B ) − β f ( B − A ) ,

where A and B represent a set of characteristic features of objects or events.

A ∩ B represents the features common to A and B, whereas A − B and B − A
represent the features distinctive to A or B, respectively. The scaling fac-
tor (f) reflects the salience of any of the (common or distinctive) features,
and a set of weights ( θ, α, and β ) that change how heavily the common
or distinctive features factor into the similarity judgement. According to
Tversky, the salience of features is determined, in part, by diagnostic fac-
tors. Diagnostic factors refer to the importance of certain features depend-
ing on context. For example, ‘the feature “real” has no diagnostic value in
a set of actual animals since it is shared by all actual animals and hence

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 247 ]
 

cannot be used to classify them. This feature, however, acquires consider-

able diagnostic value if the object set is extended to include legendary ani-
mals, such as a centaur, a mermaid, or a phoenix’ (Tversky, 1977, p. 342).
This diagnosticity principle predicts that the features which are most salient
in any given similarity judgement will change depending on the features
made salient by the other objects present. More generally, depending on
the context, different features will take on different weight.
The results of West and colleagues’ (2014) research on cross-​racial inter-
actions exemplify aspects of this analysis. They demonstrated that only
particular types of common features (i.e. important information about
the self) have a functional impact on similarity. The experiment also dem-
onstrated a way in which assessments of similarity depend on context. In
circumstances of low self-​revelation, the students’ responses were largely
dictated by the salient feature of race. However, circumstances of high self-​
revelation increased the salience of common features and thereby made
race less salient.
We can now apply this analysis to a literary example. Consider the char-
acter Gabriel from A Nearly Perfect Copy (Amend, 2013). Gabriel is male and
in his mid-​thirties. He is an artist who believes that his work is underap-
preciated. He is a Spaniard who feels cultural alienation among his Parisian
peers. All of these attributes could enter into the equation as common or
distinctive features to determine how similar readers feel to Gabriel. Quite
plainly, the features will count differently as common or distinctive for
different readers. In addition, as the novel unfolds, feelings of similarity
will likely change as the salience of features changes. For example, at one
moment in the novel, Gabriel offers very strong opinions about Damien
Hirst: ‘Of all the contemporary posers who seemed to have charmed the
establishment, Damien Hirst seemed the most vile, mercenary, talentless
of the bunch’ (p. 79). For readers who do not know that Damien Hirst is a
real artist (the novel also includes fictional ones), this moment might have
little impact on their assessment of similarity. However, readers with art
world expertise will likely find themselves somewhere on the dimension of
agreement versus disagreement with Gabriel’s sentiments. Thus, different
readers will experience this attitude as a common or distinctive feature.
For some it will be highly diagnostic whereas for others it will not have any
particular salience. This one moment—​and the other myriad moments in
which Gabriel thinks, emotes, and acts—​may have a dramatic impact on
how readers experience their similarity to him. Thus, it is possible to assert
in a general way that similarity affects readers’ narrative experiences with-
out being able to assert what features will matter most for any particular
reader’s experience of similarity.

[ 248 ]  Literature and Cognition in Cognitive Science

 

As we noted earlier, theories of narrative engagement have often sug-

gested that the likelihood that readers will experience empathy towards
particular characters varies with the perception of similarity (e.g. Keen,
2006, 2007; Hogan, 2011). These analyses often acknowledge, however,
that empathy itself remains ill-​defined. In an extensive review of the psy-
chological literature, Jamil Zaki (2014) suggested that empathy has three
components (pp. 1608‒1609):

• Mind perception: Observers’ detection that targets have internal states.

• Experience sharing: Observers’ ‘tendency to take on the sensory, motor,
visceral, and affective states they encounter in targets’.
• Mentalizing: Observers’ ‘capacity to draw explicit inferences about tar-
gets’ intentions, beliefs, and emotions’.

These components are separable in particular instances. For example, read-

ers may ‘take on’ a character’s emotional state without having an explicit
understanding of the basis of that state.
In the context of these components, Zaki (2014) articulated an account
of patterns of approach and avoidance motivation that cause people to
seek or shun opportunities to experience empathy. He articulated three
major phenomena that motivate people to approach empathy (namely,
positive affect, affiliation, and social desirability) and three that motivate
people to avoid it (suffering, material costs, and interference with compe-
tition). A good deal of Zaki’s analysis could be applied to understanding
how particular readers make decisions about the narratives with which to
engage as well as their responses to those narratives. For example, Zaki
observed that empathy is often unpleasant, because it requires people to
engage with others’ distress. That potential for ‘suffering’ is a type of avoid-
ance motivation. However, people’s need to feel connected (i.e. the need
for affiliation) prompts people to take on that suffering. The question then
becomes, when does that need for affiliation trump the avoidance of (per-
sonal) pain? Literary theorists have considered questions of this sort. Keen
(2007, p. 19), for example, described circumstances in which readers disen-
gage because the emotional costs of empathy are too high.
To see how these considerations may help account for variability in read-
ers’ participatory responses, we once again consider events from The Slippage
(Greenman, 2013). While attending a trade show in Chicago, William has
a one-​night fling with a woman named Emma. About a year later, by some
major quirk of fate, Emma and her husband buy a house across the street
from William and Louisa. There is clear potential for the fling to turn into
a more extended affair. We imagine that readers will have a diversity of

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 249 ]
 

responses to this possibility as a function, in part, of perceptions of their

own similarity to William and Emma. The Slippage narrates unhappiness in
both William’s and Emma’s marriages. For some readers, who perhaps have
had unfortunate relationships, this unhappiness will provide a context for
empathy, and allow tacit approval for an affair. Others will likely disap-
prove of the affair, no matter what the circumstances. Among that group,
some will likely have moral objections (i.e. affairs are unethical) whereas
others might have pragmatic objections (i.e. the lovers will likely be caught,
with disastrous consequences). As we suggested earlier, different readers
will encode participatory responses consistent with their preferences (e.g.
‘You deserve this!’ or ‘Bad idea!’).
In the event, William and Emma do undertake an affair, including a
tryst in the art studio of William’s brother-​in-​law, Tom. After the affair has
ended, Tom makes the correct inference that William has behaved inap-
propriately in his studio. Tom’s proximal realization is that William has
intruded into Tom’s private cache of art works, but that leads to a greater
truth: ‘How does a man come to be in a place like this, looking at things he
shouldn’t?’ He paused, as if at the top of a hill, and then started down it.
‘And not looking at them alone, either’ (p. 233).
We imagine that readers would assimilate this discovery quite dif-
ferently as a function of their prior commitments (Jacovina and Gerrig,
2010). For example, those who disapproved for moral reasons might expe-
rience pleasure at William’s anticipated distress (e.g. they might hear them-
selves thinking ‘I told you so!’). Those who approved might experience guilt
because they didn’t do enough to warn William away from his actions (see
Gerrig and Prentice, 1996).
In summary, we suggest that readers encode types of mental responses
that parallel those they would encode were they real participants in events.
We have illustrated how readers’ life events will influence their perceptions
of their similarity to characters by determining, for example, which attri-
butes are most salient. We have followed other theorists in accepting the
conclusion that similarity affects empathy (e.g. Keen, 2006, 2007; Hogan,
2011; see also Komeda, Tsunemi, Inohara, Kusumi, and Rapp, 2013). Our
extension is to cite Zaki’s (2014) larger theory of forces that change the
probability that readers will approach or avoid empathy. Finally, all these
forces help determine the content of readers’ participatory responses and,
therefore, the idiosyncratic properties of their narrative experiences. To be
clear, not all participatory responses require empathy. Readers may be able
to develop an understanding of villains’ emotional states and still diligently
root against them. We have focused on similarity and empathy in large part
because they are so well established in theories of readers’ responses.

[ 250 ]  Literature and Cognition in Cognitive Science

 

The cognitive processes we have described suggest why readers’ experi-

ences may depart from those that authors intended. Research in narratol-
ogy has identified possible causal relationships between textual features
and readers’ experiences (e.g. Herman, 2002, 2013; Fludernik, 2009). For
example, The Slippage is focalized through William’s perspective such that,
all things being equal, the novel would appear to make the case that his
affair is justified. Our purpose has been to describe the forces that make
matters unequal. Ultimately, it will be important to understand the bal-
ance between narrators (i.e. how features of narration prompt overlapping
experiences) and readers (i.e. how features of readers prompt individual-
ized experiences). For example, we speculated earlier (in the context of
Louisa’s disappearance from the party) that readers’ experiences might
overlap more as a narrative unfolds. We wonder whether that claim is cor-
rect, and what other claims we could make to capture the limits on the
variability of readers’ experiences.
In the final section of the chapter, we turn to another variable in read-
ers’ experiences: their judgements of their immersion in narrative worlds.

TRANSPORTATION

Our analyses have focused largely on the details of readers’ moment-​by-​

moment experiences:  the ways in which their collections of memories
inform their response to narrative information. However, we suggest that
the same processes also influence the extent to which readers find them-
selves immersed in particular narrative worlds (Gerrig, 1993). The concept
of immersion has appeared in various guises in literary and psychologi-
cal theory, including transportation (Gerrig, 1993; Green and Brock, 2000),
narrative engagement (Busselle and Bilandzic, 2009), and story world absorp-
tion (Kuijpers, Hakemulder, Tan, and Doicaru, 2014). Here, we will use the
term transportation.
To address the claim that individual readers have different experiences,
Melanie Green and Timothy Brock (2000) developed the transportation
scale. They intended their scale to capture various aspects of transporta-
tion: ‘emotional involvement in the story, cognitive attention to the story,
feelings of suspense, lack of awareness of surroundings, and mental imag-
ery’ (p.  703). The transportation scale asks readers to answer questions
such as ‘While I was reading the narrative, I could easily picture the events
in it taking place’ and ‘I was mentally involved in the narrative while reading
it’ (p. 704). Participants respond on seven-​point scales that are anchored
by ‘not at all’ and ‘very much’. In their original research, Green and Brock

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 251 ]
 

demonstrated that self-​reports of transportation were positively corre-

lated with the extent to which readers’ beliefs were changed in response
to narrative information. The more they were transported, the more read-
ers accepted the narrative’s conclusions. In fact, research using self-​report
scales has yielded consistent evidence that greater transportation is associ-
ated with greater belief and attitude change as well as greater enjoyment
(e.g. Green, Brock, and Kaufman, 2004; Appel and Richter, 2010; Tal-​Or
and Cohen, 2010; Murphy, Frank, Moran, and Patnoe-​Woodley, 2011; van
Laer, de Ruyter, Visconti, and Wetzels, 2014). These data support the claim
that readers’ self-​reports provide valid information about the qualities and
consequences of their narrative experiences.
We wish to describe how the factors we have explored could explain
individual differences in readers’ self-​reports of transportation. To do so,
we will review empirical research on factors that affect transportation.
For example, Green (2004) had participants read a story that was about
a gay man attending a fraternity reunion. Readers who knew gay people
or had familiarity with fraternities reported higher levels of transporta-
tion. Green conjectured that readers with greater familiarity may have
been able to produce more vivid mental images of the situations. She also
suggested that familiarity could lead to greater motivation to engage with
the narrative. Finally, she argued that familiarity might provide more con-
tact with readers’ own memories. These theoretical constructs are consis-
tent with the perspective on individual differences that we have developed
here. However, there may be circumstances in which familiarity leads some
readers to become alienated from a text or a character. For example, when
Gabriel in A Nearly Perfect Copy (Amend, 2013) maligns Damien Hirst, read-
ers’ familiarity with Hirst may, as we suggested earlier, change their assess-
ments of similarity to Gabriel and, at the same time, affect the extent of
their immersion in the text. Thus, we agree that familiarity likely has an
impact on readers’ experiences of having been transported. We only won-
der whether that relationship is positive for all readers.
Readers’ reports of transportation are also influenced by their responses
to particular characters. The fourth season of the television programme
Desperate Housewives features a six-​episode arc in which a central charac-
ter develops lymphoma. In an experiment involving women who regularly
watched the series, participants reported their involvement with the char-
acter by providing ratings of ‘how much they liked, how similar they were
to, how much they felt like they knew, and how much they would like to
be like’ the character (Murphy, Frank, Moran, and Patnoe-​Woodley, 2011,
p.  416; see also Murphy, Frank, Chatterjee, and Baezconde-​Garbanati,
2013). The data indicated that greater involvement with the character (with

[ 252 ]  Literature and Cognition in Cognitive Science

 

respect to a measure combining the four questions) led to greater transpor-

tation. Ellis Van den Hende and colleagues (2012) also demonstrated that
readers were more transported when they perceived the main character to
be similar to them. These results highlight the importance of readers’ per-
ceived similarity to characters when it comes to their experiences of trans-
portation. Thus, factors that affect readers’ tacit judgements of similarity
are likely to explain some individual variation in transportation.
With respect to narrative events, suspense appears to have a positive
impact on transportation (Tal-​Or and Cohen, 2010; Krakowiak and Oliver,
2012; Bezdek and Gerrig, in press). For example, Tal-​Or and Cohen (2010)
had participants view a film in which a man faced the temptation to cheat
on his wife. They manipulated what they called ‘time of deeds’ and ‘char-
acter valence’. In the past condition, participants were told the protagonist
was either loyal or a serial cheater. In the future condition, participants
were told that the protagonist would eventually cheat on his wife or that
he would stay faithful. They found the future conditions caused higher
transportation than the past ones. They suggested that the future fram-
ing increases suspense, which more solidly transports readers. We note,
however, that readers will experience suspense more or less intensely as a
function of, for example, similarity and empathy. Andrew Ortony, Gerald
Clore, and Allan Collins (1988) argued that suspense requires ‘a Hope emo-
tion and a Fear emotion’ (p. 131) in the context of uncertainty between two
(or more) outcomes. Recall our discussion of how readers might respond to
the possibility, in The Slippage (Greenman, 2013), that William’s affair will
be revealed. For some readers, this will be a low-​stakes outcome, yielding
very little experience of hope or fear. Other readers may experience higher
stakes, but those readers may differ with respect to which outcome inspires
hope and which outcome fear.
Basic cognitive processes should also help explain why readers’ experi-
ences will change over time. Accumulations of new memories will affect
which content feels familiar, which characters appear similar, and which
outcomes inspire hope and fear. As they grow older, readers may undergo
radical shifts in the aspects of texts that successfully transport them to
narrative worlds.

CONCLUSION

We began this chapter by citing a scene from The Slippage (Greenman,

2013) in which William and Christopher endeavour to fly a kite. We chose
that scene because we believed it allowed room for readers to respond

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 253 ]
 

differently, as a product of their own individual memories of kites, parks,

childhood, and adult authority. We declared as uncontroversial the claim
that individual memories matter. We have attempted to bolster that claim
by describing basic memory processes that act in a largely automatic fash-
ion to individuate readers’ experiences. We have seen how those processes
affect fundamental responses such as readers’ judgements of the normality
of characters’ behaviours and their similarity to those characters. Readers’
judgement of similarity, in turn, can affect the empathy they feel towards
particular characters. We have argued, in addition, that readers’ familiarity
with aspects of narrative worlds as well as, again, their similarity to par-
ticular characters will help determine the extent to which they are trans-
ported into those worlds. Thus, the specifics of readers’ lives help structure
their participation in narrative worlds to yield unique experiences.

REFERENCES

Alabastro, A., Rast III, D. E., Lac, A., Hogg, M. A., and Crano, W. D. (2013). Intergroup
bias and perceived similarity:  Effects of success and failures on support for
in-​and outgroup political leaders. Group Processes and Intergroup Relations,
16, 58–​67.
Allbritton, D. W., and Gerrig, R. J. (1991). Participatory responses in prose understand-
ing. Journal of Memory and Language, 30, 603–​626.
Amend, A. (2013). A nearly perfect copy. New York: Nan A. Talese.
Appel, M., and Richter, T. (2010). Transportation and need for affect in narrative per-
suasion: A mediated moderation model. Media Psychology, 13, 101–​135.
Bezdek, M. A., Foy, J. E., and Gerrig, R. J. (2013). ‘Run for it!’: Viewers’ participatory
responses to film narratives. Psychology of Aesthetics, Creativity, and the Arts, 7,
409–​416.
Bezdek, M. A., and Gerrig, R. J. (in press). When narrative transportation narrows
attention: Changes in attentional focus during suspenseful film viewing. Media
Psychology.
Brock, T. C. (1965). Communicator-​recipient similarity and decision change. Journal of
Personality and Social Psychology, 1, 650–​654.
Busselle, R., and Bilandzic, H. (2009). Measuring narrative engagement. Media
Psychology, 12, 321–​347.
Cook, A. E., and O’Brien, E. J. (2015). Passive activation and instantiation of inferences
during reading. In E. J. O’Brien, A. E. Cook, and R. F. Lorch (Eds.), Inferences dur-
ing reading (pp. 19–​41). New York: Cambridge University Press.
Filik, R. (2008). Contextual override of pragmatic anomalies: Evidence from eye move-
ments. Cognition, 106, 1038–​1046.
Filik, R., and Leuthold, H. (2008). Processing local pragmatic anomalies in fictional
contexts: Evidence from the N400. Psychophysiology, 45, 554–​558.
Filik, R., and Leuthold, H. (2013). The role of character-​based knowledge in online nar-
rative comprehension: Evidence from eye movements and ERPs. Brain Research,
1506, 94–​104.

[ 254 ]  Literature and Cognition in Cognitive Science

 

Fincher-​Kiefer, R., Post, T. A., Greene, T. R., and Voss, J. F. (1988). On the role of prior
knowledge and task demands in the processing of text. Journal of Memory and
Language, 27, 416–​428.
Fludernik, M. (2009). An introduction to narratology. London: Routledge.
Foy, J. E., and Gerrig, R. J. (2014). Flying to Neverland: How readers tacitly judge
norms during comprehension. Memory & Cognition, 42, 1250–​1259.
Gerrig, R. J. (1993). Experiencing narrative worlds. New Haven, CT: Yale University Press.
Gerrig, R. J., and Jacovina, M. E. (2009). Reader participation in the experience of nar-
rative. In B. H. Ross (Ed.), The psychology of learning and motivation (Vol. 51, pp.
223–​254). Burlington, MA: Academic Press.
Gerrig, R. J., and O’Brien, E. J. (2005). The scope of memory-​based processing.
Discourse Processes, 39, 225–​242.
Gerrig, R. J., and Prentice, D. A. (1996). Notes on audience response. In D. Bordwell
and N. Carroll (Eds.), Post-​theory:  Reconstructing film studies (pp. 388–​403).
Madison, WI: University of Wisconsin Press.
Gerrig, R. J., and Wenzel, W. G. (2015). The role of inferences in narrative experiences.
In E. J. O’Brien, A. E. Cook, and R. F. Lorch (Eds.), Inferences during reading (pp.
362–​385). New York: Cambridge University Press.
Green, M. C. (2004). Transportation into narrative worlds: The role of prior knowledge
and perceived realism. Discourse Processes, 38, 247–​266.
Green, M. C., and Brock, T. C. (2000). The role of transportation in the persuasiveness
of public narratives. Journal of Personality and Social Psychology, 79, 701–​721.
Green, M. C., Brock, T. C., and Kaufman, G. F. (2004). Understanding media enjoyment: The
role of transportation into narrative worlds. Communication Theory, 14, 311–​327.
Greenman, B. (2013). The slippage. New York: Harper Perennial.
Griffin, T. D., Jee, B. D., and Wiley, J. (2009). The effects of domain knowledge on meta-
comprehension accuracy. Memory & Cognition, 37, 1001–​1013.
Herman, D. (2002). Story logic:  Problems and possibilities of narrative. Lincoln, NE:
University of Nebraska Press.
Herman, D. (2013). Storytelling and the sciences of mind. Cambridge, MA: MIT Press.
Hogan, P. C. (2011). Affective narratology. Lincoln, NE: University of Nebraska Press.
Jacovina, M. E., and Gerrig, R. J. (2010). How readers experience characters’ decisions.
Memory & Cognition, 38, 753–​761.
Kahneman, D., and Miller, D. T. (1986). Norm theory: Comparing reality to its alterna-
tives. Psychological Review, 93, 136–​153.
Keen, S. (2006). A theory of narrative empathy. Narrative, 14, 207–​236.
Keen, S. (2007). Empathy and the novel. New York: Oxford University Press.
Kintsch, W. (1988). The role of knowledge in discourse comprehension: A construction-​
integration model. Psychological Review, 95, 163–​182.
Kintsch, W. (1998). Comprehension:  A  paradigm for cognition. New  York:  Cambridge
University Press.
Komeda, H., Tsunemi, K., Inohara, K., Kusumi, T., and Rapp, D. N. (2013). Beyond
disposition: The processing consequences of explicit and implicit invocations of
empathy. Acta Psychologica, 142, 349–​355.
Krakowiak, K. M., and Oliver, M. B. (2012). When good characters do bad things:
Examining the effect of moral ambiguity on enjoyment. Journal of Communication,
62, 117–​135.
Kuijpers, M. M., Hakemulder, F., Tan, E. S., and Doicaru, M. M. (2014). Exploring
absorbing reading experiences: Developing and validating a self-​report scale to
measure story world absorption. Scientific Study of Literature, 4, 89–​122.

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 255 ]
 

MacInnis, C. C., and Page-​Gould, E. (2015). How can intergroup interaction be bad
if intergroup contact is good? Exploring and reconciling an apparent paradox
in the science of intergroup relations. Perspectives on Psychological Science, 10,
307–​327.
McKoon, G., and Ratcliff, R. (1992). Inference during reading. Psychological Review, 99,
440–​466.
McKoon, G., and Ratcliff, R. (2015). Cognitive theories in discourse-​ processing
research. In E. J. O’Brien, A. E. Cook, and R. F. Lorch (Eds.), Inferences during
reading (pp. 42–​67). New York: Cambridge University Press.
McNamara, D. S., and Magliano, J. (2009). Toward a comprehensive model of compre-
hension. In B. H. Ross (Ed.), The psychology of learning and motivation (Vol. 51, pp.
298–​384). Burlington, MA: Academic.
Meyer, D. E., and Schvaneveldt, R. W. (1971). Facilitation in recognizing pairs of
words:  Evidence of a dependence between retrieval operations. Journal of
Experimental Psychology, 90, 227–​234.
Mitchell, J. P., Macrae, C. N., and Banaji, M. R. (2006). Dissociable medial prefron-
tal contributions to judgments of similar and dissimilar others. Neuron, 50,
655–​663.
Murphy, S. T., Frank, L. B., Chatterjee, J. S., and Baezconde-​Garbanati, L. (2013).
Narrative versus nonnarrative:  The role of identification, transportation, and
emotion in reducing health disparities. Journal of Communication, 63, 116–​137.
Murphy, S. T., Frank, L. B., Moran, M. B., and Patnoe-​Woodley, P. (2011). Involved,
transported, or emotional? Exploring the determinants of change in knowledge,
attitudes, and behavior in entertainment-​education. Journal of Communication,
61, 407–​431.
Myers, J. L., and O’Brien, E. J. (1998). Accessing the discourse representation during
reading. Discourse Processes, 26, 131–​157.
Nieuwland, M. S., and van Berkum, J. J. A. (2006). When peanuts fall in love: N400
evidence for the power of discourse. Journal of Cognitive Neuroscience, 18,
1098–​1111.
O’Brien, E. J., and Myers, J. L. (1999). Text comprehension: A view from the bottom
up. In S. R. Goldman, A. C. Graesser, and P. van den Broek (Eds.), Narrative com-
prehension, causality, and coherence: Essays in honor of Tom Trabasso (pp. 35–​53).
Mahwah, NJ: Lawrence Erlbaum.
Ortony, A., Clore, G. L., and Collins, A. (1988). The cognitive structure of emotions.
Cambridge: Cambridge University Press.
Rapp, D. N., and Gerrig, R. J. (2002). Readers’ reality-​driven and plot-​driven analyses
in narrative comprehension. Memory & Cognition, 30, 779–​788.
Rapp, D. N., and Gerrig, R. J. (2006). Predilections for narrative outcomes: The impact
of story contexts and reader preferences. Journal of Memory and Language,
54, 54–​67.
Rieger, C. J. (1975). Conceptual memory and inference. In R. C. Schank (Ed.), Conceptual
information processing (pp. 157–​288). New York: Elsevier.
Shkurko, A. V. (2015). Cognitive mechanisms of ingroup/​outgroup distinction. Journal
for the Theory of Social Behaviour, 45, 188–​213.
Spilich, G. J., Vesonder, G. T., Chiesi, H. L., and Voss, J. F. (1979). Text processing of
domain-​related information for individuals with high and low domain knowl-
edge. Journal of Verbal Learning and Verbal Behavior, 18, 275–​290.
Tal-​Or, N., and Cohen, J. (2010). Understanding audience involvement: Conceptualizing
and manipulating identification and transportation. Poetics, 38, 402–​418.

[ 256 ]  Literature and Cognition in Cognitive Science

 

Turner, J. C., and Reynolds, K. J. (2003). The social identity perspective in intergroup
relations: Theories, themes, and controversies. In R. Brown and S. L. Gaertner
(Eds.), Blackwell handbook of social psychology: Intergroup processes (pp. 133–​152).
Malden, MA: Blackwell Publishing.
Tversky, A. (1977). Features of similarity. Psychological Review, 84, 327–​352.
van den Hende, E. A., Dahl, D. W., Schoormans, J. P.  L., and Snelders, D. (2012).
Narrative transportation in concept tests for really new products:  The mod-
erating effect of reader–​protagonist similarity. Journal of Product Innovation
Management, 29 (S1), 157–​170.
van Laer, T., Ruyter, K. D., Viconti, L. M., and Wetzels, M. (2014). The extended
transportation-​imagery model: A meta-​analysis of the antecedents and conse-
quences of consumers’ narrative transportation. Journal of Consumer Research,
40, 797–​817.
Wecker, H. (2013). The golem and the jinni. New York: Harper Collins.
Wenzel, W. G., and Gerrig, R. J. (2015). Convergent and divergent thinking in the con-
text of narrative mysteries. Discourse Processes, 52, 489–​516.
West, T. V., Magee, J. C., Gordon, S. H., and Gullett, L. (2014). A little similarity goes
a long way: The effects of peripheral but self-​revealing similarities on improv-
ing and sustaining interracial relationships. Journal of Personality and Social
Psychology, 107, 81–​100.
Wolf, W. (2009). ‘Illusion (aesthetic).’ In P. Hühn, J. Pier, W. Schmid, and J. Schönert
(Eds.), Handbook of narratology (pp. 144–​159). Berlin: Walter de Gruyter.
Zaki, J. (2014). Empathy: A motivated account. Psychological Bulletin, 140, 1608–​1647.

R e a de r s ’ L i v e s a n d N a r r at i v e E x p e r i e n c e s  [ 257 ]
 
 

CHAPTER 13

On Truth and Fiction

KEI T H OATLEY

DISTRUST OF FICTION

In The Republic, Plato (375 b.c.e./​1955) told a story of how we humans

are like prisoners chained to a bench in a cave, able to see only shadows
on a wall. These shadows are of people walking back and forth behind us,
between our backs and a fire. Our task in life is to struggle up out of the
cave, to reach the light by which we can see the truth.
Plato thought poetry was mimesis, which for him meant ‘copying’ or
‘representation’. To copy is to produce a shadow of a shadow. Plato devotes
Part 10 of The Republic to arguing that, for this reason, works of literary art-
ists are of ‘no serious value’ (602b). This was a radical move because, for the
Greeks of his time, Homer’s epics were secular scripture, a foundation of
society (Powell, 2002). Plato proposed a system in which truth exists only
on the plane of unchanging ideals of the kind embodied in Pythagoras’s
theorem, as he depicts in Meno (402 b.c.e./​1956). In his ideal society, Plato
said that poetry (fiction) would be banned because it invites us to fall
‘under the spell of a childish and vulgar passion’ (608a).
To illustrate his idea that our everyday relation to the truth is distorted,
Plato offered the story of shadows in the cave. To teach his system of ide-
als, he wrote stories in which the long-​dead Socrates showed people how
they went wrong in their thinking. Plato did not turn his critical eye on
himself; his use of fictional stories to convey his teachings contradicted his
own precepts. Despite this, a residue of his idealization has remained, and
his contempt of poetry, as well as of fiction other than his own, has melded
with a certain amount of popular distrust.
 

As it derives from Greek, the word ‘poetry’ means ‘something made’, and
‘fiction’, which derives from Latin, means the same. With an almost unno-
ticed segue, poetry and fiction have become not just something made, but
something made up. Why should we take notice of such stuff? In English
we hear such phrases as ‘fact or fiction’. Of course one knows the problem.
In the newspaper one wants to read what actually happened, not some-
thing that someone has made up. But a corollary has formed in the minds
of some that fiction is of little or no value, so that now some philosophers
have described ‘the paradox of fiction’.
The paradox of fiction was discussed by Colin Radford (1975). It goes like
this. How is it that we experience emotions as we read novels and watch
plays or films when we know the characters in them are not real? Radford
says it makes sense to feel moved by a friend who is in pain, but it does not
make sense if the friend is putting it on. In the same way we might properly
be moved by historical accounts of the sufferings of real people, but

what seems unintelligible is how we could have similar reaction to the fate of
Anna Karenina, the plight of Madame Bovary, or the death of Mercutio. Yet we
do. We weep, we pity Anna Karenina, we blink hard when Mercutio is dying and
wish that he had not been so impetuous. (p. 69)

Radford discusses how we might think about the paradox: perhaps we are

behaving like children, perhaps we suspend disbelief, perhaps we imag-
ine the same kind of thing happening to someone we love. He ends up by
stating that although it may seem natural to be moved by works of art, it
‘involves us in inconsistency and so incoherence’ (p. 78).
In his treatment of the paradox of fiction, Kendall Walton (1990) has
argued that if a story moves us, the emotions we feel are not those of real
life. He illustrates this with a vignette of Charles, who watches ‘a horror
movie about a terrible green slime … afterwards he confesses that he was
“terrified” of the slime’ (p.  196). Walton says that Charles was suffering
from ‘quasi-​fear’. He explains that ‘Charles does not imagine merely that
he is afraid; he imagines being afraid’ (p.  247). In a (1997) follow-​up, he
says that in fiction ‘we need to recognize that we imagine feeling fear, pity,
and admiration—​fear, pity and admiration of the kinds we might actually
feel in “real life” ’ (p. 48). Walton is right, of course, to argue that imagina-
tion is involved, but to call the emotions of engagement in fiction ‘quasi-​
emotions’ seems to imply that there is something wrong with them.
Gregory Currie (2011) doubts the value of fiction. He says, ‘When we
engage with great literature we do not come away with more knowledge,
clarified emotions, or deeper human sympathies’ (p. 15); the only writings

[ 260 ]  Literature and Cognition in Cognitive Science

 

we should take seriously in understanding others or ourselves are scientific

psychology and neuroscience. If you must read Middlemarch, he says, you
should do so ‘only when Nature Neuroscience is ready as an antidote’ (p. 15).
Fiction, he says, is mere pretence. Although Currie is right to point out
that empirical work in psychology and neuroscience is informative about
human life, he fails to see that artistic fiction enables us, in ways that indi-
vidual empirical findings do not, to think for ourselves about interactions
among knowledge of the mind, emotions, sympathy, and so on.
I wonder whether those who think that fiction is merely made up, or
paradoxical, or pretence, are gazing at shadows like the ones on the wall
of Plato’s cave. What is needed is a psychological theory of truth in fiction,
and evidence that bears on it.

FOUR BASES OF A PSYCHOLOGY OF FICTION

I propose four bases for a psychology of fiction: a theoretical base that fic-
tion is mental simulation, a developmental base of how fiction derives from
childhood play, an evidential base of effects of reading fiction on people’s
understandings of others, and a second kind of evidential base about the
kinds of changes that can occur to one’s sense of self from engaging with
fiction.

A Theoretical Base: Simulation

We humans are members of a future-​orientated social species, dependent

on plans made jointly with others and on understandings of those others
(Dunbar, 2004). Fiction is about thinking and feeling beyond the imme-
diate, into worlds of the possible (Bruner, 1986; Gavins, 2007). Fiction
is not best characterized as description. It is better thought of as a set
of simulations of selves and their interactions in a range of social cir-
cumstances (Oatley, 1992, 1999). It is about inward truth, of others and
oneself.
In the last 50 years, the cognitive approach to psychology has replaced
the emptiness of behaviourism. It is about the mind: about how conscious
and unconscious knowledge is organized and used to perceive, remember,
think, and converse, as well as to experience emotions and other mental
states. Recently a cognitive science of fiction has developed (Oatley, 2012).
Important for this development is understanding how we humans can
know what we ourselves, and what other people, know and feel.

O n T r u t h a n d Fi c t i o n   [ 261 ]
 

If fiction were principally description, it would fail tests on which psy-

chologists insist—​tests of reliability and validity—​and be of dubious value.
But fiction is not primarily description. Cognitive science (including cogni-
tive psychology) has been advanced by the idea of simulation. Fiction is a
kind of simulation: one that runs not on computers but on minds (Oatley,
2011). It is possible, indeed, that it was the very first kind of simulation.
Stories, told orally, were simulations and it is likely that they have run on
minds for tens of thousands of years. In this chapter, I use the term fic-
tion to mean both oral and written stories, as well as plays and movies.
Melanie Green and colleagues (2008) have found that although the effects
are not all the same, engagement in stories occurs in a somewhat similar
way across different media.
A working idea of simulation can be gained from modern weather fore-
casts, which are outputs of simulations that run on computers. We humans
are good at understanding mechanisms one at a time. We can understand,
for instance, that when a mass of cold air meets a mass of warm air, the cold
air cools the warm air so that water vapour tends to condense. This conden-
sation then falls as rain or snow or hail. We are far less good at thinking in
complexes, for instance about how such cooling of warm air interacts with
barometric pressure, with winds, with the influence of mountains and seas,
and so on. Simulations allow us to put processes that we understand well,
one at a time, together into complexes, so that we can explore, and better
understand interactions of these processes.
Similarly in social life, we can know that if Abigail is angry with Beatrice,
she may try and get even with her. Many people will have this intuition.
But what if Beatrice is Abigail’s three-​year-​old daughter? What if she is
Abigail’s lover? What if Beatrice acted as she did to jog Abigail out of a cer-
tain complacency? Fiction is a set of simulations that enable us to explore
complexes of these kinds.
Clearly in any simulation there are descriptive elements, and one wants
facts to be right. For instance, in simulations of the weather one needs
individual mechanisms to be right, so that, for instance, the cooling of
warm air does produce precipitation if the warm air is saturated with water
vapour. To make a forecast, one also needs to know current facts about air
temperature, wind force and direction, and so on, from weather stations.
The issues are similar in fiction. When facts or processes are wrong in fic-
tion, the narrative mode can make them too easy to accept (Gerrig, 1993).
Thus, Deborah Prentice, Richard Gerrig, and Daniel Bailis (1997) asked stu-
dents to read narratives that contained weak or unsupported assertions
about either their home school or a school they did not know. Readers’
beliefs on unsupported assertions about their home school did not change,

[ 262 ]  Literature and Cognition in Cognitive Science

 

but they did change about the away school. Serious writers of fiction usu-
ally do a lot of research on their subject matter. If they get things wrong, if
they write in a way that is unsupported, or if their work contains untruths
or propaganda, it is damaged.
I suppose one might say of a weather forecast: ‘This isn’t weather, it’s
just quasi-​weather’, but would it help our understanding to do so? Should
we say:  ‘Because its processes are not real weather, it is incoherent to
take any notice of a forecast?’ Principal questions for a simulation-​based
weather forecast are:  Does it clarify outcomes of interactions among its
components, and does it produce better predictions than those from single
observations such as, ‘red sky at night, shepherd’s delight’? In the case of a
fictional literary simulation, we can ask: Does it run well, does it resonate
with individual readers’ understandings of the social world, does it enable
us to experience emotions and insights about selves and their interactions?
Inner truths about others and oneself are as important as correspondences
of beliefs with events of the physical world (Oatley, 1999).
When he used the term mimesis (for instance in his discussion of art in
Part 10 of The Republic, 595c) Plato meant ‘imitation’, or ‘representation’,
and these are among the usual English translations. Stephen Halliwell
(2002) has shown, however, that in the time of Plato and Aristotle mimesis
also had a second family of meanings, which is less widely known, and sel-
dom appears in translations. This second set of meanings relates to mimesis
as ‘world-​making’, or ‘model-​building’, or—​in modern terms—​‘simula-
tion’. As Halliwell puts it:

Reduced to a schematic but nonetheless instructive dichotomy, these varieties

of mimetic theory and attitude can be described as encapsulating a difference
between a ‘world-​reflecting’ model (for which the mirror has been a common
though far from straightforward metaphorical emblem), and, on the other side,
a ‘world-​simulating’ or ‘world-​creating’ conception of artistic representation.
(p. 23)

If one reads Aristotle’s (330 b.c.e./​1970) Poetics with this in mind, and
substitutes ‘simulation’ for its usual translations of ‘imitation’ and ‘copy-
ing’, it becomes clear that it is mimesis as simulation with which the Poetics
is principally concerned. Although narrative does indeed have world-​
reflecting purposes, arguably this second sense, of world-​creation, is more
important (Oatley, 1999). Simulations do require a mental leap from the
world of day-​to-​day experience to that of created, imagined, worlds and
the emotions we might experience in them, but there is nothing paradoxi-
cal about that leap.

O n T r u t h a n d Fi c t i o n   [ 263 ]
 

Knowing others’ minds is central to human life. But because what goes
on in these minds is not immediately visible, we make inferences. We are
good at inferring social truths but, because the social world is complex, we
are not that good. We can always improve (Nickerson, 1999), and fiction is
a means by which we can do so to develop our understanding of others, an
ability that has become known technically as ‘theory of mind’ (Astington,
Harris, and Olson, 1988). Most experiments on theory-​of-​mind are about
inferring what another person may know at a certain time, but the issue
is deeper. The whole of social life—​the means by which we can cooperate
with others or compete with them and by which we feel for them or against
them—​is based on making mental models of their character: their proclivi-
ties and concerns, their abilities and trustworthiness. As Robin Dunbar
(2004) has shown, the ability to make and remember such models is the
main reason why we humans have such large brains.
In detective stories a protagonist tries to solve a crime by inferring what
various suspects know while they try to keep their knowledge secret. In
Georges Simenon’s detective novels, for instance, Maigret concentrates on
conversations with the story’s characters to make theory-​of-​mind infer-
ences, while being often perfunctory about police procedure (e.g. Simenon,
1971). Fiction is read for enjoyment, sometimes for escape, sometimes
for other reasons, but Lisa Zunshine (2006) has argued that it is read to
exercise our theory-​of-​mind ability. We are good at using this ability and,
she says, we like doing what we are good at. More generally, the idea of
understanding other minds as being central to fiction has become impor-
tant in literary theory (e.g. Lauer, 2009; Leverage, Mancing, Schweickert,
and William, 2011).
In social psychology there is an effect called the actor-​observer differ-
ence (Jones and Nisbett, 1971). If we see someone walk across a room and
trip over a toy that has been left on the floor, we might think that person
was clumsy. If we were to do the same thing, we would experience ourselves
as affected by the situation and say the toy shouldn’t have been left there.
In the first case, we act as an observer and make a personality judgement
about the other person. In the second case, we experience ourself as an
actor, in relation to circumstances of the world. A critical component of the
understanding of others that derives from fiction is that rather than simply
making observer-​based personality judgements about others from their
behaviour, as we often do in the day-​to-​day world, we can come to know
others from the inside, as actors, in the way we think about ourselves.
A fundamental process here is empathy:  feeling with another person
(e.g. Keen, 2007). We can identify with a protagonist by putting aside our
own concerns and taking on those of the character (Trabasso and Chung,

[ 264 ]  Literature and Cognition in Cognitive Science

 

2004; Oatley, 2012, 2013a). The emotions we thereby feel are not those of
the character; they are emotions of identification and empathy, of our own
self in the character’s situation.
As Alvin Goldman (2006) has explained, empathy has two parts. In day-​
to-​day life, one part is that, by a process of simulation within ourselves,
from the flow of an interaction and understanding the concerns of another,
we infer what emotion another person is feeling and impute it to that per-
son. The second part, which occurs at the same time, is that we feel the
corresponding emotion in our self. Feeling an emotion with a protagonist
in fiction has two similar parts. In one part (within our simulation of the
story), we come to know a character’s concerns and understand what emo-
tions might occur as the character’s intentions are affected by events. In
the second part, having taken on the character’s concerns, we experience
corresponding emotions of our own. So, unlike those who find the emo-
tions of fiction paradoxical in seeming to be about something that is not
real, this explanation is based on a fundamental psychological process
of making mental models of others that is the same in fiction as in the
interactions of real life (Kotovych, Dixon, Bortolussi, and Holden, 2011). It
enables us to feel with others in empathy. The truer our mental models are,
the better (see also Caracciolo, 2014). One might even say that by means of
its analyses of other people’s intentions and their implications, fiction may
augment everyday processes of empathy. A useful piece of evidence here is
that Raymond Mar (2011) has shown that several parts of the brain that
are used to comprehend stories are the same as those used to understand
other people.
A consequence of thinking of fiction as simulation is that from it we
can learn how the social world works from a variety of perspectives, in a
variety of circumstances. If we were learning to fly a plane, we would do
well to spend time in a flight simulator. Fiction is the social mind’s flight
simulator.

The Developmental Base: Imaginative Play

While it is valuable, as Walton (1990) has pointed out, to think of fiction as

having certain properties of games, his discussion of Charles at the horror
movie includes this: ‘In many ways Charles is like a child … playing a game
of make-​believe’ (p.  242). Although Walton (personal communication,
20 March 2013) has said that he uses ‘make-​believe’ as a technical term,
the above quotation seems to imply that fiction is a regression to a childish
state. Fiction is not a regression. It is a development of childhood play, a

O n T r u t h a n d Fi c t i o n   [ 265 ]
 

development that is central to our ability to create the abstractions of art

and science.
Culture (including language, art, and science) depends on abstractions
that derive from the imagination (Harris, 2000; Mar and Oatley, 2008;
Oatley and Olson, 2012). Paul Harris and his colleagues (Dias, Roazzi, and
Harris, 2005) followed up Alexander Luria’s (1976) studies carried out in
the 1930s, when the USSR was beginning to introduce literacy to remote
communities. Luria found that people in Uzbekistan who had not attended
any of the new reading classes were not able to answer questions based on
syllogisms like: ‘In the Far North, where there is snow, all bears are white.
Novaya Zemlya is in the Far North. What colour are the bears there?’ These
respondents were literal-​minded. They answered by saying such things as
that they did not know what bears were like in Novaya Zemlya because they
had never been there. By contrast, Luria found that people who had begun
classes in literacy could reason with such syllogisms; they could make infer-
ences from what the words meant.
In thinking about this result, Harris considered that the literacy pro-
grammes that Luria’s respondents had attended in Uzbekistan were very
elementary, so their ability to answer syllogisms could not have been from
a wide base of knowledge. He wondered whether one of the effects of the
programmes had been to introduce people to the idea of using their imagi-
nation to think about things they did not know directly. Taking this idea
forward, Dias, Roazzi, and Harris report a study from Recive in Brazil,
where interviews were conducted with 24 people who had been to literacy
classes two or three times a week for two years, and 24 people who had not
attended any classes.
Dias and colleagues constructed some syllogisms which started with gen-
eralizations that would be unfamiliar to participants, such as ‘All leucocytes
are white’, and some with generalizations that would be familiar, such as
‘All blood is red’. Half the participants who were interviewed were assigned
to one group and tested in the way Luria had tested his participants. To
these people, the researchers said: ‘I am going to read you some little sto-
ries about things that will sound funny. But let’s pretend that everything
in the stories is true. Okay, now I’m going to tell you the first story …’
The other half of the participants were assigned to a second group, who
were introduced to the problems in a different way. For these people the
instructions were:

I am going to read you some little stories about things that will sound funny. But
let’s pretend that I am telling you all about another planet. Everything in that planet
is different. Okay, now I’m going to tell you the first story about that planet …

[ 266 ]  Literature and Cognition in Cognitive Science

 

For those in the first group, tested in the matter-​of-​fact way that Luria
had used, Dias and colleagues found what Luria found. For instance, when
asked to reason from the premise, ‘All blood is blue’, they generally could
not do so. Instead they tended to give answers from their own experience.
For those in the second group, who were asked to think imaginatively about
another planet, both those who were illiterate and those who had received
some literacy training did better on syllogisms that had either familiar or
unfamiliar content than did those in the first group. Syllogisms require the
imagination, and it was imagination that had been invoked by asking par-
ticipants to think about another planet. It is the imagination that is fos-
tered in play, and in play’s development into fiction.
A number of writers have discussed how fiction is based on child-
hood play (including Freud, 1908, and Huizinga, 1955). More recently,
Brian Boyd (2009) has argued that play is the origin of stories. A piece of
empirical evidence that bears on the question is the finding that profes-
sional writers of fiction are more likely than members of the ordinary
population to have had imaginary playmates when they were children
(Taylor, Hodges, and Kohányi, 2002‒2003). One might be reminded of
the Brontë children—​Charlotte, Branwell, Emily, and Anne—​who devel-
oped and played in the imagined romantic worlds of Angria and Gondal
(e.g. Cox, 2011).
Steven Mithen (1996) has argued that metaphor is at the centre of all
art, and this is a useful way of thinking. In metaphor a ‘this’ is a ‘that’. A set
of paint marks on a canvas is Mona Lisa. I suggest that in literary art we as
readers and audience members take a further metaphorical step. We can be
ourselves and also Anna Karenina. As Anna, we enter a simulated world, we
take on her concerns, and we experience emotions—​not Anna’s emotions
but our own real emotions—​in circumstances that Anna enters.
We can see the beginnings of this in children’s play that involves roles.
Here, for instance, from the opening of Judy Dunn’s (2004) book Children’s
Friendships:  The Beginnings of Intimacy, is a scene in which two boys are
playing in a room with some dressing-​up clothes, some toys, and a table.
The boys have been friends for about a year.

First they are pirates sailing on a search for treasure, then their ship is wrecked,
and they are attacked by sharks; they reach the safety of an island, and build a
house (under the table). What to eat and how to cook it are problems that are
ingeniously solved. Their elaborate adventure, their quickly solved disputes (are
they being attacked by sharks or by crocodiles?), their extended conversations
about what happens next—​all are captured by our video camera in the corner
of the room.

O n T r u t h a n d Fi c t i o n   [ 267 ]
 

In my 2011 book Such Stuff as Dreams, I juxtapose this scene with one in
which my daughter, aged about five, was sitting with three friends to watch
a movie.

‘I’ll be x’, said my daughter, naming one of the characters in the movie, which the
friends had watched before. ‘You can be y’, she said to the girl who sat next to
her. There was some discussion among the four girls, until each had chosen who
would be who. Then they watched the film. (p. 24)

Children can move effortlessly between play and absorption in a story, as

if both are forms of the same activity. The taking of roles in a narratively
structured game of pirates is not very different than the taking of roles in
identifying with characters as one watches a movie. It might be thought
that, as they grow towards adolescence, people give up childhood play, but
this is not so. Instead, the bases and interests of this activity change and
develop to playing and watching sports, to the fiction of plays, novels, and
movies (Oatley, 2013a), and nowadays to video games. In fiction, one can
enter possible worlds. When we experience emotions in such worlds, this is
not a sign that we are being incoherent or regressed. It derives from trying
out metaphorical transformations of our selves in new ways, in new worlds,
in ways that can be moving and important to us. As Donald Winnicott
(1971) has argued, the developments of play become bases of selfhood.
Writers on the paradox of fiction suppose that there is something
wrong either with fiction itself, or about the ways in which we think about
it, because its emotions are not those of real life. As Stevenson argued in
1884, fiction is an abstraction, like mathematics. As Harris (2000) and his
colleagues have shown, play is the imaginative entry into possible worlds,
and this becomes the basis of the abstract thinking of art and science. Like
mathematics, fiction is not the same as real life, but that does not mean it is
untrue or without value. The truths of fiction, and the emotions of fiction,
concern mental models (abstractions) of other people and ourselves, which
are the very centres of our social lives.

Evidence on Understanding Others

In The Republic, Plato says that whereas a craftsman has knowledge (epis-
teme) and skill (techne), ‘the artist knows little or nothing about the subjects
he represents’ (602b). A Platonic commentary now might be that an actor
who plays a doctor in a television series knows nothing about medicine.
This may seem compelling until one thinks that medical series on television

[ 268 ]  Literature and Cognition in Cognitive Science

 

are usually less about medicine than about characters and their emotions
as they enter roles such as doctor, nurse, patient, hospital administrator,
and so on. The actors in the series will have thought deeply about emotion
and entry into roles in social life. Insofar as such a series is about medicine,
one or more doctors will have advised on the episodes, and both actors and
viewers can also learn something about medicine.
Dunbar (2004) has found that the chief content of day-​to-​day conver-
sation is of what goes on as people’s plans and projects are worked out in
interaction with others. As well as developing our relationships, conversa-
tion enables us to enjoy finding out more about ourselves and those we
know. We can nearly always benefit from learning more about how selves
and their emotions can be affected in new situations that we have not yet
thought about, and fiction offers us this kind of possibility.
A colleague (a graduate student at the time) Raymond Mar said: if this
idea of fiction as simulation of the social world is right, then people who
read a lot of fiction might have better understandings of the social world.
(The same would be true of people who live in oral cultures and regu-
larly have stories told to them.) Led by Mar, we did a study (Mar, Oatley,
Hirsh, dela Paz, and Peterson, 2006) in which we measured the amount of
fiction-​reading that people did. Our measurement was based on the Author
Recognition test: a list of names, some of authors and some of non-​authors
(Stanovich, West, and Harrison, 1995). Keith Stanovich and his colleagues
report that the number of authors recognized in such lists is a very close
proxy for the amount of reading people do, as measured by diary meth-
ods and behavioural means. We modified Stanovich and colleagues’ lists to
separate writers of fiction from those of non-​fiction. We found that par-
ticipants who were good at recognizing authors of fiction were also good
at two outcome tests of social understanding. One test, which measures
empathy and theory of mind, was the Mind in the Eyes test (Baron-​Cohen,
Wheelwright, Hill, Raste, and Plumb, 2001):  a set of 36 photographs of
people’s eyes (as if seen through a letter box). To take the test, people
choose for each photograph from four descriptors, for instance, ‘joking’,
‘flustered’, ‘desire’, ‘convinced’. The second outcome measure we used was
the Interpersonal Perception test (Constanzo and Archer, 1993), which is
a set of 15 video clips of ordinary people interacting together. A  person
taking the test has to answer a question about what is going on among the
participants in each of the clips.
Our main finding was that the more fiction people read, the better they
were at the Mind in the Eyes test; this effect was significant. By compari-
son, people who read a lot of non-​fiction were less good at this test. Also,
the more fiction people read, the better they were at the Interpersonal

O n T r u t h a n d Fi c t i o n   [ 269 ]
 

Perception test, though this effect narrowly missed significance at the

5% level. The reason for our main finding seems to be one of expertise.
Because the subject matter of fiction is of selves and others in interaction,
people who read a lot of it become more empathic and better at thinking
about what people are up to. By contrast, readers of non-​fiction are likely
to become better at what they read about—​genetics perhaps, or medieval
history. (Our original segmentation was between fiction and non-​fiction,
but of course certain kinds of non-​fiction such as biography, some history,
and so on, are also narratives about selves in interaction. It might be that
‘narrative’ and ‘expository’ would be better terms for the conditions that
led to different effects, see Mar and Rain, 2015.)
Mar, Oatley, and Peterson (2009) replicated the finding that people who
read more fiction were better at the Mind in the Eyes test, and ruled out
the possibility that the effect was due to individual differences in which,
for instance, people who were more empathic, or more interested in the
social world, might have preferred to read fiction. Mar, Tackett, and Moore
(2010) found in pre-​school children that the more stories the children had
read to them, and the more movies they watched, the better they were at
five theory-​of-​mind tasks. Simply watching television was not associated
with any measure of theory of mind. Mar and Rain (2015) found, in adults,
that reading narrative fiction is better than reading expository non-​fiction
in improving verbal reasoning generally.
In terms of what kinds of fiction are most effective in promoting social
understanding, Katrina Fong, Justin Mullin, and Raymond Mar (2013)
separated four genres: love stories, thrillers, family stories, and science fic-
tion. Reading of love stories and thrillers was most closely associated with
empathy and social understanding (as measured by the Mind in the Eyes
test). Reading science fiction showed no such association. One reason is
probably that in a love story a protagonist needs to work out whether a
certain someone is a suitable person for a long-​term commitment, and in a
thriller the protagonist has to work out what the antagonist is up to when
this character is trying to conceal it. By contrast, the content of science fic-
tion is often focused on technical matters or fantasies that are not primar-
ily interpersonal.
The original studies that showed associations between fiction reading
and improved social understanding were correlational, but more recently
people have undertaken experimental tests. Two such tests reported that
the more transported readers were into a fictional story, the greater was
their affective empathy, and the more likely they were to help someone
pick up some pencils that had been dropped on the floor (Johnson, 2012;
Johnson, Cushman, Borden, and McClune, 2013). Another found that

[ 270 ]  Literature and Cognition in Cognitive Science

 

people low on the personality trait of Openness who read a fictional short
story had better self-​reported empathy than those who read a non-​fictional
literary essay (Djikic, Oatley, and Moldoveanu, 2013). David Kidd and
Emanuele Castano (2013) did a set of experiments, in the first of which
they randomly assigned participants to read one of three literary short
stories (including Chekhov’s ‘Chameleon’) or one of three non-​fictional
essays (including Mann’s ‘How the Potato Changed the World’). Those who
read a fictional piece significantly improved their scores on the Mind in
the Eyes test, as compared with people who read one of the essays. In their
other experiments, they found that literary fiction was better at producing
these effects than popular fiction. Jessica Black and Jennifer Barnes (2015)
replicated Kidd and Castano’s experimental result on the comparison of
fiction and non-​fiction, and showed, too, that reasoning about the social
world was improved, but that there was no effect on reasoning about the
physical world.
The finding that fiction enables people to be better at empathy and at
understanding others is predicted by the theory that fiction is a simulation
of the social world. It is not predicted by opinions that fiction is without
value, or that it should be thought of as a paradox in which what we read
about is unreal, and in which our emotions are inappropriate.

Evidence on Transformation of Self hood

What of the effects of reading on ourselves? Maja Djikic and I, with col-
leagues Sara Zoeterman and Jordan Peterson (2009), randomly assigned
people to read either Chekhov’s most famous story, ‘The Lady with the Little
Dog’, about a man who meets a lady at the seaside resort of Yalta and starts
an affair with her, or a control text that was a non-​fictionalized version: a
report from a divorce court. The non-​fictionalized control version was the
same length, had the same information, and had the same level of reading
difficulty. Readers rated it as just as interesting, though not as artistic as
Chekhov’s story. Before we asked participants to read the text, we admin-
istered the Big Five measure of personality, a standard test that assesses
the traits of Extraversion, Neuroticism, Agreeableness, Conscientiousness,
and Openness. We also asked people to rate how strongly they were feeling
on 10 emotions. After reading we administered these same two measures
again. As compared with those who read the control text, those who read
Chekhov’s story were found to experience small but significant changes in
their personality. The changes were not all in the same direction. Each per-
son changed in his or her own way. These changes were mediated by the

O n T r u t h a n d Fi c t i o n   [ 271 ]
 

amount of emotion people experienced while reading; they were perhaps

temporary, but with a lot of artistic reading, they might cumulate.
We believe the effect is due to what Kierkegaard (1846/​1968) called
‘indirect communication’. Here is how he put it.

The indirect mode of communication makes communication an art in quite a dif-

ferent sense than when it is conceived in the usual manner. … To stop a man on
the street and stand still while talking to him, is not so difficult as to say some-
thing to a passer-​by in passing, without standing still and without delaying the
other, without attempting to persuade him to go the same way, but giving him
instead an impulse to go precisely his own way. (pp. 246‒247)

Chekhov knew that artistic literature was not about getting people to think
or feel in pre-​planned ways. In a letter of 1888 to his friend and mentor
Alexei Suvorin (Hellman, 1955, p. 57), he wrote that there are two things
one must not mix up:

the solution of the problem and a correct presentation of the problem. Only the latter
is obligatory for the artist. In Anna Karenina and Onegin not a single problem is
solved, but they satisfy you completely just because all their problems are cor-
rectly presented. The court is obliged to submit the case fairly, but let the jury do
the deciding, each according to his own judgment. (emphasis in original)

In another letter, written two years later, Chekhov said that in his writing
he assumed that his readers would ‘add the subjective elements that are
lacking in the story’ (Yarmolinsky, 1973, p. 395).
To see whether our finding was not just peculiar to Chekhov’s story,
Djikic and I, with another colleague, Matthew Carland (2012), did another
experiment in which we asked people to read one of eight literary short
stories or one of eight literary essays. (The data set was the same as that of
the study by Djikic, Oatley, and Moldoveanu, 2013.) The stories included
Frank O’Connor’s ‘My Oedipus Complex’ and Jean Stafford’s ‘A Country
Love Story’. The essays included John Galsworthy’s ‘Castles in Spain’ and
Rabindrath Tagore’s ‘East and West’. As in our earlier experiment, we mea-
sured readers’ personality traits and emotions before and after they read
the text to which they had been assigned. We kept the stories as written,
but modified the essays by small amounts, to keep their sense but to ensure
that their average length and ease of reading was the same as for the sto-
ries. We had expected that people who read a piece of fiction would show
most change to their personalities, but the genre of the text—​fiction or
non-​fiction—​did not make much difference. Those who read a story or

[ 272 ]  Literature and Cognition in Cognitive Science

 

essay that they judged to be artistic showed significant more personality

change than those who judged what they read to be less artistic.
Although there are, of course, literary essays, prose writers who aspire
to art have tended to put their best efforts into short stories and novels.
Each such artistic writer of fiction has a vision of the world, and much
fiction derives from personal, social, or political circumstances that have
prompted the writer towards an emotionally intense exploration, and the
urge to offer readers critical issues to think about and have feelings about.
It is, however, not the main goal of artistic literature to persuade readers
or audiences to some particular point of view, or to feel, think, or dispose
themselves in some particular way in relation to such issues.
R. G.  Collingwood (1938) proposed that art should be distinguished
from craft. In craft an outcome is planned in advance. When making lasa-
gne, for instance, the cook plans to produce a certain result. Pre-​planned
effects in fiction can include induction of anxiety in thrillers, or encourage-
ment towards a political attitude, for instance accepting the importance of
civil rights for all. Such motivations among fiction writers are common and,
of course, all art has aspects of craft. Collingwood’s argument however is
that art properly so-​called is not conceived to have a pre-​planned effect but
is an exploration of some issue that the writer feels is important but has
not yet fully understood. Often the exploration is based on emotion. The
story-​reader or audience member is invited to undertake an accompanying
exploration in the process of bringing the story-​simulation alive. Typically,
he or she will experience emotions of the kind that are being explored. But
the reader of a piece of artistic literature does not experience emotions
planned in advance by the writer, and does not experience the emotions of
the characters. The reader experiences his or her own emotions.
Goals of persuasion are common in writing. For instance, although
science is always provisional in that it depends on evidence, all of which
may not yet have been gathered, a writer in science wants the reader to
agree with particular inferences from the evidence or theory that he or she
has presented, as I want you to do now with the chapter you are reading.
Attempts at persuasion occur when a politician gives an election speech.
It occurs in advertising, and in its pernicious cousin, propaganda. Such
goals also occur in artistic literature; for instance, in his first 17 sonnets,
Shakespeare (see Vendler, 1997) is concerned to persuade the young man
to whom most of the sonnets are addressed not to be so self-​involved (in
several senses), but to beget children. But in artistic literature, motives of
persuasion are outweighed by the writer leaving it open to the reader’s or
audience member’s particular imagination of the social world as suggested
in what is written. Because people are different, because their tastes are

O n T r u t h a n d Fi c t i o n   [ 273 ]
 

different, because the memories on which they draw to create their imag-
ined worlds are their own, their imaginings of the storyworlds of fiction will
be partly idiosyncratic. In science, by contrast, a writer aims for a shared
understanding among all his or her readers. The artistic writer invites read-
ers or audiences to enter imagined worlds, to think their own thoughts and
to have their own emotions, in the circumstances that are suggested.
This leads to a further conclusion, and to a proposal about the psychology
of fiction. There is a large social psychology of how people can be persuaded,
for instance in voting or making purchases (e.g. Green and Brock, 2005).
The psychology of fiction heads in a different direction, towards a psychol-
ogy of ‘indirect communication’ (as Kierkegaard called it), to enable people
to go not in some way that a writer might prefer, but in their own way.
The idea of indirect communication is not unfamiliar. With our children
or lovers, we hope to influence them in a way that enables them not to be
what we want but to be themselves. And in non-​directive forms of psycho-
therapy a therapist seeks to enable a client to make his or her own decisions.
This idea is less familiar for artistic writing, but it is comparable (Djikic
and Oatley, 2014; Oatley and Djikic, 2014). Artistic literature is not pri-
marily about influencing people to feel and behave in some particular way
according to the purposes of the influencer. The kind of writing that exerts
indirect influence can be non-​fiction, but perhaps most often it is artistic
fiction in the form of poetry, plays, novels, short stories, films. Works of
this kind influence us and enable us to change by inviting us into situa-
tions that are often different from those we ordinarily enter, and express-
ing them in phrases and ideas we would not have thought of ourselves. In
fiction we can lead many lives. In relation to the circumstances we enter
in a piece of fiction, and only if we want to, we can experience ourselves in
new ways, and change, at least temporarily, in ways that are our own.

CONCLUSION

The writings of science are provisional because they depend on something

outside themselves: evidence. The writings of fiction are provisional because
they depend on something outside themselves: emotion, imagination, and
the judgements of readers. Much has been written about the methods of
science. Less has been written about the methods of artistic fiction. These
methods are based on simulations of the social world, dependent on the
imagination and developed from our ability, as children, to play. The emo-
tions one may experience in fiction are not paradoxical. Like mathematics
and science, fiction is an approach to truth. It is a kind of truth that is social

[ 274 ]  Literature and Cognition in Cognitive Science

 

and empathic. It enables us to understand others in their inwardness and

also to think and to experience emotions for ourselves.

ACKNOWLEDGEMENT

I offered some of the arguments proposed here in a conference talk entitled

‘Beyond Fictionality: Cognition and the Psychology of Fiction’ (Oatley, 2013b).

REFERENCES

Aristotle. (1970). Poetics (G. E. Else, Trans.). Ann Arbor, MI: University of Michigan
Press.
Astington, J. W., Harris, P. L., and Olson, D. R. (Eds.). (1988). Developing theories of
mind. New York: Cambridge University Press.
Baron-​Cohen, S., Wheelwright, S., Hill, J., Raste, Y., and Plumb, I. (2001). The ‘Reading
the Mind in the Eyes’ test revised version:  A  study with normal adults, and
adults with Asperger’s syndrome or high-​functioning autism. Journal of Child
Psychology and Psychiatry, 42, 241–​251.
Black, J., and Barnes, J. L. (2015). Fiction and social cognition: The effect of viewing
award-​winning television dramas on theory of mind. Psychology of Aesthetics,
Creativity, and the Arts, 9(4), 423–​429.
Boyd, B. (2009). On the origin of stories. Cambridge, MA: Harvard University Press.
Bruner, J. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press.
Caracciolo, M. (2014). Beyond other minds: Fictional characters, mental simulation,
and ‘unnatural’ experiences. Journal of Narrative Theory, 44, 29–​53.
Collingwood, R. G. (1938). The principles of art. Oxford: Oxford University Press.
Constanzo, M., and Archer, D. (1993). The Interpersonal Perception Task-​15 (IPT-​
15):  A  guide for researchers and teachers. Berkeley, CA:  University of California
Extension Center for Media and Independent Learning.
Cox, J. (2011). Brief lives: Charlotte Brontë. London: Hesperus Press.
Currie, G. (2011, 2 September). Let’s pretend. Times Literary Supplement, pp. 14–​15.
Dias, M., Roazzi, A., and Harris, P. L. (2005). Reasoning from unfamiliar prem-
ises: A study with unschooled adults. Psychological Science, 16, 550–​554.
Djikic, M., and Oatley, K. (2014). The art in fiction: From indirect communication to
self-​development. Psychology of Aesthetics, Creativity, and the Arts, 8, 498–​505.
Djikic, M., Oatley, K., and Carland, M. (2012). Genre or artistic merit:  The effect of
literature on personality. Scientific Study of Literature, 2, 25–​36.
Djikic, M., Oatley, K., Zoeterman, S., and Peterson, J. (2009). On being moved by
art: How reading fiction transforms the self. Creativity Research Journal, 21, 24–​29.
Dunbar, R. I.  M. (2004). The human story:  A  new history of mankind’s evolution.
London: Faber.
Dunn, J. (2004). Children’s friendships: The beginnings of intimacy. Oxford: Blackwell.
Fong, K., Mullin, J., and Mar, R. (2013). What you read matters:  The role of fiction
genres in predicting interpersonal sensitivity. Psychology of Aesthetics, Creativity,
and the Arts, 3, 370–​376.

O n T r u t h a n d Fi c t i o n   [ 275 ]
 

Freud, S. (1908/​1985). Creative writers and day-​dreaming. In A. Dickson (Ed.), Pelican

Freud library, 14: Art and literature (pp. 130–​141). London: Penguin.
Gavins, J. (2007) Text world theory:  An introduction. Edinburgh:  Edinburgh
University Press.
Gerrig, R. J. (1993). Experiencing narrative worlds: On the psychological activities of read-
ing. New Haven, CT: Yale University Press.
Goldman, A. (2006). Simulating minds:  The philosophy, psychology, and neuroscience of
mindreading. New York: Oxford University Press.
Green, M. C., and Brock, T. C. (2005). Persuasiveness of narratives. In T. C. Brock and
M. C. Green (Eds.), Persuasion: Psychological insights and perspectives (2nd ed., pp.
117–​142). Thousand Oaks, CA: Sage.
Green, M. C., Kass, S., Carrey, J., Herzig, B., Feeney, R., and Sabini, J. (2008).
Transportation across media:  Repeated exposure to text and film. Media
Psychology, 11, 512–​539.
Halliwell, S. (2002). The aesthetics of mimesis:  Ancient texts and modern problems.
Princeton, NJ: Princeton University Press.
Harris, P. L. (2000). The work of the imagination. Oxford: Blackwell.
Hellman, L. (Ed.) (1955). The selected letters of Anton Chekhov. New York: Farrar, Strauss.
Huizinga, J. (1955). Homo ludens:  A  study of the play-​element in culture. Boston,
MA: Beacon.
Johnson, D. R. (2012). Transportation into a story increases empathy, prosocial behav-
ior, and perceptual bias toward fearful expressions. Personality and Individual
Differences, 52, 150–​155.
Johnson, D. R., Cushman, G. C., Borden, L. A., and McClune, M. S. (2013). Potentiating
empathic growth:  Generating imagery while reading fiction increases empa-
thy and prosocial behavior. Psychology of Aesthetics, Creativity, and the Arts, 7,
306–​312.
Jones, E. E., and Nisbett, R. E. (1971). The actor and the observer: Divergent perceptions
of the causes of behavior. New York: General Learning Press.
Keen, S. (2007). Empathy and the novel. New York: Oxford University Press.
Kidd, D. C., and Castano, E. (2013). Reading literary fiction improves theory of mind.
Science, 342, 377–​380.
Kierkegaard, S. (1846/​1968). Concluding unscientific postscript (D. F. Swenson and
W. Lowrie, Trans.). Princeton, NJ: Princeton University Press.
Kotovych, M., Dixon, P., Bortolussi, M., and Holden, M. (2011). Textual determinants of
a component of literary identification. Scientific Study of Literature, 1, 260–​291.
Lauer, G. (2009). Going empirical: Why we need cognitive literary studies. Journal of
Literary Theory, 3, 145–​154.
Leverage, P., Mancing, H., Schweickert, R., and William, J. M. (2011). Theory of mind and
literature. West Lafayette, IN: Purdue University Press.
Luria, A. R. (1976). Cognitive development: Its cultural and social foundations. Cambridge,
MA: Harvard University Press.
Mar, R. (2011). The neural bases of social cognition and story comprehension. Annual
Review of Psychology, 62, 103–​134.
Mar, R., and Oatley, K. (2008). The function of fiction is the abstraction and simulation
of social experience. Perspectives on Psychological Science, 3, 173–​192.
Mar, R. A., Oatley, K., Hirsh, J., dela Paz, J., and Peterson, J. B. (2006). Bookworms
versus nerds: Exposure to fiction versus non-​fiction, divergent associations with
social ability, and the simulation of fictional social worlds. Journal of Research in
Personality, 40, 694–​712.

[ 276 ]  Literature and Cognition in Cognitive Science

 

Mar, R. A., Oatley, K., and Peterson, J. B. (2009). Exploring the link between reading
fiction and empathy: Ruling out individual differences and examining outcomes.
Communications: The European Journal of Communication, 34, 407–​428.
Mar, R. A., and Rain, M. (2015). Narrative fiction and expository nonfiction differen-
tially predict verbal ability. Scientific Studies of Reading, 19(6), 419–​433.
Mar, R. A., Tackett, J. L., and Moore, C. (2010). Exposure to media and theory-​of-​mind
development in preschoolers. Cognitive Development, 25, 69–​78.
Mithen, S. (1996). The prehistory of the mind:  The cognitive origins of art and science.
London: Thames and Hudson.
Nickerson, R. (1999). How we know—​ and sometimes misjudge—​ what others
know: Imputing one’s own knowledge to others. Psychological Bulletin, 125, 737–​759.
Oatley, K. (1992). Best laid schemes: The psychology of emotions. New York: Cambridge
University Press.
Oatley, K. (1999). Why fiction may be twice as true as fact: Fiction as cognitive and
emotional simulation. Review of General Psychology, 3, 101–​117.
Oatley, K. (2011). Such stuff as dreams: The psychology of fiction. Oxford: Wiley-​Blackwell.
Oatley, K. (2012). The cognitive science of fiction. Wiley Interdisciplinary Reviews:
Cognitive Science, 3(4), 425–​430.
Oatley, K. (2013a). Worlds of the possible:  Abstraction, imagination, consciousness.
Pragmatics and Cognition, 21, 448–​468.
Oatley, K. (2013b). Beyond fictionality: Cognition and the psychology of fiction. Paper pre-
sented at the Fiction in historical and cultural contexts conference, Faculdade de
Filosofia, Letras e Ciências Humanas, University of São Paulo, Brazil, 19–​21 March.
Oatley, K., and Djikic, M. (2014, 19 December). How reading transforms us. In Sunday
Review, Gray Matter, New York Times. Retrieved from http://​www.nytimes.com/​
2014/​12/​21/​opinion/​sunday/​how-​writing-​transforms-​us.html
Oatley, K., and Olson, D. R. (2010). Cues to the imagination in memoir, science, and
fiction. Review of General Psychology, 12, 56–​64.
Plato. (375 b.c.e./​1955). The republic. Harmondsworth, UK: Penguin.
Plato. (402 b.c.e./​1956). Meno, in Protagoras and Meno (W. K.  C. Guthrie, Trans.).
Harmondsworth, UK: Penguin.
Powell, B. (2002). Writing and the origins of Greek literature. Cambridge:  Cambridge
University Press.
Prentice, D., Gerrig, R. J., and Bailis, D. S. (1997). What readers bring to the processing
of fictional texts. Psychonomic Bulletin and Review, 4, 416–​420.
Radford, C. (1975). How can we be moved by the fate of Anna Karenina? Proceedings of
the Aristotelian Society, Supplementary Volumes, 49, 67–​80.
Simenon, G. (1971). Maigret et l’indicateur. Paris: Presses de la Cité.
Stanovich, K. E., West, R. F., and Harrison, M. R. (1995). Knowledge growth and
maintenance across the life span:  The role of print exposure. Developmental
Psychology, 31, 811–​826.
Stevenson, R. L. (1884/​1992). A humble remonstrance. Longman’s Magazine, December.
Reprinted in C. Harman (Ed.), R. L. Stevenson: Essays and poems (pp. 1179–​1188).
London: Dent Everyman’s Library.
Taylor, M., Hodges, S., and Kohányi, A. (2002‒2003). The illusion of independent
agency: Do adult fiction writers experience their characters as having minds of
their own? Imagination, Cognition, and Personality, 22, 361–​380.
Trabasso, T., and Chung, J. (2004). Empathy: Tracking characters and monitoring their
concerns in film. Paper presented at the Winter Text Conference, Jackson Hole,
WY, 23 January.

O n T r u t h a n d Fi c t i o n   [ 277 ]
 

Vendler, H. (1997). The art of Shakespeare’s sonnets. Cambridge, MA:  Harvard

University Press.
Walton, K. (1990). Mimesis as make-​believe: On the foundations of the representational
arts. Cambridge, MA: Harvard University Press.
Walton, K. (1997). Spelunking, simulation, and slime: On being moved by fiction. In
M. Hjort and S. Laver (Eds.), Emotion and the arts (pp. 37–​49). New York: Oxford
University Press.
Winnicott, D. W. (1971). Playing and reality. London: Tavistock.
Yarmolinsky, A. (Ed.). (1973). Letters of Anton Chekhov. New York: Viking.
Zunshine, L. (2006). Why we read fiction:  Theory of mind and the novel. Columbus,
OH: Ohio State University Press.

[ 278 ]  Literature and Cognition in Cognitive Science

 

CHAPTER 14

Under Pressure
Norms, Rules, and Coercion in Linguistic
Analyses and Literary Readings
ALEXANDER BERGS

INTRODUCTION

A remarkable feature of human cognition is pattern recognition. From our

earliest days on this earth, we seek to establish patterns, regularities, sche-
mata in the input we get (see Tomasello, 2006, and elsewhere). Pattern rec-
ognition as such is a central and important ability, since most of our daily
interactions and cognitive processes are in some way or another based on
it: language, social interaction, navigation and orientation, food prepara-
tion and consumption, and so much more. Without any (recognizable) pat-
terns, everything we say, hear, or do would be unpredictable, new, and in
need of computationally costly interpretation. In that sense, pattern rec-
ognition is, of course, also related to some principles of Gestalt psychology
(Neisser, 2004). Gestalt, in a nutshell, refers to the percepts formed in the
human mind, which are often not just based on the actual elements per-
ceived, but rather influenced or even formed by the idea of ‘the whole’ in
the mind, which transcends the sum of its parts. So when we see a figure
like the one in Figure 14.1, we not only see the actual parts (three black
pacman-​like objects) but also ‘see’ (or believe that we perceive) three black
circles with a white triangle in the middle and foreground so it that covers
three pie-​slice shapes of the circles. So the whole of the picture and how we
perceive it is actually quite different from the simple sum of its parts.
 

Figure 14.1  Gestalt psychology.

These mental processes are made possible by a number of different prin-

ciples or laws that guide the construction of gestalts. Some of these laws
are labelled Proximity, Similarity, Continuity, and Closure. The Law of
Proximity says that elements which are closer to each other are more likely
to be perceived to form a group than those which are at some greater dis-
tance. The Law of Similarity claims that elements in an arrangement are
more likely to be seen as a group when they are similar to each other. The
Law of Continuity states that objects are more likely to perceived as percep-
tual wholes if they are aligned, so that two overlapping objects, for example
as in Figure 14.1, are more likely to be perceived as independent circles and
an overlapping triangle, rather than as discontinuous objects with sharp
directional changes. Finally, the Law of Closure states that well-​known
objects are perceived as such, even when their representation has, for exam-
ple, broken lines or interruptions, so that regularity in objects is maximized.
It seems intuitively plausible that Gestalt psychology and pattern recog-
nition go hand in hand, not least since many laws and principles of Gestalt
psychology operate on the basis of perceived or constructed patterns. It
should be noted, though, that pattern recognition is more extensive than
the laws of Gestalt psychology, as patterns do not necessarily depend on
any kind of (natural) gestalt, but can in fact be very irregular in themselves,
as long as the whole as such repeats itself and thus becomes schematized.
Both Gestalt psychology and pattern recognition are based on expecta-
tions, though. While we perceive and process incoming data in the here and
now, we also do so dynamically in time (i.e. in relation to past experiences
and also with certain predictions about what will follow). These predic-
tions can be based on past experiences, Gestalt laws, pattern recognition,
or even probabilistic thinking, but they all lead to expectations about what
will happen next.

[ 280 ]  Literature and Cognition in Cognitive Science

 

While pattern recognition and schematization are a crucial feature

in human cognition, not everything we encounter follows a given pat-
tern or schema. What happens when we come across some mismatch,
some unexpected data that does not fit in with our predictions? This,
essentially, is the topic of this chapter. What exactly counts as a mis-
match? How are mismatches resolved, if at all? This chapter will inves-
tigate the role and nature of mismatch and ‘coercion’ (the technical
term for the resolution of mismatch) in three hitherto mostly indepen-
dent domains: linguistic analyses, literary readings, and neuroscience.
On the basis of some exemplary analyses of English-​language poetry
and prose texts from the 17th to the 20th centuries, I will argue that
some fundamental principles of literary practice (such as foreground-
ing or deviation) can actually be seen as coercive (i.e. sense-​making,
or mismatch-​resolving; see Ziegler, 2007, p. 992) linguistic activi-
ties which in turn have their basis in neuropsychology. This is par-
ticularly interesting insofar as coercion itself, like pattern recognition,
might prove to be another universal mechanism of human cognition,
which may be shown to have significant neurocognitive underpinnings
(Pylkkänen, 2008, and elsewhere; Kuperberg et al., 2010; Husband,
Kelly, and Zhu, 2011; de Almeida and Riven, 2012). If this were indeed
the case, it would not be surprising to find that coercion plays at least
some role in literary practices.

MISMATCH AND COERCION IN LINGUISTICS

For more than 25 years, linguists have studied the phenomenon of coer-
cion (or type-​shifting), in which an apparent mismatch between linguistic
elements lies at the heart of a (new) reading for a given utterance, as in the
infamous examples (1) and (2).

(1) She smiled herself an upgrade. (Adams, Hitchhiker’s Guide to the

Galaxy, first mentioned by Goldberg, 1995)
(2) The ham sandwich wants to pay. (Nunberg, 1979. The context
is a restaurant scene, with one waiter saying this to the
other.)

In both cases, semantically or grammatically incompatible elements

seem to combine and render a perfectly interpretable utterance. ‘Smile’
is an intransitive verb, so it cannot be combined with any object,
let alone a reflexive complement as in (1). And yet, native speakers will

N or m s , R u l e s , a n d C o e r c i o n   [ 281 ]
 

automatically and without objection interpret (1) as ‘She smiled in such

a way that she was given an upgrade’. In (2) the inanimate subject noun
phrase ‘the ham sandwich’ is disallowed by verbs such as want which
require an animate (agentive) subject. And yet, (2) is easily interpretable
(by the second waiter in the particular context) as ‘The customer who ate
the ham sandwich wants to pay’. Even though these are two examples
of clear mismatches, speakers/​hearers experience no trouble in finding
plausible readings for both utterances (though one has to acknowledge
that the second example needs more context to be understandable than
the first example). These new readings are said to be ‘coerced’ out of
those particular mismatches. The term ‘coercion’ (for related terms, see
below) signifies that in order to arrive at a meaningful interpretation,
items need to be ‘forced’ into a new and commonly not available reading
(by the speaker/​hearer)—​‘ham sandwiches’ cannot pay unless we inter-
pret them metonymically, and you cannot smile something, unless we
(are forced to) reinterpret smile semantico-​syntactically as a verb like
buy. But how does that work?
At first sight, it appears that coercion should provide an ‘anything goes’
kind of licence. However, this does not seem to be the case. Even with coer-
cion as a mechanism to resolve mismatch, some utterances remain (almost)
uninterpretable, such as (3).

(3) ain’t me without don’t “Adventures, matter You about you

have Tom know read a by the that name The of book Sawyer,”
but no of.

This is the first, well-​known line of The Adventures of Huckleberry Finn

(Twain, 1885/​1994, p. 1), jumbled at random: ‘You don’t know about me,
without you have read a book by the name of “The Adventures of Tom
Sawyer,” but that ain’t no matter.’ The way it is presented here makes it
(almost) incomprehensible. Even with a lot of effort, a plausible interpre-
tation probably cannot be coerced out of example (3), at least not without
an extensive amount of context and imagination. In any case, it should be
obvious that the mismatch in (3) is much harder to resolve, if at all, than
the one in (1) and (2). It is interesting to consider a possible third, inter-
mediate level between utterances like (1) and (2), which are more or less
easily interpreted through coercion, and example (3), which is practically
impenetrable. Consider ­example (4):

(4) The coffee table read the albatross.

[ 282 ]  Literature and Cognition in Cognitive Science

 

Example (4) in contrast to (3) is not ungrammatical in the technical sense.

In other words, apart from semantic considerations, the syntax of this sen-
tence is well formed. And indeed, there is the impression that an utter-
ance like (4) might receive some more or less plausible interpretation in
a particular context (such as an imaginary world, like we find in Alice in
Wonderland, in which certain semantic constraints do not seem to apply).
But it also seems plausible to assume that (4) requires more effort and a
particular context, whereas (1) and (2) are perhaps more generalized. The
conclusion could be that there is a continuum between mismatches in
utterances that are ‘easily’ resolved, and those which are not susceptible to
coercion, with utterances like (4) somewhere in the middle. I will return to
this question of a continuum below.
Unfortunately, the exact boundaries and limits of coercion are still
unclear, so that it is not easy to clearly distinguish a priori between mis-
matches that are susceptible to coercion and those which are not and thus
either remain meaningless, or alternatively require a substantial amount
of interpretive work. The discussion of neurocognitive aspects later in this
chapter may shed some more light on this question. Similarly, the exact
relations of coercion to other phenomena, such as metaphor, metonymy,
or blending, are still somewhat unclear and hotly debated (for an up-​to-​
date exposition and discussion of coercion, see the papers in Lauwers and
Willems, 2011). I will come back to this below.
Coercion, alternatively known as type-​shifting (Partee and Rooth, 1983),
accommodation (Goldberg, 1995), enriched composition (Jackendoff,
1997), forçage (Gadet, Léon, and Pécheux, 1984), implicit conversion
(Talmy, 2000), or simply mismatch (Francis and Michaelis, 2003) has been
defined in linguistics as ‘a compromise between the combinatorial con-
straints imposed by the language system and the flexibility (and creativity)
allowed by the same system’ (Lauwers and Willems, 2011, p. 1219). Ziegler
succinctly summarizes the idea:  ‘coercion is thus the resolution of mis-
match’ (Ziegler, 2007, p. 992). I will follow this general idea in the present
chapter and say that mismatch is the actual observable phenomenon while
coercion is the mechanism that can resolve this mismatch. The following
examples illustrate this idea.

(5) She must swallow her anger. (Corpus of Contemporary

American English [COCA], True Sisters, 2012)
(6) Two empty hours were a sinus in which infections bred.
(Franzen, The Corrections, 2001, p. 1)
(7) The White House said more than 1 million U.S. jobs could be
created over the next decade. (COCA, Associated Press, 2012)

N or m s , R u l e s , a n d C o e r c i o n   [ 283 ]
 

  (8) I was the only one who thought she drank the whole bottle.
(COCA, The Atlantic, 1998)
  (9) Smith began the book after eight days at Ground Zero.
(COCA, USA Today, 2002)
(10) He sneezed the napkin off the table. (Goldberg,
Constructions, 1995)
(11) I’m lovin’ it. (McDonald’s)
(12) James Bond suddenly knew that he was tired. (COCA,
Style, 1990)

In (5)  the immaterial emotion anger is the complement of the physical

action verb swallow. In grammatical theory at least, this should not be
possible. However, in this case the actually attested utterance is perfectly
interpretable, as a new metaphorical reading is coerced out of this particu-
lar mismatch. Similarly, in (6), the immaterial hours are metaphorically
interpreted as a physical sinus (a place) in which infections can actually
breed. In (7)  and (8)  we find a typical metonymic expression, with The
White House standing for ‘US politics’ generally, and the whole bottle for
the contents of that bottle. In (9) the verb began does not combine with
another verb as in begin reading, begin eating … , but rather with a noun,
which is technically impossible. Nevertheless, the mismatch of begin the
book coerces the interpretation begin reading the book. Similarly, in (10)
the intransitive verb sneeze combines with a complex object construction
the napkin off the table. Coercion leads to the reading ‘sneezed so hard that the
napkin flew off the table’. In (11), the stative verb love—​which usually does
not occur in a progressive BE-​V-​ing construction—​is used in exactly such
a construction to signify intensity and dynamicity (Huddleston, 2002,
p. 170). Finally, in (12) the adverbial suddenly combines with the stative
verb know. But facts can only be realized suddenly, not suddenly known,
which leads to a certain kind of mismatch. Note that all these utterances
are attested and that mismatch is apparently anything but rare. In fact, it
seems to be an all-​pervasive facet not only of literary language or planned
discourse, but of everyday speech. In fact, it seems so deeply ingrained
and familiar that many speakers do not even notice certain mismatch as
such (e.g. begin the book usually does not sound ‘wrong’ or ‘mismatched’ to
most native speakers).
Also note that not all sorts of mismatches are susceptible to coercion,
as already mentioned above. Examples (13) and (14), for instance, also dis-
play semantic (13) or grammatical (14) mismatches. In (13), the intransi-
tive verb sleep is complemented by the object ‘the book’, and in (14) the
third-​person singular subject Peter obviously does not agree with the verb

[ 284 ]  Literature and Cognition in Cognitive Science

 

write. Yet, these mismatches are not as readily interpretable as examples

(5)‒(12) above.

(13) Mary slept the book.

(14) Peter write the book.

Usually instances such as these are categorized as ‘anomalous’ and are not
treated as examples for mismatch and coercion. We can thus find sets of
sentences such as (15a‒c) below that illustrate regular versus mismatched
versus anomalous constructions.

(15) a. The journalist wrote the article before his coffee break.

b. The journalist began the article before his coffee break.
c. The journalist astonished the article before his coffee break.

Example (15a) is a perfectly regular, semantically and grammatically

well-​formed, compositional sentence with no mismatch and therefore
also no need for coercion of any kind. Example (15b) shows a semantico-​
grammatic mismatch in that the verb begin actually requires a verbal rather
than a nominal complement. Yet, through coercion we can easily arrive at
a contextually appropriate interpretation: began writing/​reading. Example
(15c), however, which also shows some apparent mismatch (the verb aston-
ish requires an animate object), would require a considerable amount of
context and effort to be interpretable at all. It is therefore usually classified
as anomalous. We will come back to this distinction later.
Mismatches such as (5)‒(12) and (15b) can also be classified into differ-
ent categories. We need to distinguish between nominal coercion (16a, b),
aspectual coercion (16c, d) and complement coercion (16e, f).

(16) a. She had a beer.

b. You have apple on your shirt.
c. For months, the train arrived late.
d. Peter is believing in ghosts these days.
e. Sam squeezed the ball inside the jar.
f. The ham sandwich wants to pay.

In (16a) the actually uncountable noun beer is used as countable noun, in

(16b) the countable noun apple is used as an uncountable substance noun.
In (16c) the simple present verb form arrive, which actually denotes punc-
tual events, is combined with a time adverbial that suggests a durative
event (as in He stayed there for months). Consequently, we get an iterative

N or m s , R u l e s , a n d C o e r c i o n   [ 285 ]
 

reading across that time span. In (16d) the stative verb believe is used in the
progressive, which usually signals active, ongoing action. Here, the inter-
pretation is rather that the state is only temporary and currently also rel-
evant. In (16e) the verb squeeze is incompatible with a directional adverb,
and in (16f) the want requires a human, or at least an animate, subject.
Just as we find numerous names for mismatch and coercion, there are
also quite a number of linguistic accounts of the mechanism itself. One of
the most central questions in this regard is whether coercion (i.e. the reso-
lution of mismatch) is a pragmatic, context-​based operation or not.

Formal Approaches to Coercion

Coercion has been studied extensively in the domain of Montague gram-

mars and similar formal approaches to language structure. Since this is not
the right place to discuss the details of these approaches, suffice it to say
that an influential account such as found in James Pustejovsky’s (1995)
book The Generative Lexicon claims that coercion is type-​shifting, that is, a
semantic operation that shifts the type of a (mismatched) argument to the
one expected by the function of the predicate. This semantic operation, also
called ‘enriched composition’, is based on the abstract qualia features1 of the
elements in question and needs to be strictly constrained in the grammatical
system in order not to overgenerate (i.e. produce ungrammatical sentences).
Constraints on this operation are (a) that it only applies with a very limited
number of verbs (or items) as functors/​predicates, (b) that it always affects
the arguments, never the functors/​predicates (e.g., the interpretation of
begin remains constant, while that of the book is enriched appropriately to
arrive at an event-​based reading), and (c)  that the lexicon should contain
a list of very well-​defined reinterpretations for the affected arguments; for
example, He began the book usually evokes reading the book or writing the book
but not colouring the book, unless there are very strong contextual triggers for
this reading (which is then probably not part of the lexicon or grammar, but
rests on pragmatic ad-​hoc processes). Consider ­example (17).

(17) The three-​year-​old had been colouring books all morning.

Only one was left. He began the book after his nap.

1. The structure of so-​called qualia features stems from the generative lexicon
(Pustejovsky, 1995). Pustejovsky defines them as ‘modes of explanation associated with
a word or phrase in the language’ and distinguishes between the following aspects: for-
mal (what a given element is), constitutive (what the element is made of), telic (what
the function of the element is), and agentive (how the element came into being).

[ 286 ]  Literature and Cognition in Cognitive Science

 

In (17), the interpretation began the book as began colouring the book is con-
textually licenced and could also be seen in the context of elision, rather
than mismatch and coercion. Certain reinterpretations are excluded alto-
gether: astonish the book, for example, is not susceptible to regular gram-
matical or semantic mechanisms such as coercion. Pustejovsky thus
suggests a very strong and detailed generative lexicon that can contain this
kind of information. The actual competition between contextual pressure
and semantic default properties, however, is still an unresolved issue. The
point is that there is a rich body of research offering formalized and more
or less context-​free mechanisms that plausibly explain how mismatches of
all kinds can get resolved by the hearer.

Pragmatic Approaches to Coercion

The alternative view is that the resolution of mismatches mostly depends

on ‘pragmatic inferences computed over underspecified semantic repre-
sentations’ (de Almeida, 2004, p.  249). In other words, coercion as such
is not a semantico-​syntactic mechanism, but the resolution of mismatch
is based on more general cognitive principles (such as pragmatic inferenc-
ing), which in turn are invited (or necessitated) by underspecified semantic
representations. It has been claimed (de Almeida, 2004, p. 259) that sen-
tences such as John began the dictionary are not semantically ill-​formed, but
actually violate one or more Gricean maxims of conversation (Grice, 1975;
Horn, 2012). In John began the dictionary, we find violations of both the
maxim of quantity (‘make your contribution as informative as is required’;
this utterance lacks important pieces of information) and the maxim of
manner (‘avoid obscurity of expression’, ‘be orderly’; this utterance is
not clear). These violations motivate actual interpretation as inference,
whenever possible. But hearers may also conclude with a reaction such as
‘began doing WHAT with the dictionary?’ On the other hand, given enough
contextual clues, the violations can also disappear or at least be reduced,
resulting in a regular and easily interpretable expression.

Coercion as Metaphor and Metonymy

Cognitive linguistics has offered yet another account for at least some of
the mismatch and coercion phenomena already discussed. Instances such
as (5)‒(9) have also been analysed as metaphor (5)‒(6) and metonymy
(7)‒(9).

N or m s , R u l e s , a n d C o e r c i o n   [ 287 ]
 

Needless to say, cognitive linguistics has not been the first approach
to treat examples like these as instances of metaphor and metonymy.
Philosophy, rhetoric, and literary studies offered complex discussions as
early as Greek and Roman antiquity. Cognitive linguistics added to these
well-​established perspectives in that it brought to the fore the cognitive
underpinnings and general cognitive mechanisms responsible for the pro-
duction and reception of metaphorical and metonymical constructions
(this is not the right place to discuss this in any greater detail; see e.g.
Gibbs, 2008; Ortony, 2008; and Fludernik, 2011). What is interesting for
the present discussion is that some cognitive-​linguistic analyses of exam-
ples such as (5)‒(9) need little or no recourse to concepts such as mismatch
or coercion (Ziegler, 2007, would be one such example).
Metaphors and metonymies such as these are treated as features of
everyday language, often motivated by embodied cognition and expli-
cable through mechanisms such as conceptual integration (or ‘blend-
ing’, see Fauconnier and Turner, 2002)  and conversational inferencing
(see Ziegler, 2007, p.  1003). Francisco Gonzálvez-​García (2011) takes
a more moderate position and convincingly argues that metaphor and
metonymy do not render coercion superfluous, but that these concepts
are actually compatible with each other. With reference to Peter Harder
(2010, p. 247), he explains that metaphor and metonymy can be seen as
bottom-​up, cumulative conceptual processes, while coercion (and syntax
generally) are rather top-​down processes that assign syntactic functions.
In particular, Gonzálvez-​García argues that grammatical constructions
(as in the frameworks of Berkeley Construction Grammar, Sign Based
Construction Grammar, or the Lexical Constructional Model) can ele-
gantly provide us with constructional templates (constrained by syn-
tax) operating as top-​down mechanisms in combination with metaphor
and metonymy (as bottom-​up meaning construal) in order to arrive at
a ‘proper understanding of the division of labor between lexical mean-
ing and grammatical meaning’ (2011, p.  1348). Coercion is then the
essential mechanism that allows for the interpretation of ‘mismatches’
within a certain ‘interpretive latitude’ (p.  1348) determined by syntax.
Furthermore, Gonzálvez-​García argues that there is a group of construc-
tions (the so-​called subject-​transitive in English, as in You think him guilty
or They called me a Frankenstein) which does not ‘appear to be amenable to
an explanation in terms of metaphor or metonymic extension alone (at
least synchronically), thus pointing to the inevitability of retaining the
mechanism of coercion’ (2011, p. 1350). Reasons for this include the com-
plex semantico-​pragmatic constraints that affect this group of construc-
tions, as well as the fact that these constraints lead to a variety of related

[ 288 ]  Literature and Cognition in Cognitive Science

 

sub-​constructions. The latter call into question the explanatory power of

an all-​purpose mechanism such as metonymy.
For present purposes, this particular controversy must remain unre-
solved. However, future studies on music and art, for instance, could dis-
cuss the notion of mismatch and coercion in their own domains, such as
when certain notes or colours unexpectedly ‘don’t add up’—​that is, when
pattern-​based expectations of hearers or spectators are not met. One
example can be found in Mozart’s Symphony No. 31 (KV 297), with a sur-
prising shift to F major (instead of staying in A major) in the first repetition
in development of the first movement. Abstract art or atonal music might
also be seen from the perspective of mismatch, though one would have
to ask whether this is always mismatch that requires coercion, or whether
there could also be anomalies that require other interpretive strategies. In
any case, if it can be shown that similar effects like mismatch and coercion
can also be observed in other domains, we have further reason to believe
that coercion might be a very basic cognitive principle.
An idea that stems from the discussion in Gonzálvez-​García and which
will also be important in what follows is the question of whether coercion is
a binary or a gradient phenomenon. Gonzálvez-​García (going back to Boas
[2008] and Langacker [2009]) argues that we need to distinguish between
highly entrenched and conventionalized constructions (such as He sent a
package to her uncle, where send is coerced into a caused-​motion reading)
and novel, highly constrained expressions such as She sneezed the napkin
off the table, where sneeze also receives a caused-​motion reading. It seems
intuitively plausible and clear that while both structures require some sort
of coercion, send is much more easily used and requires less effort in pro-
cessing than sneeze. The notion of ‘more’ or ‘less’ of course suggests some
gradient in coercion, rather than a binary distinction. How far this claim
can be substantiated by findings from other domains will be discussed in
the following sections.

MISMATCH AND COERCION IN LITERATURE

It would be an exaggeration to claim that mismatch lies at the heart of

all poetic and artistic work. As I have shown, mismatch is closely related
to pattern recognition and Gestalt psychology. Patterns, often based on
Gestalt laws, form the basis of our expectations, and mismatch breaks
with patterns, presents something unexpected, and thus creates the
effect of novelty. This can be expressed in terms of other important
Gestalt principles such as figure-​ground, which forms the basis of many

N or m s , R u l e s , a n d C o e r c i o n   [ 289 ]
 

works of art (e.g. the use of ground in Vermeer’s single-​figure paintings

from 1660s, where he used specific elements in the ground to reduce any
impression of physical movement; and the intentional ‘cancellation’ of
figure-​ground relations in 20th-​century art). The simple and yet cen-
tral idea is, in a nutshell, that aesthetic experience can arise when input
is unexpected, at least to some degree (Silvia, 2014, p.  265). This idea
can already be found expressed in the early 18th century, when Joseph
Addison took Horace’s famous dictum that art should ‘prodesse et delec-
tare’ (‘instruct and delight’) and instead claimed that art should (also)
‘surprise and delight’ (Addison, 1712). And the role and nature of ‘sur-
prise’ in art has been a matter of debate ever since (see the comprehen-
sive outline and discussion in Miller, 2015). For our present purposes,
suffice it to say that aesthetic effects can arise when there is a pattern, a
background, against which something new and unexpected (a mismatch)
stands out. This unexpected input is the cause of additional processing
and computing, which in turn has sometimes been implicated in the
cognitive-​psychological basis of aesthetic pleasure (e.g. Schmidhuber,
2009). Here are just two examples:

(18) He came back without the fan, only with the casual
observation that he couldn’t find it. As he dropped this
cynical confession he looked straight and hard at the
candidate for the honour of taking his education in hand.
(Henry James, The Pupil, 1891/​1909, p. 511)
(19) Mine eye and heart are at a mortal war
How to divide the conquest of thy sight;
Mine eye my heart thy picture’s sight would bar,
My heart mine eye the freedom of that right.
My heart doth plead that thou in him dost lie—​
A closet never pierced with crystal eyes—​
But the defendant doth that plea deny
And says in him thy fair appearance lies.
Shakespeare, sonnet 46, c. 1609/​2002

These are just two very simple examples that help to illustrate the occur-
rence and use of mismatch in the verbal arts. In Henry James (18), the verb
drop is combined with the object noun phrase his confession. In the previ-
ous section we classified this as complement coercion, since drop actually
requires a physical, material object that can be dropped. Similarly, take in
hand also requires a physical, material object, but in this case it is combined
with his education, again requiring complement coercion in order to make

[ 290 ]  Literature and Cognition in Cognitive Science

 

sense. In Shakespeare’s sonnet (19) we see an interesting combination of

what are traditionally termed metaphors and metonymies. Eye and heart
are used metonymically for the two senses of sight and feeling. Both of them
being at war is a metaphor for a discrepancy or conflict between what the
lyrical I sees and what it feels. The use of metaphor and metonymy (which
have been described in terms of mismatch and coercion in the previous sec-
tion) here leads to a powerful poetic effect and a great deal of compression.
Metaphor and metonymy are common devices in literature to create
artistic effects, and we could illustrate this further with an endless num-
ber of different examples. What is more interesting, however, is the ques-
tion of whether metaphor and metonymy are independent mechanisms,
or whether they can be tied in with other, perhaps more general, linguistic
principles. As we have already discussed, the jury is still out on this ques-
tion, but studies are beginning to suggest that metaphor and metonymy
can be elegantly captured as the ‘bottom’ or meaning level of the respec-
tive processes, while coercion adds the equally important ‘top’ or syntax
level that constrains metaphoric or metonymic processes. Coercion, as
Gonzálvez-​Garcia argues, determines and constrains in syntax what kind
of metaphors and metonymies are grammatically possible and impossible.
He illustrates this on the basis of examples such as (20).

(20) Te hací-​a más de tom-​ar cerveza Fede !!!

acc.2sg do-​imprpret.1sg more of take-​inf beer Fede [name]
‘I thought you were more the kind of guy who drinks beer, Fede.’
(http://​twitter.com/​ManoloArmani/​status/​
37303790386290688 [WebCorp], taken from Gonzálvez-​
Garcia, 2011, p. 1311)

Gonzálvez-​Garcia sees (20) as a ‘double metonymy’ or ‘metonymic com-

plex’: ‘habitually drinking beer’ stands for the person habitually drinking
beer, which stands for the person’s habit of drinking beer. That is, we inter-
pret source > target/​source > target, or action for agent for agent’s
habit of performing the action. This metonymic interpretation is
made possible, Gonzálvez-​Garcia argues, by the syntactic configuration
(the coercive context) of hacía in the imperfect (which here signals some
counterfactuality) and its complement, the infinitival phrase in a position
where we usually would expect a regular nominal object. With reference to
Harder (2010), he explains that this illustrates a conceptual build-​up on the
one hand (the more semantic, metonymic aspects) in tandem with a top-​
down assignment of functions and roles through mismatch and coercion.

N or m s , R u l e s , a n d C o e r c i o n   [ 291 ]
 

So what has traditionally been described as metaphor and metonymy in

literary theory and rhetoric can also be captured by mismatch and coercion
accounts. This does not necessarily mean that the latter are superior to
previous accounts. But perhaps they can offer a more general account of
language use.
In particular, from a literary point of view, it is interesting to think about
mismatch and coercion beyond the single sentence (i.e. with larger units).
For example, when we consider Wilfred Owen’s ‘Anthem for Doomed
Youth’ (21), we see that this poem was written as a Petrarchan sonnet with
an English, or Shakespearean, rhyme scheme (ABABCDCD EFFEGG) and
in iambic pentameter.

(22) Anthem for Doomed Youth

What passing-​bells for these who die as cattle?
—​Only the monstrous anger of the guns.
Only the stuttering rifles’ rapid rattle
Can patter out their hasty orisons.
No mockeries now for them; no prayers nor bells;
Nor any voice of mourning save the choirs,—​
The shrill, demented choirs of wailing shells;
And bugles calling for them from sad shires.

What candles may be held to speed them all?

Not in the hands of boys but in their eyes
Shall shine the holy glimmers of goodbyes.
The pallor of girls’ brows shall be their pall;
Their flowers the tenderness of patient minds,
And each slow dusk a drawing-​down of blinds.
Owen, 1917/​1990, p. 74

This particular pattern is usually (but not necessarily) associated with

amorous love, strong emotions, a relationship between lovers, and the
perception and expression of sentiments. We get something very dif-
ferent in ‘Anthem for Doomed Youth’. Owen described the atrocities of
World War I  in quite some gruesome detail. So in some sense, we can
speak of a mismatch here between a form that is commonly associated
with love and related emotions, on the one hand, and content that is
about the atrocities of war, on the other. This mismatch, however, does
not render the product uninterpretable, but rather seems to underline
the strong message of the content by contrasting it with the common
understanding of the form.

[ 292 ]  Literature and Cognition in Cognitive Science

 

Yet another aspect that we need to consider is when poetic language

steps beyond traditional metaphors and metonymy and becomes, at least
superficially, nonsensical. Some examples are given in (22) and (23).

(22) The fall (bababadalgharaghtakamminarronnkonnbronnton

nerronntuonnt-​hunntrovarrhounawnskawntoohoohoord
enenthurnuk!) of a once wallstrait oldparr is retaled early in
bed and later on life down through all christian minstrelsy.
(Joyce, Finnegans Wake, 1939/​2012, p. 3)

(23) Gadji Beri Bimba
gadji beri bimba glandridi laula lonni cadori
gadjama gramma berida bimbala glandri galassassa
laulitalomini
gadji beri bin blassa glassala laula lonni cadorsu sassala bim
gadjama tuffm i zimzalla binban gligla wowolimai bin beri ban
o katalominai rhinozerossola hopsamen laulitalomini hoooo
gadjama rhinozerossola hopsamen
bluku terullala blaulala loooo
Hugo Ball, composed 1916, first published 1928

Do (22) and (23) still count as mismatch and coercion, or should they be
classified as ‘anomalous’? One might argue that, since these utterances
cannot be parsed by regular cognitive and linguistic means, as can exam-
ples (13), (14), and (15c), they may not fall under the rubric of mismatch
and coercion. There is no regular, constrained, and predictable mechanism
that helps to coerce a new meaning out of these combinations. This is not
to say that they are not meaningful. Both Joyce’s Finnegans Wake and Ball’s
Gadji Beri Bimba are, of course, interpretable. But these interpretations
must have recourse to aspects beyond language, or even contextualized
language, itself. The ‘language’ that is used here is not the same as the con-
ventionalized tool of communication that is used in the other examples
(18) and (19) above.
Having looked at mismatch and coercion from the viewpoint of cogni-
tive linguistics and literary studies, in the next section I will turn to cog-
nitive neuroscience, for two main reasons. First, this is one of the fields
where mismatch and coercion have been studied extensively in order to
get a better picture of language processing in the brain. And secondly, the
discovery of unique neurophysiological correlates to coercion would sub-
stantially strengthen the suggestion made earlier, that coercion is actually
a very basic cognitive principle.

N or m s , R u l e s , a n d C o e r c i o n   [ 293 ]
 

COERCION IN NEUROSCIENCE

Mismatch and coercion have been studied extensively in psycholinguistics,

neurolinguistics, and neuroscience. We can broadly distinguish between
three different types of study that have extensively dealt with coercion:

(a) Magnetoencephalography (MEG; see Pylkkänen and McElree, 2007;

Pylkkänen, 2007, 2008; Brennan and Pylkkänen, 2008),
(b) Event-​related potential (ERP; see Kuperberg, Choi, Cohn, Paczynski,
and Jackendoff, 2010),
(c) Functional magnetic resonance imaging (fMRI; see Husband, Kelly,
and Zhu, 2011).

We will look at each of these in turn.

Magnetoencephalography (MEG)

MEG is used for non-​invasive, functional neuroimaging. Magnetic fields,

which are produced by natural electric currents in the brain, are recorded
and mapped as closely as possible onto brain regions in real time. MEG
thus allows for both temporal and spatial brain activity studies. The major
drawback of MEG studies is that MEG signals are extremely small and that
tracing them not only requires a substantial amount of technology but also
is very susceptible to errors of any kind.
Liina Pylkkänen and Brian McElree (2007) used MEG to study possible
differences between regular expressions (as in the (α) examples below),
coerced constructions (β), and anomalous sentences (γ).

(24) The journalist (α) wrote (β) began (γ) astonished the article
before his coffee break.
(25) The baby (α) ate (β) tried (γ) panicked the banana before the
short nap.
(26) The housewife knew that the guests (α) ate (β) tried (γ)
displeased the salmon after the music started.
(27) The nanny said the toddler (α) used (β) mastered (γ) alarmed
the seesaw before his second birthday.

Interestingly, the coerced (β) examples did not modulate any activity in
the traditional language areas such as Broca’s or Wernicke’s area. Rather,
they correlated with increased activity in the ‘anterior midline field’ (AMF),

[ 294 ]  Literature and Cognition in Cognitive Science

 

which came from a midline source in the ventromedial prefrontal cortex

(vmPFC; Pylkkänen and McElree, 2007, p.  1916). The vmPFC is usually
not implicated in language-​related activities, which suggests some extra-​
linguistic factors in the processing of coercion. The vmPFC is active, how-
ever, in the processing of non-​literal meaning in cases of irony and sarcasm,
and has also been discussed in the context of social cognition and theory
of mind. Moreover, the AMF did not show any irregular activity when the
subjects were confronted with anomalous sentences. Still, anomalous sen-
tences triggered an M350 effect. M350 is a reaction in the left temporal
cortex with a peak at around 300‒400 ms. It is usually associated with
automatic lexical retrieval. An N400 effect (i.e. negative peak around 400
ms) also associated with word processing (but strongly affected by factors
such as frequency, predictability, etc.) was not clearly present. This is par-
ticularly interesting given that we usually expect a sharp increase in the
N400 amplitude for lexical items which do not fit into their syntactic con-
text (Hagoort and Brown, 1999, p. 280).
Edward Husband and colleagues (2011), however, report that they had
difficulties in replicating these results. In particular, they did not find the
vmPFC localization of the AMF. They also point out that the localization of
the AMF is still a matter of debate.

Event-​R elated Potential (ERP)

Gina Kuperberg and colleagues (2010) use ERP in their investigation of

coercion. Event-​related potentials are studied through electroencepha-
lography (i.e. the non-​invasive measuring of electric activity in the brain).
This method is related to MEG in that ERP simply measures activity, while
MEG looks at event-​related fields (ERF). In ERP studies, subjects are pre-
sented with certain stimuli (events), and the electrical reaction of the brain
is plotted through time. Just like Pylkkänen and McElree, the authors
presented participants with sets of minimal pair sentences, one of which
contained a transparent, regular expression (The student read the book),
one a construction that requires coercion (The student began the book), and
one an anomalous sentence (The student astonished the book). What they
found was that coerced constructions showed a larger N400 effect than
normal, non-​coerced sentences. This is relatively unsurprising, as the so-​
called N400 effect (a negative peak at about 400 ms) has frequently been
associated with any kind of lexical-​semantic processing. The amplitude of
the N400 is inversely correlated with factors such as frequency and contex-
tual fit: that is, the more frequent and the more plausible (expected) the

N or m s , R u l e s , a n d C o e r c i o n   [ 295 ]
 

stimulus in question, the smaller the amplitude, and vice versa. It is thus
not unexpected that coerced constructions would tend to elicit a higher
N400 amplitude than non-​coerced constructions. Much more interesting
is the fact that Kuperberg and colleagues found no significant differences
in amplitude for coerced constructions within a dominant context (e.g.
The author began the manuscript, where the interpretation began to write is
strongly suggested) and open interpretations (e.g. The man began the book,
which is more open and could plausibly include began to write, to read, to
study …). Anomalous constructions (The author astonished the book) also
showed a similar N400 effect as the coerced stimuli. However, anomalous
stimuli also showed a robust late P600 effect (i.e. a marked positive peak at
about 600 ms when the final word of the sentence—​SFW—​is presented),
which the coerced constructions and the controls did not show. This P600
effect is usually associated with syntactic processing, reanalysis, and high
levels of syntactic complexity and indeterminacy. Kuperberg and colleagues
speculate that the observed N400 effect might be due to the participants
noticing the mismatch between the verb and its complements, and making
(more implicit, automatic) attempts to resolve this mismatch (by coercing
a new reading out of the complement). The observed P600 effect at the
SFW, however, might then be due to more explicit attempts to resolve the
(anomalous) mismatch by constructing unstated specific activities that
could have been implied by the verb-​argument combination.
The N400 is known to be sensitive to a wide array of factors, such as
categorical feature-​based, animacy-​based, and association-​based relation-
ships, including those grounded in real-​world expectations, as well as in
pragmatic relationships. Kuperberg and colleagues therefore also suggest
that the observable effects could be signs of attempts to retrieve unstated
meaning that may lead to plausible interpretations (2010, p. 2698).

Functional Magnetic Resonance Imaging (fMRI)

Functional MRI is also a non-​invasive, dynamic technology, but instead

of electrical currents and related magnetic fields it uses a powerful exter-
nal magnetic field to influence atomic nuclei in the participant’s body, for
instance in the brain. The spatial resolution of this method today is less
than 1mm, and it is thus one of the most precise tools available. Functional
MRI rests on the so-​called blood oxygenation level dependent response.
The key idea is that the brain needs oxygen to function, and the more
work it has to do, the more oxygen is required. This means that we see an
increased blood flow into regions of higher activity. Since oxygenated and

[ 296 ]  Literature and Cognition in Cognitive Science

 

de-​oxygenated blood have different magnetic properties, the distribution

and flow of these two types of blood in the brain can be traced across time
with the use of the external magnetic field. While the spatial resolution for
this method is excellent, the temporal resolution is still relatively poor, as
the signal dependent on blood oxygen level takes up to six seconds to reach
its peak (so that the activity in the brain takes place several seconds before
we are able to trace it). Another important caveat to keep in mind is that,
even though the pictures derived from fMRI look very natural, they are only
abstract representations based on statistical analyses of the different con-
ditions, and a calculated likelihood that these differences are significant. So
we do not actually see brains in action, but rather trust that the mathemat-
ical models accurately reflect ongoing activity. However, even when we see
this activity, this does not necessarily mean that the active area is involved
in the process we want to study. So while fMRI studies have shown some
(surprising) right-​hemisphere activity during language processing, damage
to these particular regions usually does not have dramatic consequences
for linguistic performance. Nevertheless, even with these substantial draw-
backs, fMRI today counts as one of the most popular methods in neurolin-
guistic research (for an excellent and more comprehensive discussion, see
Schlesewsky and Bornkessel-​Schleswesky, 2009, pp. 14‒18).
Husband and colleagues (2011) used fMRI to investigate any differences
in the processing of coerced, non-​coerced, and anomalous constructions,
such as (30)‒(33).

(27) The novelist began the book. (mismatch and coercion)

(28) The novelist wrote the book. (no coercion necessary)
(29) *The novelist sleeps the book. (anomalous, semantic
violation)
(30) *The novelist write the book. (anomalous, syntactic violation)

While reading sentence (30) correlates with stronger activity in Brodman’s

area (BA) 45 in the left inferior frontal gyrus than (31), it is not associated
with any increased activity in the regions typically associated with seman-
tic or syntactic violations (e.g., BA 6, 8, 41, 44). These show a higher level
of activity with anomalous sentences such as (32) and (33), which contain
semantic or syntactic violations. This is particularly interesting in that BA
45 (and BA 44) constitute what is traditionally known as Broca’s area. One
part of this area seems to be affected by coercion (BA 45), the other one not
(BA 44). The authors nevertheless suggest that coercion seems to be part of
the core language faculty. BAs 6 and 8 have been implicated in a relatively
wide range of different tasks, including motor planning and execution,

N or m s , R u l e s , a n d C o e r c i o n   [ 297 ]
 

novel problem-​solving, calculation, and deduction (BA 6), and governance

of eye movements, planning, processing uncertainty, reasoning, calcula-
tion, and logic (BA 8). Some of these more general cognitive skills (novel
problem-​solving, deduction, processing uncertainty, and reasoning) could
obviously also be related to the resolution of mismatch.
Husband and colleagues (2011) conclude that coercion can be interpreted
as a complex compositional operation that naturally leads to greater activ-
ity in language regions, such as the left inferior frontal gyrus. Syntactic
and semantic violations, however, lead to differential activity in a much
broader network of brain regions, including the left ATC and bilateral AG.
This suggests, on the one hand, that left ATC and bilateral AG are sensitive
to these particular aspects (grammaticality and plausibility, respectively),
but perhaps not to compositionality (which is a key issue for the coerced
sentences). On the other hand, this finding also seems to suggest that mis-
matches (but not violations) are parsed as regular linguistic expressions
in the usual linguistic areas, while syntactic and semantic violations and
anomalies are a fundamentally different kind of phenomenon.

Summary of Neurolinguistic Results

In the previous sections I discussed coercion studies using MEG, ERP, and
fMRI. Unsurprisingly, the resulting picture is not entirely uniform or con-
clusive. Table 14.1 summarizes the results.
MEG studies pointed towards increased activity in the ‘anterior mid-
line field’ (AMF), triggered by the ventromedial prefrontal cortex (vmPFC).
The vmPFC is generally implicated not in linguistic reasoning, but rather
in social cognition. These studies also found some M350 effect, but were
unclear about the role of N400. Ultimately, this suggests that coercion
might be a process related to other general cognitive processes outside lan-
guage, in particular social cognition. ERP studies found a clear N400 for
coerced sentences, but not for anomalous sentences. The latter also showed
a marked P600 effect, which is associated with post-​hoc processing of syn-
tactic complexity and anomaly. Coerced sentences did not show this effect.
This in turn points towards coercion being a regular linguistic process that
only requires some degree of additional processing in the mental lexicon.
Anomalous structures seem to be fundamentally different. Finally, fMRI
seems to complement these findings and point out that coercion correlates
with increased activity in the normal language areas, while anomalous sen-
tences are associated with a much broader network of activation, which
could also mean that they need recourse to general cognitive mechanisms.

[ 298 ]  Literature and Cognition in Cognitive Science

 

Table 14.1   SUMMARY OF NEUROLINGUISTIC STUDIES

MEG ERP fMRI

Coercion • no traditional • N400 effect (but • Broca’s area (BA 45) active

language area no difference for • coercion is a compositional
• AMF, vmPFC context) operation
• social cognition,
theory of mind
Anomaly • M350 • N400 effect • BA 6, 8, 41, 44 active
(lexical retrieval) • P600 effect • involves non-​linguistic
knowledge
• non-​compositional operation

SUMMARY AND CONCLUSION

When we try to combine the three aspects of mismatch and coercion that
have been discussed in this chapter (linguistic approaches, literature, neu-
roscience), a number of interesting findings surface. First, it seems plau-
sible to assume that coercion is a very general cognitive mechanism which
helps to resolve mismatch. Mismatch, in turn, is a phenomenon which
tends to be dispreferred (in the sense that mismatch needs to be resolved
as quickly as possible) and leads to an increase in cognitive activity in the
search for meaning. A similar point was already made by Frederic Bartlett in
the 1930s, when he discussed his ‘effort after meaning’ principle (Bartlett,
1932). The idea is that, instead of discarding incongruous information as
nonsense, we tend to perform operations that lead to some sort of sense
even in the face of mismatches and anomaly. We often perceive these cog-
nitive operations as pleasant in some sense, and they seem to be part of the
aesthetic experience. Coercion as a very basic and general cognitive opera-
tion does not make other notions such as metaphor or metonymy redun-
dant. Rather, it seems to form the basis for these phenomena.
Secondly, there seems to be a fundamental difference, linguistically
and neurophysiologically, between mismatches that are susceptible to
coercion, and what I  have termed anomalous structures (i.e. structures
with grammatical or semantic incongruences which cannot be resolved by
coercion). Problem-​solving for these kinds of structures involves other
cognitive structures, a much wider network in the brain, and more
language-​independent thought processes than coercion. Whether the aes-
thetic effects are different for mismatch plus coercion and anomaly remains
to be seen, though at first sight one would expect anomalies to pose greater

N or m s , R u l e s , a n d C o e r c i o n   [ 299 ]
 

problems and perhaps be more quickly rejected than mismatched struc-

tures. Coercion, just like metaphor and metonymy, might also trigger some
reactions in the more generalized cognitive and neural domains, but these
are far less significant than the ones we find in anomalous contexts.

REFERENCES

Addison, J. (1712, 21 June to 3 July). The pleasures of imagination. The spectator,

411–​421.
Ball, H. (1928). Gadji beri bimba. De Stijl (Leiden), 85/​86.
Bartlett, F. (1932). Remembering:  A  study in experimental and social psychology.
Cambridge: Cambridge University Press.
Boas, H. C. (2008). Resolving form‒meaning discrepancies in Construction Grammar.
In Jakko Leino (Ed.), Constructional reorganization (pp. 11–​36). Amsterdam: John
Benjamins.
Brennan, J., and Pylkkänen, L. (2008). Processing events:  Behavioral and neu-
romagnetic correlates of aspectual coercion. Brain and Language, 106(2),
132–​143.
De Almeida, R. G. (2004). The effect of context on the processing of type-​shifting
verbs. Brain and Language, 90(1‒3), 249–​261.
De Almeida, R. G., and Riven, L. (2012). Indeterminacy and coercion effects: Minimal
representations with pragmatic enrichment. In A. M. Di Sciullo (Ed.), Towards
a biolinguistic understanding of grammar:  Essays on interfaces (pp. 277–​301).
Amsterdam: John Benjamins.
Fauconnier, G., and Turner, M. (2002). The way we think. New York: Basic Books.
Fludernik, M. (Ed.). (2011). Beyond cognitive metaphor theory:  Perspectives on literary
metaphor. London: Routledge.
Francis, E. J., and Michaelis, L. A. (Eds.). (2003). Mismatch: Form‒function incongruity
and the architecture of grammar. Stanford, CA: CSLI Publications.
Franzen, J. (2001). The corrections. New York: Picador.
Gadet, F., Léon, J., and Pécheux, M. (1984). Remarques sur la stabilité d’une construc-
tion linguistique: La complétive. Linx, 10(1), 23–​47.
Gibbs, R. (2008). The Cambridge handbook of metaphor and thought. Cambridge: Cambridge
University Press.
Goldberg, A. (1995). Constructions: A construction grammar approach to argument struc-
ture. Chicago: Chicago University Press.
Gonzálvez-​García, F. (2011). Metaphor and metonymy do not render coercion super-
fluous: Evidence from the subjective-​transitive construction. Linguistics, 49(6),
1305–​1358.
Grice, H. P. (1975). Logic and conversation. In D. Davidson and G. Harman (Eds.), The
logic of grammar (pp. 64-​75). Encino, CA: Dickenson.
Hagoort, P., and Brown, C. M. (Eds.). (1999). The neurocognition of language.
Oxford: Oxford University Press.
Harder, P. (2010). Meaning in mind and society: A functional contribution to the social turn
in cognitive sociolinguistics. Berlin: Walter de Gruyter.
Horn, L. R. (2012). Implicature. In L. R. Horn and G. Ward (Eds.) The handbook of prag-
matics (pp. 3-​38). Malden, MA: Blackwell.

[ 300 ]  Literature and Cognition in Cognitive Science

 

Huddleston, R. (2002). The verb. In R. D. Huddleston and G. K. Pullum (Eds.), The

Cambridge grammar of the English language (pp. 71–​212). Cambridge: Cambridge
University Press.
Husband, E. M., Kelly, L. A., and Zhu, D. C. (2011). Using complement coercion to
understand the neural basis of semantic composition: Evidence from an fMRI
study. Journal of Cognitive Neuroscience, 23(11), 3254–​3266.
Jackendoff, R. (1997). The architecture of the language faculty. Cambridge, MA: MIT Press.
James, H. (1909). What Maisie knew; In the cage; The pupil. London: Macmillan.
Joyce, J. (2012). Finnegans wake. Ware, UK: Wordsworth.
Kuperberg, G. R., Choi, A., Cohn, N., Paczynski, M., and Jackendoff, R. (2010).
Electrophysiological correlates of complement coercion. Journal of Cognitive
Neuroscience, 22(12), 2685–​2701.
Langacker, R. (2009). Cognitive (construction) grammar. Cognitive Linguistics, 20(1),
167–​176.
Lauwers, P., and Willems, D. (Eds.). (2011). New reflections on coercion. Special issue,
Linguistics, 49(6).
Miller, C. (2015). Surprise: The poetics of the unexpected from Milton to Austen. Ithaca,
NY: Cornell University Press.
Neisser, U. (2004). Pattern recognition. In D. A. Balota and E. J. Marsh (Eds.), Cognitive
psychology (pp. 125–​150). New York: Psychology Press.
Nunberg, G. (1979). The non-​uniqueness of semantic solutions: Polysemy. Linguistics
and Philosophy, 3(2), 143–​184.
Ortony, A. (Ed.). (2008). Metaphor and thought. Cambridge: Cambridge University Press.
Owen, W. (1990). The poems of Wilfred Owen (J. Stallworthy, Ed.). London: Chatto &
Windus.
Partee, B., and Rooth, M. (1983). Generalized conjunction and type ambiguity. In R.
Bauerle, C. Schwarze, and A. von Stechow (Eds.), Meaning, use and interpretation
of language (pp. 361–​383). Berlin: Walter de Gruyter.
Pustejovsky, J. (1995). The generative lexicon. Cambridge, MA: MIT Press.
Pylkkänen, L., and McElree, B. (2007). An MEG study of silent meaning. Journal of
Cognitive Neuroscience, 19(11), 1905–​1921.
Pylkkänen, L. (2008). Mismatching meanings in brain and behavior. Language and
Linguistics Compass, 2(4), 712–​738.
Schlesewsky, M., and Bornkessel-​Schlesewsky, I. (2009). Processing syntax and morphol-
ogy. Oxford: Oxford University Press.
Shakespeare, W. (2002). Complete sonnets and poems (C. Burrow, Ed.). Oxford: Oxford
University Press.
Silvia, P. J. (2014). Human emotions and aesthetic experience: An overview of empirical
aesthetics. In A. P. Shimamura and S. E. Palmer (Eds.), Aesthetic science: Connecting
mind, brains, and experience (pp. 250–​275). Oxford: Oxford University Press.
Schmidhuber, J. (2009). Simple algorithmic theory of subjective beauty, novelty, sur-
prise, interestingness, attention, curiosity, creativity, art, science, music, jokes.
Journal of SICE, 48(1), 21–​32.
Talmy, L. (2000). Toward a cognitive semantics. Cambridge, MA: MIT Press.
Tomasello, M. (2006). Construction grammar for kids. Constructions, SV1-​11/​2006.
Retrieved from http://​journals.linguisticsociety.org/​elanguage/​constructions/​
article/​view/​26.html
Twain, M. (1994). The adventures of Huckleberry Finn. New York: Dover Thrift.
Ziegler, D. (2007). A word of caution on coercion. Journal of Pragmatics, 39(5),
990–​1028.

N or m s , R u l e s , a n d C o e r c i o n   [ 301 ]
 
 

CHAPTER 15

Affective and Aesthetic Processes

in Literary Reading
A Neurocognitive Poetics Perspective
ART HUR M. JACOBS

Metaphorically, the page thus stands to the reader as a flower to a bee, and not, as
one might have argued, as a picture of a flower to a bee: flat, odorless, and hopelessly
unchanging.
Nell, 1988, p. 38

INTRODUCTION

While the field of (psycho)linguistics has traditionally neglected emo-

tional aspects of language from de Saussure to Chomsky, most emotion
theories in psychology are silent with regard to language-​related processes
(Schrott and Jacobs, 2011). This ‘double neglect’ is astounding given that
Karl Bühler’s (1934) magnificent but almost forgotten ‘Language Theory’
already specified affective aspects of language as found in his better-​known
‘organon model’, later extended by Roman Jakobson (1960). Moreover, my
own major scientific playground, experimental reading research, has—​ever
since the days of James McKeen Cattell in Wundt’s lab—​shed a lot of light
on the information processing that goes on while people move their eyes
about 3‒5 times per second across printed symbols they often took years
to learn. What remains much more in the shade, though, are the affective
 

and aesthetic processes that without doubt constitute a significant part

of the reading act (Jacobs et  al., 2015). On the other hand, a large body
of literature from the humanities, published in outlets like Poetics, Poetics
Today, Scientific Study of Literature (SSOL), and numerous book chapters
and monographs, provides valuable insights and empirical data on affec-
tive and aesthetic processes in literary reading, and a number of recent
works have started to integrate this body of knowledge with neurocog-
nitive research on reading (e.g., Burke, 2011, 2015; Jacobs, 2011, 2015a,
2015b; Schrott and Jacobs, 2011).
In this chapter, I  shall outline a neurocognitive poetics perspective on
literary reading which has emerged from my interactions with scholars
from the humanities such as the Austrian poet Raoul Schrott and has led
to a number of cross-​disciplinary publications like our book on Brain and
Poetry (Schrott and Jacobs, 2011). I start with a ‘mountain climb’ presenting
a series of studies on progressively larger text units, many of which were run
in the context of the Free University of Berlin’s transdisciplinary research
cluster ‘Languages of Emotion’ (Gebauer and Edler, 2014). These studies
answer some key questions about literary reading and raise some new ones.
After reaching the ‘summit’ of this climb, I  end with a note on theoreti-
cal and methodological desiderata of the scientific study of literary reading,
inspired by two recent comprehensive and complementary frameworks.

A MOUNTAIN CLIMB: FROM WORD VALLEYS

TO THE SUMMITS OF POEMS AND NOVELS

If word recognition is the central process underlying the reading skill, then
it makes sense to start any theory of literary reading with that miraculous
performance of the human mind (Jacobs, 2001, 2011). If word recognition is
impaired, as in dyslexic and alexic patients who can only read very slowly or
letter by letter, respectively, reading longer pieces of text becomes difficult
or impossible and the pleasures offered by reading poems or novels become
aloof (Jacobs, 2001; Jacobs and Ziegler, 2015). The key to these pleasures is
acquired early in life and depends on genetic as much as on environmental
factors. The fact that the meaning of words is first learned through the ear
has important consequences for visual word recognition and literary read-
ing in general: it is highly likely that even in silent reading most words are
automatically phonologically (and prosodically) recoded even in proficient
readers, and that lexical meaning construction requires a prior activation
or coactivation of the sound of syllables and words (Conrad, Stenneken,
and Jacobs, 2006; Jacobs and Ziegler, 2015). If the sound echo of written

[ 304 ]  Literature and Cognition in Cognitive Science

 

words is so important for meaning construction, it is easy to understand

Valéry’s adage that poetry is a ‘prolonged hesitation between sound and
meaning’ (1941/​1960, p. 636).
The sound gestalt of words is what stabilizes their mental representa-
tion in working memory for purposes of reading and thinking when the
eyes have moved to the next word in the text and the orthographic repre-
sentation of the previously fixated ones has already decayed; it thus con-
stitutes the basis for making meaning from word pairs, phrases or verses
(Jacobs and Kinder, 2015). But word sound echoes too have a limited men-
tal life during reading unless they are refreshed by stylistic tricks of asso-
nance, alliteration, or rhyme (see below). Larger text units are coded by
the brain in a semantic way using event or situation models (Bailey and
Zacks, 2011). It is generally agreed that text information is coded at three
levels:  the surface form (i.e. verbatim memory of the exact word form,
sound, and order), the textbase (i.e. a memory for the abstract proposi-
tions/​ideas that were contained in the text itself), and the situation model
(i.e. a referential representation of the events described by the text (van
Dijk and Kintsch, 1983). Each level appears to have its own forgetting
curve (Copeland, Radvansky, and Goodwin, 2009). Thus, different mecha-
nisms seem at work when short and long texts are processed, and research
on literary reading should strive to integrate micro-​, meso-​, and macro-
scopic aspects in a comprehensive theoretical framework. My approach
here resembles a mountain climber: I start from the bottom, move up to
the summit, and look down to the valley again. The approach parallels my
own scientific pathway: starting with visual psychophysics, visual search,
and eye-​movement control, spending a long period as a ‘word nerd’, and
finally moving up to text processing and poetry reception. The ultimate
goal remained the same during all these periods, namely, understanding
this perhaps greatest achievement of human civilization and most complex
performance of the human brain: literary reading.

WORD VALLEYS

LIBELLE is the German word for dragonfly and was once selected from
among many thousands of proposals from German-​ speaking people
all over the world as the most beautiful word of the year for children
(Limbach, 2004). In Limbach’s book, a nine-​year-​old child describes
the micropoetry hidden in this three-​syllable word in a very clear, sim-
ple statement—​at the most studied of all levels of observation in liter-
ary reading studies, the subjective experiential one assessed by explicit

A Neurocogni tive Poetics Perspective  [ 305 ]

 

verbal reports. He mentions three elementary cues for the beauty

of words:  (1)  an articulatory-​phonological cue (the gliding l’s), (2)  a
sensorimotor-​perceptual cue (the wobbling of the dragonfly), and (3) an
affective-​semantic cue (no fear of this beautiful animal; for details, see
Jacobs, 2015b). All three cue types can be quantified with the Berlin
Affective Word List (BAWL; Võ, Jacobs, and Conrad, 2006; Võ, Conrad,
Kuchinke, Hartfeld, Hofmann, and Jacobs, 2009). This supports the view
that associations with both discrete emotions and embodied cognitions
play a role in the ‘normal’ processing and the aesthetic appreciation of
words, as already hypothesized in Bühler’s (1934) notion of the ‘spheric
fragrance’ of words (Jacobs, 2011, 2015b). A study involving adults con-
firmed the child’s intuitions empirically:  Libelle was rated as the most
beautiful word in the BAWL corpus containing 450 beautiful and ugly
German words (Jacobs et al., 2015). As argued elsewhere, I believe that
this micropoetic experience with single words is the beginning of later
immersive and aesthetic experiences with larger pieces of text, and that
understanding the underlying word recognition process(es) is a key ele-
ment of any theory of prose or poetry reception (Schrott and Jacobs,
2011; Jacobs and Kinder, 2015). However, despite their descriptive accu-
racy and explanatory power for many aspects of reading, current compu-
tational models of word recognition are still unable to explain the how
and why of this phenomenon, because they ignore affective and aesthetic
aspects (e.g. Grainger and Jacobs, 1996; Hofmann and Jacobs, 2014).
Unfortunately, in contrast to affective word processing (for reviews,
see Kuchinke, Trapp, Jacobs, and Leder, 2005; Citron, 2012; Ponz et al.,
2014), there seems to be only one pilot study on aesthetic processes in
single word recognition which could inform and constrain such models
(Jacobs et al., 2015).
DUFTGESANG (perfume chant) is a German noun-​noun compound
(NNC) and was part of the novel metaphor group of 200 NNCs studied
by Bálint Forgács and colleagues (2012) to discover the neural correlates
of combinatorial semantic processing of literal and figurative stimuli.
Together with the other 49 words of its group, Duftgesang triggered
much more brain activity in a region known to be involved in reading,
the left inferior frontal gyrus (LIFG), than words like Glasaffe (glass
monkey) in the novel literal NNC group, Augapfel (apple of the eye)
from the conventional metaphor NNC group, or Alarmsignal (alarm sig-
nal) in the conventional literal NNC group. ABSCHAUMKNOSPE (scum
bud) is an example of an affectively bivalent NNC (i.e. a compound
whose first constituent has a negative valence, while its second has a
positive one), as assessed by the BAWL (Jacobs et al., 2015). Using all

[ 306 ]  Literature and Cognition in Cognitive Science

 

four combinations of such NNCs (negative-​positive, negative-​nega-

tive, etc.), Michael Kuhlmann and colleagues (2016) recently confirmed
and extended the results by Forgács and colleagues (2012) in showing
that the LIFG is also a key structure in affective meaning-​making (i.e.
when subjects have to decide on the overall valence of a bivalent word).
Making meaning out of novel or ambiguous verbal stimuli is part and
parcel of the wonders of literary reading and poetry reception in par-
ticular, but we are only at the beginning of understanding its neuronal
and cognitive-​affective bases.
Word pairs can have metaphoric and bivalent relations, but they also
feature rhyme, like ‘double trouble’, or alliterations, like ‘whisper words of
wisdom’. This provides them with increased memorability and pleasantness
even when being separated across larger units of text as in Shakespeare’s
popular sonnet 12:

When I do count the clock that tells the time,

And see the brave day sunk in hideous night;
When I behold the violet past prime,
And sable curls, all silvered o’er with white

The assonance to the sound gestalt of the base word (time) provides a
rhyme with the creative power of evoking—​by association—​that other
word’s semantic field and contrasting or fusing it with its own (Schrott
and Jacobs, 2011). This associative process likely constitutes the most
basic skill underlying creativity and poetic experiences in (figurative)
language reception and production, namely the ability to discover hid-
den similarities in word pairs, idioms, proverbs, puns, metaphors, or
verses. Koestler (1964) called it bisociative thinking, a process allowing the
discovery of a relationship between one object or pattern and another
object or pattern. Perhaps the activity uncovered in the LIFG during the
processing of novel metaphoric NNCs (Forgács et al., 2012; Kuhlmann et
al., 2016) is a neural marker of bisociative thinking. This possibility is sup-
posed by a study by Barbara Rutter and colleagues (2012) on conceptual
expansion (i.e. the extension of an existing concept to include new fea-
tures and attributes, thereby widening its original definition) during the
processing of metaphoric sentences like ‘The clouds have danced over the
city’: again, the LIFG showed significantly increased activation relative to
control conditions. Given that the study by Isabel Bohrn and colleagues
(2012b) discussed next also found increased LIFG activity for defamiliar-
ized proverbs, the LIFG bisociative thinking hypothesis appears to warrant
further research.

A Neurocogni tive Poetics Perspective  [ 307 ]

 

SENTENCE SLOPES: IDIOMS AND PROVERBS

JEMANDEM SEIN HERZ AUSSCHÜTTEN (to pour out one’s heart to some-
one) is a German idiom from the PANIG corpus (Citron, Cacciari, Kucharski,
Beck, Conrad, and Jacobs, 2015)  meaning ‘to talk openly with someone
about one’s problems’ and part of a study in which we wanted to learn more
about the role of figurative language in conveying affect. The results sup-
ported the idea that figurative expressions are more emotionally engaging
than literal expressions (Citron and Goldberg, 2014) and add another step-
ping stone to our mountain-​climbing adventure from affective-​aesthetic
single word processing to literary experiences with entire poems or novels.
WER WAGT, GEWINNT (who dares, wins) is a familiar German proverb
used in a study we ran (Bohrn et al., 2012b; Bohrn, Altmann, Lubrich,
Menninghaus, and Jacobs, 2013; see also Menninghaus, Bohrn, Knoop,
Kotz, Schlotz, and Jacobs, 2015) in order to discover the neural correlates
of defamiliarization effects, a key element of foregrounding theory (van
Peer, 1986). ‘Wer klagt, gewinnt’ (who laments, wins) is a defamiliarized,
artful variation of this proverb (keeping rhyme and rhythm of the original,
but changing the meaning by way of substituting only two letters), called
anti-​proverb (Mieder, 2008). Contrary to our expectations, when partici-
pants rated groups of familiar proverbs together with ‘anti-​proverbs’ (and
other control conditions), overall they preferred the former over their art-
ful adaptations (Bohrn et al., 2012b). This confirmed the standard finding
from empirical and theoretical (neuro)aesthetics that familiarity is a key
element of beauty and aesthetic liking (Leder, Belke, Oeberst, and Augustin,
2004; Leder, Gerger, Dressler, and Schabmann, 2012; Leder, Markey, and
Pelowski, 2015; Reber, Schwartz, and Winkielman, 2004; Kuchinke et
al., 2009; Leder, 2013). However, since in our (anti-​)proverb corpus, only
about 30% of variance in beauty ratings was accounted for by familiarity
(Bohrn et al., 2013), that leaves about 70% of variance unexplained and
thus a lot of space for theorizing in neurocognitive poetics (Jacobs, 2015b).
When correlating the individual beauty ratings with functional neuroim-
aging data, we discovered that some spontaneous aesthetic evaluation
takes place during reading, even if not required by the task (silent reading).
Positive correlations were found in the dorsal striatum of the basal ganglia
(i.e. the caudate nucleus, a key structure of the dopaminergic system) and
in medial prefrontal cortex, likely reflecting the rewarding nature of sen-
tences that are aesthetically pleasing. Interestingly, a study on sentences
containing functional shifts (i.e. the use of a semantically appropriate word
in a syntactically inappropriate role) taken from plays by Shakespeare, like
‘He was no longer alone in the world; he was wived to a kind and beautiful

[ 308 ]  Literature and Cognition in Cognitive Science

 

woman’, also found increased caudate nucleus activity in such sentences

(Keidel, Davis, Gonzalez-​Diaz, Martin, and Thierry, 2013). The authors
interpreted this as a sign of disrupted automatic syntactic and semantic
integration.
Together with results from other studies, those by Bohrn and colleagues
(2012b, 2013) support the Panksepp-​Jakobson hypothesis of my ‘neurocog-
nitive poetics model’ (NCPM; Jacobs, 2011, 2014, 2015b), which is needed
to bridge the language-​emotion gap mentioned in the Introduction (see
also Panksepp, 2008; Schrott and Jacobs, 2011). The hypothesis states
that since evolution had no time to invent a proper neuronal system for
art reception, even less so for literary reading, the affective and aesthetic
processes we experience when reading (as in Jakobson’s ‘poetic function’)
must be linked to the ancient emotion circuits we share with all mam-
mals, as perhaps best described by Panksepp (1998). Thus, when partici-
pants experience and rate words or text passages as ‘fearful’, ‘disgusting’,
or ‘beautiful’, neuronal networks systematically associated with fear and
disgust (e.g. the amygdala and insula) or reward and pleasure (e.g. the stri-
atum, orbito-​frontal cortex, and medial prefrontal cortex) should be more
active than in appropriate (neutral) control conditions, which is what was
found in our proverb study.
As evidenced by the above examples, we know a bit about affective
and aesthetic processing of figurative sentences (for a review, see Bohrn
et  al., 2012a), and we also know a lot about oculomotor and cognitive
processes underlying literal sentence processing, such as word frequency
and predictability effects on ERPs and eye movements (e.g. Dambacher,
Kliegl, Hofmann, and Jacobs, 2006; Dambacher, Rolfs, Göellner, Kliegl,
and Jacobs, 2009; Dimigen, Sommer, Hohlfeld, Jacobs, and Kliegl, 2011;
Dambacher, Dimigen, Braun, Wille, Jacobs, and Kliegl, 2012). Experimental
studies on emotional processing of literal (isolated) sentences are becoming
more popular (e.g. Scott, O’Donnell, and Sereno, 2012; Sheikh and Titone,
2013; Lüdtke and Jacobs, 2015), and some studies have also looked at emo-
tional processing of sentence combinations (see my section ‘Passage Hills’).

VERSE LIFTS

ES IST EIN BRAUCH, VON ALTERS HER: WER SORGEN HAT, HAT AUCH
LIKÖR! (From ancient times it has been true, /​He who has cares, has
liquor, too) is a couplet from the popular German humorist and poet
Wilhelm Busch (1832‒1908), whose rhymed and metred narratives have
been published in several languages including English. As shown in a study

A Neurocogni tive Poetics Perspective  [ 309 ]

 

by Winfried Menninghaus and colleagues (2014), the special rhyme and

metre features of this and similar verses evoke humour and positive
affective responses, facilitate understanding, and are better memorized
than de-​rhetoricized control stimuli, in which rhyme and/​or metre were
removed while keeping the overall meaning constant. To account for
the non-​semantic effects of funniness that are based on the processing
of formal incongruities between the historically predominant proto-
type of ‘good’ verse-​making and Busch’s humorous verses, the authors
offer what I would like to call here the Bergson hypothesis, establishing an
analogy between the metrical pattern of verses and dance movements
(Menninghaus, 1991): it states that ‘mechanically’ degrading the phono-
logical and prosodic gestalt of prototypical poetic verses (i.e. Busch’s tech-
nique of deliberate attacks on a reader’s aesthetic expectations as shaped
by the canonical art of poetry) has similar effects to body movements
which appear to degrade a living organism into a mere ‘mechanically’
moving object, namely laughter, regardless of whether such movements
are performed unintentionally or as a deliberate clownish performance
(Bergson, 1911). It would be interesting to examine to what extent sen-
sorimotor networks of the brain associated with rhythmic patterns in
general are responsive to such poetic stimuli, and thus to see how deeply
the embodied qualities of literary reading reach into neural circuitry (e.g.
Hogan, 2003; Burke, 2011; Schrott and Jacobs, 2011).

STANZA RISES

Figure 15.1 shows my eye movements and gaze fixations while reading the
first part of the love poem ‘Wo hast du all die Schönheit hergenommen’
(Where did you get all this beauty) by the German poet Ricarda Huch on
a computer screen in one of the eye-​tracking labs of the Dahlem Institute
for Neuroimaging of Emotion (D.I.N.E.). The places where my gaze stopped
to allow the brain to take in the visual information required for achieving
the ultimate goal of reading, making meaning, are indicated by the circles.
Their size codes the duration of these fixations. The lines indicate the sac-
cades that propel the gaze forwards or backward to the next stop. During
these saccades I was virtually blind, while during the roughly 50 stops on
the eight lines, 61 words, and 84 syllables (11 or 10 per line), lasting 250 ms
on average, my brain not only went through the highly automated routines
of word recognition and sentence comprehension but also computed the
next landing point on the line, anticipated and preprocessed the next word
and/​or sentence, re-​activated (and partly re-​enacted) memories generating

[ 310 ]  Literature and Cognition in Cognitive Science

 

Figure 15.1  Eye-​movement pattern (of this author) while reading the love poem ‘Wo hast
du all die Schönheit hergenommen’ (Where did you get all this beauty) by the German poet
Ricarda Huch (see text for details).

emotions and (reflective) thoughts, and did a myriad of things I was not
aware of and will never be.
A lot of—​as yet unanswered—​questions can be generated from the
‘gaze blobs’ pattern of Figure 15.1 at all levels of text, context, and reader
analysis. For example, what did I  know about Ricarda Huch and what
mood was I in when I chose the poem? What would change if I read the
poem from a book in my favourite chair at home? How different would
my son’s gaze pattern look (context and reader analysis)? Why did the
poet choose a poem form with an alternating 11/​10 syllables per line or
ABAB rhyme structure (text analysis)? Why did I read this poem with a
rate of approximately 250 words per minute (wpm) (i.e. in a relatively
slow mode)? Why did my gaze involuntarily (i.e. without my conscious
control) stop only twice on the word BEAUTY but four times on the word
LIEBESANGESICHT (face of love). Why do I like the poem, especially line
three (reader response analysis)?
What is needed to answer these and related questions are studies on
poetry reading using eye-​movement recording technology, but very few
exist (though see Koops van t’Jagt, Hoeks, Dorleijn, and Hendricks, 2014),
and even in my own lab, we are only beginning to adopt a more system-
atic approach (for a study measuring pupil size variation while listening to
limericks, see Scheepers, Mohr, Fischer, and Roberts, 2013). There is some
literature, though, reporting behavioural and neuronal measures during
the reading of stanzas (e.g. Carminati, Stabler, Roberts, and Fischer, 2006;
O’Sullivan et al., 2015) and peripheral-​physiological measures during the

A Neurocogni tive Poetics Perspective  [ 311 ]

 

reading of entire poems (Jacobs et al., 2016). The latter will be discussed in
the section ‘Poem Mountains’.

PASSAGE HILLS

(a) Screams of terror rent the air:  the fighters scattered, Death Eaters and
Hogwartians alike, and red and green jets of light flew into the midst of the
oncoming monsters, which shuddered and reared, more terrifying than ever.
Rowling, 2007/​2014, p. 528
(b) Hagrid helped Harry on the train that would take him back to the
Dursleys, then handed him an envelope.
‘Yer ticket fer Hogwarts’, he said, ‘First o’ September—​Kings Cross—​it’s all
on yer ticket. …’
Rowling, 1997/​1999, p. 87

When participants read passages like the ones taken from the Harry Potter
series, a specific part of their brain—​in the mid-​cingulate cortex—​showed
selective activity which may be a neural correlate of the perhaps most amaz-
ing of all phenomena related to the reading act, immersion (Hsu, Conrad,
and Jacobs, 2014). The observation that this activity is higher in ‘fear-​
inducing’ passages (a) than in emotionally ‘neutral’ ones (b) is evidence for
two key hypotheses of the NCPM: the Panksepp-​Jakobson hypothesis, men-
tioned earlier, and the fiction feeling hypothesis (Jacobs, 2015b). The latter
states that narratives with emotional contents invite readers more to be
empathic with the protagonists and to become more immersed in the text
world, including through engagement of the affective empathy network of
the brain (mainly the anterior insula and midcingulate cortex), than do sto-
ries with neutral contents. The hypothesis was tested in several studies from
my group using short narratives that were constructed to induce empathy
and emotions like fear and joy, as compared to neutral passages:

(c) Florian and his father are making a model air plane fly together. The airplane
crashes down and is broken. Florian starts to cry. The father ignores him.
(d) Jens is standing at a river and can’t get to the other side. He takes a saw
and cuts down a tree. Jens carries the heavy tree to the edge. He lays the tree
over the river and balances to the other side.

In a first study, Brink et al. (2011) had four-​to eight-​year-​old children lis-
ten to a series of ‘micro-​stories’ like the ones in (c) and (d), either eliciting
affective and cognitive empathy (c) or depicting neutral scenes which relied

[ 312 ]  Literature and Cognition in Cognitive Science

 

on the understanding of physical causalities (d), while their brain activity

was measured with functional Near-​Infrared Spectroscopy. Children’s pro-
cessing of stories of type (c) was associated with medial and bilateral orbi-
tofrontal cortex activation, brain regions known to play a role in Theory of
Mind (ToM) and emotions.

(e) A farmer steered his harvester into a cornfield where his children were play-
ing hide-​and-​seek. Suddenly the machine seemed stuck, so he got off to find the
fault. When he realized that he had run over his children, he took his own life.

In a second study comparing neutral narratives with those like the one
in (e), taken from the popular game Black Stories, Ulrike Altmann and
colleagues (2012, 2014) looked at whether readers’ affective mentalizing
networks were more likely to be activated in short stories with negative
emotional contents than in stories with neutral valence. The results cor-
roborated both aforementioned hypotheses, showing that with increas-
ingly negative content, stories engaged the bilateral inferior frontal gyrus
and additional subcortical structures commonly involved in emotion pro-
cessing, such as the caudate body, or (left) amygdala. Interestingly, in
this study we also discovered that empathy and immersion can depend
on reader personality: the bilateral anterior insula and the right poste-
rior cingulate cortex showed a stronger activity (coupled with medial pre-
frontal cortex) for individuals who reported a stronger tendency to feel
concern for other people, as assessed by a self-​report scale of empathy.

STORY KNOLLS

(f) As my old father now stooped down to the fire, he looked quite another man.
A frightful convulsive pain seemed to have distorted his mild reverend features
into a hideous repulsive diabolical countenance. He looked like Coppelius: the
latter was brandishing red hot tongs, and with them taking shining masses bus-
ily out of the thick smoke, which he afterwards hammered. It seemed to me, as
if I  saw human faces around without any eyes—​but with deep holes instead.
‘Eyes here, eyes!’ said Coppelius in a dull roaring voice. Overcome by the wildest
terror, I shrieked out, and fell from my hiding place upon the floor. Coppelius
seized me, and showing his teeth, bleated out, ‘Ah—​ little wretch,—​
little
wretch!’—​then dragging me up, he flung me on the hearth, where the fire began
to singe my hair. ‘Now we have eyes enough—​a pretty pair of child’s eyes.’ Thus
whispered Coppelius and taking out of the flame some red-​hot grains with his
fists, he was about to sprinkle them in my eyes.
Hoffman, 1816/​1844, p. 144

A Neurocogni tive Poetics Perspective  [ 313 ]

 

(g) Now will I tell you what has befallen me. I must do so, that I plainly see—​
but if I only think of it, it will laugh out of me like mad. Ah, my dear Lothaire,
how shall I begin it? How shall I make you in any way sensible that that which
occurred to me a few days ago could really have such a fatal effect on my life? If
you were here you could see for yourself, but now you will certainly take me for
a crazy ghost-​seer. In a word, the horrible thing which happened to me, and the
painful impression of which I in vain endeavour to escape, is nothing more than
this; that some days ago, namely on the 30th of October, at twelve o’clock at
noon, a barometer-​dealer came into my room and offered me his wares. I bought
nothing, and threatened to throw him down stairs, upon which he took himself
off of his own accord.
You suspect that only relations of the most peculiar kind, and exerting the
greatest influence over my life can give any import to this occurrence, nay, that
the person of that unlucky dealer must have a hostile effect upon me. So it is,
indeed. I collect myself with all my might, that patiently and quietly I may tell
you so much of my early youth as will bring all plainly and clearly in bright images
before your active mind. As I am about to begin I fancy that I hear you laughing
and Clara saying: ‘Childish stories indeed!’ Laugh at me I beseech you, laugh with
all your heart. But, heavens, my hair stands on end, and it seems as if I am asking
you to laugh at me, in mad despair, as Franz Moor asked Daniel. But to my story.
Hoffman, 1816/​1844, pp. 140‒141

The texts in (f) and (g) are two segments from a total of 65 representing

a shortened version of the narrative ‘The Sandman’ (totalling 6,859 from
12,232 words), a prominent example of a Romantic story devoted to the
darker sides of emotional life. It relates events in the life of the student
Nathaniel who, traumatized by the early death of his father—​has been
haunted since childhood by the mysterious Sandman. The story was chosen
because of its suspenseful character and uncanny atmosphere, famously
discussed in Freud’s (1919/​1953‒1974) essay ‘The Uncanny’.
In a first study on neural correlates of suspense, one of three ‘univer-
sals’ of narrative (Sternberg, 2003; see also Jacobs, 2015b),1 Lehne and
colleagues (2015) had participants read the entire narrative, segment by
segment, while their brain activity and peripheral physiological indicators
(heart and respiration rate) were recorded in the scanner. Participants
rated each segment on a suspense scale to allow for correlations of the

1.  According to Sternberg (2003), suspense, curiosity, and surprise constitute the
universals of narrative, each encoding a distinct functional operation of the mind
within narrative’s overall intersequencing (i.e. the dynamics of prospection, retrospec-
tion, and recognition, respectively).

[ 314 ]  Literature and Cognition in Cognitive Science

 

subjective experiential data with the neural ones. In subsequent studies

in our lab the same text segments were rated on several other scales, such
as immersion and fear (Jacobs and Lüdtke, 2016). Text (f) was found to
be the most immersive, suspenseful, and fear-​inducing, while (g) scored
lowest on these three scales. Individual ratings of experienced suspense
were found to be related to activation in the medial frontal cortex, bilat-
eral frontal regions (along the inferior frontal sulcus), and lateral premo-
tor cortex, as well as posterior temporal and temporoparietal areas. The
results suggest that the ‘emotional’ experience of suspense depends on
brain areas associated with social cognition and predictive inference. The
high correlations between suspense, fear, and immersion ratings cross-​
validate our results obtained for the Black Stories summarized earlier and
thus are additional support for the fiction feeling hypothesis of the NCPM.
In other studies, we have investigated affective and aesthetic processes
during the reading of texts by Jean Genet (Lubrich, Knoop, and Jacobs,
2014) and Heinrich von Kleist (Engel, Jacobs, Lehne, Menninghaus, and
Koelsch, 2016), and it has become clear that interlexical and supralexical
variables play as big a role as do properties of single words, making it nec-
essary to operationalize key intervening factors at all levels of structural
description (Jacobs, 2015b).

POEM MOUNTAINS

Vom Taue glänzt der Rasen; beweglicher

Eilt schon die wache Quelle; die Buche neigt
Ihr schwankes Haupt und im Geblätter
Rauscht es und schimmert; und um die grauen

Gewölke streifen rötliche Flammen dort,

Verkündende, sie wallen geräuschlos auf;
Wie Fluten am Gestade, wogen
Höher und höher die Wandelbaren.

Komm nun, o komm, und eile mir nicht zu schnell,

Du goldner Tag, zum Gipfel des Himmels fort!
Denn offner fliegt, vertrauter dir mein
Auge, du Freudiger! zu, solang du

In deiner Schöne jugendlich blickst und noch

Zu herrlich nicht, zu stolz mir geworden bist;

A Neurocogni tive Poetics Perspective  [ 315 ]

 

Du möchtest immer eilen, könnt ich,

Göttlicher Wandrer, mit dir!—​doch lächelst

Des frohen Übermütigen du, daß er

Dir gleichen möchte; segne mir lieber dann
Mein sterblich Tun und heitre wieder
Gütiger! heute den stillen Pfad mir.

These 20 lines form Friedrich Hölderlin’s poem ‘Des Morgens’ (‘In the
morning’, 1799), a remarkable five (stanza) by four (line) ode of the alcaic
type with alternating rising and falling rhythmic periods and numerous
enjambements, no rhyme, rather vivid personifying adjectival imagery,
strongly amplifying valence, archaizing tone (language/​grammar), and a
‘me-​subjectivity’. In a recent study on the reception of 24 German ‘mood
poems’ from three centuries (Jacobs et al., 2016), this was the one poem
that induced conspicuously increased heart rate variability in readers, an
indicator of emotional intensity and suspense during the processing of spo-
ken and written narratives (Wallentin, Nielsen, Vuust, Dohn, Roepstorff,
and Lund, 2011; Lehne et al., 2015). Whether this change in (para-​)sym-
pathetic activity is due to increased efforts of meaning-​making and/​or a
heightened aesthetic feeling is a hot open question for future studies.
In related work on the reception of poems from the volume ‘verteidi-
gung der wölfe’ (defence of the wolves) by the German poet Hans Magnus
Enzensberger (b. 1957), we succeeded for the first time in predicting the
basic affective tone of poems (e.g. sad or friendly) based on an operational
definition of (internal) sound that allows a quantitative, statistic valida-
tion by use of the Emophon algorithm (Aryani, Jacobs, and Conrad, 2013;
Aryani, Kraxenberger, Ullrich, Jacobs, and Conrad, 2015). We interpret
this as evidence that the iconic associations of foregrounded phonologi-
cal units contribute significantly to the emotional and aesthetic percep-
tion of a poem by the reader and the author, as assumed by Jakobson
(1960).
Comparing the processing of different variants of prose (functional vs.
evocative) and poetry (accessible, difficult, and self-​selected, i.e. brought
to the lab by participants themselves) using fMRI, Adam Zeman and col-
leagues (2013) found that brain activation increased with increasing ‘liter-
ariness’ in predominantly left-​sided regions, including the LIFG and areas
of the basal ganglia. The differential activation in the left hemisphere by
literariness was interpreted in line with evidence that these structures
are engaged by complex syntax and semantic ambiguity, and supports the
above-​mentioned role played by the LIFG.

[ 316 ]  Literature and Cognition in Cognitive Science

 

NOVEL SUMMITS

As far as I  know, no study has really yet investigated the reading of an

entire novel, but some come close to it. In Victor Nell’s (1988) famous
series of studies on ludic reading, 33 participants ‘sampled’ the first 50
pages of three enjoyable books they had brought to the lab, and then read
on for about 30 minutes while their reading speed and various peripheral-​
physiological variables (e.g. heart and respiration rate, electrodermal activ-
ity) were monitored (studies 2 and 4). Of the many results of these studies,
the increase in heart rate (relative to some control conditions) during the
reading of most-​liked pages is of particular interest, since it parallels find-
ings from our lab indicating that poems inducing the mood of a morning
(Lüdtke, 2013; Lüdtke, Meyer-​Sickendiek, and Jacobs, 2014), which yield
high aesthetic liking ratings, also produced higher heart rates and electro-
dermal activity (Jacobs et al., 2015). Another notable result of Nell’s (1988)
study is that the so-​called flexibility ratio, that is, the quotient of reading
speed (in wpm) for the fastest-​read passage divided by that for the slowest-​
read passage, was 2.63 on average (average speed being 409 wpm), sug-
gesting that readers do indeed ‘bolt’ and ‘savour’, a marker of ludic reading.
However, although Nell found a significant decrease in speed for most-​liked
pages, this finding is inconclusive, because with his data it was not possible
to determine whether the slowing arose because of a reduction in read-
ing rate, or because these passages were reread once or more. Thus, more
research—​ideally using eye tracking—​is needed to test the hypothesis that
pleasurable passages slow down reading compared to neutral ones. In this
context, Nell’s (1988) distinction between ‘rauding’ (reading with full com-
prehension of each thought) and ‘skimming’ (partial or disengaged reading
with speeds of 600‒800 wpm) is important:

Because the primary vehicle for ludic reading is formulaic fiction … —​that is,
long, continuous texts of moderate difficulty … and high predictability—​‘bolt-
ing’ the text is feasible, because experienced readers have little difficulty captur-
ing the gist of the material by skimming it. (p. 20)

In contrast, my own reading speed from Figure 15.1 above (250 wpm)
would thus qualify as ‘rauding’ or even as intensified ‘rauding’ during the
reading of mainly foregrounded poem verses.
In the most extended of empirical novel reading studies that I  know
of, an eye-​movement study by Ralph Radach (1996; see also Radach and
McConkie, 1998), four participants read the first two parts of the book
Gulliver’s Travels (about 160 book pages). The study focused on issues of

A Neurocogni tive Poetics Perspective  [ 317 ]

 

eye-​movement control during reading—​ignoring the main topics of this

chapter—​but deserves mentioning, because it presented strong evidence
for the hypothesis that eye-​movement control during reading is a discrete,
word-​based process, involving the selection of a target word and the some-
what error-​prone process of moving the eyes to that word.
The few available studies on reading longer texts make it difficult to
evaluate to what extent the results discussed earlier obtained with smaller
text units generalize to extended reading sessions lasting over hours. Some
preliminary results obtained by Sebastian Wallot and colleagues (2013; see
also Wallot, 2014)  with a prose text of about 14,000 words suggest that
there are significant differences. However, much more research is needed
before any conclusions can be drawn with regard to this issue.
In sum, although some (poem) mountains have been climbed, the (novel)
summit is not yet reached, as empirical studies providing data about the
reading of an entire novel still seem to be the biggest desideratum in this
field of research.

FROM WORDS TO POEMS AND NOVELS: TOWARDS

INTEGRATIVE MODELS OF MICRO-​ AND MACROPROCESSES
OF LITERARY READING

It is difficult to make an accurate and complete diagram of what happens when one reads.
Burke, 2011, p. 159

Can the results of empirical studies such as those discussed in this chap-
ter, examining all kinds of micro-​, meso-​, and macroscopic aspects of
literary reading, with all kinds of methods, producing a heterogeneous
wealth of data and effects, be integrated under one theoretical roof? The
easy answer is ‘not yet’, the hard one: perhaps ‘not at all’. The literature
on empirical studies of literary reading offers various well-​founded and
empirically supported hypotheses, which basically focus on one or few
selected aspects of the reading act, such as the foregrounding hypothesis
(van Peer, 1986), David Miall and Don Kuiken’s (2001) defamiliarization-​
reconceptualization cycle, or Keith Oatley’s (1994) model of emotional
literary responses. However, general theories or computational process
models like those dominating mainstream experimental reading research
in cognitive psychology—​dealing with non-​literary text materials and
being of only limited validity for the study of literary experiences (Miall
and Kuiken, 1994)—​are still a second major desideratum (for an over-
view, see Jacobs, 2015b).

[ 318 ]  Literature and Cognition in Cognitive Science

 

There are first attempts at comprehensive theoretical and method-

ological frameworks (e.g. Bortolussi and Dixon, 2003), and two recent
ones merit a special mention. Michael Burke’s (2011) model of emotion
in literary reading, which—​at least to me—​really looks more like a gen-
eral theory of literary reading than a specific model, postulates seven stages
of reading, covering a vast space-​time continuum from pre-​reading events
to online processes and post-​reading experiences. In contrast, the NCPM,
distilled from the more general framework developed in our book (Schrott
and Jacobs, 2011), focuses on relatively microscopic aspects, that is, online
processes occurring at a time scale of seconds to (few) minutes of liter-
ary reading (Jacobs, 2011, 2015a). Burke’s model deals with the reading of
novels like The Great Gatsby, while the NCPM describes processes under-
lying the reading of words, phrases, passages, short stories, and poems.
A detailed or even superficial treatment of both models being beyond the
scope of this chapter, I would simply like to propose that both frameworks
are quite complementary in many respects and also could be partially inte-
grated. For example, Burke’s (2011) first and second stages of reading (‘long
before’ and ‘shortly before’ the reading event) fit with the NCPM’s ‘offline’
box entitled ‘reading motivation /​perspective’ (Jacobs, 2014), while stages
three to five (‘during reading’) overlap with the NCPM’s two ‘online routes’,
that is, the upper, fluent route primarily responding to text units relatively
bare of foregrounding elements facilitating immersive processes, and the
lower, dysfluent route getting into operation and facilitating aesthetic
feelings whenever the foregrounding/​backgrounding quotient becomes
greater than 1. It is important to note that the NCPM does not postulate
a categorical, black-​and-​white line separating texts that induce (necessar-
ily and exclusively) either immersion or aesthetic feelings. Certainly, pieces
of text from novels or even newspaper columns can evoke affective and
aesthetic responses, as can single words, adages and aphorisms, and many
other text units—​from the one-​letter word ‘O’ or the Jewish one-​syllable
interjection ‘Oy’ (Schrott and Jacobs, 2011) to entire passages from a novel
(see my section ‘Novel Summits’). On the other hand, it is not unlikely that
readers can also immerse into three-​line poems like Quasimodo’s famous
‘Ed è subito sera’ (‘And suddenly it’s evening’; Schrott and Jacobs, 2011).
Still, if the NCPM is correct, poems full of foregrounding elements should
have a higher likelihood of inducing aesthetic feelings than passages from
novels full of suspenseful action scenes (Jacobs, 2015b).
A central notion of Burke’s (2011) model, ‘disportation’, at first glance
resembles neither the immersion nor the aesthetic feeling concepts of the
NCPM, but may possess elements of both processes. Disportation is a height-
ened emotional state that occurs in affectively-​engaged individuals while

A Neurocogni tive Poetics Perspective  [ 319 ]

 

reading literature, in particular at the moment of closure—​such as when

reading the very last words of The Great Gatsby: ‘So we beat on, boats against
the current, borne back ceaselessly into the past’ (1926/​2000, p. 166). It is
characterized by a distinct feeling that a reader undergoes for a few seconds
wherein a person feels that he/​she is in motion, even though this is not the
case. It is a felt motionless movement through space. In many ways, dispor-
tation is a simulated, embodied affective-​cognitive event that must include
mirror-​neural activation (Burke, 2011, pp. 231‒233). It would thus be inter-
esting to examine the neural underpinnings of ‘disportative’ experiences at
moments of closure during the reading of poems or passages from novels to
see to what extent they correspond with neural activity that has been shown
to correlate with immersive processes in reading parts of novels (Hsu et al.,
2014) or aesthetic processes in reading poetry (Zeman et al., 2013).
In conclusion, even though the truth value of Burke’s line at the begin-
ning of this section is ‘1’, promising frameworks (‘diagrams’) at both the
micro-​and macroscopic levels of literary reading already exist, and they
should be used to generate much more interdisciplinary, ecologically
(more) valid research examining all of its aspects, from the word valleys to
the poem mountains and novel summits.

REFERENCES

Altmann, U., Bohrn, I. C., Lubrich, O., Menninghaus, W., and Jacobs, A. M. (2012).
The power of emotional valence-​from cognitive to affective processes in reading.
Frontiers in Human Neuroscience, 6, 192. doi: 10.3389/​fnhum.2012. 00192
Altmann, U., Bohrn, I. C., Lubrich, O., Menninghaus, W., and Jacobs, A. M. (2014). Fact
vs fiction—​how paratextual information shapes our reading processes. Social
Cognitive and Affective Neuroscience, 9, 22–​29. doi: 10.1093/​scan/​nss098
Aryani, A., Jacobs, A. M., and Conrad, M. (2013). Extracting salient sublexical units
from written texts: ‘Emophon’, a corpus-​based approach to phonological iconic-
ity. Frontiers in Psychology, 4, 654. doi: 10.3389/​fpsyg.2013.00654
Aryani, A., Kraxenberger, M., Ullrich, S., Jacobs, A. M., and Conrad, M. (2015). Measuring
the basic affective tone in poetry using phonological iconicity and subsyllabic
salience. Psychology of Aesthetics, Creativity, and the Arts, 10(2), 191–​204.
Bailey, H., and Zacks, J. M. (2011). Literature and event understanding. Scientific Study
of Literature, 1(1), 72–​78. doi: 10.1075/​ssol.1.1.07bai
Bergson, H. (1911). Laughter: An essay on the meaning of the comic. New York: Macmillan.
doi: 10.1037/​13772-​000
Bohrn, I. C., Altmann, U., and Jacobs, A. M. (2012a). Looking at the brains behind
figurative language—​ A quantitative meta-​ analysis of neuroimaging studies
on metaphor, idiom and irony processing. Neuropsychologia, 50, 2669–​2683.
doi: 10.1016/​j.neuropsychologia.2012.07.021
Bohrn, I. C., Altmann, U., Lubrich, O., Menninghaus, W., and Jacobs, A. M. (2012b).
Old proverbs in new skins—​an FMRI study on defamiliarization. Frontiers in
Psychology, 3, 204. doi: 10.3389/​fpsyg.2012.00204

[ 320 ]  Literature and Cognition in Cognitive Science

 

Bohrn, I. C., Altmann, U., Lubrich, O., Menninghaus, W., and Jacobs, A. M. (2013).
When we like what we know—​A parametric fMRI analysis of beauty and famil-
iarity. Brain and Language, 124, 1–​8. doi: 10.1016/​j.bandl.2012.10.003
Bortolussi, M., and Dixon, P. (2003). Psychonarratology:  Foundations for the empirical
study of literary response. Cambridge: Cambridge University Press.
Brink, T. T., Urton, K., Held, D., Kirilina, E., Hofmann, M., Klann-​Delius, G., Jacobs,
A. M., and Kuchinke, L. (2011). The role of orbitofrontal cortex in process-
ing empathy stories in 4-​to 8-​year-​old children. Frontiers in Psychology, 2, 80.
doi: 10.3389/​fpsyg.2011.00080
Bühler, K. (1934). Sprachtheorie:  Die Darstellungsfunktion der Sprache [Language the-
ory: The representational function of language]. Jena: G. Fischer.
Burke, M. (2011). Literary reading, cognition and emotion: An exploration of the oceanic
mind. London: Routledge.
Burke, M. (2015). The neuroaesthetics of prose fiction: Pitfalls, parameters and pros-
pects. Frontiers in Human Neuroscience, 9, 442. doi: 10.3389/​fnhum.2015.00442
Carminati, M. N., Stabler, J., Roberts, A. M., and Fischer, M. H. (2006). Readers’
responses to sub-​genre and rhyme scheme in poetry. Poetics, 34, 204–​218.
doi:10.1016/​j.poetic.2006.05.001
Citron, F., Cacciari, C., Kucharski, M., Beck, L., Conrad, M., and Jacobs, A. M. (2015).
When emotions are expressed figuratively: Psycholinguistic and affective norms
of 619 idioms for German (PANIG). Behavior Research Methods, epub ahead of
print. doi: 10.3758/​s13428-​015-​0581-​4
Citron, F. M. (2012). Neural correlates of written emotion word processing: A review of
recent electrophysiological and hemodynamic neuroimaging studies. Brain and
Language, 122, 211–​226. doi: 10.1016/​j.bandl.2011
Citron, F. M. M., and Goldberg, A. E. (2014). Metaphorical sentences are more emotion-
ally engaging than their literal counterparts. Journal of Cognitive Neuroscience,
26, 2585–​2595. doi: 10.1162/​jocn_​a_​00654
Conrad, M., Stenneken, P., and Jacobs, A. M. (2006). Associated or dissociated effects
of syllable frequency in lexical decision and naming. Psychonomic Bulletin and
Review, 13, 339–​345.
Copeland, D., Radvansky, G., and Goodwin, K. (2009). A novel study: Forgetting curves
and the reminiscence bump. Memory, 17, 323–​336.
Dambacher, M., Dimigen, O., Braun, M., Wille, K., Jacobs, A. M., and Kliegl, R. (2012).
Stimulus onset asynchrony and the timeline of word recognition:  Event-​
related potentials during sentence reading. Neuropsychologia, 50, 1852–​1870.
doi: 10.1016/​j.neuropsychologia.2012.04.011
Dambacher, M., Kliegl, R., Hofmann, M., and Jacobs, A. M. (2006). Frequency and
predictability effects on event-​related potentials during reading. Brain Research,
1084, 89–​103. doi: 10.1016/​j.brainres.2006.02.010
Dambacher, M., Rolfs, M., Göellner, K., Kliegl, R., and Jacobs, A. M. (2009). Event-​
related potentials reveal rapid verification of predicted visual input. PLoS One,
4, e5047. doi: 10.1371/​journal.pone.0005047
Dimigen, O., Sommer, W., Hohlfeld, A., Jacobs, A. M., and Kliegl, R. (2011). Co-​registra-
tion of eye movements and EEG in natural reading: Analyses and review. Journal
of Experimental Psychology: General, 140, 552–​572. doi: 10.1037/​a0023885
Engel, P., Jacobs, A.M., Lehne, M., Menninghaus, W., and Koelsch, S. (2016).
Manipulating Jakobson’s poetic function: Affective and aesthetic effects of syn-
tactic complexity in a text by Kleist. Manuscript in preparation.
Fitzgerald, F. S. (2000). The great Gatsby. London: Penguin.

A Neurocogni tive Poetics Perspective  [ 321 ]

 

Forgács, B., Bohrn, I. C., Baudewig, J., Hofmann, M. J., Pleh, C., and Jacobs, A. M.
(2012). Neural correlates of combinatorial semantic processing of literal and fig-
urative noun-​noun compound words. Neuroimage, 63, 1432–​1442. doi: 10.1016/​
j.neuroimage.2012.07.029
Freud, S. (1953‒1974). The uncanny. In J. Strachey (Ed. and trans.), The standard edi-
tion of the complete psychological works of Sigmund Freud (Vol. 17, pp. 217–​252).
London: Hogarth Press.
Gebauer, G., and Edler, M. (2014). Sprachen der Emotion:  Kultur, Kunst, Gesellschaft
[Languages of emotion: Culture, art, society]. Frankfurt: Campus.
Grainger, J., and Jacobs, A. M. (1996). Orthographic processing in visual word rec-
ognition:  A  multiple read-​ out model. Psychological Review, 103, 518–​565.
doi: 10.1037//​0033-​295X.103.3.518
Hofmann, E. T. A. (1844). The Sandman. In J. Oxenford and C. A. Feiling (Trans.), Tales
from the German, comprising specimens from the most celebrated authors, pp. 140–​
164. London: Chapman and Hall.
Hofmann, M. J., and Jacobs, A. M. (2014). Interactive activation and competition
models and semantic context:  From behavioral to brain data. Neuroscience &
Biobehavioral Reviews, 46, 85–​104. doi: 10.1016/​j.neubiorev.2014.06.011
Hogan, P. C. (2003). Cognitive science, literature and the arts. New York: Routledge.
Holenstein, E. (1976). Roman Jakobson’s approach to language: Phenomenological struc-
turalism. Bloomington, IN: Indiana University Press.
Hsu, C.-​T., Conrad, M., and Jacobs, A. M. (2014). Fiction feelings in Harry
Potter:  Haemodynamic response in the mid-​cingulate cortex correlates with
immersive reading experience. Neuroreport, 25, 1356–​ 1361. doi:  10.1097/​
WNR.0000000000000272
Jacobs, A. M. (2001). The cognitive psychology of literacy. In N. J. Smelser and P. B.
Baltes (Eds.), International encyclopedia of the social and behavioral sciences (pp.
8971–​8975). Amsterdam: Elsevier.
Jacobs, A. M. (2011). Neurokognitive Poetik: Elemente eines Modells des literarischen
Lesens [Neurocognitive poetics: Elements of a model of literary reading]. In R.
Schrott and A. M. Jacobs (Eds.), Gehirn und Gedicht: Wie wir unsere Wirklichkeiten
konstruieren [Brain and poetry: How we construct our realities] (pp. 492–​520).
Munich: Carl Hanser.
Jacobs, A. M. (2014). Affektive und ästhetische Prozesse beim Lesen: Anfänge einer
neurokognitiven Poetik [Affective and aesthetic processes in reading: Towards
a neurocognitive poetics]. In G. Gebauer and M. Edler (Eds.), Sprachen der
Emotion: Kultur, Kunst, Gesellschaft [Languages of emotion: Culture, art, society]
(pp. 134–​154). Frankfurt: Campus.
Jacobs, A. M. (2015a). Towards a neurocognitive poetics model of literary reading.
In R. Willems (Ed.), Towards a cognitive neuroscience of natural language use (pp.
135–​159). Cambridge: Cambridge University Press.
Jacobs, A. M. (2015b). Neurocognitive poetics: Methods and models for investigating
the neuronal and cognitive-​affective bases of literature reception. Frontiers in
Human Neuroscience, 9, 186. doi: 10.3389/​fnhum.2015.00186
Jacobs, A. M., and Kinder, A. (2015). Worte als Worte erfahren:  Wie erarbeitet das
Gehirn Gedichte [Experiencing words as words:  How the brain constructs
poems]. In A. Pompe (Ed.), Kind und Gedicht [Child and poem] (pp. 57–​76).
Berlin: Rombach.
Jacobs, A. M., and Lüdtke, J. (2016). Immersion into narrative and poetic worlds: A
neurocognitive poetics perspective.

[ 322 ]  Literature and Cognition in Cognitive Science

 

Jacobs, A. M., Lüdtke, J., Aryani, A., Meyer-​Sickendiek, B., and Conrad, M. (2016).
Mood-​empathic and aesthetic responses in poetry reception: A model-​guided,
multilevel, multimethod approach. Scientific Study of Literature.
Jacobs, A. M., Võ, M. L.-​H., Briesemeister, B. B., Conrad, M., Hofmann, M. J., Kuchinke,
L., Lüdtke, J. and Braun, M. (2015). 10 years of BAWLing into affective and aes-
thetic processes in reading: What are the echoes? Frontiers in Psychology, 6, 714.
doi: 10.3389/​fpsyg.2015.00714
Jacobs, A. M., and Ziegler, J. C. (2015). Visual word recognition, neurocognitive psy-
chology of. In James D. Wright (Ed.), The international encyclopedia of social and
behavioral sciences (2nd ed., pp. 214–​219). Oxford: Elsevier.
Jakobson, R. (1960). Closing statement: Linguistics and poetics. In T. A. Sebeok (Ed.),
Style in language (pp. 350–​377). Cambridge, MA: MIT Press.
Jakobson, R. (1979). Hölderlin, Klee, Brecht: Zur Wortkunst dreier Gedichte [Hölderlin,
Klee, Brecht: On the verbal art of three poems]. Frankfurt: Suhrkamp.
Keidel, J. L., Davis, P. M., Gonzalez-​Diaz, V., Martin, C. D., and Thierry, G. (2013). How
Shakespeare tempests the brain: Neuroimaging insights. Cortex, 49(4), 913e919.
http://​ dx.doi.org/​10.1016/​j.cortex.2012.03.011
Koops van’t Jagt, R., Hoeks, J. C., Dorleijn, G. J., and Hendriks, P. (2014). Look before
you leap: How enjambment affects the processing of poetry. Scientific Study of
Literature, 4(1), 3–24.
Kuchinke, L., Jacobs, A. M., Grubich, C., Võ, M. L.-​H., Conrad, M., and Herrmann,
M. (2005). Incidental effects of emotional valence in single word processing: An
fMRI study. Neuroimage, 28, 1022–​1032. doi: 10.1016/​j.neuroimage.2005.06.050
Kuchinke, L., Trapp, S., Jacobs, A. M., and Leder, H. (2009). Pupillary responses in art
appreciation: Effects of aesthetic emotions. Psychology of Aesthetics, Creativity,
and the Arts, 3, 156–​163. doi: 10.1037/​a0014464
Kuhlmann, M., Hofmann, M. J., Briesemeister, B. B., and Jacobs, A. M. (2016). Mixing
positive and negative valence: Affective-​semantic integration of bivalent words.
Scientific Reports, in press.
Leder, H. (2013). Next steps in neuroaesthetics: Which processes and processing stages
to study? Psychology of Aesthetics, Creativity, and the Arts, 7, 27–​37.
Leder, H., Belke, B., Oeberst, A., and Augustin, D. (2004). A model of aesthetic appre-
ciation and aesthetic judgments. British Journal of Psychology, 95, 489–​508.
doi: 10. 1348/​0007126042369811
Leder, H., Gerger, G., Dressler, S., and Schabmann, A. (2012). How art is appreciated.
Psychology of Aesthetics, Creativity, and the Arts, 6, 2–​10. doi:10.1037/​a0026396
Leder, H., Markey, P. S., and Pelowski, M. (2015). Aesthetic emotions to art—​What
they are and what makes them special. Comment on ‘The quartet theory of
human emotions:  An integrative and neurofunctional model’, by S.  Koelsch
et al. Physics of Life Reviews, 13, 67–​70.
Lehne, M., Engel, P., Rohrmeier, M., Menninghaus, W., Jacobs, A. M., and Koelsch,
S. (2015). Reading a suspenseful literary text activates brain areas related to
social cognition and predictive inference. PLoS One, 10, e0124550. doi: 10.1371/​
journal.pone.0124550
Limbach, J. (2004). Das schönste deutsche Wort [The most beautiful German word].
Freiburg: Herder.
Lubrich, O., Knoop, C., and Jacobs, A. M. (2014). Jean Genet und die Ästhetisierung
des abweichenden: Ein interdisziplinäres Experiment [Jean Genet and the aes-
theticisation of deviance: An interdisciplinary experiment]. In M. N. Lorenz and
O. Lubrich (Eds.), Jean Genet und Deutschland [Jean Genet and Germany] (pp.
393–​406). Berlin: Merlin.

A Neurocogni tive Poetics Perspective  [ 323 ]

 

Lüdtke, J. (2013). Eine Frage der Empirie: Zum emotionalen Erleben bei der Rezeption
von Stimmungsgedichten [An empirical question: On emotional experience in
poetry reception]. In B. Meyer-​Sickendiek and F. Reents (Eds.), Stimmung und
Methode [Mood and method] (pp. 119–​138). Tübingen: Mohr Siebeck.
Lüdtke, J., and Jacobs, A. M. (2015). The emotion potential of simple sentences: Additive
or interactive effects of nouns and adjectives? Frontiers in Psychology, 6, 1137.
doi: 10.3389/​fpsyg.2015.01137
Lüdtke, J., Meyer-​Sickendiek, B., and Jacobs, A. M. (2014). Immersing in the stillness
of an early morning: Testing the mood empathy hypothesis in poems. Psychology
of Aesthetics, Creativity, and the Arts, 8, 363–​377. doi: 10.1037/​a0036826
Menninghaus, W. (1991). Dichtung als Tanz—​Zu Klopstocks Poetik der Wortbewegung
[Poetry as dance—​On Klopstock’s poetics of word movement]. COMPARATIO.
Revue internationale de littérature comparée, 2‒3, 129–​150.
Menninghaus, W., Bohrn, I. C., Altmann, U., Lubrich, O., and Jacobs, A. M. (2014).
Sounds funny? Humor effects of phonological and prosodic figures of speech.
Psychology of Aesthetics, Creativity, and the Arts, 8, 71–​76. doi: 10.1037/​a0035309
Menninghaus, W., Bohrn, I. C., Knoop, C., Kotz, S. A., Schlotz, W., and Jacobs, A. M.
(2015). Rhetorical features facilitate prosodic processing while handicapping
ease of semantic comprehension. Cognition, 143, 48–​60.
Miall, D. S., and Kuiken, D. (1994). Beyond text theory:  Understanding literary
response. Discourse Processes, 17, 337–​352.
Miall, D. S., and Kuiken, D. (2001). Shifting perspectives: Readers’ feelings and literary
response. In W. van Peer and S. Chatman (Eds.), New Perspectives on Narrative
Perspective (pp. 289–​301). Albany, NY: State University of New York Press.
Mieder, W. (2008). Proverbs speak louder than words: Wisdom in art, culture, folklore, his-
tory, literature and mass media. New York: Peter Lang.
Nell, V. (1988). The psychology of reading for pleasure:  Needs and gratifications.
Reading Research Quarterly, 23, 1, 6–​50.
Oatley, K. (1994). A taxonomy of the emotions of literary response and a theory of
identification in fictional narrative. Poetics, 23, 53–​74. doi:  10.1016/​0304-​
422X(94)P4296-​S
O’Sullivan, N., Davis, P., Billington, J., Gonzalez-​Diaz, V., and Corcoran, R. (2015).
‘Shall I compare thee’: The neural basis of literary awareness, and its benefits to
cognition, Cortex, 73, 144–​157. doi: 10.1016/​j.cortex.2015.08.014
Panksepp, J. (1998). Affective neuroscience: The foundations of human and animal emo-
tions. New York: Oxford University Press.
Panksepp, J. (2008). The power of the word may reside in the power of affect. Integrative
Psychological and Behavioral Science, 42, 47–​55. doi: 10.1007/​s12124-​007-​9036-​5
Ponz, A., Montant, M., Liegeois-​Chauvel, C., Silva, C., Braun, M., Jacobs, A. M., and
Ziegler, J. C. (2013). Emotion processing in words: A test of the neural re-​use
hypothesis using surface and intracranial EEG. Social Cognitive and Affective
Neuroscience, 9, 619–​627. doi: 10.1093/​scan/​nst034
Radach, R. (1996). Blickbewegungen beim Lesen: Psychologische Aspekte der Determination
von Fixationspositionen [Eye movements in reading: Psychological aspects of the
determination of fixation positions]. Münster: Waxmann.
Radach, R., and McConkie, G. W. (1998). Determinants of fixation positions in words
during reading. In G. Underwood (Ed.), Eye guidance in reading and scene percep-
tion (pp. 77–​100). Oxford, Elsevier.

[ 324 ]  Literature and Cognition in Cognitive Science

 

Reber, R., Schwartz, N., and Winkielman, P. (2004). Processing fluency and aesthetic
pleasure:  Is beauty in the perceiver’s processing experience? Personality and
Social Psychology Review, 8, 364–​382. doi: 10.1207/​s15327957pspr0804_​3
Rowling, J. K. (1999). Harry Potter and the philosopher’s stone. New York: Scholastic.
Rowling, J. K. (2014). Harry Potter and the deathly hallows. London: Bloomsbury.
Rutter, B., Kröger, S., Stark, R., Schweckendiek, J., Windmann, S., Hermann, C., and
Abraham, A. (2012). Can clouds dance? Neural correlates 110 of passive concep-
tual expansion using a metaphor processing 111 task: Implications for creative
cognition. Brain and Cognition, 78, 114–​122. doi: 10.1016/​j.bandc.2011.11.002
Scheepers, C., Mohr, S., Fischer, M. H., and Roberts, A. M. (2013). Listening to lim-
ericks:  A  pupillometry investigation of perceivers’ expectancy. PLoS One, 8,
e74986. doi:10.1371/​journal.pone.0074986
Schrott, R., and Jacobs, A. M. (2011). Gehirn und Gedicht: Wie wir unsere Wirklichkeiten
konstruieren [Brain and poetry: How we construct our realities]. Munich: Hanser.
Scott, G. G., O’Donnell, P. J., and Sereno, S. C. (2012). Emotion words affect eye fixa-
tions during reading. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 38(3), 783–​792. doi: 10.1037/​a0027209
Shakespeare, W. (2002). Complete sonnets and poems (C. Burrow, Ed.). Oxford: Oxford
University Press.
Sheikh, N. A., and Titone, D. A. (2013). Sensorimotor and linguistic information atten-
uate emotional word processing benefits: An eye-​movement study. Emotion, 13,
1107–​1121. doi: 10.1037/​a0032417
Sternberg, M. (2003). Universals of narrative and their cognitivist fortunes (I). Poetics
Today, 24, 297–​395.
Valéry, P. (1960). Oeuvres II. Paris: Pléiade.
van Dijk, T. A., and Kintsch, W. (1983). Strategies of discourse comprehension.
New York: Academic.
van Peer, W. (1986). Stylistics and psychology:  Investigations of foregrounding.
London: Croom Helm.
Võ, M. L.-​H., Conrad, M., Kuchinke, L., Hartfeld, K., Hofmann, M. J., and Jacobs, A.
M. (2009). The Berlin Affective Word List reloaded (BAWL-​R). Behavior Research
Methods, 41, 534–​539. doi: 10.3758/​BRM.41.2.534
Võ, M. L.-​H., Jacobs, A. M., and Conrad, M. (2006). Cross-​validating the Berlin affective
word list. Behavior Research Methods, 38, 606–​609. doi: 10.3758/​B. F. 03193892
Wallentin, M., Nielsen, A. H., Vuust, P., Dohn, A., Roepstorff, A., and Lund, T. E.
(2011). Amygdala and heart rate variability responses from listening to emo-
tionally intense parts of a story. Neuroimage, 58, 963–​ 973. doi:  10.1016/​
j.neuroimage.2011.06.077
Wallot, S. (2014). From ‘cracking the orthographic code’ to ‘playing with lan-
guage’:  Toward a usage-​based foundation of the reading process. Frontiers in
Psychology, 5, 891. doi: 10.3389/​fpsyg.2014.00891
Wallot, S., Hollis, G., and van Rooij, M. (2013). Connected text reading and differences
in text reading fluency in adult readers. PLoS One, 8, e71914. doi: 10. 1371/​jour-
nal.pone.0071914
Zeman, A., Milton, F., Smith, A., and Rylance, R. (2013). By heart:  An fMRI study
of brain activation by poetry and prose. Journal of Consciousness Studies, 20,
132–​158.

A Neurocogni tive Poetics Perspective  [ 325 ]

 
 

INDEX

Emboldened page ranges refer to chapters; page numbers in italics refer to figures or
tables.
2001: A Space Odyssey (film)  56, 77 Aldama, Frederick L.  20
4E paradigm of mind (embodied, Alexander, M. 74
embedded, enactive, Algom, D. 81
extended)  139, 143, 169 Alice in Wonderland (Carroll)  283
aliens (in science fiction)  79, 80
absorption. See immersion Allbritton, D.W.  245
abstraction  5, 74–​76, 86, 98, 266, 268 allegorical literature  227–​234
abstract events  225 alliteration 305
accommodation 283 allusions, intertextual  147
action-​compatibility effect  224 ‘All You Zombies’ (Heinlein)  78
Action in Perception (Noë)  140 Alony, R. 82
actor-​observer difference  264 Alps 55, 57, 61
Addison, Joseph  290 alternate realities  136
Adler, Hans  20, 22, 29 alternative history genre  80
aesthetic experience  32, 290, 299 Altmann, Ulrike  137–​138, 308, 313
aesthetic illusion  9, 139 Amend, A. 248
aesthetic reading processes  303–​325 amygdala  309, 313
aesthetic response  120n6 analogies 98
aesthetics, neuro-​  308 Anderson, M. 155
aesthetics of text  48 Anderson, Poul  77
aesthetic theory  55 Anderson, T. 169
aesthetic trends  12 Andric, M. 226
Affective Narratology (Hogan)  35 animals  12, 22, 23, 44, 94, 95, 118, 180,
affective reading processes  303–​325 195–216
affective responses  38. See also anomalous sentences  295–​298, 299, 299
emotions, literary anorexia nervosa  170, 172–​173, 175,
affiliation, need for  249 185, 191
After the Fall (Miller)  113–​133 Anstey, F. 78
agency, human  5, 38, 80, 82, 85 anterior insula  312
and narrative  100–​101 anterior midline field (AMF)  294–​295,
Age of Insight: The Quest to Understand the 298, 299
Unconscious in Art, Mind and Brain, ‘Anthem for Doomed Youth’ (Owen)  292
from Vienna 1900 to the Present Anthologist, The (Baker)  219, 232
(Kandel) 152 anthropology  6, 9, 13, 136n1
Alabastro, A. 247 anti-​proverbs  308
 

Appel, M. 252 BAWL (Berlin Affective Word List)  306

Archer, D. 269 Baxandall, Michael  155
Aristopia (Holford)  80 Baxter, Stephen  77, 79
Aristotle 142, 263 Bayesian cognition  151, 155, 165, 188
Armstrong, Paul B.  156 Bayesian inferences  158
art, psychology of  154 Bear, Greg  79
art, visual  8, 152–155, 157–158, 222, Beat (eating disorders charity)  177
289–290 Beaty, R.E.  69
Art and Illusion: A Study in the Psychology beauty of words, cues for  306
of Pictorial Representation Beck, L. 308
(Gombrich)  153–​154, 164 behaviourist paradigms  261
art distinguished from craft  273 ‘beholder’s share’  153–​154
artificiality in fiction  147 Belke, B. 308
artists  153–​154 bell curve  45n2
Aryani, A. 316 Benford, Gregory  77
Asimov, Isaac  81 Bergen, B. 223
assonance 305 Bergh, C. 191
Astington, J.W.  264 Bergs, Alexander  5–​9, 279–​301
Äström, C.J.  171 Bergson, Henri  310
Attebery, Brian  82 Berkeley Construction Grammar  288
attention 104 Berlin Affective Word List (BAWL)  306
attitude change  252 Bernini, M. 170
attribution, mental-​state  12, 213 Bezdek, M.A.  245, 253
Attridge, D. 57 biases  117, 185–86, 176, 177
audience, sense of  118 Bible 59
Augustin, D. 308 ‘Big Dumb Objects’  77
Author Recognition test  269 Big Five personality measures  271
authors’ misunderstanding of own Big Six topics  5
works 129n11 Bilandzic, H. 251
automatic memory processes  242–​244 bilateral anterior insula  313
autonomy, principle of  4, 6, 18, 25–​29, 32 bilateral inferior frontal gyrus  313
awareness  165–​166 Binder, J. 226
Ayduk, O. 76 binocular rivalry  157, 159n5
‘Birds, The’ (Du Maurier)  205
Baezconde-​Garbanati, L.  252 bisociative thinking  307
Bailey, H. 305 Black, Jessica  271
Bailis, Daniel  262 Blackfish (film)  214
Baker, J.A.  205–​206 Blackwell, N. 230
Baker, L. 175 Blaszczynski, A.  175–​176
Baker, Nicolson  219 blending  21, 283, 288. See also
Balakian, J.N.  115n3 metonymy
Ball, Hugo  293 Bless, H. 80
Banaji, M.R.  247 blindness, inattentional  153
Bar-​Anan, Y.  81 Blindsight (Watts)  80
Barnes, Jennifer  271 Blood, A.J.  70
Baron-​Cohen, S.  269 Blood Music (Egan)  78
Bartlett, Frederic  299 Boas, H.C.  289
basal ganglia  308, 316 Boden, Margaret  47
Bateson, Gregory  22 body dissatisfaction  175–​176
Baudelaire, Charles  70 body dysmorphia  185

[ 328 ] Index
 

body image  175, 179–​180 caring, paradox of  25n8

Boers, E. 227 Carland, Matthew  272
Bohn, W. 78 Carlson, L.  121n7
Bohrn, Isabel  307–​309 Carminati, M.N.  311
Bolens, G. 84 Carney, James  5, 7, 10–​12, 188
Bor, Daniel  93–​95, 98, 104 on CLT and science fiction  73–​92
Borden, I.A.  270 Carroll, Joseph  24, 48
Bornkessel-​Schlesewsky, I.  297 Carroll, N.  98n1
Bortolussi, M.  265, 319 Carroll, Noël  105
Bould, M. 85 Carter, R. 71
Boulenger, V. 225 Caruso, D.R.  46
Bourrit, M.T.  55 Castano, Emanuele  137, 271
Boyd, Brian  5, 7–​8, 12, 267 catastrophe theory  174, 188
on cognitive patterns  93–​109 categorization  84, 121, 204
on ‘natural’ narratology  48 Cattell, James M.  303
Bozena, J.-​D.  76 caudate nucleus  308–​309, 313
Brain and Poetry (Schrott & Jacobs)  304 Cavanagh, P.  153–​156
brain imaging  27n10 Cazotte, Jacques  11, 156–​166
brains  24–​25, 152, 154 Cenis, Mont  58–​59
Braun, M. 309 Central Park  241–​242, 244
Brennan, J. 294 Chabon, Michael  80
Brink, T.T.  312 characterisation  8, 74, 82, 131, 137
Brisch, R. 191 characters, fictional  38, 117, 187
Broca’s area (BA)  294, 297–​298, 299 engagement with  181
Brock, Timothy  137, 247, 251–​252, 274 Chatterjee, J.S.  252
Brodman’s area. See Broca’s area (BA) Chekhov, Anton  272
Brown, C.M.  295 Chesters, T. 170
Brown, P. 84 Chiang, Ted 78
Bruehlman-​Senecal, E.  76 Chiesi, H.L.  241
Bruner, J.  165, 261 Children’s Friendships: The Beginnings of
Bühler, Karl  303, 306 Intimacy (Dunn)  267
bulimia nervosa  174 Choi, A. 294
Burke, Edmund  57, 77 Chung, J. 264
Burke, Michael  3, 17, 35–36, 84, 222, Churchland, P.S.  171
304, 310, 318–​320 Citron, F.M.  306, 308
on cognitive and affective narratology  36 City and the Stars, The (Clarke)  79
Burroughs, John  212 City at the End of Time, The (Bear)  79
Busch, Wilhelm  309 Clark, Andy  6, 152, 188, 191
Busselle, R. 251 Clarke, Arthur C.  77, 79
Buswell, B.N.  84 Clore, Gerald  253
Butler, Octavia E.  79 closure  222, 280, 320
Buzsaki, Gyorgy  148 Codispoti, M. 67
Byatt, A.S.  143 coercion  7, 10. See also type-​shifting
‘By His Bootstraps’ (Heinlein)  78 complement 290
formal approaches to  286–​287
Cacciari, C. 308 in linguistics  281–​286
Cantwell Smith, B.  47 in literature  289–​293
Caracciolo, M.  47, 140, 169–​170, 265 as metaphor and metonymy  287–​289
Caramazza, A. 225 in neuroscience  294–​299
Card, Orson Scott  79 pragmatic approaches to  287

Index  [ 329 ]
 

cognition  35, 139, 169 altered state of  60–​61, 71

4E (embodied, embedded, enactive, another’s 49
extended)  13, 139, 169 embodied 46
abstract modes of  85 experiencing, collective  49n5
action-​related  138 landscape of 204
Bayesian 151 Consciousness Explained (Dennett)  198
broad 36, 40, 46 Constanzo, M. 269
contentless 141n3 construal level theory (CLT)  5, 8
and context  75 and science fiction  73–​92
embodied 288 constructionism  41–​42
enactive 135 construction process in reading 
fantastic 10, 151–​167 240–​241
interpersonal 138, 163 context  7, 13, 39, 75, 153, 311
models of 86 contextualization, narrative  162
narrow  36, 38–​40, 44, 46 contextual pressure  287
and patterns  94 continuity, law of  280
situated 47, 50 conversation 114, 269
social. see social cognition Cook, A.E.  240n1
cognitive behavioural therapy (CBT)  174 Copeland, D. 305
cognitive literary science  4, 13, 136, 192 Coriolanus (Shakespeare)  66
appeal of 30 Corvus (Woolfson)  202
heterogeneous foundations  19–​20 counterfactual thinking  76, 80
and mind-​body problem  27–​28 Cowan, D.E.  82
and ‘reality’ of reading and writing  21 Cowperthwaite, Gabriela  214
cognitive neuropsychology  140 Cox, J. 267
cognitive neuroscience  1, 293 Craighero, L. 225
cognitive science  47, 115 Crane, M.T.  155
and fiction  135–​150 Crano, W.D.  247
and literature  17–​34 Crist, E.  200–​201
and science fiction  73–​92 critical period  120, 125–​129
Cohen, J.  252–​253 criticism, literary. See literary criticism
Cohen, P. 18 Cromwell, Howard Casey  148
Cohen, R.  175–​176 Csicsery-​Ronay, I., Jr.  146
coherence  18, 25–​29 cues for beauty of words  306
Cohn, N. 294 culture 29, 266
Coleridge, S.T.  142 curiosity 314n1
collaboration 4, 26 Currie, Gregory  25, 86, 260–​261
Collingwood, R.G.  273 Cushman, G.C.  270
Collins, Allan  253
Colston, H.  223, 225, 227 Dainton, B. 60
comedy, romantic  95–​96 Damasio, Antonio R.  36–​37, 39, 41
communication, indirect  272, 274 on cognitive patterns  95
‘competing timeline’ theme  80 on literary emotions  45, 120
composition, enriched  283, 286 on ‘natural’ narratology  50
comprehension 95, 242 Dambacher, M. 309
computational model of mind  136, 139 Danta, C. 136
Conant, L. 226 d’Aquili, E.G.  69
conceptual expansion  307 Darwin, Charles  75, 200
Conrad, M.  304, 306, 308, 312, 316 Davies, J. 169
consciousness  1, 27, 50, 171–​172, 200n2 Davis, J.I.  84

[ 330 ] Index

Davis, Mark H.  138
Davis, Philip  66–​67, 309
deactivation  69–​71
de Almeida, R.G.  281, 287
Debek, M. 76
Decety, J.  121n7
decision-​making processes  36–​37
deconstruction 19
defamiliarization  65–​66, 308
defamiliarization-​reconceptualization
cycle 318
dehabituation 67
deictic centre  66
dela Paz, J.  269
Denis, M. 135
Dennett, Dan  172, 196–​198, 200n2
density, heterophenomenological  211
Desai, R. 226
Descola, P. 213
‘Des Morgens’ (Hölderlin)  316
De Soussa, R.  37
Desperate Housewives 252
Dhar, R. 76
Diable Amoureux, Le (Cazotte)  156–​166
diagnosticity principle  248
dialogue, cognitive-​literary  2
Diamond Age, The (Stephenson)  78
Dias, M. 266
Dick, Philip K.  80
Dietrich, A.  24n7, 28
Different Flesh, A (Turtledove)  81
differing emotions problem  40, 49
Dimigen, O. 309
Di Paolo, E.A.  139
disambiguation devices  162
discourse domains  195–​216
disengagement, readers’  249
disportation 319–320
disquietude, sublime  187
dissociation 187
distance, psychological  5, 74–​75
and construal level theory  81,
83–​84, 86
perceptions of 76
probabilistic 80, 82
in science fiction  77–​80, 85
distraction 181
Divine Invasion, The (Dick)  80
Dixon, P.  265, 319
Djikic, Maja  236, 271–​272, 274
dogs, companion  199–​200
Doherty, M.  114, 118
Dohn, A. 316
Doicaru, M.M.  251
dopamine system  191
Dorleijn, G. 311
Dorsch, T.S.  61–​62
Dovidio, J.F.  76, 83
Dressler, S. 308
Dr. Jekyll and Mr. Hyde (Stevenson)  147
dualism, mind-​body  47, 200
dual vision of fiction  135–​150, 141
Du Maurier, Daphne  205–​206
Dunbar, Robin  261, 264, 269
Dunn, Judy  267
Dutton, D. 48
eating disorders  9, 169–​194
Eco, Umberto  80
Écrits (Lacan)  127n10
Edelman, Gerald  93–​94
Edler, M. 304
‘effort after meaning’ principle  299
Egan, Greg  78, 81
Ekman, P.  36, 39, 41
electrodermal activity  317
electroencephalography 295
elevation (emotion)  62–​63
elision 287
Elizabethan sonnets  102
Elster, J.  37, 41
embodied dynamics  219–​237
embodiment  5, 36, 46, 47, 63–​65, 131,
149–140, 169–​170, 199, 306, 310,
320
and enactive perception  139
Emmott, C. 74
Emophon algorithm  316
emotionality  83–​84
emotional knowledge  45
emotional resonance  99, 101, 105
emotional responses  42, 190
emotion contagion  120
emotionology 41
emotion regulation  39–​40
emotions 50, 274
appraisal theory of  191
basic 41, 49
evoked by fiction  260
literary  5–​6, 35–​42, 49, 318
Index  [ 331 ]

emotions (Cont.)
in literary reading  221, 319
moral  62–​63
and ‘natural’ narratology  47–​51
and neurocognitive poetics  313
in possible worlds  268
quasi-​  40
secondary 62
and self-​transformation  272–​274
shared 49
shifts of 107
and simulation  113, 263, 265
emotion theories  42, 303
empathy  5, 37–​38, 42, 46, 83, 120,
138–​139
components of 249
and construal level theory  76
critical period  125–​129
evoked by art  152
for fictional characters  240, 246–​247,
250, 253–​254
and neurocognitive poetics  312–​313
and simulation  264–​265
and transportation  270
and understanding others  269–​271
enactive paradigm  136, 139–​143
Ender’s Game (Card)  79
End of Eternity, The (Asimov)  81
engagement, readerly  7, 143, 154, 170,
187–​188, 243
and embodied dynamics  221
emotional 99
and enactive perception  138–​139, 142
with narrative  251
with reality  147
and simulation  262
theories of 249
Engel, P. 315
enjoyment, literary  222, 252
Enzensberger, Hans Magnus  316
equilibrium in feedback
systems  171–​172, 174
erotectic force  105
Eskine, K.J.  69
Esrock, E.J.  140
ethology  6, 201, 208
evaluation, aesthetic  308
Evans, Dylan  37, 41
Evans, J. 75
‘Eveline’ (Joyce)  36, 42–​46
[ 332 ] Index
event-​related potentials (ERP)  67,
294–​296, 298, 299
evolutionary theory  37
exclamations, palilogical  44
expansion, conceptual  307
expectations, readers’  124, 147, 153,
188, 190, 280
arising from life experiences  244
experience 
fiction-​derived  49
lived  41, 48–​50, 118
personal 11
experiments, laboratory  9, 117
experiments, thought  116
expertise, interdisciplinary  25–​26
expertise, readers’  241–​242, 244, 270
explanation, levels of  22–​25, 28, 30, 32
explanatory gaps  28, 31
expressions, emotional  120
Eyal, T.  82–​83
eye movements  310, 311, 318
Eysenck, M. 126
Fabb, N. 98
Faber, Jean-​Henri  200
Facebook  175–​176
Fairburn, C.G.  173–​174
familiarity 252, 308
families, psychosomatic  175
family stories  270
family therapy  175
fantasies  152, 154, 165
Fantastic, The (Todorov)  159
fantastic genre  5, 7, 86
and cognition  152, 159–​160, 164–​165
and double vision of fiction  143, 147
Fantastic Voyage (Fleischer)  78
Fantasy Scale  138
fatalism 85
Fathallah, Judith  179
Fauconnier, Gilles  21, 288
Fayn, K. 69
fear 315
feedback  5, 9, 38, 139–140, 151–152,
158, 160, 169–194
Feeling Beauty (Starr)  155
Feeling of What Happens: Body and
Emotion in the Making of
Consciousness (Damasio)  50
Fekete, J. 85
 

Fiasco (Lem)  79 Friston, K.  151–​152, 154, 160

fiction  7, 76, 135–​150, 165, 188–​189 Frith, Chris  151–​152, 154–​155, 164
distrust of  259–​261 frontal cortex  69, 71
eating-​disorders genre  186 Frost, Robert  227, 229
formulaic 317 Fujita, K. 82
historical 203 functional magnetic resonance imaging
horror 206 (fMRI)  294, 296–​298, 299
paradox of  40, 260 data  9, 64, 138, 316
patterns of 96 functional shift  6, 66–​67, 308
psychology of 274
reading  176–​186 Gadamer, H.-​G.  158n3
and truth  259–​278 Gadet, F. 283
fictionality  39, 135–​141 Gadji Beri Bimba (Ball)  293
and The Prestige  143–​148 Galactic Center (Benford)  77
fiction feeling hypothesis  312, 315 Gallagher, S.  139, 199
fiction–​nonfiction divide  196, 202, Gallese, V.  136n1
205, 207 Gapenne, O. 139
Filik, R. 244 Gaussian distribution  45n2
Fincher-​Kiefer, R.  241 Gavins, J. 261
Finnegan’s Wake (Joyce)  293 Gebauer, G. 304
Fischer, M.H.  311 Generative Lexicon, The (Pustejovsky)  286
Fisher, C.A.  175 Genet, Jean  315
Fitzgerald, F. Scott  222 Genette, G. 208
Fleischer, Richard  78 genres  7, 95, 209
flexibility ratio  317 Georgiadis, J.R.  70
Fludernik, Monika  30, 35–​36, 46–​47 Gerger, G. 308
on coercion  288 Gerrig, Richard  6–​8, 11, 48, 191
on construal level theory (CLT)  84 on readers’ lives and narrative
on ‘natural’ narratology  48–​51 experiences  239–​257
on readers’ lives and narrative on simulation in fiction  262
experiences 251 Gestalt psychology  279, 280, 289
Fong, Katrina  270 Gibbs, Raymond  7, 9, 11, 219–​237, 288
foregrounding  65–​67, 77, 308, 318–​319 Gibson, J.J.  153
Forever War, The (Haldeman)  78 Gilligan, J. 127
Forgács, Bálint  306–​307 Gilmore, P. 86
Förster, J. 81 Glenberg, A. 224
Forsyth, A.  115n3 Göellner, K.J.  309
‘found science’  154, 156, 162–​163, Goldberg, A.E.  281, 283, 308
165–​166 Golding, William  79
Fowler, R.  202n3 Goldman, Alvin  265
Foy, J.E.  244 Golem and the Jinni, The (Wecker)  241,
frames, cognitive  8, 13, 48 244–​245
Francis, E.J.  283 Gombrich, Ernst  153–​156, 158, 164
Frank., G.K.W.  191 Gomel, Elana  82
Frank, L.B.  252 Gonzalez, A.L.  175
Frankenstein (Shelley)  147 Gonzalez-​Diaz, V.  66, 309
free indirect discourse  43, 204 Gonzálvez-​Garcia, F.  288–​289, 291
Freeland, C. 86 Goodwin, K. 305
Freud, Sigmund  267, 314 Goren, A. 76
Frijda, Nico  37, 62, 191 Gothic style  156n1

Index  [ 333 ]
 

Gottschall, J.  18n2, 20, 28, 29n13 Hemingway, Ernest  208

Gould, J. 226 Hendriks, P. 311
Grabe, S. 175 Herbert, C. 67
Grainger, J. 306 Herman, David  6–​10, 12, 251
grammar 114, 284 on animal minds  195–​216
Gravity (2013 film)  56 on construal level theory (CLT)  84
Great Gatsby, The (Fitzgerald)  222, 320 on literary emotions  41, 46
Green, Melanie  137, 251–​252, 262, 274 on ‘natural’ narratology  47–​48
Greene, T.R.  241 on transdisciplinary research  31
Greenman, B.  239, 243 hermeneutic circle  156, 158n3
Grethlein, Jonas  50 heroic characters  82
Grice, H.P.  287 heterophenomenology  6, 9, 196, 209, 213
Griffin, T.D.  241 rethought  197–​201
Groisman, E.A.  171 Hetrick, S.E.  175
Gropius, W. 86 Hickok, G. 225
Gross, J.J.  84 Hill, J. 269
Gross, Sabine  20, 22, 29 Hirsh, J. 269
Grosso, M. 69 His Master’s Voice (Lem)  79
Groth, H. 136 Hitchcock, Alfred  205
Gulliver’s Travels (Swift)  317 Hodges, S. 267
Hodgson, William Hope  78
Hagoort, P. 295 Hoeks, J. 311
Haidt, Jonathan  62–​63 Hoffman, A. 121
Hakemulder, F. 251 Hoffmann, E.T.A.  313–​314
Haldeman, Joe 78 Hofmann, M.J.  306, 309
Halliwell, Stephen  142, 263 Hofstadter, Douglas  23
Hamburger, Käte  47n3 Hogan, Patrick  6–​7, 9, 18, 32
Hamilton, C.A.  74 on cognitive and affective
Hamm, A.O.  67 narratology 35
Hancock, J.T.  175 on cognitive patterns  98
Handbook of Affective Sciences (Haidt)  62 on collaboration  26
Harder, Peter  288, 291 on construal level theory (CLT)  73
Harris, P.L.  264, 266, 268 on embodied reading  310
Harrison, Harry  81 on levels of explanation  23–​24, 30n14
Harrison, M.R.  269 on literary emotions  36–​37, 39–​42
Harry Potter (Rowling)  312 on Miller’s After the Fall  113–​133
Hart, F.E.  47 on ‘natural’ narratology  51
Hart, P.S.  84 on readers’ lives and narrative
Hartfeld, K. 306 experiences  247, 249–​250
Hartner, Marcus  6–​7, 10, 12, 195 Hogg, M.A.  247
on literature and cognitive Hohlfeld, A. 309
science  17–​34 Hohwy, Jacob  152, 155,
Hauk, O. 225 157–​159, 159n5
heart rate variability  316–​317 Holden, M. 265
‘He Built a Crooked House’ (Heinlein)  78 Hölderlin, Friedrich  316
Heilmann, A.  143n5 Holford, Castello  80
Heinlein, Robert  78–​79 Hollinger, V. 73
Helkama, K. 40 Holloway, John  86
Hellman, L. 272 Holstege, G. 70
von Helmholtz, Hermann  152 Homer 259

[ 334 ] Index
 

Horace 290 and suspense  315

Horn, L.R.  287 unconscious  9, 118, 152, 154,
Horwood, William  202–​203, 206 159–​162, 166
How Authors’ Minds Make Stories inferior frontal sulcis  315
(Hogan) 115, 120 information, paratextual  137
How Literature Plays with the Brain information, textual  240
(Armstrong) 156 information chunks  104
How to Do Things with Fictions information compression  98
(Landy) 141 information pathways  95
Hsu, C.-​T.  312, 320 Ingarden, Roman  158n3
Huch, Ricarda  310, 311 Inheritors, The (Golding)  79
Huddleston, R. 284 innovation, mobile tech  13
Huizinga, J. 267 Inohara, K. 250
human-​animal relationships  195, 200, insight in reading  187
209, 214 insight value  20
humanities 30n14, 136 integration, conceptual  288
Husband, Edward  281, 294–​295, integration process in reading  240–​241
297–​298 intentional stance  37, 199
Hutto, D.D.  141n3, 142, 199 interdisciplinarity  4, 8, 10, 19–20, 22,
Huynh, H.K.  70 26, 26n9, 30, 31, 320
Hyltenstam, K.  120n5 Interpersonal Perception test  269–​270
Hyman, Steven  191–​192 Interpersonal Reactivity Index (IRI)  138
‘hyper-​priors’  160, 162–​163, 165 interpretation  170, 187, 190
intersequencing 314n1
Ibsch, Eldrud  17 Interstellar (film)  77
iconography 155 intersubjective experience  170
idioms  308–​309 introspection 4, 114
idiosyncrasies, readers’  40 Inverted World, The (Priest)  78
illusion  9, 148, 154, 157 ‘I-​Origo’  47n3
imagery  63–​65, 98, 100, 176 irony 295
imagination  1, 9, 55, 60, 118–​119, 140 Iser, Wolfgang  20, 29, 38, 142n4, 158n3
embodied 220
and the fantastic  165 Jackendoff, R.  283, 294
and fiction  135–​136, 274 Jackson, M.P.  104
‘picture in the head’ type  140 Jackson, Tony  17, 26–​27, 29–​30
and play  266–​267 Jacobs, Arthur  6–​7, 11, 191, 236
and simulation  121, 260 on neurocognitive poetics  303–​325
immersion  5–​6, 138, 191, 319 Jacovina, Matthew  245–​246, 250
and fear  315 Jajdelska, E. 140
in fictional world  137 Jakobson, Roman  303, 309, 316
neural correlates of  312–​313 James, Henry  30, 290
in novel reading  320 James, S.P.  199
indirection 233 James, William  153
individual differences  252 Jameson, Frederic  78, 83, 85
inferences  158, 266, 287 jargons 84
explicit 249 Jee, B.D.  241
intellectual 190 Jenkins, J.M.  41, 119
readers’  242–​244 Jensen, M.S.  159
relational 138 Jeschke, Wolfgang  79
and simulation  264 Johnson, D.R.  270

Index  [ 335 ]
 

Johnson, M.  36, 165, 227 Komeda, H. 250

Jones, E.E.  264 Koops van ‘t Jagt, R.  311
Jones, Gwyneth  82 Kotovych, M. 265
Journal of Literary Semantics 1 Kotz, S.A.  308
Journey to the Centre of the Earth Krakowiak, K.M.  253
(Verne) 77 Kraxenberger, M. 316
Joyce, James  11, 36, 42–​46, 293 Kreuziger, F.A.  82
Junghofer, M. 67 Kubrick, Stanley  77
Kucharski, M. 308
Kacinik, N.A.  69 Kuchinke, L.  306, 308
Kagan, J. 196 Kuhlmann, Marco  307
Kahneman, Daniel  36–​37, 39, 75, 243 Kuhn, Thomas  19
on science fiction  80 Kuijpers, M.M.  251
Kandel, Eric  152, 154–​155 Kuiken, Don  187, 318
Kant, Immanuel  55–​58, 60 Kukkonen, Karin  5, 7–​12
Kanwisher, N. 64 on enactive perception  141
Kaschak, M. 224 on fantastic cognition  151–​167
Katz, R. 175 on ‘natural’ narratology  47
Kaufman, G.F.  252 on reading and eating
Kaye, W.H.  191 disorders  169–​170, 190
Keen, S.  247, 249–​250, 264 Kuperberg, G.R.  281, 294–​296
Kehler, A.  187–​188 Kurzweil, R. 93
Keidal, J.L.  309 Kusumi, T. 250
Kelly, E.F.  69 Kuzmicova, A.  140, 170
Kelly, L.A.  281, 294
Keltner, D.  41, 84, 119 Lac, A. 247
Kenny, R.  121n7 Lacan, Jacques  127, 131
Kerry, S. 185 Lady Chatterley’s Lover (Lawrence)  232
Kesner, Ladislaw  155 Lakoff, G.  36, 165, 227, 229
Keysers, C.  136n1 Land and Overland (Shaw)  81
Kidd, David Comer  137, 271 Landy, Joshua  141–​142, 147
Kierkegaard, S.  272, 274 Langacker, R. 289
‘Killers, The’ (Hemingway)  208 language 131
Kim, E.Y.  76 acquisition 120n5
kinaesthetics 13 action-​related  225
Kinder, A.  305–​306 comprehension 121n7
King Lear (Shakespeare)  66 everyday 288
Kintsch, Walter  240–​241, 305 figurative 308
Kirsh, Steven J.  175 games  207–​208
Kissler, J. 67 literary 284
Klein, J.T.  26n9 poetic 292
Kliegl, R. 309 processing 229, 236
Knauer, V. 127 vernacular 201
Knight, Eric  209–​210, 212 written 157, 225
Knoop, C.  308, 315 language-​emotion gap  309
knowledge  188, 242, 244 Lassie, Come Home (Knight)  209, 212
Koelsch, S. 315 Last Day of Creation, The (Jeschke)  79
Koestler, Arthur  307 lateral premotor cortex  315
Kohányi, A. 267 Latzer, Y. 175
Kohn, E. 213 Lauer, G. 264

[ 336 ] Index
 

Lauwers, P. 283 as cultural or political

Lawrence, A. 127 phenomenon 28
Lawrence, D.H.  232–​233 mismatch and coercion in  289–​293
Leder, H.  306, 308 value for understanding emotion and
LeDoux, J. 120 simulation  115–​119
LeDoux, Joseph  39, 41, 95 Liviatan, I. 82
“Leerstellen’ (empty places, gaps)  158n3 Lodge, D. 84
Left Hand of Darkness, The (Le Guin)  79 London, J. 212
left inferior frontal gyrus (LIFG)  297, Longinus  7, 61, 65, 67, 70
298, 306–​307, 316 Lord of the Rings (Tolkien)  156n1
Le Guin, Ursula K.  79, 82 Lorenz, Konrad  200–​201
Lehne, M.  314–​316 love stories  270
Lehrer, J. 113 Lubrich, O.  308, 315
Lem, S.  79–​80 Luca, C. 76
Léon, J. 283 ‘lucid self-​delusion’  142, 147
Leuthold, H. 244 Lüdtke, J.  236, 309, 315, 317
Leverage, P. 264 Luguri, J.B.  76, 83
Levine, Joseph  28n11 Lund, T.E.  316
Levinson, Jerrold  40 Luria, Alexander  266
Levinson, S.C.  84, 207 lyric  98–​99, 108, 188
Lévi-​Strauss, Claude  131 lyric–​narrative distinction  7–​8
Levy, R.  187–​188
Lewis, W. 86 M350 effect  295, 298, 299
Lexical Constructional Model  288 Maas, A. 224
Liberman, N.  74, 76, 81–​83 MacBeth (Shakespeare)  59, 65
on construal level theory (CLT)  86 MacInnis, C.C.  247
Libet, Benjamin  28 Macrae, C.N.  247
Lilith’s Brood (Butler)  79 Madden, Deirdre  233
Limbach, J. 305 Magliano, J.  240, 242
line length  104 magnetoencephalography (MEG)  294–​
linguistic cues  158 295, 298, 299
linguistic register  74, 83 Mahon, B. 225
linguistics, cognitive  6, 48, 287–​288 Majkut, P. 190
literacy programmes  266 make-​believe  136n1, 265
literal expressions  308 Malevich, K. 86
literariness 316 Mancing, H. 264
literary analysis  117 Manifold (Baxter)  77
literary criticism  221–​222, 234–​235 Man in the High Castle, The (Dick)  80
literary experience  219–​237 manner, maxim of  287
literary fiction  137 Mano, Q. 226
Literary Reading, Cognition and Emotion Mansour, J. 27
(Burke) 35 Mar, Raymond  265–​266, 269–​270
‘literary reading loop’  222 March-​Russell, P.  86
literary theory  264 Marinetti, F.T.  86
literature 25 Markey, P.S.  308
and affective science  113 Martin, C.D.  66, 309
artistic 274 Martin, George R.R.  156n1
as a biological phenomenon  24 marvellous genre  156n1, 159, 164
and cognitive science  17–​34 Marxist criticism  85
as cognitive training  8 Massey, I. 222

Index  [ 337 ]

mathematics 268
Mathewson, K.E.  159
Mayer, J.D.  46
Mayer-​Salovey-​Caruso emotional
intelligence scale  46
McCarthyism 115n3
McClune, M.S.  270
McComas, K.A.  84
McConachie, B. 47
McConaha, C. 191
McConkie, G.W.  317
McElree, B.  294–​295
McHale, B. 86
McKoon, Gail  242
McNamara, D.S.  240, 242
meaning construction  48, 304
media, thin-​ideal  175–​176
medial frontal cortex  315
medial prefrontal cortex  308, 313
memories, accumulated  240, 243,
252–​253
memory 64, 135
autobiographical 2
emotional 5, 113–​133
and narrative
experiences  240–​245, 254
working 98, 104
memory-​based processing  242
Menary, R.  139, 169
Menninghaus, Winfried  308, 310, 315
mental imagery  5, 170
mentalizing 138, 199
mental models  264–​265, 268
mental states  25, 141, 200
attribution of  12, 204, 207, 209
paradoxical 142
Méquinion, M. 191
metaphor  106n2, 165, 186–​187, 299
‘career of’ theory  227
and coercion  283, 287–​289, 291–​292
conceptual 230
and embodied dynamics  223, 226
and imaginative play  267
in literary criticism  235
meanings 229
and poetic language  293
and transformation  268
metarepresentation 21
meta-​theoretical overview  7
metonymy  283–​284, 289, 299
[ 338 ] Index
and coercion  287–​289, 291–​292
and poetic language  293
Meyer, D.E.  240n1
Meyer-​Sickendieck, B.  236, 317
Miall, David  5–​7, 11–​12, 35
on defamiliarization  318
on literary emotions  38
on reading and eating disorders  170
on sublime experiences  55–​72
Michaelis, L.A.  283
micropoetry  305–​306
mid-​cingulate cortex  312
Midgley, M. 27
Mieder, W. 308
Miéville, C. 85
Miller, Arthur  113–​133
Miller, C. 290
Miller, Dale  243
Mills, L. 127
Milner, A. 73
Milton, John  59, 65
mimesis  142, 154, 158, 158n3, 259, 263
mind, computational model of  136, 139
mind, life of  47
mind, literary  165
mind, philosophy of  201
mind, theory of  9, 21, 114, 169, 264,
270, 295, 299
and neurocognitive poetics  313
and understanding others  269
mind-​ascribing acts  197, 200,
208, 213
mind-​body continuum  27–​28, 46,
169–​170, 172
mind–​brain  200n2
Mind in the Eyes test  269–​271
mind–​narrative nexus  213
minds, animal  195–​216
minds, non-​human  7–​8
minds, other  118, 137
minds, social  21
minds, species of  196, 208, 214
‘mind style’  202n3
Minuchin, Salvador  175
mirror neurons  135, 225, 320
misdirection 144
mismatches, grammatical  281, 299
in linguistics  281–​286
in literature  289–​293
in neuroscience  294, 296, 298

Mitchell, J.P.  247
Mithen, Steven  267
Mitrophanov, A.Y.  171
models, mental. See mental models
moderation  18, 25–​29
Modernism 7, 86
Mohr, S. 311
Moldoveanu, M.C.  271–​272
Molé, P.A.  172
Molly Fox’s Birthday (Madden)  233
moments, mental  98
Monkey Taming (Fathallah)  179
Montague grammars  286
Montalván, Luis Carlos  209, 211
Mont Blanc  55, 65, 71
mood  179–​180, 184
Moore, C. 270
morality  40, 82–​84
Moran, M.B.  252
Moretti, F. 86
Morgan, Lady  57–​61, 63–​64
motivation 38
Moylan, Tom 83
Mozart, Wolfgang Amadeus  289
Mullin, Justin  270
Mumper, Micah  6–​8, 11, 191
on readers’ lives and narrative
experiences  239–​257
Murphy, B.  127n10
Murphy, G. 121
Murphy, S.T.  252
Murray, P.  61–​62
Murray, R.M.  171
music  69–​70, 289
Myers, J.L.  240
Myin, E.  141n3, 142
mysteries, narrative  243
mystic experiences  69–​70
Myyry, L. 40
N400 effect  295–​296, 298, 299
Nagel, T.  196–​198
Nagel–​Dennett debate  197, 199, 201
nanotechnology 78
Napier, J.L.  76, 83
Napoleon 59
narration, scene-​setting  204
narrative  93–​109, 123, 159, 165
and eating disorders  188
experiences of  239–​257
rhythms of 140
universals in  314n1
narratives, animal  205
narratives, emotional  126, 158
narrativity 50
narratology  9, 196, 251
analysis 214
and animal minds  196–​197,
201–​207
cognitive 19, 27n10
cognitive and affective  35–​53
natural  8, 36, 47–​51
theory formation  46
narrators, intradiegetic  201
nature  55, 59–​60, 63–​64
Nearly Perfect Copy, A
(Amend) 248, 252
Neisser, U. 279
Nell, Victor  303, 317
Neural Sublime, The (Richardson)  155
neuroaesthetics 19, 308
neurocognitive poetics model
(NCPM)  308–​309, 312, 315, 319
neurolinguistic studies  299
neuropsychology  57, 139–​140
neuroscience 148, 261
cognitive 293
computational 135
second-​person  13
‘Neuroscience Delusion, The’ (Tallis)  22
Newberg, A.B.  69
Nicholls, Peter  77
Nickerson, R. 264
Nielsen, A.H.  316
Nieuwland, M.S.  244
Niffeneger, Audrey  78
Night Land, The (Hodgson)  78
Nisbett, R.E.  121, 264
Niven, Larry  77
Noë, Alva  139–​141, 153
on animal minds  200–​201
Nolan, Christopher  77, 143n5
non-​fiction  7, 188–​189
norm theory  243, 254
Norris, D. 188
novel-​reading studies  317
novels 114, 156n1
novelty 289. See also surprise
Nunberg, G. 281
Nusbaum, E.C.  69
Index  [ 339 ]

Oatley, Keith  6–​9, 11, 227, 236
on emotion and simulation  119
on literary emotions  36–​37, 41,
114, 318
on truth and fiction  259–​278
O’Brien, E.J.  240, 240n1, 242
Ochsner, K.N.  84
O’Craven, K.M.  64
O’Donnell, P.J.  309
Oeberst, A. 308
Oliver, M.B.  187, 253
Olson, D.R.  264, 266
Olson, Greta  30
online processing  319
On the Origins of Stories (Boyd)  93, 98
ontology, cultural  213
openness, interpretive  188
openness trait  271
orbitofrontal cortex  313
O’Regan, J.K.  153
organon model  303
Orthogonal (Egan)  81
Ortony, Andrew  253, 288
O’Sullivan, N. 311
outcome preference  245–​246
overreaction, emotional  121–​125
Owen, Wilfred  292
oxford-​scholarship.com  5
P600 response  9, 66–​67, 296, 298, 299
Paczynski, M. 294
Page-​Gould, E.  247
Palimpsest (Stross)  81
Palmer, Alan  21, 49n5
Panksepp, Jaak  148
Panksepp-​Jakobson hypothesis  309, 312
Panofsky, Erwin  155
paradigm shifts  19
Paradise Lost (Milton)  59
paradox of caring  25n8
paradox of fiction  40, 268
parallelism 98
parallel processing  1
Paratime (Piper)  81
parietal lobe  69
Parrinder, P. 83
Partee, B. 283
participatory responses,
readers’  245–​251
Paterson, Don  94, 105
[ 340 ] Index
Patnoe-​Woodley, P.  252
pattern recognition  8, 279–​281, 289
patterns, cognitive  48, 93–​109
patterns, cross-​cultural  41
patterns, story  36
Paulson, W. 23
Pécheux, M. 283
Pelowski, M. 308
penetrability, cognitive  158–​159, 165
perception  64–​65, 119–​120
enactive  139–​143
false 191
and fantastic cognition  151
illusions 157
shortcuts 154
visual 135
of visual art  152–​153
perceptual distortion  185
Peregrine, The (Baker)  205–​206
personality  69, 114, 272–​273, 313
perspective  40, 113, 187
persuasion 6, 273
Peterson, Jordan  269–​271
Petrarchan sonnets  102, 188
Peyk, P. 67
phenomenology  139, 197–​198, 201
Picturesque Tour through the Oberland
(anon.) 60
Pilkington, A. 84
Piper, H. Beam  81
Pirlet, Caroline  5–​7, 11–​12
on cognitive and affective
narratology  35–​53
on reading and eating disorders  170, 191
Plato  259, 263, 268
play, imaginative  265–​268, 274
pleasure, aesthetic  290
plot  50, 95, 186, 190
Plumb, I. 269
poetic function  309
poetic licence  22, 30–​31
Poetics (Aristotle)  263
poetics, cognitive  19, 35–​36
poetics, neurocognitive  304, 308
poetic texts  170
poetry  259–​260
micro-​  305
reading 311, 320
reception 307
variants of 316
 

poets 98, 104 quantity, maxim of  287

‘point’ in narrativity  50, 104 quasi-​emotions  40, 260
point of view  121 Quasimodo, Salvatore  319
politeness  83–​84
Polvinen, Merja  6–​9, 191 Rachman, S. 185
on cognitive science and Radach, Ralph  317
fiction  135–​150 Radford, Colin  260
Ponz, A. 306 Radvansky, G. 305
Possession (Byatt)  143 Rain, M. 270
possible worlds  201, 263, 268 Rama (Clarke)  77
Post, T.A.  241 Ramsey, W.  24n7
Powell, B. 259 Rapp, D.N.  246, 250
praxis 141 Raste, Y. 269
prediction  9, 155, 158, 192 Rast III, D.E.  247
predictive models  162, 190 Ratcliff, Roger  242
predictive processing  151–​159, 164–​166, ‘rauding’ 317
188, 191 reader parameters, cognitive  49
prefocusing, narrative  98n1 reader personality  313
Prentice, Deborah  250, 262 reader response  12, 36, 86, 311
Prestige, The (Priest)  143–​148 readers 7
Priest, Christopher  78, 80, 143–​148 blank-​slate  4
priming 13, 240n1 changed by literary
Prinz, J.J.  36, 69 engagement 187, 236
probabilistic models  8, 151–​152 empathic 45
‘probability design’ of narratives  160 of fiction  136
problem-​solving instructions  245 judgements of 274
prose, variants of  316 naïve  9, 222, 236
Proust, Marcel  113 recreational 234
proverbs  308–​309 self-​reports  137, 252
proximity, law of  280 uniqueness of narrative
psychoanalysis  57, 127, 129n11, 131 experiences  239–​257
psycholinguistics 303 reading 7, 10
psychologists, discursive  41 as challenge  182
psychology, cognitive  239, 261, 318 close 234
psychology, developmental  6, 136n1 embodied 46, 221
psychology, personality  45n2 emotion-​based  45
psychology, social  6, 73, 264, 274 and feedback systems  186
psychology of art  154 literary  303–​325
psychology of fiction  261 and mental health  192
psychotherapy 274 non-​scientific aspects of  29
Pulvermüller, F. 225 professional vs. recreational  190
Pupil, The (James)  290 research  303–​304, 318
pupil size variation  311 solitary vs. group  190
purposiveness (in science fiction)  74, 81, 84 stages in 221
Pustejovsky, James  286–​287 study of 164
Pylkkänen, L.  281, 294–​295 subjective 28
Pythagoras’s theorem  259 realism 147
reason, powers of  55–​56
qualia features  286n1 reasoning, verbal  270
qualities, emergent  23–​24 Reber, R. 308

Index  [ 341 ]
 

reception theory  74 Ruyter, K.D.  252

Reeve-​Tucker, A.  83 Ryan, Marie-​Laure  19, 21, 27n10, 29
relativity theory  78
relevance theory  227 Salovey, P. 46
Renaissance Extended Mind, The Sanders, Scott  82
(Anderson) 155 ‘Sandman, The’ (Hoffmann)  314
Republic, The (Plato)  259, 263, 268 Sanford, A.J.  74
research, cognitive-​literary  3 Sappho  7, 12, 61, 63, 71
research, empirical  148 sarcasm 295
research, field  116 Schabmann, A. 308
research, transdisciplinary  31 Scheepers, C. 311
Reynolds, K.J.  246 Scheff, T. 127
Rezaie, R. 66 schematization 48, 281
rhetoric  1, 6, 30, 186 schizophrenia 191
Rhetoric of Fictionality (Walsh)  142 Schlesewsky, M. 297
Rhine Falls  67–​69, 71 Schlotz, W. 308
rhyme 98, 305 Schmidhuber, J. 290
rhythm 98, 140 Schmidt, S.J.  29
Richardson, Alan  19, 136, 138, 155, 165 Schneider, R.  21, 74
Richter, T. 252 Schrott, Raoul  303–​304, 306–​307,
Ricoeur, Paul  142n4, 158n3, 208 309–​310, 319
Rieger, C.J.  242 Schuerewegen, Franc  165
right posterior cingulate cortex  313 Schupp, H.T.  67
rigour and fidelity  4 Schvaneveldt, R.W.  240n1
Rinaldi, S. 188 Schwartz, N. 308
Ringworld (Niven)  77 Schwarz, N. 80
Riven, L. 281 Schweickert, R. 264
Rizzolatti, G.  136n1, 225 science  19, 221–​222
‘Road Not Taken, The’ (Frost)  227, 232 Science and Literary Criticism,
Roazzi, A. 266 symposium (2012)  3
Roberts, Adam  79, 311 science fiction (SF)  5, 73–​92, 146, 270
Roberts, N. 66 Scott, G.G.  309
Robinson, Jenefer  38–​40 Scott, Ridley  79
Robinson, Kim Stanley  80 scripts 48
Robu, C. 77 Seidenberg, M. 226
Roepstorff, A. 316 self, annihilation of  55–​72
Rohde, H.  187–​188 self–​confidence  172
Rolfs, R. 309 self–​esteem  175, 178–​180
Romantic Sublime, The (Weiskel)  56 selfhood 268, 271
Rooth, M. 283 self–​limiting events  183
Rosman, B.L.  175 self–​other dichotomy  69
Ross, A. 169 self–​other relationships  196
Ross, L. 121 self-​reports, readers’  137, 252
Rowling, J.K.  312 self–​report studies  116
Royal, D.  115n3 self–​revelation  247–​248
rules, categorization  121 semantic comprehension  187
rules in art, lack of  57 semantic networks  126
Rushford, N. 175 semantic processing  306
Russo, A. 224 semiotic feedback loops  188
Rutter, Barbara  307 sensitivity, emotional  45n2

[ 342 ] Index
 

sensorimotor perception  140 social cognition  5, 136n1, 295, 298,

sensory properties  119 299, 315
sentences 309 social comparison theory  175
Separation, The (Priest)  80 social descriptions  76
Sereno, S.C.  309 social interaction  247
settings 95, 101 socialisation 39, 264
sexuality  69–​70 social norms  117
Shafran, R. 185 social phenomena  24
Shakespeare, William  7, 12, 94, 273, sociolects 84
307–​308 sociology of science  201
and coercion  290–​291 Socrates 259
and the sublime  59, 66 Södersten, P. 191
and Twelfth Night  95–​98 Solaris (Lem)  79–​80
Shakespeare’s Brain (Crane)  155 Sommer, W. 309
Shapiro, K.J.  199 Song of Ice and Fire (Martin)  156n1
shared emotions problem  42, 49 Sonnets (Shakespeare)  94, 99, 101–​108
Shaw, Bob 81 sound echoes  304–​305
Shaw, Philip  65 sound gestalt  307
Sheenan, Paul  18, 20 space, possibility  108
Sheikh, N.A.  309 space, sense of  69
Shelley, Mary  147 spaces, fictional  140
Shelley, Percy Bysshe  65–​66, 71 species boundary  214
Shibley Hyde, J.  175 spectroscopy, near-​infrared  313
Shkurko, A.V.  246 speech 202, 284
Sign Based Construction Grammar  288 Speer, N.K.  27n10
signifiers 127, 131 Sperry, Roger  24
Silvia, P.J.  69, 290 Spiegel, S. 146
Simenon, Georges  264 Spilich, G.J.  241–​242
similarity, judgements of  246–​250, Spivak-​Lavi, Z.  175
253–​254 Spolsky, E. 22
similarity, law of  280 Stabler, J. 311
simile 187 Staddon, J. 75
simplifications 154 Stanfield, R. 223
simulation, embodied  222–​227 Stanovich, K.E.  75, 269
and allegorical literature  227–​234 Stapledon, Olaf  79, 81
in literary criticism  234–​235 Star Maker (Stapledon)  79, 81
simulation in fiction  5, 9, 113–​133, Starr, G. Gabrielle  155–​156
136n1, 261–​265, 269, 271, 274 Starship Troopers (Heinlein)  79
Sitarenios, G. 46 Stearns, C. 41
situation models  305 Stearns, P. 41
situations  95, 106n2, 165 Stedman, R.C.  84
‘skimming’ 317 Stenneken, P. 304
Skinner, B.F.  75 Stephan, E. 83
Skins (TV programme)  183 Stephenson, Neal  78
Slaughterhouse Five (Vonnegut)  78 Sternberg, M.  35, 314
Slippage, The (Greenman)  239, 243–​244, Stevens, J.  121n7
246, 249–​251, 253 Stevenson, Robert Louis  147, 268
Smit, D.  115n3 Stewart, J. 139
Smith, J.L.  185 Stockburger, J. 67
Snow, C.P.  20, 196 Stockwell, Peter  35–​36, 48, 84

Index  [ 343 ]

Stonor Eagles, The (Horwood)  202
stories  48, 129, 262
animals in 202
story comprehension  39
‘Story of Your Life’ (Chiang)  78
storytelling  35, 195, 213–​214
Stross, Charles  81
structuralism 19
structure, emergent  27–​28, 32
style  84, 152, 155
stylistics, cognitive  6, 202
subjectivities, non-​human  207, 213
subjectivity, ascriptions of  199, 202
subject-​transitive construction  288
sublime, literature of the  61
sublime, spatial  77
sublime experiences  5–​7, 9–​10
cognitive challenge of  55–​72
in science fiction  77
subvocalization  113–​114
Such Stuff as Dreams (Oatley)  268
suffering and empathy  249
‘super-​priors’  161
surprise  62, 290, 314n1
suspense  253, 314–​316
Suvin, Darko  73, 83
syllogisms  266–​267
sympathy  37–​38, 42, 45–​46, 126
syntactic anomalies  66
syntactic processing  188
Tackett, J.L.  270
Taft, C. 127
Tallis, R. 22
Talmy, L. 283
Tal-​Or, N.  252–​253
Tan, E.S.  251
Tau Zero (Anderson)  77
Taylor, L. 224
Taylor, M. 267
Teachman, B. 185
templates 108
tenses 68
terror 60, 71
text comprehension  187, 240
text information  305
textual analysis  21
textual cues  35
‘thick descriptions’  47
Thierry, G.  66, 309
[ 344 ] Index
thinking  22, 138, 180–​181, 185, 268
Thompson, Evan  139–​140, 198n1
thought, free indirect  204
thought presentation  196, 202, 213
thoughts, content of  64
‘thought-​shape fusion’  185
Three Stigmata of Palmer Eldritch, The
(Dick) 80
time, narrative  4, 108
timeless, sense of  60
Time Machine, The (Wells)  78
Time Traveller’s Wife, The
(Niffeneger) 78
Tinbergen, Nikolaas  200
Titone, D.A.  309
Todorov, Tzvetan  76, 151–​152, 156,
159–​160
Tolkien, J.R.R.  156n1
Tomasello, M. 279
topologies of space  78
Tourmalin’s Time Cheques (Anstey)  78
Trabasso, T. 264
transcendence 69, 82
transdisciplinarity  10, 195, 197, 214
transference, psychoanalytic  131
transformation of selfhood  271
transformation sets  131
transforming experiences  69
transportation  191, 251–​253
and empathy  270
measures of  8, 137–​138
to narrative worlds  240
Trapp, S. 306
Trevarthen, C. 199
Trope, Y.  74, 76, 81–​83
on construal level theory (CLT)  86
Troscianko, Emily  3, 5–​7, 9, 12,
17, 73
on enactive perception  140, 148
on narrative and metaphor  106n2
on ‘natural’ narratology  47
on reading and eating
disorders  169–​194
truth and fiction  259–​278
Tsunemi, K. 250
Tsur, R. 70
Turner, J.C.  246
Turner, Mark  21, 47, 165, 229, 288
Turtledove, Harry  81
Tversky, Amos  80, 247–​248
 

Twelfth Night Vô, M.L.-​H.  306

(Shakespeare)  66–​67, 94–​98 von Helmholtz, Hermann  154, 159, 161
type-​shifting  281, 283, 286. See also von Kleist, Heinrich  315
coercion Vonnegut, Kurt  78
von Uexküll, Jacob  198
Ullrich, S. 316 Voss, J.F.  241
Umwelt 198 Vuust, P. 316
‘Unbestimmtheitsstellen’ (spots of
indeterminacy) 158n3 Waddell, N. 83
‘Uncanny, The’ (Freud)  314 Wakslak, C.J.  82
uncanny genre  159 Walker, J. 127
unconscious, cognitive  4, 152 Wallace, B. 169
understanding, linguistic  222–​227 Wallentin, M. 316
understanding others  268–​271 Wallot, Sebastian  318
unity of self and universe  70 Walsh, Richard  142, 188
unreality, expectation of  147 Walton, Kendall  40, 136n1,
Until Tuesday (Montalván)  209 260, 265
utopianism 83 Wänke, M. 80
Ward, L.M.  175
vacillation 8 War of the Worlds (Wells)  79
Valéry, Paul  305 Watson, J. B.  75
validity, representational  116–​117 Watts, Peter  80
VALIS (Dick)  80 Waugh, Evelyn  208
van Berkum, J.J.A.  244 Waugh, Patricia  18–​19, 31
van den Hende, E.A.  253 Wecker, H.  241, 244
van Dijk, T.A.  305 Wegner, P.E.  86
van Laer, T.  252 Weike, A.I.  67
van Peer, W.  308, 318 Weiskel, Thomas  56–​57
variability, interpersonal  4 Wells, H.G.  78–​79
variation, individual  12 Wenzel, W.G.  243
Varieties of Presence (Noë)  140 Wernicke’s area  294
Vendler, Helen  100, 104, 220–​221, 230, West, R.F.  269
234, 273 West, Tessa  247–​248
ventromedial prefrontal cortex West of Eden (Harrison)  81
(vmPFC)  295, 298, 299 Wetzels, M. 252
verb phrase, future-​directed  43 ‘What Is It Like to be a Bat?’
verisimilitude 84, 141 (Nagel) 197
Vermeer, Johannes  290 What Literature Teaches Us About Emotion
Verne, Jules  77 (Hogan)  113, 115, 119
verse  93–​109 Wheelwright, S. 269
Vesonder, G.T.  241 Why Lyrics Last (Boyd)  93, 98–​100
Viconti, L.M.  252 Wilden, A. 24
View from Nowhere, The (Nagel)  198 Wiley, J. 241
Vile Bodies (Waugh)  208 Wille, K. 309
violations, grammatical  298 Willems, D. 283
visual art 8 William, J.M.  264
visualization, reader’s  113–​114 Williams, Helen Maria  63, 67–​69, 71
visual processing  64 Wilson, E.O.  75
Viveiros de Castro, E.  213 Wilson, N. 225
vividness 140 Winkielman, P. 308

Index  [ 345 ]

Winnicott, Donald  268
Wirag, Andreas  5–​7, 11–​12
on cognitive and affective
narratology  35–​53
on reading and eating
disorders 170, 191
Wittgenstein, Ludwig  141, 207
Wolf, W. 139
Woltin, K.A.  76, 83
wonder  62, 63, 65, 68
Wood, James  235
Woolf, Virginia  59, 235
Woolfson, Esther  202–​203
word pairs  305, 307
word processing, affective  306
word recognition  304, 306
Wordsworth, William  56, 70
working memory  98, 104
world-​creation  201, 263
worlds, nested  202
worlds, possible  268
writers, professional  267
[ 346 ] Index
Xeelee (Baxter)  79
Yarley, R. 223
Yarmolinsky, A. 272
Years of Rice and Salt, The (Robinson)  80
Yiddish Policeman’s Union, The (Chabon)  80
Young, R. 84
Zacks, J.M.  27n10, 305
Zahavi, D. 139
Zaki, Jamil  249–​250
Zatorre, R.J.  70
Zeeman, E.C.  174
Zeki, Semir  152
Zeman, Adam  316, 320
Zhu, D.C.  281, 294
Ziegler, D.  281, 283, 288
Ziegler, J.C.  304
Zillmann, Dolf  40
Zoeterman, Sara  271
Zunshine, Lisa  21, 73, 264
Zwaan, R.  223–​224