Вы находитесь на странице: 1из 521



The Routledge Companion to Philosophy of Literature is an in-depth examination of literature

through a philosophical lens, written by distinguished figures across the major divisions
of philosophy. Its 40 newly commissioned essays are divided into six sections:

•• historical foundations
•• what is literature?
•• aesthetics and appreciation
•• meaning and interpretation
•• metaphysics and epistemology
•• ethics and political theory.

The Companion opens with a comprehensive historical overview of the philosophy of

literature, including chapters on the study’s ancient origins up to the 18th–20th
centuries. The second part defines literature and its different categories. The third
part covers the aesthetics of literature. The fourth and fifth sections discuss the meaning
and consequences of philosophical interpretation of literature, as well as epistemological
and metaphysical issues such as literary cognitivism and imaginative resistance. The
sixth section contextualizes the place of philosophy of literature in the “real world” with
essays on topics such as morality, politics, race and gender.
Fully indexed, with helpful bibliographies at the end of each chapter, this Companion
is an ideal starting point for those coming to philosophy of literature for the first time as
well as a valuable reference for readers more familiar with the subject.

Noël Carroll is Distinguished Professor of Philosophy at the Graduate Center, City

University of New York. He is the author of many books and essays in the philosophy
of literature, aesthetics, and the philosophy of film. Recent books include Art in Three
Dimensions, On Criticism, Humor: A Very Short Introduction, Living in an Art World, and
Minerva’s Night Out: Philosophy, Pop Culture, and Moving Pictures.

John Gibson is Associate Professor of Philosophy at the University of Louisville. His

research focuses on topics in the philosophy of literature and aesthetics. Among his
publications are Fiction and the Weave of Life (2007) and, as editor, The Philosophy of
Poetry (2015). He is currently working on a book titled Poetry, Metaphor & Nonsense:
An Essay on Meaning.
Routledge Philosophy Companions

Routledge Philosophy Companions offer thorough, high quality surveys and assessments of
the major topics and periods in philosophy. Covering key problems, themes and thinkers,
all entries are specially commissioned for each volume and written by leading scholars
in the field. Clear, accessible and carefully edited and organised, Routledge Philosophy
Companions are indispensable for anyone coming to a major topic or period in philosophy,
as well as for the more advanced reader.

The Routledge Companion to Aesthetics, Third Edition

Edited by Berys Gaut and Dominic Lopes
The Routledge Companion to Philosophy of Religion, Second Edition
Edited by Chad Meister and Paul Copan
The Routledge Companion to Philosophy of Science, Second Edition
Edited by Martin Curd and Stathis Psillos
The Routledge Companion to Twentieth Century Philosophy
Edited by Dermot Moran
The Routledge Companion to Philosophy and Film
Edited by Paisley Livingston and Carl Plantinga
The Routledge Companion to Philosophy of Psychology
Edited by John Symons and Paco Calvo
The Routledge Companion to Metaphysics
Edited by Robin Le Poidevin, Peter Simons, Andrew McGonigal, and Ross Cameron
The Routledge Companion to Nineteenth Century Philosophy
Edited by Dean Moyar
The Routledge Companion to Ethics
Edited by John Skorupski
The Routledge Companion to Epistemology
Edited by Sven Bernecker and Duncan Pritchard
The Routledge Companion to Philosophy and Music
Edited by Theodore Gracyk and Andrew Kania
The Routledge Companion to Phenomenology
Edited by Søren Overgaard and Sebastian Luft
The Routledge Companion to Philosophy of Language
Edited by Gillian Russell and Delia Graff Fara
The Routledge Companion to Philosophy of Law
Edited by Andrei Marmor
The Routledge Companion to Social and Political Philosophy
Edited by Gerald Gaus and Fred D’Agostino
The Routledge Companion to Ancient Philosophy
Edited by Frisbee Sheffield and James Warren
The Routledge Companion to Eighteenth Century Philosophy
Edited by Aaron Garrett
The Routledge Companion to Bioethics
Edited by John Arras, Rebecca Kukla, and Elizabeth Fenton
The Routledge Companion to Virtue Ethics
Edited by Lorraine Besser-Jones and Michael Slote
The Routledge Companion to Hermeneutics
Edited by Jeff Malpas and Hans-Helmuth Gander
The Routledge Companion to Islamic Philosophy
Edited by Richard C. Taylor and Luis Xavier López-Farjeat
The Routledge Companion to Philosophy of Literature
Edited by Noël Carroll and John Gibson

The Routledge Companion to Sixteenth Century Philosophy
Edited by Benjamin Hill and Henrik Lagerlund
The Routledge Companion to Seventeenth Century Philosophy
Edited by Dan Kaufman
The Routledge Companion to Medieval Philosophy
Edited by Richard Cross and JT Paasch
The Routledge Companion to Philosophy of Race
Edited by Paul C. Taylor, Linda Martín Alcoff, and Luvell Anderson
The Routledge Companion to Environmental Ethics
Edited by Benjamin Hale and Andrew Light
The Routledge Companion to Free Will
Edited by Meghan Griffith, Neil Levy, and Kevin Timpe
The Routledge Companion to Philosophy of Technology
Edited by Joseph Pitt and Ashley Shew Helfin
The Routledge Companion to Philosophy of Medicine
Edited by Miriam Solomon, Jeremy Simon, and Harold Kincaid
The Routledge Companion to Feminist Philosophy
Edited by Ann Garry, Serene Khader, and Alison Stone
The Routledge Companion to Philosophy of Social Science
Edited by Lee McIntyre and Alex Rosenberg
The Routledge Companion to Aesthetics

“This is an immensely useful book that belongs in every college library and on the
bookshelves of all serious students of aesthetics.”—Journal of Aesthetics and Art

“The succinctness and clarity of the essays will make this a source that individuals not
familiar with aesthetics will find extremely helpful.”—The Philosophical Quarterly

“An outstanding resource in aesthetics . . . this text will not only serve as a handy reference
source for students and faculty alike, but it could also be used as a text for a course in the
philosophy of art.”—Australasian Journal of Philosophy

“Attests to the richness of modern aesthetics . . . the essays in central topics—many of

which are written by well-known figures—succeed in being informative, balanced and
intelligent without being too difficult.”—British Journal of Aesthetics

“This handsome reference volume. . .belongs in every library.” – CHOICE

“The Routledge Companions to Philosophy have proved to be a useful series of high quality
surveys of major philosophical topics and this volume is worthy enough to sit with the
others on a reference library shelf.”—Philosophy and Religion

The Routledge Companion to Philosophy of Religion

“. . . A very valuable resource for libraries and serious scholars.”—CHOICE

“The work is sure to be an academic standard for years to come . . . I shall heartily
recommend The Routledge Companion to Philosophy of Religion to my students and
colleagues and hope that libraries around the country add it to their collections.”
—Philosophia Christi

The Routledge Companion to Philosophy of Science

A CHOICE Outstanding Academic Title 2008

“With a distinguished list of internationally renowned contributors, an excellent choice

of topics in the field, and well-written, well-edited essays throughout, this compendium
is an excellent resource. Highly recommended.”—CHOICE

“Highly recommended for history of science and philosophy collections.”—Library

“This well conceived companion, which brings together an impressive collection of distin-
guished authors, will be invaluable to novices and experienced readers alike.”—Metascience

The Routledge Companion to Twentieth Century Philosophy

“To describe this volume as ambitious would be a serious understatement . . . full of

scholarly rigor, including detailed notes and bibliographies of interest to professional
philosophers. . . . Summing up: Essential.”—CHOICE

The Routledge Companion to Philosophy and Film

“A fascinating, rich volume offering dazzling insights and incisive commentary on every
page . . . Every serious student of film will want this book . . . Summing Up: Highly

The Routledge Companion to Philosophy of Psychology

“This work should serve as the standard reference for those interested in gaining a
reliable overview of the burgeoning field of philosophical psychology. Summing Up:

The Routledge Companion to Metaphysics

“The Routledge Philosophy Companions series has a deserved reputation for impressive
scope and scholarly value. This volume is no exception . . . Summing Up: Highly

The Routledge Companion to Nineteenth Century Philosophy

A CHOICE Outstanding Academic Title 2010

“This is a crucial resource for advanced undergraduates and faculty of any discipline
who are interested in the 19th-century roots of contemporary philosophical problems.
Summing Up: Essential.”—CHOICE

The Routledge Companion to Ethics

“This fine collection merits a place in every university, college, and high school library
for its invaluable articles covering a very broad range of topics in ethics[.] . . . With its
remarkable clarity of writing and its very highly qualified contributors, this volume
is must reading for anyone interested in the latest developments in these important
areas of thought and practice. Summing Up: Highly recommended.”—CHOICE

The Routledge Companion to Philosophy and Music

“Comprehensive and authoritative . . . readers will discover many excellent articles in

this well-organized addition to a growing interdisciplinary field. Summing Up: Highly
“. . . succeeds well in catching the wide-ranging strands of musical theorising and thinking,
and performance, and an understanding of the various contexts in which all this takes
place.”—Reference Reviews

The Routledge Companion to Phenomenology

“Sebastian Luft and Søren Overgaard, with the help of over sixty contributors, have
captured the excitement of this evolving patchwork named ‘phenomenology.’ The
Routledge Companion to Phenomenology will serve as an invaluable reference volume for
students, teachers, and scholars of phenomenology, as well as an accessible introduc-
tion to phenomenology for philosophers from other specialties or scholars from other
disciplines.”—International Journal of Philosophical Studies

The Routledge Companion to Epistemology

A CHOICE Outstanding Academic Title 2011

“As a series, the Routledge Philosophy Companions has met with near universal acclaim.
The expansive volume not only continues the trend but quite possibly sets a new
standard . . . Indeed, this is a definitive resource that will continue to prove its value for
a long time to come. Summing Up: Essential.”—CHOICE

The Routledge Companion to Philosophy of Language

“This collection presents more than 65 new essays by prominent contemporary figures
working in the philosophy of language. Collectively, they represent the cutting edge
of philosophical research into issues surrounding the use, understanding, and study
of language . . . the book constitutes an invaluable current resource for students and
scholars alike. It will appeal to anyone interested in the current state-of-play within
this important area of philosophical research. Summing Up: Highly recommended.”

The Routledge Companion to Social and Political Philosophy

“This 15th book in the Routledge Philosophy Companions series is also the most
comprehensive, both chronologically and conceptually. . . . The polish and high quality
of the essays provide a multifaceted mirror of the passions and interests of contemporary
academic Anglophone philosophy. Summing Up: Highly recommended.”—CHOICE

The Routledge Companion to Ancient Philosophy

“This excellent reference will be useful to faculty and students alike. The essays are of
uniformly high quality.”—CHOICE

Edited by
Noël Carroll and John Gibson
First published 2016
by Routledge
711 Third Avenue, New York, NY 10017
and by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2016 Taylor & Francis
The right of the editors to be identified as the authors of the editorial material,
and of the authors for their individual chapters, has been asserted in accordance
with sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other means,
now known or hereafter invented, including photocopying and recording,
or in any information storage or retrieval system, without permission
in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification
and explanation without intent to infringe.
Library of Congress Cataloging in Publication Data
The Routledge companion to philosophy of literature / edited by
Nöel Carroll and John Gibson.
pages cm. — (Routledge philosophy companions)
1. Literature—Philosophy. 2. Literature—Aesthetics.
I. Carroll, Nöel, 1947- II. Gibson, John, 1969-
PN45.R636 2015

ISBN: 978-0-415-88972-8 (hbk)

ISBN: 978-1-315-70893-5 (ebk)

Typeset in Goudy Oldstyle Std

by Swales & Willis Ltd, Exeter, Devon, UK

Notes on Contributors xiv

Introduction xxi

Historical Foundations 1

  1 Ancient Beginnings 3

  2 Philosophy of Literature in the Eighteenth Century 13


  3 Philosophy of Literature in the Nineteenth Century 30


  4 Twentieth-Century European Philosophy of Literature 40


  5 Analytic Philosophy of Literature 53


What is Literature?

  6 The Idea of Literature 67


  7 The Novel 83

 8 Poetry 97

  9 Reading Plays as Literature 107


10 Popular Fiction 117


11 Screenplays 127

12 Evolutionary Approaches to Literature 137


13 Canon and Tradition 147



Aesthetics and Appreciation

14 Creativity 163

15 Authorship 173

16 Expression 184

17 Literary Style 195


18 Theme 205

19 Character 217

20 Empathy 234

21 The Paradox of Fiction 247


22 Fiction and Negative Emotions 259


23 Neuroscience and Literature 269


Meaning and Interpretation

24 Narrative 281


25 Narrative Understanding 291


26 Interpretation 302

27 Criticism 313

28 The Poetic Imagination 323


29 Metaphors in Literature 334


PART V 347
Metaphysics and Epistemology

30 The Ontology of Literary Works 349


31 Fiction 359

32 Fictional Truth and Truth through Fiction 372


33 Literary Cognitivism 382


34 Imagination 394

35 The Problem of Imaginative Resistance 405


36 Literature and “Theory of Mind” 419


Ethics and Political Theory

37 Literature and Morality 433


38 Literature and Marxism 451



39 Literature and Race 462


40 Literature and Gender 474


Index 483

To Sam Toperoff
My most unforgettable English teacher
– Noël Carroll

To Aaron Mason,
For all the stories
– John Gibson

Kristin Boyce is Assistant Professor in the Department of Philosophy and Faculty

Fellow in Shackouls Honors College at Mississippi State University. Her primary
research interests are in philosophy of art, history of early analytic philosophy, and
Wittgenstein. Her work has been published in The Journal for Philosophical Research,
The Henry James Review, and the volume Thinking Through Dance (Dance Books, 2013).
She has essays forthcoming in the volumes Wittgenstein and Literary Modernism
(Chicago University Press, forthcoming) and Ethics and the Limits of Sense: Essays on
Wittgenstein and Value (in preparation).

Elisabeth Camp is Associate Professor of Philosophy at Rutgers University, New

Brunswick. Her research focuses on thoughts and utterances that don’t fit standard
propositional models, including figurative speech, slurs and insinuation, animal
cognition, and maps. Her work has appeared in journals including Midwest Studies in
Philosophy, Nous, Philosophy and Phenomenological Research, Philosophical Quarterly,
and Philosophical Studies. She is currently editing a volume on philosophical responses
to the poetry of Emily Dickinson.

Noël Carroll is Distinguished Professor of Philosophy at the Graduate Center, City

University of New York. He is the author of many books and essays in the philosophy
of literature, aesthetics, and the philosophy of film. Recent books include Art in
Three Dimensions (Oxford University Press, 2010), On Criticism, Humor: A Very
Short Introduction (Oxford University Press, 2010), Living in an Art World (Routledge,
2009), and Minerva’s Night Out: Philosophy, Pop Culture, and Moving Pictures
(Wiley-Blackwell, 2013).

Robert Chodat is Associate Professor of English at Boston University. His research

focuses on twentieth-century American fiction and the relations between philosophy
and literature. He is the author of Worldly Acts and Sentient Things: The Persistence of
Agency from Stein to DeLillo (Cornell University Press, 2008). He is currently finish-
ing a book about naturalism, narrative, and moral thought in post-World War II
authors such as Marilynne Robinson, Stanley Cavell, and David Foster Wallace.

Gregory Currie is Professor of Philosophy and Head of Department at the University

of York. He is a Fellow of the Australian Academy of the Humanities and an editor
of Mind & Language. He is the author of six books and numerous essays, including The
Nature of Fiction (Oxford University Press, 1994), Arts and Minds (Oxford University

Prtess, 2004), and Narratives and Narrators (2010). He is currently writing a book
titled This Unintelligible World (unpublished).

E.M. Dadlez has a PhD in Philosophy from Syracuse University and is Professor of
Philosophy at the University of Central Oklahoma. Her work is mainly on the
philosophy of literature, and on topics at the intersection of aesthetics, ethics, and
epistemology. She is the author of various articles on aesthetics and applied ethics, as
well as What’s Hecuba to Him? Fictional Events and Actual Emotions (Penn State Press,
1997) and Mirrors to One Another: Emotion and Value in Jane Austen and David Hume
(Wiley-Blackwell, 2009).

David Davies is Professor of Philosophy at McGill University. He is the author of Art

as Performance (Wiley-Blackwell, 2004), Aesthetics and Literature (Continuum, 2007),
and Philosophy of the Performing Arts (Wiley-Blackwell 2011), editor of The Thin Red
Line (Routledge 2008), and co-editor of Blade Runner (Routledge 2015). He has
published articles on general metaphysical, epistemological, and axiological issues
about the arts, and issues relating specifically to film, photography, literature, music,
theatre, dance, fiction, and the visual arts.

Stephen Davies is a Distinguished Professor of Philosophy at the University of Auckland.

He is the author of many articles. His most recent books are Philosophical Perspectives
on Art (Oxford University Press, 2007), Musical Understandings and Other Essays on
the Philosophy of Music (Oxford University Press, 2011), and The Artful Species
(Oxford University Press, 2012). He is a former President of the American Society
for Aesthetics.

A.W. Eaton is an Associate Professor of Philosophy at the University of Illinois –

Chicago. She works on topics in feminism, aesthetics and philosophy of art, value
theory, and Italian Renaissance painting. She is the editor of the Aesthetics and
Philosophy of Art section of Philosophy Compass. Her work has appeared in Ethics,
Journal of Aesthetics and Art Criticism, Revue Internationale de Philosophie, and Hypatia:
A Journal of Feminist Philosophy, among other publications.

Susan L. Feagin is Visiting Research Professor of Philosophy at Temple University. She

is the author of Reading with Feeling: The Aesthetics of Appreciation (Cornell University
Press, 1996) as well as of numerous articles and book chapters on issues concerning
art and emotions, tragedy, literature, empathy, interpretation, art and knowledge, and
painting and drawing. Her current research involves issues in the philosophy of theater.
She was editor of The Journal of Aesthetics and Art Criticism from 2003 to 2013.

Damien Freeman lectures on ethics and aesthetics at Pembroke College, Cambridge,

England, and the Art Gallery of New South Wales in Sydney, Australia. His recent
publications include Art’s Emotions: Ethics, expression and aesthetic experience
(Acumen, 2012), Roddy’s Folly: R. P. Meagher QC—art lover and lawyer (Connor
Court Publishing, 2012), The Aunt’s Mirrors: Family experience and meaningfulness—a
memoir (Brandl & Schlesinger, 2014), and, with Derek Matravers, an edited collection,
Figuring Out Figurative Art: Contemporary philosophers on contemporary paintings
(Routledge 2014).


Tamar Szabó Gendler is Dean of the Faculty of Arts and Sciences at Yale University,
the Vincent J. Scully Professor of Philosophy, and Professor of Psychology and
Cognitive Science. Her books include Intuition, Imagination and Philosophical
Methodology (Oxford University Press, 2013) and Thought Experiments: On the Powers
and Limits of Imaginary Cases (Routledge, 2000), as well as a number of edited volumes,
including Conceivability and Possibility and Perceptual (Oxford University Press, 2002).

John Gibson is Associate Professor of Philosophy at the University of Louisville. His

research focuses on topics in the philosophy of literature and aesthetics. Among
his publications are Fiction and the Weave of Life (Oxford University Press, 2007) and,
as editor, The Philosophy of Poetry (Oxford University Press, 2015). He is currently working
on a book titled Poetry, Metaphor & Nonsense: An Essay on Meaning.

Jonathan Gilmore teaches philosophy at Baruch College, City University of New York.
He is the author of The Life of a Style: Beginnings and Endings in the Narrative History
of Art (Cornell University Press, 2000) and articles in the philosophy of art, law, and
the imagination. His art criticism appears in Artforum, Art in America, ArtNews, Tema
Celeste, and Modern Painters. He was awarded a 2013–14 National Endowment for
the Humanities Fellowship in support of his current book project on emotions, fictions,
and discontinuities between art and life.

Kristin Gjesdal is Associate Professor of Philosophy at Temple University. Her work

covers the areas of German idealism, phenomenology, and aesthetics. Her book
Gadamer and the Legacy of German Idealism appeared with Cambridge University Press
in 2009. She is the editor of The Oxford Handbook to German Philosophy in the
Nineteenth Century (with Michael Forster) (Oxford University Press, 2015) and Key
Debates in European Nineteenth-Century Philosophy (forthcoming with Routledge).

Paul Guyer is Jonathan Nelson Professor of Humanities and Philosophy at Brown

University. His interests include all areas of the philosophy of Kant, modern philosophy
more generally, and the history of aesthetics. He is the author of the recent
three-volume set A History of Modern Aesthetics (Oxford University Press, 2014) and
nine books on Kant, including Kant’s Groundwork for the Metaphysics of Morals
(Bloomsbury Academic, 2007), Knowledge, Reason, and Taste: Kant’s Response to
Hume (Princeton, 2008). He is also the general co-editor of the Cambridge Edition of
the Works of Immanuel Kant (Cambridge University Press).

Garry L. Hagberg is the James H. Ottaway Professor of Philosophy and Aesthetics at

Bard College. Author of numerous articles, his books include Meaning and
Interpretation: Wittgenstein, Henry James, and Literary Knowledge (Cornell University
Press, 1994); Art as Language: Wittgenstein, Meaning, and Aesthetic Theory (Cornell
University Press, 1995); and Describing Ourselves: Wittgenstein and Autobiographical
Consciousness (Oxford University Press, 2011). He is editor of Art and Ethical Criticism
(Wiley-Blackwell, 2013), co-editor of A Companion to the Philosophy of Literature
(Wiley-Blackwell, 2010), and editor of the journal Philosophy and Literature.

Stephen Halliwell is Professor of Greek and Wardlaw Professor of Classics at the

University of St Andrews, Scotland. He has published widely on Greek literature,


philosophy and culture: his books include Plato Republic 5 (Aris and Phillips Ltd.,
1993), The Aesthetics of Mimesis: Ancient Texts and Modern Problems (Princeton
University Press, 2002), Greek Laughter: a Study of Cultural Psychology from Homer to
Early Christianity (Cambridge University Press, 2008), and Between Ecstasy and Truth:
Interpretations of Greek Poetics from Homer to Longinus (Oxford University Press, 2011).

Espen Hammer is Professor of Philosophy at Temple University, Philadelphia. He is

the author of numerous essays and books, including Stanley Cavell: Skepticism,
Subjectivity, and the Ordinary (Polity, 2002), Adorno and the Political (Routledge,
2005), Philosophy and Temporality from Kant to Critical Theory (Cambridge University
Press, 2011), and Adorno’s Modernism: Art, Experience, and Catastrophe (Cambridge
University Press, 2015). His edited collection German Idealism: Contemporary
Perspectives came out in 2007 (Routledge).

James Harold is Professor of Philosophy at Mount Holyoke College. He has published

widely in aesthetics and metaethics. His work has appeared in journals such as
Philosophical Psychology, Dao, The British Journal of Aesthetics, Midwest Studies in
Philosophy, Philosophy and Literature, The Journal of Aesthetics and Art Criticism, and
American Philosophical Quarterly, among other places.

Wolfgang Huemer teaches philosophy at the University of Parma (Italy). His

research interests are in the philosophy of literature, philosophy of the mind, epis-
temology, and early twentieth-century philosophy in both the analytic and the
phenomenological traditions. In addition to many essays in these areas, he is the
author of The Constitution of Consciousness. A Study in Analytic Phenomenology
(Routledge, 2005), and co-editor of The Literary Wittgenstein (Routledge, 2004),
A Sense of the World. Essays on Fiction, Narrative, and Knowledge (Routledge,
2007), and Wittgenstein Reading (De Gruyter, 2013).

Daniel D. Hutto is Professor of Philosophical Psychology at the University of

Wollongong. His most recent books include: Wittgenstein and the End of Philosophy
(Palgrave, 2006) and Folk Psychological Narratives (Cambridge University Press, 2008).
He is co-author of the award-winning Radicalizing Enactivism (MIT Press, 2013) and
editor of Narrative and Understanding Persons (Cambridge University Press, 2007) and
Narrative and Folk Psychology (Imprint Academic, 2009). He regularly speaks at
conferences and expert meetings for clinical psychiatrists, educationalists, narratologists,
neuroscientists and psychologists.

Eileen John is an Associate Professor of Philosophy at the University of Warwick. Her

publications are primarily in aesthetics, with a special focus on philosophy of literature,
and broad interests in how art and creative practices contribute to cognitive and
ethical development. She has served as the Director of Warwick’s Centre for Research
in Philosophy, Literature and the Arts, and is the co-editor of the Blackwell anthology,
The Philosophy of Literature (Wiley-Blackwell, 2000).

Matthew Kieran is Professor of Philosophy and the Arts at the University of Leeds. He
is the author of a number of books, edited collections, and articles on art, creativity,
literature, and morality. His publications include the co-edited Imagination, Philosophy


and the Arts (Routledge, 2003), and his book Revealing Art (Routledge, 2005). He is
currently writing a book on creativity.

Peter Lamarque is Professor of Philosophy at the University of York, UK. His books
include Truth, Fiction, and Literature, with Stein Haugom Olsen (Oxford University
Press, 1994), Fictional Points of View (Cornell University Press, 1996), The Philosophy
of Literature (Wiley-Blackwell, 2009), Work and Object: Explorations in the Metaphysics
of Art (Oxford University Press, 2010), and The Opacity of Narrative (Rowman and
Littlefield, 2014). He was editor of the British Journal of Aesthetics from 1995 to 2008.

Ernie Lepore is a Board of Governors Professor of Philosophy at Rutgers, New Brunswick.

He is the author of many books and papers on the philosophy of language, philo-
sophical logic, metaphysics, and philosophy of the mind, including, with Matthew
Stone, Convention and Imagination (Oxford University Press, 2014); with Herman
Cappelen, Language Turned on Itself (Oxford University Press, 2007) and Insensitive
Semantics (Blackwell, 2004); and, with Jerry Foder, Compositionality Papers (Oxford
University Press, 2002) and Holism: A Shopper’s Guide (Wiley-Blackwell, 1992).

Shen-yi Liao works on the interface between mind and value. He is an Assistant Professor
of Philosophy at the University of Puget Sound and Marie Curie International
Incoming Fellow at the University of Leeds. His work has been published or is forth-
coming in journals such as The Journal of Philosophy, Philosophy and Phenomenological
Research, and British Journal of Aesthetics.

Paisley Livingston is Chair Professor of Philosophy at Lingnan University in Hong

Kong. He previously held teaching or research positions at McGill University, the
University of Copenhagen, l’Ecole Polytechnique, Paris, and Aarhus University. His
books include Cinema, Philosophy, Bergman: On Film as Philosophy (Oxford University
Press, 2012), and Art and Intention: A Philosophical Study (Oxford University Press,
2012). He has published in a variety of academic journals and reference works.

Aaron Meskin is Associate Professor of Philosophy at the University of Leeds. He is the

author of numerous journal articles and book chapters on aesthetics and other
philosophical subjects. His current work focuses on experimental approaches to
philosophical aesthetics. He co-edited The Art of Comics: A Philosophical Approach
(Wiley-Blackwell, 2012), Aesthetics and the Sciences of Mind (Oxford University
Press, 2014), and Philosophical Aesthetics and the Sciences of Art (Cambridge University
Press, 2014).

Ted Nannicelli is Lecturer in Film and Television Studies at the University of Queens­
land, Australia. His publications include A Philosophy of the Screenplay (Routledge,
2013), Appreciating the Art of Television: A Philosophical Perspective (Routledge,
forthcoming), and, with Paul Taberham, Cognitive Media Theory (Routledge, 2014).

Stein Haugom Olsen is at present Pro-Rector at Østfold University College in Norway.

He has held chairs at the Universities of Oslo and Bergen in Norway and at Lingnan
University, Hong Kong. He has written extensively on problems in the philosophy of
literature and literary aesthetics as well as a number of articles of literary criticism.


He is at the moment working on a book about the origin of literary criticism as an

academic discipline.

Anna Christina Soy Ribeiro is Associate Professor of Philosophy at Texas Tech

University. Her work on the philosophy of poetry has appeared in numerous journals
and various edited collections. She has been awarded fellowships from the National
Humanities Center and the Mellon and Woodrow Wilson Foundations. Ribeiro is
currently working on a book that offers an empirically informed philosophical
account of poetry. She is the editor of The Bloomsbury Companion to Aesthetics
(Bloomsbury, 2012, 2015).

Stephanie Ross is Professor of Philosophy at the University of Missouri-St. Louis. She

has written on a range of topics in the philosophy of art and environmental aesthetics.
She published a book on landscape aesthetics, What Gardens Mean (Chicago
University Press, 1989), and is currently working on a new book project, Two Thumbs
Up: How Critics Aid Appreciation.

M.W. Rowe is Honorary Researcher in Philosophy at the University of East Anglia, UK,
and the author of three books: Philosophy and Literature: A Book of Essays (Ashgate,
2004), Heinrich Wilhelm Ernst: Virtuoso Violinist (Ashgate, 2008), and Philip Larkin:
Art and Self (Palgrave, 2011). He is currently overseeing a complete recording of
Ernst’s compositions on seven CDs for Toccata Classics, and writing a biography of
J.L. Austin.

William P. Seeley is a Lecturer in Philosophy at the University of New Hampshire. He

holds a PhD in philosophy from CUNY-The Graduate Center, an MFA in sculpture
from Columbia University, and a BA in philosophy from Columbia University. His
research interests lie at the intersection of philosophy of art, cognitive science, and
embodied cognition. His welded steel constructions have been exhibited in New
York City and at a number of colleges and university galleries.

Allen Speight is Associate Professor of Philosophy and Chair of the Department of

Philosophy at Boston University. He is the author of Hegel, Literature and the Problem
of Agency and The Philosophy of Hegel (Routledge), co-translator/editor of Hegel:
Heidelberg Writings, and editor of Narrative, Philosophy and Life. He writes on issues in
aesthetics and the philosophy of literature, with a particular focus on figures from
German Idealism and nineteenth-century philosophy.

Matthew Stone is Associate Professor of Computer Science at Rutgers University, New

Brunswick. His research focuses on problems of meaning in human–human and
human–computer conversation. His papers on conversational agents have appeared in
a range of conferences and journals, including ACL, AAAI, SIGGRAPH, and
Cognitive Science. His book with Ernie Lepore, Imagination and Convention: Distinguishing
Grammar and Inference in Language, appeared in 2015 (Oxford University Press).

Paul C. Taylor is Associate Professor of Philosophy and African American Studies

at Pennsylvania State University, and a 2014–15 Edmond J. Safra Network Fellow at
Harvard University. In addition to many essays, he is the author of Race: A Philosophical


Introduction (Polity, 2003) and Black is Beautiful: A Philosophy of Black Aesthetics

(Wiley-Blackwell, forthcoming). He was one of the founding editors of the journal
Critical Philosophy of Race.

Amie L. Thomasson is Professor of Philosophy and Cooper Fellow at the University of

Miami. She is the author of Ontology Made Easy, Ordinary Objects, and Fiction and
Metaphysics (Oxford University Press, 2014), and co-editor (with David W. Smith)
of Phenomenology and Philosophy of Mind (Oxford University Press, 2005). In addition,
she has published more than 50 book chapters and articles on topics in metaphysics,
metaontology, fiction, philosophy of mind and phenomenology, the philosophy of art,
and social ontology.

Mary Bittner Wiseman is the author of The Ecstasies of Roland Barthes (Routledge,
1989) and co-editor of Subversive Strategies in Contemporary Chinese Art (Brill, 2011).
She has written on aesthetics, ethics, feminism, and the philosophies of art and

Noël Carroll & John Gibson

Of all the Muses, the first to receive sustained philosophical attention was poetry, which,
for our purposes, may stand for literature in general. First Plato and then Aristotle
concentrated their penetrating analytic skills especially on tragedy, although with
insights into other precincts of literature as well. Thus was the philosophy of literature
born even if in Plato’s case—with reference to his Ion and his Republic—the operative
preposition here might better be “against,” as in philosophy against literature.
However, in order to be against literature, Plato was obliged to propose a philosophy
of literature, since he had to clarify the nature of what he opposed, while also explaining
why—metaphysically, epistemologically, ethically, and politically—he rejected it. That
is, Plato had to advance an ontology of literature before interrogating its claims to
cognitive, moral, and political value. Plato questioned the value of literature in all of
these domains, but in order to do so, he had to arraign literature before the full court
of philosophical inquiry, including metaphysics, epistemology, philosophical psychology
and the philosophy of mind, ethics and political theory.
Aristotle, of course, pursued this line of inquiry not only more sympathetically but
more systematically and, for that reason, has the best claim to the title of “father of the
philosophy of literature,” at least in the Western philosophical tradition. In many
respects, much philosophy of literature still remains, if not a footnote to Aristotle, then
nevertheless deeply indebted to him. In his Poetics he effectively asks two questions:
what is literature? And why do we value it? With these questions, we have the beginning
of philosophy of literature performed in a charitable key. Literature, for Aristotle,
engages mind and emotion in a distinctive manner—it offers emotion “catharsis” and
mind an insight into general features of human behavior—and understanding how it
manages to accomplish this requires studying the mechanics and formal features of
literature itself. While the philosophy of literature is today animated by a vastly broader
range of issues and concerns, its Aristotelian inheritance is still very much in evidence.
But what is the philosophy of literature. Very roughly, it is the examination of litera-
ture from the viewpoint of the major divisions of the field of philosophy: metaphysics,
the philosophy of mind, epistemology, ethics, political philosophy and, of course,
aesthetics. In addition, the philosophy of literature addresses the various paradoxes,
anomalies, and questions that arise in the course of attempting to systematically work
out an account of the arts of literature in relation to our broader conceptions of human
nature and the world.
In the larger academic world the philosophy of literature is at times confounded with
literary theory (sometimes simply called “Theory”). It is nearly impossible to distinguish
literary theory and the philosophy of literature in a way that is accurate yet does not

deeply insult one or the other field. We will not try to define “literary theory” here,
except to say that it is an immensely inclusive term that brings under its extension
theoretically minded work in English, Comparative Literature, Cultural Studies, and
much else. Philosophy of literature is a narrower field, not restricted to but in practice
often coextensive with the work produced by professors of philosophy who pursue their
interests in literary aesthetics in conversation with debates in the core areas of profes-
sional philosophy. Whatever philosophy of literature precisely designates as a field of
research in the Anglophone academic world, this volume acts as a very good representa-
tive of its concerns and boundaries.
The philosophy of literature contrasts straightforwardly with something else: philosophy
in literature. The latter discourse primarily discusses philosophical themes found in
individual literary works, such as the challenges to utilitarianism presented in Dickens’s
Hard Times and Aldous Huxley’s Brave New World or, taking “philosophy” in a more
expansive sense, the critique of modern culture found in William Wordsworth’s “The
World is Too Much With Us” or T.S. Eliot’s The Waste Land. The presentation of
philosophical themes in literary works may range from illustrating said themes to
attempts to advance original philosophical theses by means of a literary text. But, with
the exception of literary works that endeavor reflexively to raise philosophical questions
about literature by means of literature, the study of philosophy in literature differs from
philosophy of literature in that the former typically takes as its object a particular work,
or genre or author, while the philosophy of literature takes all literature or great swathes
of it in its purview, examining it in terms of the animating questions of metaphysics,
epistemology, philosophical psychology, ethics, political theory, and aesthetics.
Virtually all of the entries in this volume are exercises in the philosophy of literature,
though readers should expect the odd foray into philosophy in literature, too. Thus,
there are chapters not only on the ontology of literature but also on the nature of the
novel, poetry, narrative, authorship, style, fiction, and metaphor. Epistemological issues
in relation to literature are examined in entries on fictional truth and literary cognitivism.
Literature is viewed from the perspective of the philosophy of mind and philosophical
psychology in contributions on the imagination, literature and the theory of mind,
neuroscience and literature, literature and evolution, the paradox of fiction, the paradox
of negative emotion, empathy, and the problem of imaginative resistance. Ethics is not
only given an entry of its own, but is also addressed in the chapters on literary cognitivism
and imaginative resistance. Literature meets politics in an essay on Marxist critique and
criticism as well as in essays on literature and race and literature and gender. There is,
naturally, overlap, and many of the fundamental issues concerning fiction, imagination,
and the relation between the ethical, aesthetic, and cognitive values of literature make
numerous appearances in the following chapters. This repetition is of the productive and
healthy sort, and it demonstrates the extent to which work in the philosophy of litera-
ture is interlocking and mutually informed.
Also covered are such general aesthetical topics as literature and expression, interpre-
tation, criticism, style, and canon, while issues in the aesthetics of literature in
particular are dealt with in essays on character, theme, literary adaptations and screen-
plays, theatrical literature, and popular fiction. Of course, discussions of literature occur
in specific contexts, and in order to supply an introductory grasp of some the leading
contexts for the Western conversation of the philosophy of literature, we have included
articles on the philosophy of literature in the ancient world, in the eighteenth, nineteenth,
and twentieth centuries, as well as an overview of analytic aesthetics. (If one wonders


why there is no chapter on literary aesthetics between antiquity and the modern era, it
is because there is virtually no contemporary philosophical research on this. Graduate
students interested in the history of aesthetics should take note when devising dissertation
topics.) Needless to say, more topics could be added to our table of contents, though the
fact that forty chapters were required to cover the essentials shows that contemporary
philosophy of literature is flourishing indeed. Nor is our exclusion of articles on
philosophy in literature intended in any way to demean that enterprise. There is much
more to the relation between philosophy and literature than what we are able to canvass
here, and it is heartening to know that there are still unexplored frontiers. It is our hope
that we have presented enough of the field to inspire readers to discover the rest and to
chart out new territories in their own research.
We made a few felicitous discoveries in the process of putting this volume together.
The most difficult task editors face is that of selecting, from all the available talent, the
leaders in the field. We suspect that even twenty years ago, forty chapters would have
sufficed to include almost all the brightest lights in the philosophy of literature. Though
it made our job unenviable, we were often struck by how many excellent philosophers
of literature we simply did not have the space to invite. For many topics—on narrative,
fiction, or imagination, for example—the list of figures producing cutting-edge work was
at times more than ten-deep. We were also struck by the number of first-rate philoso-
phers working in core areas of philosophy and their histories who have taken an interest
in the philosophy of literature. Since Aristotle, philosophers of literature have known
that our field raises important general issues for the philosophy of language, metaphysics,
epistemology, and the philosophy of mind. It is good to know that many luminaries in
these fields now know this, too. Lastly, we were surprised to see just how much competi-
tion we faced from other handbooks, companions, and surveys of the philosophy of
literature; there is clearly a sizable market for books on the topic. This placed considerable
pressure on us to raise our game, and we hope the result is a volume that anyone with
an interest in the philosophy of literature will find comprehensive, au courant, and essential.
Our primary goal was to produce a companion to the philosophy of literature that will
be attractive to both researchers and university instructors (for use in graduate and
upper-level undergraduate courses). We asked contributors to strike a balance between
survey and originality, and to write in such a way that readers new to the field will
acquire a sense not only of the showcase debates, but also of how one might go about
making innovative contributions to them. In a volume of this sort, originality typically
takes the form of at least outlining novel positions and suggesting new lines of inquiry,
and we think the chapters here promise to do an excellent job of guiding and inviting
new work in the field.
We are grateful to the many people who have helped to bring this volume to the light
of day. At Routledge Tony Bruce, Andrew Beck, and Laura Brickman offered much help
along the way. Two of our graduate students did an extraordinary amount of work both
proofreading the manuscript as well as preparing it for submission. At the University of
Louisville, Nick Curry (now at the University of Illinois-Chicago) formatted and
proofed thirty-odd chapters, and at the CUNY Graduate Center, Chloé Cooper Jones
brought the remaining ten to perfection and took care of all of the notes on contribu-
tors. We are grateful to both of them.

Part I

Stephen Halliwell

Was the philosophy of literature an invention of classical antiquity? Since all the major
phases and movements of Greco-Roman philosophy, from presocratic origins to the
Neoplatonism of late antiquity, saw reason to engage with practices we now count as
“literature”, it is hard to avoid some kind of affirmative answer to that question. But the
answer needs modifying in the light of historical and conceptual complexities. Antiquity
brought into being not one, but several (potential) philosophies of literature, and it did
so with results which were, and remain, intrinsically contentious.
While the culture of Greece and Rome lacked a category corresponding straight-
forwardly to “literature” as we now claim to know it, ancient philosophy often defined
and sustained itself in part by dialectic with (and sometimes in opposition to) forms
of discourse, thought, and sensibility which have subsequently helped to shape
conceptions of the literary. This is true above all in relation to poetry (including
drama), which held a privileged place in ancient education and society and accordingly
provided a vital reference-point for philosophy’s own aspirations to a position of cultural
authority. In addition, the study and evaluation of poetry in antiquity became associated
with wider interests in the workings of language. This gave rise to modes of analysis –
among them stylistics, rhetoric, hermeneutics, and criticism – in which prose texts too,
notably oratory and history (but also philosophy itself), were taken into consideration.
Such analysis (already visible in Aristotle’s Poetics 20­–22 and Rhetoric 3.1–12) was
generally known as grammatikê in Greek and came to be called lit(t)eratura in
Latin. Thus antiquity possessed, in essence, two overlapping paradigms of “literature”:
one, poetry in all its generic varieties (narrative, dramatic, lyric); the other, a domain
marked out by special, heightened uses of language (thus a kind of “literariness,” in
modern terms). Philosophy of literature emerges from shifting reactions to these
two paradigms.
The most obvious gap between ancient and modern thinking in this area concerns
prose fiction. Although antiquity produced a tradition of prose novels/romances, this
left virtually no trace in philosophical or critical theory. The significance of this gap,
however, is debatable, since many ancient theories encompassed questions of (for
example) representation, style, audience psychology, and even fiction (Halliwell 2015)
which are equally applicable to prose works. In Plato’s Republic, when Socrates proposes
Stephen Halliwell

a proto-narratological schema of the psychological and ethical implications of “diegetic”

form and viewpoint, he indicates in passing that his criteria are applicable to narrative
discourse in general (Republic 3.397c, Plato 1997). Aristotle, indeed, taking mimesis
rather than verse-form to be a necessary condition of poetry, was prepared to include
some prose genres within the latter category (Poetics 1.1447b9–11, Aristotle 1995). The
conceptual boundaries of ancient philosophies of literature are therefore far from fixed
and the relevant terrain is extensive. The following sections will focus on just a selection
of the most prominent landmarks.

In a famous but perplexing passage of Plato’s Apology (22a–c) Socrates recounts how, in
his search for someone wiser than himself, he interrogated a number of poets, among
them tragic playwrights. Before approaching them, he “took up” (written copies of)
works he thought particularly well crafted. He makes, then, a prior judgement of poetic
merit. But he does not declare his criteria. Nor does he spell out why he needed to inter-
rogate the poets in person: what is he looking for which the poetry itself cannot supply?
He proceeds, in fact, to invite them to explain or interpret their own poems, asking
them “what they meant” but finding them unable to do so. The context shows that he
is not seeking domain-specific knowledge (e.g. mastery of poetic language or metres) but
a general, life-guiding wisdom of the kind he had earlier called knowledge of “human
and political excellence” (ibid.: 20b) and will later equate with knowledge of “the most
important things” (ibid.: 22d).
In inquiring what the poets, or their works, meant, Socrates follows the same
principle as with the Delphic oracle, which had stated that no one was wiser than him.
“Whatever does the god mean?”, he had asked himself insistently (ibid.: 21b, 22a). Yet
the semantic surface of the oracle is not obscure; Socrates pursues meaning at a (puta-
tively) deeper level. Likewise, it seems, he wants the poets to disclose some larger import
of their works that he cannot find for himself. In the oracle’s case, he assumed that the
meaning might be cryptic, like a “riddle”. Does he make the same assumption about
poetry? Certainly he does, perhaps ironically, at Republic 1.332b, when suggesting that
Simonides might have been “speaking in riddles” about the nature of justice. But how
could the hermeneutics of decoding a riddle furnish an adequate model for the task of
establishing what, for instance, a whole tragedy might “mean”?
In the case of the poets, unlike Apollo’s oracle, it needs stressing that Socrates is
doing something subtly different from attempting to get behind the text to a separately
identifiable authorial intention. He does not ask the poets to specify such an intention,
but to undertake an act of interpretation on the words of the poems themselves. If it
were the poets’ (separately identifiable) intentions at issue, it would not make sense for
Socrates to claim that “almost all those present could have spoken better than the
poets themselves about their compositions” (ibid.: 22b). In a piquant historical irony,
Socrates prefigures the Schleiermachian principle of understanding an author better
than he can understand himself (Schleiermacher 1998: 23, 33). But whereas
Schleiermacher’s hermeneutics ties this to the psychological difference between what
remains unconscious in the author and what interpretation of his text makes explicit,
Socrates contemptuously counts it as a symptom of something badly lacking on the
poets’ part.


Socrates’ disappointment with the poets leads him to infer that they

do not compose by expert knowledge but by some kind of natural talent and in
an inspired state of mind, just like seers and oracle-reciters, for these too say
many beautiful things but possess no knowledge about what they say.
(Apology 22c)

The hypothesis of “inspiration” supposedly explains what is inaccessible to discursive

reason. But this leaves us with the puzzle of how Socrates himself finds value (“many
beautiful things”) in the poets’ works. Moreover, later in the Apology Socrates will have
no trouble adopting the role of poetic interpreter when using the Iliadic Achilles as a
paragon of virtuous disdain for death (ibid.: 28b–d) and in the process supplementing
Homer’s discourse with his own (by appealing to a vocabulary of “justice” that is absent
from Achilles’ words in the relevant passage). At that later stage, Socrates moves con-
fidently from inside to outside the text, connecting the psychological drama of the poem
with a general conception of how to live (ibid.: 28d). It transpires there that poetic
interpretation cannot be neutral exegesis. It is a form of ethical dialectic with the text –
and with other interpreters. Socrates’ philosophical premise appears to be that every
poem is incomplete without the supplementary discourse of ethical interpretation.
Interpretation assimilates the poem into a larger framework of life-values applicable both
to the characters in the text and to the reader. Poetry, on this view, needs philosophy.
The poets themselves, of course, would have given a very different account of their
encounters with Socrates. Rejecting what Nietzsche, in The Birth of Tragedy, called
“aesthetic Socraticism” – “everything must be conscious in order to be beautiful”
(Nietzsche 1999: 64) – the poets might have protested that their work is complete
without philosophical interpretation. As Tolstoy put it, subconsciously inverting a
Socratic stance: “An artist’s work cannot be interpreted. Had it been possible to explain
in words what he wished to convey, the artist would have expressed himself in words.
He expressed it by his art” (Tolstoy 1930: 194).
Socrates’ interrogation of the poets voices dissatisfaction with the discursively non-
transparent nature of their works, but it also betrays a residual attraction to the beauties
of these same works. In that regard it is emblematic of the ambivalence which colours
Plato’s lifelong obsession with poetry. Plato’s relationship to poetry is complex; it should
not be reduced to the textbook clichés of unqualified hostility. This is true even of Book
10 of the Republic, where Socrates notoriously speaks of an “ancient quarrel between
philosophy and poetry” (ibid.: 607b). He does so, it is too rarely noticed, to defend
himself against the possible accusation of insensitivity to poetry’s intense allure; and he
proceeds to declare that “we would happily welcome poetry back” into the city (and
the soul) if “lovers of poetry”, of whom he confesses he was once one himself, can
harmonise the pleasure of poetry with the good of both individual lives and the community
as a whole (Republic 10.607c–e). This is not implacable hostility, but conflicted uncer-
tainty translated into a challenge to the Republic’s own readers. Socrates demands that
poetry’s distinctively imaginative and emotional powers of expression, capable as they
are of resisting rational control, be held to account to philosophical standards of
meaning and value. Yet Plato’s dialogues never purport to supply a definitive answer to
the question of whether or how far that is achievable. In that sense, we might say, Plato
conceives of the “philosophy of literature” as a permanently unsolved problem.

Stephen Halliwell

At a late stage in Milan Kundera’s novel Immortality, the semi-autobiographical narrator
pauses to reflect on Aristotle’s conception of narrative unity in the Poetics (Kundera
1991: 338–40). “Aristotle did not like episodes”, he states (in something of a simplification).
“According to him, an episode . . . is outside the causal chain of events which is the
story.” Yet “Life is as stuffed with episodes,” he remarks, “as a mattress is with horsehair.”
Shortly afterwards, the narrator concedes that while every episode has the potential to
turn into a story, only some of them fulfil that potential. So is he, or is he not, disagreeing
with Aristotle?
Aristotle’s Poetics is the single most widely cited work in the history of the philosophy
of literature. It has been the object of continuous debate – along a spectrum running from
canonisation, via contestation, to outright opposition – since its “rediscovery” by
sixteenth-century Italian literary theorists. As my example from Kundera illustrates, it is
hard now to argue with Aristotle without arguing with oneself about assumptions deeply
embedded in views of literature. The Poetics’ compressed arguments contain suggestive
observations on such fundamental topics as genre, form, plot, character, emotional
expression, audience psychology (including the notoriously elusive concept of katharsis),
and metaphor (Rorty 1992). Though partly a response to Platonic promptings, the treatise
stays conspicuously silent about Plato’s writings as such – except, ironically (and
provocatively), to classify Socratic dialogues like Plato’s as themselves “poetic” in virtue
of their mimetic, i.e. dramatic-cum-fictive, status (Poetics 1.1447b9–11).
Aristotle does not provide any easy answer to the question of how philosophy should
construe its relations to poetry/literature – or vice versa. Indeed, at a certain level he
displays his own ambivalence on the point. In the first chapter of the Poetics he is
anxious to separate poetry and philosophy: Homer and Empedocles (who had written in
verse), he insists, “have nothing in common except their metre; so one should call the
former a poet, the other a natural scientist” (ibid.: 1.1447b17–20). Yet in chapter 4 he
explains the pleasure derived from poetry (as well as from visual and other art-forms) as
cognitive and quasi-philosophical: “understanding gives great pleasure not only to
philosophers but likewise to others too” (ibid.: 4.1448b13–14). And in chapter 9 he
famously pronounces poetry “more philosophical” than history on the grounds that “it
speaks more of universals, history more of particulars” (ibid.: 9.1451b5–7). If Aristotle
anticipates certain modern views of literature’s own philosophical capacities, he does so
equivocally. Taking chapters 1 and 9 together, it looks as though mimesis is contrasted
with the making of overt truth claims but can nonetheless be a medium for serious
discernment of some of the underlying structures of human experience.
Everything in the Poetics is underpinned by the master-concept of mimesis, the
narrative/dramatic projection of a hypothetical world (Halliwell 2002). Aristotle
happily grants mimesis a wide, flexible scope, capable of moving between three basic
perspectives: known states of affairs; the whole realm of culturally shared beliefs and
suppositions (including mythology); and the ideal (ibid.: 25.1460b9–11). Even the first
of those perspectives requires imaginative re-making of the world: if a poet draws on
historical events, he must still convert and shape them into the substance of a new
artefact that will be poetry not history (ibid.: 9.1451b29–32). Aristotle’s account of
mimetic “making” accordingly implies, in some degree, the operations of fiction.
Most interpreters maintain that the Poetics ascribes to poetry its own intrinsic, even
formalist, principles of value. But does Aristotle claim pure autonomy for poetry? In


chapter 25, which anatomises (il)legitimate reasons for criticism of poetry, Aristotle
asserts that “poetry does not have the same standard of correctness as politics, or as any
other art” (ibid.: 1460b13–15). But “not . . . the same” need not mean entirely discon-
nected. Later in this chapter Aristotle declares:

when the issue is whether or not someone [i.e. in poetry] has spoken or acted
well, one should examine not only whether the actual deed or utterance is good
or bad, but also the identity of the agent or speaker, to whom he acted or spoke,
when, with what means, and for what end.
(ibid.: 25.1461a4–9)

Patently this is some sort of contextualism. But might it be a partly ethical contextualism?
Comparable formulations do in fact occur in Aristotle’s ethical writings.
Admittedly, that Poetics passage must be stressing more than the context-sensitive
nature of ethical judgement in general. If it were not, it would fail to satisfy the need for
distinctive principles of poetic “correctness”. Yet it does suggest that judgement of poetry
cannot simply dispense with ethical considerations; rather, it will harmonise them with
appreciation of the individual work’s design and dynamics. Hence Aristotle’s point that
depiction of moral evil in a poem is unjustified if it contributes nothing to the achieve-
ment of the poem’s goal (ibid.: 25.1461b19–20). That goal is the pleasure of a certain
kind of intensely emotional response. But we have seen that this pleasure is also
cognitively grounded. Aristotle’s psychology indeed takes emotions to be informed by
“judgements”. The experience of poetry entails, therefore, a kind of emotionally
inflected understanding which cannot be dissociated from ethical value. The Poetics, I
submit, seeks an “ethicist” mean, avoiding the extremes both of moralism (which would
identify poetry’s “standard of correctness” with the ethical tout court) and of outright
poetic formalism (which would assert a pure autonomy severed from the ethical).
The terseness of the Poetics leaves much to unpack. One example of this, relevant to
my quotation from Kundera at the start of this section, is a tension between chapter 6’s
description of tragedy (but not tragedy alone) as “mimesis of life” (ibid.: 1450a16–17)
and the emphatic statement that unity of plot – with the causal-cum-logical stringency
and non-redundancy of the beginning-middle-end formula – is sharply distinct from
the fragmented, untidy contingency of life itself (ibid.: 1451a16–22). Aristotle takes
narrative/dramatic poetry to be a representation of “life”, yet to differ from life in some
of its essential conditions. Unity of plot presupposes the causal intelligibility of life, yet
involves a degree of intelligibility which life itself lacks. This paradox, not unique to the
Poetics, points to a process of mimetic concentration and transformation. But it is a
paradox that cannot be fully resolved, only negotiated delicately: that is a task Aristotle
has bequeathed to the philosophy of literature.

The history of Epicureanism, as perhaps befits a philosophy suspended between a physics
of cold materialism and an ethics of tightly circumscribed pleasure, threw up an unstable
mixture of attitudes to poetry (Asmis 1995). Epicurus himself called poetry a “lethal
ensnarement by myths/stories”, evidently thinking of it as a vehicle of much of the
inherited Greek conglomerate of beliefs (about the gods, death, and so forth) which his

Stephen Halliwell

own philosophy sought to dismantle. He urged his followers to have little to do with
poetry and pronounced that the wise man would not compose it. He reserved particular
contempt for the study of poetry that was at the core of a “liberal” Greek education and
was pursued at a higher level by scholar-critics of various kinds. In a deliberately shocking
piece of ostentatious philistinism, Metrodorus, one of Epicurus’ closest associates, out-
did the master by remarking that it would not matter if one did not know even the
opening lines of the Iliad and Odyssey. Epicureanism, it seems, makes available an
anti-philosophy of literature.
Yet this same philosophical school produced both poets and poetic theorists of its own.
The former include one of the supreme figures of Latin literature, Lucretius (c.95–55 BC).
This is not the place to rehearse the significance of De rerum natura, but two brief points
are worth making. First, the form of Lucretius’ didactic poem reaches back beyond
Epicureanism itself to the earliest traditions of Greek philosophy, in which several major
thinkers – Xenophanes, Empedocles, Parmenides – wrote in verse (though Aristotle, we
have seen, denied that this made their work poetry). Second, Lucretius diverges from
the ostensible spirit of Epicurus’ own “poetics” but need not contradict its fundamental
presuppositions. Epicurus regarded poetry as capable of expressing values in a way that
could impinge (negatively) on the lives of its audiences. Under the pressure of an ethical
psychology averse to “unnecessary” pleasures (especially those of elaborate, variegated
kinds), he simply did not conceive of the possibility that someone might tap this power
in the positive service of philosophy.
On the theoretical or critical side, developments in Epicurean thinking about
poetry now have to be painstakingly reconstructed above all from the badly damaged
Herculaneum papyri of Philodemus’ On Poems (Janko 2000, 2011; basic overview
in Gutzwiller 2010: 360–4). Philodemus of Gadara (c.110–40 BC), who moved in
high cultural circles at Rome (his patron was Julius Caesar’s father-in-law and he
knew the poets Vergil and Horace), was a scathing polemicist. On Poems, in five
books, unpicked the theories of numerous non-Epicurean thinkers, frequently pour-
ing scorn on their views.
That typically Epicurean penchant for belittling the beliefs of others leaves
Philodemus’ own position somewhat elusive. He denies that there is any “natural”
value in poetry and is particularly harsh on the views of euphonist critics who tried to
reduce poetic excellence to a quasi-musical beauty of sound and therefore to a matter
of purely sensory pleasure. Philodemus takes good poetry to engage, indeed enthrall,
the mind by a fusion of language and thought: he even seems to hold the subtle view
that poetic “thought” is inseparable from its verbal expression. But he is not prepared
to allow poetry, as many Greeks did, any cognitively or ethically serious value: its aim
is not truth or the enrichment of life-values, only some kind of mentally satisfying
manipulation of expressive language in the delineation of a mimetically coherent
world. There is even a hint of something like a concept of literary mimesis as consti-
tuted by fictive speech-acts: that is at any rate one possible interpretation of his
principle that poetry’s language is itself mimetic of, and should “resemble”, language
that communicates in an instructive manner.
Philodemus’ philosophical poetics insists, against euphonists and other “formalists”,
on poetry’s possession of the full signifying powers of language. Yet he seems to want
largely to disconnect those powers from an expressiveness whose function might extend
beyond the self-contained experience of the poetic moment. Curiously, Philodemus
follows in Epicurus’ footsteps by recognising some poetry as harmful in its embodiment


of damagingly false beliefs about the world while showing reluctance to subscribe to a
matchingly positive counterpart of beneficial poetry. Like Epicurus himself, Philodemus
did not envisage a full-blown Epicurean poet like Lucretius. Unlike Epicurus, Philodemus
could actually have known the De rerum natura (composed during his own lifetime).
Exactly how he would have made sense of it remains for conjecture. Epicureanism
generated an internal tension between philosophy and literature, which gives it a
uniquely ambiguous place in the philosophy of literature.

In the eighth of his Moral Letters, written around 62–65 AD, the Roman Stoic Seneca
(the younger) exclaims at one point: “how many things poets say which philosophers
themselves either have said or should say!” (Seneca 1917–25). This comes from a
passage where Seneca is justifying a habit of reinforcing his convictions with quotations
from non-Stoic sources (including Epicurus, whose sayings he treats not as personal
utterances but as “common property”). He goes on to explain that it is not just serious
poetic genres like tragedy from which edification can be derived, but even comic forms
like mime (a kind of light popular drama): he quotes, among other things, an aphoristic
line from one mime (“Nothing that comes from wishing belongs to us”), which he takes
to chime with the Stoic tenet that virtue is the only truly valuable possession and every-
thing else is “indifferent”.
Seneca aligns philosophy and poetry in terms of their capacity to give expression to
true conceptions of reality. What he has in mind in the first instance is sententious or
gnomic wisdom encapsulated in neatly turned apophthegms of a sort which were indeed
a shared resource of ancient philosophy and literature. Seneca’s thesis is Janus-faced: he
ascribes (partial) philosophical credentials to poetry but at the same time perceives
philosophy as a type of understanding not confined to the pronouncements of the
recognised philosophical schools.
If we probe a little further, that gerundive (things philosophers “should say”) becomes
intriguing. Is there a suggestion here that poetry has independently contributed to the
achievement of philosophical insight? Seneca appears to imply that one can find
philosophical thoughts in poetry which philosophers themselves have not so far
produced, or not in quite the same form. Philosophy itself could learn, he seems to say,
from the range and nature of poetic expression.
That last inference is made more plausible by a passage from another letter
(ibid.: 108.10) where Seneca endorses the view of Cleanthes (third century BC, the
second head of the Stoa) that poetic form, including the constraints of metre, could lend
intensified force to the expression of ideas. Seneca speaks there in terms of the “looseness”
of prose and the “tightness” or compression of poetic form; he urges philosophers to
emulate, as well as borrow from, the latter. Although neither Cleanthes nor Seneca
elaborate the point very far, they display a clear recognition, all the more telling for
belonging to a Stoic perspective, that there may be distinctively poetic/literary qualities
of expressive form which in turn affect the communication of thought.
That recognition, along with the notion that poetry can itself find room for philo-
sophical ideas, potentially complicates the scope of both philosophy and “literature”. It
is no accident that Cleanthes himself wrote poetry, including a Hymn to Zeus which
expresses Stoic ideas of cosmic order and providence. Seneca too was a poet, though in
a very different vein (as author of some nine tragedies). His letters themselves are a

Stephen Halliwell

philosophico-literary hybrid: they quote poetry frequently, weaving it into their own
verbal texture; they also use an epistolary voice to create their own peculiarly eloquent
and intimate style of meditation (hence their later appeal to Montaigne and others).
But there is an important caveat to add. The letters define a project caught somewhere
between compelling statements of philosophical wisdom and, on the other hand, the
conviction that philosophy can find true fulfilment not in words at all but only in a
certain way of living.
Seneca addresses that point directly in letter 33, rejecting individual maxims as too
commonplace and too superficial (like eye-catching goods in a shop, he says) to capture
the real substance of Stoic philosophy. Such maxims, he points out, are available outside
philosophy in such writings as poetry and history; but here he uses this point to play
down, not up, the value of literature. Equally interesting, however, is that in developing
his case in letter 33, Seneca ascribes to Stoic philosophy itself properties akin to those
of a unified literary or artistic work: it has a continuous fabric, he indicates, whose
strands are all bound together inseparably and from which no element can be removed
without destroying the whole. So in the very act of stressing the status of philosophy as
a work of “life” not words (and Stoics traditionally called virtue “the art of life”, ars
vitae), Seneca appeals to a conception and value of unity which was primarily familiar
from the principles of literary and artistic criticism. That opens up the possibility,
glimpsed rather than elaborated in Seneca’s terms of reference, that poetry/literature
might have something special of its own, and something more than well-turned apho-
risms, to offer philosophy.
Though Seneca is far from a seamlessly orthodox Stoic thinker, he is here in line with
mainstream Stoicism, which was always fundamentally hospitable to the value of poetry
(De Lacy 1948). Many of the major Greek Stoics, not least Chrysippus (third century
BC), are known to have cited poetry abundantly in their works and to have tried strenu-
ously, where necessary by employing types of allegorical interpretation (Gutzwiller 2010:
354–9), to find anticipations and/or corroborations of Stoicism itself (e.g. its ideas of
cosmic providence or of the nature of virtue) in Greek literature all the way back to
Homer. The larger story of Stoicism’s attitudes to poetry inevitably prompts the question
whether philosophy can go this far in seeing itself reflected in literature without distort-
ing the latter by imposing heavily doctrinaire demands on it. By attributing to poetry
resources of language and thought capable of expressing even philosophical ideas with
distinctive cogency, Stoicism was at any rate paying it a sort of compliment. But it was a
compliment that had to be hedged round. Stoicism could never, for instance, be easily
reconciled to varieties of literary experience which gave a central role to strong emotional
responsiveness (Nussbaum 1993). Thus Stoic reactions to tragedy, when not simply
hostile, involved radical revisionism about the genre’s justification. As a totalising
worldview, Stoicism adumbrated a conception of literature which respected its powers of
imagination and expression but could not afford to grant it a raison d’être of its own.

Plutarch’s How the Young Man Should Study Poetry (Plutarch 1927, Hunter and Russell 2011),
part of the collection of essays known as the Moralia (whose general influence in the
early modern period was immense), is an instructive specimen of some of the tensions
which can arise from attempts to apply a philosophical agenda to the interpretation of


literature. Plutarch (c. AD 45–125), an eclectic Platonist whose views also incorpo-
rate elements of Aristotelianism and Stoicism, expounds a set of reading habits to be
inculcated in the young as preparation for philosophy. But very importantly he
acknowledges, as few ancient philosophers did, that there is more than one kind of
reader of poetry. He distinguishes three main classes (Moralia 30c–e): the “story-lover”
(who reads for narrative and dramatic interest), the “language-lover” (who reads for
nuances of style and rhetoric), and the “lover of the noble/beautiful”, Plutarch’s own
desired type, who reads for philosophical enlightenment. This last reader needs to
acquire critical equipment with which to minimise the risk of suffering harm from too
much absorption in poetry’s pleasures and to maximise the possibility of extracting
valuable insights from literary works.
Plutarch’s stance rests on a pair of notionally complementary premises stated at the
outset (ibid.: 15e–f): one, that philosophy should be “mixed” into the reading of poetry;
the other, that (the best) poetry itself contains elements of philosophy which it has assim-
ilated into its own fabric. Plutarch is open to more than one way in which literature can
convey philosophical content. In addition to supposedly direct authorial judgements
(though, like many ancients, he does not consistently distinguish between narrator/per-
sona and author), he allows for ethical lessons which can be learnt “from the events
themselves” (ibid.: 19e). He refers to the latter as “silent” teaching which the reader,
through reflective contemplation (Plutarch’s whole model of reading entails self-
conscious processing of the text), can convert into useful ethical “examples” or
“paradigms”. But it is clear that Plutarch feels the need to bridge a gap between philosophy
and literature. He contrasts the status of “examples” in the two kinds of discourse: philosophers
use ready-made ones (from life?), he thinks, and they employ them in an explicitly didac-
tic manner, whereas poets not only operate through fictional invention (the Greek verb
plattein, “mould” or “fashion”, is common in this sense) but elaborate their materials into
forms of “story” (muthos) which carry with them a world of their own.
Plutarch is concerned, as he puts it in his own somewhat literary trope, to find
philosophical “fruit” hidden amidst the lush textual foliage of poetry (ibid.: 28d–e).
But one difficulty with his position is that its results may be self-generating: the
strategy of reading itself brings into being what it purports to discover. The trope in
question obscures this point. Plutarch is constantly aware – and this is what makes his
views worth attention, beneath their prima facie naivety – that the texts he deals with
have their own complex dynamics. His treatise refers to qualities of compositional
intricacy, multiplicity, expressive intensity, narrative unpredictability, emotional
impact, beauty, and various other features that Plutarch groups together under two
broad, conventional headings of words and subject-matter, style and story. He is alert,
then, to the manifold strands which can combine to form the value of a literary text.
But he is nonetheless committed to isolating individual strands at the expense of the
larger fabric.
One particularly revealing remark in this regard is Plutarch’s statement that “to connect
and harmonise poems in this way with philosophical teachings takes them out of their
narrative and dramatic sphere and adds seriousness to the useful things they say”
(ibid.: 36d). Plutarch comes perilously close here to admitting that the price of har-
monising poetry with philosophy in the manner he proposes is at least partly to alienate
poetry from itself. The issue at stake here is one which remains a challenge for the
philosophy of literature.

Stephen Halliwell

Restrictions of space prohibit further extension of the historical scope of this chapter. It
should not go unmentioned, however, that in the last great phase of ancient philosophy
the Neoplatonists of the third to sixth centuries AD attempted – in a strange compound
of sophistication and ideological wilfulness – to reconcile philosophical and poetic
ideals by new forms of allegorical and symbolist interpretation. Those forms encom-
passed, among much else, the possibility of seeing plural levels of meaning in a narrative/
dramatic text and also conceiving such a text as itself a sort of hierarchical “world”,
kosmos (Russell 1989).
The problems foregrounded in the preceding sections might be thought of as
variations on a central question. Ancient philosophy produced a range of approaches to
classes of work we now count as literary. But was it more concerned to use literature as
a mirror to itself or to open itself up to objects and kinds of value that differed from its
own capacities? That question can be posed with specific force in each of the cases I
have considered. But its ramifications are of wider and more lasting consequence for any
enterprise that might call itself a philosophy of literature.

Aristotle (1995) Poetics, trans. S. Halliwell, Loeb Library, Cambridge, MA: Harvard University Press.
Asmis, E. (1995) “Epicurean Poetics”, in D. Obbink (ed.), Philodemus and Poetry, New York: Oxford
University Press.
De Lacy, P. (1948) “Stoic Views of Poetry”, American Journal of Philology 69: 241–71.
Gutzwiller, K. J. (2010) “Literary Criticism”, in J. J. Clauss and M. Cuypers (eds), A Companion to Hellenistic
Literature, Malden MA: Wiley-Blackwell.
Halliwell, S. (2002) The Aesthetics of Mimesis: Ancient Texts and Modern Problems, Princeton: Princeton
University Press.
Halliwell, S. (2015) “Fiction”, in P. Destrée and P. Murray (eds), A Companion to Ancient Aesthetics, Malden
MA: Wiley-Blackwell.
Hunter, R. and Russell, D. (2011) Plutarch: How to Study Poetry, Cambridge: Cambridge University Press.
Janko, R. (2000) Philodemus On Poems Book 1, Oxford: Oxford University Press.
Janko, R. (2011) Philodemus On Poems Books 3–4, Oxford: Oxford University Press.
Kundera, M. (1991) Immortality, Eng. trans. P. Kussi, London: Faber and Faber.
Nietzsche, F. (1999) The Birth of Tragedy, Eng. trans. R. Speirs, Cambridge: Cambridge University Press.
Nussbaum, M. C. (1993) “Poetry and the Passions: Two Stoic Views”, in J. Brunschwig and M. C. Nussbaum
(eds), Passions and Perceptions, Cambridge: Cambridge University Press.
Plato (1997) Complete Works, ed. J. M. Cooper, Indianapolis/Cambridge: Hackett Publishing Company.
Plutarch (1927) Moralia, vol. 1, Eng. trans. F. C. Babbitt, Cambridge, MA: Harvard University Press.
Rorty, A. O. ed. (1992) Essays on Aristotle’s Poetics, Princeton: Princeton University Press.
Russell, D. A. (1989) “Neo-Platonic Interpretation”, in G. A. Kennedy (ed.), The Cambridge History of
Literary Criticism, vol. 1 (Classical Criticism), Cambridge: Cambridge University Press.
Schleiermacher, F. (1998) Hermeneutics and Criticism, Eng. trans. A. Bowie, Cambridge: Cambridge
University Press.
Seneca (1917–25) Ad Lucilium Epistulae Morales, Eng. trans. R. M. Gummere, 3 vols, Cambridge, MA:
Harvard University Press.
Tolstoy, L. (1930) What is Art?, Eng. trans. A. Maude, London: Oxford University Press.

Further Reading
Halliwell, S. (2011) Between Ecstasy and Truth: Interpretations of Greek Poetics from Homer to Longinus,
Oxford: Oxford University Press.
Heath, M. (2012) Ancient Philosophical Poetics, Cambridge: Cambridge University Press.

Paul Guyer

1 Introduction
Philosophy of literature was not a named subfield of aesthetics in the eighteenth century.
Indeed, aesthetics itself had only been so named in Germany in 1735 by Alexander
Gottlieb Baumgarten (see Baumgarten 1954, §CXVI), and the name for the more
general field came into use in Britain only in the nineteenth century, but still without
a name for the subfield. Moses Mendelssohn did draw a distinction between the “fine
arts” and “fine sciences” (schöne Künste or beaux artes and schöne Wissenschaften or belles
lettres), defining the former as those that use “natural signs,” such as their appearance,
to “express” or refer to their objects and the latter as those that use “arbitrary signs,” such
as words, which refer to their objects only by convention, and said that the latter include
“poetry and rhetoric,” while the former would include not only the visual arts but also
arts employing natural tones and gestures, thus music and dance as well (Mendelssohn
1997, pp. 177–8). But Mendelssohn did not argue that the philosophy of the arts that
employ artificial signs constitutes a special discipline within aesthetics, and in any case
his terminological distinction was not widely adopted—we find no trace of it in Kant’s
classification of the arts, for example, though Mendelssohn influenced Kant’s aesthetics
in many other ways.
Yet philosophy of literature did exist in the eighteenth century. For not only did many
aestheticians base their supposedly general theories almost entirely on the case of
literature (usually understood to be comprised of lyric and epic poetry and drama, which
was itself usually assumed to be a form of poetry, thus neglecting the novel in spite of its
increasing popularity throughout the century), such as Baumgarten, Kant, and between
them Henry Home, Lord Kames, whose 1762 Elements of Criticism was one of the most
popular and influential works of aesthetics in the whole century; several of the topics
most hotly debated in aesthetics as a whole throughout the century, such as the reasons
for our pleasure in tragedy or the comparison between painting and poetry, were obviously
in whole or in large part ways of philosophizing about literature. Those two debates will
be the focus of this chapter, but before turning to them, several connected issues about
the goals of literature and the means of attaining those goals must first be considered.
Throughout, we will see that the distinction between natural and artificial signs, made
Paul Guyer

prominent by Mendelssohn but going back to Augustine and even to Plato’s Cratylus,
was often at the heart of discussion.

2 The Function of Literature

Since antiquity, many writers and readers accepted Horace’s statement that the function
of poetry is to “instruct” and “delight,” prodesse et delectare, but preferably to do both at
once, simul iucundu et idonea dicere (“to say something both appropriate and delightful”
(De arte poetica, 333)). Horace’s work was available in French by 1541, and his thought
that poetry must have both instructive content and pleasing form was a commonplace
taken up, for example, by Nicolas Boileau in his own Art of Poetry (1674), where he
instructs the poet, “Whether your subject matter is sublime or amusing, be sure that good
sense and rime always agree” (Boileau 1965, p. 12). Essentially the same view is still
expressed at the end of the eighteenth century by Kant, when he argues that the “spirit”
of fine art lies in its expression of “aesthetic ideas,” and that among the arts poetry ranks
first in its capacity for the expression of aesthetic ideas. In his initial analysis of the “pure”
judgment of taste and its object, “free” beauty, Kant had argued that our pleasure in
beauty comes from a “free play” between imagination and understanding triggered by the
form of the representation of an object, for example by spatial design rather than color in
the case of painting or by composition rather than instrumentation in the case of music,
and that such free play takes place independently of the subsumption of the object under
any concept, a fortiori independently of any content that could be conveyed only through
concepts (Kant 2000, Introduction, section VII, and §§2, 5, 9, 13, and 14).1 But in the
“Analytic of the Beautiful,” where this analysis is found, Kant is primarily concerned with
the beauty of individual objects in nature, such as flowers and birds, and, although he
does not put it this way, is primarily concerned to identify only a necessary condition for
all beauty and aesthetic judgment. The beauty of art, however, is impure and more
complex than the beauty of nature, especially the beauty of “fine” or “beautiful art”
(schöne Kunst), a term which Kant, unlike Mendelssohn, uses comprehensively for the
visual arts of painting, sculpture, architecture, and garden design; for music and dance,
and for literature, as well as for various combinations of these media, such as opera. When
Kant finally offers his account of fine art in the form of his theory of genius, only after the
intervening “Analytic of the Sublime” and “Deduction” of judgments of taste, he argues
that works of artistic genius must present aesthetic ideas, that is, weighty, paradigmati-
cally moral content, such as “rational ideas of . . . the kingdom of the blessed, the kingdom
of hell, eternity, creation . . . death, envy, and all sorts of vices, as well as love, fame, etc.,”
by means of “a representation of the imagination that belongs to its presentation, but
which by itself stimulates so much thinking that it can never be grasped in a determinate
concept, hence which enlarges the concept itself in an unbounded way,” so that “in this
case the imagination is creative, and sets the faculty of intellectual ideas (reason) into
motion” (Kant 2000, §49).2 In other words, a successful work of fine art must both present
important content and yet give the imagination room for free play, thus it must both
instruct and delight. For this reason Kant further says that:

The poet announces merely an entertaining play with ideas, and yet as much
results for the understanding as if he had merely had the intention of carrying
on its business. . . . The poet . . . promises little and announces a mere play with
ideas, but accomplishes something that is worthy of business, namely providing


nourishment to the understanding in play, and giving life to its concepts through
the imagination.
(Kant 2000, §53)

And because poetry does this through “word,” that is artificial signs, capable of referring
to anything and everything, but also through “gesture” and “tone” (perhaps especially,
although Kant does not say this, when spoken or performed on stage), poetry can express
aesthetic ideas better than any other art form, and thus ranks more highly than any
other with regard to what is most important to art.
But although poetry on Kant’s own account deals with the weightiest of human ideas,
Kant says nothing about the emotional impact that such ideas might be expected to
have, and the communication of which might have been thought to be among the most
important aims of art. In fact, Kant says that any art that needs “charm and emotion”
(Reiz und Rührung) to please is “barbaric” (Kant 2000, §13), and although he later allows
that music can please by conveying an “affect,” which might be interpreted as an emotion,
he insists that it does this without communicating any definite idea and thus “it has,
judged by reason, less value than any other of the beautiful arts” (Kant 2000, §53). But
other writers of the period were not hesitant to identify the production or communica-
tion of emotion as among, or even as chief among, the aims of art in general and
literature in particular. In one of the earliest but most enduringly influential works in
aesthetics of the eighteenth century, his Critical Reflections on Poetry, Painting, and Music
published in French in 1719 and translated into English in 1748, the Abbé Jean-Baptiste
Du Bos argued that boredom is one of our most painful conditions, that boredom is best
dispelled by the arousal of intense emotions, but that while some ways of arousing
emotions, such as by gambling for high stakes, often have great costs, the arousal of
emotions through works of art has great benefit but little cost. He wrote that:

Since the most pleasing sensations that our real passions can afford us, are
balanced by so many unhappy hours that succeed our enjoyments, would it not
be a noble attempt of art to endeavour to separate the dismal consequences of
our passions from the bewitching pleasure we receive in indulging
them? . . . Might not art contrive to produce objects that would excite artificial
passions, sufficient to occupy us while we are actually affected by them, and
incapable of giving us afterwards any real pain or affliction?
(Du Bos 1748, vol. I, p. 21)

Indeed, on Du Bos’s account the passions produced by art are artificial only in that they
are not produced by the ordinary means and do not have their usual effect; they feel real
and are fully engaging while we are having them, and that is why we value them and
value art for producing them.
Du Bos, a diplomat and historian, was an inspired amateur in philosophy, but more
professional philosophers took up his ideas. Even before his work was translated into
English, British philosophers such as David Hume were very familiar with it.3
Mendelssohn initially criticized Du Bos in the 1755 first edition of his Letters on the
Sentiments, but in the Rhapsody that he added to this work in his 1761 Philosophical
Writings, he conceded that he had not done Du Bos justice (Mendelssohn 1997,
pp. 136–7) and that Du Bos’s theory of artificial passions anticipated aspects of his own
theory of “mixed sentiments.” On this theory, the presentation of intellectual content

Paul Guyer

in works of art through their beautiful form is only one source of our pleasure in them,
and we also enjoy the arousal of our emotions by works of art, even painful ones, as well
as the exercise of our moral capacity for approbation and disapprobation—in other
words, works of art exercise our emotional, intellectual, and moral capacities, and allow
for complex responses because of that fact (Mendelssohn 1997, pp. 133–6, 172–4). And
in his Elements of Criticism, published the year after Mendelssohn’s Philosophical Writings,
Henry Home, Lord Kames, Scotland’s leading jurist but also a more sophisticated
philosopher than Du Bos, indicated by the very title of his work that the effects of art
cannot be reduced to any single one (Kames 2005a, vol. I, p. 19), but also devoted his
second-longest chapter to “Emotions and Passions” (Kames 2005a, chapter II, vol. I,
pp. 32–140). Although works of art can please us through formal properties such as
beauty (Kames 2005a, chapter III, vol. I, pp. 141–9), we value them most highly because
they “open a direct avenue to the heart of man,” and even insofar as we also value them
because they convey knowledge, it is chiefly “a thorough knowledge of the human heart”
that the arts afford (Kames 2005b, vol. I, p. 32)—for Kames, the cognitive and
emotional impact of art are not really to be separated. Indeed, Kames’s longest chapter,
“On the Beauty of Language,” concerns the emotional impact of language as much as it
does its formal pleasures (Kames 2005a, chapter XVI, vol. II, pp. 373–494): “Language
may be considered as the dress of thought” (ibid., p. 386), but one of its “chief pleasure[s]”
also “proceeds from having . . . concordant emotions combined in perfect harmony, and
carried on in the mind to a full close” (ibid., p. 431).
Kant’s effort to keep “charm and emotion” out of the experience of art was thus a
protest against the prevailing view of the aim of the arts in general and of literature in
particular in the eighteenth century. But the importance of the emotional impact of art,
especially literature, for most other philosophers of the century—indeed, even the
supposed cognitivist Baumgarten had claimed that it is “most poetic to arouse affects”
(Baumgarten 1954, §XXVI)—now raises the question of how literature achieves its
emotional impact through a medium that consists of “artificial signs” and fictions.

3 The Emotional Impact of Literature

The issue can be phrased in terms of the distinction between natural and artificial signs.
It might seem only natural, after all, that natural signs, connected with their objects by
resemblance or causation, should also trigger emotions typically connected with those
objects, thus that the sight or smell of smoke, caused by a fire and triggering the idea of
it, should also trigger fear of that fire, or even that an as we say “naturalistic” picture of
a beautiful child should trigger the emotions we might feel at the sight of an actual
beautiful child, which in some ways it resembles. But how can artificial signs, such as
words, connected to their objects not by resemblance or causation but only by conven-
tion (which could be entirely different a few miles away), trigger the same emotions that
their intended objects would trigger? There were two main approaches to this problem,
both suggested by ideas from John Locke, the font of so much eighteenth-century phi-
losophy not only in Britain but throughout the rest of Europe as well. Locke’s theory of
the meaning of words, according to which words refer to objects through the medium of
ideas in the minds of speakers and hearers, could suggest that words trigger ideas or
images of their objects in the minds of hearers or readers, which images, resembling their
objects, then trigger emotions by the same pathways, whatever those might be, as would
their objects (Locke 1975, Book III, Chapter II); while Locke’s theory of the association


of ideas, although introduced by him only to explain aberrant thoughts, could suggest
more generally that the ideas (the sounds or shapes) of words that have become
connected to certain objects could acquire the causal power to trigger emotions associated
with those objects without going through the medium of ideas or images of those objects
(Locke 1975, Book II, Chapter XXXIII).4 For an example of the first of these approaches,
we can turn again to Kames; for an example of the second, we can turn to the influential
Part Five of Edmund Burke’s Philosophical Enquiry into the Origin of our Ideas of the Sublime
and Beautiful (1958).
Kames claimed that:

A lively and accurate description of an important event, raises in me ideas no

less distinct than if I had been originally an eye-witness: I am insensibly
transformed into a spectator; and have an impression that every incident is
passing in my presence.

He called the production of such vivid images “ideal presence,” and thus claimed that
“ideal presence supplies the want of real presence; and in idea we perceive persons acting
and suffering, precisely as in an original survey.” He then inferred that emotions can be
triggered by ideal presence just as by real presence: “if our sympathy be engaged by the
latter, it must also in some degree be engaged by the former, especially if the distinctness
of ideal presence approach to that of real presence.” Thus, he continued, “The power of
language to raise emotions, depends entirely on the raising such lively and distinct
images as are here described” (Kames 2005a, vol. I, pp. 68–9). In fact, Kames did not
limit ideal presence to the effects of spoken or written language; since in the case of the
depiction of an object by a work of visual art, such as a painting, the object depicted is
not actually before us, Kames thought that in this case too our emotion is produced by
ideal presence: the painting triggers an image of its object, and that image in turn trig-
gers an emotion. Ideal presence can thus be created “by speech, by writing, or by
painting” (ibid., p. 68), and thus is the pathway to emotional impact in all of the arts.
And that paintings as well as speech raise emotions in the same way suggests that the
theory of ideal presence as an explanation of the emotional impact of artificial signs
actually works by piggy-backing on the emotional impact of natural signs: paintings
produce emotional impact because they produce images that resemble those that would
be produced by their intended objects and naturally produce the same emotional effects
the latter images would produce, at least to some degree; and if literature raises the same
sorts of images that painting does, then it produces emotion in the same way.
That literature and art produce emotions by the same pathway, although not neces-
sarily to the same degree, leads Kames to a comparison of the emotional impact of
different artistic media: he argues that words, the medium of literature, does not produce
such vivid images as paintings do, and therefore does not produce quite the same degree
of emotional impact, while theater, or the staged performance of literature, combines
words with visual depiction, thereby achieving a greater degree of ideal presence than
either words or painting alone and a correspondingly greater degree of emotional impact.
In his words,

Of all the means for making an impression of ideal presence, theatrical repre-
sentation is the most powerful. That words independent of action have the
same power in a less degree, every one of sensibility must have felt: a good

Paul Guyer

tragedy will extort tears in private, though not so forcibly as upon the stage.
That power belongs also to painting: a good historical picture makes a deeper
impression than words can, tho’ not equal to that of theatrical action. Painting
seems to possess a middle place between reading and acting.
(Kames 2005a, vol. I, pp. 71–2)

Kames’s three-way comparison among painting, written literature, and theater was not
unusual, although we will see later that a more popular subject for debate was just the
comparison between poetry and painting. But before we come to that, we must first
consider the other approach to the emotional impact of literature: that based on Locke’s
conception of the association of ideas, and also a specific issue referred to in the last
quotation from Kames, namely the debate over the effects of tragedy.
As already mentioned, Burke based his account of the emotional impact of litera-
ture in Part 5 of his Philosophical Enquiry, first published in 1757, thus five years before
Kames’s Elements, on the theory of the association of ideas. In the previous parts of his
book, Burke explained the experience of the beautiful and the sublime as consisting
in the perception of images of that which we find terrifying or lovable, respectively
(Parts 2 and 3), and then provided a physiological explanation of the pleasure or
delight (distinguished in Part 1) that we take in such imagery as due to the invigoration
or relaxation of our organs and especially our nervous system by such images (Part 4).5
In his account of the sublime, Burke uses passages from the poetry of Virgil,
Shakespeare, and Milton, but his theory of the beautiful as well as the sublime remains
primarily a theory of the effects of the experience of nature and of works of visual art
that imitate nature.6 In Part 5, however, Burke turns explicitly to the case of words,
and argues that they “affect us in a manner very different from that in which we are
affected by natural objects, or by painting or architecture,” although they may “have as
considerable a share in exciting ideas of beauty and of the sublime as any of those,
and sometimes a much greater share than any of them” (Burke 1958, p. 163). In
particular, Burke claims, “The common effect of POETRY” is “not by raising ideas of
things” (ibid.), but comes simply from words having, “by use” (p. 167), certain emotional
associations, such that hearing or seeing the words immediately triggers the associate
ideas without any intervening imagery. This must be the case, Burke argues, because
even in those cases where words can call up imagery—he illustrates this possibility
with an elaborate description of the course of the Danube from the mountains of
Bavaria to the Black Sea—the emotional impact comes more quickly than the images
do, and in other cases the words that trigger emotions have no images associated with
them at all. This is so in the case of “compounded abstract words,” such as “honour,
justice, liberty, and the like” (ibid., p. 166), but also in the case of words that designate
particular beings but ones we never have and never can experience, such as “God,
angels, devils, heaven and hell, all of which however have a great influence over our
passions” (ibid., p. 174).7 In spite of the fact that literary words achieve their emotional
impact without images, however, Burke, contrary to Kames, holds that “we find by
experience that eloquence and poetry are as capable, nay indeed much more capable
of making deep and lively impressions than any other arts, and even than nature itself
in very many cases” (ibid., p. 173). As Burke’s appeal to experience implies, he correctly
recognizes that the degree of emotional impact possible for various media of art can
only be determined empirically, not a priori.


It seems an undeniable fact that fictions produce emotions.8 But neither Kames’s
theory of ideal presence nor Burke’s theory of the association of ideas seems an
adequate explanation of this fact. As regards ideal presence, there appears to be too
much variation among individuals in whether they have visual imagery while reading
for the theory to be sufficiently general, and the theory also seems to run the risks of
circularity, that the only evidence for the vivacity (to borrow Hume’s term) of the
supposed imagery that causes emotions would be the vivacity of the emotions them-
selves. As regards the association of ideas, it is unclear how emotions could come to
be associated with the names for things like justice and honor unless they were first
caused by actual instances of those things, a step in the explanation of emotion that
Burke omits even if it would not be impossible for him to take it. At the very least,
his theory seems incomplete. Nevertheless, the theory of the association of ideas
became the dominant model for the explanation of emotions in response to art in
subsequent eighteenth-century British aesthetics, reaching its full flower in Archibald
Alison’s 1790 Essays on the Nature and Principles of Taste (see Alison 1811). But
instead of pursuing the details of Alison’s account, which have been recounted by
others,9 we can now turn to the extensive debate about a specific case of emotion
triggered by literature that raged in the eighteenth century, namely the debate
over the nature of the emotions triggered by tragedy and the reasons for our pleasure
in this literary genre.

4 The Paradox of Tragedy

Du Bos commenced his Critical Reflections with what he called the paradox that

the more our compassion would have been raised by such actions as are
described by poetry and painting, had we really beheld them; the more in pro-
portion the imitations attempted by those arts are capable of affecting us.
(Du Bos 1748, vol. I, p. 2)

Or, as he also puts the paradox, one might think that we would take greater pleasure in
depictions of characters enjoying themselves than in ones of persons suffering, but in
fact it is the other way around. Why? The first part of Du Bos’s answer follows from his
general theory that we enjoy the arousal of any emotions by fictions because they arouse
us from our boredom without any great costs, but that the emotions aroused by tragedy
are more intense than those aroused by any alternative, and thus afford us greater
stimulation and relief from ennui. As he puts it, “the terror and pity, which the picture
of tragical events excites in our souls, engages us much more than all the laughter and
contempt excited by the several instants of comedies” (ibid., p. 51). But that is only part
of Du Bos’s answer, and he makes two further claims. One is that while the emotions
aroused by tragedy may be necessary to engage our attention, we also learn from tragedy,
namely we learn to try to avoid the kind of conduct that has such tragic results for the
depicted characters: “Now a faithful image of the passions is sufficient to strike us with
horror, and to induce us to determine resolutely to avoid them” (ibid., p. 355). Second,
Du Bos makes the important observation that the emotions that spectators or readers of
tragedy experience are not the same as the painful emotions that the characters in the
tragedy are depicted as undergoing, thus that while the latter may be depicted as

Paul Guyer

experiencing unbridled pain, the audience instead may feel the very different emotion
of compassion, which is by no means unpleasant.

Tragedy pretends indeed, that the passions it represents should move us; but it
does not intend that our emotion should be the same as that of a person
tortured by his passion, or that we should espouse his sentiments. Its aim most
frequently is to excite opposite sentiments to those, which it lends to the
personages it depicts.
(Ibid., p. 358)

And that we are sensitive to the suffering of others, and respond with an urge not to flee
but to help, is in Du Bos’s view the foundation of society:

Nature has thought proper to implant this quick and easy sensibility in man as
the very basis of society. . . . We are moved by the tears of a stranger, even
before we are apprized of his weeping. The cries of a man, to whom we have no
other relation than the common one of humanity, make us fly instantly to his
assistance, by a mechanical movement previous to all deliberation.
(Ibid., p. 32)

In the case of a fiction, of course, there is nothing we can do to aid the sufferer, and we
know that perfectly well, but the exercise of our disposition to socially beneficial
emotions through art nevertheless prepares us to act when in real life we can do so.
It should come as no surprise that Du Bos’s ultimate explanation that our pleasure in
tragedy comes from its arousal of the sentiments that are the condition of the possibility
of society was taken up by the adherents of the “moral sense” school of British philoso-
phers, and in a 1751 essay on “Our Attachment to Objects of Distress,” Kames similarly
argues that “tragedy is allowed to seize the mind with all the different charms which arise
from the exercise of the social passions” (Kames 2005b, p. 20), although he adds a refine-
ment to Du Bos’s third point. Kames distinguishes between pleasure and pain on the one
hand and “affection” and “aversion” on the other, and argues that “there are many emotions
or passions, some of them of the most painful sort, that raise no aversion” (ibid., p. 14).
The sight of the suffering of others is painful, but it raises no aversion in us; “On the
contrary, [it] draw[s] us to them, and inspire[s] us with a desire to afford relief”
(ibid., p. 15). This is true in the case of fictional suffering as well, and thus tragedy does
not excite in us the painful emotions supposed to be undergone by its characters and
aversion to those emotions, but a separate set of “social passions,” some of which may be
painful but do not produce aversion and others of which may in fact be pleasurable:
“tragedy engages our passions, no less than true history. Friendship, concern for the virtuous,
abhorrence of the vitious, compassion, hope, fear, and the whole train of the social
passions, are roused and exercised by both of them equally” (ibid., p. 18).
Kames’s cousin David Hume took a different approach to the paradox of tragedy. In
his 1757 essay “Of Tragedy,” substituted in a volume of Four Dissertations at the last
minute, along with “Of the Standard of Taste,” in lieu of more controversial essays on
the moral permissibility of suicide and the indemonstrability of the immortality of the
soul,10 Hume offers what we might think of as a refinement of Du Bos’s first resolution
of the paradox of tragedy, that turning on the sheer pleasure of any arousal of artificial
passions, rather than the third, that turning on the special pleasures of social passions.


Hume begins with the claim that “It is impossible not to admit” that Du Bos’s thesis that
the experience of any passion “is still better than that insipid languour, which arises from
perfect tranquillity and repose,” is, “at least in part, satisfactory” (Hume 1987, p. 217).
What he adds to Du Bos’s account is the argument that we take pleasure in the medium
of tragedy, that is, its language, and the genius required to produce that—a kind of
pleasure that tragedy shares with non-fictional “eloquence”—and we can also take
pleasure in the kind of imitation that is specific to tragedy as a kind of fiction, for “imi-
tation is always of itself agreeable.” He then argues that the intense feelings aroused
by the content of tragedy, though not initially pleasurable, “receives a new direction
from the sentiments” of these forms “of beauty,” as the intensity of the pain triggered by
the depiction of tragic events is redirected and reinvested into the pleasure triggered
by eloquence and imitation, and that

The latter, being the predominant emotion[s], seize the whole mind, and
convert the former into themselves, at least tincture themselves so strongly as
totally to alter their nature. . . . The affection, rouzing the mind, excites a large
stock of spirit and vehemence, which is all transformed into pleasure by the
force of the prevailing movement.
(Hume 1987, pp. 220–1)

Hume thus explains away the paradox of tragedy by converting its pain into pleasure.11
While Hume recognizes that our response to tragedy is complex because it is a
response to literary form and to the fact of imitation as well as to the content of the
literary work, his solution to the paradox of tragedy might still be regarded as simplistic
because it ultimately does away with the reality of any enduring pain in our experience
of such work. Mendelssohn’s contemporaneous theory of “mixed sentiments” might be
seen as a more complex resolution of the paradox than either Hume’s or Kames’s.
Mendelssohn’s aesthetics is based on the premise from Leibniz and Christian Wolff that
all pleasure must be a response to the perception of some perfection (see Leibniz 1969,
pp. 425–6, and Wolff 1719, §404), so on his view we cannot directly take pleasure in
anything that we perceive as an outright imperfection in the world, such as undeserved
suffering. In the original, 1755 edition of his Letters on the Sentiments, this leads him to
reject Du Bos’s theory as one that finds pleasure in the arousal of any emotions, no mat-
ter what their objects (Mendelssohn 1997, p. 71). But Mendelssohn finds numerous
sources of pleasure in the experience of tragedy, even though it does depict unpleasant
events that are imperfections in the grand scheme of things (or would be if they were
not fictional). Mendelssohn emphasizes the difference between a representation and an
object represented by it (he should distinguish two kinds of representation, an artistic
representation of an object and a mental representation of an object, which might in
turn be triggered by an artistic representation), and he holds that we can take pleasure
in a representation, as a kind of perfection, for example in the unity and clarity of the
representation (Mendelssohn 1997, pp. 14–15), even when the object represented is not
a perfection. As Mendelssohn clarifies in his 1757 essay on “The Main Principles of the
Fine Arts and Sciences” and in the Rhapsody added to the 1761 republication of
the Letters, in such a case we are in fact taking pleasure in the exercise of our own cognitive
powers through such a representation, which is certainly a case of perfection. As
Mendelssohn adds in the “Conclusion” to the Letters, we can also take pleasure in the
fact of imitation and the human skill it takes to produce imitation (Mendelssohn 1997,

Paul Guyer

pp. 71–2), but his conception of our pleasure in imitation is more complex than this
Humean point alone; he also observes that a certain amount of pain actually heightens
pleasure, as “a few bitter drops . . . mixed into the honey-sweet bowl of pleasure . . . 
enhance the taste of the pleasure and double its sweetness,” and argues that “the recollection”
that a work of fiction “is nothing but an artistic deception lessens our pain to some extent
and leaves only as much of it as is necessary to lend our love the proper fullness”
(ibid., pp. 74–5). Mendelssohn also recognizes that a work of art such as a tragedy exercises
not only our cognitive capacities but also our moral powers, our capacity for judgment
and approbation or disapprobation, and thus even though evil actions and tragic events
are themselves imperfections in the world, “recognizing an evil action and disapproving of
it are affirmative features of the soul, expressions of the mental powers of knowing and
desiring, and elements of perfection which, in this connection, must be gratifying and
enjoyable” (ibid., pp. 133–4). Finally, and even more specifically, Mendelssohn makes an
Aristotelian point that seems to have escaped other participants in the debate over
tragedy, namely, that since the tragic hero must be a person with great qualities as well as
some great flaw, “the theatrical presentation of innumerable episodes of ill fortune, to
which someone virtuous succumbs, enhance[s] our love for his perfection and make[s] him
worthier in our eye!” (ibid., p. 75). Our response to a tragic hero is not just compassion, as
Kames suggested, but involves admiration as well. On Mendelssohn’s account, then, our
response to tragedy is complex, involving both cognitive capacities and moral sentiments
and judgments, and a mix of compassion and affection, in which pain is not simply
transformed into pleasure but both pleasure and pain are subtly combined. Mendelssohn’s
theory is thus as sophisticated a theory of response to art in general and to the literary art
of tragedy in particular as can be found in eighteenth-century aesthetics.12

5 Poetry versus Painting (and Music)

The distinctive character of literature is explicitly thematized in the comparison of the
sources of beauty and emotional impact in poetry and painting in which many authors
indulged, some also including music or even other arts in their comparison as well.13
Kames’s comparison of theater, written literature, and painting with regard to their
potential for triggering “ideal presence” has already been noted. But the comparison
between poetry and painting especially was a well-established theme long before Kames’s
Elements. Du Bos had already raised the question “Whether the effect which painting
produces on men, be greater than that of poetry,” but although he initially announced
that the effect of painting “surpasses that of poetry,” he actually reached a split decision.
He first maintained that “sight has a much greater empire over the soul than any of the
other senses,” and that since painting not only works through sight but also makes
use of “natural signs, the energy of which does not depend on education,” as opposed to
the “arbitrary or instituted” signs “employed by poetry” (Du Bos 1748, vol. I, pp. 321–2),
from which he inferred that paintings have greater emotional impact than poetry. At
this point in his argument, he also made the same comparison between staged and
merely read tragedy: since the former uses natural signs that appeal to sight as well as
artificial sounds that influence us through hearing, “A tragedy represented on the stage,
produces its effect by means of the eye,” while “Tragedies that are read in private, very seldom
make us weep; especially when we read them without having seen them previously acted”
(ibid., p. 328). However, Du Bos next argues that while painting is ordinarily confined
to the representation of a single moment in an action, thus “A painter who draws the


sacrifice of Iphigeneia, represents only one instant of the action,” a poem can represent
successive moments of an action, as “Racine’s tragedy exhibits to our sight several
instants of this action,” and the poet can thus achieve a greater cumulative effect even
if the effect of each of his descriptions of individual moments has less impact than a
painting of it would. Thus “The poet presents us successively with fifty pictures, as it
were, which lead us gradually to that excessive emotion, which commands our tears”
(ibid., p. 329). Of course a staged tragedy, which might consist of fifty tableaux instead
of the single one presented by a painting, will have a greater emotional impact than that
one painting; but even a read tragedy, where the images presented to the mind’s eye are
not as vivid as those a painting presents to the actual eye, can still have a greater cumu-
lative emotional effect than one painting, powerful as the effect of the latter considered
in isolation may be. “Hence we are more moved by a poem than a picture, tho’ painting
has a greater empire over us than poetry” (ibid., p. 330).
In spite of this split decision, the grounds for Du Bos’s comparison between poetry
and painting are simple, involving solely his suppositions about the singular and cumu-
lative effects of images and words. In A Discourse on Music, Painting, and Poetry, first
published in 1744, James Harris, the nephew of Anthony Ashley Cooper, third Earl of
Shaftesbury, offers a more complex comparison. First he makes a three-way comparison
among music, painting, and poetry, arguing that since both music and painting use
natural signs, what they can represent is restricted to that for which there are natural
signs, while poetry, using words, which “stand by Compact for the various ideas with
which the Mind is fraught,” can, “in a manner, include all things” (Harris 1801, vol. I,
p. 35) (of course music with words would combine the expressive power of both the
natural signs of music and the artificial signs of words). Harris also thinks that all the
arts are “mimetic, or imitative,” so they are to be compared first of all by their capacity to
represent different sorts of objects.
Painting can represent colors and figures (ibid., p. 37), while the “fittest subjects of
imitation” for music will be motion and sound, but Harris does not infer that one of these
media of art is superior to its emotional impact to the other from this difference. He does,
however, argue that poetry is superior to both music and painting. It is superior to painting
because while “of necessity every Picture is a Punctum Temporis or Instant” (ibid., p. 38):

The Subjects of Poetry, to which the Genius of Painting is not adapted, are—all
Actions, whose Whole is of so lengthened a Duration, that no Point of Time, in
any part of that While, can be given Fit for painting.

But this does not just mean that a poem can have a greater cumulative effect than any
single painting; because action “gives us an Insight into Characters, Manners, Passions,
and Sentiments” of people, and these are certainly the most affecting subjects for art, “the
most improving; and such of which the Mind has the Strongest Comprehension”
(ibid., p. 50), the emotional impact of poetry is also deeper as well as cumulatively
greater than that of a painting. The same would hold of poetry in comparison to music,
at least music without words, and thus Harris concludes that:

Poetry is therefore, on the whole, much superior to either of the other mimetic
Arts; it having been shewn to be equally excellent in the Accuracy of its Imitation;
and to imitate Subjects, which far surpass, as well as in Utility, as in Dignity,
those of the other arts.
(Ibid., p. 55)

Paul Guyer

Music, however, has a power to arouse affections that does not depend upon mimesis,
and music with words can combine this agency with the agency of poetry, thus “these
two Arts can never be so powerful singly, as when they are properly united” (ibid., p. 59).
But since “Music, when alone, can only raise Affections, which soon languish and
decay, if not fed by the nutritive Images of Poetry” (ibid., pp. 59–60), the emotional
impact of poetry, even if not as great when poetry is alone as when it is combined with
this other art, remains the necessary condition of the enduring emotional impact of all
other art.
However, the most famous comparison between poetry and painting, in this case
without reference to music, was undertaken by Gotthold Ephraim Lessing in his 1766
work Laocoön, or on the Limits of Painting and Poetry.14 His friend Mendelssohn had
touched upon the already well-known point that a painting can represent only one
moment of an action in his 1757 essay on the “Main Principles,” observing that:

Since the painter and sculptor express beauties that are alongside one another,
they must choose the instant that is most favorable to their purpose. They must
assemble the entire action into a single perspective and divide it up with a great
deal of understanding.
(Mendelssohn 1997, pp. 180–1)

adding in his typical fashion that if they do so, then through their works “our senses
are all at once animated, all the capabilities of our soul suddenly become lively, and
the imagination can fathom the past from the present while reliably anticipating the
future” (ibid.).
Lessing embedded this point in a larger polemic with Johann Joachim Winckelmann,
in the course of which also he argued against Winckelmann’s explanation of the beauty
of classical Greek art, refined the conception of poetry as the depiction of temporally
extended action, and argued for the superiority of poetry over painting. In his 1755
Reflections on the Painting and Sculpture of the Greeks, Winckelmann had memorably
claimed that the beauty of Greek art, even in the case of the face and torso of the central
figure of the famous statue of the Trojan priest Laocoön and his sons being strangled by
Neptune’s serpents, was an expression of the “noble simplicity and sedate grandeur” of
the Greek sensibility (Winckelmann 1765, p. 30). Lessing rejected this account, arguing
instead that “among the ancients beauty was the supreme law of the visual arts” (Lessing
1984, p. 15), thus that a visual presentation of anything ugly had to be beautified, there-
fore Laocoön’s moment of supreme pain could not be represented in a sculpture and his
“scream had to be softened to a sigh” (ibid., p. 17). Since poetry works with artificial
rather than natural signs, its representation of something ugly does not itself have to be
ugly, and this is already one mark of the superiority of poetry over painting. However,
because the staged presentation of a drama shares some of the character of visual art, it
is limited in how it can present painful subject-matter, and in this way is more limited
than poetry as such (ibid., p. 24). In this way, Lessing reverses the judgment of Kames,
for example, that theater is superior to poetry because it combines the impact of painting
with that of poetry.
Lessing also puts the point that visual art is confined to the presentation of a single
moment in an action to work in his own explanation of the appearance of the Laocoön
statue. He shared Mendelssohn’s view that this limitation of visual representation means
that the “single moment and the point from which it is viewed” in a sculpture or painting


“cannot be chosen with too great a regard for its effect. But only that which gives free
rein to the imagination is effective. The more we see, the more we must be able to imag-
ine.” And from this he inferred that, “In the full course of an emotion, no point is less
suitable for this than its climax” (Lessing 1984, p. 19). Thus the moment of Laocoön’s
extreme pain also could not be represented in the statue because it is the climax of the
action being depicted, and would leave nothing further for the imagination. Poetry does
not suffer this limitation either; it always leaves much to the imagination. And now
Lessing adds yet another point to the contrast between visual and poetic art. Whereas
Burke had argued that poetry had to work by the association of ideas because it refers to
abstract objects or qualities, Lessing allows that poetry, like painting or sculpture, can
depict bodies in action, but because poetry describes a succession of moments rather
than a single moment, it is best at depicting bodies through the depiction of action, in
particular the action of making the bodies in question:

For example, if Homer wants to show us Juno’s chariot, he shows Hebe putting it
together piece by piece before our eyes. . . . And when Homer wants to show us
how Agamemnon was dressed, he has the king put on his garments, one by one.
(Ibid., p. 80)

Most famously, when Homer wants to depict the shield of Achilles, he “does not paint
the shield as finished and complete, but as a shield that is being made” (ibid., p. 95). On
Lessing’s account, then, the range of poetry is virtually unlimited: it can describe the
ugly as well as the beautiful, the concrete as well as the abstract, and engage us and our
emotions through its unparalleled capacity for the depiction of that which engages us
the most, namely the actions of human beings or very human gods. Even when it comes
to the depiction of physical beauty, poetry wins the palm: when it describes a beautiful
woman, for example,

Her bosom enchants us, not so much because its whiteness and delicate shape call
to mind milk and ivory apples as because we see it gently rise and fall, like the
waves on the farthest edge of the shore when a playful zephyr contends with the sea.
(Ibid., p. 113)

In his early Groves of Criticism (1768; Herder 2006) and subsequent essay on sculpture
(1778; Herder 2002), Johann Gottfried Herder criticized Lessing for failing to distin-
guish between the appeal of painting to sight and of sculpture to touch, and likewise for
failing to distinguish between the two arts that appeal to hearing, namely music as well
as poetry, and for failing to grasp that music is more purely an art of succession than
poetry. Yet because he upheld “the supremacy of hearing over the other senses” (Herder 2006,
p. 250), contrary to Du Bos, who had maintained “the empire of sight,” Herder regarded
both music and poetry as superior to either visual or plastic art; and while he held that
music is unparalleled in its capacity to express passion and emotion, he still held that
poetry could express an even wider range of ideas, thus that in this regard poetry is
superior to music, although “the emotion-laden language of poetry supported by music”
is the greatest art of all (ibid., p. 255).
Herder thus reached the same conclusion as had James Harris (whose work he knew;
see Herder 2006, pp. 155–7), but at the same time pointed forward to Richard Wagner’s
conception of a Gesamtkunstwerk. Here we cannot follow Herder into the nineteenth

Paul Guyer

century, but will conclude by returning to Kant, who perhaps surprisingly also points the
way toward this characteristic nineteenth-century conception. As already noted, once
he had stated his theory of aesthetic ideas, Kant turned to his own comparison among
the arts. The comparison was biased in favor of literature from the outset, for once Kant
had argued that art pleases by its content as well as its form and by the interplay between
them, it was natural for him to compare the arts to language: “if we wish to divide the
beautiful arts, we can, at least as an experiment, choose no easier principle than
the analogy of art with the kind of expression that people use in speaking in order to
communicate with each other.” Through speech, Kant claims, people communicate to each
other “not merely their concepts, but also their sentiments” (Empfindungen), that is,
their feelings. His argument is then, to put it bluntly, since nothing is better than speech
for communicating both concepts and emotions through “word,” which communicates
the former, and “gesture” and “tone” (Kant 2000, §51), which add the latter, no artistic
medium can communicate more and have more impact than the medium of speech
itself, transformed into art in literature.
Thus “The art of poetry . . . claims the highest rank of all,” because it can most fully
exhibit aesthetic ideas and thus most fully expand

the mind by setting the imagination free and presenting, within the limits of a
given concept and among the unbounded manifold of forms possibly agreeing
with it, the one that connects the presentation with a fullness of thought to
which no linguistic expression is fully adequate.

Drawing perhaps on the Affektenlehre of Johann Mattheson15 (there is no evidence that he

ever read his former student Herder’s fourth Grove, which was not published until 1846),
Kant conceded that music was very good at stimulating emotions and communicating the
affect associated with an idea, but not at communicating the idea itself, and thus offers
“more enjoyment than culture” (ibid., §53)—for as he had argued in the “Analytic of the
Beautiful,” the arousal of emotion is not an important part of our experience of art. And
about painting, Kant says only that “it can penetrate much further into the region of ideas”
than the other “pictorial” or visual arts, but he does not even bother to compare its impor-
tance to poetry or music: the outcome of that comparison should be obvious.
Kant does recognize that poetry can be combined with music in “song” and “this, in
turn, with a painterly (theatrical) representation in an opera” (ibid., §52), so that the
impact, both cognitive and emotional, of several art forms combined can be greater than
that of any one, even poetry, alone—“can,” that is, not “must,” because for Kant, like
other eighteenth-century aestheticians, there can be no hard-and-fast rules for judg-
ments of taste, and a great poem by itself might have more impact than the combination
of lesser work from multiple arts. A Gesamtkunstwerk is not automatically better than a
work in one artistic medium alone, though it might be contingently so. Contingency is,
after all, especially at home in the realm of the aesthetic. But leaving the possibility of
Gesamtkunstwerk aside, it is clear that the prominence of both painting and poetry in
Kant’s initial exposition of the formalism of pure or free beauty (ibid., §§14, 16) is a
misleading artifact of Kant’s order of exposition, and that when it comes to the more
complex case of fine art proper, there is no other medium of art that can, by itself, hold
a candle up to the expressive power of poetry.


  1 For my interpretation of Kant’s account of free play, see especially Guyer 2005.
  2 For a more detailed account, see Guyer 1997, chapter 12.
  3 See Jones 1982.
  4 This chapter was added in the fourth edition of Locke’s Essay Concerning Human Understanding, thus in
1700. It is fitting that the concept of the association of ideas, which would have so much influence
throughout the eighteenth century, itself first appeared in (what at least some would count as) the first
year of that century.
  5 Burke’s theory of invigoration or relaxation by the beautiful and sublime would be transformed into a
theory of two types of beauty, “energizing” and “melting,” by Friedrich Schiller; see Schiller 1967,
Sixteenth and Seventeenth Letters.
  6 The theme of the sublime became particularly prominent in the eighteenth century through the ancient
treatise Peri Hypsous or “On the Sublime” attributed to Longinus, which was widely available in both
French and English translations by the beginning of the century. Pseudo-Longinus was particularly
concerned with the sublime style in poetry, so the sublime could well seem to be an important topic in a
discussion of eighteenth-century philosophy of literature. But since most authors turned the sublime into
a quality of nature or our response to it, culminating with Kant who held, at least in the Critique of the
Power of Judgment, that only nature could induce an experience of the sublime (see Kant 2000, §26), the
limited space available here will not be used for a further discussion of the sublime. For such discussion,
see among many other works, Hipple 1957, Monk 1960, and Kirwan 2005.
  7 It should be noticed here that there is considerable overlap between this list of objects that mere mention
of which is supposed to have such a strong emotional impact on us and Kant’s list of the paradigmatic
contents of aesthetic ideas, the emotional impact of which, however, Kant fails to mention.
  8 In contemporary aesthetics, the chief critic of this view has been Kendall Walton, who argues that fiction
can produce what he calls “quasi-emotions,” that is, some of the usual physiological concomitants of
emotion, on the basis of which subjects can then “make believe” or play that they are having the relevant
emotions (see Walton 1990). This view does not seem to explain the conviction with which most people
describe the emotional impact of art or the importance they attach to it.
  9 And for commentary, see Dickie 1995, chapter 3, and Kivy 2003, chapter XI.
10 Actually, what Hume argues is that there is no rational basis for belief in the immortality of the soul, so
it can be supported only by revelation (Hume 1987, p. 598). But given his arguments against revelation
elsewhere, e.g. the famous chapter “Of Miracles” in the Enquiry Concerning Human Understanding (Hume
2000, Section 10), this could only have been intended ironically.
11 Alex Neill has defended the ascription of the “conversion theory” to Hume while criticizing the theory
itself; see Neill 1992.
12 Mendelssohn also developed his views about tragedy in a series of epistolary exchanges with his friends
Gotthold Ephraim Lessing and Friedrich Nicolai, which unfortunately have not been translated into
English; for the German texts, see Lessing/Mendelssohn/Nicolai 1972. Mendelssohn’s theory of tragedy
is discussed in Beiser 2009, chapter 7, and Pollok 2010, chapter II.3.
13 The comparisons between literature, painting, and music are discussed in Lipking 1970 (with primary
reference to Britain) and by David Marshall and Dean Mace in Nisbet and Rawson 1997, chapter 29.
14 On Lessing’s Laocoön, see Wellbery 1984 as well as Beiser 2009, chapter 8.
15 On Mattheson, see Kivy 1980, chapter V, Kivy 1984, and Kivy 2001, pp. 51–3.

Primary Sources
Alison, Archibald (1811). Essays on the Nature and Principles of Taste. Second edition. 2 vols. Edinburgh:
Bell & Bradfute.
Baumgarten, Alexander Gottlieb (1954). Reflections on Poetry: A.G. Baumgarten’s Meditationes philosophicae
de nonnullis ad poema pertinentibus. Latin text and English translation by Karl Aschenbrenner and
W.B. Holther. Berkeley and Los Angeles: University of California Press.
Boileau-Despréaux, Nicolas (1965). Selected Criticism. Translated by Ernest Dilworth. Indianpolis:
Bobbs-Merrill, 1965.
Burke, Edmund (1958). A Philosophical Enquiry into the Origin of our Ideas of the Sublime and Beautiful. Edited
by James T. Boulton. London: Routledge & Kegan Paul. (First edition 1757, second edition 1759).

Paul Guyer

Du Bos, Jean-Baptiste (1748). Critical Reflections on Poetry, Painting, and Music. Translated by Thomas
Nugent. 3 vols. London: John Nourse.
Harris, James (1801). The Works of James Harris, Esq., with an Account of his Life and Character by his Son, the
Earl of Malmesbury. 2 vols. London: F. Wingrave.
Herder, Johann Gottfried (2002). Sculpture: Some Observations on Shape and Form from Pygmalion’s Creative
Dream. Translated by Jason Gaiger. Chicago: University of Chicago Press. (German original 1778).
Herder, Johann Gottfried (2006). Selected Writings on Aesthetics. Translated and edited by Gregory Moore.
Princeton: Princeton University Press.
Hume, David (1987). Essays Moral, Political, and Literary. Edited by Eugene F. Miller. Revised edition.
Indianapolis: Liberty Fund.
Hume, David (2000). An Enquiry Concerning Human Understanding. Edited by Tom L. Beauchamp. Oxford:
Clarendon Press.
Kames, Henry Home, Lord (2005a). Elements of Criticism. Edited by Peter Jones. 2 vols. Indianapolis: Liberty
Fund. (Based on sixth edition, 1785).
Kames, Henry Home, Lord (2005b). Essays on the Principles of Morality and Natural Religion. Edited by Mary
Catherine Moran. Indianapolis: Liberty Fund.
Kant, Immanuel (2000). Critique of the Power of Judgment. Edited by Paul Guyer, translated by Paul Guyer
and Eric Matthews. Cambridge: Cambridge University Press. (First edition 1790, second edition 1793).
Leibniz, Gottfried Wilhelm (1969). Philosophical Papers and Letters. Edited by Leroy E. Loemker. Second
edition. Dordrecht: D. Reidel.
Lessing, Gotthold Ephraim (1984). Laocoön: An Essay on the Limits of Painting and Poetry. Translated by
Edward Allen McCormick, with a foreword by Michael Fried. Baltimore: The Johns Hopkins University
Press. (Original German edition 1766).
Lessing, Gotthold Ephraim, Moses Mendelssohn, and Friedrich Nicolai (1972). Briefwechsel über das
Trauerspiel. Edited by Jochen Schulte-Sasse. Munich: Winkler Verlag.
Locke, John (1975). An Essay Concerning Human Understanding. Edited by P.H. Nidditch. Oxford: Clarendon
Longinus (1965). On the Sublime. In T.S. Dorsch, editor. Classical Literary Criticism. Harmondsworth: Penguin.
Mendelssohn, Moses (1997). Philosophical Writings. Edited by Daniel O. Dahlstrom. Cambridge: Cambridge
University Press.
Schiller, Friedrich (1967). On the Aesthetic Education of Man in a Series of Letters. Translated by Elizabeth M.
Wilkinson and L.A. Willoughby. Oxford: Clarendon Press.
Winckelmann, Johannn Joachim. Reflections on the Painting and Sculpture of the Greeks: With Instructions for
the Connoisseur, and an Essay on Grace in Works of Art. Translated by Henry Fusseli. London: A. Millar,
1765. Facsimile edition in Winckelmann, Essays on the Philosophy and History of Art, edited by Curtis
Bowman. Vol. I. Bristol: Thoemmes Press, 2001.
Wolff, Christian (1719). Vernünftige Gedancken von Gott, der Welt, und der Seele des Menschen. Halle: Renger;
German text republished in Christian Wolff, Metapfisica Tedesca, edited by Raffaele Ciafardone. Milan:
Bompiani, 2003.

Secondary Sources
Beiser, Frederick C. (2009). Diotima’s Children: German Aesthetic Rationalism from Leibniz to Lessing. Oxford;
Oxford University Press.
Dickie, George (1995). The Century of Taste: The Philosophical Odyssey of Taste in the Eighteenth Century.
New York: Oxford University Press.
Guyer, Paul (1997). Kant and the Claims of Taste. Second edition. Cambridge: Cambridge University Press.
Guyer, Paul (2005). “The Harmony of the Faculties Revisited.” In Guyer, Values of Beauty: Historical Essays
in Aesthetics. Cambridge: Cambridge University Press.
Hipple, Walter J. (1957). The Beautiful, the Sublime, and the Picturesque in Eighteenth-Century British Aesthetic
Theory. Carbondale: Southern Illinois University Press.
Jones, Peter (1982). Hume’s Sentiments: Their Ciceronian and French Context. Edinburgh: Edinburgh University
Kirwan, James (2005). Sublimity. London: Routledge.
Kivy, Peter (1980). The Corded Shell: Reflections on Musical Experience. Princeton: Princeton University Press.
Kivy, Peter (1984). “Mattheson as philosopher of art.” Musical Quarterly 70: 248–65.


Kivy, Peter (2001). New Essays on Musical Understanding. Oxford: Clarendon Press.
Kivy, Peter (2003). The Seventh Sense: Francis Hutcheson and Eighteenth-Century British Aesthetics. Second
edition. Oxford: Clarendon Press.
Lipking, Lawrence I. (1970). The Ordering of the Arts in Eighteenth-Century England. Princeton: Princeton
University Press.
Monk, Samuel (1960). The Sublime: A Study of Critical Theories in XVIII-Century England. Second edition.
Ann Arbor: University of Michigan Press.
Neill, Alex (1992). “Yanal and Others on Hume on Tragedy.” Journal of Aesthetics and Art Criticism 50: 151–4.
Nisbet, H.B. and Clyde Rawson, editors (1997). The Cambridge History of Literary Criticism, Volume IV: The
Eighteenth Century. Cambridge: Cambridge University Press.
Pollok, Anne (2010). Facetten des Menschen: Zur Anthropologie Moses Mendelssohns. Hamburg: Felix Meiner
Walton, Kendall L. (1990). Mimesis and Make-Believe: On the Foundations of the Representational Arts.
Cambridge, MA: Harvard University Press.
Wellbery, David E. (1984). Lessing’s Laocoon: Semiotics and Aesthetics in the Age of Reason. Cambridge:
Cambridge University Press.

Allen Speight

The nineteenth century marked the emergence of literature’s philosophical importance

in several crucial new ways. In the wake of post-Kantian developments in both romantic
and Idealist thought occurring at the end of the eighteenth century, literature emerged
(in the words of Hegel and others) as “art itself,” “art par excellence”—not simply one
genre of art among many (or, for that matter a form of discourse removed from
comparison with other artistic genres), but the epitome of artistic experience. Perhaps
even more strikingly, given the literary output of the century, these new philosophical
construals of literature more than met their match—were in fact in many ways overtaken
by the artistic development of new forms which owed their origins to the eighteenth
century but their development to the nineteenth. In particular, the novel, which until
the late eighteenth century had been written in a primarily epistolary form, came to
acquire the sprawling and narrative form associated with many of the great nineteenth-
century Russian, French, and English novels and raised new questions about the relation
between literature and a rapidly changing social and political world.
In what follows, these developments will be traced in terms of three large consider-
ations. In the first section, I will explore the general question of what overarching
philosophical frameworks were relevant for the construal of literature over the course of
the nineteenth century and to what extent these frameworks still have relevance for
contemporary philosophy of literature. Two of these approaches—lyric expressivism and
the conception of the objectivity or independence of the literary artwork—will be of
particular interest. In the second section, I will take up the status of literature within
the shifting contours of the nineteenth century’s accounts of aesthetics and artistic
genres—in particular, how the reflection of Schopenhauer and Nietzsche on the non-
mimetic character of music had broader consequences for the philosophy of art and
literature. Finally, in the third section, I explore in light of these considerations of
philosophical and aesthetic approaches to literature a tension that becomes especially
important over the course of the century between literature’s internal importance as a
work of art and its relation to the world—the question of whether and to what extent
art is engaged with social and cultural realities, for example, or is concerned with the
creation of something independent from them.

I Two Nineteenth Century Philosophical Frameworks

for Considering Literature
It is clear that a consideration of literature was an important issue for many nineteenth
century philosophers, from Hegel to Mill to Kierkegaard to Nietzsche. The roots of this
new philosophical interest in literature can be traced in many ways back to currents
within late eighteenth-century literature and philosophy. As Dieter Henrich has
noticed, for example, there was a remarkable explosion in German-language literary
production—both the publication of new literary works and philosophical and critical
reflections on them—in the last two decades of the eighteenth century (roughly following
the time of the publication of the Critique of Pure Reason; see Henrich 2003, 74). In this
section, I will argue that there were two especially important and opposing approaches
to the philosophical construal of literature that developed in the nineteenth century
that have remained important for contemporary discussion: the first, lyric expressivism,
which had a loosely romantic origin, and the second the objectivity or independence of
the literary artwork, which had somewhat more emphatically Idealist roots, although
some heritage among the early German romantics, as well. (This issue of the artwork’s
independence is distinct from the related and much-discussed issue of the autonomy of
literature itself.)
Before we examine these two approaches, it should be noted that the prevailing terms
of philosophical discourse about literature in the nineteenth century were frequently
different from those used in contemporary discussions in the field: not only had the
academic sub-discipline known as “philosophy of literature” not come into being as
such, but “literature” had not yet come to be widely used as a broad term referring to the
category of “significant imaginative or aesthetic writing in general” including both
poetry and the novel. The primary eighteenth-century meaning of “literature” was still,
as Johnson’s Dictionary had defined it, “learning; skill in letters” (see Speight 2013b);
according to the Oxford English Dictionary, the first indications of the use of the word in
general to mean “written work valued for superior or lasting artistic merit” without
reference to a specific national or technical literature such as “German” or “scientific”
literature, come in the mid-nineteenth century, and the strongest evaluative uses of it
to distinguish writing that matters aesthetically or ethically are from twentieth-century
writers. Most of the philosophical writers from the nineteenth century to be considered
in what follows mark off their territory of philosophical interest in this area not as
“literature” but rather as “poetry”—and what is serious in prose fiction is often treated
by them on a somewhat ad hoc basis, not always neatly fitting into the larger aesthetic
structures within which they work.
Whether this difference in usage matters only historically or philosophically is a good
question. Peter Lamarque is certainly right to say (in the context of a discussion of the
institutional theory of literature) that we should not confuse the history of a word with
the history of a concept (Lamarque 2009, 64). There are, however, I would argue, both
linguistic as well as conceptual questions relevant in this context as one looks at
nineteenth-century contributions to the philosophy of literature.
Philosophically, one question of particular relevance that might be asked is whether
a general philosophical approach to literature construed around poetry (as are those of
many of our nineteenth-century philosophers) does in fact pose different conceptual
issues than one framed in terms of literature (as is the case with most contemporary
discussions in the philosophy of literature). One obvious difference that a contemporary

Allen Speight

philosopher would remark between the two construals is that the poetry-centered
nineteenth-century discussions seem to focus perhaps more immediately on questions of
form and origin (what account can we give of what makes a poem come to be, as opposed
to “ordinary” prose?) rather than, say, the contemporary question of value (what makes
something significant enough—ethically or otherwise—to be considered literature?).
These are, of course, not exclusive questions: as we will see shortly, the nineteenth-
century concern with poetry also—especially in Hegel—never leaves aside the question
of value, (especially, as it turns out, the value of truth), and the literary perspective of
contemporary philosophers of literature is certainly not unconcerned with issues of form
and origin. The difference, however, is one of emphasis, and it is a question, as it turns
out, which will prove illuminating not only for the issue we are interested in at the
moment—under what general philosophical framework should works of literature be
viewed?—but also for the other two central questions of our discussion, concerning
aesthetic genre theories and the relation of literature to the world.
Let us begin with two famous yet quite different discussions of poetry—and its antith-
eses—that stem from the first third of the nineteenth century: those of Hegel and Mill.
The contrast that interests Hegel in his discussion in the Lectures on Fine Arts is one
between poetry and what he calls, in a memorable phrase, “the prose of the world.” Poetry,
in Hegel’s view, “is the original presentation of the truth,” which does not divide its
subject matter as prose does but rather presents it “as a totality complete in itself and
therefore independent” (Hegel 1975, 973). The real origin of poetry is thus not to be
found in considerations of diction, rhyme, or rhythm, but in “the form which the way
of imagining things [Vorstellen] must take in order to be expressed poetically” (ibid., 1000).
Hegel gives as an example of poetry’s independence and distinctness from its prosaic
surroundings the elegiac distich commemorating the Greeks who fell at Thermopylae,
as reported by Herodotus: “Here four thousand from the Peloponnese fought against three
myriads.” Such a commemorative expression is “of such a high dignity that it tries to
distinguish itself from any other mode of speech,” and thus poetry can be said to have
“the vocation of being a sphere of its own” (ibid., 974). Hegel’s point here is worth
contrasting with many views of the poetic (or the literary, for that matter): it’s not that
the Thermopylae distich is “fine writing” and therefore poetic, but rather that it some-
how stands out with some independence from other writing. Within the sphere of
independence visible even in a distich, a work of poetry has an organic unity which “gives
in its parts the appearance of close connection and coherence and, in contrast to the
world of mutual dependence, stands there for its own sake and free on its own account”
(ibid., 965); the whole which it articulates (even in the relatively short compass of a
distich, but obviously more so over the course of an entire epic):

may be rich and may have a vast range of relations, individuals, actions, events,
feelings, sorts of ideas, but poetry must display this vast complex as perfect in
itself, as produced and animated by the single principle which is manifested
externally in this or that individual detail.
(Ibid., 973)

Poetry thus must avoid merely retailing the endless particular details that prose by its
nature cannot satisfactorily unify (or so claimed Hegel, writing before the appearance
of many of the great nineteenth-century novels). But poetry must likewise avoid attempt-
ing to unify details under the abstract universals found in philosophy or scientific


thinking: it must instead present them within a unity that is “animated, manifest,
ensouled, determining the whole, and yet at the same time expressed in such a way that
the all-comprising unity, the real animating soul, is made to work only in secret from
within outwards” (ibid., 973).
Mill’s discussion of poetry in his two well-known essays (“What Is Poetry?” and “The
Two Kinds of Poetry”) begins with Wordsworth’s contrasting of poetry with its “logical
opposite”—in this case, not prose in the broad sense, but elements that Hegel would
consider central to the world’s prosaicness: the realm of fact and science (Mill 1833).
But Mill then moves to differentiate poetry from fiction, narrative, and the novel more
broadly in terms of their origin. An interesting difference emerges: in Mill’s view, poetry
derives at its depth from a concern with heart and feeling, whereas fiction derives from a
concern with incident (the details of stories—from the latest gossip to travel narratives
to novels—which can attract a reader’s attention. Mill acknowledges that novels can
have a poetry to them—and not merely be in verse—but maintains the underlying
contrast in this distinction between concern for feeling and concern for incident.)
The different interests in poetry taken by Hegel and Mill generate differing histories
of literature, both on the level of society and the individual character that is drawn to
literature: for Hegel, phylogenetically speaking, poetry is somehow more original or
primal within the development of the human race. For Mill, both ontogenetically and
phylogenetically—i.e. both in one’s own individual childhood and in the childhood
of the human race—narrative prose, thirst for gossipy details and incidents, comes
first, whereas poetry comes later; these are for him above all the result of different
kinds of character.
The differing approaches that Hegel and Mill take to poetry correlate closely with
the final two of the four general approaches to literature outlined in M. H. Abrams’
well-known typology of critical approaches to literature and art. As Abrams presents it,
critical approaches can be either mimetic (deriving ultimately from Platonic and
Aristotelian understandings of art as primarily a form of representation or imitation);
pragmatic (concerned with the effect a work of art or literature has on a reader or spectator);
expressive (concerned with the romantic impulse of the poet to express himself above all); or
objective/autonomous (construing works of art and art itself as having no purpose but its
own creation). It is not surprising that the two of Abrams’ approaches that we have
associated with Hegel and Mill were perhaps most important in the nineteenth century.
The mimetic tradition had been roundly attacked in the eighteenth century and left (by
many) for dead (although, as Halliwell among others have argued, that tradition has
persistent elements that continue even into the twentieth century [see Halliwell 2002,
as well as his contributions to this volume]). Meanwhile the pragmatic tradition’s
prominent eighteenth-century exponents had in turn been criticized especially by the
post-Kantian Idealists who turned away in their philosophies of art from the Kantian
and pre-Kantian discussion of aesthetics and aesthetic reaction. But both the expressivism
underlying Mill’s approach and the objectivity of the artwork articulated by Hegel
remain important across the nineteenth and into the twentieth century. Mill’s view is
well represented across the nineteenth century in a romantic strain that can be said to
run from Wordsworth and the English romantics at the beginning of the century to such
different figures as Tennyson, Carlyle, Wagner, and Nietzsche. Hegel’s understanding of
the work of art or literature as independent or objective has further articulation in his
fellow Idealist Schelling (and to some extent as well in the early German romantic
Friedrich Schlegel); this approach suffers perhaps more diminution over the nineteenth

Allen Speight

century than does the lyric, but it nonetheless remains an important approach up
through the early and mid-twentieth century in figures like A. C. Bradley, T. S. Eliot,
and the American New Critics.
Both the romantic and the Idealist traditions are decisive as well in the nineteenth
century’s approach to literature in the context of aesthetic theory more broadly, which
is the topic of the next section. On the Idealist side, as we will see, one of the prime
motivations behind considering works of art as independent lies in a broader aesthetic
commitment that is implicit in their emphasis on the objectivity of the art work: a
commitment to the unity of form and content in art more generally. The systematic genre
theory motivated by this commitment, however, was challenged by new approaches to
genre stemming from an essentially lyric impetus, as Schopenhauer, Nietzsche, and
others discovered newfound importance in music. Yet, as will be seen, neither approach
still engaged fully such rapidly developing nineteenth-century genres as the novel.

II Literature and Artistic Genre: Changing Nineteenth

Century Approaches
Although certainly not all literary critics have approached their field with an eye for
how literature fits within a larger theory of art, recent work within analytic philosophy
of literature has placed a useful emphasis on doing so—and hence on seeing how litera-
ture as a distinct genre connects with other artistic genres in a broader theory of
aesthetics (Lamarque 2009, 12–16). When we look at literature from the perspective of
broader aesthetic theory in the nineteenth century, we see a division of approaches that
in many ways follows loosely (though with some qualifications) the lyric and Idealist
models we saw in the first section.
Most of the prominent treatments of genre in the early part of the nineteenth
century still worked within a set of assumptions that had largely been inherited from
the eighteenth century. The classification of the fine arts (the beaux arts or the schöne
Künste) as a group distinct from the useful or mechanical arts—and including especially
the five arts of architecture, sculpture, painting, music, and poetry—had reached its
canonical form in the mid-eighteenth century in Batteux and others, and represented,
at least according to the view of Kristeller, a decisive “modern” break from the ancient
Greek and Roman conceptions of the arts (see Kristeller 1951–2, as well as recent
debate about his claims concerning the importance of these eighteenth-century devel-
opments in Kivy 2012 and Porter 2002). At the same time, central questions about how
the arts within this broader genre classification relate to one another were influenced
by important distinctions such as that of Lessing’s between the visual arts and poetry
(Lessing 1984).
If we look at the most prominent philosophically-inspired theories of aesthetics at
the start of the nineteenth century—for example, the extensive and influential lecture
series on aesthetics given by the Idealists Hegel and Schelling as well as those of the
critic A. W. Schlegel—all sketch a view of the individual artistic genres that draws on
these eighteenth-century views with impressive similarity. In each of the three lecture
series, for example, it is poetry that is placed at the height of the scheme of genres (as
somehow offering a unification or reconciliation of the visual and musical arts) and
drama that is regarded as the supreme form of poetry (as offering a recapitulated version
of the same unification within poetry itself, bringing together both lyric and the visual
qualities of epic).


The Idealist account of this genre scheme follows in important ways on a commitment
that underlies the emphasis we have seen on an artwork’s unity and objectivity: a
commitment to the unity of form and content in aesthetics more generally (Zuckert 2010).
Hegel and Schelling (despite their differences) both saw this unity as most visible, for
example, in the representation of the Greek gods, who were both universal and particular
at once (for the Greeks, they claimed, Athena did not simply mean or stand for divine
wisdom but in some sense was divine wisdom itself as a particular god). Hegel’s notion
of classical beauty is construed with this unity of the Greek gods in mind in terms of a
union of universal meaning (Bedeutung) and particular shape (Gestalt) according to
which each of the two must harmoniously “interpenetrate” the other—something most
fully visible in the anthropomorphic sculptures of the Greek deities. And such a
commitment to the unity of form and content makes clear the Idealist need to work out
a genuinely systematic account of the genres: specific artistic media could not to be
regarded somehow just as natural kinds but rather must be construed in terms of a more
thoroughly unified approach to art.
Within this account of the genres, it is above all drama that achieves the highest
unity of form and content, according to both Hegel and Schelling. Thus Schelling in
his Lectures on the Philosophy of Art calls drama “the highest manifestation of the nature
and essence of all art” (Schelling 1989, 247) and Hegel valorizes drama because of its
ability to represent above all human action. While all the artistic genres are somehow
concerned with the adequate (formal) representation (of the essential content) of the
human, it is drama, Hegel says, in which “the whole man presents, by reproducing it, the
work of art produced by man,” thus rendering the whole man “fully alive” onstage,
“made into an animated work of art” (Hegel 1975, 627, 955). Drama as a kind of “living
sculpture” thus most fully unifies aesthetic form and content.
Hegel’s praise of drama here is, however, not one which can be read as free of
tension—if only because his engagement with the question of unity in art’s high human
task as a whole must include some consideration of that side of his aesthetics that is
often taken up in terms of his supposed “end of art” thesis—that in the modern (post-
romantic) world “art, considered in its highest vocation, is and remains for us a thing of
the past” (ibid., 11). A correct construal of this side of Hegel’s view of literature would
notice that the obverse side of Hegel’s having a sculptural ideal of tragedy is that he has
a tragic ideal of sculpture—that is to say that the unity of form and content captured by
the ancient Greeks in sculptural renderings of anthropomorphic divinity is a unity
which we in the modern world experience necessarily retrospectively, from a distance at
which we are always aware of the downfall of the civilization that gave us the beauty of
Greek art (Speight 2013a). From this perspective, Hegel’s own appeal to the lyric as that
literary mode which might best voice aspects of modernity (as opposed to ancient beauty)
offers a presentiment of the decline over the nineteenth century of the general project
of looking to art to find the adequate rendering of human significance in human shape.
So it is of course true enough that the broadly holistic approach to works of art that
stems from Hegel’s Idealist commitments retains an importance for many later philo-
sophical views of aesthetics stretching into the twentieth century—and not only for
those, like A. C. Bradley (Bradley 1901), who worked within an essentially Idealist
frame of reference but also for others like Arthur Danto, for whom the notion of embod-
ied meaning became central for other reasons (Danto 1997). But—as the tension in
Hegel’s account of artistic modes suggests—the Idealist approach to genre had already
begun to fade in crucial ways by the middle of the nineteenth century and was

Allen Speight

challenged in particular by the new stress from the lyrical side to be found in the conflict
between music and the rest of the arts in Schopenhauer and Nietzsche.
If the eighteenth-century views of aesthetics the Idealists had drawn upon had
particularly compared poetry as a genre with painting and the visual arts, culminating
in the Idealist sculptural view of tragedy, it is more typical of the later nineteenth
century—on the other side of the romantic movement and the critique of mimetic
poetry—to view poetry as having what Abrams calls its “most profound affinity” with
music (Abrams 1953, 50). The emphasis in the title of the first edition of Nietzsche’s
first book, The Birth of Tragedy out of the Spirit of Music, is the result not only of
Schopenhauer’s appeal to music’s non-mimetic character, but also to a radical way of
re-envisioning what the tragic spectator experiences. Tragic drama may still bring
together the lyrical and the epic as it did in the systematic aesthetic theories of Hegel,
Schelling, and Schlegel, but Nietzsche’s revaluation of aesthetics in terms of the
Apollinian–Dionysian opposition construes the grounds for such unity in tragedy in a
different way. A remark of Nietzsche’s in The Case of Wagner makes clear why the Idealist
view of tragic action as unifying the spectator’s experience of the drama needs to be
replaced with the more lyrical experience of pathos or feeling:

It has been a real misfortune for aesthetics that the word drama has always been
translated ‘action’ [Handlung]. It is not Wagner alone who errs at this point, the
error is world-wide and extends even to philologists who ought to know better.
Ancient drama aimed at scenes of great pathos—it precluded action (moving it
before the beginning or behind the scene).
(Nietzsche 1968)

But the lyric expressivist strain behind Wagnerian opera and Nietzsche’s appeal to the
Dionysian are not the only aesthetic forces which represent the needs for potential shifts
within nineteenth-century accounts of literature. Philosophical attention to how the
great nineteenth-century novels should be thought about in terms of genre theory was,
for example, scant. Partly this is because systematic genre theory itself was in real disar-
ray by the later nineteenth century; in any case, the boldest new theoretical attempts to
construe, for example, the nineteenth-century Russian novel were made in the early
twentieth century (by, e.g. Lukács and Bakhtin).

III Literature, Philosophy and the World: Nineteenth-Century

Influences in Contemporary Debates?
Given the important nineteenth-century shifts we have seen in philosophical conversa-
tions both about the status of literature and its place within aesthetic genre theory, what
assessment can we make about the role of these approaches from the perspective of
contemporary debates about the philosophy of literature? Are there elements of the
nineteenth-century debate about these topics that have continued to persist with influence
in discussion in our own day, or has philosophy of literature simply moved past the con-
siderations of greatest importance to nineteenth-century thinkers?
The landscape of the contemporary debate—both in current literary theory and in
analytic philosophy of literature—certainly seems to be cast in quite different terms. In
literary theory, the most recognizably Idealist cousin left in twentieth-century theory,
New Criticism, was largely supplanted mid-century by successive waves of structuralist,


post-structuralist, and ideologically grounded (Freudian, Marxian, feminist) approaches

to theory, which have in their turn by now each gone past a certain heyday. For a
moment, there was a declaration—by some parties, at least—of an “end of theory,” with
a new interest in the writing of memoir and the explicitly autobiographical satisfactions
of individual critics in their own reading. If the social and cultural forms of ideological
criticism have come to be seen by some as outmoded, the newest theoretical turns after
the end of the heyday of theory have sought a more naturalized home for literary theory
in contemporary work in evolutionary biology—a project which, of course, bears a
proper nineteenth-century point of origin in Darwin, although the author of the Origin
of Species appears to have written little about literature, and contemporary theorists have
charted somewhat different lines connecting evolutionary and literary theory than
Darwin himself may have contemplated (see, among others, Boyd 2009).
In analytic philosophy of literature, the landscape has been well described in terms
largely of reactions to Frege’s insistence that in listening to poetry we are “interested
only in the sense of the sentences and the images and feelings thereby aroused,” not the
question of truth (Frege 1970). For many contemporary philosophers of literature—
including Lamarque and Olsen (1994), Walton (1990), and Moran (1994)—accepting
Frege’s claim means that we may of course be concerned with compelling aesthetic and
imaginative issues in our engagement with literature—may even, as Lamarque and
Olsen put it, look to literature for some kind of insight or understanding—but cannot
hope to find in it knowledge of what our world is actually like. If literature concerns
questions of truth, these must refer rather to fictional worlds (in Walton’s formulation,
for example, fictional representations serve precisely as props in “games of make-believe”).
In light of the contemporary contour of these conversations, what if anything is left
of the nineteenth-century approaches that might offer resources for current work in the
philosophy of literature? Can either of the general nineteenth-century approaches to
literature we have explored here, for example—given their generally pre-Fregean,
pre-Marxist, pre-Freudian commitments—still have an influence in contemporary
philosophical thinking about literature?
I would argue that both of these nineteenth-century approaches still have interesting
contributions to make within the contemporary context. The Idealists’ approach to the
literary work of art offers, for example, a way of preserving what Gibson calls the
“humanist intuition”—that literature can present the reader with an “intimate and
intellectually significant engagement with social and cultural reality” (Gibson 2007)—
while at the same time not losing a second intuition—that literature can present other
worlds of aesthetic creation rather than this-worldly representation. While Hegel does
not frame the issue precisely in terms of these two competing intuitions, or concern
himself (as the post-Fregean discussion has) particularly with the role of fiction as
central to the debate, his account of the form and origin of poetry nonetheless engages
directly the question of literature’s relation to worldliness in a way which might be
contemporarily productive.
As we have seen, Hegel does praise the ability of the poetic imagination to sketch a
world that has its own independence and coherence over against the prosaic reality of
our this-worldly lives—and it is precisely this essentially creative function of the imagi-
nation that makes him think of poetry as the touchstone for art as a whole: imaginative
poetry is thus not only the highest artistic genre, as we have seen, but in fact that genre
which Hegel thinks can clarify what makes the other genres forms of art themselves
(Hegel 1975, 967). Yet however strong Hegel’s praise of the independent and creative

Allen Speight

imagination, he makes clear that two important facets of art’s historicity and cultural
valence need to be part of the story as well: that (on the side of content) literature’s
significance in terms of what Hegel calls the “deepest interests of mankind, and the most
comprehensive truths of the spirit” (ibid., 7) is something that has in fact changed over
time and (on the side of form) that literary artworks have in fact been useful at crucial
moments in human development in helping prompt the discovery of the inadequacy of
certain forms of life (ibid., 967). In both functions, literature is of course not to be
understood in mimetic terms as representing the world in some way but rather as having
a more constitutive role, in showing us what it could mean to espouse or avow the norms
that matter for us in modern life (Pippin 2010). We might note, further, in the wake of
the post-ideological age of literary criticism, that Hegel’s overall approach to how social
and cultural reality is figured in literature—given his commitment to the unity of form
and content—might allow a stance in which contemporary critics may avoid the separation
of form and content often associated with the attempt to strip some ideological “meaning”
from a work’s aesthetic fabric.
From the lyric expressivist strain in nineteenth-century philosophical engagement
with literature, one might expect—following, for example, Rorty’s (1989) contrast
between opposing philosophical tendencies toward perfection and community—less of
an engagement with social and cultural realities than perhaps a more perfectionist
exploration of the freedom and creativity of imagination. Yet here too we might notice
important features of our shared way of life in the modern world that remain important
from the perspective of this philosophical stance toward literature. It might be argued,
for example, that it is only after Nietzsche’s engagement with tragedy that certain literary
terms have broken away from the narrower consideration of art and its genres to have a
larger social and cultural impact, as can be glimpsed from the various appeals that have
been made in a post-Nietzschean context to “the tragic” as a phenomenon in a direction
that both Scheler (1954) and Szondi (2002), in different ways, for example, have also
articulated, or, for that matter, to the “Dionysian” itself (Porter 2000). From the vantage
point of peculiarly twentieth-century horrors in social and cultural experience, however,
it is perhaps the more plaintive and persistent, distinctively private lyric voice—in
figures as different as Celan, Kafka, Rilke, and Beckett—that may continue to offer some
of the most provocative re-considerations of literature’s connection to the world.
Whether these lyric provocations are best heard, however, in the context of a largely
Nietzschean approach to literature or in the context of that strain of lyric modernism
that emerges from Hegel’s reflections on the “end of art” is a question that must be left
for another occasion.

Abrams, M. H. (1953), The Mirror and the Lamp: Romantic Theory and the Critical Tradition. Oxford: Oxford
University Press.
Bakhtin, M. M. (1984), Problems of Dostoevsky’s Poetics, trans. Caryl Emerson. Minneapolis: University of
Minnesota Press.
Boyd, B. (2009), On the Origin of Stories: Evolution, Cognition and Fiction. Cambridge: Harvard University
Boyle, Nicholas (2010), “Goethe’s Theory of Tragedy,” The Modern Language Review, 104, 5 (October),
Bradley, A. C. (1901), Poetry for Poetry’s Sake. Oxford: Clarendon Press.
Danto, A. (1997), After the End of Art: Contemporary Art and the Pale of History. Princeton: Princeton
University Press.


Eldridge, R. (forthcoming), “The Question of Truth in Literature: Die poetische Auffassung der Welt.”
Frege, G. (1970), “On Sense and Reference,” in Philosophical Writings of Gottlob Frege, trans. and ed. Peter
Geach and Max Black. Oxford: Oxford University Press.
Gibson, J. (2007), Fiction and the Weave of Life. Oxford: Oxford University Press.
Halliwell, S. (2002), The Aesthetics of Mimesis: Ancient Texts and Modern Problems. Princeton: Princeton
University Press.
Hegel, G. W. F. (1975), Aesthetics: Lectures on Fine Art, trans. T. M. Knox. Oxford: Oxford University Press.
Henrich, D. (2003), Between Kant and Hegel: Lectures on German Idealism, David S. Pacini, ed. Cambridge:
Harvard University Press.
Kivy, P. (2012), “What Really Happened in the Eighteenth Century: The ‘Modern System’ Re-examined
(Again),” British Journal of Aesthetics, 52, 1 (January), 61–74.
Kristeller, P. O. (1951–1952), “The Modern System of the Arts: A Study in the History of Aesthetics,”
Journal of the History of Ideas, 12, 4 (October 1951), 496–527, and 13, 1 (January 1952), 17–46.
Lamarque, P. (2009), The Philosophy of Literature. Oxford: Blackwell.
Lamarque, P. and Stein Haugom Olsen (1994), Truth Fiction and Literature: A Philosophical Perspective. Oxford:
Clarendon Press
Lessing, G. E. (1984), Laocoön: An Essay on the Limits of Painting and Poetry, trans. Edward Allen McCormick.
Baltimore: Johns Hopkins University Press.
Lukács, G. (1971), Theory of the Novel: A Historico-Philosophical Essay on the Forms of Great Epic Literature,
trans. Anna Bostock. Cambridge, MA: MIT Press.
Mill, J. S. (1833), “What Is Poetry?” and “The Two Kinds of Poetry,” Monthly Repository, VII (January),
60–70; and (October), 714–724.
Moran, R. (1994), “The Expression of Feeling in Imagination,” The Philosophical Review, 103, 1 (January),
Nietzsche, F. (1968), Basic Writings of Nietzsche, trans. Walter Kaufmann. New York: Modern Library.
Pippin, R. (2010), “The Paradoxes of Power in the Early Novels of J. M. Coetzee,” in J. M. Coetzee and Ethics,
ed. A. Leist and P. Singer. New York: Columbia University Press.
Porter, J. I. (2000), The Invention of Dionysus: An Essay on the Birth of Tragedy. Stanford: Stanford University
Porter, J. I. (2009), “Is Art Modern? Kristeller’s ‘Modern System of the Arts’ Reconsidered,” British Journal of
Aesthetics, 49, 1 (2009), 1–24.
Rorty, R. (1989), Contingency, Irony and Solidarity. Cambridge: Cambridge University Press.
Rutter, B. (2010), Hegel on the Modern Arts. Cambridge: Cambridge University Press.
Scheler, M. (1954), “On the Tragic,” trans. Bernard Stambler, in Crosscurrents, 4 (1954): 178–191.
Schelling, F. W. J. (1989), The Philosophy of Art, trans. Douglas W. Stott. Minneapolis: University of
Minnesota Press.
Schopenhauer, A. (1958), The World as Will and Representation, trans. E. F. Payne. New York: Dover.
Szondi, P. (2002), An Essay on the Tragic, trans. Paul Fleming. Stanford: Stanford University Press.
Speight, A. (2013a), “Tragedy and the Human Image: German Idealism’s Legacy for Theory and Practice,”
The Impact of Idealism: The Legacy of Post-Kantian German Thought, vol. 3, ed. Nicholas Boyle. Cambridge:
Cambridge University Press.
Speight, A. (2013b), “Philosophy and Literature in the Eighteenth Century,” in Routledge Companion to
Eighteenth Century Philosophy, A. Garrett, ed. (forthcoming).
Walton, K. (1990), Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge:
Harvard University Press.
Zuckert, R. (2010), “The Aesthetics of Schelling and Hegel,” in Dean Moyar, ed., The Routledge Companion
to Nineteenth Century Philosophy. New York: Routledge, pp. 165–193.

Kristin Gjesdal

Within the field of European literature, it would be no exaggeration to suggest that the
twentieth century begins with the modernist pantheon of the late 1800s (Baudelaire,
Rimbaud, Verlaine all deserve mentioning). Philosophy follows in due course, and
asks how modernist literature, with its stylistic earnest, its attacks on bourgeois morality,
and its relentless will to transcend existing literary conventions can yield a unique per-
spective on modern life. In seeking to provide a roadmap to twentieth-century
philosophy of literature in Europe, this chapter hones in on three different theoretical
paradigms and their responses to new literary forms: Marxism and critical theory
(Section II), phenomenology and hermeneutics (Section III), and the synthesis of polit-
ical and phenomenological thought in existentialist theory of literature (Section IV).
European twentieth-century philosophy of literature is, admittedly, much richer than
these three paradigms or theory formations. It includes structuralism, psychoanalytic
theory, post-structuralism, post-colonialist theories, and so forth. Yet these develop-
ments are not, typically, taught in anglophone philosophy departments. Critical theory,
phenomenology, and existentialism, by contrast, hold an uncontested place within
aesthetics and twentieth-century European philosophy classes. Further, the develop-
ment of post-structuralism and other, more recent contributions in Continental
philosophy can only be understood against the background of the central philosophical
positions of the twentieth century. It will be my suggestion that these positions not only
respond to the new forms of modernism, but that they also, in an important sense, shape
their philosophical form and direction in and through the encounter with modernist art.

The cultural vibrancy of the early twentieth century cannot fail to impress. The first
decades of the century alone saw the emerging of Schönberg’s atonal music, Picasso’s
transformation of painting, and the publication of groundbreaking works by Proust,
Joyce, Rilke, Kafka, Lawrence, and Trakl. New ways of writing require new ways of
thinking about literature. And the search for a vocabulary that can help us understand

the nature and value of these new expressions was one of the challenges with which
twentieth-century philosophy of literature was faced.
Looking for new ways to think and talk about contemporary art is not the same as to
uncritically endorse it. Sometimes the development of new forms reinforces the value
of more traditional literature. This, at least, was the conviction of Georg (Györy)
Lukács, whose work provides a profound defense of the novel as form.
In Soul and Form (1911), the young Lukács follows Hegel in seeing literature as a
way of making sense of the world. In modernity, however, such sense-making seems
increasingly difficult. In the face of secularization and a loss of shared cultural and ethical
values, the modern individual is left with a feeling of homelessness. According to Lukács,
the romantics first became aware of this feeling and the way it shapes modern literature.1
For the romantics, however, it was an unspoken premise that modern life—and, by
implication, modern literature—must be understood in contrast to ancient, and especially
classical Greek culture. As far as literature goes, classical Greek culture found its initial
form in the grand, Homeric epos. The Homeric epos, the romantics assumed, afforded a
religious and ethical anchoring of Greek culture; it expressed a set of ethical and religious
values that was, for centuries, left unquestioned. For us moderns, by contrast, the great
epos can no longer play this role. It is, for us, no more than a work of art, objects of
aesthetic appreciation. But what is more, no artwork, not even the works of our own
time, can offer binding guidelines for a meaningful life. Nor can such guidelines be found
in the extra-aesthetic realm. In this sense, the modern individual—and with it, modern
art—is and must be homeless.
Lukács’s Theory of the Novel (1914–15), the other main pillar of his pre-Marxist phase,
hones in on the idea that, amongst all artforms, the novel provides the most adequate
expression of the modern quest for an existential–ethical anchor-point. After his turn
to Marxism, however, Lukács revises his analysis of the modern world—and, in turn, his
view of what can qualify as an adequate response to the predicament that is uniquely
expressed by the novel. Yet Lukács’s contribution, his early as well as his later work, is
held together by a strong wish to defend the relevance of the novel in general and its
realist form in particular.
For the early Lukács, the realist novel, with its focus on individual character develop-
ment and a chronological ordering of events, is a medium in which the unease with
modern culture can reach its most adequate expression. But Lukács does not only display
a preference for realism as a literary genre. In his view, all novels deserving of their name
are realist in that they respond to the reality in which they emerge and of which they
are a part. Hence, Lukács adds to his notion of realism as a genre the idea of realism as
key to the understanding of modern literature across its different genre contributions.
At the end of the day, the commitment to realism is, in his view, the raison d’être of the
novel as form. Only against this commitment can realism, as a literary genre, be adequately
understood and appreciated.
After his turn to Marxism, Lukács moves from an understanding of literature in terms
of its capacity to express the modern condition to an interest in its more direct political
potential. The novel, he now argues, can trigger a will to societal change. However, even
if literature can initiate a will to social change, it cannot (as an aesthetic expression)
bring it to completion. At the end of the day, literature, even realist literature, must be
replaced by conceptual–analytical reasoning: the cold, hard light of scientific Marxism.
It is in this period that we find Lukács’s most aggressive disclaimer of modernist
literature. Whether it is the Russian masterpieces of Dostoyevsky and Tolstoy, the work

Kristin Gjesdal

of Balzac, or the contemporary novels of Thomas Mann, the realist novel is reflective of
the inner contradictions of bourgeois life (and can thus usher a move from bourgeois
decadence to more wholesome, post-revolutionary life-forms). In the modernist works
of Kafka and Beckett, by contrast, Lukács finds only the self-indulging narcissism of the
educated upper middle classes (though Kafka in this context gets a more charitable reading
than Beckett does).2 In the realist novel, form and objective content are one. Modernism,
by contrast, represents a false polarization of form and content. Thus modernism is
unable to express society as an objective whole; it can only present a splintered, frag-
mented, and tormented reality in which there is no ultimate retreat for the individual,
let alone redemption of homelessness through collective action.
In spite of the political rhetoric that saturates his critique of modernism, Lukács
never retorted to a defense of propaganda literature. It is indeed difficult to imagine a
more subtle analysis of the historical conditions and societal place of the novel than
the one we find in Lukács’s work. Further, Lukács initiates a line of criticism whose
anglophone descendants include Erich Auerbach (though he was German by birth,
Auerbach emigrated to the US in the wake of the Nazi horrors), but also, more recently,
Fredric Jameson.
The complexity of Lukács’s position—his insistence on the philosophical depth and
historical possibilities of the novel as a form—was never a point of appreciation for
Adorno, who was soon to emerge as Lukács’s starkest critic from within the quarters of
left-wing intellectuals. Likewise, Lukács, ever suspicious of Adorno’s negative dialectics,
had sought to sideline him as the last resident of the “grand hotel abyss.”3 The ideological
and philosophical differences between Lukács and Adorno culminate in their respective
readings of the works of aesthetic modernism. Beckett, the modernist writer par excel-
lence for Adorno, is for Lukács a living proof of how literature is doomed to decline once
it lets go of its commitment to realism (in both meanings of the term). Beckett’s
painfully passive characters, his minimal, only reluctantly expressive language, and his
turning his back on the representative function of literature add up to little more than
a perversion.4 In Beckett’s poetic aporia—“you must go on, I can’t go on, I’ll go on,” as
the Unnamable (1953) closes—Lukács sees nothing but a helpless reveling in the misery
of human existence.
For Adorno, by contrast, the distortions that Beckett voices are by no means his or
those of literary modernism, but symptoms of the dire condition of post-War society as
such. In Adorno’s view, Beckett responds to his historical world and society and is indeed
right in abandoning the (quasi-)Hegelian idea of an artistic totality that fully expresses
and mediates an absolute meaning that holds society together. In late modernity, Adorno
insists, such a totality can no longer be found. And to the extent that literature mimics
life, modern life must be presented in its true colors: as fragmented, atomized, and devoid
of meaning. That is, art certainly offers some form of meaning, but it is utopian, indirect,
and by way of hints and gestures, rather than insights that can be translated into direct
political action. For Adorno, this follows from the autonomy of modern art.
If Lukács and Adorno differ in their understanding of literary modernism, this, in
turn, reflects diverging readings of pre-modern art, even of the Homeric epos. Lukács, I
have mentioned, saw the Homeric epos as a testimony to an undivided, wholesome
society—one in which the homelessness of us moderns was still an unknown concept.5
Odysseus is, arguably, adrift, but homeless he is not. Adorno has a different reading of this
pre-modern work. In an early study co-authored with Max Horkheimer, Adorno focuses
on Odysseus’s sly and cunning strategizing in the face of mythical and natural powers.6


This, Adorno and Horkheimer suggest, is the dawn of Western rationality, its power over
nature, and, with this power, its peculiar dialectics of freedom and constraints. While
Odysseus calculates and conquers his surroundings, he concomitantly develops a strate-
gic reasoning that comes to dominate ever larger shares of the human life-world. Nature,
and with it human nature, is disenchanted: explained, perhaps, but also deprived of
meaning. This is a nature that fits the models of modern science, but not one where a
human agent can recognize itself and find a home. For Adorno and Horkheimer, this
one-dimensional rationality, this considerable engine of modernizing, will eventually
prepare for the crisis of meaning that preceded the National Socialists’ rise to domi-
nance in the 1930s.7
Responding to the horrors of the Second World War, Adorno asks how modern art
can find meaning in this predicament. Like Lukács, Adorno is socially and politically
motivated. And like Lukács, Adorno moves freely between a descriptive and a norma-
tive level of analysis. Yet the answer he provides is as far as it can get from Lukács’s
realist credo and its accompanying rejection of aesthetic modernism.
It was clear to Adorno that after the human tragedy of the War, art could not continue
as if nothing had happened. Any effort to retrieve traditional art represents, in his view,
a denial of the horrors that have taken place. A totalizing art in the face of a shattered
reality is betrayal—betrayal in the name of art, but also in the name of humanity. Art
that emerges from within and is expressive of a society that has proved rotten to its core
must reject all hope of societal consolation. It must reject the lukewarm bourgeois
humanism that had failed to stand up against the atrocities of the War. In Adorno’s
words, “[a]rt must turn against itself, in opposition to its own concept, and thus become
uncertain of itself right into its innermost fiber.”8 The art that does this, that displays its
staggering uncertainty in the most convincing way, is, for Adorno, the works of aesthetic
modernism. Whereas the bourgeois culture of the nineteenth century begged voluptuous
art and ascetic life, Beckett and other modernist writers produce literature that is as
ascetic as can be. This kind of literature cannot be enjoyed in an immediate or tradi-
tional way. It yields no direct aesthetic gratification. For the ever-uncompromising
Adorno, however, whoever concretely enjoys art is a philistine.9 Only as autonomous,
only as it turns its back to the very culture from which it came, can art disclose a glimmer
of a utopian dimension. From this point of view, Lukács’s commitment to realism is but
the last contraction of a loyal apparatchik. As Adorno puts it, Lukács managed to
“philosophize right past the aesthetic import of even his favorite texts.”10
Is there, then, no middle way between the dour endorsing of realism, on the one
hand, and the equally uncompromising commitment to modernism, on the other?
Walter Benjamin, it might be said, shares fundamental aesthetic sensibilities with
both Lukács and Adorno. Indeed, as Arendt once put it, Adorno, Benjamin’s junior
by a short decade, was perhaps his one and only disciple.11 But in the years from his
Marxist conversion to his tragic suicide (Benjamin took his own life near the Spanish
border when trying to escape his Parisian exile after France had fallen to the
Germans), Benjamin also drew on—and came to shape—Lukács’s outlook. So while
Benjamin, chronologically, belongs in between Lukács and Adorno, systematically
speaking his position can be understood as mediating the two extremes of Marxist
theory of literature.
Like Lukács, Benjamin starts out with an interest in Goethe and philosophical
romanticism. In Benjamin’s case, it is the romantic notion of criticism and, in a separate
work, Goethe’s Elective Affinities (1809), that draw his attention.12 Further, Benjamin is

Kristin Gjesdal

the author of a thoughtful philosophical-historical study of German tragic drama in the

baroque period.13 In all these works, he asks how literature can represent a reality from
which it is increasingly alienated, but also how we, as modern readers, should respond
to and critique art. How do traditional artforms answer the challenges of moderniza-
tion? How can we experience art in an age of mass reproduction and new media? Has
art, in this state, become a thing of the past?
In this context, Benjamin’s notion of the storyteller emerges as particularly
significant—and perhaps also as a link between his romantic beginnings and the Marxist
orientations that would color the final years of his life. Benjamin here develops a notion
of temporality that deviates from both Lukács’s interest in the historical-progressive
force of realist literature and Adorno’s turn to the promesse du bonheur of modernism. For
Benjamin, storytelling is the core of literature. Storytelling draws on an ability to share
experience.14 This sharing of experience, Benjamin suggests, is the glue of culture. The
storyteller’s secret is the capacity to make the content real and alive to his or her
audience. The storyteller has the audience share the experience of the story to the
extent that they make it their own. This takes a certain skill, yet it also requires an
audience that is able to listen with the appropriate degree of openness. Only to the
extent that the audience is absorbed and forgets itself in the story will the narrative
come to life and the experiences it conveys be made real.15 While the meaning of the story
is handed down from generation to generation, it exists in time, but is, all the same,
timeless. As Benjamin elaborates, the story “does not expend itself. It preserves and
concentrates its strength and is capable of releasing it even after a long time.”16 But in
order for it to live on in this temporality—its storage of ever new “nows”—the story
must be preserved, cared for, and passed on.17
Benjamin fears that the art of storytelling—which rests at the heart of the epos, but
also of modern literature from Cervantes’s Don Quixote, via Flaubert’s Sentimental
Education (1869), to Poe and Stevenson—has come to an end. As an oral tradition,
storytelling lives independently of the art of print, the stock in trade of the novel. The
novel, by contrast, nourishes itself from the printed media. As such, it competes with
another form of communication: that of the news story. The news story, though, reflects
a time consciousness that differs from the epos and the novel. It only lives the very
moment it is new, and it is as such that it evokes interest and curiosity. The fascination
with the new—with the ephemeral “now” of novelty—is why the art of storytelling “is
becoming unraveled at all its ends after being woven thousands of years ago in the ambi-
ence of the oldest forms of craftsmanship.”18 Storytelling has lost its magic, its ability to
draw its listeners into its world—its aura, as Benjamin calls it. In a world where the art
of listening is ever more rare, storytelling, and with it the traditional novel, is an endan-
gered species.
While exploring the form of storytelling, Benjamin lays out a landscape where the
novel is not (as Lukács had argued) threatened by modernism, but finds a last shelter in
the broken, modernist forms of a writer like Kafka. Kafka is a storyteller in a world where
the art of listening has perished. He gives expression to lives that would otherwise not
be chronicled and voices that would otherwise not be heard. Yet his characters, though
emerging in the strangest shapes of insects and furry animals are expressions of modern
life and its experience of alienation.19
Benjamin was not only writing philosophy and criticism. He also authored a number
of personal recollections and memoirs,20 as well as well as an incomplete poetic-
philosophical experiment, the Arcades Project (which survived the war thanks to a copy


of the manuscript that Benjamin had left in the care of his friend Georges Bataille).21 In
this work, Benjamin seeks to document the pace and temporality of the nineteenth-
century city focusing on its emblematic expressions in the wrought-iron works, the
shopping arcades, and urban characters like the flâneur, the prostitute, the scrap collec-
tor, and the avant-garde poet. In laying out the anatomy of the budding metropolis,
Benjamin collects prismatic glimpses of modern life, thus turning himself into the sto-
ryteller of an age whose hallmark it is to resist the temporality of conventional narratives,
be they in the form of the epic or the novel. In this way, Benjamin not only contrib-
utes to the philosophy of modern literature, but also reflects on the form and manner
in which such philosophy can and should be conducted.

In its different forms and permutations, Marxist philosophy theorizes the relationship
between literature and society, be it with reference to the realist novel or the modernist
work of art. It looks at how literature can voice the contradictions of modern life and
point beyond existing societal forms by offering glimpses of a worthy, human
existence. Phenomenology approaches literature from a different point of view. It does
not concern itself with political questions, but addresses the ontological status of
literature. What is a work of art? What is literary meaning? How can we understand
the kind of truth we experience in reading? Again, historicity and time emerge as key
categories, though these are now subject to a different philosophical import and are
not related to the societal function of art, but to the initiation of tradition as a
communal space or horizon of meaning that ultimately discloses the shared world of
author, work, and reader.
Heidegger starts teaching philosophy around the time Lukács publishes Soul and
Form. Initially, Heidegger did not express much interest in literature. What worried the
young phenomenologist was a meta-philosophical issue: how philosophy had turned into
a stifling, academic discipline and thus failed to engage the great works of the past.
Heidegger was a student of Edmund Husserl and part of the phenomenological move-
ment and its call for a return to the things, or in the young Heidegger’s case, the great
books—themselves. Husserl had parceled out the various phenomenological subfields to
his students, leaving the arena of literature to Roman Ingarden, whose study The Literary
Work of Art (1931) is not much read these days. Heidegger was assigned the field of
theology, which he would quickly leave behind. His magnum opus, Being and Time
(1927), is, as one might say, an existential-ontological study of the way we lead our lives
as understanding and interpreting beings. The fundamentally hermeneutic character of
human existence will be an important premise when Heidegger turns to literature in the
1930s, though his philosophy, at the time, has changed from an analysis of individual
existence to a study of being as such. As part of this framework, literature becomes
entirely crucial to Heidegger. Philosopher-poets like Hölderlin, Rilke, and Trakl offer
insights that jaded intellectuals no longer care about, insights into the difficulty of
addressing, in modernity, the question of the meaning of human existence.22
The short yet important essay The Origin of the Work of Art (1936) marks Heidegger’s
turn to philosophy of art. Right from the beginning, however, it is clear that Heidegger
wants none of what traditional aesthetics has to offer. For Heidegger, what matters, as a
point of departure for a study of art and its origins, is the work’s capacity for disclosure
of truth and meaning. What Heidegger has in mind is not representational truth or truth

Kristin Gjesdal

as correspondence, but the opening up of a field of meaning, a “world,” with reference

to which the human being can understand herself and her being.
Traditional aesthetics has traced art and literature back to creation, imagination,
psychology, and feeling. Here, the artist is seen as the origin of the work. Heidegger deems
this a subjectivization of art. It is not the artist that is the origin of the work, be it paining,
architecture, or literature, but, vice versa, the work that is the origin of the artist.
In bringing forth, articulating, and thus realizing the truths and values of a given
community, the work of art opens up a space of practice and self-understanding. The
standard example of such a truth-disclosing work is the Greek temple. With its solid
blocks of marble, the temple is a towering manifestation of what is good, true, and valu-
able for the Greeks (what Heidegger terms their “world”) yet, all the same, the mute,
physical presence of the stone marks a reminder of the limits of the space of understand-
ing and interpretation.23 In modern society, the tension between the world of human
meaning and the mute presence of the material world is expressed in less traditional
ways. Heidegger exemplifies his point through his analysis of van Gogh’s painting of a
pair of farmer’s shoes (that would later inspire some of Jacques Derrida’s ventures into
the ontology of painting). In this context, however, what matters more is the reflection,
towards the end of Heidegger’s essay, on how all art, in an emphatic sense, is or lives as
poetry, in the original meaning of poiesis.24
According to Heidegger, art establishes a non-subjective truth. Great works of art
create a standard of meaningfulness; they do not so much draw on tradition as emphatically
establish it. This is ultimately a poetic gesture; it is no means to an end, but an end in
itself: no contribution within the field of utilitarian purposefulness, but a sublime
disclosure of a horizon in which things are given meaning and purpose in the first place.
In modern society, however, we no longer gather at the steps of the temple. In
Heidegger’s elegiac phrase, the gods have fled. According to the later Heidegger—and
this is a point that might shed light on his inexcusable affiliation with National
Socialism—we no longer allow ourselves to have the kind of values that would yield an
ultimate point of orientation in life. When the temple no longer retains its status as a
place of worship and truth, when it is no more to us than an object of sheer aesthetic
pleasure, only the poets, marginalized as they are, can shelter the most profound meaning
of human existence by preserving a capacity for truth and giving expression to the dire
condition in which such truths are no longer viewed as fundamental to human life.
If the human being is a creature that lives understandingly, and if understanding is,
for Heidegger, mediated through language, then human existence is, in a certain sense
poetic: we dwell poetically, as Heidegger puts it, citing the late Hölderlin.25 Yet in
modernity, this dwelling has been occluded. Only the poets can point beyond the
forgetfulness at the core of such existence. Hölderlin’s “The Ister” is one example of
poetry that has retained this force.26
Within the framework of hermeneutic philosophy, Heidegger’s thoughts on litera-
ture, his redefining of poetry from a subjective artform to the very center of human
being and meaning-production, has influenced the work of Hans-Georg Gadamer,
though Gadamer filters the late Heidegger’s interest in poetry through the phenom-
enological vocabulary of Being and Time and an orientation towards historicity,
tradition, and Bildung that he reads out of Hegel. What Gadamer wants from
Heidegger is the notion of the non-subjective, meaning-disclosive truth of literature,
though this truth is now realized within the framework of a continuous, historical


Like Heidegger, Gadamer’s contribution to philosophy of literature consists in a

general theory of interpretation, as well as concrete analyses of works by Rilke, Celan,
and others. Like Heidegger, Gadamer does not see poetry as one artform amongst others,
but views it as offering a privileged access to art. In his view, literature yields the kind of
insight into human life and meaning that philosophy strives for yet, with its commit-
ment to discursive rationality, seems ever unable to grasp.27
Following Heidegger, Gadamer opens Truth and Method (1960) with a critique of the
post-Kantian tradition in aesthetics. But where Heidegger delivers a fundamental attack
on modern life and society, Gadamer is more concerned with the status of the human
sciences and the way in which they have changed within the climate of post-Kantian
philosophy. In his view, the romantics and their followers misunderstood Kant’s third
Critique, which, in turn, led to the view that art and poetry should be approached in
light of subjective intentions, psychological life, and genius. Gadamer, by contrast, is
interested in another dimension of literature and interpretation. He argues that the
meaning of literary works must be understood along the lines of their continuous appli-
cation and the way in which they make sense to readers within ever new contexts of
interpretation. The work of art is a totality of meaning, yet this totality is only realized
in and through the manifold of its concrete realizations.28 In its depending on ever new
productions, drama exemplifies the temporal dynamic of art. The meaning of a given
play, such as Shakespeare’s Hamlet, cannot be led back to an original, authorial inten-
tion or historical meaning, but lives in and by the different ways that the work gains life
and concretion through myriads of performances. Its meaning is a totality that is only
realized as a unity in manifold.
What, then, of modernist literature? What of the works that, as Adorno and Benjamin
point out, seem to live by rejecting rather than affirming the tradition and society in
which they emerge? While Adorno remained critical of (if not straight out hostile to)
the phenomenological tradition and its “jargon of authenticity,” Gadamer, throughout
Truth and Method, occasionally refers to Critical Theory and its insights. Moreover, he
is aware that for his philosophy of interpretation to be valid universally—which, no
doubt, is the goal of Truth and Method—he has to tackle the aesthetic challenges of
modernism. This discussion, however, is not part of Truth and Method, but takes place in
a number of essays published under the title The Relevance of The Beautiful, as well as a
series of readings of Paul Celan.
For the German intellectuals of the post-War era, Paul Celan’s hermetic poetry had
become a litmus test for philosophy of literature as well as for society and poetry’s role
and function in it. Adorno had famously declared that “[t]o write poetry after Auschwitz
is barbaric.”29 Yet Celan, in his view, had shown how poetry could be written even after
the horrors of the Shoah: by making clear the struggle with words, the pains of expres-
sion, the denial of a wholesome totality of meaning with which it is possible directly to
identify. It is in this abrupt style that the impossible, barren beauty of Celan’s poetry
finds shape.30 Benjamin, likewise, had viewed modern literature as a preparation for
hermetic modernism; in his study of the baroque tragic drama, he finds no great and
living tradition, but a melancholy outlook that is unable to invest in life and existence,
unable to see history and its forms as more than a landscape of dead meaning and petrified
symbolic forms.31
Gadamer, by contrast, does not accept the notion of such a fundamental break within
tradition.32 There are, to be sure, changes in our relation to—or failing to relate to—
traditional forms of poetry. But even hermetic poetry like Celan’s must be understood

Kristin Gjesdal

with reference to the larger framework of Western literature. Hence, Gadamer finds the
poetry of Celan, in whom Adorno saw a refusal to speak, to be a “covering of words (. . .)
like a roof over us.” As Gadamer continues, the words of Paul Celan “secure the
familiar (. . .). Syllable by syllable, that is, laboriously and tirelessly, the poet—or is it
each of us?—seeks to dismantle what is covering.”33
This interpretation of modernism makes it possible for Gadamer to incorporate post-
War art and poetry into his notion of a synthesizing, ever-expanding, and ongoing
tradition that always precedes and includes reader as well as author, laying down a
common ground that enables understanding and interpretation. Whether Gadamer
really does justice to the complex, literary status of modernist poetry, however, is a different
matter, one that, in fact, has led another German hermeneutician, Peter Szondi, to call
for a return to the romantic aesthetics that he, along with the early Lukács and Benjamin,
sees as particularly well equipped to account for the challenges of modern life.34

Within a French context, phenomenological and Marxist approaches to literature con-
verge in the work of Jean-Paul Sartre and Maurice Merleau-Ponty. While Sartre and
Merleau-Ponty initially worked closely together, they were soon to part ways. In
particular they disagreed on how Marxism and phenomenology could be best combined.
Sartre had been a student of phenomenology, both in its Husserlian and its
Heideggerian forms. However, writing in the wake of the Second World War, Sartre is
convinced that literature must have a political commitment: it must be engaged. And
the way it can be engaged is, for Sartre, more or less a function of its social involvement.
Thus Sartre’s philosophy of literature can be seen as an attempt to show how art can be
both an expression of freedom and, at the same time, socially committed. For Sartre, this
is far more than a theoretical challenge. Sartre is not only the author of philosophical
works like Being and Nothingness (1943), but also of literary texts and drama such as
Nausea (1938) and No Exit (1944). The challenge of unifying phenomenology and
Marxist politics saturates both his literary and his theoretical production.
In the build-up to the War, Sartre’s interest in the nature of consciousness and
subjectivity led to a study of imagination.35 In this context, he pursued a quasi-transcendental
approach and did not say much about the societal situatedness in which the freedom of
imagination is realized. After the War, however, this became a pressing issue. The question
“What is literature?”, he now felt, must be answered with reference to the society in
which literature is produced and appreciated.
In his 1947 essay—entitled, precisely, What is Literature?—Sartre addresses the nature
of literature through three fundamental questions: “What is writing?” “Why write?”
And, “For whom does one write?” In answering each of these questions, he draws, in
equal measure, on his philosophical work in phenomenology and his experience as a
writer and critic.
Sartre contrasts literature with painting, music, and poetry, claiming that not only is
literature committed but, indeed, that it also harbors a commitment that is exclusive to
it. If color and sound (music and painting) is penetrated with significance, only language
displays a structure that refers to the world.36 Yet, literature is not only characterized by
the fact that certain things are said, but also by the fact that they are being said in a
certain way.


Upon asking why writing is worthwhile, Sartre immediately turns to the Leitmotif of
Being and Nothingness—that of human freedom. What makes writing worthwhile is the
freedom it involves for both author and reader: the freedom to not only shape an
experiential totality, but also to take a stance on the social reality of which author and
reader are both a part. A book is, as Sartre sees it, an expression of authorial freedom
and, all the same, an appeal to the freedom of the reader. In making the story come alive,
in realizing its potential, the reader co-creates the book. Reading is a synthesis of percep-
tion and creation; it is a concrete manifestation of freedom in the historical world.
Sartre takes an interest in colonial literature and politics—which is further explored
in essays such as “Black Orpheus” (1948)—and emphasizes the impossibility of writing
for the oppressed. To write for the oppressed is already to anticipate their liberation, to
address them as free individuals. This is further explored as Sartre turns to the issue of
readership. Ultimately, it is the phenomenology of readership that allows Sartre to
emphasize both freedom and engagement in literature.
The reader is typically viewed as an anonymous, faceless entity. Sartre wishes to get
away from the abstract and universal category of readership and develop the notion of
a historically situated reader. The reader’s subjectivity is permeated with interests and
desires. No literature can be written without having a particular reader in mind. Sartre
exemplifies his point by reference to Richard Wright’s Black Boy (1945). Wright, he
claims, does not appeal to an abstract subjectivity and freedom. The subjectivity he
addresses is chained, colored, and longing for liberation. And, what is more, it is precisely
in this shared historical-political predicament that freedom and engagement concur—
not only in Wright’s work, but also in literature as such.37
From the point of view of Sartre’s engaged literature, both Adorno’s turn to high-
modernism and Heidegger’s focus on the truth of great art may look like a nostalgic
mourning of a bourgeois world whose values are void of meaning in post-Second World
War society. Whereas Adorno and Heidegger, though for different reasons, reject
engaged literature, Sartre sees in literature a tool for freedom and liberation, hence also
a set of new values that he would later endorse in his role as a public intellectual.
Yet, Sartre never abandoned his interest in stylistic issues. This is clear from his many
critical essays (which include careful readings of Baudelaire, Gide, and Camus),38 but
also from his late analysis of Flaubert in the multi-volume study The Family Idiot (1971–2).39
Sartre approaches Flaubert from the point of view of his social and historical context,
yet never loses sight of the transparent and fluid character of Flaubertian prose, his
acute and detailed descriptions of landscapes and social situations. Throughout his
work, Sartre’s voice as a craftsman and writer always shines through his criticism—and
vice versa.
This, however, was not how Merleau-Ponty saw Sartre’s focus on art and literature.
Merleau-Ponty had initially been co-editing the monthly Les temps modernes with Sartre.
In this period, he emerged as an eager spokesman for existentialist literature. In essays
such as “Metaphysics and the Novel” (1945) and “A Scandalous Author” (1947),
Merleau-Ponty not only defended the work of Sartre and Simone de Beauvoir, but also
drew up the contours of an existentialist philosophy of literature. According to Merleau-
Ponty, literature is about bringing ideas into existence. This, he claimed, is what the
works of Balzac, Stendhal, and Proust are all about.40 Yet literature does not explain the
world (or ideas about it), but realizes it in concrete experience. At the end of the day,
the tasks of philosophy and literature are one: “One will not only witness the appearance

Kristin Gjesdal

of hybrid modes of expression, but the novel and the theatre will become thoroughly
metaphysical, even if not a single word is used from the vocabulary of philosophy.”41
According to Merleau-Ponty, this is a new situation. Only in modernity does the
human being free itself of a notion of universal essence. Thus, the novel no longer has
a metaphysical grounding on which to build. In this way, each work must provide its own
justification (its own metaphysics, as Merleau-Ponty puts it). In doing so, however, it
evades the pale of pre-formed morality and becomes, as it were, amoral, expressive of the
kind of freedom that is presupposed in all decision-making. In the words of Merleau-
Ponty, “this freedom, the condition of all morality, is equally the basis of an absolute
immoralism because it remains entire, in both myself and others, after every sin and
because it makes new beings of us at every instant.”42
Throughout the 1950s, the political and philosophical differences between Sartre and
Merleau-Ponty became more and more evident. Existentialism, Merleau-Ponty argued,
is a kind of subjectivism, a quasi-Cartesian position that ascribes too much power to
subjectivity. It leads to a misunderstanding of (Marxist) politics and, in turn, a naïve
celebration of Soviet Communism. As Merleau-Ponty now sees it, art can help us see
why existentialism does not hold. Art—painting, but also literature and music—
presents us with the mute, trans-subjective foundation of the world, the “prose” or
ground of meaning that is prior to individual (discursive) language use.43 While Merleau-
Ponty in this period comes close to both Schelling and the Kant of the third Critique,
he retains an interest in the concrete being and nature of art, literature included, and
delivers brief yet powerful readings of works by Claude Simon and Michel Butor.44 With
Merleau-Ponty, existentialist philosophy of literature returns to its phenomenological
basis, though this basis is now oriented towards the craft, style, and materiality of literature,
thus avoiding the Teutonic Heideggerian language of world-disclosure and the absence
of gods in the modern world.

Twentieth-century philosophy of literature involves a rich and complicated roster of
positions that are intertwined, opposed, and in constant communication and develop-
ment. Common to these positions, however, is the reflection on a body of literature
that is, by and large, overlapping. Modernism, I have suggested, is one such shared
point of reference. The realist novel is another one. Other works, and other ways of
responding to these works, could also have been mentioned: psychoanalysis, struc-
turalism, deconstruction, and feminism, to list only a few. However, as mentioned in
the introduction to this chapter, I have wanted to focus on the figures of the twentieth
century who are read and discussed across philosophy departments and literary stud-
ies and show how the encounter with (modernist) literature informs, develops, and
challenges their positions, influencing their contribution to aesthetics, but also feed-
ing into their philosophy as such. In this way, philosophy of literature is not an
auxiliary branch of twentieth-century thought. Indeed, twentieth-century European
philosophy cannot be understood without taking into account the ongoing dialogue
with and reflection on art and literature. And in this landscape, aesthetic modernism,
presenting a profound stylistic and philosophical challenge, plays a central role:
Whether as affirmative or critical, it is as a reflection on modernism that twentieth-
century philosophy of literature finds its ultimate form.


  1 See Georg Lukács, “On the Romantic Philosophy of Life: Novalis,” Soul and Form, trans. Anna Bostock
(London: Merlin Press, 1974), 42–54.
  2 Georg Lukács, “The Ideology of Modernism” and “Franz Kafka or Thomas Mann?” in Realism in Our
Time: Literature and the Class Struggle, trans. John and Necke Mander (New York: Harper & Row, 1962),
26, 44–46, and 47–92.
  3 Lukács, The Theory of the Novel, trans. Anna Bostock (London: Merlin Press, 1971), 22.
  4 For Lukács, modernism is, by definition, anti-realism. See “The Ideology of Modernism,” Realism in Our
Time, 17. For his comment on Beckett, see ibid., 32.
  5 Theory of the Novel, 29–56.
  6 Theodor W. Adorno and Max Horkheimer, Dialectic of Enlightenment, trans. John Cumming (London:
Verso, 1979), 43–80.
  7 Dialectic of Enlightenment, 182–9.
  8 Theodor W. Adorno, Aesthetic Theory, ed. and trans. Robert Hullot-Kentor (Minneapolis: University of
Minnesota Press, 1997), 2.
  9 Aesthetic Theory, 13.
10 Adorno, “Extorted Reconciliation: On Georg Lukács’ Realism in Our Time,” in Notes to Literature vol. I,
trans. Shierry Weber Nicholsen (New York: Columbia University Press, 1991), 236.
11 Hannah Arendt, “Walter Benjamin 1892–1940,” trans. Harry Zohn, in Men in Dark Times (London:
Harcourt Brace & Co, 1983), 154.
12 Walter Benjamin, Goethe’s Elective Affinities, trans. Stanley Corngold, in Walter Benjamin (ed.), Selected
Writings, vol. I, ed. Marcus Bullock and Michael W. Jennings (Cambridge, MA: Harvard University
Press, 1996), 297–361.
13 Walter Benjamin, The Origin of German Tragic Drama, trans. John Osborne (London: Verso, 2003).
14 Walter Benjamin, “The Storyteller,” in Illuminations, ed. Hannah Arendt, trans. Harry Zohn (New York:
Schocken Books, 1969), 84.
15 “The Storyteller,” Illuminations, 91.
16 “The Storyteller,” Illuminations, 90.
17 Thus Jameson classifies Benjamin’s position as a Marxist hermeneutics. See Frederic Jameson, Marxism
and Form (Princeton: Princeton University Press, 1971), 60–83.
18 “The Storyteller,” Illuminations, 91.
19 “Franz Kafka: On the Tenth Anniversary of his Death,” 111–40 and “Some Reflections on Kafka,” in
Illuminations, 111–40 and 141–45.
20 See for example A Berlin Chronicle and One-Way Street (selections), in Water Benjamin, Reflections, ed.
Peter Demetz, trans. Edmund Jephcott (New York: Schocken Book, 1987), 3–60 and 61–96.
21 Walter Benjamin, The Arcades Project, ed. Rolf Tiedemann, trans. Howard Eiland and Kevin McLaughlin
(Boston: Harvard University Press, 2002).
22 Martin Heidegger, “Why Poets?,” in Off the Beaten Track, ed. and trans. Julian Young and Kenneth
Haynes (Cambridge: Cambridge University Press, 2002), 200­–42.
23 Martin Heidegger, The Origin of the Work of Art, in Off the Beaten Track, 19–27.
24 The Origin of the Work of Art, 44.
25 Martin Heidegger, “. . . Poetically Man Dwells . . .,” in Poetry, Language, Thought, trans. Albert Hofstadter
(New York: Harper & Row, 1975), 213.
26 Martin Heidegger, Hölderlin’s Hymn, “The Ister,” trans. Julia Davis and William McNeill (Bloomington:
Indiana University Press, 1996).
27 Gadamer, “Aesthetics and Hermeneutics,” in Philosophical Hermeneutics, ed. and trans David E. Linge
(Berkeley: University of California Press, 1977), 99–104. See also “Are the Poets Falling Silent?,” in On
Education, Poetry, and History: Applied Hermeneutics, ed. Dieter Misgeld and Graeme Nicholson, trans.
Lawrence Schmidt and Monica Reuss (Albany, NY: SUNY Press, 1992), 81.
28 This is brought out in Gadamer’s analogy between play and the experience of art. The play, as Gadamer
reads it, “renews itself in constant repetition.” Further, play (and this is part of Gadamer’s critique of
modern aesthetics) is prior to the consciousness of the players. See Hans-Georg Gadamer, Truth and
Method, trans. Joel Weinsheimer and Donald G. Marshall (New York: Continuum, 2003), 103.
29 Theodor W. Adorno, “Cultural Criticism and Society,” in Prisms, trans. Samuel and Shierry Weber
(Cambridge, MA: The MIT Press, 1997), 34.
30 Aesthetic Theory, 322.

Kristin Gjesdal

31 See for example Benjamin, The Origin of German Tragic Drama, 174–77. Lukács borrows from Benjamin’s
analysis of allegory in “The Ideology of Modernism.” Even though he sees Benjamin, in his study of the
allegory, as “the first critic to attempt a philosophical analysis of the aesthetic paradox underlying
modernist art,” Lukács does not accept Benjamin’s more balanced assessment (Realism in Our Time, 41).
For Lukács, the force of realist literature is the way that it “assumed change and development to be the
proper subject of literature” (ibid., 35). Hence modernism, with its static allegorical form, “leads not only
to the destruction of traditional literary forms; it leads to the destruction of literature as such” (ibid., 45).
32 Hans-Georg Gadamer, “The Relevance of the Beautiful,” in The Relevance of the Beautiful and Other
Essays, ed. Robert Bernasconi, trans. Nicholas Walker (Cambridge: Cambridge University Press, 1986),
27–9 and 45–6.
33 Hans-Georg Gadamer, “Who Am I and Who Are You?,” in Gadamer on Celan, ed. and trans. Richard
Heinemann and Bruce Krajewski (Albany, NY: SUNY Press, 1997), 81.
34 See Peter Szondi, “Schleiermacher’s Hermeneutics Today,” On Textual Understanding and Other Essays,
trans. Harvey Mendelsohn (Minneapolis: University of Minnesota Press, 1986), 95–115. See also Peter
Szondi, Introduction to Literary Hermeneutics, trans. Martha Woodmansee (Cambridge: Cambridge
University Press, 1995).
35 Jean-Paul Sartre, The Imagination, trans. Kenneth Williford and David Rudrauf (London: Routledge, 2012).
36 Jean-Paul Sartre, What is Literature? And Other Essays, trans. Bernard Frechtman (Cambridge, MA:
Harvard University Press, 1988), 30.
37 What is Literature?, 78–81.
38 Some of these essays are included in Sartre, Modern Times: Selected Non-Fiction, ed. Geoffrey Wall, trans.
Robin Buss (London: Penguin, 2000).
39 Sartre, The Family Idiot, vols 1–5, trans. Carol Cosman (Chicago: University of Chicago Press, 1981–1993).
40 Maurice Merleau-Ponty, “Metaphysics and the Novel,” in Sense and Non-Sense, ed. and trans. Hubert L.
Dreyfus and Patricia Allen Dreyfus (Evanstone: Northwestern University Press, 1964), 26.
41 “Metaphysics and the Novel,” in Sense and Non-Sense, 28.
42 “Metaphysics and the Novel,” in Sense and Non-Sense, 38.
43 Referring to the poet Francis Ponge, Merleau-Ponty writes that “[i]n the art of prose, words carry the
speaker and the listener into a common universe by drawing both toward a new signification through
their power to designate in excess of their accepted definition or the usual signification that is deposited
in them from the life they have had together in us. This is what Ponge very well described as the ‘semantic
thickness’ of words and Sartre as their ‘signifying soil.’ ” Maurice Merleau-Ponty, The Prose of the World,
trans. John O’Neill (Evanston: Northwestern University Press, 1973), 87.
44 See Maurice Merleau-Ponty, Texts and Dialogues, ed. Hugh J. Silverman and James Barry, Jr., trans.
Michael B. Smith (London: Humanities Press, 1992), 140–45.

Further Reading
Bürger, Peter. The Decline of Modernism. Trans. Nicholas Walker. University Park, PA: The Pennsylvania
State University Press, 1992.
Gutting, Gary. French Philosophy in the Twentieth Century. Cambridge: Cambridge University Press, 2001.
Jameson, Fredric (ed.). Aesthetics and Politics: The Key Texts of the Classic Debate within German Marxism.
London: Verso, 1980.
Kearney, Richard and David Rasmussen (eds). Continental Aesthetics: Romanticism to Postmodernism. London:
Blackwell, 2001.

Kristin Boyce

The history of analytic philosophy of literature could be written as a narrative of the

efforts to formulate and solve a series of interrelated paradoxes. Indeed, such efforts have
been so central to the development of the discipline that the editors of a recent
anthology of philosophy of literature single out attention to puzzles and paradoxes as
definitive of the discipline. Philosophers of literature may differ widely in their assump-
tions and ambitions, but they all seek to “help us better understand the nature of [the
relevant] paradoxes—their goals, the entities with which they deal, the standards that
govern our participation in them, and the broadly ethical questions to which they give
rise” (Davies and Matheson 2008, xiv). This kind of approach will strike many as uncon-
troversially part and parcel of an analytic philosophy of literature. After all, as Mary
Mothersill has argued, one can trace back to the giants of the early analytic tradition
such as Russell (if not also to Wittgenstein) the conviction that philosophy itself takes
the form of resolving puzzles and paradoxes. I will argue, though, that this is neither the
only nor the best form that a distinctively “analytic” philosophy of literature can take.
Instead of writing a survey of paradoxes formulated and solutions attempted, I shift
attention in this chapter to what I will call “the paradox of philosophy and literature.”
What I have in mind is this. On the one hand literature (along with the arts more
generally) has consistently been of marginal importance as an object of philosophical
reflection. On the other, or so I will argue, it has been of special methodological
importance for analytic philosophy from the beginning.
The marginalization of aesthetics within analytic philosophy is no secret. By
contrast, the methodological centrality of the literary arts for analytic philosophy has
not until recently been recognized at all. From its inception, the analytic tradition has
worked hard to disentangle itself from other humanistic enterprises, especially art and
religion, and to secure its proximity to modern science. Indeed, when art first registers
for analytic philosophy, it is not as an object of reflection at all but rather as an object
of comparison, the function of which is precisely to underscore that proximity. Recent
work in the history of early analytic philosophy by Cora Diamond and others, though,
has made it possible to see how deeply tied the “analytic” ways of doing philosophy that
emerged were not just to developments in the sciences, but to those in the literary arts
and criticism as well. I argue that this work has important implications for aesthetics
generally, and for philosophy of literature in particular, that have not yet been recognized
or explored.
Kristin Boyce

In Section 1 of this chapter, I argue that the marginal importance that art has had as
an object of philosophical reflection is a function of the way that it first registered for
the emerging analytic tradition. The assumption about art upon which philosophers
such as Gottlob Frege and Rudolf Carnap rested their developing conception of
philosophy did not constitute serious theoretical reflection about art. Nevertheless, I argue,
those assumptions play a serious role in shaping the course of such theoretical reflection
when it does begin to develop within the analytic tradition and they help to explain why
such reflection has continued to be so marginal within that tradition as a whole. Because,
and insofar as theoretical reflection about art has been shaped by and bound up with the
project of disentangling philosophy as an enterprise from artistic ones, I argue, there is
good reason to suspect that a re-evaluation of that project should have important impli-
cations for aesthetics in general and philosophy of literature in particular.
In Section 2, I turn to such a re-evaluation. I begin by exchanging one object of
comparison—the assumptions about art that exercised the early analytic imagination—
for another: developments within the literary arts and criticism that were contemporaneous
with the emergence of the analytic tradition. With this new object of comparison in
view, I argue, it becomes clear that the enterprise of philosophy neither can nor should
be disentangled from literature. I make this argument by focusing on a particular case
study, the literary project of Henry James. James is a contemporary of Frege’s and his
importance as a progenitor of high literary modernism rivals that of Frege’s for analytic
philosophy.1 The salient comparison is that between the literary project, which led
James to develop the new literary form and style for which he is famous, and the
philosophical one, which led Frege to develop his new form of expression, Begriffsschrift.
From the vantage created by this comparison, I argue, it becomes clear just how deeply
related the emerging “analytic” ways of doing philosophy were not just to the aims and
methods of science, but to those of the literary arts and criticism as well.
Making this argument involves writing philosophical history of a particular kind.
Michael Kremer has recently distinguished between history of philosophy, which func-
tions to ratify the concerns of the present, and historical philosophy, which returns to
the past for the purposes of, as Nietzsche put it, providing an “untimely perspective”
upon those concerns (Kremer 2013, 23). In this particular case, returning to the past
provides a critical vantage from which to see that one cannot do full justice to the
philosophical possibilities that were opened by the development of an analytic tradition
without taking the measure of philosophy’s proximity to both science and art while
conflating it with neither. The pay-off for aesthetics of this return to the past is that a
philosophical self-understanding, which more adequately reflects the proximity of the
work of philosophy to the work of literature, should make possible new and, by some
measures, better ways of reflecting philosophically on art.

Section 1: Art and Literature as Objects of Analytic

Nowhere has the tendency of analytic philosophers to pay too little attention to art
been better identified than in Mary Mothersill’s Beauty Restored. Mothersill does not
share that tendency; she aims at nothing less than an analytic theory of art that can
stand comparison with great theories of the pre-analytic past such as those of
Immanuel Kant and John Dewey. In order to clear the way for this project, though,


she begins her ambitious monograph by asking why so little progress has yet been
made on such a project. A good part of the answer she gives to this question bears
on particular reasons why some analytic aestheticians have mistakenly thought such
a project misguided and how those who are by contrast interested in developing such
a theory have misunderstood what doing so would entail. But she also points to a
more general lack of interest in art that, she argues, is both persistent and largely
unexplained. Perhaps, she speculates, the explanation lies in the assumption, which
C. I. Lewis makes explicit, that the arts are “payday goods not workday goods” that,
“although very well in their place” fit at best into the “interstices” of a serious life
(Mothersill 1984, 65). After all, philosophy certainly takes itself to be a serious busi-
ness, so one could see it standing to reason that aesthetics or philosophy of literature
would fit at best into its interstices.2 A fuller explanation, though, can be found if
one considers how and why early analytic philosophers first turned their attention
to art.
Early analytic philosophers such a Gottlob Frege and Rudolf Carnap do not ignore art,
but they draw attention to it when and insofar as it is useful as a special sort of object of
comparison, one which clarifies what philosophy is by throwing into sharp relief the
common concerns that define both philosophy and science and reinforce the proximity
between them. Frege and Carnap both draw attention to art in the course of separating
out the wheat of real philosophy from the chaff of pseudo or debased versions of it. Like
mathematics and the sciences, Frege argues, philosophy is concerned with truth, that is,
with advancing human knowledge. The problem, Frege argues, is that a commitment to
truth is hard to sustain and the temptation to avoid or short-circuit the demands that
such a commitment makes on our thinking is ever-present.
Frege directs attention to multiple variations of this temptation throughout his
writings.3 Consider, for instance, Frege’s criticism of formalist mathematicians. One
of the most difficult demands that truth makes on the mathematician, he argues, is
that she carefully establish that the basic terms she uses refer. A pretense to formal-
ism is tempting precisely because it serves as a means of circumventing this

The formal concept is a shield held up so long as questions about the reference
of the signs threaten. This danger over, the shield is allowed to fall. In fact, the
conception of numbers as figures is really used only at the outset, where that
obligation is oppressive.
(Frege 1960, 217; emphasis mine)

Another place where Frege makes especially clear the connection between the confusions
with which philosophy is principally concerned and the temptation to avoid the difficulty
of living up to the demands of truth is found in section 39 of the Foundations of Arithmetic.
Here Frege argues that our use of the term “unit” functions to conceal, and to allow us
to avoid, conceptual difficulty. He writes,

The difficulty is so well hidden under the word “unit”, that those who have any
suspicion of its existence must surely be few . . . [and] that is the real, though no
doubt unconscious reason why we prefer it.
(Frege 1979, 50–1)

Kristin Boyce

What is important for our purposes is that Frege describes philosophers who have
succumbed to it as “lapsing into fiction” (Frege 1984, 362 and 367). The only difference
between such a failed philosopher and fiction writer is that the fiction writer makes no
pretense to a concern for truth. It is for this reason, Frege argues, that neither the writer
nor the reader of the Odyssey care whether the word ‘Odysseus’ refers to a real person or
not: for questions about the referent of a proper name arise if and only if a concern for
truth is operative. The philosopher, by contrast, is confused: he has “gone astray into the
sphere of fiction without knowing it or wanting to” (Frege 1984, 163). By Frege’s lights,
the central task of philosophy is to “conduct unceasing struggle” against the vulnerability
of the “human spirit” to suffering this kind of confusion, to protect it from its tendency
to come unmoored from the concern for truth that properly orients it (Frege 1979, 6–7;
1967, 7).
Carnap inherits this central task and reinterprets it in his own way. He, too, seeks to
protect philosophy from a form of degeneration that he characterizes through comparison
with art. Like science, he argues, genuine philosophy is a matter of theory construction—
of producing “a system of statements” which “relate as premises and conclusions”
(Carnap 1932, 147). Pseudo philosophy, by contrast, confuses the “medium theoretical”
with those of artistic expression. Pseudo philosophers, i.e. metaphysicians, he writes
(ibid., 147),

Are musicians without musical ability. Instead they have a strong inclination
to work within the medium theoretical, to connect concepts and thoughts.
Now, instead of activating, on the one hand, this inclination in the domain of
science, and satisfying, on the other hand, the need for expression in art, the
metaphysician confuses the two and produces a structure which achieves nothing
for knowledge and something inadequate for the expression of attitude.

In Carnap’s hands, the task of liberating the human spirit from confusion evolves into
the project of “portioning all uses of language into one of two categories”: scientific uses
and emotive uses; those with logical form and those without it; sense and nonsense.
When Frege and Carnap make these remarks about art, they are not (nor do they take
themselves to be) developing a theory of art. Instead, they are relying upon assumptions
about what art is in order to make a point about what philosophy is. It might seem natural,
though, for a tradition that begins by resting its self-understanding on unreflective
assumptions about art would be content to let art remain an afterthought—a “special
case” to be handled once adequate theories of mind and language are all-but-complete—
and to allow the course of what theoretical reflection about art does arise to be
disproportionately shaped by those unreflective assumptions.
The legacy of the analytic tradition’s unreflective starting point has different impli-
cations for philosophy of literature than it does for aesthetics more generally. For
aesthetics, the most immediate and important starting point is logical positivism. In
light of Carnap’s sorting project, one central agenda that takes shape is that of “getting
clear about the logic of critical talk about art” (Weitz 1956, 207). As M. H. Abrams
characterizes it, this required developing the capacity to discriminate “a variety of typi-
cal linguistic usages,” which the positivists had lumped together into the dustbin of
nonsensical, emotive uses, thus clarifying their distinctive logical forms (Abrams 1972, 5).
For philosophical thought about literary art, which is itself constituted of “linguistic
usages,” the focus of attention falls less on the logic of talk about art than on the logic


of those units of discourse that comprise the work itself. Here the initial philosophical
orientation comes from Frege’s discussion of sense without reference as well as from
Russell’s analysis of empty definite descriptions. The starting point is the analysis of
apparent “assertions” within a fictional work, which contain what appear to be proper
names. Questions about the reference of such apparent proper names develop into ques-
tions about the “fictional worlds” to which a network of apparent assertions that contain
them seem to refer—particularly questions about the relation between such a “fictional
world” and the real one. How can such a network of apparent assertions, which consti-
tute a fictional work and fail systematically to refer to individuals in the actual world,
afford insight into that world? How can we explain our emotional responses to such a
fictional world and the individuals within it: our weeping with or for Oedipus, or the
pounding of our hearts in response to murderous green slime on a movie screen?
I’ve been arguing that the marginal importance that literature has had as an object
of analytic reflection is a function of how (and how closely) early analytic philosophers
sought to identify the aims and methods of philosophy with those of science. This might
seem tantamount to arguing that literature, and art more generally, just is and should
remain of marginal importance from an analytic point of view. The particulars of a posi-
tivist viewpoint may no longer be widely shared, but it is still fairly uncontroversial to
assume that the aims and methods of philosophy are intimately bound up with those of
science. This assumption is shared even by those like Mothersill, who seek to challenge
the marginality of art and to develop an analytic aesthetics that can stand comparison
in scope and insight with theories of art of the pre-analytic past. As Mothersill under-
stands it, philosophy, like science, seeks to contribute to the slow but steady progress of
human knowledge by “transmuting perplexities”—that is, “sensible sounding questions
to which there seem to be no sensible answers”—into “problems” that are “tractable,”
i.e. that admit of solution (Mothersill 1984, 22–4). In this way, she argues, philosophy
contributes to the slow but steady progress of human knowledge.4
In the next section I will argue that a closer look at the early stages of the analytic
tradition suggests that such an exclusionary identification of philosophy with science
cannot be justified. If we shift attention from the assumptions about art that exercised
the early analytic imagination to developments within the literary arts and criticism
themselves that were contemporaneous with the emergence of that tradition, what we
see is how intimately bound up the transformations within philosophy were with those
developments. My point in making this argument is not to steer away from the Scylla of
identifying philosophy with science and towards the Charybdis of conflating it with
literature instead. It is rather to suggest that one cannot do justice to the distinctive
possibilities that the analytic tradition opens without successfully navigating the prox-
imity of philosophy to both literature and science, while collapsing it into neither. Once
the need for such ongoing methodological negotiation is recognized, it becomes clear
that a different kind of philosophical reflection on literature is both possible and
called for.

Section 2: Philosophical Work and the Work of Literature

We can begin to see how deeply bound up the development of analytic philosophy is
with developments in the literary arts and criticism if we place that development within
the context of what Clement Greenberg calls a modernist turn within philosophy and
the arts. Greenberg motivates that turn the following way. In the wake of newly

Kristin Boyce

developed scientific method, the arts, along with philosophy and also religion, were
in danger of losing the authority they had traditionally had as important sources of
knowledge. In response, they turned to what Greenberg calls “immanent critique”, crit-
icizing “from the inside, through the procedures themselves of that which is being
criticized” (Greenberg 1993, 86). The purpose of such critique was to “justify” these
humanistic enterprises by showing that they each made some contribution that could
not be made in any other way. Kant is Greenberg’s paradigmatic example. In Logic, he
argues, Kant uses the methods of philosophy to: (1) demonstrate that there are forms of
knowledge which those methods, and only those methods, could make available; and
(2) clarify the limits of those methods, that is, to show that they cannot provide certain
kinds of knowledge they had been taken to provide, namely, knowledge beyond
experience—whether of God or freedom or the world as a whole. In the arts, such cri-
tique involved an individual art in using its own procedures to (1) produce the kind of
experience or effect that it alone could produce, and (2) show that it alone could pro-
duce it by demonstrating that the conditions for the possibility of that experience were
these possibilities unique to its medium. What this involved, Greenberg argues, was
eliminating “any effect that might conceivably be borrowed from or by the medium of
any other art” (ibid.).
The remarks that Frege and Carnap make about art are best understood as a moment
in the dialectical unfolding of this modernist turn, a moment when the task of self-
justification had intensified into a felt danger of degeneration and a corresponding need
for purification. As we have already seen, Frege and Carnap understood this task to
involve distancing philosophy from those other humanistic enterprises whose value had
also been called into question, eliminating from it any effects that were not proper to
the “medium theoretical.” But if attention is focused on actual developments within the
literary arts, instead of on the unexamined assumptions about art that figured in their
self-understanding, it becomes evident that philosophy and literature cannot be as
neatly and completely prized apart as they suggest.
The connections between modernist literature and philosophy come most sharply
into focus if we compare the minute and anxious scrutiny that both direct toward
ordinary language. Within both literature and philosophy, the felt danger of degeneration
takes the form of a deep and abiding worry about ordinary forms of expression. Within
philosophy, this worry turns on the limitations of ordinary language as a tool for
expressing thoughts in a logically perspicuous way. With its powerful capacity to convey
feeling and tone, as well as to paint vivid pictures in the imagination, Frege argues,
ordinary language often covers that which is relevant to a concern for truth, even as
colorful clothing covers and obscures the articulated structure of the human body. It
is in effect the medium of our vulnerability to degenerating in the way that so concerns
Frege and which is, in his view, a standing threat to the integrity of those enterprises
that are, or should be, concerned with truth. For the imprecision of ordinary language
plays into the persistent temptation to keep ourselves from being very clear about what
a commitment to truth really requires of us in any given case. Frege’s response is
Begriffsschrift, a notation for expressing thoughts in such a way that everything of
relevance to a concern for truth is perspicuously displayed. Begriffsschrift is, he argues, an
important tool for the philosopher because once someone’s thought is represented to
them in this way, it is much more difficult to remain conveniently in the dark about
what a commitment to truth requires.


Modernist philosophy has nothing on modernist literature when it comes to an acute

awareness of the shortcomings of ordinary forms of expression. Consider, for example,
the literary project of Henry James, a writer who is contemporaneous with Frege and
whose importance as a precursor of modernist literature rivals that of Frege’s as a
progenitor of analytic philosophy. James is as deeply concerned as is Frege with the way
that ordinary forms of expression foster a tendency to degenerate. As early as The Portrait
of a Lady, James is preoccupied with ways that such conventional forms lend themselves
to a pervasive tendency to keep ourselves in the dark—albeit a tendency to keep
ourselves in the dark not (or not only) about what our commitment to truth requires,
but about what is required of us by our commitments to other people or even ourselves.
In his seminal reading of Portrait, Richard Poirier argues that this early masterwork is an
extended study of how pervasively and costly is the tendency to allow “convenient
typologies of literary and social convention” to compromise or “deaden” genuine intimacy.
The centrality of this preoccupation, he argues, is reflected in the novel’s title:

The movement towards the end of the novel puts into juxtaposition the varieties
of death and deadness which concerned James so deeply and all of those aspects
of life which are tainted and deadened by the predictable, conventional. . . . All
the characters are at last known to us in some final way, as if they have become
“portraits” indeed . . . rather than remained imaginatively alive and changing.
(1960, 241)

This preoccupation stays with James throughout his career, deepening and darkening
until, as Leo Bersani puts it, the novels from the major phase show us what Leo Bersani
calls the “peculiar spectacle of [one of] the two most social talkers” in all of fiction
exploring the possibility that language itself (not just particular ways we use it) precludes
genuine knowledge of another and “giving the highest value in [his] work to the secret
inexpressible passions of unworldly heroes and heroines” (Bersani 1976, 130).
James’s literary project, no less than Frege’s philosophical one, engages in an
“unceasing struggle” against the tendency to keep ourselves from getting clear about
what our most basic commitments require of us—what James calls “the forces that make
for muddlement” (James 1908, xiv). He does not develop a new script, one which might
be taken for an alternative to or even replacement for natural language, but he develops
new literary forms that make it harder for us to stay in the dark. Poirier draws attention,
for example, to how James’s famously difficult and abstract writing style contributes to
this end. In his reading of Portrait, Poirier argues that James anticipates that his reader
will be inclined to assume too quickly that he or she knows all about his protagonist,
Isabel Archer. Such a reader is all too primed to bring to bear the “convenient typologies
of literary and social convention” in order to draw premature inferences from the shape
of the dramatic action to “too exclusive definitions” of Isabel’s character (Poirier 1960,
237–8). James, Poirier argues, does his best to intervene on Isabel’s behalf. By employing
language to describe her that is “general to the point of being abstract” he undermines
the authority of any of the more specific interpretations of her that the reader is likely
to have ready to hand (ibid., 237). By frustrating the expectations with which we
approach Isabel, he seeks to show us what we are doing—how the habitual ways that we
bring words to bear with respect to others compromise our appreciation for and respon-
siveness to their aliveness.

Kristin Boyce

A certain interpenetration of early analytic philosophy and high literary modernism

has not gone unremarked. Richard Shusterman, for example, has argued that the
relationship between Bertrand Russell and T. S. Eliot extended beyond the personal in
ways that have not been sufficiently recognized. In a similar vein, Morritz Weitz and
Stanley Cavell lament the impact of logical positivism on criticism. “It is one irony of
recent literary history,” Cavell (2003, 226) argues, that:

The New Criticism, with one motive fixed on preserving poetry from what it
felt as the encroachment of science and logical positivism (repeating as
academic farce what the nineteenth century went through as a cultural tragedy),
accepted undemurringly a view of intention established, or pictured, in that
same philosophy.

And Weitz, Mothersill argues, goes so far as to picture critics as distinguished as Clive
Bell and Northrop Frye as “driven by the desire to emulate Carnap” (Mothersill 1973, xxx).
I would not dispute the connections to which Shusterman, Weitz, Cavell, and others
draw attention, but the significance of those connections needs to be re-evaluated.
For one thing, once we do fuller justice to the deep connection between the modernist
literary and philosophical enterprises that were unfolding, we can see the influences that
philosophy exercised on literature and criticism in a different light. Consider, for exam-
ple, Shusterman’s argument that Russell influenced Eliot’s work as well as his person. In
making this argument, Shusterman carefully teases out multiple ways in which Eliot’s
critical project resonates with Russell’s philosophical one, pointing particularly to Eliot’s
operating assumption that “the misleading nature of language” generates confusion
within the practice of criticism (Shusterman 1982, 170). What is now clearer is that this
assumption resonates with not only the philosophical enterprises with which Russell was
bound up, but was also internal to the modernist literary and critical enterprises that
Eliot inherited and developed. This makes it less tempting to construe Eliot’s openness
to Russell’s influence in terms of the imposition of a bullying, aggressive “pants wearing”
discipline by a more delicate and less directed one.5
But the change of aspect that is needed does not just involve changing how we view
particular instances of literature’s openness to the influence of philosophy. It also
involves becoming cognizant of how deeply connected certain ways of doing philosophy,
which the emergence of the analytic tradition made possible, are to the literary arts. The
most important case here is that of Wittgenstein. Wittgenstein’s investment in the arts,
especially music and literature, is common knowledge. And like Russell, he was on
intimate terms with critics of his day as important as F. R. Leavis. Although he rarely
makes art the explicit focus of philosophical reflection, that writing is remarkable for its
literary quality. Like James, he is set apart from his predecessors by the unusual form and
style he develops.
Because that form cannot be easily parsed into view and argument—the form which
many analytic philosophers have taken to be constitutive of philosophical writing—it
has often registered as, at best, an obstacle to getting at the philosophical value of
Wittgenstein’s work. It is here that recent work in the history of early analytic philosophy
by Cora Diamond and James Conant is of special importance. The form and style that
Wittgenstein developed, they argue, were philosophically necessary—the ones that were
needed in order for Wittgenstein to develop central aspects of the philosophical project
that he inherited from Frege. In other words, in order to understand what Wittgenstein


is doing, it is necessary to understand why he wrote the way he did. And one cannot under-
stand this, it turns out, without appreciating the connections between the philosophical
work Wittgenstein is doing and the work of literature.
Diamond is especially explicit about this connection. Consider for example the
argument she makes in “Wittgenstein, Mathematics and Ethics: Resisting the Attractions
of Realism.” In this article, Diamond seeks to clarify the sense in which the Tractatus
has an “ethical intent.” Most often, she argues, this has been taken to mean that,
appearances to the contrary, the book must contain some views about morality. As
Diamond interprets it, by contrast, both the ethical intent and the philosophical
achievement of the book are better understood on the model of Laura Ingalls Wilder’s
novel for young adults, The Long Winter. Like the Tractatus, The Long Winter contains
little if any obviously “ethical” vocabulary or thought. But that does not mean the work
is not ethical through and through:

Suppose that what is brought into contact with a situation is the story in The
Long Winter of the two men struggling through the snow. . . . The person
making this connection acts in the light the story casts. To understand the
sense that the person’s action has is to see in it its relation to the story. . . . Whole
sentences, stories, images, the idea we have of a person, words, rules: anything
made of the resources of ordinary language may be brought into such a relation
to our lives and actions and understanding of the world that we might speak of
the thinking involved in that connection as “moral.”
(1999, 247–8)

In other words, the book’s ethical significance isn’t a matter of some ethical content
embodied in it but of the way a reader brings the book to bear on any number of situations.
Similarly, the ethical intent of the Tractatus is best understood as a matter of the way
a reader can bring it to bear on her life. The particular use for it that Diamond singles
out is the use which Frege saw for his Begriffsschrift­—as a tool in the philosophical task
of liberating his interlocutors from confusion, that is, as we saw, from their tendency to
stay unclear about something. Only what Wittgenstein strives to make it harder for us
to stay unclear about is not so much what either truth (Frege) or others (James) require
of us as it is what life or reality does. If one makes the right philosophical use of a tool
like the Tractatus, Diamond argues, it can help us overcome the temptation to short-
circuit the difficulties of reality by looking to books for a kind of “answer” that they
cannot give us:

At the end of the Tractatus (6.52), we are in a position to imagine a whole

library of books containing all sorts of true learning about the world and what
is in it, the answers to every imaginable question. But all these books have in
them is absence. There is nothing in them that answers, or bears on, the
problems of life. But the very fact that in these books, as we may imagine them,
there are answers to every imaginable question can help us to transform our own
desire for an answer to the problem of life. This can be done by transforming
also our desires with relation to the Tractatus, the “philosophy book” before us,
that is, by giving up our desires to read its propositions as providing a kind
of knowledge.
(2004, 127)

Kristin Boyce

For our immediate purpose, what is important about Diamond’s interpretation is how
clear she makes it that coming to so much as understand the kind of philosophical work
that Wittgenstein is doing—philosophical work that depends for its possibility on the
emergence of an analytic tradition—depends on grasping how closely that philosophical
work is related to the work of literature.
My point in drawing attention to the proximity between philosophical and literary
work here is not to invert the arguments of Frege and Carnap by arguing that the
realness of philosophy is a function of how closely its aims and methods track those
of literary art on the one hand and how clearly distinguished they can be from those
of science on the other. It goes without saying that a great deal of powerful philo-
sophical progress has taken (and with luck will continue to take) the form of
advancing views and argument, or even constructing the kind of systematic theory
that so preoccupied Carnap. The point is rather that not all genuine philosophical
progress takes this form. Nor does all “work” that does take this form make genuine
philosophical progress. In order to do full justice to the possibilities that were opened
by the emergence of a distinctively analytic tradition in philosophy, what is needed,
rather, is an ongoing willingness to recognize both science and literature as useful
objects of comparison as well as to think patiently through both the similarities and
differences that these objects of comparison throw into relief in any given case. In
other words, what is needed is a return to the ancient commitment to the ongoing
and open-ended process of self-knowledge with which philosophy began. Clearly this
commitment is no easier to live up to in modern than in ancient times. It is easier to
obscure the importance of literature (and art more generally) as an object of philo-
sophical reflection and minimize the difficult work involved in doing justice to it.
But if we avoid this temptation and face the challenges more squarely, the pay-off,
both for philosophy and for a philosophical understanding of literature, is substantial.
For there is no philosophical tradition better positioned to illuminate literature in a
powerful and distinctive way; nor is there a tradition better positioned to benefit
deeply from doing so.6

1 The claim is not that Frege directly influenced James (or vice versa). There is no evidence for such a claim
nor even reason to suspect that they were aware of each other’s work.
2 If one feels the force of this explanation, it will not seem mysterious that continental philosophy should
strike most philosophically minded literary critics and theorists as a more promising resource. Even the
hermeneutics of suspicion, which philosophers such as Marx, Nietzsche, and Foucault have been
understood to license or require, reflects a deep appreciation for the power of art to shape our individual
and collective self-understandings (even if it is more often for ill than for good).
3 Conant 2003 argues that an ethical commitment to thinking through, rather than avoiding, conceptual
difficulty, animates Frege’s thought as a whole and deeply influenced Wittgenstein.
4 Michael Kremer has persuasively argued that to conceive philosophy as “progressing” in this way is to
model it on the mathematical sciences (Kremer 2013, 12–13).
5 Mothersill draws attention to the tendency that analytic aestheticians have historically had to picture
philosophy as something that literature needs to be protected from. Cf. especially Mothersill 1984,
38 and 52.
6 This research was assisted by a New Faculty Fellows award from the American Council for Learned
Societies, funded by the Andrew W. Mellon Foundation. I am grateful for comments on earlier drafts
by John Bickle, Mike Bruno, John Gibson, Alicia Hall, Chris Snyder, J. Robert Thompson, and
Danielle Wyle.


Abrams, M. H. (1972): “What is the Use of Theorizing about the Arts?,” in The Search for Literary Theory,
ed. Morton W. Bloomfield. (Ithaca: Cornell University Press), 3–54.
Beardsley, Monroe (1958): Aesthetics: Problems in the Philosophy of Criticism (New York: Harcourt, Brace &
Bersani, Leo. (1976): A Future for Astyanax: Character and Desire in Literature (Boston: Little Brown).
Carnap, Rudolph (1932): “The Elimination of Metaphysics Through Logical Analysis of Language,” in
Logical Positivism, ed. A. J. Ayer (New York: Free Press), 60–81.
Cavell, Stanley (2003): Must We Mean What We Say? (New York: Cambridge University Press).
Conant, J. (2003): “On Going the Bloody Hard Way in Philosophy,” in The Possibilities of Sense, ed.
J. Whittaker (New York, Macmillan).
Davies, David and Carl Matheson (2008): “Introduction: What is Philosophy of Literature?,” in Contemporary
Readings in the Philosophy of Literature (Broadview Press), xi–xvii.
Diamond, Cora (1999): “Wittgenstein, Mathematics, and Ethics: Resisting the Attractions of Realism,” in
The Cambridge Companion to Wittgenstein, ed. H. Sluga (Cambridge: Cambridge University Press), 226–61.
Diamond, Cora (2004): “Introduction to ‘Having a Rough Story About What Moral Philosophy Is’ ,” in The
Literary Wittgenstein, eds. J. Gibson and W. Huemer (London and New York: Routledge), 127–32.
Frege, Gottlob (1960): Translations from the Philosophical Writings of Gottlob Frege, ed. P. Geach (Oxford: Basil
Frege, Gottlob (1967): “Begriffsschrift,” in From Frege to Goedel: A Source Book in Mathematical Logic:
1879–1931, ed. J. Van Heijenoort (Cambridge: Harvard University Press), 1–83.
Frege, Gottlob (1979): Posthumous Writings (Chicago: The University of Chicago Press).
Frege, Gottlob (1984): Collected Papers on Mathematics, Logic and Philosophy (Oxford, Blackwell).
Frege, Gottlob (1997): The Frege Reader, ed. Michael Beaney (Oxford: Basil Blackwell).
Greenberg, Clement (1993): The Collected Essays and Criticism: Modernism with a Vengeance, 1957–1969
(Chicago: The University of Chicago Press).
James, Henry (1908): “Preface,” in What Maisie Knew (New York: Scribners & Sons), xiv.
Kremer, Michael (2013): “What is the Good of Philosophical History?,” in The Historical Turn in Analytic
Philosophy, ed. E. Reck. (Palgrave MacMillan). Page numbers taken from version available at http://
historyjune%201.pdf (accessed August 10, 2015).
Mothersill, Mary (1973): “Introduction,” in Aesthetics and the Theory of Criticism, eds. L. C. William
Calaghan, C. Hempel, S. Morgenbesser, M. Mothersill, Ernest Nagel, and Theodore Norman (Chicago
and London: The University of Chicago Press), xix–xxxix.
Mothersill, Mary (1984): Beauty Restored (Oxford: Clarendon Press), 1984.
Poirier, Richard (1960): The Comic Sense of Henry James: A Study of the Early Novels (New York: Oxford
University Press).
Shusterman, Richard (1982): “Eliot and Logical Atomism,” ELH 49:1, 164–78.
Weitz, Morris (1956): “The Philosophy of Criticism,” Proceedings of the Third International Congress on
Aesthetics, 207–16.

Part II

M.W. Rowe

The word ‘literature’ has three uses in contemporary English: it can refer to linguistic
works of outstanding artistic merit (“She’s studying French Literature”); it can refer to
printed promotional pamphlets and catalogues produced, for example, by damp-proofing
companies and evangelical churches (“Would you like to take some of our literature?”);
and it can refer to established written material used in a field of academic study (“He’s
read widely in the immunology literature”). In this article, I shall only be concerned
with ‘literature’ in the first sense.
Within literature of this kind, it’s useful to make two distinctions. First, most lit-
erature is fictional, and the very greatest literary works (the Canterbury Tales, Macbeth,
Middlemarch, The Waste Land etc.) tend to be fictional or fictionalized; but a signifi-
cant body of important literature is intended to be straightforwardly argumentative
and factual – including many essays by Bacon, Hazlitt, and Orwell. Second, non-
argumentative literature includes both drama and narrative. A purely dramatic work
is one in which the scenes, characters, and events are directly shown and presented;
a purely narrative work is one in which the scenes and story are described and told. In
sections II–IX of this chapter, I shall concentrate on works of fiction in narrative form;
but in section X, for ease of argument, I shall focus on works of fiction in dramatic

A work of literature is a work of art; and a work of art is something which can sustain,
and is usually intended to sustain, aesthetic attention. But what kind of attention is
aesthetic? Let me take the example of sight. I can look at something to create or increase
desire for some activity (such as drinking, eating, sex, or physical exercise); and I can
look at something out of a desire for factual knowledge. This latter motive can be divided
in two. I can look either out of pure curiosity (such as when, out of historical interest, I
look at Roman wheel-ruts in order to estimate the width of Roman wheel-bases), or from
a desire to acquire information needed for a practical project (such as when I look at a
desk in a showroom to see if it will fit through my study door). However, I can also look
at something simply because it gives me pleasure to do so, and it is this form of attention
which we term ‘aesthetic’.
The differences between aesthetic attention and looking as a prelude to practical
activity become clearer when we consider the two attitudes’ satisfaction conditions. If I

am looking at something or someone in order to encourage appetite, sexual desire, or

physical exercise, then the natural outcome of this looking is eating, sex, or exercise.
The gaze and the activity are both motivated by the same desire; trying to fulfil the
desire, we might say, falls into two ordered parts, the first subordinate to the second. If
this kind of looking is not followed by activity, then it will be considered a failure and
some explanation will be required (“The cake was too expensive”, “She didn’t fancy
me”, “There wasn’t time”).1 But aesthetic attention does not fall into two ordered parts:
it is not simply a preliminary to an activity that involves more than mere looking; we
feel no temptation to explain why we have not moved into the practical phase (there is
no practical phase); and we do not think our desire unfulfilled if it is not followed by
some kind of action.
The difference between the aesthetic gaze and the two forms of looking which aim
to acquire knowledge comes into clearer focus when we realize that these latter
motives are satisfied when the desired knowledge has been gained. Once I’ve looked
at the wheel-ruts and assessed their width then, if my motive was simple curiosity
about wheel-bases, I shall stop looking and turn to some other activity. Similarly,
when I’ve decided that the desk looks too wide to fit through my study door, I shall
stop looking and possibly turn to other desks which might fit through it rather better.
In addition, it is quite possible that the knowledge I sought from looking could be
acquired in some other way. For example, I might find that reading Roman documents
gives me a more accurate idea of Roman wheel-bases than looking at certain ruts
(which may have been affected by later usage); and I may find that consulting the
piece of paper on which I wrote down the dimensions of my door, in conjunction with
the furniture shop’s catalogue, is a more reliable method for establishing whether the
desk will fit through the door.
Neither of these features is found in the aesthetic case. If I am looking at a landscape
because it gives me pleasure, then it is clearly not the case that reading about its
dimensions would satisfy my desire equally well or better; and aesthetic pleasure in an
object can never require me to dispense with the object and pass on to something else.
Aesthetic attention can slowly wither or suddenly disappear, or some more pressing need
may override it, but it does not have a second stage to which merely looking is a
preliminary, and it cannot be replaced by some other source of knowledge.2

Sometimes attending is something I do; sometimes it is something that happens to me.
In a boring lecture, or reading a set of instructions, I may well make myself attend, and
in this case attending is an action explained by beliefs and desires. But if some Roman
artefact immediately captures my curiosity, or some side-street strikes me as immensely
charming, then I might well find my attention held and sustained by either object.
There is an internal relation between the aesthetic attitude and pleasure, since plea-
sure is a form of non-voluntary attention.3 It would be strange to be totally absorbed by
a film – feel every emotion of the characters, anticipate the next twist in the plot with
baited breath, be utterly self-forgetful and completely unaware of the cinema and audi-
ence for two hours – and yet claim that the film was without any artistic merit. Similarly,
it would be very odd, when seated in the cinema, to spend two hours doodling on pieces
of paper, looking at the people around you, worrying about your motor insurance,


thinking about what you might eat afterwards, and yet claim that you enjoyed every
moment of the film.
Of course, some kind of back-story could make these claims just about intelligible:
perhaps, in the first case, the film was produced in Nazi Germany and you found it
fascinating only insofar as it revealed insights into Nazi psychology; perhaps, in the
second, you could see that the film would be gripping and compelling under ordinary
circumstances but an unpleasant row at work had entirely destroyed your powers of
concentration. But in both circumstances, I think we can agree, aesthetic attention was
not present, either because the attitude was not aesthetic or because it was not attention.
The quality of a work is determined by the intensity and durability of my pleasur-
able experience. A work can capture my attention on first sight, but prove incapable
of sustaining it on subsequent viewings. Conversely, I might find a work initially off-
putting, but after reading some relevant critical books, and holding my attention on
it, I might later find the intensity and durability of the attitude it supports is greater
than that supported by works which initially appeared more inviting. Which works
can sustain attention will vary from time to time and from one person to another. But
when we consider unbiased people with the relevant practice and theoretical under-
standing, we do tend to find that their tastes converge and that there is some agreement
over which authors, for example, are of central importance (and even greater agree-
ment, of course, over which are not). This convergence and intersection is the basis
for the notion of ‘the canon’, and the fact that it is subject to endless revision and
argument does not show that the notion is in some way bogus, but that it is sustained
by living tastes informed by changing social outlooks.4
The word ‘pleasure’ is deeply problematic. It seems accurately to capture an appropriate
response to A Midsummer Night’s Dream, the Rape of the Lock, Schubert’s Scherzo in Bb,
and Auden’s “On the Circuit”; but does it also capture an appropriate response to King
Lear, Tess of the D’Urbervilles, Brahms’ C minor piano quartet, and Rembrandt’s late
self-portraits? We certainly find this latter group gripping, interesting, absorbing,
compelling – but pleasurable?
The linguistic difficulty we face here is one aspect of a larger problem we face with
the word ‘pleasure’ in all areas of life. It seems natural to describe chatting with friends,
swimming, drinking champagne, and dancing as pleasures, but it applies much less
comfortably to more demanding and protracted activities undertaken as ends in themselves.
Were conquering Mount Everest, proving Fermat’s last theorem, or writing Ulysses
pleasurable? In many ways they were no doubt stressful, miserable, exhausting, and
utterly preoccupying, yet they were undertaken for their own sake, and for the satisfaction
of those who accomplished them. Indeed, it is precisely this kind of pleasure which is
generally thought to give life meaning and point.
Unfortunately, there is no suitable word to describe the psychological concomitants
of these endeavours, so I’ll use the phrase ‘strenuous pleasures’ to contrast with the ‘idle
pleasures’ of chatting, champagne-drinking and so forth.5 Any life which consisted just
of strenuous pleasures would be insupportable; any life which consisted just of idle
pleasures would be boring and frivolous; and it is clear that in any sane life the two kinds
of pleasures have to alternate. Similarly, there is something unbalanced about someone
who can only enjoy artistic entertainments and frivolities, or someone who can only
respond to harrowing, tragic, difficult, and demanding works. Both profundity and
charm, we feel, should have their place.


I have already implied that the aesthetic attitude does not involve scanning texts
in order to pick up facts, but it strikes me as obvious that literature provides –
amongst other benefits – the strenuous pleasure of acquiring knowledge. If this is so,
what kind of knowledge can literature provide, and how is this knowledge related to
propositional information?
Science attempts to show how the world is in itself, and not how it appears to any
observer. Consequently, its explanations (as opposed to its observations) are restricted
to the so-called primary qualities (solidity, extension, figure, motion, rest, and number)
and do not mention properties which depend on the constitution and position of any
particular creature.6 These include the so-called secondary qualities (colours, sounds,
sensations, smells, and tastes); indexical terms and phrases (‘I’, ‘now’, ‘here’ etc.); emotions
(love, anger, nostalgia, etc.); and values (rudeness, sensitivity, good, and evil, etc.).
Having prioritized the primary qualities, physical science tries to explain the behaviour
of physical objects by positing general covering laws, by referring to past and present
conditions, and by framing explanations in terms of causes.7
Since the seventeenth century, this kind of reductive explanation has proved remark-
ably successful, but it should not blind us to the fact that there are other forms of
explanation that are equally valuable. An obvious instance is the form of explanation
used when trying to account for the experience and behaviour of a conscious creature.
In this case, we try to explain its behaviour by imaginatively reconstructing its life-
world. This means we must imagine how the universe seems to the creature, not how
the universe is in itself; we must reconstruct all of its experience (including its experi-
ence of secondary qualities, values, emotions, indexical thoughts etc.); we must imagine
it in as much concrete detail as possible; we must refer to its future goals as well as its
current and past experience; and we should not restrict our explanations to causal rela-
tionships alone. This kind of subject matter and explanation is the ambit of the
humanities: history, human geography, anthropology, social science, and literature.8
To put the matter in a nutshell: science tries to give knowledge of the world from the
objective point of view; the humanities try to give knowledge of the world from the
subjective point of view.9 Abstract scientific truths like ‘6 × 6 = 36’ and ‘Water is H20’
are not simply true for me from over here now, they are true for everyone everywhere at
all times. But literature’s subjective point of view, its account of how the world seems,
necessarily involves reconstructing a number of different views from a number of differ-
ent spatio-temporal points. The perceptual experience I am currently enjoying can only
be seen from over here; the particular anxiety I feel about my court appearance tomor-
row can only be experienced by me now; and these experiences can only be identified
by the use of indexical expressions.
Many people have the impression that a work of literature can be about any subject
treated in any kind of way, but it soon becomes evident that works of literature can only
explore the Lebenswelten of conscious, rational creatures, most usually human beings.
Literature is concerned with how the world appears to the senses and emotions, the way
it stirs memories and associations, the way it meshes with future plans and projects, the
way all experiences are woven together and interfused, and basically every aspect of
being a rational creature inhabiting a segment of the universe at a certain point in time.
Accordingly, a book of logarithms, a physics textbook, even a set of instructions for
assembling a flat-pack hi-fi unit, cannot be works of literature and cannot be read as


works of literature even if they are put into verse. The reason for this is that such works
lack all implication of a subjective point of view, and the interest of the language
alone – however beautiful and ingenious – seems insufficient for a full-blooded literary
response. Of course, a work which explored emotional reactions to the icy perfection of
mathematics, a novel about a widely admired physics professor failing to write his seminal
paper, or a poem about the frustrations of assembling flat-pack units are quite conceivable.
They may even contain substantial sections of logarithms, scientific explanations, and
instructions. But these treatments are potentially literary precisely because they embed
these intrinsically objective topics within a subjective and emotional viewpoint. Just
as there is something which it is like to see certain objects and feel certain emotions,
so there is something which it is like to think certain abstract propositions.10
We might contrast the difficulty of imagining a book of logarithms alone as a work of
literature, or read as a work of literature, with the ease of imagining some historical tale
as a work of literature or read as a work of literature. If there is a conscious experience
and a narrative point of view then the change of aspect is effortless; indeed, there are
many works which began life as contributions to history, anthropology, sociology, and
scripture, which are now largely read as works of literature.

The idea that a work of literature must embody a point of view is something of a cliché,
but the usual accounts of this viewpoint, and the theories of narrative based on these
accounts, are often unclear.
There is an important difference between perceiving and imagining perceiving. In the
case of perception, the centre of consciousness and the perceptual point will usually be
found in a single physical object, namely, my body, specifically my head and brain.
Sometimes, however, the centre of consciousness and the point of view can be divided
between two physical objects. If I am looking through a periscope, down a fibre-optic
cable, or at TV pictures transmitted from a drone, then the point of view will be the lens
at the top of the periscope, the far end of the cable, or the TV camera, and for vicarious
perception of any kind to take place, there must be a physical object of some kind at the
locus of the point of view.
Imagining perceiving is different. Although my consciousness must still be located
in my body, there is no need to imagine my body or any other kind of physical object
at the point of view. To establish this fact, consider the following example: suppose I
am asked to imagine what a town I know looks like when seen from a height of 200 feet.
In order to do this, do I first (or also) have to imagine myself or some other person in
a plane, or a camera mounted in a drone, at the imagined point of view? Clearly not. I
can imagine a point of view moving over the town without first (or also) imagining
some person or a camera moving over the town. In the same way, I can imagine myself
flying over a town without imagining I am in a plane rather than a helicopter, or one
particular kind of plane rather than another. Imagining a viewpoint moving over a
landscape is therefore a different kind of task from imagining the viewpoint of some
person or physical object (even an imaginary person or physical object) travelling over
the same landscape.11
The usual way of thinking about narrative – dividing it into first-person and
third-person – is evidently confused, because first-person narrative is from the first-person


point of view, while third-person narrative is of people described in the third-person; it’s
a bit like dividing paintings into watercolours and seascapes. In addition, the phrase
‘first-person narrative’ suggests that we must imagine events from a person’s point of
view. This, as we’ve seen, is false on two counts: first, in the case of actual perception,
the point of view could be that of a physical object; second, in the case of imagined
perceiving, we can imagine a point of view without imagining it to be the point of view
of a real or imaginary person or object. Finally, ‘third-person narrative’ suggests that all
narrative must be about or of people, whereas this is clearly not the case: one could well
imagine a poem which is simply about moving through an uninhabited landscape.
A better way to think about narration is to see that its basic forms are impersonal
narrative and first-person narrative. Impersonal narrative is akin to the way the camera is
normally used in non-documentary films: it has a point of view on, shall we say, three
people in a room, but we do not have to imagine that the camera’s point of view is the
perspective of some assumed fourth person or of a spy camera deployed by one of the
characters; it simply gives us vicarious perceptual information about what is going on in
the room. Consequently, the author of an impersonally narrated novel has the same role
in the fictional world of the novel as the director and cameraman have in the fictional
world of an impersonally framed film: i.e. none.
In certain kinds of film, however, the camera can pretend to take on a character’s
point of view ­– often signalled by the use of wobbly hand-held cameras, and occluded
views of important events – and this is analogous to first-person narration in literature.
Impersonal narration is the most basic variety of narration, and every child is familiar
with its apparently magical or self-contradictory powers (I can imagine a place where I
am not, or seeing heaven when I have no physical eyes, or the Earth before life evolved),
and first-person imagining – which involves additional elements of pretence – is the
more evolved, circumscribed, and complex form.

Imagining part of the world from a point of view is a completely commonplace form
of thought, and reading or watching literature is a natural enhancement and extension
of it.
Sometimes we entertain fictional storylines because they give us the pleasure of
imagining we are richer, more attractive, exciting, and successful than in fact we are;
sometimes we entertain fictional storylines simply because they are funny. But on other
occasions we use imagination to act as a way of probing and rehearsing for the future;
and here our goal is truth and effective action.12 Suppose you have a job interview
tomorrow. It would be rational to imagine who might be interviewing you, and what
kind of questions you might be asked. You can rehearse possible replies and consider how
these might steer the panel onto your areas of strength, and what strategies you might
employ to make them oblivious to areas of weakness. If you want to steel your nerves
you might try to imagine the room and the interviewers’ faces in detail, the blended
smell of coffee and leather chairs, and the quiet rattle of a keyboard in an anteroom.
Clearly, how well you do in the actual interview could depend on how accurately you
imagined the scene, how insightfully you predicted the questions, and how effectively
you managed to quell your nerves.
We can see that we engage in similar exercises when we imagine an entirely fictional
interview; or imagine, after a real interview, what we could or should have said; or envisage


what someone else is going through when they are having a real interview, and so on.
None of these interview scenarios are straightforwardly true (none of them shows what
is true or was true) but they do try to show truths of another kind: what might be and
could be true, or what might have been and could have been true13 – even if fiction tends to
disguise the status of its truths by use of the present and past tense. If we couple these
observations with what has already been said about the subjective point of view, then
we can see that the imagination is well equipped to discover modal truths about
subjective experience.
What kinds of knowledge might be derived from imagining an interview? The first is
experiential knowledge: you have established what some experience might be like.
Second, you have acquired a certain amount of knowledge of how to do something –
perhaps simple skills such as how to answer a question truthfully without revealing the
whole truth. Third, you have acquired knowledge of what to do (what Aristotle called
phronesis):14 the ability to weigh up and correctly evaluate a whole mass of often conflicting
human data and arrive at a sensible and just solution. Fourth, you might have acquired
a certain amount of empathic understanding of what, for example, your interviewers’
backgrounds might lead them to look for in a candidate. Finally, you may even be able
to summarize some of this information in propositional form.
There is no general rule about how much and how accurately propositions can capture
the four kinds of knowledge outlined above. In the case of experiential knowledge, one
can be completely at a loss as to how to put one’s knowledge into verbal form (e.g. your
knowledge of what blue looks like); sometimes it’s possible to say rather more (what
driving a car is like). The same is true of the other forms of knowledge: it’s fairly straight-
forward to say how you should shut down a computer, rather more difficult to say how
hard the engine should be pulling before you release the handbrake on a hill start; it’s
easier to give rules of thumb for calming cantankerous customers than it is for making
people love you; it is more straightforward to say why being an exile should make some-
one anxious than it is say why someone feels so insecure about their completely normal
and unexceptional background. In many cases, your being unable to say anything about
one or more of the four forms of knowledge outlined above is not a reason for doubting
that you have such knowledge; and even when you can summarize some of your knowl-
edge in propositional form, no one supposes that this is somehow equivalent to
possessing the knowledge itself.15
It might be objected that the outcomes of the interview thought-experiment hardly
amount to knowledge. Do I really know what the interview will be like or merely what
I think it will be like? Do I really know how tactfully to discourage a line of questioning?
And so forth. In the light of these objections, shouldn’t I describe the outcomes as, at
best, belief or opinion?
These objections could also be raised about learning from any kind of simulation.
Might not my experience on an artificial climbing-wall give me the illusion that I can
climb a real rock face? Might mastering the 747 flight-simulator give the completely
misleading impression that I can fly a 747? Of course, both might be deeply inaccurate
and misleading, but it’s easy to see how errors can be spotted and corrected. The same
considerations apply to our own thoughts. In the general case, it would be very surprising
if millions of years of evolution had left us with a fundamentally flawed simulation
system; and in the individual case, experience can teach you where your imaginings tend
to be unrealistic, and you can learn to correct for this. No simulation can infallibly
guarantee future performances (even past practice in real situations cannot guarantee


that), but experience still shows that someone who has only practised on a 747 simulator
is more likely to fly the real plane better than someone with no practice of any kind; and
that thinking through a problem in advance, soberly and carefully, is more likely to have
a happy outcome than not thinking at all.
The idea that having less than an infallible justification for a knowledge-claim
demotes it to belief or opinion is based on knowledge-claims about individual proposi-
tions, and is not a reliable guide to all knowledge-claims for exactly this reason. I
currently don’t know, for example, that Newcastle has a larger population than York, and
I would be inclined to say that I think or believe that Newcastle has the larger population.
But in the four cases of knowledge described above, there is no falling back to a weaker
cognitive state like belief or opinion, and knowledge is a matter of degree. In tennis, you
can know how to play some shots but not others; your knowledge of Paris can be super-
ficial; your advice to others about their love lives may occasionally be obtuse; your grasp
of what it means to be an exile can be inadequate, but in all these cases you can still be
said to have some knowledge of tennis, Paris, and so on. It’s also worth observing that,
in the last two cases, we might be tempted to describe our cognitive state as one of
understanding rather than knowledge, a temptation which does not exist in the case of
asserting a proposition.
Aesthetic interest in a work requires that you should want to linger with or return
to it. If, therefore, a desire to pick up straightforward facts from a work were part of
aesthetic interest, then it would be impossible to explain why you should want to
linger with or return to a work when you already know all the facts it has to impart.
Factual knowledge is an all-or-nothing affair – you either know it or you don’t – but
the four kinds of knowledge sketched above are matters of degree, and they can always
be sharpened, deepened, honed, and made more intense by continued exposure to a
rewarding work. This is why curiosity of this kind can be an important element in
aesthetic interest.

I might decide to write up some of the interview scenarios described above: perhaps as
a series of short stories. Clearly, in writing up an imagined scene, I will have to be
extremely selective: any scenario can contain an infinite number of potential features
(including the weave of the interviewers’ clothes, the pattern of dust motes in the air,
the grain of the desk’s wood) and part of being a competent fiction-writer is selecting
just those elements which will instantly evoke countless other salient features in the
reader’s mind. Thus if I write well, my text should allow other people to reconstruct
relevantly similar imaginings on the basis of my description, and perhaps benefit from
any knowledge I’ve achieved.
The text is merely a means to an end, and it is the reader’s imaginative engagement –
his ability to fill out, colour, concretize and realize the action, people and places
described – that possesses the genuine probative power. At most, a work of literature
offers a bare framework in which to imagine, a series of limited prompts; it is not a report
of knowledge achieved (although some works also contain this) but a heuristic device
designed to allow the reader to achieve knowledge. These prompts, however, can enable
the reader to imagine possibilities of his own being he could not have imagined unaided;
and they allow him to imagine with a freshness, depth, duration, and intensity that
would have been quite impossible without the collaboration of the author.


There is ample empirical evidence to support the idea that engaging with literature
involves imagining the scenes and people described, and that this phenomenon can lead
to an expansion of imaginative possibilities. Consider this passage from Richard Church’s
autobiography, Over the Bridge, which describes the occasion when, as a small child, he
suddenly discovered he could read:

I picked [the book] up, opened it at the first page, and began to read The Swiss
Family Robinson. It is an understatement to say that I began to read. I passed
into another life. I was one of that family on the wrecked ship, passing through
the barrier of words, enlarging my small suburban existence by this new dimen-
sion. I did not know what was happening, or the scope of this vast inheritance.
I heard the sea breaking on the shore of that fortunate island and shared with
Fritz, Ernest, Jack and Franz in establishing ourselves under the palm trees, and
in offering up thanks for our safety. [. . .] At that stage the enlargement was not
literary. The words themselves were not endeared to me. I had merely caught
the knack of using them. [. . .] Mother found me sitting at the table with the
book in front of me, and my breathing body in front of the book.16

The elderly Erich Heller stands at the greatest possible distance, in terms of literary and
philosophical sophistication, from the young Richard Church, and yet the critic
evidently derived an equally intense imaginative experience from reading Knut
Hamsun’s Hunger; indeed, the experience was so intense that it stayed in his mind long
afterwards like a sequence of experiential memories:

[The novel] vigorously attacks our senses with Christiania’s sights, sounds and
smells. For days on end after closing the book, we feel we have actually met its
shopkeepers, pawnbrokers, policemen and prostitutes; climbed staircases to
deserted addresses on streets given by name, trying to find a long neglected
friend who may help us, desperately hungry as we are, to some food; or sit on a
bench at the harbour, watching the ships on the mother-of-pearl coloured sea.17

The reader’s thought has to be generated spontaneously and unselfconsciously; he

cannot bring it about through an act of will. If I ask you now to imagine a tiger, then the
image you produce will be utterly feeble. If, on the other hand, I ask you to recall how
completely absorbed you can became in a long delightful fantasy, or a series of horrible
imaginings, then we remember that imagination can be utterly engrossing, vivid, and
disequilibriating. Indeed, when disturbed in the middle of such imaginings, we can feel
ourselves jolted out of another world, and can be quite shocked and disorientated to
discover ourselves amidst our ordinary surroundings. Perhaps the most intense kind of
imaginative experience is the dream or nightmare; the probable reason for this being
that our normal perceptual engagement with the external world is temporarily switched
off, and there is no possibility of self-consciousness or any other form of distraction from
the experience we are currently undergoing. Not being aware you are imagining, seems
to be a prerequisite for imagining vividly.
Thus a work of literature will only act as a probative instrument when you are
completely absorbed in it; which is another way of saying that it will only act as such an
instrument when you are taking pleasure in it. Pleasure is a kind of non-voluntary
absorption, and it is only in this state that imagination fires and knowledge can be


achieved. I think this gives some understanding of the traditional idea that the arts give
knowledge through pleasure: it is only when we are engrossed, absorbed, ‘lost in a book’,
‘taken out of ourselves’, that we experience the full cognitive benefit of literature.
These considerations might explain why we find the Harvard Man in Catch-22 ­–
“who knew everything about literature except how to enjoy it” – troubling.18 You’ll
certainly be more likely to be good at maths or archaeology if you enjoy these subjects,
but an excellent mathematician or archaeologist who doesn’t enjoy his work is not an
impossibility. However, a professor of literature who had never become engrossed and
absorbed in what he reads seems like a man who had never imaginatively engaged with
what he reads, and thus one who could, amongst other things, never learn anything
distinctive and significant from literature.

The object of music is aural; the objects of dance, drawing, and painting are visual; the
objects of opera and ballet are clearly a combination of the two; and the object of
sculpture can contain both visual and tactile components. But to which sense, or senses,
does literature appeal? Films and theatre plays are both aural and visual, but non-dramatized
literature has a bifurcated object: its linguistic component can be encoded in different
media (spoken language, print, Braille, etc.)19 but its other – and more important –
component is the thoughts evoked in the mind of the reader. Some of these thoughts
will be abstract and conceptual; but, because literature is concerned with the subjective
point of view, many of them will be thoughts about what something looks like or sounds
like. To think what something looks like is to have a visual image; to think what it
sounds like is to have an auditory image; and the same applies to thoughts about the
objects of the other senses. There is thus some truth in another traditional idea – that
literature gives us ‘the pleasures of the imagination’.
Although we can still experience the sense of being jolted back into the real world
when our engrossed reading is interrupted, I think it has to be admitted that our
imaginative engagement with a text only rarely achieves the kind of intensity found in
fantasies and horrified forebodings. This is partly because our thoughts and mental
imagery are subject to continual modification and revision as we read further into a text,
but mainly because imaginatively responding to literature is a more complex psychological
state than fantasizing.
In the literary case, we must be absorbed in our imaginings while being simultane-
ously responsive to the words which guide and generate them; somehow or other, we
must be imagining one thing, and perceiving something else, at the same time. There
are two kinds of split consciousness described in the literature which make useful staring
points for considering this complex state:

1. Philosophers of mind commonly distinguish between two states of consciousness.

Consider a truck driver on a long, boring but complicated journey who is engrossed in
an important mobile-phone conversation throughout.20 At the end of the journey, he
might be quite unable to remember slowing to stop at a certain set of traffic lights or
waiting to turn across the oncoming lane, but he nonetheless accomplished all these
manoeuvres, avoided collisions with other traffic, and did not run over any pedestrians.
On the other hand, he might recall every detail of the phone conversation. We might
describe the state of awareness he is in when on autopilot – responsive but unaware – as


consciousness2; and we might describe the state of consciousness he directed towards

the phone call – where he is both aware and responsive – as consciousness1.21
2. There are cases where an outside stimulus is transmuted and incorporated into a
dream. For example, in a nightmare you are blinded by car headlights, and you
immediately wake to realize that someone has just switched on the bedroom lights;
or – a case mentioned by Freud – someone has water dripped on his forehead and
dreams he is sweating violently and drinking Orvieto wine.22

The state of consciousness I am in while engrossed in a narrative is more complex still.

The truck driver’s consciousness1 and consciousness2 are quite independent; the phone
call and the road conditions may have absolutely nothing to do with one another. But
my literary imaginings are clearly the result of the words I read, even though I can – at
the level of consciousness1 – be unaware of them. In one way the dream case is closer
to the reading experience because the outside stimulus does affect the content of the
dream. There are, however, several important differences: there is a serious physical as
well as psychological divide between waking and sleeping, which is not found when we
pass from pure imagining to awareness of the words which produced it; dream-responses
to external stimuli, unlike a reader’s imaginative responses to words, are unpredictable
and partial; and imaginative responses to words, unlike dream-responses, are clearly
based on convention.
How conscious we are of the words we read while absorbed in them can depend on
many factors, but age and genre are particularly significant. Richard Church, in the
passage quoted above, says that his youthful engagement was “not literary[,] the words
themselves were not endeared to me[,] I had mere caught the knack of using them”, and
I think we can all recall reading 1984, or The Three Musketeers, or the second half of A
Tale of Two Cities for the first time, when they seized us by the scruff of the neck, and
seemed to haul us, willy-nilly, through a sequence of appalling and appallingly exciting
events. We had no concern for, and paid no conscious attention to, style or artistry, but
it was of course a feature of their style and artistry that both qualities seemed to disappear
to reveal the events depicted with film-like immediacy and clarity. Church’s description
of his own experience as “not literary” is not dismissive – indeed, I’m sure he thinks it’s
deeply enviable – but emphasizes that it was the imaginative experience, rather than the
words or the way the words gave rise to the imaginative experience, which was the focus
of his interest.
Characteristically, the adult experience of reading poetry is rather different. In this
case, the intensity of the imaginative experience can still sometimes take the breath
away, but at the same time we seem to be equally focused on the language, and how the
language gives rise to the imaginative experience. Thus I find Heaney’s lines, “ [. . .] the
squelch and slap / Of soggy peat, the curt cuts of an edge [. . .]”23 deeply evocative; I can
almost feel the moist squelchiness of the turf bog, and hard sharp edge of the spade
cutting through it. And one way I’d want to explain the almost palpable sense of reality
the lines evoke is by noting the onomatopoeia: the moist, whole-mouth-involving
alliterative syllables of “squelch”, “slap” and “soggy”; and then, by contrast, the short,
clipped, front-of-mouth, thin-lipped syllables of “curt cuts of an edge”. This experience
is literary in Church’s sense; it not only involves pleasure in the intensity and freshness
of the experience, but pleasure in the words and how they function.
The ease with which these states can be brought about is aided by how boring the
printed page looks; people who cannot read cannot begin to imagine what pleasure


could possibly be derived from looking at complex patterns of grey marks for hour after
hour. (The same applies to listening to a long story in a language you don’t understand.)24
Like the truck driver’s numbing road conditions, this enforced state of sensory depriva-
tion makes the imagination more lively, and supplies a wonderfully undistracting
background against which our imaginings can play out. Of course, when you understand
what the words mean, and when they evoke experience for you, they too can come to
seem fascinating (like the Japanese paper crumbs mentioned in Proust which, when
dropped in water, reveal themselves to be delicate artificial flowers, people, houses, and
so forth).25

The literary experience has two phases. On the one hand, there is the phase when we
should just try to give ourselves up to a work, to become absorbed and lost in it. It is
clearly a criticism of a work if this proves impossible: if it is unevocative, dead, unre-
warding, a permanent chore to read. On the other hand, there is a critical and
reflective phase, and this will often be spent away from the text – pondering on the
work while looking out of a window, talking to friends about it, or reading relevant
critical commentary.
The kind of criticism which is appropriate to a work will depend on the kind of
absorption it supports. Some criticism will restrict itself to the level of the imaginative
experience – how coherent it is, and how well it corresponds with reality – and will have
very little to say about the physical words and patterns on the page.26 Criticism which
restricts itself to examining character, plot, atmosphere, and setting in this way tends to
focus on novels and short stories, and often applies equally well to their adaptions as
plays, films, operas, ballets, and so on. Another kind of criticism focuses not only on the
imaginative experience but on how the words on the page give rise to it. Discussions of
poetry, for example, will characteristically examine not only the imaginative experience,
but also the diction, alliteration, onomatopoeia, assonance, rhythm, rhyme, stanza
structure, form etc. which brought it about.
A work might absorb us on first reading, but critical conversation or reading after-
wards convinces us it is false, clichéd, incoherent, unamusing, implausible, over-simple,
clumsy, and so forth. In this case, we might find ourselves unable to read the work with
pleasure again; or, if it continues to grip, we can either question the critical responses
we’ve heard, or acknowledge that the work is very seductive but should be treated with
caution for that very reason – it should be regarded as a fantasy (droll or otherwise)
rather than a thought-experiment.
With any significant work, we are likely to pass through these two phases on many
occasions in the course of a lifetime.

The discussions of pleasure and knowledge in previous sections help us to give a satisfying
answer to the so-called paradox of fiction (why do we care about fictional events and
people?) and the so-called paradox of tragedy (why do we take pleasure in the depiction
of horrific and harrowing events?).
Humans did not develop amidst an environment of mirrors, simulators, holograms,
and IMAX cinemas, and there was thus no reason for our senses to develop a fine-tuned


capacity to discriminate between real events and accurate simulacra; our emotional
responses, which are closely connected to the senses, are therefore not good at dis-
criminating between real events and simulations. Although, when watching a fictional
play, we know intellectually that what we see is a representation, our senses and emotions
still tend to function as if we were witnessing real events: we flinch at pretend dagger-
thrusts and smile when delightful fictional characters marry. It is also worth pointing out
that the normal theatre and cinema environment and conventions deliberately prevent
normal probative behaviour: the lights focus the senses’ attention by illuminating the
action but leaving the audience in darkness; and we are prevented from consulting,
picking things up, sniffing and tapping them, as well as moving about so that things can
be seen from the back as well as the front.
Our involuntary responses are hair-triggered, undiscriminating, and hard to disable for
good evolutionary reasons. A person who sees a snake-like object and jumps, only to
discover moments later that he has responded to a piece of rope, has expended a certain
amount of unnecessary energy, and may feel rather foolish, but he has lived to tell the
tale. A person who does not jump when he sees a snake-like object, but decides to check
whether action is really warranted before jumping, could well end up dead. Conversely,
our voluntary and deliberate actions tend to come about more slowly, but are easily dis-
engaged and much more sensitive to small perceptual – as well as intellectual – differences
between environments.27 In the theatre, therefore, we may well gasp, jump, clench our
knuckles, and cry at a stage murder; but we feel no temptation to call the police or warn
the management that a murder is taking place.28 It must be admitted, however, that even
reactions and emotional responses operate in a slightly damped-down form: we can be
shocked by a stage murder, but not as shocked as by a real murder; and the misery caused
by the death of an engaging character may linger long after we have left the theatre, but
it will not be as intense or protracted as that caused by the death of a real friend.29
The evolutionary aberration which makes our reactions to what the senses tell us
insensitive to the difference between real events and simulations, has the unexpected
advantage that it allows us to feel some emotion even when just imagining events, and
thus gives us important hints at what emotions we are likely to feel in a real case, and how
best they might be dealt with. This looks like a case of evolution turning an old
drawback into a new advantage.30
From evolution’s point of view, nothing is more natural than that we should want to
avoid horrific and dangerous situations. On the other hand, nothing is more natural
than that we should want to know about horrific and dangerous situations;31 in fact,
knowing about this kind of situation is more important than knowing about any other
kind of situation. Thus our attraction to and repulsion from danger have the same
motivation: we want to avoid all intimacy with danger, but we can only do this effectively
by knowing danger intimately.32
Our curiosity about horror and danger is not left to chance or means–end motivation;
we have a deep-seated instinct for such knowledge. When a fight is beginning to develop
in a bar, you can sense all the other people in the bar involuntarily losing the threads of
their own conversations and focusing in on the group where trouble is brewing. If you
are in the vicinity of a nasty accident, then there is an almost insatiable desire to know
what kind of accident it was, what caused it, how many people were hurt, and so on,
even if you’re not in any position to help.
In witnessing a real case of suffering or danger, we are likely to feel torn. On the one
hand, we are prone to be overwhelmed by curiosity and a desire to see how events play


out. But the desire to indulge this impulse is likely to be interfered with by a number of
more practical concerns: we will think how we can offer comfort or practical help; there
may well be imminent danger to ourselves, in which case we need to think about flight
or protection; staying around to watch may well hinder the rescue; we might feel vulner-
able to the accusation of gloating over others’ suffering. Watching a newsreel of some
horrific event in the past or some distant place solves all these practical problems except
the possible accusation of feeding off others’ misery; watching a fictional case of horrific
suffering resolves all of them. Thus fiction alone allows us to indulge pure and unfettered
curiosity about human vulnerabilities.
In order to understand our response to fictional tragedies we need to grant three
things which I have already argued for: first, that things which can grip, absorb, exhaust,
interest, and appall can be pleasures, in particular, strenuous pleasures; second, part of
the importance of art – specifically literature – is that it gives experiential and practical
knowledge; third, the things which humans most want to avoid are the things they also
feel the deepest need to know about (and thus to seek them out). Once we see our
attraction to fictional tragedies in this light, then tragedy ceases to be a paradoxical case,
requiring various additional theories to explain it away, and becomes an absolute
paradigm of art and our response to it.33

  1 Looking hungrily or lustfully at pictures of food or attractive people is clearly an inferior substitute for
looking at food or attractive people, and such substitutes automatically frustrate the desires which
prompted interest in their objects. This is clearly not the case when we take an aesthetic interest in
pictures or sculptures.
  2 This section relies on my “The Definition of ‘Art’ ” in M.W. Rowe, Philosophy and Literature: A Book of
Essays (Aldershot: Ashgate, 2004). This earlier article goes into considerably more detail about the
distinction between aesthetic and non-aesthetic looking. It considers, for example, why looking
prompted by pride, or by curiosity about objects associated with historically important events, is not
(purely) aesthetic.
  3 The idea that pleasure is a form of attention was suggested by Aristotle. See his Nicomachean Ethics, Book
VII, (trans.) J.A.K. Thompson and Hugh Tredennick, (ed.) Jonathan Barnes (Harmondsworth: Penguin,
2004), pp. 167–199. The notion has been explored in modern times by, amongst others, Gilbert Ryle
(“Pleasure,” in Moral Concepts, (ed.) Joel Feinberg (Oxford: Oxford University Press, 1969), pp. 19–28),
and Terence Penelhum (“The Logic of Pleasure,” in H. Gustafson (ed.) Essays in Philosophical Psychology
(London: Macmillan, 1964), pp. 227–247).
  4 The twin notions of a competent judges and convergence are examined in my “ ‘Of the Standard of
Taste’: Decisions, Rules and Critical Argument,” in Alan Bailey and Dan O’Brien (eds) The Continuum
Companion to Hume (London: Continuum, 2012), pp. 349–363.
  5 Wordsworth also finds it necessary to distinguish between two kinds of pleasure. On the one hand he
deplores those “who talk of Poetry as a matter of amusement and idle pleasure; who will converse with
us gravely about a taste for Poetry; as they express it, as if it were a thing as indifferent as a taste for Rope-
dancing, or Frontiniac or Sherry” (“Preface to the Lyrical Ballads (1802)” in Stephen Gill (ed.) William
Wordsworth, The Oxford Authors (Oxford: Oxford University Press, 1984), pp. 595–615). On the other,
he urges us not to think of the “necessity of producing immediate pleasure [. . .] as a degradation of the
Poet’s art. It is far otherwise. [. . .] It is a homage paid to the naked and native dignity of man, to the grand
elementary principle of pleasure, by which he knows, and feels, and lives, and moves” (ibid., p. 605). This
latter conception of pleasure comes to seem even grander when we realize his is alluding to Acts: 17:28,
where St Paul tells us that “we live, move, and have our being” in God.
  6 Limitations of space make me confine my discussion to a purely Newtonian conception of physics and
  7 For an excellent discussion of these issues see Colin McGinn, The Subjective View: Secondary Qualities
and Indexical Thoughts (Oxford: Clarendon Press, 1983).


  8 As Wordsworth puts it: “The appropriate business of poetry (which nevertheless, if genuine, is as
permanent as pure science) her appropriate employment, her duty, is to treat of things not as they are,
but as they appear, not as they exist in themselves, but as they seem to exist to the senses and the passions”
(“Essay, Supplementary to the Preface to Poems (1815)”, p. 641).
  9 Both phrases require some care: the objective point of view is neither a point nor a view; and there is
clearly more than one subjective point of view – in fact, an enormous number.
10 The topics of the last two paragraphs are treated in greater depth in “Poetry and Abstraction” in Rowe,
Philosophy and Literature, pp. 165–181.
11 For a more extended discussion of these issues, focusing on a single example, see my “ ‘Here’: Avoiding
the Subject,” in M.W. Rowe, Philip Larkin: Art and Self (Basingstoke: Palgrave, 2011), pp. 6–43.
12 Clearly, a work of art can be valuable for a number of reasons simultaneously: it might, for example, be
both truthful and funny.
13 I draw on Aristotle: “[The] poet’s function is to describe, not the thing that has happened, but a kind of
thing that might happen [. . .].” See The Art of Poetry, section 9, (trans.) Ingram Bywater (Oxford:
Clarendon Press, 1920), p. 43.
14 See Aristotle, The Nicomachean Ethics, (trans.) J.A.K. Thompson and Hugh Tredennick, (ed.) Jonathan
Barnes (Harmondsworth: Penguin, 2004), Books VI and VII, pp. 144–199.
15 For a more detailed discussion of the various kinds of knowledge we might derive from literature, and
which are compatible with the aesthetic attitude, see my “Literature, Knowledge, and the Aesthetic
Attitude”, Ratio, vol. XXII, no. 4, December 2009, pp. 375–397; reprinted in Philosophy of Literature (ed.)
Severin Schroeder (Oxford: Wiley-Blackwell, 2010), pp. 1–23.
16 Richard Church, Over The Bridge (London: Heinemann, 1955), p. 78.
17 Erich Heller, ‘Knut Hamsun’ (pp. 163–176) in his, In the Age of Prose: Literary and Philosophical Essays
(Cambridge: Cambridge University Press, 1984), p. 167.
18 Joseph Heller, Catch-22 (London: Vintage, 2010), p. 78.
19 It is possible to imagine language represented in tastes and smells (in a form of Morse code for instance)
and this suggests that literature, as well as having the imaginary objects of all the senses (amongst other
things) as its objects, is the only art which could be accessed by each of the senses. There could be some
aesthetic loss in passing from one sense-modality to another, but there is some aesthetic loss when a poem
is heard rather than read (and vice versa): certain puns, for example, disappear.
20 See D.M. Armstrong, “What is Consciousness?” in John Heil (ed.) Philosophy of Mind: A Guide and
Anthology (Oxford: Oxford University Press, 2004), pp. 610–611.
21 My terminology follows Daniel C. Dennett’s ‘awareness1’ and ‘awareness2’ in “Reply to Arbib and
Gunderson,” (pp. 23–38) in his Brainstorms: Philosophical Essays on Mind and Psychology (Brighton:
Harvester Press, 1979), especially pp. 30–33.
22 Sigmund Freud, (trans.) James Strachey, The Interpretation of Dreams (Harmondsworth: Penguin, 1986), p. 86.
23 Lines 25–26 of “Digging” (pp. 3–4), from Seamus Heaney, New and Selected Poems 1966–1987 (London:
Faber and Faber, 1990), p. 4.
24 It’s particularly hard for a literate native English speaker to say what English sounds and looks like – we
seem to hear and see straight through to the meaning.
25 Marcel Proust, In Search of Lost Time, Volume 1, Swann’s Way, (trans.) Charles Scott Moncreiff, Terence
Kilmartin and D.J. Enright (London: Vintage, 2005), p. 55.
26 The look and sound of the words will always be relevant to a literary discussion, but such considerations
are more frequently mentioned in some contexts than others.
27 This distinction between the intellectual and the instinctual is based on the view argued for by Jenefer
Robinson in Deeper than Reason: Emotion and its Role in Literature, Music and Art (Oxford: Clarendon
Press, 2005), especially pp. 28–99. This is one of the best and most stimulating books on aesthetics and
the philosophy of mind published in the last sixty years.
28 The difference between the two kinds of response is emphasized in Kendall Walton’s “Fearing Fictions”,
Journal of Philosophy 75 (1978), pp. 5–27.
29 From this paragraph on, my thoughts have been so heavily influenced by discussions with Katherine Allen
that I can no longer clearly distinguish between her ideas and mine. For a detailed and thoroughly worked
out account of human responses to horror and tragedy along the lines sketched below, see her unpublished
D.Phil thesis, “What Joy from Misery: The Pleasures of Horror” (University of East Anglia, 2012).
30 In this section, I have largely concentrated on theatre and film, but I hope it’s obvious that I think the
sensory imagination which is engaged by non-dramatic literature has closer connections with the senses,
and thus the emotions, than what is intellectually and abstractly known.


31 Again, Aristotle lies behind this view: “[Though] the objects themselves may be painful to see, we delight
to view the most realistic representations of them in art [. . .] The explanation is to be found in a further
fact: to be learning something is the greatest of pleasures [. . .]” The Art of Poetry, section 4, p. 28.
32 We can find the same kind of conflicted attitudes, and for the same reason, in non-aesthetic contexts.
It’s something of a mystery, for example, why we should want to ride in frightening roller coasters when
we normally go to great lengths to avoid frightening experiences. As in the case of tragedy and horror, I
suspect we seek out experiences which the senses and emotions tell us are terrifying, but which at the
same time we know intellectually to be perfectly safe. We can thus acquire experiential knowledge of
danger while avoiding danger.
33 If your theory of art also requires a theory of tragedy, then there’s something wrong with your theory
of art.

Robert Chodat

Philosophers have been interested in novels but seldom in the novel. Fiction and
narrative have earned considerable attention, the former in philosophy of language
from Frege and Russell to Kripke and Lewis, and the latter in recent philosophy of the
self.1 But which kinds of fiction, which sorts of narrative? These questions have tradi-
tionally not played a substantial role, whether the issue is semantics or virtue. And the
result is that, while philosophers have regularly referred to particular novelists, and at
times even offered extended commentaries on them—lively disputes have come to
swarm around James, Proust, and Coetzee in particular—they have often betrayed what
Northrop Frye once called a “novel-centered view of prose fiction” (1957: 304),
ignoring not only, say, fables, tales, folklore, medieval mystery plays, staged dramas,
closet dramas, Renaissance romances, horror stories, fantasy, young-adult fiction, and
other forms, but also the broader question of why these other forms should seem to
speak to so many fewer philosophical topics than does the novel, and should provide
so many fewer examples.2
This marginalization of genre can often be warranted; it may reflect a healthy respect
for concrete examples and anomalous cases. But to understand what profit may be had
from speculating on the logic of the genre, what might be gained from asking about the
conditions and procedures and subjects and goals of composition that “novels” inherit
and share, imagine one meets a young person who introduces herself as a writer, and
who, when asked what she’s currently writing, says: “An epic.” One might be forgiven
for responding with a giggle or raised eyebrow. An epic—like Homer, Virgil, Milton?
Here? Now? Imagine, by contrast, that she says she’s writing a novel. Why would this be
less comical? Why would we nod and wish her luck?3
Versions of this question have been common at least since Henry Fielding claimed
in the preface to Joseph Andrews (1742) to be writing a “comic epic in prose.” “Epic,”
for starters, is what we call texts that habitually portray gods and supernatural
beings, who sometimes appear before human characters and often intervene into
human affairs—causing, say, the wind to alter the course of ships. Epics also focus
on characters of extraordinary stature, “heroes,” figures who exist somewhere
between the gods and the rest of us, and whose experiences determine the fate of
entire peoples. They are the kings, aristocrats, and warriors whose lofty character
and social station reflect a purposeful natural order rather than the contingent jumble
of genes, social environment, and dumb luck that we nowadays use to explain
Robert Chodat

behavior. Moreover, audiences are encouraged to admire the virtues of these personages
and to appreciate the magnitude of their actions, even when vicious. Accordingly,
the language of the epic is “high,” with august invocations to the gods and lavish
acclaim for exceptional deeds.
No doubt such texts can help us elucidate the concepts “fiction” and “narrative,” but
surely their full philosophical significance is bound to be different from a more or less
fictional narrative in which such features are often absent—as, very roughly speaking,
they tend to be in works we call “novels.” As I’ll be suggesting here, novels can be under-
stood as the kinds of narratives in which each of the elements I’ve just sketched has, to
varying degrees and in various ways, receded: the clarity of the metaphysical order, the
clarity of the human order, the clarity of the linguistic order. In their place novels tend
to express a deep agnosticism about ultimate questions, a radical uncertainty about our
politico-moral norms, and a keen awareness of dizzyingly different languages, dialects,
and tones.4
Such traits are sometimes said to make the novel quintessentially “modern.” One
version of this claim was popularized by Ian Watt in The Rise of the Novel (1957), which
took eighteenth-century British novelists as the model, arguing that Defoe, Richardson,
and Fielding registered the shift to an increasingly bourgeois and empiricist culture.
Earlier incarnations of the idea are found in Walter Benjamin, who contrasted novelists
with pre-modern storytellers, and Georg Lukács, whose Theory of the Novel (1920)
moved from a famously lyrical account of “integrated” classical civilizations, where “the
structures made by man for man” are “his necessary and native home,” to the fractured
modern world of Don Quixote (1605–15), where “the extensive totality of life is no
longer directly given” and the “problematic individual” struggles with “the absence of
any manifest aim, the determining lack of direction of life as a whole” (1971: 56, 62).5
Other commentators, however, resist the modernity thesis. Here, the most sophisticated
case comes from Mikhail Bakhtin, who identified the novel with those moments across
Western history when political, cultural, and linguistic hierarchies began to fray: not just
the seventeenth and eighteenth centuries, but also the late medieval texts of Rabelais
and, even further back, such ancient fictions as The Golden Ass (c.150 CE) and Aethiopica
(c.250 CE). For Bakhtin, the novel is less a set a set of formal features than a centrifugal
force that periodically gains momentum, undermining the dominant culture’s efforts at
unification and homogenization. Several recent scholars have embraced Bakhtin’s
sweep, claiming that epics and novels are not discrete categories but areas on a spectrum
comprising both “modern” and “pre-modern” texts, including ancient Egyptian narra-
tives, medieval romances, and Renaissance dramas.6
Such historical arguments can quickly calcify. Obviously “novel” is not a neatly
bounded concept, and with Bakthin and his heirs we should recognize that its various
exemplars over the last few centuries have an indefinite number of precedents. At the
same time, it’s telling that philosophical aesthetics from Aristotle to Kant said nothing
consequential about any novel, ancient or modern, and that the genre becomes an
explicit philosophical theme only between the 1790s, when Friedrich Schlegel called
novels “the Socratic dialogues of our time” (1968: 123), and the 1820s, when Hegel
referred to the genre as the “bourgeois epic” (1975: 1092).7 It’s similarly telling that one
of Nietzsche’s expressions of contempt for modernity involved denouncing the novel as
a descendent of the Platonic dialogue, “an infinitely enhanced Aesopian fable”
(1999: 69), and dismissing one of its great figures, George Eliot, as one of “the little moral
females” (2005: 193). Novels, that is, may have no single essence or ancestor, and their


development may not yield a tidy historical narrative, but when read from a certain
altitude, the books of Cervantes, Defoe, and Fielding—and later of Austen, Flaubert,
Eliot, Woolf, Mann, Joyce, Morrison, McEwan—do offer something importantly differ-
ent than most earlier forms, and this difference has something to do with the gradual yet
decisive changes to our conceptions of the natural world, language, knowledge, personal
identity, moral goodness, socio-political authority, and the divine. The genre’s most
judicious histories, then, will have a Hegelian ring: the forces that help engender the
novel have always been incipient, but over time these forces have gained self-consciousness,
intensifying to the point that recognizably novelistic forms are not just marginal artifacts
but instead, as Anthony Trollope put it in 1870, “in the hands of all, from the Prime
Minister . . . down to the last-appointed scullery maid” (1938: 108).
What matters most here is less the history of the novel per se than the family of features
that virtually all sides in the argument recognize as its leading characteristics—namely,
the various types of irony and ambivalence that I noted above. It’s this quality that
chiefly attracted Schlegel, but it’s also what arguably leads contemporary philosophers
to avoid speculating on the novel genre, even as they mine particular novels for examples.
For this quality reflects the novel’s capacity to extend and even radicalize the ancient
quarrel between poetry and philosophy. (Not for nothing is Schlegel’s status as a
philosopher controversial.) Philosophers have frequently employed narratives, Plato
most famously, but these stories are typically placed within a framework of assertion and
argument, which constrain the wayward ideas and emotional responses that narratives
potentially excite. The novel is the genre in which these discursive frameworks most
visibly corrode: the “point” of a given story becomes less available to definitive explica-
tion, candidates for permanent or universal truth grow ever harder to discern, and
scrupulous descriptions of actions and events tend to muffle any assertoric claims that—
as in Tolstoy, Proust, or Musil—might be offered. The groping individual reader is left
to draw the conclusions that Socrates once elucidated with dialectical authority. That
such authority is in dispute is perhaps why the philosophers who appear in novels come
off little better than Socrates does in The Clouds: Eliot’s Casaubon, Woolf’s Mr. Ramsey,
Bellow’s Herzog, Murdoch’s Razamov. Whatever self-criticisms have filtered into recent
philosophy, the ideals of permanence and universality continue to be central to the
disciplinary self-image. Of concern is not the mind or language of any particular person
or culture, but “mind” and “language” as such—which, today, often means rearticulating
what’s discovered by the cognitive and evolutionary sciences. Hence the most sustained
reflections on the novel genre remain those in the post-Hegelian tradition and the
various literary-theoretical schools this tradition has inspired, whereas analytic aesthetics
has mostly limited itself to the type of trans-historical categories that Anglo-American
philosophy as a whole tends to esteem.

At the end of The Golden Ass—one of the classical “novels” that Bakhtin sought to
retrieve—a goddess changes the picaresque protagonist from a donkey back into a
human, and in return for his transformation he is initiated into the cults of both Isis and
Osiris. Near the end of the Song of Roland (c.1100), the Franks are aided in their pursuit
of the Saracens when the sun miraculously fails to set, and in its closing passage the
angel Gabriel visits Charlemagne in a dream to tell him of future battles he must wage.
Half a millennium later, Macbeth (1605) opens with three witches prophesying, and the

Robert Chodat

morning after King Duncan is murdered, horses are said to break from their masters and
“darkness does the face of earth entomb” (II.iv.9).
Such happenings are what led commentators from Watt onwards to associate the
novel with the development of modern empiricism. In the century after the founding of
the Royal Society, little of the palpably supernatural is on display in Defoe, Richardson,
and Fielding, and the same is largely true for most of their nineteenth- and twentieth-
century British descendents. When efficient causes come to take precedence in the
order of explanation, the moral and the natural orders are no longer easily correlated,
and one of the many results is a shift in the notion of “realism,” from the medieval belief
in universals to the modern aesthetic notion of something causally explicable, plausible
within a baldly naturalistic universe.8 Lukács is thinking chiefly of continental fiction,
and his idiom is more dramatic than Watt’s, but the idea is similar: if in earlier genres,
he says, “a god always plots the hero’s paths and always walks ahead of him” (1971: 86),
the novel’s universe is one in which “the immanence of being” (1971: 71) fails “to enter
into empirical life. . . . The novel is the epic of a world that has been abandoned by
God” (1971: 88).
Clearly most contemporary Anglo-American philosophy is a product of the shift that
Watt and Lukács identify. But while novels share this basically naturalistic tempera-
ment, they tend to express more reservations than Russell, Carnap, and their heirs about
the scope and ambitions of modern science, and tend to be far less categorical in their
assumptions about what’s real and possible. As Lukács says, the novel is “the epic of an
age in which the extensive totality of life is no longer directly given . . . yet which still
thinks in terms of totality” (1971: 56). As a genre, that is, it may not always make open
avowals of faith or overtly align the terrestrial and heavenly realms, but it expresses an
interest in the teleologies and explanations that both science and scientific philosophy
have largely banished.
Such agnosticism can take several forms, the most obvious of which is the novel’s
persistent thematization of religious questions. From Melville and Tolstoy to Gide and
Joyce to John Updike and Marilynne Robinson, the struggle for faith in a desacralized
world is a recurring preoccupation. At times these religious impulses can lead a novelist
into pledges of doctrinal allegiance. “Actually I do not consider myself a novelist,”
Walker Percy once admitted, “but a moralist or a propagandist” trying “to tell people
what they must do and what they must believe if they want to live” (Tolson 1992: 300); and
comparable proclamations are found in, say, Dostoevsky and Flannery O’Connor. But
struggles for belief in novels themselves—including in those of Dostoevsky, Percy, and
O’Connor—tend remain just that: a struggle. For most novelists most of the time, the
pull of modern disenchantment is too potent to be ignored entirely, its naturalistic
notions of plausibility too deeply ingrained; and the result is the liminal position that
Lukács evokes: a meaningful “totality” is unreachable, even as it hovers over the
characters and narrators as an ideal or aspiration.
A second type of metaphysical agnosticism is visible in the haunted mansions, ghosts,
paranormal creatures, ritualism, decadence, and intense violence that mark certain texts
from the mid-eighteenth century onwards. Such features arose in the “Age of Reason”
to contest our post-Galilean assumptions about what’s precisely explicable, and these
ways of dramatizing the limits of human cognition and rationality deeply shape the work
of Dickens and Dostoevsky, not to mention novels about vampires or Mr. Hyde.
Americans are sometimes said to be particularly given to this mode (Poe, Hawthorne,
Faulkner, Morrison) (Chase 1980), as are the works of magical realism, where historical


chronologies are disrupted and lines between human and animal are regularly blurred
(Quayson 2006). As with religious novels, however, what’s striking is that such texts
seldom abandon a naturalistic vision outright, even while probing its (possible) limits.
In the preface to The Castle of Otranto (1764), often considered the first “gothic” novel,
Horace Walpole admits that, for all the “boundless realms of invention” in his work, he
also wished “to conduct the mortal agents in his drama according to the rules of prob-
ability,” and Nathaniel Hawthorne claimed that, in composing his “romance” The House
of Seven Gables (1851), he strove not only to “manage his atmosopherical medium as to
bring out or mellow the lights, and deepen and enrich the shadows,” but also to “make
a very moderate use” of such privileges, “to mingle the marvellous rather as a slight,
delicate, and evanescent flavor.” When the narrator of Rushdie’s Shame (1983) claims
to be writing a “modern fairy tale,” not a full-blown account of modern Pakistani poli-
tics and society—“nobody need get upset, or take anything I say too seriously”—his
qualifying adjective deserves emphasis as much as his generic noun.9
Ambivalence toward modern science, however, can pervade even the most prosaic,
least God-haunted of novels. Part of the empiricist mission, after all, is a reductivist
drive to discover the basic constituent pieces of the universe, and advances in technology
have over the centuries generated exactingly fine-grained accounts of the material
world. But it’s not clear how the most salient entities discussed by modern chemistry or
physics (cells, molecules, atoms, quarks, quantum fields) could ever be a novel’s central
focus. Of primary concern to novelists are not the imperceptible forces and entities
posited by scientists for purposes of explanation and prediction, but the variety of bodies
available to everyday human perception, some of which—those called “characters”—
undertake intentional actions justified by reasons. The genre is oriented less toward
what Wilfrid Sellars called the “scientific image,” which understands all perceptible
bodies as “a swirl of physical particles, forces, and fields” (1991: 20), than toward the
“manifest image,” which takes middle-sized dry goods to be primary. To be sure, many
novelists have evoked the kind of supra-personal forces that social scientists use to
explain human behavior (culture, society, class, etc.), and recent novelists are keenly
aware of the sub-personal explanations offered by neuro- and evolutionary science. But
even authors well-versed in scientific theories—Emile Zola, Richard Powers—seem to
preserve the possibility of meaningful action, however attenuated. As modern science
comes increasingly to view the universe in strictly non-intentional terms, the novel
dramatizes what Lukács called the “productivity of spirit”: the human capacity to
perceive and create intentional orders (1971: 33). This “productivity” is no doubt one
reason the genre had such a powerful appeal to Sartre, de Beauvoir, and other phenom-
enologists. But even beyond these specific cases, the novel as a genre has tended to be
less Humean than Kantian, striving to understand the relation between actions and
events rather than subsuming one under the other.10

The “productivity of spirit” left Lukács melancholic. As the “primaeval images have lost
their objective self-evidence for us,” we have “invented the creation of form”; hence
“everything that falls from our weary and despairing hands must always be incomplete”
(1971: 34). “Forms”—coherent systems, meaningful patterns—are not expressions of the
universe, but products of reason, desire, or other human faculties. And foremost among
these “forms” are the moral vocabularies that define and articulate our ends. Purposes in

Robert Chodat

the epic are given to the hero “with immediate obviousness,” but in the novel they are
diminished to mere “ideals,” “subjective facts,” and the “conflict between what is and
what should be” is never abolished (1971: 80).
Again, we need not accept Lukács’s tone or terms to grant his point. When our moral
concepts—what’s good and bad, just, worthy, contemptible, shameful—seem mere
inventions, then agreement about how and when to apply them is liable to wane, and
the novel is one of the chief records of this precarious condition. As commentators have
recognized, a major symptom of this shift is the genre’s marginalization of aristocratic
figures. Epics are usually unapologetic in identifying virtue with social privilege, but
even the early novelistic texts celebrated by Bakhtin stage unambiguous social hierar-
chies: Aethiopica and The Golden Ass each revolve around nobles and priests. The focus
of the modern novel is by contrast the unheroic, the commonplace, its characters
defined less by valorous deeds than by ordinary labor, particularly among the emerging
professional classes, and by the life of the family, which is typically a less tribal unit than
in earlier narratives. In brief, the novel reflects and extends the “affirmation of the
ordinary” that Charles Taylor has identified in Christianity (1989: 211–302), a connec-
tion Bakhtin underscores when he counts the Gospels among the genre’s prototypes
(Problems 135–37).
The leveling spirit of such changes can be invigorating: “honorable” comes to desig-
nate personal character rather than social status, and women come consistently to the
fore as worthy subjects of narrative—a phenomenon that was bolstered still further as
literacy rates among women gradually rose.11 But such shifts can also be disquieting, as
all that is solid melts into air and traditional affiliations vie with newly instrumentalized
social relationships.12 It’s only appropriate, in such a context, that the culture’s most
representative narratives should increasingly seem to be those of youth. Achilles and
Ulysses are mature men, and the speaker of the Inferno is “midway upon the journey of
our life,” but the novel’s paradigmatic characters are inexperienced and untutored:
Wilhelm Meister, Elizabeth Bennett, Julien Sorel, Dorothea Brooke, Stephen Daedalus,
Holden Caulfield (Moretti 1987). Philosophical wisdom may be associated with white
beards and measured contemplation, but the novel’s unseasoned protagonists tend to
avoid monastic or scholarly meditation, finding their way in a bustling, fragmented
world where exhilarating freedoms contend with the pressures of the marketplace and
modern bureaucracy. Loosened from family, village, and church, they are frequently on
the make, hunting for opportunity in the quickly expanding cities or the increasingly
colonized territories.
The sheer volatility of this condition, both expressed in and reinforced by the novel,
is perhaps one reason why so much modern philosophy has cherished singular moral
concepts: the “duty” of deontology, the “pleasure” of utilitarianism. Novels have
certainly been aware of such keywords, but they are chiefly interested in how these kinds
of principles fit into the course of particular lives, and whether they’re in fact adequate
responses to the atomistic, disintegrating tendencies of modern life. Richly detailing
specific situations and environments, the genre tends—particularly in the realist variant
that emerges in the nineteenth century—to challenge the habit among moral theories
to isolate specific actions and choices, placing these moments back into the stream of a
life, where particular decisions are taken by particular persons in particular circum-
stances, each with their own distinct tangle of traits. In doing so it tends to subordinate
the application of principles and calculations to the messier, more unsettled process of
balancing a diversity of goods, a dynamic that has led some commentators to compare


novelists to Aristotle.13 That such practical reasoning is often shared—think of the

incessant conversations in Austen, Eliot, James, or Wharton—is a further indication
of the novel’s distinctiveness. Whereas deontological and utilitarian theories
foreground the reasoning of individual agents, many novels have encouraged a more
robustly social conception of moral thought, underscoring the Sittlichkeit that often
structures agents’ aims and goods. Accordingly, particularly in nineteenth-century
texts, moral debates focus not infrequently on marriage, the trope of reconciliation that
novels often adopt from Renaissance comedy, and the domain that for Hegel and others
seems to exemplify non-manipulative relationships, a counterbalance to the dead cold
letter of modern institutions.
Nietzsche, of course, regarded such accommodations to modernity as herd-like (recall
his gibe at Eliot, cited earlier), and many late nineteenth- and twentieth-century novels
seem to agree, combating modern anomie not with fuzzy notions of Geist or fellow
feeling but with the strenuous perfectionism that Nietzsche himself extolled. Beginning
especially with Flaubert, novelists come increasingly to regard art itself as the fullest
human expression; thematizing and dramatizing the plight of the artist in a reductivist,
profanely commercial, complacently mediocre, middle-class world becomes a central
theme from Gide, Mann, and Joyce through Beckett and Don DeLillo. In this tradition,
artistic realization becomes the central aspiration in much the way that social recogni-
tion had been in earlier works, and marriage is supplanted as an ideal either by ascetic
isolation or an enclave of like-minded souls (“bohemia”). Modern moral life had always
inspired counter-reactions; for every dissenting Defoe eager to affirm the ordinary, there
was a tory Swift ready to ironize it.14 But by the start of the twentieth century, the artist
had become the aristocrat, and the quintessentially bourgeois genre had increasingly
taken on an anti-bourgeois cast.15
Again, however, what marks even the novel, even in these instances, is its essen-
tially ambivalent stance. The Nietzschean narrator of Gide’s The Immoralist (1904)
spurns his wealth and learning in favor of a more “natural” life of north Africa, but
his heated journey out of Europe hastens the death of his fragile, loyal wife. Ulysses
(1922) opens with the erudite aspiring poet Stephen, but quickly turns its attention
to the advertising salesman Bloom and eventually to the sprawling late-night rumi-
nations of Molly—each of whom are ironized less than the self-important young
poet.16 And in DeLillo’s Mao II (1992) it remains fundamentally unclear whether the
reclusive novelist Bill Gray ever finds any “pure” artistic space beyond the reach of
the global market and global politics. Such novels are ultimately too alert to the
prose of the world—Hegel, again—to lock themselves into the radical forms of perfec-
tionism that they entertain. And in doing so they make visible the irony that tends
to characterize the genre’s moral vision generally. The novel’s earliest readers may
have been bred on didactic texts—sermons, guidebooks—and early novels may
indeed strike us as wearily moralizing, but there’s a reason Samuel Johnson found
Fielding’s Tom Jones (1749) to be “mischievous or uncertain in its effects,” insuffi-
ciently clear in its exhibition of virtue (1969: 19–25).17 And even in the works of
Austen and George Eliot, with their witty and knowing narrators, characters achieve
happiness only insofar as they grasp the “productivity of spirit”: there may not be
much dirt beneath the fingernails of Darcy or Ladislaw, but Pride and Prejudice (1813)
and Middlemarch (1871–72) are studded with people who fail to recognize that their
social environments are improvised human products, always open to contentious
revision and only partially navigable by principles and traditions.18 Flaubert was

Robert Chodat

being unusually austere when he wrote to George Sand that he recognized no “right
to blame anyone” in fiction, and that the novelist should never “express his own
opinion of the things of this world” (McEnzie 1922: 99). But in many respects the
“impersonality” that he inaugurated simply intensifies the moral uncertainty into
which novels have always thrown their readers, and makes explicit their distance
from the single-minded philosopher.

Flaubert’s coolness landed him in hot water. His account in Madame Bovary (1857) of a
young woman’s overactive imagination led to one of the major court trials in literary
history, its author accused of the “glorification of adultery.” As critics have long noted,
however, the prosecutor could make his case only by being an incompetent reader: the
problematic passages he cited express the deliberations not of the author, but of the
character. That this could happen, that Emma could be taken for Flaubert, introduces a
final form of ambivalence that the novel genre has foregrounded, one appreciable at the
level of language.
Consider a moment just after Emma Bovary’s first tryst, which the prosecutor introduced
as evidence:

She repeated: “I have a lover! a lover!”, delighting at the idea as if a second

puberty had come to her. So at last she was going to possess those joys of love,
that fever of happiness of which she had despaired. She was entering upon
something marvelous where all would be passion, ecstasy, delirium.19

In the first sentence, the voices of narrator and character are clearly distinct, a separation
signaled orthographically by a colon and quotation marks. But by the time we arrive
at the final sentence, this line has disintegrated: the third-person voice remains, as it
does for most of the novel—hence no “she thought,” no quotation marks—but the
description of something “marvelous, where all would be passion, ecstasy, delirium” is
best ascribed not to Flaubert or his narrator, but to the young Emma, whose style of
thought momentarily controls the narration. The boundary between the author and
Emma becomes suddenly porous, the words half in the mouth of one and half in the
mouth of the other.
Such a technique—free indirect discourse, style indirect libre, erlebte Rede—is
frequently seen as the novel’s premier stylistic contribution to fiction, and sometimes
creates unusual test-cases for philosophers of language. When in Lawrence’s Women in
Love (1915) we read, “Tomorrow was Monday, Monday, the beginning of another
school week!” we may be struck by how the indexical and the verb tense are combined
in ways that standard speech would regard as nonsensical.20 More important, however,
are the variable effects that such sentences generate. In Madame Bovary the technique
sometimes allows a sympathetic intimacy with the character, as readers come temporarily
to inhabit Emma’s own language. Indeed, this potential for sympathy is what got
Flaubert into trouble: Emma often seems a meretricious young woman, yet Flaubert’s
language encourages readers to identify with her as much as he himself sometimes did
(“Madame Bovary, c’est moi,” he famously quipped), and it’s perhaps unsurprising that
the judge in the trial took the occasion to warn against works whose “realism” would
be “the negation of the beautiful and the good.”21 In other contexts, however, such


techniques are used to distance us from characters, undermining them without blanket
condemnation. Here the master is Austen, who, several decades before Flaubert,
introduces Sir William Lucas in Pride and Prejudice (1813) by saying he had “been
formerly in trade in Meryton, where he had made a tolerable fortune,” but had now
“removed with his family to a house about a mile from Meryton, denominated from
that period Lucas Lodge.” As James Wood observes, tolerable and denominated are not
Austen’s terms, but part of Lucas’s self-description, slyly inserted to expose his outsized
vanity—a strategy Austen perhaps adapted from the mock-heroic poems of her
Augustan predecessors (2008: 19–21).22
We should be careful, however, not to overstate how transparent such effects can be.
As Gregory Currie notes, there is no “simple relation between categories of speech and
narrative effects,” not least because such “character-focused narration” can have varying
degrees of intensity (2010: 134–35). Sir Lucas inflects Austen’s language far less than
Emma’s does Flaubert’s, but Flaubert maintains more boundaries than descendants such
as Woolf:

Oh, but she never wanted James to grow a day older or Cam either. These two
she would have liked to keep for ever just as they were, demons of wickedness,
angels of delight, never to see them grow up into long-legged monsters. Nothing
made up for the loss.

Or Bellow:

His achievements were not only scholarly but sexual. And were those achieve-
ments? It was his pride that must be satisfied. His flesh got what was left over.

Fierce critical disagreements often turn on how one should “take” such passages—how
much empathy or irony they encourage, which attitude the author intimates and wants
readers to adopt. Is Mrs. Ramsay a celebration of the female voice, or is she unhealthy
in her emotional demands? Is Herzog sanctioned or censured?23
This uncertainty can deepen when the author’s voice seems to disappear entirely—
is Humbert Nabokov, or Meursault Camus?—and intensifies still further when we
notice how the language of the novel puts in play more than just authors and characters.
Recall Flaubert’s sentence: “She was entering upon something marvelous where all
would be passion, ecstasy, delirium.” These are partly Emma’s words, partly Flaubert’s,
but they are not theirs alone; they’re culled from the popular fiction of the period, with
its exorbitant promises of erotic romance. What the sentence expresses, that is, is not
simply the mind of a particular character or author, but styles of thought pervading an
entire culture. It’s often said, rightly, that novels have a psychological intensity
untapped by other narrative forms, and certainly they have helped popularize Freudian
and other models of mind. But they should not for that reason be psychologized. They
regularly call attention to the social nature of thought and language, its narrators and
characters always embedded in particular milieus, dialects, economic classes, educa-
tional systems, professional jargons, generational cohorts, and so on. Wittgenstein
suggested that the different regions of our language are fundamentally heterogeneous,
ordered and hierarchized only for specific purposes, and a strikingly similar vision lies
at the heart of Bakhtin’s work. “The word in language,” says Bakhtin, “is half someone
else’s,” and the novel is the form that most richly organizes this “heteroglossia,” that

Robert Chodat

most radically pulls together the wildly diverse modes that our common words can take.
It is the genre of genres, exemplifying, quoting, parodying, and mimicking all ranges of
speech and thus all types of socio-cultural life. Moby Dick (1851) juxtaposes not only
Ahab and Ishmael, but also biblical, scientific, and a host of other discourses; the
“Nausicaa” section of Ulysses puts on display not just Gerty McDowell, but the culture
of early twentieth-century women’s magazines; grasping Pale Fire (1962) requires under-
standing not merely Shade and Kinbote, but the entire language and institution of
literary criticism. How such phenomena fit into our philosophical accounts of language
is unclear. Philosophers have long debated the relative priority of semantics and
pragmatics—“word-meaning” and “speaker-meaning,” etc.—but these arguments tend
to break down rather tidily between large-scale, atemporal, stable systems of meaning
on the one hand, and eccentric speakers or moments of utterance on the other.24 The
novel’s ability to coordinate various idiolects and sociolects, both within and between
sentences, may force theorists to consider other, mid-level categories not fully captured
by such crude dichotomies as the “community” and the “speaker,” an impersonally
ordered Language and an occasionally unruly I.25

Broad and breathless as my survey here has been, it should be enough to suggest why
we should avoid conflating novel with fiction and narrative, and what some of the
novel’s specific philosophical contours might be. To talk about a novel is to talk not
about persons, environments, actions, and events per se, but about a particular type of
description of persons, environments, actions, and events. The criteria for this type
may be relatively loose, but identifying the characteristic face of the genre can help
us recognize anew how an underlying perspective or ethos can be manifest in literary
texts in ways that extend well beyond the manifest claims of its narrators and characters.
Making this perspective or ethos explicit is one way to avoid misreading particular
works, but more broadly it suggests why both artists and audiences might be drawn to
particular forms rather than others, and why whole genres might fall flat to a particular
person and even an entire culture. More broadly still, it helps illuminate why genres
flourish and ebb, and why they cannot be fully understood apart from the social
environment in which they evolve. How one regards the novel—whether one reads
novels, whether one believes they’re a living and significant form of expression—will
reflect not just a certain “taste,” but something important about how one understands
metaphysics, science, psychology, morals, politics, language, modernity, and an array
other fundamental issues. Nietzsche, as we’ve seen, recognized the philosophical
stakes that genre entails, and disdained the novel accordingly; Bakhtin recognized
them as well, and chose to celebrate. We ourselves may find neither of these extremes
satisfying, and treat the novel as a phenomenon less to love or loathe than to under-
stand, not a symptom of degeneration or emancipation, but a fantastically complex
description of what we have become, the literary genre that perhaps most vividly
displays to us our own unsettledness. Such a response would be appropriate to a genre
that has absorbed so many other genres, that convenes so many voices and arranges
so many angles of vision, and which so frequently frustrates the quest for certainty,
singular forms, and clear and distinct ideas.


  1 Recent collections such as John and Lopes 2004 or Schroeder 2010 include no essay on the novel genre,
though several of their authors discuss fiction and narrative. Lamarque 2009 raises the issue of genre and
the concept of “literature” in his first chapter, and the book has a later chapter entitled “Fiction,” but he,
too, includes no extended discussion of the novel. This relative indifference to genre is not, however,
strictly a philosopher’s habit. A focus on “narrative” has been a hallmark of literary theory ever since
Roland Barthes and Tzvetan Todorov, building on Russian Formalism, developed “narratology” in the
1960s, and this habit continues apace today, in the range of journals, seminars, and institutes devoted to
narrative and “narrativity.”
  2 This indifference to specific narrative genres has perhaps begun to fade, as we see with Goldman 2013,
Woodruff 2008, and the 2013 University of Warwick conference on philosophy and the short story. Two
earlier exceptions to the philosophical indifference to genre are Martha Nussbaum and Richard Rorty,
each of whom has important things to say about the particularities of the novel genre. But neither
Nussbaum nor Rorty are (nor wishes to be) in the mainstream of Anglo-American philosophy, and even
in their work, the recurring distinction tends to be between the novel and philosophy, rather than the
novel and other literary forms. In speaking here of the place of James, Proust, and Coetzee in recent
debate, I have in mind the work of Nussbaum, Rorty, Cora Diamond, Richard Moran, Stephen Mullhall,
Alice Crary, Robert Pippin, and others.
  3 On the way genres exhibit a “logic” and inherit conditions, procedures, subjects, and goals of composition,
see Cavell 1981, 1–42. I’m thankful to Kristin Boyce for asking me to think more carefully about the idea
of a genre, and Cavell’s discussion in particular.
  4 For reasons that I’ve indicated already, I’ll be ignoring the issue of fiction, but the relation of fiction to
the novel is indeed an important topic. For some reflections on fictionality and the novel genre, see
Gallagher 2006.
  5 For a sense of Lukács’s place in current discussions of the novel, see Cascardi 2009.
  6 On Greek novels, see Doody 1996. For the more ambitious claims about the novel-hood of Gilgamesh
and Egyptian fiction, see Moore 2010. A subtler version of this broad conception of the novel is offered
in Pavel 2006.
  7 Obviously I don’t mean to ignore the reams of prefaces, letters, and essays that reflect the novel form
throughout the seventeenth and eighteenth centuries. I simply mean that the major philosophical voices
prior to Hegel—not just Aristotle and Kant, but also figures as late as Shaftesbury, Lessing, Herder, Vico,
and Schiller—typically took their examples from painting, sculpture, drama, and poetry when they were
discussing aesthetics. Most of the earliest reflections on the novel were written by literary authors and
critics, and most were polemical efforts either to justify the form (e.g. Fielding, Clara Reeves, Fanny
Burney, Ludwig Tieck) or to interrogate it (e.g. Samuel Johnson). Hume cites Don Quixote in “Of the
Standard of Taste,” but less to explicate the novel than to ponder a particular remark made by Sancho
Panza about the taste of experienced judges.
  8 This change is central to chapter 1 of Watt 1957.
  9 My references here to the Gothic allow me to mention, in passing, the contributions that Queer Theory
has recently made to debates about the modern novel, particularly texts absorbed with tropes of the
unnamable, the unspeakable, the not-wholly-legible, the uncanny, the liminal, the marginal, and so
forth. See, for instance, Sedgwick 1997.
10 On the relationship of the modern novel to sub- and supra-personal levels of explanation, see Chodat
2008. Cascardi 2009 also uses this line from Lukács to develop a distinction between the novel’s type of
empiricism and the philosophical empiricism of Locke and Hume; see 173–75.
11 On “honor” and the “crisis of status inconsistency” that helped shape the English novel, see McKeon
1987: 194. On literacy rates between 1600 and 1800 in England and other countries, see Hunter 1990:
12 As my nod to the Communist Manifesto here suggests, it’s not an accident that Marx took such an inter-
est in the modern novel, especially Balzac. All of which raises the question of how much the genre
reflects specifically Western socio-historico-economical developments. For an introduction to this large
and oft-discussed question, see part 14 of McKeon 2000.
13 Most notably Nussbaum 1990.
14 On Defoe, Swift, and the larger tension they helped exemplify, see McKeon 1987: 150–71.
15 For an interesting twist on this theme, see Robbins 2006.

Robert Chodat

16 It’s telling that, as he read early drafts of Ulysses, Ezra Pound—not a friend of the bourgeois—thought
Joyce should abandon Bloom and push Stephen back into the foreground. It’s also telling that Joyce
disagreed, saying that he could do nothing more with Stephen’s character. See Ellmann 1982: 459.
17 On the didactic tradition that preceded and informed the novel, see Hunter 1990: 225–47.
18 Cascardi 2009 makes this connection between Lukács and Austen; see 175. The comment about dirt and
fingernails is from Eagleton 2005, who uses it (not unjustifiably) to question Bakhtin’s insistence on the
novel’s endlessly disruptive and centrifugal energy.
19 This passage is addressed in Jauss 1982: 41–45.
20 This example is drawn from Schlenker 2004: 280.
21 Cited in Jauss 1982: 45.
22 On the “empathic” and “ironic” poles of free indirect discourse, see Cohn 1978.
23 On Mrs. Ramsay as the embodiment of the female voice, see Lidoff 1986. For less flattering views of her,
see Pederson 1958 and Millett 2000.
24 This was Bakhtin’s complaint already in the 1920s. Existing theories of language, he said, restrict
themselves to “the system of a unitary language” (e.g. Saussure’s langue) and existing poetics to “the
individual speaking in this language” (e.g. parole) (1981: 270). Both ignore the historically variable
genres of speech that stand between these poles conceptually, as well as the genre that assembles them
most thoroughly.
25 Do these linguistic phenomena—free indirect discourse, Bakhtinian “polyphony”—depend upon the
existence of print and texts? Possibly. Don Quixote appears about a century and a half after Gutenberg,
whereas epics—and fairy tales, fables, and legends—are passed down orally between generations and
often recited for public performances (recall Plato’s friend Ion). Any such argument, however, would
have to address early modern drama; Shakespeare’s plays, after all, are among other things supremely
composed documents. For Benjamin’s reflections on what he calls the novel’s “essential dependence on
the book,” see Benjamin 2002: 146. A quite different way of describing this dependence comes from
Ann Banfield, who, drawing on Chomsky, claims that free indirect discourse does not “exist in the spoken
language,” and has, “unlike language itself, . . . a determinable historical origin (McKeon 2000: 519).

Bakhtin, M. M. 1981. The Dialogical Imagination: Four Essays. Trans. Caryl Emerson and Michael Holquist.
Austin: University of Texas Press.
Bakhtin, M. M. 1984. Problems in Dostoevsky’s Poetics. Trans. Caryl Emerson. Minneapolis: University of
Minnesota Press.
Benjamin, Walter. 2002. “The Storyteller.” Selected Writings, Volume 3: 1935–1938. Ed. Howard Eiland and
Michael W. Jennings. Cambridge: Harvard University Press.
Cascardi, Anthony. 2009. “The Novel.” The Oxford Handbook of Philosophy and Literature. Ed. Richard
Eldridge. New York: Oxford University Press.
Cavell, Stanley. 1981. Pursuits of Happiness: The Hollywood Comedy of Remarriage. Cambridge: Harvard
University Press.
Chase, Richard. 1980. The American Novel and its Tradition. Baltimore: Johns Hopkins University Press.
Chodat, Robert. 2008. Worldly Acts and Sentient Things: The Persistence of Agency from Stein to DeLillo. Ithaca:
Cornell University Press.
Cohn, Dorrit. 1978. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton:
Princeton University Press.
Currie, Gregory. 2010. Narratives and Narrators. New York: Oxford University Press.
Doody, Margaret Anne. 1996. The True Story of the Novel. New Brunswick: Rutgers University Press.
Eagleton, Terry. 2005. The English Novel: An Introduction. Blackwell, 2005.
Ellmann, Richard. 1982. James Joyce. New York: Oxford University Press.
Frye, Northrop. 1957. Anatomy of Criticism. Princeton: Princeton University Press.
Gallagher, Catherine. 2006. “The Rise of Fictionality.” The Novel: Volume 1: History, Geography, and Culture.
Ed. Franco Moretti. Princeton: Princeton University Press, 2006. 336–63.
Goldman, Alan H. 2013. Philosophy and the Novel. New York: Oxford University Press.
Hegel, G. W. F. 1975. Aesthetics. Vol. 2. Trans. T. M. Knox. New York: Oxford University Press.
Hunter, J. Paul. 1990. Before Novels: The Cultural Contexts of Eighteenth-Century English Fiction. New York:
Norton, 1990.


Jauss, Hans Robert. 1982. Toward an Aesthetic of Reception. Ed. Timothy Bahti. Minneapolis: University of
Minnesota Press.
John, Eileen, and Dominic McIver Lopes, ed. 2004. The Philosophy of Literature: Contemporary and Classic
Readings. Oxford: Wiley-Blackwell.
Johnson, Samuel. 1969. The Yale Edition of the Works of Samuel Johnson. Volume III. Ed. W. J. Bate and
Albrecht B. Strauss. New Haven: Yale University Press.
Lamarque, Peter. 2009. The Philosophy of Literature. Oxford: Blackwell, 2009.
Lidoff, Joan. 1986. “Virginia Woolf’s Feminine Sentence: The Mother–Daughter World of To the Lighthouse.”
Literature and Psychology 32: 43–59.
Lukács, Georg. 1971. The Theory of the Novel. Trans. Anna Bostock. Cambridge: MIT Press.
McEnzie, Aimee L., trans. 1922. The George Sand–Gustave Flaubert Letters. London: Duckworth.
McKeon, Michael. 1987. The Origins of the English Novel, 1600–1740. Baltimore: Johns Hopkins University
McKeon, Michael, ed. 2000. Theory of the Novel: A Historical Approach. Baltimore: Johns Hopkins University
Millett, Kate. 2000. Sexual Politics. Urbana: University of Illinois Press.
Moore, Steven. 2010. The Novel: An Alternative History, Beginnings to 1600. New York: Continuum.
Moretti, Franco. 1987. The Way of the World: The Bildungsroman in European Culture. London: Verso.
Nietzsche, Friedrich. 1999. The Birth of Tragedy and Other Writings. Trans. Ronald Speirs. Ed. Raymond Guess
and Ronald Speirs. Cambridge: Cambridge University Press.
Nietzsche, Friedrich. 2005. The Anti-Christ, Ecce Homo, Twilight of the Idols and Other Writings. Ed. Aaron
Ridley and Judith Norman. Cambridge: Cambridge University Press, 2005.
Nussbaum, Martha C. Love’s Knowledge: Essays on Philosophy and Literature. New York: Oxford University
Press, 1990.
Pavel, Thomas. 2006. “The Novel in Search of Itself: A Historical Morphology.” The Novel: Volume 2: Forms
and Themes. Ed. Franco Moretti. Princeton: Princeton University Press. 3–31.
Pederson, Glen. 1958. “Vision in To the Lighthouse,” PMLA 73: 585–600.
Quayson, Ato. 2006. “Fecundities of the Unexpected: Magical Realism, Narrative, and History.” The Novel:
Volume One: History, Geography, and Culture. Ed. Franco Moretti. Princeton: Princeton University Press.
Robbins, Bruce. “The Portrait of the Artist as a Young Social Climber.” The Novel: Volume 2: Forms and
Themes. Ed. Franco Moretti. Princeton: Princeton University Press. 409–35.
Schlegel, Friedrich. 1968. Dialogue on Poetry and Literary Aphorisms. Trans. Ernst Behler and Roman Struc.
University Park: Pennsylvania State University Press.
Schlenker, Philippe. 2004. “Context of Thought and Context of Utterance: A Note on Free Indirect
Discourse and the Historical Present,” Mind and Language 19: 279–304.
Schroeder, Severin, ed. 2010. Philosophy of Literature. Oxford: Wiley-Blackwell.
Sedgwick, Eve Kosofsky, ed. 1997. Novel Gazing: Queer Readings in Fiction. Durham: Duke University Press.
Sellars, Wilfrid. 1991. Science, Perception, and Reality. Atascadero, CA: Ridgeview.
Taylor, Charles. 1989. Sources of the Self: The Making of the Modern Identity. Cambridge: Harvard University
Tolson, Jay. 1992. Pilgrim in the Ruins: A Life of Walker Percy. New York: Simon and Schuster.
Trollope, Anthony. 1938. Anthony Trollope: Four Lectures. Ed. M. L. Parrish. London: Constable.
Watt, Ian. 1957. The Rise of the Novel: Studies in Defoe, Richardson, and Fielding. London: Chatto & Windus.
Wood, James. 2008. How Fiction Works. New York: Picador.
Woodruff, Paul. 2008. The Necessity of Theater. New York: Oxford University Press.

Further Reading
Girard, René. 1965. Deceit, Desire, and the Novel: Self and Other in Literary Structure. Trans. Yvonne Freccero.
Baltimore: Johns Hopkins University Press.
Hoffman, Michael J. and Patrick D. Murphy, ed. 2005. Essentials of the Theory of Fiction. Durham: Duke
University Press.
Jameson, Fredric. 1981. The Political Unconscious: Narrative as a Socially Symbolic Act. Ithaca: Cornell University
Kundera, Milan. 1988. The Art of the Novel. Trans. Linda Asher. New York: Harper & Row.

Robert Chodat

Landy, Joshua. 2004. Philosophy as Fiction: Self, Deception, and Knowledge in Proust. New York: Oxford
University Press.
McHale, Brian. 1988. Postmodernist Fiction. London: Routledge.
Murdoch, Iris. 1997. Existentialists and Mystics: Writings on Philosophy and Literature. Ed. Peter Conradi.
New York: Penguin.
Pippin, Robert B. 2001. Henry James and Modern Moral Life. Cambridge: Cambridge University Press.
Rorty, Richard. 1989. Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press.

Anna Christina Soy Ribeiro

What can a philosopher have to say about poetry? One may rightly be skeptical of the
chances of success of such an endeavor. After all, philosophers and poets have been
butting heads for over 2,500 years. Ever since Plato challenged the poets for the meaning
of their works in the Ion, and excluded nearly all poetry from his ideal city in the
Republic, philosophers have been trying to show that they are better, at least where the
acquisition and conveyance of knowledge are concerned. Truth be told, poets have
seldom taken much notice of philosophers’ attacks.1 By the time the quarrel started,
poets had long been the sage guardians of their communities’ knowledge; as Socrates and
Plato well knew, it would not be easy to dethrone Homer. That said, philosophy’s attack
on poetry was timely. Philosophers would not have had a chance in the time of Homer
himself (whoever he was), or even for centuries afterwards; but by Socrates’ time, the
power of poetry as arbiter and conveyor of values was waning, and sophists as well as
philosophers had a fighting chance to take its place—though they attacked the lesser
rival, the rhapsode, rather than the poet himself.2
Not every philosopher since Plato has persisted on his exclusionary task, and even
his own most famous pupil, Aristotle, tried to give poets their philosophical due. It is
fair to say that various versions of the Platonic and the Aristotelian stances have held
sway over the centuries and continue to do so to this day. Indeed, the widely discussed
philosophical and methodological differences between what is today called ‘analytic/
Anglo-American’ and ‘Continental’ philosophy, may be seen through the lens of
philosophical approaches to poetry. On the Anglo-American, anti-poetry side, we may
look to one of its greatest forebears, John Locke, who, in his Essay Concerning Human
Understanding (1690) made it very clear that poetic language and truth-seeking were
inherently at odds with one another:

Figurative speech . . . an abuse of language . . . I confess, in discourses where we

seek rather pleasure and delight, than information and improvement, such
ornaments . . . can scarce pass for faults. But yet, if we would speak of things as
they are, we must allow, that all the art of rhetoric, besides order and clearness,
all the artificial and figurative application of words eloquence hath invented,
are for nothing else but to insinuate wrong ideas, move the passions, and
thereby mislead the judgment; and so indeed are perfect cheats: and therefore,
however laudable or allowable oratory may render them in harangues and
popular addresses, they are certainly, in all discourses that pretend to inform or
instruct, wholly to be avoided; and where truth and knowledge are concerned,
cannot but be thought a great fault, either of the language or person that makes
use of them.3
Anna Christina Soy Ribeiro

Much as Plato thought, on Locke’s view poetic uses of language are suitable only for
pleasure and deceit, for they distort the truth. Poetry is not the place to go to if we are
seeking ‘information and improvement,’ if we wish to know of ‘things as they are.’ Much
as Plato did, in modern times A. J. Ayer discarded not only poetry but all of the arts as
a proper philosophical subject, contending that “aesthetic words such as ‘beautiful’ and
‘hideous’ are employed, as ethical words are employed, not to make statements of fact,
but simply to express certain feelings and evoke a certain response. It follows, as in
ethics, that there is no sense in attributing objective validity to aesthetic judgments, and
no possibility of arguing about questions of value in aesthetics, but only about questions
of fact.”4 The descendants of Plato, Locke, and Ayer are alive and well; one need go no
farther than this very volume to find them:

Poetry brings insights. . . . The poem’s insights are not, we would say, a matter
of meaning, as we normally think of it in the philosophy of language. Meaning
is transparent. . . . The model of meaning we have in mind is the model of
inquiry. . . . Meaning, on this view, is a tool we use to reach agreement on matters
of interest to us. Through meaning, we associate linguistic units with objects
and properties in the world. . . . And we seek the truth. . . . Contrast this with
the poetic imagination. . . . It is expansive, open-ended, and dynamic . . . there
are insights prompted by poetry, in our view, but no poetic meanings.5

What a different picture is drawn by Aristotle, and by the forebears of the Continental
tradition. For Aristotle, poetry could be a source of insight into human nature, since the
poet needs to speak of what courses of thought and action are possible for a human
being, and that requires knowledge of essences and possibilities: the poet is to show what
“a certain kind of person may well say or do in accordance with probability or neces-
sity.”6 Moving closer to Locke’s period, Kant’s Critique of Judgment typifies how highly
German philosophers of the time regarded the art of poetry:

Among all the arts poetry holds the highest rank. (It owes its origin almost
entirely to genius and is least open to guidance by precept or examples.) It
expands the mind: for it sets the imagination free, and . . . rises aesthetically to
ideas. Poetry fortifies the mind: for it lets the mind feel its ability—free,
spontaneous, and independent of natural determination—to contemplate and
judge phenomenal nature as having aspects that nature does not on its own
offer in experience either to sense or to the understanding.7

For Hegel, poetry is “the universal art of the spirit which has become free in itself and
which is not tied down in its realization to external sensual material; instead, it launches
out exclusively in the inner space and the inner time of ideas and feelings”; unlike other
art forms, which are restricted to their media, “poetry is adequate to all forms of the
beautiful and extends over all of them, because its proper element is beautiful imagina-
tion, and imagination is indispensable for every beautiful production, no matter to what
form of art it belongs.” For these reasons, poetry crowns Hegel’s hierarchy of the arts.8
For Schopenhauer, poetry is “the art of bringing into play the power of imagination
through words,” an art form whose works “exercise a much stronger, deeper, and more
universal effect than pictures and statues do.”9 Schopenhauer, too, ordered the arts in a
hierarchy, and for him tragic poetry was the highest of the representational arts, second


only to music, a non-representational art form he considered an immediate copy of the

Will itself (the Will, in his philosophical system, being the underlying reality of all there is).
In Continental philosophy, thinkers such as Heidegger, Adorno, and Lyotard all granted
poetry a special status as a discloser of truths not available to the logico-scientific
method. Yet, despite these defenses, the kind of truth philosophers such as Heidegger
ascribed to poetry would seem to reinforce the Plato-Locke-Ayer approach, insofar as
this truth is described as a kind of profound, quasi-mystical insight that is not amenable
to the logico-scientific method. So poetic language has gradually and inexorably been
denied the status of a language of knowledge.
It was not always like this. Not only in Plato’s time but also long after, the language
of knowledge has, in fact, often been poetic. Treatises of all sorts were written in the
kind of language we now consider the prerogative of poetry, conceived as an artistic
rather than a scientific practice (where these in turn are distinguished in terms of knowledge-
building). The Greek philosophers Parmenides and Empedocles in the fifth century BC
wrote their philosophical treatises in verse (the same dactylic hexameter used by
Homer), as did the Roman Lucretius four centuries later; the Indian Aryabhata wrote
his mathematico-astronomical treatises in verse in the fifth century AD, and the Coptic
physician Al Mufaddal Ibn Majid Ibn Al-Bishr his medical treatise in the thirteenth
century in the same manner. These works were often just as filled with metaphors,
similes and symbolism as the works we now call poetry (whether we should consider
them poems in the now prevalent ‘artwork’ sense of the term is another question). There
were various reasons why these treatises were written in poetic language. Certainly
among them was the practical one of enabling listeners to commit another’s ideas to
memory, since language to which a regular rhythm is added (as in metrical language) is
more easily memorized. There were also reasons having to do with status, a status in turn
having to do with knowledge. There was something quasi-religious and highly special,
and specialized, in knowledge that was conveyed by means of verse. Learning the
mysteries of the universe, of mind, body, numbers, and the stars took on the status of a
ritualistic initiation, something available for the few, not the average folk.10
Sometimes, however, knowledge was put into verse precisely to make it memorable
for the average folk. The Code of Health of the School of Salernum, used as a medical guide
for some seven centuries, is entirely in rhymed hexameter verse and, though sometimes
it is more literal than one might wish, it abounds in figurative speech. Here are its opening
lines, followed by a nineteenth-century verse translation and a twentieth-century
translation that strives to avoid figurative language:

Anglorum Regi scribit schola tota Salerni.

Si vis incolumem, si vis te redderem sanum,
Curas tolle graves, irasci crede profanum,
Parce merum, coenato parum, non sit tibi vanum
Surgere post epulas, somnum fuge meridianum,
Non mictum retine, nec comprime fortiter anum.
Haec bene si serves, tu longo tempore vives.
Si tibi deficiant medici, medici tibi fiant
Haec tria, mens laeta, requies, moderata dieta.
Salerno’s School, in conclave high, unites
To counsel England’s King, and thus indites:
If thou to health and vigour wouldst attain,

Anna Christina Soy Ribeiro

Shun weighty cares—all anger deem profane,

From heavy suppers and much wine abstain.
Nor trivial count it, after pompous fare,
To rise from table and to take the air.
Shun idle, noonday slumber, nor delay
The urgent calls of Nature to obey.
These rules if thou wilt follow to the end,
Thy life to greater length thou mayst extend.
Shouldst Doctors need? Be these in Doctors’ stead—
Rest, cheerfulness, and table thinly-spread.11
The whole school of Salerno wrote for the English king:
If you want to be healthy, if you want to remain sound,
Take away your heavy cares, and refrain from anger,
Be sparing of undiluted wine, eat little, get up
After eating fine food, avoid afternoon naps,
Do not retain your urine nor tightly compress your anus.
Do these things well, and you shall live a long time.
Should you need physicians, these three doctors will suffice:
A joyful mind, rest and a moderate diet.12

As its nineteenth-century American translator said,

though its merit is not enhanced by the Leonine verses in which the subject was
set, it would be unjust to suppose that even this masculine, unvarnished measure,
without any quality to recommend it saves its sonorous cadences, had no part
in introducing it to popular favor.13

The poetic form in which the Code is set may not make its precepts more apposite, but
neither do they make it less so; that is, neither do they make them wrong or false or
merely ‘insight’ in some derogatory sense. In other words, the poetic form neither adds
to nor subtracts from the truth of the matter in this case. Neither do the two poetic
figures seem a challenge: to deem anger profane is to deem it a bad thing to be avoided;
to flee from afternoon sleep is likewise to shun it.
Other kinds of poetry also seem to challenge any truth/insight distinction. The
amusing Ecloga de Calvis (In Praise of Bald Men), in the praise-poem tradition, seems
clearly to “speak of things as they are,” accomplishing this moreover while using only
words that begin with the letter ‘c’ (as in ‘calvis’, ‘bald’) throughout its 146 lines. Finally,
in oral cultures the poet is often the conveyor of news, and surely the news are facts,
whether or not they are conveyed in poetic fashion. (A contemporary version of the
traditional news-poet is now found on the internet: Senegalese rappers Keyti and Xuman
have a popular channel on YouTube, Journal Rappé, where they rap the news of the day
in French and in their native Wolof.)14
On the one hand, the long history of philosophical and scientific inquiry conducted
in poetic language might make us question any too-sharp distinctions between scientific
‘truth’ and poetic ‘insight.’ On the other hand, given that the cryptic nature of much of
this intellectual work typically made necessary the aid of a commentator, who could be


the same person as the poet but who spoke in prose, one could argue that the distinction
has always been there, although the scales of meaning and value were clearly tipped in
favor of poetic language as the ‘true’ conveyor of knowledge. In other words, the prose
commentaries were merely a pedagogical tool on the path to knowing the meaning of
the poetic treatise, to be dispensed with upon arrival.15
History, schmistory, one might say. The empirical and theoretical sciences all gained
by being emancipated from the confines of both mnemonic and religious needs (which
emancipation, incidentally, happened concurrently). The need to memorize because the
recording means were few in kind and scarce in number meant not only that few people
could devote their time to learning, but also that learning took a very long time (surely
the ritualization of learning was a clever way to protect the practice from the criticism
of those with more pragmatic inclinations). Books made the need to memorize obsolete,
and meant that ideas could now be expressed in less intricate prose. Freedom from ritu-
alistic memorization meant that knowledge spread far and wide and down from the
castes that kept it to themselves. All hail Guttenberg. Moreover, this historical change
seems to constitute clear evidence of a bona fide distinction between poetic and non-
poetic uses of language, and of the specific ways in which they convey and promote
knowledge—one regimented, relying on meanings that are “transparently compositional
and systematic,” the other “expansive, open-ended and dynamic,” relying on meanings
whose significance must be reassessed on each encounter (Lepore and Stone, Chapter
28 in this volume). The distinction may be explained in various ways, but it is evident
wherever one looks. Here is an example taken from the Poetry Foundation page on the
philosopher and poet John Koethe, that was probably meant as a compliment: “As critic
Robert Hahn notes, ‘Koethe’s poetry is ultimately lyrical, and its claim on us comes not
from philosophy’s dream of precision but from the common human dream that our lives
make some kind of sense’.”16 In other words, whatever Professor Koethe is doing in his
poems, it is not philosophy, because it does not aim at precision. It is, perhaps, “expansive,
open-ended and dynamic.”
Lepore and Stone argue that “poetry is like quotation in that both poems and
quotations privilege and problematize linguistic form in relationship to meaning”
(Chapter 28, this volume, p. 323). Furthermore, it is because poems are about their own
articulation that they are able to offer ‘insights,’ but not ‘truth.’ Truth requires “meanings
[that] are transparently compositional and systematic” (Chapter 28, this volume,
p. 330); the idea is that an utterance that draws attention to its own articulation is made
opaque, and that opacity threatens the systematicity and compositionality of meaning:

when we understand an utterance as ‘poetic’ in a special way, we may draw

distinctive insights from our interpretive efforts, but these special insights
should not be understood as contents of any level of linguistic meaning—not
even a ‘pragmatic’ level of meaning concerned with propositions that go beyond
conventional meaning but the speaker obviously intends to communicate
anyway. (Lepore and Stone, Chapter 28, this volume, pp. 323–4)

It seems true that we typically approach utterances we consider poetic differently from
utterances we do not consider poetic. However, since we do have many scientific, phil-
osophical, medical, and practical treatises written in formal and figurative language, it
also seems that an utterance’s being about its own articulation is no obstacle to

Anna Christina Soy Ribeiro

knowledge-building. Furthermore, I am not sure that one can take such a ‘reader-
response’ approach to utterances across the board as Lepore and Stone think. They
characterize the poetic imagination as “a search for significance in the formal organiza-
tion of the poem itself” and claim that “the same utterance [can be] understood in two
different ways, as a function of two different interpretive practices we have at our dis-
posal” (ibid., p. 326) and that “The poetry . . . lies not in the words themselves so much
as the way we approach them as readers” (ibid., p. 329). This may be true of some poems,
especially those in the ‘found poetry’ category, but I wonder whether the following lines
can be approached or understood in anything other than a poetic manner:

The art of losing isn’t hard to master;

so many things seem filled with the intent
to be lost that their loss is no disaster.
Lose something every day. Accept the fluster
of lost door keys, the hour badly spent.
The art of losing isn’t hard to master.
Then practice losing farther, losing faster:
places, and names, and where it was you meant
to travel. None of these will bring disaster.17

Even with found poetry, what is ‘found’ is selected precisely because one recognizes
something poetic in it. So, if some set of words ‘works’ as a poem even when the words
were not originally presented as such, is it not because there is something in that set
that is inherently poetic, and all it takes is a poetic ear to notice it? We cannot make
Lord Byron’s Don Juan not poetry (or its lines not “about their own articulation”) by
approaching them with a non-poetic imagination. Some utterances are so dense with
the prosody of the language in which they occur, or filled with such evocative images,
that we cannot help being struck by them as poetic. Once we are familiar with the
concept of poetry and with poetic practices, then we come prepared to utterances or
texts that label themselves as such. But making the underlying prosodic structure of a
natural language salient is something that even children untutored in poetry naturally
do and delight in. Surely they “find new significance in the very workings of language
itself” (Lepore and Stone, Chapter 28, this volume, p. 329), but before that they just
find pleasure in it.
It is certainly true that part of the reason poetry cannot be paraphrased lies in its
formal and figurative aspects. But poetry is not special in being occasionally unpara-
phrasable. There is a sense in which the simple expressions Go!, Leave!, and Get out
of here! are all synonymous, but they can mean rather different imperatives as well.
Or, to make it even simpler, think of the word yes and all the different meanings that
can accompany it merely by the intonation in which we utter it: excited, resigned,
hesitant, assertive, aggressive. We could easily have had different words to express
what is behind all these intonations. But the fact that we speak them makes different
‘yeses’ unnecessary. That is why writing must regiment meaning—that is why it is so
easy to be misunderstood in writing. It is important to bear in mind that language
is, first and foremost, speech. We have regimented language in the course of becom-
ing a writing and print culture—standardizing spellings, vocabulary meanings, and


grammatical structures. That is a recent, and a contingent fact in our history. This is
not to say that there was no distinction to be drawn between meaning and use before
the invention of writing and of the printing press. It is to draw attention to language
as speech, and to speech as where meaning primarily lies. It is to put the horse before
the cart, and to realize that if that cart gets somewhere it is only because the horse
keeps on galloping.
There is another important aspect to the process of becoming writers and readers and
not only speakers and hearers of language. Making someone’s words available to more
than those who might be there at the time when they were spoken is making them
accessible to infinitely many rather than an absolute few. The process of democratization
of anything, of making anything accessible to more people, is invariably a process of
simplification, of pruning something down to its lowest common denominator. Seeing
that final product that is now easily available to the majority as the ideal state of the
product is tantamount to seeing the factory-produced polyester suit as superior to the
handmade wool bespoke, or the white bromated-flour sliced bread as the ideal version
of the hand-kneaded loaf made with stone-ground flour and a sourdough starter. The
Health Code of Salerno may have taken more work to compose, but its verse is what made
it last from the twelfth century right up until the nineteenth, that ‘Age of Wonders’
century that gave us so much and left us with so little. The language of learning became
simpler because resources did not allow for everyone who wished to learn to spend years
in the shadow of a master memorizing cryptic verses; it became simpler not because
simpler was better, but because simpler was easier, more accessible to those without
training. The language of knowledge can and has been poetic. The language of inquiry
may have lost its versification and become prosaic, but, that, too, is an accident of
history, not a logical necessity.
The notion of semantic innocence espoused by Locke’s heirs, and its concomitant
idea that truth requires “meanings [that] are transparently compositional and sys-
tematic,” seems, on the face of it, simply false. More likely, we agree on meanings
as circumstances require, and our lexicon, as has recently been proposed, is
‘dynamic.’18 If this is true, then the fact that we expand the meanings of words in
figurative speech becomes considerably less mysterious, and a new understanding of
figurative language emerges. For instance, we would now have to explain metaphor
not as some sort of deviation from literal language, whether as a ‘filter’ or any other
device that is itself understood metaphorically, but as perhaps the occasional
extreme on a continuum of semantic possibilities, not all of which can be conjec-
tured in advance.
Poetry makes salient the various potentialities of language: the phonetic, the syntactic,
and the semantic. It typically makes salient the phonetic potentialities of language by
means of meter, rhyme, and other rhythmic devices. It often makes salient the syntactic
potentialities of language by producing sentences that do not fit the everyday syntax of
speech. It makes salient the semantic potentialities of language by using words in novel
senses, via various figures of speech. By these means, poetry with artistic rather than
merely didactic aspirations at its best also succeeds in making salient the potentialities
of thought and feeling. It shows us what can be thought and felt in ways we may not
have thought or felt before, but now discover, or in ways that we may have, and now
recognize and find felicitously expressed.19 Poetry thereby enlarges our own potentialities
for thought, feeling, and expression. Epic and dramatic poetry in addition reveal to us

Anna Christina Soy Ribeiro

the potentialities of action, character, circumstance, and outcome, and how all of these
inform and influence one another.
It was because he recognized in poetry all these powers that Plato exiled most poets
from his ideal city. Neither Plato nor Aristotle were concerned solely with meaning.
Indeed, the effects of poetic tropes and schemes on listeners was something marvelous
(and ‘natural’) for Aristotle and dangerous because effective for Plato. Poetry is made of
words, and words can be used for good or ill: clearly, what could effect nefarious
consequences could also produce beneficial ones, and the moral educational value of
poetry was not lost on Plato, who, eventually and somewhat begrudgingly, admitted
poetry into the city so long as it could be cleansed of any of its (as he saw them) reprehen-
sible aspects. Contemporary philosophy of literature has largely focused on the potential
value of novels as instruments of moral education,20 but voices begin to be heard that
defend the value of lyric poems in that regard.21 Indeed, if it is true that often the lyric
invites a kind of identification, of ‘trying on’ another’s thoughts and feelings (as I have
argued elsewhere),22 then its potential for moral education is perhaps even greater than
that of novels. Can poems memorized in school, for instance (what was a common
practice until a couple of generations ago in many countries), serve as guides later in life,
as reminders of what is important or suggestions for how to construe a particular situa-
tion? Even if the first-person lyric is found not to be better than novels when it comes
to moral education, there seems to be no reason to think it might fare worse; we may
conclude that their relative potentials in this respect lie simply in different approaches.
That said, the therapeutic value of lyric poems in moments of difficulty is hardly to be
doubted;23 if that value may serve as a springboard to ethical edification, or, more
broadly, a humanistic enrichment, then it seems there is a special value to be accorded
to the lyric in that regard. If so, this would constitute a return to poetry’s original value,
insofar as it was, from its beginnings, regarded as the repository of the wisdom of the
culture whence it emerged—the role, as we saw at the opening of this essay, challenged
by Plato and Socrates.
Contrary to what may be expected, then, one may surmise from these pages that
philosophers have yet a great deal to say about poetry, and not only about the age-old
question of whether it can be the conveyor of truth. The answer to that, as the examples
discussed herein reveal, is clearly yes. For its part, poetry, given its syntactic, semantic,
and phonetic peculiarities, which in turn open up avenues for thought, feeling, and
expression, may help illuminate existing philosophical debates as well as generate new
topics for philosophical discussion once we pay it the theoretical attention it deserves
as the ancient and universal human practice that it is.

  1 Two notable exceptions are Sir Philip Sidney, An Apology for Poetry (1595) and Percy Bysshe Shelley,
A Defence of Poetry (1840), both found in Critical Theory Since Plato, edited by H. Adams and L. Searle
(Boston: Thomson Wadsworth, 2005), pp. 186–206 and 538–41 respectively.
  2 An Essay Concerning Human Understanding, Edited, with an Introduction, by Peter H. Nidditch
(Oxford: Oxford University Press, 1979 [1690]), §34, “Of the Abuse of Words,” p. 508. O. J. Todd.
Loeb Classical Library 168 (Cambridge, MA: Harvard University Press, 1997), pp. 559–60.
  3 Essay Concerning Human Understanding, “Book III, Chapter X, of the Abuse of Words,” London, 1690,
p. 34. As Susan Haack has noted, “if figurative use of language is indeed, at least where ‘dry truth and
real knowledge’ are concerned, an abuse of language, then it is an abuse of which Locke himself is hardly
innocent. . . . And Locke’s use of metaphor is not always . . . purely in the service of vividness; certain


metaphors play a role in his philosophy much deeper than mere picturesqueness of speech”; most notable
among these is Locke’s metaphor of the mind as a tabula rasa at birth. Locke is not alone on the list of
philosophers who deplore figurative language while making important use of it in their works: Haack
further amuses us with passages from Thomas Hobbes and John Stuart Mill, “Dry Truth and Real
Knowledge: Epistemologies of Metaphor and Metaphors of Epistemology,” pp. 69–71; Manifesto of a
Passionate Moderate: Unfashionable Essays (University of Chicago Press, 2000), pp. 69–89. I am grateful
to Paisley Livingston for bringing Haack’s essay to my attention.
  4 A. J. Ayer, Language, Truth and Logic, 2nd ed. (Dover, 1952), p. 113.
  5 Ernie Lepore and Matthew Stone, “The Poetic Imagination,” this volume, Chapter 28, pp. 323–33.
  6 Aristotle, Poetics I with Tractatus Coislinianus, A Hypothetical Reconstruction of Poetics II, The Fragments
of the On Poets, translated with notes by Richard Janko (Indianapolis/Cambridge: Hackett Publishing
Company, 1987), p. 12.
  7 Critique of Judgment, translated by W. S. Pluhar (Indianapolis: Hackett, 1987 [1790]), §53, p. 327
  8 Hegel’s Aesthetics: Lectures on Fine Art, translated by T. M. Knox (Oxford: Clarendon Press, 1975), Vol. I,
pp. 89–90. See also Vol. I, pp. 82–90 and Vol. II, Chapter III.
  9 The World as Will and Representation, translated by E. F. J. Payne (New York: Dover Publications, 1958),
Vol. II, p. 424. See also Vol. I, §51 and Vol. II, §37.
10 “After the pupils had memorized a text by repeating it word for word, time and again until it was ‘clasped
to their bosoms’ (until they had learned it by heart), the teacher would provide them, orally, with the
illustrations, demonstrations and calculations that the text concealed. This was the key that opened up
the paths of knowledge and an instrument of spiritual fulfilment [sic].” Francis Zimmerman, “Lilavati,
Gracious Lady of Arithmetic.” UNESCO Courier (November 1989), p. 21.
11 Regimen Sanitatis Salernitanum: Code of Health of the School of Salernum. Translated into English Verse,
with an Introduction, Notes, and Appendix by John Ordronaux, LL.D., M.D. (Philadelphia:
J. B. Lippincott & Co, 1871). The original Latin and the translation that follows are on pp. 46 and 47
12 A Critical Edition of Le Regime Tresutile et Tresproufitable pour Conserver et Garder la Santé du Corps Humain
by Patricia Willet Cummins, published by the North Carolina Studies in the Romance Languages and
Literatures (Chapel Hill, 1976).
13 The negative opinion of the Leonine verse in which the poem is set is due to the fact that it is
rhymed. Classical Latin verse was not rhymed; the Leonine rhyme is a medieval invention. Ordronaux
(supra), p. 12.
14 See http://www.npr.org/2015/01/15/377527029/rapping-the-news-in-west-africa (accessed August 12, 2015).
15 Interestingly, this is precisely the opposite of what Haack suggests is the epistemological usefulness of
metaphors (which we may extend to figurative or poetic language in general), namely, that they “can
be the training wheels of inquiry” (“Dry Truth and Real Knowledge”, p. 87). For Haack, “a
metaphorical presentation can be helpful to the goal of instruction if it makes a theory comprehensible
to an audience unfamiliar with its technical vocabulary, or insufficiently sophisticated in its logical,
mathematical, or experimental techniques, to understand it in a literal presentation” (ibid., p. 81).
But as we see from several of the examples given here, it was literal language, in the form of commentaries,
which was used so that one could understand the metaphors, not the other way around (cf., e.g.
interpretations of biblical verses). The Code of Salerno seems to fit Haack’s picture a bit better, although
it is not meant as an exploratory stage to something more abstract or difficult; it is simply what is
meant to be conveyed.
16 Http://www.poetryfoundation.org/bio/john-koethe (accessed September 7, 2015).
17 Excerpt from “One Art” from Poems by Elizabeth Bishop, published in 2011 by Farrar, Straus and Giroux,
LLC, and Chatto & Windus. Reproduced by permission of Farrar, Straus and Giroux, LLC, and The
Random House Group Ltd.
18 See Peter Ludlow, Living Words (Oxford: Oxford University Press, 2014).
19 For more on poetry serving as a vehicle for our thoughts and feelings, see Anna Christina Ribeiro,
“Toward a Philosophy of Poetry,” Midwest Studies in Philosophy 33 (2009), pp. 61–77 and “Heavenly Hurt:
The Joy and Value of Sad Poetry,” in Suffering Art Gladly: The Paradox of Negative Emotions in Art, edited
by Jerrold Levinson (Palgrave-Macmillan, 2014), pp. 186–206, as well as Kendall Walton,
“Thoughtwriting—In Poetry and Music,” New Literary History 42 (2011), pp. 455–76. For more on
poetry offering us new possibilities for thought and feeling, see Eileen John, “Poetry and Directions for
Thought,” Philosophy and Literature 37: 2 (2013), pp. 451–71.

Anna Christina Soy Ribeiro

20 See Martha C. Nussbaum, Love’s Knowledge: Essays on Philosophy and Literature (Oxford: Oxford
University Press, 1990); for the opposing view that we do not learn ethics from novels, see Gregory
Currie, “Does Fiction Civilize Us?”, The New York Times, June 2, 2013, p. SR12. Available online as
“Does Great Literature Make Us Better?”, at http://opinionator.blogs.nytimes.com/2013/06/01/does-
great-literature-make-us-better/?_r=0 (accessed August 28, 2015).
21 See Bridget Vincent, “The Example of Poetry,” Philosophy and Literature 37: 1 (2013), pp. 53–71.
22 See references in Note 19 above.
23 “Heavenly Hurt,” pp. 197–203.

Susan L. Feagin

1 Play Scripts and Dramatic Literature

The earliest known reference to reading a play for pleasure occurs in Aristophanes’ The
Frogs, first presented in 405 BCE, where Dionysus says that he was sitting on the deck
of a ship one day, “reading once more that dear Andromeda,” a (now lost) play by
Euripides. Assuming that what he was doing was not unusual at the time, we can
conclude that people have been reading plays for at least 2,400 years. Indeed, public
demand from readers has long driven the publication of scripts—legitimate and
pirated—in the Western world.
There are two main ways to identify what playwrights write: as a particular type of
literature, dramatic literature, and as scripts. As dramatic literature, the written works
that playwrights produce are conceived on a par with novels, short stories, travel essays,
and dialogues, which are all literature in being written to be read in the sense that they
are written to be appreciated in the reading. Scripts, in contrast, are written to be
performed. Written scripts are thus like some other categories of written works in being
written for another form of presentation. Examples are screenplays, which are to be
filmed; song lyrics, which are to be sung; and speeches, which are for oral delivery.
Dialogues—such as those written by Plato, Hume, and Berkeley—are a type of literature
written in the basic form of a play, so that they look to an extent like scripts, but to be
read and appreciated in the reading rather than to be performed. It has been argued that
Percy Bysshe Shelley’s Prometheus Unbound is in fact a dialogue and not a script.
Some works are written both to be read and to be performed, both as dramatic litera-
ture and as script. Many playwrights through Western history—such as, notably, George
Bernard Shaw and Ben Jonson—have expressed frustration with the incompetence of
performers and the appreciative incapacities of audiences, and as a result welcomed print
publication of their works, usually altered somewhat by editorial or their own authorial
intervention designed to enhance the experience of readers (Wells 1991). Shaw’s reader-
directed intentions are clearly manifested in the wit and style of the scene-setting at the
beginning of each act and in sketches that accompany the first appearance of each
character, features he notes are needed in what he identifies as the art of writing plays
to be read, as opposed to the art of the “practical dramatist” who has an eye to the stage
(Shaw 1963 [1898]: xxvi). Given that frequently, for obvious reasons, only the printed
versions produced for readers have been preserved, it is unfortunately often difficult to
Susan L. Feagin

tell if original scripts from earlier eras differed from preserved versions and, if so, what
they would have been like.
Shaw identified two arts of writing plays (I use “plays” as a neutral term between
dramatic literature and scripts); in contrast, it is currently a widespread philosophical
view that if a written play is a work of art, it is in virtue of being literature, like a novel
or short story, as something to be read and appreciated in the reading (e.g. Carroll 2001;
Hamilton 2007; Nannicelli 2011). The idea that playwrights are artists who write plays
that are dramatic literature has been well entrenched since the early twentieth century
(Worthen 2010). Philosophies of literature have generally followed suit, with their
various accounts—of the ontology of the work, of a work’s meaning, of the ontological
status of fictional entities, of relevance of emotional responses, and of criticism and
evaluation—taken without argument to apply to plays as well as to novels and short
stories. If written plays are artworks only insofar as they belong to a particular genre of
literature, dramatic literature, their merits and defects would have to be of a sort that
can be appreciated in reading them. And it is certainly undeniable that one can appre-
ciate many qualities of plays to a significant degree by reading them: expressive,
figurative, and other aesthetic qualities of the language, artistic control over character
development and how the plot unfolds, and all in relation to the work’s major and
minor themes.
Yet, it does not follow from the fact that one appreciates a given quality of a written
work by reading it that one is appreciating it as literature, since one may not be appreci-
ating it as something written to be read. One might instead read a script with a
sensitivity to the performance potential of the use of language, the way the characters
develop and how the plot unfolds. After all, before a play is performed it typically is read
in order to see if it is worth performing. Playwriting is subject to special demands due to
the constraints and potentialities of performance, described copiously in books and
courses on the subject, and playwrights typically welcome opportunities to “workshop”
their plays with theater companies and to do talk-backs with an audience to see
how their scripts work out on the stage and how audiences respond to them. For
example, the pace of the action needs to accommodate the fact that the audience is
trapped in their seats for a set period of time and unable to speed things up or slow them
down to their own taste. One way to keep the audience interested is to employ dialogue
that raises questions or suggests hypotheses about what will happen subsequently, what
are sometimes called “forwards.” It is a crucial point, however, that it may not be possible
to tell just by reading a script whether, when performed, the dialogue, use of language,
and unfolding of the action will meet the demands of a “captive” audience. The fact is
that a script may have merits (or defects) as something written for performance that are
not merits (or defects) as something to be read, and vice versa. It follows that relevant
types of considerations for the appreciation and evaluation of scripts and of dramatic
literature may be different. A play may be appreciated and judged as different kinds of
things, as something to be performed (i.e. as a script), or as dramatic literature (as what
at least some scripts also are). There is an additional question of whether to judge a work
as a script is to judge it as a work of art.
In what follows, I discuss considerations advanced in favor of and against reading
written plays as dramatic literature and as scripts. In Section 2, I describe how
Aristotle and the New Critics have been taken to support the view that written plays
are to be appreciated as dramatic literature, that is, in reading them. In Section 3, I
discuss a more general, empirical argument in support of this view, based on the idea


that reading as dramatic literature maximizes a work’s potential for appreciation. In

Section 4, I explain some difficulties and possibilities for appreciating scripts as works
of art.

2 Theory-based Views: Aristotle and New Criticism

According to Aristotle in the Poetics, whether a tragedy is a good tragedy depends on
whether it produces its “proper effects.” These clearly include pity and fear, though he
also mentions empathy with the suffering of the characters and the catharsis of these
emotions (whatever, notoriously, a catharsis might be). Tragedies can produce their
proper effects, he claims, through plot, development of character, “thought”—the
presentation of an argument or opinion in the dialogue—and metered language or
diction. The “manner” of presentation for tragedy is “live performance” (Poetics,
1449b30–32) in a theater (the Greek word for which is related to the word for watching),
so audiences would see the performers and their movements, masks, and costumes
(though, given the distances that audience members might be from the stage, they
would not necessarily have seen them clearly). Nevertheless, in several passages—
notably in Chapters 6, 14, and 26—Aristotle discounts the importance of acting
(or “movement”) and spectacle (basically the visual components of a production) for
producing the proper effects of tragedy. His dismissal of acting and spectacle has led
some to interpret him as supporting the view that to appreciate tragedies we should read
them as dramatic literature rather than see them performed.
To see whether Aristotle supported such a view, and whether his reasoning can plau-
sibly be extended to our own circumstance, it is useful to look at the scope and context
of his claims. First, the topic of the Poetics is not the theater in general but the particular
genre of tragedy. It is unlikely that he intended his claims about acting and spectacle to
be extended to all forms of theater, especially if one includes what Aristotle identified
as its “lower” forms, such as comic lampoons of actual people. A great deal of comic
effect in his day as well as our own is due to clowning and use of props. Many comedic
portions of Aristophanes’ plays, for example, are incomprehensible without the assump-
tion of certain sorts of stage action. There is an indication in Chapter 4 (9b21–22) that
he has an analysis of a form of comedy as an art analogous to what he provides for trag-
edy, but since his work on comedy is lost except for fragments, this is a matter for dispute
(Halliwell 1986: 268, 271).
Second, the topic of the Poetics is the nature of the writer’s art. It is Aristotle’s response
to Plato’s charge that there was no “techne” or systematic art, involving knowledge and
understanding, that enabled the tragic playwrights to do what they did well. As an
account of the art or techne of the tragic playwright, the Poetics is concerned with what
is under the logical or rational control of the playwright, which might theoretically
include making a space for certain kinds of stage action, but would not include specifically
what makes such action effective, such as acting or costuming. He claims in Chapter 6
that spectacle is not part of the art of poetry (1450b16–17) and, in Chapter 14, that
spectacle requires material resources and is less artistic (1453b7–8). From the perspec-
tive of the writer, it is also useful to keep in mind that Greek tragedies of the fifth
century BCE were written for competitions and the winners were selected for presenta-
tion at civic or religious festivals (though performance contexts during Aristotle’s own
time were more varied). Judges made their selections by reading submissions and not by
seeing them performed.

Susan L. Feagin

A third consideration is that Aristotle’s claim that a tragedy’s quality can be judged
by “reading” it (anagignõskõ, 1462a12, 1462a17) should probably be understood as “read-
ing aloud.” In Chapter 14 Aristotle explicitly contrasts seeing and hearing (rather than
reading) a play when he discounts the role of acting and spectacle in relation to seeing.
“Reading aloud” includes possible effects of the words as spoken and eliminates a sig-
nificant indeterminacy of the written word: the way in which the lines are read.
Nevertheless, poets of the fifth century BCE neither expected nor depended on special
talents of professional performers to bring the lines to life. Performers of that era were
not professional actors but amateurs or “ordinary citizens” who would not have engaged
in the myriad acting styles available today. By Aristotle’s time, however, in the fourth
century BCE, there were professional actors who performed speeches from the plays in
their own interpretive style. Aristotle seems to be steering his readers away from the
relevance of such practices for understanding tragedy’s “true nature.”
A fourth and a related point is that Greek tragedy is written in meter, which when
read aloud establishes the rhythms of the speeches, which are embedded in traditions of
delivery for metered verse. Different meters were associated with different emotional
tones and hence the ability to write in meter carried an important share of the burden
of ensuring a tragedy fulfills its emotive function, something that is typically lost in
English translations. In sum, Aristotle appears to have assumed that the tragic play-
wrights were writing for performance. His goal of establishing the tragic playwright’s art
as a “techne,” as something governed by knowledge and understanding, placed acting,
movement, and visual elements of performance outside the art of the poet—even though
it was assumed that the mode of access to the words would have been through hearing
them. The way such factors shaped his argument should caution against generalizing to
other types of theater or to contemporary drama as read to oneself on the basis of the
claims he makes.
Closer to our own day, originating in the 1930s, New Criticism rose up to challenge
what its proponents saw as an excessive use of biographical details about an author, often
gossipy and prurient, to aid in interpreting a work. New Criticism also filled the desire
of many literary scholars to have a distinctive methodology for reading literature—one
that would define the discipline, involving a close reading of the text in a way that
divorced it not only from authorial biography but also from political, social, economic,
and other contexts. In Understanding Drama (1945) Cleanth Brooks, one of the founding
fathers of the movement, along with Robert Heilman, applied the methods of New
Criticism to drama, explaining how to read and appreciate plays as literature, divorced
from the biographical and contextual fact that they are written to be performed.
One problem this approach raises is how to read the non-dialogic components of a
script, such as speech prefixes, stage directions, and act and scene divisions. They are
clearly not expendable, but they are rarely “literary,” with the primary exceptions being
works written, like those of Shaw, for both readers and performance. Brooks and Heilman
propose simply that such components should be kept to a minimum. Drama as a genre
of literature is effectively taken to be one where certain constraints are placed on how
plot and character are allowed to develop: by the quality of the dialogue alone, with
minimal assistance from the occasional phrase indicating where the action takes place
or how a line is uttered. The assumption seems to be that authors are not responsible for
the literary quality of writing in the non-dialogic components of the script, though the
rationale for this assumption is elusive, given that these components enable readers to
make sense of at least some of the dialogue. It is hard to get around the fact that scripts


are written for performance and hence would reasonably include these types of instructions
that, as components of scripts, are to be judged by their usefulness for performance rather
than for being read and appreciated just in the reading of them.
The idea that non-dialogic features of a script should be kept to a minimum in order
to meet the demands of literature is also tested by counter-examples in the form of good
plays that have a great quantity of mundanely written stage direction and comparatively
little dialogue. For example, Eugène Ionesco’s Amédée, Or How to Get Rid of It (1954)
is an outstanding play but contains an abundance of stage direction that is difficult to
process when read (and difficult to produce: Ionesco even provided an alternative
ending in case a theater did not have the technical apparatus to carry out the staging as
written; one wonders how to treat the alternative ending when reading it “as literature”).
Samuel Beckett wrote two plays that have no dialogue at all.
Another attempt to deal with non-dialogic components of scripts denies that they
are necessary or essential components of scripts at all. For readers, this is difficult, since
these components are often crucial to understanding not only how the dialogue is to
be read, but also what is happening while the characters speak. For directors, however,
the situation is different. It is common to treat stage directions as merely advisory, in
principle expendable, when mounting productions of plays. A director can have the
benefit of reading the stage directions and then discarding them in favor of something
else. In 1984, JoAnne Akalaitis of the American Repertory Theater (ART) attempted
to stage a production of Samuel Beckett’s Endgame (1957) set in a tunnel. When sued
by Beckett for not abiding by the license granted to produce the play, ART argued that
the stage directions were not necessary or essential elements of the script and hence
that the choice of setting was not controlled by the license. An agreement was reached
out of court, depriving us of the knowledge of how the law would have settled the matter.
In any case, the expendability and in principle replaceable character of such compo-
nents is congenial to staging and stage practices, and awkward at best for reading, whose
practices do not typically include reading something and then ignoring it in favor of
something one thought of that is “better.”

3 The Maximization Argument

Aristotle and the New Critics may be seen as providing broadly theory-based consider-
ations that have been used to bolster the view that plays are to be read as literature. A
different type of argument relies on the empirical claim that reading plays, compared
with viewing performances, maximizes their potential for appreciation. This potential
has been facilitated at least since the turn of the eighteenth century in Europe by stan-
dardization of the appearance of plays on the page—a list of characters at the beginning,
the use of speech prefixes, stage directions set off typographically in some way, and
divisions into acts and scenes (Worthen 2010: 3). In addition, as already indicated,
many skills employed in reading literature—such as sensitivities to plot and character
development, use of imagery, and so on—are employed in appreciating a play. Finally,
the potential for appreciation when reading a play is thought to be greater because of
the nature of the process of reading itself: reading at one’s own pace, easily refreshing
one’s memory about what went before (with respect to plot, character, use of metaphor,
different speech styles for different characters, and other qualities of language), and
rereading portions or the whole multiple times at will and quickly or slowly as the need
arises. In contrast, a performance occurs at a fixed speed and audience members are

Susan L. Feagin

unable to review what went before to compensate for a momentary lapse of attention or
inadequate memory. Repeat viewings may be inconvenient or impossible, separated by
time spans that also risk a fading memory. Performances themselves may be inept or
interpretatively shallow. It would seem that there is everything to gain and nothing to
lose by reading as opposed to watching a play, except for the visual pleasures of the
staging itself. In short, reading maximizes the play’s appreciative potential.
The maximization argument does not recognize variations among different types of
appreciators: some people may be more visual or auditory than others, and hence better
able to appreciate performances than what they read. But let us assume for the sake of
argument that such persons are relatively rare and that we can in any case abstract from
such differences. Control over speed and accessibility of parts of the play for readers as
opposed to viewers of performances are undeniable. But we should regard other alleged
advantages of reading with some suspicion. The extensive merits of a play like Hamlet,
for example, cannot be appreciated in a single reading any more than in a single viewing.
Actors’ movements, interactions with other characters, and vocal inflections (along
with set design and costuming) can clarify what is merely written on the page, providing
insight into the character who says such a thing, highlighting points that reinforce an
image or theme, and supplying a pace and rhythm that keep the action moving. Multiple
viewings may reveal as much as or more than multiple readings: different productions of
the same play often present different interpretations of it, including interpretations you
might never discover for yourself no matter how many times you read the script. They
also may provide different points of emphasis for the same interpretation, with one
production revealing its force and significance in a different way from others. Thus, even
if reading enhances appreciation in some respects, it does not follow that it is an
unalloyed benefit or a self-sufficient replacement for viewing a performance.
There has been a long history of denigrating performers, performances, and
spectators—what has come to be known as the anti-theatrical prejudice—that feeds
into inflating the importance and sufficiency of reading (Barish 1981). One source of
this prejudice is the equally long history of taking vision and visual imagery to be
epistemically inferior to the written word: less reliable as a source of knowledge, less
susceptible to the operations of reason, less capable of conceptual sophistication, and
more prone to co-option by irrational emotions. A related factor is that, historically, the
visual and aural components of theatrical performances, what Aristotle called “spectacle,”
appealed to the uneducated and served as a “second best” means of access to the play for
those who couldn’t or wouldn’t read. Another source of prejudice is more overtly social:
for major periods of human history, professional actors came from the lower socioeco-
nomic classes and were not considered “respectable.” They might also have been
illiterate. An exploration of these claims goes beyond the scope of this essay, but it is
useful to remember that there continues to be room underground for vestiges of views
that are “officially” discredited, as in this case, to feed skepticism about the status and
sophistication of performers, performance, and the visual as opposed to the written.
From the 1950s and 1960s, performance artists of assorted stripes—with backgrounds
in the visual arts, poetry, dance, theater, and music—experimented with varieties of
performances that demonstrated their independence from the written word, both as a
source—insofar as they were unscripted or improvisational—and as content, since spoken
words could play a minor role in the performances themselves. The rise of “performance
studies” in the 1970s and 1980s championed the idea that performances are autonomous
works of art that need not be tied to scripts. Traditional scripts came to be used in a variety


of non-traditional ways, such as, on the more conservative side, giving speeches to
characters other than those indicated in the script and, more radically, cutting up the
lines of the text to be recited in random order. Words have also been used to generate
music or movement instead of speeches uttered by characters (Hamilton 2007: 43–50).
The recognition of performance as an “independent” art form has a tendency to
bifurcate performing and writing as forms of theater as an art, with the former constituting
the art of theater and the latter a literary art. Scripts, if not dramatic literature, become
a mere handmaid to performance. Along these lines, the philosopher James Hamilton
(2007) proposes that a script is only one potential ingredient of a performance; plays
may be read as dramatic literature and appreciated as such, but do not have any special
status in the art of theater. Neither dialogue nor non-dialogic components of scripts have
any special weight when it comes to performing the play that is scripted, even if they do
for reading the play as literature. The script does not provide a “standard of fidelity” for
performances and is not itself to be evaluated as a component of the art of theater.
According to Paul Woodruff (2008), theater is the art of watching or of making one’s
actions worth watching; these are arts of performing, not writing.
In response, it should be observed that performances may be generated by using
scripts in non-traditional ways, but it does not follow that what takes place is a perfor-
mance of the play that was scripted. Celebrating performances as art is compatible with
celebrating scripts, written for performance, as art, as long as there are some constraints
on what qualifies as a performance of the play that is scripted. Nevertheless, many
questions surround whether and how one can appreciate a script—something written
not as an end in itself but to be performed—as a work of art. In particular, are there roles
for both reading and performance in this process, and what might they be?

4 Written for Performance

Scripts are generally taken to be sets of instructions for putting on a performance. I
assume in what follows that when something is written for performance it is written
to be performed for an audience, rather than to be performed for the sake of the
performers or for the sake of the performance itself. This is merely for purposes of
clarification; we should not foreclose the possibility that some scripts are written to
be appreciated not merely by an audience but also by performers, especially in what is
known as “participatory theater,” though there the distinction between performer and
audience tends to get dissolved.
There is a general problem about how to appreciate something as itself a work of art
when it is created not as an end in itself but to be used to generate something else. As
indicated in Section 1, a fundamental problem with appreciating scripts by reading them
has to do with the exigencies and potentialities of performance in general. Features of a
script may just not work on stage no matter how effective they are when read, and others
may work on stage but not read well. Trying to tell whether a script is good just by reading
it is a bit like trying to tell if the spout of a teapot will drip just by looking at it. Sometimes
you can tell by reading or looking, even though it was not made to be read or to be
looked at. Past experience may be beneficial in both cases: a practiced eye and hand may
see and appreciate more in the spout than an unpracticed one, just as a theatergoer with
much experience of works in the same genre or by the same author (or performed by the
same company: there are many relevant categories for appreciation) may be able to
appreciate various qualities by reading it.

Susan L. Feagin

But the analogy goes just so far. Ultimately, the proof of the teapot’s spout is in the
pouring, but the proof of a script is not in the performance. Performances cannot
be relied on to show what needs to be shown, even beyond the possibilities already
mentioned of bad acting, bad set design, bad direction, and bad interpretation.
Performances cannot be relied on because, with few exceptions, scripts find their way
into performance via productions. Productions are sets of decisions about how to interpret
a script, how actors will deliver their lines, how they will stand and move around the
stage in relation to each other and in relation to the audience (“blocking”), what the
sets and costumes will look like, how the sound and lighting will be handled, and so on.
Productions are typically in development anywhere from a month to about a year prior
to a run of performances. During this time the various aspects of a production—costum-
ing; set, sound, and lighting design; as well as acting—can be thought through in detail
so that each component is coordinated with the others and its contribution to the whole
understood. In fifth-century BCE Greece, the playwright generally oversaw rehearsals
for a single performance given at a civic or religious festival. A mere century later, plays
were being “interpreted” for personal and political purposes—so much so that Lycurgus
of Athens had official versions of Euripides, Aeschylus, and Sophocles placed in a public
depository and forbade performances that incorporated any changes. He went so far as
to commission a clerk whose job it was to read along with performances to ensure
compliance (Lives of the Ten Orators: 841–843). Since the late nineteenth century, it has
been common practice for a production to be overseen by a director who is responsible
for coordinating its various elements, and who may also be seen as an artist.
It is considered standard practice in productions to reinterpret scripts and to treat
them as malleable. The practice is not because performers want to treat scripts as mere
ingredients for the performing arts. Productions are developed over time and provide for
the creative integration of elements of the script in different ways. Further, the fact that
productions are developed at different times and places opens up opportunities for irony
and reference to events unknown to or unintended by the author but fresh in the minds
of the director and the audience. Thus, in reinterpretations, the action may be set at a
different time or place and references may be added to current events or psychological
or sociological theories that the playwright could have known nothing about. In treating
scripts as malleable, lines may be added or deleted here and there, and scenes rearranged.
Such alterations may serve a reinterpretation or alter the pace, shape of the action, or a
more localized interpretation of a character or contribution of a line. A production may
also take advantage of the fact that it will be performed in a place with an explicit or
latent social, political, or cultural significance for a group of people at a given time. In
1944 Nazi-occupied France, Jean Anouilh rewrote Sophocles’ Antigone with a focus on
resistance to power. In 1965, the South African Serpent Players Theatre Company
performed Sophocles’ version in England; a member of the company, Norman Ntshinga,
was later arrested and sent to South Africa’s notorious Robben Island, where he
performed it for his cell mates. The venue hardly functioned as a neutral frame for a text
whose meaning was independently given, but referenced the earlier production and
itself, highlighting the contrast in freedom and resistance to authority. The creative use
of time and place and treatment of reinterpretation and malleability as standard is not
arbitrary; it recognizes the power that lies within live performances as events that occur
at given times and places. Performances of productions can thus become social or
cultural events that play roles in the art world and in the larger community.


Not all scripts have been written with the expectation that performances will be
mediated by productions. It has been argued that professional practice in Shakespeare’s
day did not include mounting productions: actors had little time to learn their lines
(at most a week); they chose their own costumes and coordinated fight scenes
among themselves; performances were one-off and not in a run as with today’s pro-
ductions; and there was at most one full run-through before performing in front of
an audience (Stern 2000). Perhaps most astonishingly, each actor had only his own
lines, his own “part,” and not the entire script. Numerous aspects of Shakespeare’s
scripts—such as the use of cue words, the handling of entrances and exits and
deaths—read quite differently when read as something to be performed under the
constraints of Elizabethan stage practices rather than simply as something to be
read. Other aspects, such as the use of imagery and the poetry of the speeches, can
be appreciated both in reading and when attending a performance. In Shakespeare’s
world, the systematic reinterpretation and malleability associated with productions
did not exist. Quite the contrary, given that actors did not have access to each
other’s lines and were dependent on specific cue words to know when they were to
speak, departing from the scripts could bring the entire performance to a halt.
Though it may still be difficult to read a script and appreciate how it would work in
performance, in a world without productions it is reasonable at least to think of a
play script as something to be performed as written, within an expected range of
relatively constrained possibilities for performance.
This is not our world, where the practice is for performances not to be of scripts but
of productions and where productions may involve not insignificant alterations of the
original script. But the fact that productions treat scripts as reinterpretable and
malleable—not strictly as instructions but, I would suggest, as “somewhat advisory”
instructions for performance—has implications for the importance of reading them as
scripts, even if not for reading them as literature. In particular, the practice of treating
scripts as reinterpretable and malleable increases the value of reading scripts, both for
appreciating their own qualities and for appreciating productions and performances of
them. With regard to their own qualities, there may be features of language and char-
acter and plot development that end up not seeing the light of day in a particular
production. This is not to bemoan the inadequacies of such productions; quite the
contrary, it is because productions and performances of productions have themselves
become (potentially) works of art that they cannot be assumed to manifest the qualities
we might appreciate in the script as employed in some other production and its per-
formances. Reading a script can also help appreciation of the artistry of a production
by knowing what is not in a script, though productions may borrow from other
productions, so an absence in the script does not ensure the creative insight of any
particular production.
Given the practice of mounting productions, there is no general description of how
to appreciate a script—as something written for performance—the way there is to
appreciate a literary work. For the latter, there is a determinate mode of access to the
work: reading it. Even without a determinate mode of access, however, it does not
follow that scripts could not be works of art. The reading of scripts and the viewing and
hearing performances of productions are mutually supportive, and we do not need to
pretend that appreciation achieved through either mode will be maximized by ignorance
of the other.

Susan L. Feagin

Aristotle, Poetics, ed. and trans. S. Halliwell (1995), Cambridge: Harvard University Press, Loeb Classical
Library 199.
Barish, J. (1981) The Anti-Theatrical Prejudice, Berkeley: University of California Press.
Brooks, C. and Heilman, R. (1945) Understanding Drama, New York: Henry Holt.
Carroll, N. (2001) “Interpretation, Theatrical Performance, and Ontology,” Journal of Aesthetics and Art
Criticism 59 (Summer): 313–316.
Halliwell, S. (1986) Aristotle’s Poetics, London: Duckworth.
Hamilton. J. (2007) The Art of Theater, Malden, MA: Blackwell.
Lives of the Ten Orators, trans. C. Bancroft; rev. W. Goodwin (1878), www.attalus.org/old/orators1.html
(accessed February 7, 2012): 841–843.
Nannicelli, T. (2011) “Instructions and Artworks: Musical Scores, Theatrical Scripts, Architectural Plans,
and Screenplays,” British Journal of Aesthetics 51 (October): 399–414.
Shaw, G. B. (1963) “Preface to the First Volume of Plays: Pleasant and Unpleasant,” [1898], Complete Plays
with Prefaces, vol. III, New York: Dodd, Mead.
Stern, T. (2000) Rehearsal from Shakespeare to Sheridan, Oxford: Clarendon.
Wells, S. (1991) “To Read a Play: The Problem of Editorial Intervention,” in Scolnicov, H. and Holland, P.,
Reading Plays: Interpretation and Reception, Cambridge, UK: Cambridge University Press.
Woodruff, P. (2008) The Necessity of Theater: The Art of Watching and Being Watched, New York: Oxford
University Press.
Worthen, W. B. (2010) Drama: Between Poetry and Performance, Malden, MA: Wiley.

Further Reading
Carroll, N. (2010) “Philosophy and Drama: Performance, Interpretation, and Intentionality,” in Carroll, N.,
Art in Three Dimensions, Oxford: Oxford University Press.
Davies, D. (2009) “Works and Performances in the Performing Arts,” Philosophy Compass 4/5: 744–755.
Irvin, S. (2009) “Theatrical Performances and the Works Performed,” Journal of Aesthetic Education 43 (3):
Wolterstorff, N. (1975) “Towards an Ontology of Art Works,” Nous 9: 105–142.

Aaron Meskin

It is common to mark a distinction between popular fiction and literary fiction or, as it
is sometimes called, “serious literature”. (Two clarifications: Serious literature may
include work that is not fictional, but I shall ignore this complication for the remainder
of the essay. And although the category of fiction includes works in a wide variety of
media and forms, I shall, unless otherwise indicated, use the term to refer to prose
fiction only.) Dan Brown, E.L. James, Suzanne Collins, and James Patterson are
paradigmatic producers of the former while Jane Austen, James Joyce, Vladimir
Nabokov, and Virginia Woolf are well-known creators of the latter. Given such exemplars,
it must seem the distinction picks out a real and important difference between two
categories of writing.
Appearances, however, are often deceiving, and not all distinctions we commonly
mark are created equal – some are redundant, others are non-substantive, still others
may be pernicious (e.g. by reinforcing illegitimate hierarchies). What, then, about the
distinction between the popular and the literary? The first section of this chapter
explores this distinction and offers reasons to think that it is non-redundant, robust,
and benign, but that the two categories are neither mutually exclusive nor exhaustive
of the sphere of fiction. The section also briefly addresses the relationship between
popular fiction and some broader categories such as mass art, popular art, “low art”,
popular culture, kitsch, and entertainment. The second section of the chapter focuses
on aspects of the criticism of popular fiction – specifically the interpretation and
evaluation of works that fall into that category. Does the interpretation of popular
fiction differ significantly from the interpretation of literary fiction? I will express some
skepticism about such a view. Perhaps relatedly, popular fiction is commonly thought
to be less valuable than literary fiction. Are there distinct standards of evaluation
appropriate for popular fiction? Can such works be coherently compared with indi-
vidual works of literary fiction and, if so, are they always inferior? These are difficult
questions, but I shall argue that popular fiction and literary fiction are not evaluated in
radically different ways and that we should be careful of being too quick to dismiss the
value of popular fiction. The third section explores what Noël Carroll has called
“the paradox of junk fiction”, which arises because we seem to read popular fiction for
the sake of stories while at the same time being aware of how the story will, at least in
broad terms, turn out. The fourth section briefly addresses the relationship between
popular fiction and genre.
Before I address the first substantive topic, let me say something about terminology.
Why “popular fiction” and not the aforementioned “junk fiction” or such other common
Aaron Meskin

terms as “genre fiction”, “pulp fiction”, “commercial fiction”, “general fiction”, or “mass
market fiction”? I use “popular fiction” rather than “junk fiction”, a term popularized by
Thomas J. Roberts in his book An Aesthetics of Junk Fiction (1990), to avoid the latter’s
negative connotations. If there is something problematic about popular fiction it is a
substantive matter – we should avoid building the problem into the term itself. On the
other hand, the term “genre fiction” appears to pick out a perfectly respectable category,
and it is the category most commonly contrasted with literature in popular discourse.
Nonetheless, the category does not quite match up to our area of interest. Not all works
of genre fiction are popular, nor must all works of popular fiction be genre fiction. (And
even if they were co-extensive categories, they are clearly conceptually distinct. I will
have more to say about the relation between these two categories in the final section of
this essay.) Finally, although the categories of mass-market, pulp, general, and commer-
cial fiction significantly overlap with popular fiction, these are business or marketing
categories which are not of inherent philosophical interest. So “popular fiction” it is.

Popular Fiction and Literary Fiction

Most of the relevant philosophical work in this area has focused on popular art or
mass art generally (e.g. Novitz 1989; Shusterman 1991; Carroll 1998; Gould 1999;
Gracyk 2007a). But, unsurprisingly, many of the central philosophical concerns about
popular art also arise in the context of a consideration of popular fiction. For example, just
as one might wonder whether the high art/low art distinction is redundant (Cohen 1993),
one might also wonder this about the popular fiction/literary fiction distinction. Does
it map onto some already recognized distinction (e.g. non-literature/literature, bad
literature/good literature), or does it mark a novel distinction? Similarly, a number of
philosophers have been interested in whether the popular/fine art distinction is substan-
tive or merely social (Novitz 1989; Carroll 1998); the same sort of question may be asked
about the popular/literary fiction distinction as can the question of whether the distinc-
tion serves to reinforce social hierarchies. Finally, it is not at all obvious what the logical
relationships are between popular, mass, or low art on the one hand and fine or high art
on the other. Similarly, the fact that there is a distinction between popular and literary
fiction would not settle how these two categories are related. They might be mutually
exclusive categories, or there might be fictions that are – either at the same time or
different times – both popular and literary. The two categories might exhaust the domain
of fiction, or there may be fictions that belong to neither category.
So is the popular/literary fiction distinction redundant? As John Fisher writes, “there
is a natural line of thought that suggests that the distinction between high and low art
approximates the art/non-art distinction” (2005: 527). Similarly, if one focuses on para-
digmatic examples of literary fiction it is tempting to think that the opposed category is
simply non-literature. After all, many would resist characterizing Dan Brown’s oeuvre as
literary – ditto the novels of Jeffrey Archer. If so, the difference between popular fiction
and literary fiction might be thought to simply recapitulate the distinction between
non-literary and literary fiction.
Part of the difficulty in evaluating this proposal stems from the ambiguity of the term
“literature”. There are, plausibly, distinct wide and narrow notions of literature (Davies 2007;
Lamarque 2008) and on some wide conceptions of literature even Brown’s prose counts
as literary. On the other hand, it is plausible that not all non-literary fiction counts as
popular fiction. Literature is the sort of thing one can try but fail to produce


(compare Mag Uidhir 2013 on failed-art). So, many stories written by amateurs are not
literature in at least some narrow senses. But it does not follow that they are popular
fiction – they are, at least in many cases, simply fictions which aim at, but fail to achieve,
literary status. If this is right, then the popular/literary distinction is not the same as the
non-literary/literary distinction.
Does the popular fiction/literary fiction just recapitulate the bad fiction/good fiction
distinction? If what I have said about amateur efforts is right, we already have reason for
thinking this is not the case since the category provides plenty of examples of bad fiction
which are not popular fiction. Additionally, there are examples of paradigmatic popular
fiction which are widely recognized to have significant value – authors of popular fiction
such as Raymond Chandler, James M. Cain, John Le Carré, Stephen King, Agatha
Christie, Dorothy Sayers, and Patricia Highsmith have legitimate claims to have pro-
duced popular fiction which is good in some sense. (For recent relevant discussion see
Krystal 2012.) And surely there are bad works of literary fiction – Terry Eagleton (1983: 11)
mentions Lamb and Macaulay, and unsuccessful experimental or avant-garde literature
is also worth considering here. So the popular/literary distinction does not mirror the
good/bad distinction. Although this does not settle the issue, I think we have reason
to believe that the distinction in question does not merely duplicate some already
recognized distinction.
The distinction between the popular and the literary does not seem to be redundant.
Is it a substantive distinction? As mentioned above, this question is closely related to the
debate between Carroll and Novitz about the nature of popular art. To get a feel for the
issue consider the notion of popularity. In one, purely numerical, sense of popularity,
there is nothing more to being popular than being widely liked, appreciated, or approved
of. While there might be explanations of popularity in particular domains (there is, for
example, a psychological literature which explores various factors related to peer popularity
and sociometric status (see, for example, Newcomb et al. 2003)), there is a sense in which
popularity in the “being widely liked” sense does not mark out a robust category. If being
popular is just a matter of being well liked, then we do not gain much insight into some-
thing when we learn that a work of fiction is popular. We cannot, for example, assume
that all popular fictions share certain features over and above their popularity. A pursuit
of the hidden essence of numerical popularity would, I suggest, be futile.
This brief discussion of popularity suggests that this numerical conception of popularity
is not, in fact, what we are interested in when we are interested in popular fiction. Just
as an unsuccessful “hair metal” band does not fail to count as having produced popular
music simply because it doesn’t sell many records, so too a formulaic romance novel
does not fall outside of the category of popular fiction just because it fails to sell many
copies. Similarly, the impressive sales figures of The Satanic Verses do not show that it
is popular fiction in the sense in which that term is typically used (cf. Carroll 1998: 190
on why The Satanic Verses is not an example of mass art). Moreover, it does not seem
that we can even appeal to intended numerical popularity to mark out the category of
popular fiction. After all, it certainly seems possible for an author of popular fiction to
intend their work to be hated by most and only loved by a select few. And nothing
precludes the author of literary fiction from having the intention that it be widely
liked. Popular fiction is, then, something more than fiction which is popular or
intended to be popular.
This does not establish that there is a substantive category of popular fiction. Novitz
holds that the popular/high distinction can only be made sense of in terms of social

Aaron Meskin

relations and that it ultimately “helps mark class boundaries within a society” (1989: 227,
see also p. 224). If Novitz is right, then the same is surely true about the popular fiction/
literary fiction distinction. In response, Carroll (1998: 176–184) argues that the popular/
high distinction does not neatly map onto any class boundary (hence Novitz’s claim that
it functions to preserve and reinforce class divisions is implausible), and he further
argues that Novitz has not considered alternative, broadly formal, or structural, ways in
which popular or mass art might be distinguished from literary art.
What might those formal or structural features be which distinguish popular fiction
from literary fiction? On Carroll’s view, mass art is characterized as being

intentionally designed to gravitate in its structural choices (for example, its

narrative forms, symbolism, intended affect, and even its content) toward those
choices that promise accessibility with minimum effort, virtually on first contact
for the largest number of untutored (or relatively untutored) audiences.
(Carroll 1998: 196)

In the case of prose fiction, the relevant choices might be said to include erotetic
(i.e. question/answer) narrative (ibid.: 194), predictability which is rooted in membership in
a recognized genre, specific kinds of content such as “action/adventure scenarios”
(ibid.), sympathetic characters who display recognizable psychological traits, and the
inclusion of features which are designed to produce standard emotional responses
(Harold 2011). Note that none of these features appears to count as a necessary condition
for being popular fiction, and they may all also appear in literary fiction.
What, then, is the logical relationship between the categories of popular fiction and
literary fiction? Do they overlap or are they mutually exclusive? Are they exhaustive of
the category of fiction or are there other categories?
It is obvious that a literary work can become popular in the purely numerical sense.
It is less obvious whether a literary work which is originally outside the sphere of the
popular can become popular fiction in a more substantive sense. On the other hand, it
is often said that some works which are originally popular fiction become works of
literature over time. For example, Theodore Gracyk suggests that Charles Dickens’ nov-
els “move from popular to fine art status” (2007a: 385). We might make sense of such a
view by appealing to functional theories of fine art and literature whereby membership
in those categories depends on how a work functions at a particular time (Goodman
1978). An alternative approach proposes that some art works are “bilateral” (Cohen 1999:
141–142), by which is meant that they appeal to two different audiences (i.e. a high
audience and a low one) and, hence, may count as both high and low art. Adapting and
extending Cohen’s conception of bilaterality it may be useful to think of multifunctional
bilateral works of fiction which are intended to function as both popular and literary
works. Cohen largely focuses on film, but if this general line of thought is correct then
the popular fiction/literary fiction might not be exclusive – some works of fiction (e.g.
by such authors as Dickens, Austen and the Brontës may count as robustly bilateral and,
hence, be both popular fiction and literary fiction at one and the same time. Note that
if this is the case then we have further reason to think that the popular/literary distinc-
tion cannot be equated with other plausibly exclusive distinctions such as good/bad,
literary/non-literary, and genre/non-genre.
Popular fiction and literary fiction do not exhaust the domain of written fiction. I
have already mentioned amateur literary efforts, many of which plausibly fall into neither


category. And fiction, even prose fiction, is a very broad category and plausibly includes
such things as thought experiments and philosophical dialogues, many of which fall into
neither category. Perhaps more intriguingly, there are categories such as the middlebrow
and what has come in recent years to be called “upmarket fiction” to consider. (Consider
works by such well-known authors as Jonathan Franzen, Ian McEwan, and Ann
Patchett.) Sadly, the middlebrow has not been a topic of significant philosophical
interest (although see Carroll 1998: 232) and, hence, the question of whether this
category is distinct from both the popular and the literary or, rather, is just a sub-category
of one or the other has not been seriously investigated. Carroll, following Dwight
MacDonald’s account of “Midcult” (MacDonald 1960), treats middlebrow art as distinct
from mass art – roughly speaking it is a category of art that sits between mass art and
avant-garde art in terms of accessibility and “that imitates that structures of past avant-
garde art or traditional esoteric art” (Carroll 1998: 232). Jennifer Egan’s recent
unconventionally structured novel, A Visit from the Goon Squad, fits this description
nicely and might seem to be a paradigmatic work of middlebrow fiction, but it is not
clear whether works in this category count as popular, literary, both, or neither.
So what is the relationship between popular fiction and the various broad categories
that were mentioned above? Since not all fictions are works of art, it might seem that
popular fiction does not pick out a subset of mass, popular, or low art. So, for example,
some popular fiction is pornographic. Although it is controversial whether something
can be both pornography and art (Maes and Levinson 2012), it is much less controver-
sial that not all pornography is art. But surely whether or not a work of pornography is
art or not does not depend on whether it is a work of fiction. So it seems that some
pornographic popular fiction is not art and, hence, popular fiction is not a species of
popular, mass or low art. But this would be too quick. Some theorists argue that at least
some, perhaps all, works in these latter categories are not true art – they are, as
R. G. Collingwood put it, art “falsely so called” (1938: 11). If this is right, then popular
fiction might still be a subset of one or more of these categories even if it isn’t a subset
of art. On the other hand, theorists such as Carroll, who hold that works of mass art are,
by definition, art (1998: 196) might argue that our ordinary usage of the term “popular
fiction” picks out only cases of prose art. If either view is correct, then the question of
whether popular fiction is a sub-category of mass, popular or low art depends on whether
works in the former category meet the other substantive conditions for the latter cat-
egories of art.
What about the relationship between popular fiction and other categories such as
popular culture, kitsch and entertainment? Surely popular fiction is a subset of the
former category, and it seems to be one of the primary species of the latter. Its relationship
to kitsch is less obvious, in part because of the ambiguity of that term. Although it is
sometimes used as a catch-all term for all popular and commercial art and literature
(Greenberg 1939), it is often used in a much narrower sense (Kulka 1988). For this
reason, and also because it has historically had negative connotations, it would be
tendentious to treat popular fiction as a subset of this category.

Interpreting and Evaluating Popular Fiction

Does the interpretation of popular fiction differ in important respects from that of
literary fiction? Since the nature of interpretation is arguably the most controversial
issue in the philosophy of literature, this is a difficult question to answer. Nonetheless,

Aaron Meskin

at least one scholar has argued that works of popular art call for a different form of
interpretation than do works of fine or high art, and if this is correct, then it seems to
follow that the interpretation of popular fiction – or at least much of it – will differ from
that of serious literature. David Carrier claims that “the interpretation of comics (and
other genuine mass-culture art) thus differs in kind from analysis of museum paintings”
(Carrier 2000: 84). With the latter (and other example of high art), we attempt to “bring
close what is distant” (ibid.: 85) and this involves the reconstruction of artists’ inten-
tions. But since grasping the meaning of comics (and other mass-culture or “populist”
art works) is – according to Carrier – easy, our interpretive project is different. With such
works our aim is to “identify the ways in which it reflects the fantasies of the public”
(ibid.: 7). In fact, Carrier suggests that our different attitudes towards interpretation
might underwrite the difference between popular art and popular culture on the one
hand and high and museum art on the other – “the difference . . . is less a fact about how
the work is produced than a difference in our attitude toward its interpretation” (ibid.: 82).
So, is there a significant difference between how we do and/or should interpret
popular fiction and literary fiction? In response to Carrier, Gracyk has argued that the
widespread use of allusion in popular art (and by implication in popular fiction)
undercuts the putatively sharp distinction (2007b). Not only must allusions be
understood, according to Gracyk, by reference to the intentions of the artist or author
(ibid.: 71), but they also bring “interpretive problems” and “hermetic conditions” to
popular art (ibid.: 67). This does not provide a complete response to Carrier, but it
does suggest that there is no clear-cut difference between the interpretation of popular
and literary fictions.
On the other hand, consider the distinction between the description and the inter-
pretation of a work of fiction (Matthews 1977; Carroll 2009). Very roughly, description
involves an account of what is given or obvious in a work, while interpretation is only
called for with respect to things which are not obvious. If works of popular fiction
really are designed for “accessibility with minimum effort” (Carroll 1998: 196), then
perhaps there is no difference in how or what we interpret but, rather, merely a differ-
ence with respect to how much interpretation is required by popular fiction as opposed
to literary fiction.
What about the evaluation of popular versus literary fiction? Are there distinct
standards for evaluating works of popular fiction? Is literary fiction more valuable than
popular fiction?
With regard to the first question, it seems plausible that genre plays an especially
significant role in the evaluation of many works of popular fiction. For example, Carroll
argues that when it comes to many art works, the category or categories to which those
works of art belong play a significant role in their evaluation (Carroll 2009: 163–170).
Those categories often have certain purposes or objectives which are best achieved by a
work’s possessing certain properties. When a work belongs to that category and possesses
those properties, this is a (pro tanto) good-making feature. When it belongs to that
category and fails to possess those properties this may be a bad-making feature. Such an
account seems to capture a great deal about the evaluation of many popular fictions
which belong to established genres. So, for example, lacking a terrifying monster is
a bad-making feature in a horror novel, and the presence of a well-matched couple is a
good-making feature of a romance fiction.
But it would be a mistake to think that this underwrites a substantive difference
between the evaluation of popular and literary fiction. If Carroll is right, then categories


and their purposes are relevant to the evaluation of all works of art, not just popular
works. After all, works of literary fiction also belong to various genre categories
(e.g. absurdism, Bildungsroman, comedy, the picaresque, satire, tragedy, etc.). And
such categories are linked to purposes against which those works can be evaluated.
Moreover, genre-based evaluation does not look like it can be all there is to the
evaluation of popular or genre fiction. For example, James Harold has argued that many
works of popular fiction are properly evaluated along standards that are typically
associated with works of great literature; namely, with regard to the importance of the
themes they address and how well they handle various literary techniques (Harold 2011).
Even Carroll admits that there are aspects of evaluation which go beyond category-
relative standards (2009: 191–196).
One might attempt to distinguish the evaluation of popular fiction and literary
fiction by appeal to the broad species of value that we look for in each. It is commonly
said that our interest in literature as literature is a concern (at least in large part) for its
intrinsic merits (Lamarque 2009: 264–267), while our interest in popular literature is in
its instrumental values. Popular fiction is, we are told, often valued as a means of relax-
ation or of distracting us from work (ibid.: 262), whereas we value works of literature for
their own sake. Perhaps the fact that works of popular fiction often seem substitutable
for one another (one mystery is just as good as another in many contexts) whereas liter-
ary fictions do not typically seem substitutable lends some support to this distinction.
But we ought to be cautious about this approach, since it is controversial whether the
value of literary fictions is, in fact, primarily intrinsic rather than instrumental. Also,
some works of popular fiction are plausibly properly valued as achievements, and the
value of achievements is not solely instrumental. And it would be a mistake to think
that works of popular fiction are always substitutable for one another. In some contexts,
we are interested in a particular fantasy or detective novel and no other will do.
Is literary fiction better than popular fiction? As suggested by the discussion above,
it is clearly not the case that every example of the former is better in some sense than
every example of the latter. And, as suggested above, much of our evaluation of works
of fiction is category-relative. This might seem to suggest that comparing the value
of popular fiction to the value of literary fiction is impossible or incoherent. But this
would be too quick. In the first place, we have already seen that works in both
categories are subject to evaluation with respect to some of the same criteria. In the
second place, works of fiction belong to multiple relevant categories so popular works
may share categories with literary ones. Moreover, we certainly do make comparisons
between works in the two categories. Unless we are to count our practices as irrational,
we need to make sense of this. One way of doing this would be to appeal to something
like J. S. Mill’s distinction between higher and lower pleasures. On such a view, seri-
ous literature provides more valuable or more long-lasting pleasures than does
popular fiction (Lamarque 2009: 263–264, but see Harold 2011 for skepticism). Since
there is reason to doubt that the value of literature and popular fiction is purely
rooted in pleasure, such a view has some work to do if it is to underwrite the relative
ranking of the two categories. Another strategy might be to appeal to which sorts of
works are “more important to the life of society” (Carroll 2009: 192), but we need to
know much more about what cultural significance amounts to. After all, in one very
straightforward sense of “cultural significance”, many works of popular fiction, in vir-
tue of their capacity to capture the public imagination, possess a very high degree of
cultural significance.

Aaron Meskin

The Paradox of Junk Fiction?

Is there a distinctive paradox raised by popular fiction? Carroll presents, and proposes a
solution to, what he calls “the paradox of junk fiction” in an essay of that same name
(Carroll 2001). The puzzle arises if we accept four independently plausible propositions:

1. Ordinary readers typically consume works of popular fiction for the sake of the
stories in it.
2. Ordinary readers typically know, before reading a given work of popular fiction, how
the story will turn out.
3. If one already knows how a story will turn out in a given piece of fiction, it is
irrational to read that fiction for the sake of the story.
4. Ordinary readers are not typically irrational in their consumption of fiction.

In defense of (1), Carroll points out that we are not typically interested in the aes-
thetic features of works of popular literature. Moreover, we don’t typically like being
told the story before we read a work of popular fiction. And we typically consume
popular fictions quickly – we do not linger or puzzle over them as we often do with
serious poetry or works of modernist literature. As Carroll puts it, “what motivates
turning the page so quickly [in a work of popular fiction] is our interest in what hap-
pens next” (ibid.: 335). When it comes to popular fiction, then, it seems as if it is
story that matters. In defense of (2), it is worth pointing out that most (perhaps all,
but see below) popular fictions belong to recognized genres. And most readers of
popular fiction, at least of the time, read works in genres they know something about.
In such cases, it seems plausible to say that readers typically know, at least in broad
outline, how the story will unfold. As Carroll puts it, “in some very general sense, the
audience already knows the story in question” (ibid.: 336). Point (3) seems to follow
pretty straightforwardly from (1) and (2). After all, knowledge of what is going to
happen in a book seems to preclude a rational person being motivated by concern for
its story. With regard to (4), although some have argued that ordinary readers are
irrational in their engagement with popular fiction, Carroll argues that such a view
is implausible.
Rather than characterizing ordinary readers as irrational or denying that they read
popular fiction for the sake of stories, Carroll proposes to solve the putative paradox by
rejecting (3). Knowledge of how the story will turn out does not preclude reading for the
sake of the story. This is because engagement with a story is a significant source of what
Carroll calls “transactional value” – the value of exercising “our cognitive powers, our
powers of interpretation and inference, our powers of moral judgment and emotive
assessment” (ibid.: 344).
It is plausible that this is part of the explanation for our interest in popular fiction.
But perhaps we can provide an alternative solution to the paradox by pointing to an
ambiguity in Carroll’s articulation of it. Although it is true that ordinary readers con-
sume popular fictions for the sake of their stories, this is best understood as an interest
in individual stories; that is, story tokens; that is, an interest in how individual stories
will turn out. On the other hand, ordinary knowledge of stories is a matter of knowing
story types; that is, general knowledge about kinds of stories. But general knowledge
about kinds of stories (e.g. knowledge that the mystery will be solved or that the couple
will be happily united) does not entail detailed knowledge of individual stories (e.g. who


committed the crime, how the couple will be reunited). If this is right, then the argument
is defused since (3) is plausible only if “story” is understood unambiguously.

Genre and Popular Fiction

The vast majority of works of popular fiction fall into recognizable genres such as
romance, science fiction, horror, mystery, fantasy, or the western, but it is not clear
that they must do so unless one assumes that all fiction falls into such a genre. Since
fiction itself may count as a genre (see Friend 2012), and popular fiction itself has a
claim to count as one too, every work of popular fiction may trivially fall into some
genre. The question is whether they must also fall into some genre non-trivially. (And,
at least for the sake of argument, we shall exclude literary forms such as the novel and
short story from counting as genres.) I do not believe so, though it is difficult to offer
cases since there is a strong tendency to place works of popular fiction in some genre
or other. (Perhaps Alice Sebold’s 2002 novel, The Lovely Bones, is worth considering
here – it is popular fiction but does not seem to fall straightforwardly into any genre
although it contains elements of several.) But even if it were the case that all extant
popular fiction is genre fiction, this is not a matter of necessity. Writing in recognized
genres is merely a standard means by which authors of popular fictions aim for accessibility
and a wide readership.
More straightforwardly, not all works of genre fiction fall into the category of the
popular. Italo Calvino’s short stories from Cosmicomics are often categorized as science
fiction (and, hence, genre fiction), but these tales do not seem to be members of the
category of popular fiction. The same is true of some of J.G. Ballard’s works as well as
some of Margaret Atwood’s novels. George Saunder’s story, “Sea Oak”, is a ghost story
but not popular fiction.
Recently, some philosophers have focused on specific issues raised by popular genres
of fiction. (Note that these genres are typically transmedia.) Carroll (1990) has explored
a range of issues related to the horror genre. Brian Laetz and Joshua Johnston (2008)
have proposed a definition of fantasy. And Alan Goldman (2011) has presented a robust
defence of the appeal of mystery fiction in terms of his theory of aesthetic value. (For a
classic and entertaining criticism of mystery fiction see Wilson 1945.)

Works Cited
Carrier, D. (2000) The Aesthetics of Comics. University Park, PA: Pennsylvania State University Press.
Carroll, N. (1990) The Philosophy of Horror: Or, Paradoxes of the Heart. New York: Routledge.
Carroll, N. (1994) “The Paradox of Junk Fiction,” Philosophy and Literature 18: 225–241.
Carroll, N. (1998) A Philosophy of Mass Art. Oxford: Oxford University Press.
Carroll, N. (2009) On Criticism. New York and Oxford: Routledge.
Cohen, T. (1993) “High and Low Thinking about Art.” The Journal of Aesthetics and Art Criticism 51: 2.
Cohen, T. (1999) “High and Low Art, and High and Low Audiences,” Journal of Aesthetics and Art Criticism
57: 137–143.
Collingwood, R.G. (1938) Principles of Art. Oxford: Clarendon Press.
Davies, D. (2007) Aesthetics and Literature. London and New York: Continuum.
Eagleton, T. (1983) Literary Theory: An Introduction. Minneapolis, MN: University of Minnesota Press.
Fisher, J.A. (2005) “High Art Versus Low Art,” in B. Gaut and D.M. Lopes (eds) Routledge Companion to
Aesthetics, 2nd edition, New York: Routledge.
Friend, S. (2012) “Fiction as a Genre,” Proceedings of the Aristotelian Society 112: 179–209.
Goldman, A. (2011) “The Appeal of the Mystery,” Journal of Aesthetics and Art Criticism 69: 261–272.

Aaron Meskin

Goodman, N. (1978) “When is Art?”, in Ways of Worldmaking, Indianapolis, IN: Hackett Publishing.
Gould, T. (1999) “Pursuing the Popular,” Journal of Aesthetics and Art Criticism 57: 119–135.
Gracyk, T. (2007a) “Searching for the ‘Popular’ and the ‘Art’ in Popular Art,” Philosophy Compass 2: 380–395.
Gracyk, T. (2007b) “Allusion and Intention in Popular Art,” in W. Irwin and J.J.E. Gracia (eds) Philosophy
and the Interpretation of Popular Culture, Lanham, MD: Rowman and Littlefield Publishers, Inc.
Greenberg, C. (1939) “Avant-Garde and Kitsch,” Partisan Review 6: 34–49.
Harold, J. (2011) “Literature, Genre Fiction, and Standards of Criticism,” nonsite.org Issue # 4 (October 14,
2011). Available at: http://nonsite.org/issues/issue-3/literature-genre-fiction-and-standards-of-criticism
(accessed 28 August 2015).
Krystal, Arthur (2012) “Easy Writers,” The New Yorker 28 May 2012: 81–84.
Kulka, T. (1988) “Kitsch,” British Journal of Aesthetics 28: 18–27.
Laetz, B. and Johnston, J.J. (2008) “What is Fantasy?”, Philosophy and Literature 32: 161–172.
Lamaraque, P. (2009) The Philosophy of Literature, Malden, MA: Blackwell.
MacDonald, D. (1960) “Masscult and Midcult,” The Partisan Review 27: 203–233.
Maes, H. and Levinson, J. (2012) Art and Pornography: Philosophical Essays, Oxford: Oxford University Press.
Mag Uidhir, C. (2013) Art and Art-Attempts, Oxford: Oxford University Press.
Newcomb, A.F, Bukowski, W.M., and Pattee, L. (1993) “Children’s Peer Relations: A Meta-Analytic Review
of Popular, Rejected, Neglected, Controversial, and Average Sociometric Status,” Psychological Bulletin
113: 99–128.
Novitz, D. (1989) “Ways of Artmaking: The High and the Popular in Art,” British Journal of Aesthetics 29:
Roberts, T.J. (1990) An Aesthetics of Junk Fiction, Athens, GA: Georgia University Press.
Shusterman, R. (1991) “Form and Funk: The Aesthetic Challenge of Popular Art,” Journal of Aesthetics and
Art Criticism 31: 203–213.
Wilson, E. (1945) “Who Cares Who Killed Roger Ackroyd? A Second Report on Detective Fiction,” The
New Yorker 20 January 1945: 59–66.

Ted Nannicelli

I Introduction
Are screenplays really literature? And do they actually raise any interesting philosophical
questions? My aim in this chapter is twofold. First, I want to make a case that many,
if not all, screenplays are plausibly literature. Second, I want to show that debates
about the screenplay’s literary status have opened up a host of substantive philo-
sophical questions—specifically questions about the ontology of the screenplay:
What is the screenplay’s relationship to the finished screen work? Is it a constituent
part of that work or is it an autonomous work? Is a screenplay ever complete and,
if so, under what conditions? Is there ever a definitive version of a screenplay and, if
so, which?
Although questions about the screenplay’s literary and ontological status have
largely gone unremarked upon by contemporary philosophers and theorists, they were
identified and debated by classical film theorists such as Béla Balázs, Osip Brik, Sergei
Eisenstein, Hugo Münsterberg, and Dziga Vertov (Maras 2009: 33–36; 44–47). The
divergence in the answers these theorists offered suggests how knotty the questions
are. Brik averred, “the script is not an independent literary work [. . .] The script is
written in words. But this in no way makes the script a literary work, let alone an
autonomous one” (1974: 96). Similarly, Münsterberg wrote that in contrast to a
drama, which “is completed as a work of literature even if it never reaches the stage,”
in the case of cinema, “the work which the scenario writer creates is itself still entirely
imperfect and becomes a complete work of art only through the action of the
producer” (1916: 193). Balázs drew a similar contrast between the theatrical script and
film script, writing: “The film on the contrary [to a theatrical performance] mostly
absorbs the script completely so that it is not preserved as an independent object
which could be used again for a different film production” (1952: 247). Yet he concluded
that the film script “is just as much a specific, independent literary form as the written
stage play” (Balázs 1952: 246).
Then, as now, intuitions differ as to whether the screenplay is literature. But for both
classical film theorists and the handful of contemporary theorists who have resumed the
debate, the ontology of the screenplay seems somehow relevant (Maras 2009; Price 2010).
Evident in the remarks by both Brik and Balázs is the idea that the screenplay is not
an autonomous work, separable from the film with which it is associated, and apparent
in Münsterberg’s commentary is the suggestion that the screenplay is essentially
incomplete. For Brik and Münsterberg, if not Balázs, these putative ontological facts
Ted Nannicelli

about the screenplay preclude it from being literature. As we shall see, contemporary
philosophers and theorists have similarly appealed to the screenplay’s ontology to
explain why it is not literature (or why it is often not regarded as such) (Carroll 2008:
68–69; Maras 2009: 44–62; Price 2010: 24–62). In what follows, my discussion divides
this recent work into two categories. Claims that the ontological status of the screenplay
impinges upon its claim to literary status are supported by the “ingredient hypothesis”
or by the “incompleteness hypothesis.”

II Is the Screenplay an Autonomous Work? The Ingredient

A number of contemporary theorists have reiterated Brik and Balázs’s claims that the
screenplay is not an autonomous literary work, not an “independent object.” Echoing
Balázs’s remarks on the relationship between screenplay and film, Barbara Korte and
Ralf Schneider suggest, “a screenplay is ‘absorbed’ into one film only,” that it is “entirely
‘burnt up’ in the production process” (2000: 93; 90). Korte and Schneider also follow
Balázs in holding that the screenplay may be literature, notwithstanding this putative
fact about its ontology.
However, other contemporary scholars do take the screenplay’s supposed lack of
autonomy to problematize its claim to literary status. Steven Maras, for instance,
contends that there is “a significant issue when considering the script as literature [. . .]
which is that the existence of the script is bound to the cinematic medium, and also to
the production of films” (2009: 48). For Maras, “One general problem with the attempt
to see the screenplay as a form of literature is that it tends to take the script out of its
production context, restrict [its] intermediality and treat it as an autonomous work of
art” (ibid.). On his view, though, “the intermediality of the script complicates the extent
to which the screenplay can be considered an autonomous form” (ibid.).
The tendency in contemporary theorizing about the screenplay, evident in the work
discussed above, is to gloss over some rather thorny ontological matters. Korte and
Schneider describe the relationship between screenplay and film only in metaphorical
terms, suggesting that the screenplay somehow mysteriously disappears in production,
and Maras simply asserts that the screenplay is existentially “bound” to the film medium
without explaining what he means or offering a supporting argument. While the general
suggestion that the screenplay is not an autonomous work has some undeniable intuitive
appeal, neither Korte and Schneider nor Maras’s position is particularly compelling
because they do not offer a substantive account of the relationship between screenplay
and film that would support their claims.
However, Noël Carroll has advanced a similar argument that is bolstered by such an
account. The sole philosopher of art to discuss the screenplay—and even in this case,
only in the context of a discussion of the ontology of cinema—Carroll argues that a
screenplay is not an autonomous work, but rather a “constituent part” or an “ingredient”
of a particular movie (Carroll 2008: 68–69). In his words, screenplays “are ontologically
ingredients in the motion pictures with which they are associated rather than being
independent artworks” (ibid.: 69). Call this the “ingredient hypothesis.”
The ingredient hypothesis underpins Carroll’s objection to the idea that screenplays
can be (literary) art. Why should we think it is correct?1 Carroll argues for it in two dif-
ferent ways. First, he suggests that at least many screenplays are transcriptions of films
(ibid.). Such transcriptions are unlikely candidates for artworks, not only because they


do not seem to be created with the right sort of intentions, but also because they are
ontologically dependent upon finished films. Without a finished film, there can be no
transcription of it; hence, transcriptions are not autonomous works. This argument
seems right, but misses its intended mark because the first premise—that screenplays are
transcriptions—is erroneous. As we ordinarily conceive of them in our creative and
appreciative practices screenplays are not transcriptions of films, but rather are instruc-
tions for making films. So, accepting Carroll’s argument about transcriptions does not
threaten the idea that screenplays may be autonomous works, for it turns out that the
argument does not actually involve screenplays, properly so-called, at all.
The second way Carroll argues for the ingredient hypothesis is by contrasting the
relationship between screenplay and film with the relationship between theatrical script
and theatrical performance. In the case of theater, he writes, “The play by the playwright
is one artwork which is then interpreted like a recipe or set of instructions by a director
and others in the process of producing another artwork—the performance of the play”
(ibid.: 67). For Carroll, this makes theater and the other performing arts “two-tiered
artforms” (ibid.). As he puts it:

In music [. . .] there is the score—the artwork created by the composer; and
then there is also the performance [. . .] Similarly, there is the choreography
of the dance which is one artwork; and then there is the choreography as
performed by the troupe which constitutes a discrete artwork.
(ibid.: 67–68)

On Carroll’s view, what seems to make “recipes” like theatrical scripts artworks in their
own right is the fact that they are created to be interpreted on multiple occasions in
performance. For elsewhere he argues that theatrical scripts are art because they are
products of “the art of composing performance plans,” and not only or necessarily because
they are works of literature (Carroll 2006: 106).
In contrast, Carroll argues, screenplays are not “recipes” that are multiply interpreted:

If, in theater, the play-text type is a recipe or sketch that the director and/or the
actors interpret, and, furthermore, if the recipe and the interpretation can be
regarded as different though related artworks, in motion pictures the recipe and
the interpretations are constituents of the self-same integral artwork.
(Carroll 2008: 68)

For Carroll, the crucial difference between theatrical script and screenplay is that the
latter is not what he calls a “performance plan”; it is not intended to be interpreted in
performance on multiple occasions. This is an important difference between script and
screenplay, but, pace Carroll, it supports neither the ingredient hypothesis nor the
conclusion that screenplays are not literary art.
It is true that most screenplays are involved in the production of a single film. But
this fact entails no conclusions about the screenplay’s ontology, about the kind of thing
it is by nature. On the one hand, many screenplays are never produced. So, the claim
that “screenplays are ontologically ingredients in the motion pictures with which they
are associated” is too general and needs some immediate refinement. On the other hand,
screenplays can be (and occasionally are, as in the case of The Prisoner of Zenda
(1937/1952)) used as “recipes” to make multiple films. Therefore, even when a screenplay

Ted Nannicelli

is involved in the production of a particular movie, it is not ontologically inseparable

from it; it is not the case that “the recipe and its interpretation are presented together
in one indissoluble package.” Even if a screenplay is written for and tailored to the
specifics of a particular film production (as some theatrical scripts surely also are), it
may still be interpreted on a future occasion. So, we have reasons not to accept the
ingredient hypothesis.
The salient difference between theatrical script and screenplay is neither that the
former is a “recipe” and the other is an ingredient, nor that the former affords multiple
interpretations and the latter does not. Rather, it is this: Theatrical scripts are “work-
determinative” or “work-specifying” instructions (Davies 2001; Davies 2009). In virtue
of completing a theatrical script, the playwright brings into existence a work for perfor-
mance—a work that is distinct from the script itself and distinct from individual
performances of it, yet which requires an interpretation of the script and a performance
for its instantiation. Moreover, correctly following the instructions of the script to exe-
cute a performance will always and necessarily instantiate the same work, which the
script specifies. In contrast, screenplays are not work-specifying because no cinematic
work is created when a screenplay is completed (Nannicelli 2011b; Nannicelli 2013).
And following the instructions of the screenplay on multiple occasions will result in the
creation of multiple films rather than multiple instances of a single film. Inasmuch as
Carroll’s characterization of the theatrical script as a performance plan is able to capture
these differences between script and screenplay, it is unproblematic.
However, the theatrical script’s status as work-determinative performance plan cannot
be what secures its status as literary art. It is not sufficient to make the theatrical script
literature because other work-determinative performance plans are not literature. But,
more generally, being a work-determinative performance plan is neither necessary nor
sufficient for a set of art-instructions to be an artwork in its own right. It is not necessary
because certain architectural plans, which are not work-determinative, are plausibly inde-
pendent works of art. And it is not sufficient because some work-determinative
performance plans, like standard musical scores, are not plausibly autonomous artworks.
The important point for the present purpose is twofold: First, the fact that the theat-
rical script is a work-determinative performance plan is not what makes it art, let alone
literature. Second, the fact that screenplays are not work-determinative performance
plans thus in no way prevents them from being works of literary art. The first observa-
tion indicates that some alternative explanation for why theatrical scripts are literature
is needed. However, without committing to a particular account, we can suppose it is
plausible that what enables theatrical scripts to court literary status is the fact that they
are verbal artifacts. But then it is also plausible that this same feature of screenplays
makes them at least candidate works of literature as well. Obviously, this does not show
that screenplays are works of literature, but it does show that their ontological differ-
ences from theatrical scripts—which do not include being somehow ontologically bound
to films in the manner of constituent parts of “ingredients”—do not preclude them from
being literature.

III Is the Screenplay a Complete Work? The Incompleteness

Like the ingredient hypothesis, the incompleteness hypothesis makes a claim about the
ontology of the screenplay that may or may not be harnessed in support of an argument


against the screenplay as literature. Nathanial Kohn, for example, takes the putative
incompleteness of the screenplay to indicate that it is a “postmodern literary exemplar”
(2000: 489). Other theorists, like Steven Price, hold that recognizing that the screenplay
is incomplete in the manner of a Barthesian “text,” “helps to explain its exclusion
from the canons of literature” (Price 2010: 41). In what follows, I shall be concerned
primarily but not exclusively with versions of the incompleteness hypothesis that
supposedly problematize the screenplay’s literary status. In addition, because the
hypothesis has various formulations, my discussion will look at how it is advanced by
several different critics.
If, as the discussion in the preceding section argues, it is plausible that cinematic work
and screenplay are ontologically distinct, then it is important to distinguish between the
completion of a film and the completion of a screenplay itself. We can acknowledge that
no film is specified or completed in virtue of the writing of the screenplay, but this
should not cause us to doubt that the screenplay itself may be complete. However, not
all theorists have been sensitive to this distinction. As a result, their claims about the
screenplay’s putative incompleteness tend to be equivocal at best and deeply implausible
at worst.
Münsterberg, for example, writes ambiguously, “the work which the scenario writer
creates is itself still entirely imperfect and becomes a complete work of art only through
the action of the producer” (1916: 193). This statement is true if “the work” refers to
the cinematic work, for, as we saw, screenplays are not constitutive of cinematic works
in the same way theatrical scripts are constitutive of theatrical works. However, if “the
work” refers to the scenario or screenplay itself, though, there is no reason to accept the
claim. Why shouldn’t it be the case that the screenplay itself is complete in many, albeit
perhaps not all, instances?
A leading contemporary theorist of screenwriting, Ian W. Macdonald, is somewhat
more careful about maintaining the distinction between film-completion and screen-
play-completion, but nevertheless suggests that the screenplay itself is somehow
incomplete in virtue of the fact that it is not constitutive of a film. As he puts it, “There
are some things the screenplay is not: it is not a finished piece of work (in relation to
the screenwork—the finished film) [. . .] It is not (ever) complete, as a description of all
the aspects of the screenwork” (Macdonald 2004: 90). Macdonald’s central points are
right: The screenplay represents work that is unfinished if it is regarded qua part of a
larger project—i.e. an effort to make a “screenwork.” Relatedly, the screenplay necessarily
does not describe all the properties of the screenwork made from it. However, there is
something odd about casting these points in terms of the screenplay itself being
incomplete—despite Macdonald’s inclusion of qualifying clauses.
If we regard Macdonald’s points here as constituting a weak version of the incom-
pleteness hypothesis, then the hypothesis is innocuous except perhaps insofar as its
rhetorical flourish may lead one to believe that the claims it makes are more radical
than they actually are. However, some critics who say that the screenplay is incomplete
do have something more extreme in mind. Macdonald himself shifts from making
these rather modest, plausible claims regarding the screenplay’s completeness vis-à-vis
the finished film to making much bolder claims regarding the putative incompleteness
of the screenplay and Barthesian “texts” more generally. Indeed, one might suspect
that the rhetoric in which the truly modest claims are couched functions to imply that
those claims support the bolder thesis that the screenplay itself is somehow unfinished
or incomplete.

Ted Nannicelli

Macdonald invokes the work of Roland Barthes to suggest that the screenplay—not
only qua step towards a finished film but qua “text”—is essentially incomplete. He
endorses Nathanial Kohn’s claim, “in Barthes’s (1974) terms, screenplays are model
‘writerly texts’—open to being rewritten—as opposed to closed ‘readerly texts’ which
‘can be read but not written” (quoted in Macdonald 2004: 91). In his essay, Kohn
elaborates: “Screenplays are always works in progress—invitational, unfinished,
tempting. . . . A writerly text demands a productive collusion on our part” (2000: 495).
The idea Macdonald wants to get at here is that, as writerly text, the screenplay is being
endlessly rewritten. This, then, is a stronger formulation of the incompleteness
hypothesis—a poststructuralist-influenced claim that the screenplay is a kind of writerly-
text. However, for Macdonald, as for Kohn, this putative ontological fact about the
screenplay does not impinge upon its ability to be literature.
In his recent book, The Screenplay: Authorship, Theory and Criticism, Steven Price
(2010) also draws upon Barthes’s work to posit the poststructuralist version of the
incompleteness hypothesis. Furthermore, he suggests that the screenplay’s incompleteness
explains why it is not typically regarded as literature. After outlining and considering in
relationship to the screenplay seven “propositions” Barthes makes in his essay “From
Work to Text,” Price concludes:

Although [the screenplay] is clearly to be differentiated from the Barthesian

text, it is still in many respects the contemporary text par excellence, and at the
very least this essay can take us further in distinguishing the screenplay from
literature (or “work”). It clearly functions within “the activity of production”
rather than as a closed work; it is not concerned with validating itself as
“literature”; its meaning is deferred, since it is a text of suggestive incompletion
that demands the writerly activity of others; it is markedly intertextual,
participating within the general fields of cinema and literature, but – other than
in adaptation or historical drama – rarely concerning itself with “source” mate-
rials; it is indisputably severed from the legal conceptions of authorship outlined
by both Barthes and Foucault; and the reader of the screenplay, at least in its
industrial context, directly participates in the activity of production and, meta-
phorically and very often literally, in the “re-writing” of the text.
(2010: 41)

For Price, Barthes’s notion of the writerly text not only sheds light upon the ontology of
the screenplay, but also upon the reason it is not literature. In his words, “conceiving of
the screenplay in the ways prompted by Barthes’s essay helps to explain its exclusion
from the canons of literature” (2010: 41).
Of the poststructuralist formulation of the incompleteness hypothesis, we can ask two
questions: First, independent of the question about literary status, does it offer a plau-
sible view of the screenplay’s ontology? Second, does it offer a coherent and compelling
account of why the screenplay is often not accepted as literature or why it cannot be
literature? There are reasons to think the answer to both questions is “no.”
As a claim only about the ontology of the screenplay, the hypothesis encounters
several difficulties. First, as suggested above, its proponents tend to equivocate between
claims about film-completion and screenplay-completion. To say that the screenplay
“demands the writerly activity of others” (Price 2010: 41) is ambiguous. For it is the
completion of the film that requires further activity—not necessarily the screenplay


itself. It is, for example, very easy to imagine an independent filmmaker using a first-
draft screenplay in production. It may be the case that a given screenplay requires
revision by other writers, but this is clearly not an essential feature of all screenplays.
Price indicates that his claims are limited to the screenplay’s “industrial contexts,” in
which “the reader of the screenplay [. . .] directly participates [. . .] very often literally,
in the ‘re-writing’ of the text” (2010: 41).2 On the face of it, this statement is true, but
it implies that the re-writing of the screenplay happens through the process of reading
since it is a reader who does the re-writing. But although a reader of the screenplay may
re-write it, she does so by actually re-writing it—not reading it! Perhaps needless to say,
this does not mean that screenplays are writerly texts; it just means that screenplays, like
many other texts, are often co-authored and go through various drafts.
Second, the hypothesis assumes the validity of what is actually a highly contentious
and a deeply implausible claim—roughly, that some texts, which seem to have been
completed by their authors, in fact invite and require the reader to (re)write them and
are actually never complete. That this suggestion renders our ordinary creative and
appreciative practices utterly incoherent appears to be of little concern to the Barthesian
theorist, despite the fact that the point of theorizing those practices would seem to be
to explain them. Unattractive on its own merits, this claim becomes even less compelling
in light of the fact that Barthes himself offers no argumentation for his claims: “these
are speech-acts, not arguments, ‘hints,’ approaches which agree to remain metaphorical”
(Barthes 1989: 57).
Third, because the putative incompleteness of the screenplay owes to its status as
writerly text, the hypothesis is, at best, uninformative. On this account, incompleteness
is not a unique quality of the screenplay, but a feature the screenplay has in virtue of
being a writerly text. But then if all writerly texts are incomplete, the hypothesis tells us
nothing about the ontology of the screenplay in particular.
This third criticism of the incompleteness hypothesis’s view of the ontology of the
screenplay also indicates that the hypothesis does not fare well as an explanation of (or
as evidence to support) the screenplay’s exclusion from the realm of literature. Even if
we accepted, for the sake of argument, that screenplay is incomplete because it is a
writerly text, this fact cannot impinge upon its literary status because at least some so-
called writerly texts are uncontroversially works of literature. (For instance, Barthes
himself suggests that Mallarmé—whose poetry is indisputably literature—“wanted the
audience to produce the book” (Barthes 1989: 63)). Rather than offering an explanation
or argument for why the screenplay is putatively not literature, the incompleteness
hypothesis presents a paradox: although screenplays are supposedly not considered
literature because they are incomplete writerly texts, it is evident that other putatively
incomplete writerly texts are widely regarded as paragons of the art of literature.
None of this is to deny that many screenplays are extensively re-written (often by
multiple authors) and that some screenplays are abandoned before they have been
completed. But from these facts it hardly follows that the screenplay is essentially incom-
plete—unfinished by its very nature. True, the industrial conditions of cinema and
television production may raise questions about how we know when a screenplay is
complete and how we know which draft constitutes the finished (or definitive) version.
However, such questions involve epistemological—not ontological—matters. Moreover,
these questions are hardly exclusive to the screenplay. We may have similar queries
when studying any sort of literary work that has been substantially revised (sometimes
by different writers) on numerous occasions (e.g. Walt Whitman’s Leaves of Grass or

Ted Nannicelli

Shakespeare’s plays). And it is evident that both in these cases and even in cases when
a work seems to genuinely lack a definitive version or to have been abandoned prior to
completion, the work’s literary status is not necessarily in doubt (consider, for example,
Franz Kafka’s The Castle). So, we ought not accept the incompleteness hypothesis either
as an account of the screenplay’s ontology or as an explanation or argument regarding
its literary status.

IV Screenplays and the Definition of Literature

Thus far, my discussion has attempted to defend against two prominent objections to
the notion that screenplays can be literature. But is it possible to show that (at least
some) screenplays are literature? On the face of it, this may seem to be an insurmount-
able task which, as Price puts it, “tends to fall victim to the twin problems of defining
‘literature’ and constructing evaluative criteria that determine whether a given text, or
even a whole genre, merits entry to the canon” (2010: 27). In other words, one might
think the argument for screenplays as literature is doomed to fail because there is no
uncontested definition of literature available. However, this objection errs in assuming
that the argument requires us to assent to a single, particular definition of literature.
There are several alternative strategies open to the proponent of the screenplay as
literature, of which I shall gloss merely one.
Roughly, the thought is that rather than endorsing (or developing) a particular
definition of literature that will include screenplays, the proponent of the screenplay as
literature can proceed by showing that on any number of a variety of extant definitions
at least some screenplays (and perhaps many) will count as literature (Nannicelli 2013).
That is, we can remain agnostic about literature’s definition and simply point out that
on any definition of literature one finds plausible or compelling, at least some screenplays
will count as literary works.
Consider, for example, the perhaps intuitive and relatively prevalent objection that
screenplays are not literature because they are written in extremely prosaic language.
In Macdonald’s words, the screenplay “does not appear literary in a traditional sense,
except possibly in parts” (2004: 90). What Macdonald has in mind as the “traditional
sense” of the literary seems to be what we might follow Robert Stecker in calling a
“linguistic definition” of literature (2003: 66). On linguistic definitions, literature is
defined by particular uses of language—for example, “poetic” language in a broad
sense or in a more specific sense as, say, the Russian Formalists suggested (i.e. language
that “defamiliarizes”).
If we accept such a linguistic definition, just for the sake of argument, numerous
screenplays will unquestionably count as literature. Although it is true that many
Hollywood (and Hollywood-style) screenplays are written in workaday prose, there are
numerous instances of screenplays that employ “literary” or “poetic” language, whether
this notion is cashed out in terms of rhythm, metaphor, imagery, defamiliari­zation, or
styles of narration like free indirect discourse (Nannicelli 2010; Nannicelli 2013).
Consider, for example, Carl Mayer’s script for Sunrise: A Song of Two Humans (1927),
Walter Hill’s screenplays for Hard Times (1975) or The Driver (1978), Paddy Chayefsky’s
screenplay for Altered States (1980), or Terrence Malick’s screenplay for The Tree of Life
(2011). And this is to say nothing of screenplays written outside of Hollywood that
plausibly meet the conditions of various linguistic definitions of literature: for example,
Su Friedrich’s screenplay for But No One (1982), Marguerite Duras’s screenplay for


Hiroshima Mon Amour (1959), and Dylan Thomas’s unproduced “The Doctor and the
Devils.” However, we should also note that linguistic definitions of literature seem fun-
damentally flawed because they simultaneously exclude far too many non-controversial
instances of literature and bestow literary status upon non-literary works; that is, par-
ticular, “literary” uses of language are neither necessary nor sufficient for something to
be literature (Lamarque 2009: 47).
Of the various extant definitions of literature, perhaps the only one that may give
screenplays trouble is an institutional definition. According to Peter Lamarque’s recent
account, “the existence of literary works depends on a set of conventions concerning
how they are created, appreciated, and evaluated; in other words, on attitudes, expecta-
tions, and responses found in authors and readers” (2009: 62). But is it really plausible
that any screenplay is written with the intention that it is “produced and meant to be
read within a framework of conventions defining [literary] practice”? On the contrary,
many screenplays seem intended to be read not within the framework of the conven-
tions that define literature as an institution, but within the framework of conventions
specific to film (pre-) production. In other words, is it not the case that Hollywood
screenplays are intended to be read only within the “institution” of the Hollywood system
rather than within the institution of literature?
Surely such screenplays also need to be read within a micro-framework of Hollywood-
specific conventions (for example, “INT.” stands for “interior), but these two frameworks
do not seem mutually exclusive. Rather, it is plausible that for various film production
crew and cast members (the director, cinematographer, gaffers, sound designer, actors) to
successfully carry out their jobs, they must interpret (broadly speaking) the screenplay.
More specifically, they must attends to things like how its language creates a sense of
tone, how its structure suggests a purpose or point, how its narration focalizes the story
and, perhaps, encourages the adoption of a particular point of view. In short, what is
needed in the contexts of Hollywood studio production is to read the screenplay within
the same framework of conventions that governs the institution of literature as a whole.
On the other hand, some screenplays are written and read entirely outside Hollywood
(or any other institution of film production) and within the framework of the conven-
tions of literary practice (understood most narrowly). The best example of this is
web-based, fan-fiction screenwriting. Roughly, this practice involves fans of various
movies or television shows writing their own original teleplays or screenplays with the
specific intention of sharing them with other fans that read and respond to them. The
thought here is that these fans constitute communities of writers and readers in the most
conventional sense. The screenplays in these communities are created, appreciated, and
evaluated within the same framework of conventions as a paradigmatic community of
poets writes and reads the work of its members. Therefore, at least some screenplays will
still count as literature even if one accepts an institutional definition that construes the
conventions of literary practice in the narrowest of terms.
As indicated in the Introduction, this chapter merely attempts clear enough ground
for philosophers and literary theorists to start taking the screenplay seriously as literature.
Needless to say, there is much more work to be done.

I wish to thank Routledge for allowing me to reproduce material that was previously
published in my book, A Philosophy of the Screenplay.

Ted Nannicelli

1 For a more detailed discussion of Carroll’s position, see Nannicelli (2011a) or Nannicelli (2013).
2 One wonders why Price puts “re-writing” in scare quotes if readers literally re-write screenplays.

Works Cited
Balázs, B. (1952) Theory of the Film: Character and Growth of a New Art, translated by Edith Bone, London:
Denis Dobson.
Barthes, R. (1989) “From Work to Text,” in The Rustle of Language, translated by Richard Howard, 56–64,
Berkeley, CA: University of California Press.
Brik, O. (1974) “From the Theory and Practice of a Script Writer,” translated by Diana Matias, reprinted in
Screen 15, no. 3: 95–103.
Carroll, N. (2006) “Philosophy and Drama: Performance, Interpretation, and Intentionality,” in Staging
Philosophy: Intersections of Theater, Performance, and Philosophy. Eds. David Krasner, and David Z. Saltz.
Ann Arbor: University of Michigan Press, 104–121.
Carroll, N. (2008) The Philosophy of Motion Pictures, Malden, MA: Blackwell.
Davies, S. (2001) Musical Works and Performances: A Philosophical Study, Oxford: Oxford University Press.
Davies, S. (2009) “Notations,” in A Companion to Aesthetics, 2nd edition, edited by Stephen Davies, Kathleen
Marie Higgins, Robert Hopkins, Robert Stecker, and David E. Cooper, Oxford: Blackwell, 441–443.
Kohn, N. (2000) “The Screenplay as Postmodern Literary Exemplar: Authorial Distraction, Disappearance,
Dissolution,” Qualitative Inquiry 6, no. 4: 489–510.
Korte, B. and Schneider, R. (2000) “The Published Screenplay – A New Literary Genre?” AAA – Arbeiten
aus Anglistik und Amerikanistik 25, no. 1: 89–105.
Lamarque, P. (2009) The Philosophy of Literature, Malden, MA: Blackwell.
Macdonald, I. W. (2004) “Disentangling the Screen Idea,” Journal of Media Practice 5, no. 2: 89–99.
Maras, S. (2009) Screenwriting: History, Theory and Practice, London: Wallflower.
Münsterberg, H. (1916) The Photoplay: A Psychological Study, New York: D. Appleton and Company.
Nannicelli, T. (2010) “The Early Screenwriting Practice of Ernest Lehman,” Journal of Screenwriting 1,
no. 2: 237–253.
Nannicelli, T. (2011a) “Why Can’t Screenplays Be Artworks?” Journal of Aesthetics and Art Criticism 69,
no. 4: 405–414.
Nannicelli, T. (2011b) “Instructions and Artworks: Musical Scores, Theatrical Scripts, Architectural Plans,
and Screenplays,” British Journal of Aesthetics 51, no. 4: 399–414.
Nannicelli, T. (2013) A Philosophy of the Screenplay, New York: Routledge, 2013.
Price, S. (2010) The Screenplay: Authorship, Theory and Criticism, Houndsmills, Basingstoke: Palgrave
Stecker, R. (2003) Interpretation and Construction: Art, Speech, and the Law, Malden, MA: Blackwell.

Further Reading
Boon, K. A. (2008) Script Culture and the American Screenplay, Detroit, MI: Wayne State University Press.
Horne, W. (1992) “See Shooting Script: Reflections on the Ontology of the Screenplay,” Literature/ Film
Quarterly 20, no. 1: 48–54.
Koivumäki, M.-R. (2010) “The Aesthetic Independence of the Screenplay,” Journal of Screenwriting 2,
no. 1: 25–40.
Macdonald, I. W. (forthcoming) The Poetics of Screenwriting, Basingstoke: Palgrave Macmillan.
Millard, K. (forthcoming) Screenwriting in a Digital Era, Basingstoke: Palgrave Macmillan.
Nelmes, J. ed. (2011) Analysing the Screenplay, London: Routledge.
Sternberg, C. (1997) Written for the Screen: The American Motion-Picture Screenplay as Text, Tübingen:

Stephen Davies

I Introduction
I present this topic under two subheadings. Evocriticism involves the application of the
theories of evolutionary psychology, sociobiology, and the like to the interpretation of
literature. Literary Darwinism maintains that literary behaviors are grounded in our
evolved human nature, either as an adaptation that made its bearers more productive of
similarly endowed offspring, or as a byproduct of some other adaptation. Sometimes
“evocriticism” has been suggested as a replacement name for “literary Darwinism”
(Boyd 2009), but the latter came earlier (Carroll 2004). As I use the terms, they cover different
but related uses of biological theory applied to the discussion of literature. A proponent
of evocriticism could regard it as one among a variety of interpretative options and range
it alongside her psychoanalytic, deconstructivist, Marxist, feminist, and queer interpre-
tations of a given text. She might be agnostic about or unconcerned with the issue of
whether prehistoric literary behaviors contributed to our species’ evolution. But many
proponents of evocriticism are also literary Darwinists and combine the two approaches.
Most literary Darwinists (and most of their critics) are English literature academics.
There is often an ideological aspect to the adoption of literary Darwinism by members
of English departments. They are keen to repudiate the dominance in their discipline of
the social science model, according to which knowledge is self-referentially circular, the
human mind is a blank slate, and all concepts are culturally constructed (see Gottschall
2008, Boyd 2009, Carroll 2011). They would like to substitute for this a more scientific
approach that acknowledges biological commonalities in our shared human nature and
sees literature as playing a vital role in identifying and conveying these.
For valuable collections of academic discussions of the theories and issues, see
Gottschall and Wilson (2005), Boyd et al. (2010), and the journal Style (2008, 2/3) and
(2012, 3/4).

II Evocriticism
Evolutionary psychology argues that our minds, as well as our bodies, are products of
evolution. Many of our current behaviors and ways of thinking were inherited from our
Stephen Davies

forebears and were adaptive for them in facilitating their reproductive success. Among
the tendencies discussed by evolutionary psychologists are sexual attraction, mating
strategies, female childcare, male mate guarding, jealousy, male aggression, striving for
status, social score-keeping, cooperation, free riding, cheater detection, kin rivalry, kin
altruism, intergenerational and intragroup relations, and intergroup conflict. They
explain sociality and empathy in terms of Theory of Mind (our capacity to project into
the minds of others to understand what they are thinking, desiring, and feeling) and
the related notion of levels of intentionality (in which I know that she fears that he
will resent . . .).
Of course, much literature is concerned with graphically portraying perennially
fascinating human themes: love, ambition, redemption, war, friendship, loss, political
struggle, and the like. And these are precisely the subjects theorized by evolutionary
psychology. So it is natural to test whether great literary works match the theories and
if this explains the value we accord them.
On this last matter there is clearly room for skepticism. Ellen Dissanayake (1999)
objects that the approach reveals biological, not literary value. Others (for example,
Goodheart 2007, Kramnick 2011) have lamented how evocriticism reduces interpreta-
tion to thematics and thereby overlooks aspects of the author’s skills and style that
are crucial to assessments of literary value. Even advocates (such as Carroll 2011) have
identified some of their fellow evocritics as guilty of such charges. And there is the
concern that their focus on human universals leads evocritics to overlook what its
cultural time and location contribute to a literary work’s character and value (Easterlin 2012).
But I doubt that these worries are decisive. Evocritics need not confine their critical
appreciation exclusively to literature’s illustration of evolutionary themes, and even if
they do, they need not claim thereby to account for all the ways in which literature
might be valuable. On the positive side, they can surely claim that the author’s psycho-
logical insight and sensitivity adds value to a literary work and that the Darwinist
approach to criticism seeks evidence of this. And they can attend to the manner in
which the author individualizes and develops the theme in question, these being prime
locations of literary value. In considering the work’s individuality and originality, they
can and should take into account the work’s artworld context and cultural background.
The uses in literary interpretation of ideas from evolutionary psychology and its
cognate disciplines are too frequent and diverse to be usefully summarized here. Instead,
I offer a few illustrative cases. And because of their brief treatment, I provide little
evidence of subtle literary appreciation in the following examples.
Ernest Hemingway’s short story, “The Short Happy Life of Francis Macomber,” tells
the tale of a rich man who shows cowardice on a big-game hunt. He is scorned by his
ageing wife, Margot, who becomes sexually involved with the safari’s professional hunter.
In trying to redeem himself, Francis kills a charging wounded buffalo, succeeding only
at the last second. His wife, Margot, fires at the same moment from the car, hitting
Francis in the head and killing him.
According to Michelle Scalise Sugiyama (1996a), women should have inherited a
preference for men who are successful hunters. Meanwhile, short-term extra-pair copu-
lations with men of high reproductive quality can be to the benefit of children thereby
conceived (provided the women’s unsexy but resource-rich partners continue to support
them). So, Margot responds to Francis’ humiliation in a predictable fashion. But her
position would be undermined if he bravely overcame his fear. He then would become


more likely to ditch her for a chaste, younger woman. It is therefore likely that she
intended to kill rather than to save Francis, in order to secure his resources.
The politics of sexual attraction are explored in a broader way by Cynthia Whissel
(1996). Her study of 25 popular women’s novels and six famous romantic stories bears
out the prediction that females choose mates on the basis of economic resources and
parenting potential, while males seek sexual exclusivity and fertility (that is, youthfulness).
Male protagonists are likely to be bold, aggressive, and untrusting, while female
protagonists are likely to be friendly, timid, and astonishable. Moreover, while novels
for women are predominantly about mate choice and mating commitment, fictions for
men more often focus on inter-male aggression and adventure, and rarely involve
such commitment.
Ian Jobling (2001) notes the cross-cultural commonness of stories pitting a hero
against a semi-human ogre. Citing examples from 20 cultures, he shows that the stories
share a number of key elements: a courageous, morally upright hero defends the group
against a less-than-fully-human outsider who has injured some group members in some
way. The hero defeats the ogre in physical combat, usually by killing him. Such stories
help to reinforce innate positive biases in the perception of the self and the in-group and
they reinforce negative biases in the perception of out-groups. Jobling claims that this
is one of the principal reasons why such tales are told.
Male aggression is also a subject of Homer’s Iliad, according to Jonathan Gottschall
(2001). In nonhuman animals, male competition is mostly for status or territory and
the access these provide to mates. Though such competition is often ritualized or even
ceremonial, lethal combat is not uncommon. As Gottschall catalogs, Homer’s Iliad
makes human life out to be surprisingly similar. The Greek bard highlights how, when
we are mired in war, human animality, physical vulnerability, and mortality come to the
fore. At the same time, he shows us as apart from other animals in possessing the god-
like intelligence that allows us to comprehend the impossibility of transcending our
animal nature.
Meanwhile, an examination of ten of Shakespeare’s plays reveals that relations
between characters mimic the social dynamics of small groups (Stiller et al. 2003, Stiller
and Hudson 2005). The number of characters that are present within each scene reflects
similar numbers to those of observed human support cliques. Different scenes are linked
through keystone characters, thereby enabling the flow of information and the expansion
of the small social world of the fictional group. This approach possibly respects cognitive
limits on a large audience’s tracking of a range of characters.
Daniel Nettle (2005) takes a broad look at the dramatic genres of tragedy and
comedy. The former is typically about status competition and the latter about mate
selection. The enduring appeal of these genres is due to the way they exploit our intrinsic
interest in the strategies others of the group adopt in the attempt to maximize their
fitness. We study this for the relevance it has to our own attempts at fitness maximizing.
It is worth noting several respects in which these illustrative examples are not typical.
Cross-cultural studies (as in Jobling 2001), the use of statistical analysis (as in Whissel
1996), genre studies (as in Whissel 1996 and Nettle 2005), and comparisons across more
than a few works (as in Whissel et al.) are comparatively uncommon, as is the focus (in
Jobling 2001) on what is a predominantly oral, as against written, literary tradition.
Among literary scholars, Jonathan Gottschall (2008) and Joe Carroll (et al. 2009, 2011)
are unusual in coding and statistically testing a large body of works. Most evocritics

Stephen Davies

based in university literature departments discuss one or only a few works at a time and
these few works typically are high-quality published novels written in the past two
hundred years. The primary focus is on fictional narratives, but lyrics also have been
considered (see Boyd 2012).

III Literature as an Evolutionary Adaptation

Reading and writing originated about 5,000 years ago (Gosden 2003), but literacy
became comparatively widespread only several centuries after the invention of the
printing press in the fifteenth century. Reading and writing are not supported by
innate, task-specific neural circuits. Neither is regarded as evolutionarily adaptive
by biologists. So, if we are to identify literature as an evolutionary adaptation, the most
plausible account will take a broad view of literature, not confining it to fine writing,
and see its origins in prehistoric oral traditions of tale-telling and recitation.
Not surprisingly, the standard view is that literary behaviors are an ancient adapta-
tion and that they retain their adaptive benefits to the present day. Thus: “When I speak
of the adaptive functions of literature, I mean to signify the adaptive functions of the
oral antecedents of written stories, poems, and plays. The same arguments that apply to
these oral forms will be understood as extending also to their counterparts in written
language” (Carroll 2009, 160). But other views have been canvassed. Paul Hernadi
(2001) thinks we are now so surrounded by literature that, like our hunger for sugar and
fat, our interest in it is no longer adaptive. The opposite view is that literature has only
recently become adaptive, helping us to accommodate the mismatch between our
species’ current and ancient lifestyles (Storey 1996) or to deal with the chaos and uncer-
tainty of modern life (Argyros 1991, Austin 2007). But no evidence is offered to show,
for instance, that relevant literary behaviors have recently become heritable.
Where are we to locate the relevant adaptation? So ubiquitous and foundational is it,
the human tendency to construct narratives (about who we are, where we came from,
where we are going, what happened) is most likely adaptive. Similarly, our capacity
imaginatively to create and consider fictional scenarios is central to practical reasoning,
as when we hypothesize about the future or counterfactually consider how the past
might have been. Also adaptive is our Theory of Mind ability, which allows us to under-
stand and interact socially with other people. The creation and appreciation of literature
might involve these adaptive behaviors, but it is unlikely that literature is the primary
adaptation that gives rise to them. They are directed mainly to negotiating the real world
and the risks and challenges it poses and they tend to be well-developed in the child
before mastery of reading and widespread engagement with literature (Callaghan et al.
2005, Easterlin 2012). Indeed, it is more plausible to suggest that the creation and appre-
ciation of literature depends on the prior mastery of these skills (Astington 1990). In
other words, if literature is adaptive it is not its narrativity, fictionality, or psycho-social
insight that makes it so. Narrativity does not distinguish it from biography or history;
fictionality does not distinguish it from practical reasoning and fantasy; psycho-social
insight does not distinguish it from work in history, psychology, and the social sciences
(Davies 2012, chapter 11).
In response, it might be argued that we should not overplay the distinction between
factive and fictional discourse (Dutton 2009). Both are liable to be creatively constructive
and imaginative. I agree. But literary Darwinists are very clearly focused on the kind of
writing that belongs in university departments of literature, not departments of history


or sociology. And we are rarely indifferent to whether a story is told as fiction or fact.
We may adopt indifference to the fact that the creation stories, folk science, or religious
myths of other cultures are told as true, for instance. But most of the time it matters to
us what really happened and why, and we are keen to separate this from conjecture,
fantasy, and fictions told as entertainments. (This is appreciated and discussed in
Zunshine 2006 and Swirski 2010.)
Of course, it remains possible to argue that literature is adaptive because it refines
the relevant mind-reading and information-extracting skills in ways that would not
otherwise be possible (see Zunshine 2006, Mar and Oatley 2008, and Green and
Donahue 2009, for instance) or that it plays a special role in “fine tuning” or “calibrating”
them (see Tooby and Cosmides 2001, Carroll 2011). But I doubt that literature of the
kind under discussion is uniquely important in these regards, so this position appears
to be weak.
This is not to deny that we can learn from literature, that it allows us to explore
scenarios “off-line” without exposing ourselves to real-world costs and risks, or that it
might make us more sensitive to and understanding of people who are not like us. Many
literary Darwinists characterize this as literature’s evolutionary purpose. For instance,
Denis Dutton writes that

the features of a stable human nature revolve around human relationships of

every variety: social coalitions of kinship or tribal affinity; issues of status; recip-
rocal exchange; the complexities of sex and child-rearing; struggles over
resources; benevolence and hostility; friendship and nepotism; conformity and
independence; moral obligations, altruism, and selfishness; and so on. . . . These
issues constitute the major themes and subjects of literature and its oral
antecedents. Stories are universally constituted in this way because of the role
storytelling can play in helping individuals and groups develop and deepen
their own grasp of human social and emotional experience.
(2009, 118)

Similarly, Brian Boyd holds that

fiction . . . does not establish but does improve our capacity to interpret events.
It preselects information of relevance, prefocuses attention on what is strategi-
cally important, and thereby simplifies the cognitive task of comprehension.
At the same time it keeps strategic information flowing at a much more rapid
pace than normal in real life, and allows a comparatively disengaged attitude
to the events unfolding. . . . Fiction aids our rapid understanding of real-life
social situations, activating and maintaining this capacity at high intensity
and low cost.
(2009, 192–3)

In assessing such claims, we should bear in mind that literature is equally capable of
providing seductively misleading information, of inculcating hatred, of affirming what
is corrupt in the status quo, of reinforcing crude stereotypes, of validating what should
be questioned in the prevailing morality, and so on. And even where it is genuinely
useful and not misleading, notice that the value of literature is here alleged to lie in its
serving both as a vehicle on which we practice skills and as a repository for knowledge.

Stephen Davies

That seems to take us far from a Darwinist perspective, which should deal with inherited
traits and dispositions, not with acquired competence and learned information. If there
are evolutionary adaptations at work here, they are in the dispositions that allow us to
learn from experience and rehearsal. But as I have already suggested, those dispositions
are mainly directed to the real world and motivate engagement prior to any full-blown
interest in literature as such.
One way of mitigating this last objection would be by shifting from classical evolution
theory, in which the unit of selection is the individual (or her genes) and the means of
transmission is genetic, to what is known as multilevel selection theory, in which the unit
of selection is the group and the means of transmission is cultural. The idea then would
be that literature-enhanced groups outcompeted literature-impoverished groups and,
secondly, that intragroup competition was less significant for survival and reproductive
success than intergroup competition. And rather than genetic inheritance, the relevant
benefits would be conveyed culturally, with literary traditions playing a key role.
A few literary Darwinists explicitly adopt multilevel selection theory (Boyd 2009, for
example), though none attempts to provide the kind of evidence I just suggested is
appropriate. A few others identify group benefits, while making clear that they regard
the individual, not the group, as the primary unit of selection (Dissanayake 1988 and
Hernadi 2001, for instance). But the vast majority of the many writers who appeal to
the way that literature unifies and strengthens the group by confirming and reinforcing
its myths, hierarchies, practices, and values are apparently unaware of their departure
from the classic model of natural selection. They are also apparently unaware of the
reservations biologists hold about multilevel selection (Pinker 2007) and the weakening
of explanatory power that the shift entails (Davies 2012, 43–4).
What alternative adaptive functions are proposed for literature? A common suggestion
is that it is a form of sexual display. Both Geoffrey Miller (2000) and Denis Dutton
(2009) draw attention to the seductive power of a large vocabulary and to the possibilities
provided by language for the exhibition of intelligence, humor, imaginativeness, creativity,
and playfulness. Literary authors typically prove themselves superior in their mastery
of language.
This approach faces a number of objections. It looks as if what might be adaptive here
is intelligence and humor, rather than the production of literature as such. Literary
behaviors might implement these adaptations, but so do many non-literary alternatives.
To return the focus to literature, it would be necessary to demonstrate, for example, that
skilled authors parented more extensive lineages than non-authors and that literary
skills are heritable. (In fact, advocates of sexual selection rarely attempt such demonstra-
tions and in any case, I doubt that the attempt would be successful.) Besides, the creation
and consumption of literature is often a private affair, far removed from courtship
settings or romantic contexts. Perhaps the ancient group’s fireside storyteller benefitted
reproductively from being the focus of attention (Gottschall 2012), but in that case we
might suspect that contemporary professional literature has forfeited that earlier
adaptive function. And if face-to-face storytelling is to be made central, it should also
be recognized that this takes place more often between mothers and children, or within
groups of children, than between those in the market for a reproductive partner. This is
not to deny that males might appropriate literary behaviors for the purpose of sexual
display. (Almost any behavior that can be turned into a competition can be used in this
way, of course.) But it is to raise doubts about the claim that literary behaviors exist and
are sustained by the evolutionary need to attract mates.


A related theory argues that literature confers social, not solely sexual, rank. Michelle
Scalise Sugiyama (1996b) has defended such a view in discussing hunter-gatherer
communities. And Brian Boyd (2009) presents this as a general account of the evolu-
tionary significance of literature. He maintains that we are an ultrasocial species. As
such, we seek both to command the attention of others and to share attention. Art—
and, in particular, literature—provides the advantage of “getting along (improved
cooperation, and therefore participation in more successful groups) and getting ahead
(improved status within one’s own group)” (Boyd 2009, 108).

An evolutionary model of fiction, therefore, should focus on ways storytellers,

as active individual strategists, maximize the attention of their audience by
appealing to features that have evolved to be of interest to all human minds, to
our shared understandings of events, our shared predispositions to be interested
in and engaged by what others do and our sheer readiness to share attention.
(Boyd 2001, 201)

The concerns faced by these theories are like those raised earlier. It looks as if what is
adaptive is social status, and plainly literary skill provides only one route to this.
Meanwhile, the production and consumption of literature in the modern context take
place under circumstances that are not social. And as Pinker (2007, 174) asks, what is
adaptive about sharing attention to events that never happened? Literary creativity
might be coopted as a means to seeking status, but it is not obvious that this is literature’s
original or primary purpose.

IV Literature as a Byproduct
It may be that literary behaviors are connected to evolution not by being adaptive but
as byproducts instead. That is, they are incidental consequences of some behavior or
aptitude that is adaptive in its own right. Of course, this thesis does not secure what
literary Darwinists most desire: that literature is seen to be foundational to our human
nature by virtue of improving the reproductive success of our distant ancestors. But some
evolutionary psychologists think the byproduct thesis should be the default assumption
for the arts in general (see Tooby and Cosmides 2001, De Smedt and De Cruz 2010).
And Steven Pinker (2007), who shares this view, argues against adaptationist accounts
of literature and also against the common mistake of extrapolating beyond the observation
that literature benefits us to the conclusion that literary behaviors are adaptive for the
sake of those benefits.
What adaptive behaviors could give rise to literature? I have already mentioned some
prime candidates: our tendency to narrativize, to think and reason in terms of what is
not actual, and to understand our fellow humans as intentional agents with beliefs,
desires, and emotions other than our own. Given both how important such activities are
in ordinary life and how intrinsically pleasurable they can be to exercise, it would not
be at all surprising that we would invent pastimes that involved them. In any case, this
view is consistent with Pinker’s hypothesis: “fiction may be, at least in part, a pleasure
technology, a co-opting of language and imagery as a virtual reality device which allows
a reader to enjoy pleasant hallucinations like exploring interesting territories, conquering
enemies, hobnobbing with powerful people, and winning attractive mates. Fiction,
moreover, can tickle people’s fancies without even having to project them into a thrilling

Stephen Davies

vicarious experience. There are good reasons for people (or any competitive social
agent) to crave gossip, which is a kind of due diligence on possible allies and enemies.
Fiction, with its omniscient narrator disclosing the foibles of interesting virtual people,
can be a form of simulated gossip” (2007, 171).
William Flesch (2007) identifies a different but also plausible adaptation as the
source of literature. He argues that we admire altruists and those who punish wrongdoers,
even at their own cost. Both behaviors signal evolutionary fitness. Flesch maintains
that fictions tell interesting stories about social interactions involving altruism and
punishment, and about the score-keeping with which we monitor and judge these
behaviors: “We are fitted to track one another and to track as well how others monitor
one another and what they do when they monitor one another. What we wish to track
is past behavior in order to respond in the present to that behavior. Fiction recruits
this central capacity in human social cognition for taking pleasure in responding to
the nonactual” (Flesch 2007, 46).
As Flesch explains it, literature does not develop our sense of social justice or improve
our capacity to keep track of social relations. And though authors display their under-
standing of such matters and can be admired for this, their gain in status is not an
evolutionary purpose for which we have literature. Though he does not explicitly
advocate the byproduct thesis, Flesch’s theory is of that type. If I were to question his
theory, it would be about the narrowness of its scope. It is far from plain that all narrative
fiction is driven by our interest in monitoring altruism and punishment.

V Closing Comments
Evocriticism provides an interesting addition to the interpretative arsenal employed
by literature commentators. Provided it is used to reveal aspects of the literary work,
rather than simply to illustrate the theories of evolutionary psychology, it can account
for some aspects of literary value. Of course, if it is to be revealing it must be used
carefully, with proper respect for the work it targets, but this is no less true of other
approaches to interpretation that have as their goal the revelation of the author’s work
and achievements.
Literary Darwinism of the variety that identifies literary behaviors as evolutionary
adaptations is more problematic. First, we should note the variety of different theories
in the arena. They vary as regards when literature was adaptive, what it is an adaptation
for, whether natural or sexual selection is central, whether the unit of selection is the
individual or the group, and whether the focus should fall primarily on highly talented
individuals or on everyone, as the successful heirs of distantly past adaptations. As a
result of this variety, it is not easy to come to an all-encompassing conclusion.
Second, many literary Darwinists are careless of what is required in showing that
literary behaviors are adaptive. It is not common for them even to consider if past
authors were notably successful reproducers or if the literary behaviors they identify as
adaptive are heritable. There is a tendency to assume that it is sufficient to point to
advantages that exposure to literature can produce, whereas more is needed to demonstrate
that those behaviors were biologically selected on account of those advantages. And if
there is a shift toward locating the benefits of literature as accruing to the group, this
rarely goes with recognition of the kinds of evidence this calls for, of the differences
between organism-level and group-level selection, and of the controversy that still
surrounds the idea of multilevel selection among biologists and philosophers of biology.


Third, theorizing in the literary Darwinist mode frequently seems to be ideologically

motivated by a perceived socio-political need to rescue the academic disciplines of
English or literary studies, to ennoble the arts, or to put studies of the humanities on the
same level of those of the sciences. This ideological component is often more prominent
than a commitment to meeting scientific standards of rigor. Many hypotheses of literary
Darwinism not only are not subjected to empirical testing, but also could not be assessed
in that manner. The movement might be stronger and more convincing were there an
empirical program to support it. And some literary Darwinists (such as Gottchall 2008
and Carroll 2011) have made moves to generate such a program. But on the other hand,
others question the compatibility of such methodologies with the interpretative enter-
prise that sifts through the complex layers of meaning that are found in literary works
(Easterlin 2012) and object to reconfiguring literary criticism around the study of human
nature (Crews 2008). (On the uneasy history of the relation between literary criticism
and psychology, see Lamarque 2011.)
The version of literary Darwinism that characterizes literary behaviors as byproducts
of adaptations lying elsewhere certainly is plausible and would find more adherents than
opponents among evolutionary psychologists and biologists. But it is worth noting, as I
have not so far stressed, that this is not a less demanding option. It needs to rule out the
possibility that behaviors that were not originally adaptive could have become so later,
and it needs to agree on the adaptations that give rise to these ancillary behaviors and
to account for how they do so.

Argyros, Alexander J. 1991. A Blessed Rage for Order: Deconstruction, Evolution, and Chaos. Ann Arbor:
University of Michigan Press.
Astington, J. W. 1990. “Narrative and the Child’s Theory of Mind.” In Narrative Thought and Narrative
Language, edited by B. K. Britton and A. D. Pellegrini, 151–71. Hillsdale: L. Erlbaum.
Austin, Michael. 2007. “The Influence of Anxiety and Literature’s Panglossian Nose.” Philosophy and
Literature 31: 215–32.
Boyd, Brian. 2001. “The Origin of Stories: Horton Hears a Who.” Philosophy and Literature 25: 197–214.
Boyd, Brian. 2009. On The Origin of Stories. Cambridge: Harvard University Press.
Boyd, Brian. 2012. Why Lyrics Last: Evolution, Cognition, and Shakespeare’s Sonnets. Cambridge, MA: Harvard
University Press.
Boyd, Brian, Joseph Carroll, and Jonathan Gottschall, eds. 2010. Evolution, Literature, and Film: A Reader.
New York: Columbia University Press.
Callaghan, Tara, Philippe Rochat, Angeline Lillard, Mary Louise Claux, Hal Odden, Shoji Itakura, Sombat
Tapanya, and Saraswati Singh. 2005. “Synchrony in the Onset of Mental-state Reasoning. Evidence from
Five Cultures.” Psychological Science 16: 378–84.
Carroll, Joseph. 1999. “The Deep Structure of Literary Representations.” Evolution and Human Behavior
20 (3): 159–73.
Carroll, Joseph. 2004. Literary Darwinism: Evolution, Human Nature, and Literature. London: Routledge.
Carroll, Joseph. 2011. Reading Human Nature: Literary Darwinism in Theory and Practice. Albany: State
University of New York Press.
Carroll, Joseph, Jonathan Gottschall, John A. Johnson, and Daniel Kruger. 2009. “Human Nature in
Nineteenth-Century British Novels: Doing the Math.” Philosophy and Literature 33: 50–72.
Crews, Frederick. 2008. “Apriorism for Empiricists.” Style 42: 155–60.
Davies, Stephen. 2012. The Artful Species: Aesthetics, Art and Evolution. Oxford: Oxford University Press.
De Smedt, Johan and Helen De Cruz. 2010. “Toward an Integrative Approach of Cognitive Neuroscientific
and Evolutionary Psychological Studies of Art.” Evolutionary Psychology 8: 695–719.
Dissanayake, Ellen. 1988. What Is Art For? Seattle: University of Washington Press.
Dissanayake, Ellen. 1999. “Sociology and the Arts: An Introduction.” In Sociobiology and the Arts, edited by
J. B. Bedaux and B. Cooke, 27–42. Amsterdam: Rodopi.

Stephen Davies

Dutton, Denis. 2009. The Art Instinct. New York: Bloomsbury Press.
Easterlin, Nancy. 2012. A Biocultural Approach to Literary Theory and Interpretation. Baltimore: John Hopkins.
Flesch, William. 2007. [Comeuppance]: Costly Signaling, Altruistic Punishment, and Other Biological Components
of Fiction. Cambridge: Harvard University Press.
Goodheart, Eugene. 2007. Darwinian Misadventures in the Humanities. Edison: Transaction.
Gosden, Chris. 2003. Prehistory: A Very Short Introduction. Oxford: Oxford University Press.
Gottschall, Jonathan. 2001. “Homer’s Human Animal Ritual Combat in the Iliad.” Philosophy and Literature
25: 278–94.
Gottschall, Jonathan. 2008. Literature, Science, and a New Humanities. New York: Palgrave Macmillan.
Gottschall, Jonathan. 2012. The Storytelling Animal: How Stories Make Us Human. Boston: Houghton Mifflin
Gottschall, Jonathan and David Sloan Wilson, eds. 2005. The Literary Animal: Evolution and the Nature of
Narrative. Evanston: Northwestern University Press.
Green, Melanie and John Donahue. 2009. “Simulated Worlds: Transportation into Narratives.” In Handbook
of Imagination and Mental Simulation, edited by K. Markman, W. Klein, and J. Suhr, 241–54. New York:
Psychology Press.
Hernadi, Paul. 2001. “Literature and Evolution.” SubStance: A Review of Theory and Literary Criticism
94/95: 55–71.
Jobling, Ian. 2001. “The Psychological Foundations of the Hero-Ogre Story.” Human Nature 12: 247–72.
Kramnick, Jonathan. 2011. “Against Literary Darwinism.” Critical Inquiry 37: 315–47.
Lamarque, Peter. 2011. “On Keeping Psychology Out of Literary Criticism.” In The Aesthetic Mind: Philosophy
and Psychology, edited by E. Schellekens and P. Goldie, 299–312. Oxford: Oxford University Press.
Mar, Raymond A. and Keith Oatley. 2008. “The Function of Fiction is the Abstraction and Simulation of
Social Experience.” Perspectives on Psychological Science 3 (3): 173–92.
Miller, Geoffrey F. 2000. The Mating Mind: How Sexual Choice Shaped the Evolution of Human Nature.
New York: Doubleday.
Nettle, Daniel. 2005. “The Wheel of Fire and the Mating Game: Explaining the Origins of Tragedy and
Comedy.” Journal of Cultural and Evolutionary Psychology 3: 39–56.
Pinker, Steven. 2007. “Toward a Consilient Study of Literature.” Philosophy and Literature 31: 162–78.
Scalise Sugiyama, Michelle. 1996a. “What’s Love Got to Do with It? An Evolutionary Analysis of ‘The Short
Happy Life of Francis Macomber’.” Hemingway Review 15 (2):15–32.
Scalise Sugiyama, Michelle. 1996b. “On the Origins of Narrative: Storyteller Bias as a Fitness-enhancing
Strategy.” Human Nature 7: 403–25.
Stiller, James and Matthew Hudson. 2005. “Weak Links and Scene Cliques within the Small World of
Shakespeare.’ Journal of Cultural and Evolutionary Psychology 3: 57–73.
Stiller, James, Daniel Nettle, and Robin I. M. Dunbar. 2003. “The Small World of Shakespeare’s Plays.”
Human Nature 14: 397–408.
Storey, Robert F. 1996. Mimesis and the Human Animal: on the Biogenetic Foundations of Literary Representation.
Evanston: Northwestern University Press.
Swirski, Peter. 2010. Literature, Analytically Speaking: Explorations in the Theory of Interpretation, Analytic
Aesthetics, and Evolution. Austin: University of Texas Press.
Tooby, John and Leda Cosmides. 2001. “Does Beauty Build Adapted Minds? Toward an Evolutionary Theory
of Aesthetics, Fiction and the Arts.” SubStance 94/95: 6–27.
Whissel, Cynthia. 1996. “Mate Selection in Popular Women’s Fiction.” Human Nature 7: 427–47.
Zunshine, Lisa. 2006. Why We Read Fiction: Theory of Mind and the Novel. Columbus: Ohio State University

Stein Haugom Olsen

In the last decades the concept of a canon has figured prominently in theoretical debates
about literature. The notion of a canon has been used in this debate as a critical
primitive. While there is disagreement about whether or not it is possible to have a
canon, whether there is one canon or many canons, and about who should dictate the
canon, it has been taken for granted that the notion of a canon is itself a meaningful and
fruitful concept in literary studies, and that these questions are meaningful and can be
given answers. The idiom of ‘canon’ is relatively new in the theoretical debate. Until
the early 1960s there were debates about the nature, content, and value of the literary
tradition. Major contribution to the debate around tradition were such works as T.S. Eliot’s
“Tradition and the Individual Talent”, which proposed a thesis concerning the
relationship between tradition and new works that were meant to be a contribution to
the tradition, and F.R. Leavis’s The Great Tradition, which purported to define what
would today have been called the canon of the English novel. The concept of tradition
has not, of course, disappeared from the critical vocabulary, but the concept has ceased
to be the focus of an ongoing debate about the nature and value of works of the literary
past and their relation to the literature of the present. ‘Tradition’ like ‘canon’ was used
as a critical primitive, but where the notion of a secular canon of artworks has been
formed on the model of a canon of scripture, essentially a technical concept, the notion
of ‘tradition’ has a broad range of applications in various spheres of life which provides
a broad commonsense basis for its use in literary studies and art history.

The concept of ‘canon’ as a list of selected texts is in its origin theological. In the words
of the OED the canon is “The collection or list of books of the Bible accepted by the
Christian Church as genuine and inspired” (Oxford English Dictionary Online, “Canon”,
entry 4) The concept of ‘canon’ has its complementary in the concept of ‘apocrypha’,

A writing or statement of doubtful authorship or authenticity; spec. those books

included in the Septuagint and Vulgate versions of the Old Testament, which
were not originally written in Hebrew and not counted genuine by the Jews,
and which, at the Reformation, were excluded from the Sacred Canon by the
Protestant party, as having no well-grounded claim to inspired authorship.
(Oxford English Dictionary Online, “Apocrypha”, entry 1)
Stein Haugom Olsen

The apocrypha are texts which apparently have some claim to be included in the canon
or for which some such claim has been made.
This theological concept of canon embodies two important elements: authenticity
and authority. There is, however, a problematic relationship between these two
elements. Though the notion of ‘authenticity’ is logically independent of the notion of
‘authority’, it has no independent validity. The canon of scripture has been fixed by the
authorita­tive organs of the Church. No matter how violent the disagree­ment before
such decisions are taken, the official decision settles the matter. It is a principle recog-
nized by the Roman Catholic Church, the Orthodox Church, and the Church of
Eng­land that only the Church has a right to declare a book cano­nical. Thus, at the
Council of Trent (1546) the Roman Catholic Church dogmatically affirmed that the
entire Latin Vulgate enjoyed equal canonical status. The logical basis on which the
Church declares a work to be canonical, is genetic: it has to be a text dealing with Old
Testament or New Testa­ment history which the Church decides is inspired. After a text
has been declared canonical, it is given the attention appropriate only to canonical texts
because they are assumed to contain the hidden word of God. Only then does it become
the object of interpretation and exegesis. Prior to being declared canonical it is not
thought worthy of this attention. The connection between authority and the condition
for in­clusion in the canon of the scripture (authenticity) is close: what the authoritative
body decides is whether or not a text fulfils the condition of being inspired. One may
challenge this decision, but there is no other way of establishing a new text as canonical,
than by getting a new decision from the Church. A challenge to the list of canonical
texts either fails or forces a new decision. Thus what appears to be the basic criterion for
declaring a text canonical, i.e. the criterion of authenticity (a text must be ‘genuine and
inspired’) though it is logically independent, has no independent validity.
The notion of a scriptural canon does not embody an evaluative element. ‘Genuine
and inspired’ texts constitute a separate category (a genre) that demand a certain kind
of attention because they are sacred texts, but there is no reference to the quality of the
texts built into the notion of authenticity.

It is possible to distinguish three different concepts of ‘canon’ in literary criticism and
literary theory. ‘Canon’, according to the OED, can also mean “those writings of a
secular author accepted as authentic” (Oxford English Dictionary Online, “Canon”, entry 4).
This is the concept employed in the following paragraph about the ‘Shakespeare

There is no external evidence of value about these uncanonical plays. The fact
that a publisher declared a work to be Shakespeare’s tells us something about
his popularity but nothing about his authorship. The true canon rests upon the
Folio of 1623; and the exclusion of a play from that volume must be taken as
strong, but not necessarily irrefutable, evidence against it.
(Sampson 1961, 274)

This concept of a secular canon embodies the element of authenticity but it does not
have the implicit reference to authority. There is a good reason for this. The criterion


of ‘inspired’, which is the only criterion for inclusion in the scriptural canon, does not
in itself dictate a protocol for testing whether a text is inspired or not. An authoritative
decision is therefore indispensable in identifying an ‘inspired’ text. However, there is a
wide range of tests available for deciding whether a text has been written by a particular
individual, from eyewitness accounts and publication records to textual and stylistic
evidence. Inclusion of a Shakespeare play in the First Folio (1623) by John Hemminge
and Henry Condell, Shakespeare’s fellow actors, would be, given their motivation and
their reputation as well as what must have been generally known about Shakespeare at
the time, strong evidence of authenticity. Mention of a play as written by Shakespeare
in contemporary sources would also count as evidence of authenticity, e.g. inclusion of
a play in the list of Shakespeare’s plays given by Francis Meres in Palladis Tamia (1598),
as would an entry of a play under Shakespeare’s name in the Stationer’s Register etc.
However, this last kind of evidence must count for less and must be cross-checked with
other evidence as it was perfectly possible for someone to register a play under
Shakespeare’s name. Thus, it is quite clear in the case of literary authorship what would
constitute evidence for authenticity, while in the scriptural case the criterion of inspira-
tion provides little guidance as to what would constitute such relevant evidence.

A second concept of canon in use in literary criticism and theory is not even listed in
the OED, though it does appear in the Merriam-Webster: “a sanctioned or accepted
group or body of related works” (Merriam-Webster Online Dictionary, “Canon”, entry 3.c).
This is the concept of ‘canon’ that according to Curtius “first appears with the meaning
‘catalogue of authors’ in the fourth century of our era specifically in relation to Christian
literature” (Curtius 1953, 256, fn. 2). However, the practice of putting together a catalogue
of selected authors arose long before the concept of ‘canon’ was introduced and Curtius
traces it back to “The Alexandrian philologists” who “are the first to put together a
selection of earlier literature for the use of grammarians in their schools” (ibid.:249). A
catalogue of this kind involved the selection of ‘model authors’, or as they later were
called, ‘classics’:

The Alexandrians call them “the received (into the selection)” (έγκρινόμενοι,
έγκριτοι; in Pollux, IX, 15 κεκριμένοι). This term defied Latinization. It could
no more pass into modern usage than Quintilian’s “genera lectionum” or his
prolix circumlocutions with ordo (“auctores in ordinem redigere,” I, 4, 3) and
numerus (“in numerum redigere,” X, 1, 54). A new and convenient word had to
be found. But it was not until very late, and then only in a single instance, that
the name classicus appears: in Aulus Gellius (Noctes Atticae, XIX, 8, 15). This
learned compiler of the Antonine period discusses a large number of
grammatical problems. Are quadriga and arena to be used in the singular or the
plural? The thing to do is to follow the usage of a model author: “e cohorte illa
dumtaxat antiquiore vel oratorum aliquis vel poetarum, id est classicus
adsiduusque aliquis scriptor, non proletarius”: “some one of the orators or poets,
who at least belongs to the older band, that is, a first-class and tax-paying
author, not a proletarian.”

Stein Haugom Olsen

This concept of canon is fundamentally different from the notion of a scriptural canon
since it contains no reference to authenticity. It does, however, embody the notion of
authority. “The choice of books in our teaching institutions” (Bloom 1995, 15) involves
a selection which has to be made by people in authority. The catalogue of authors and
works is prescribed by a group of people who are or who think they are or ought to be in
a position of authority to impose this list on those who are students, in the wide sense,
of literature. However, in spite of the similar logical role which authority plays in fix-
ing the application of the concept of ‘literary canon’ and ‘canon of scripture’, there is a­
difference between the type of authority to which the two concepts appeal as well as to
the scope of that authority. The authoritative organ which determines the scriptural
canon has a status and a role within the Church which the authorities that determine
the literary canon do not have. The representatives to the National Conference on
Uniform Entrance Requirements that was responsible for drafting the list of texts to be
set for college entrance requirements in English in the United States in 1894, and which
was therefore responsible for establishing a list of canonical works which secondary
schools would adopt, can hardly be compared in its authority with the Church Council
of Trent. In fact, such authorities as have existed in the history of literature, have
always possessed their authority for a limited time, and their authority has always been
recognized only by a limited group. “Yet even as the conferences and committees of the
nineties consolidated the power of English”, says Gerald Graff in his book on the
development of English as an academic subject in the United States,

[T]hey dramatized the conflict of philosophies that was preventing any stable
conception of the new subject from materializing. Significantly, the 1894
conference adopted two separate lists, “one for ‘wide’ and the other for ‘deep’
study”, a compromise between the conflicting viewpoints of “the advocates of
disciplined study” and the “proponents of apprec­iation” and “humanistic goals”.
(Graff 1987, 100)

Even that culturally dominant group which dictated the canon of French Classicism in
the seventeenth century, dictated a canon only for a time and a place. If literature is an
institution, and there are good reasons for thinking of it in that way, it is an institution
the concepts and conventions of which are not grounded in authority. For literature has
simply not developed the characteris­tics of a universal Church with its notion of authority
and the reliance upon dogma: there is no authoritative body within the institu­tion of
literature which could establish a universally valid canon of literature through a decision
similar to official decisions taken by the authoritative body of a worldwide Church. The
literary institution is not organized in the systematic way that even the most democratic
Church is organi­zed. There is within the literary institution today a distinction between
connoisseurs or adepts and the less able, between highly trained and sensitive practitioners
who know how to read and who also have the necessary fund of knowledge about the
institution and its social setting, and those who are less skilled and less knowledgeable.
Today this distinction coincides to a great extent with the distinction between, on the
one hand, the professional teacher of literature in the academies and those who are or
have been their students and, on the other hand, those who merely read from interest
and are happy to have others guide them where they feel out of their depth. However,
there are no rules that confer upon one group an absolute authority to take decisions
about the canon.


This comes out in the other limitation on the authority that dictates the literary canon:
the limitation of scope. The construction of a canon is an attempt to answer the question,

What shall the individual who still desires to read attempt to read, this late in
history? The Biblical three-score years and ten no longer suffice to read more
than a selection of the great writers in what can be called the Western tradition,
let alone in all the world’s traditions. Who reads must choose, since there is
literally not enough time to read everything, even if one does nothing but read.
Mallarme’s grand line—“the flesh is sad, alas, and I have read all the books”—
has become a hyperbole. Overpopulation, Malthusian repletion, is the
authentic context for canonical anxieties.
(Bloom 1995, 15)

The catalogue of authors and works that constitute a literary canon is selective but not
exclusive in the way that a canon based on authenticity is. It is a selection from many
possible works and the list that is constructed is based on the variously founded prefer-
ences of those constructing the lists as well as on compromises between their interests and
ideologies. Curtius’ account of how the medieval canon was constructed illustrates the point:

About 890 Notker Balbulus rejects the pagan poets and recommends the
Christians Prudentius, Avitus, Juvencus, and Sedulius. On the other hand, a
century later students in the cathedral school at Speyer were reading “Homer,”
Martianus Capella, Horace, Persius, Juvenal, Statius, Terence, Lucan; and of the
Christian writers only Boethius. If we go a century further we come upon Winrich,
teacher in the cathedral school at Treves (ca. 1075). To his sorrow he was demoted
from the school to the kitchen. He utters his complaint in a poem which also
contains a catalogue of authors. Of pagans there appear Cato, Camillus (?), Tully,
Boethius, Lucan, Virgil, Statius, Sallust, and Terence. These nine are matched by
nine Christians: Augustine, Gregory, Jerome, Prosper, Arator, Prudentius,
Sedulius, Juvencus, Eusebius. . . . We find them in Conrad of Hirsau too, but
increased by Theodulus’ eclogue. Reviewing Conrad’s catalogue of authors we see
that Nos. 1–4 constitute a class by themselves – beginners’ reading. Nos. 5–10 are
the Christian poets. There follow three prose writers, one of whom is the Christian
Boethius, then the pagan poets, except that Terence is replaced by Ovid. . . . The
effort to establish a balance between Christians and Pagans is evident. It is a
deliberate plan of study: from the best of the pagan and Christian canons a
medieval school canon is formed. It remains the skeleton for the very greatly
enlarged catalogues of the thirteenth century.
(Curtius 1953, 260–1)

In the making of this canon there are two different forces at work: the force of authority,
which makes a selection based on interests, ideology, and compromise (“The effort to
establish a balance between Christians and Pagans is evident”), and the force of a
standard that are combined with these interests: “From the best of the pagan and
Christian canons a medieval school canon is formed” (my emphasis). The literary canon
is open in a way in which the scriptural canon is not.
In this concept of canon, the reference to a standard, to a canon in the more funda-
mental sense of “A general rule, fundamental principle, aphorism, or axiom governing

Stein Haugom Olsen

the systematic or scientific treatment of a subject; e.g. canons of descent or inheritance;

a logical, grammatical, or metrical canon; canons of criticism, taste, art, etc.” (Oxford
English Dictionary Online, “Canon”, entry 2b), replaces the criterion of authenticity as
the criterion for inclusion in the canon. The canon should comprise what is ‘best’. The
logical basis for pronouncing a work to be a canonical text of the scripture was genetic:
it had to be authentic (‘genuine and inspired’). Works are included in the literary canon
on the basis that they conform to a standard. Thus in the identification of a literary
canon there is not the same close connection between authority and the condition for
inclusion (authenticity in the sense of being ‘inspired’) as there is in the identification
of a canon of scripture. The standard of judgement and the values which a canon of
literature is intended to exemplify can be recog­nized without reference to authority. And
if the authoritative body underwriting the selection of canonical texts fails to make clear
“the nature of what [is] prescribed”, what values the canonical set of texts are aimed at
transmitting and why these values are important, it can be criticized for failing to have
a defined purpose in establish­ing a list of canonical texts:

English had become a “prescribed” study in schools and colleges, and a canonical
set of texts had been estab­lished, but this did not make the nature of what was
prescribed precisely clear. The fixing of the canon did not guarantee it would
be taught in a way that effective­ly transmitted a coherent body of values.
(Graff 1987, 100)

A literary canon in the sense of “a sanctioned or accepted group or body of related

works” is what Alastair Fowler calls ‘elastic’. “The literary canon”, he says, “varies obviously—
as well as unobviously—from age to age and from reader to reader” (Fowler 1982, 213),
and, one could add, from one nation to another and from one social group to another.
This elasticity has various sources. Because it is partly determined by a group of people
with a set of interests, tastes, goals, and ideologies exercising authority, any canon in this
sense will show traces of these interests, tastes, goals, and ideologies. These interests etc.
will manifest themselves on two levels. Given a particular standard they will, within the
parameters defined by the standard, influence which works and authors will appear in
the catalogue of authors worth reading. This is what happens when the medieval
school canon is formed: “From the best of the pagan and Christian canons a medieval school
canon is formed.” These interests etc. will have all the more influence (and the elasticity
of the canon in this sense will be all the greater) because the standards to which the
works in the canon are supposed to conform are not always clear and unambiguous. This
is the complaint registered by Gerald Graff in the last quoted paragraph. Or the standard
may be formulated in such general terms as permit a wide range of interpretations. The
present National Curriculum in Britain specifies under the heading “The English
Literary Heritage” that pupils should be taught

– how and why texts have been influential and significant [for example, the
influence of Greek myths, the Authorised Version of the Bible, the
Arthurian legends]
– the characteristics of texts that are considered to be of high quality
– the appeal and importance of these texts over time.
(The National Curriculum Online, English, Key Stage 3, En2 Reading)


The formulations are general and leave the nature of the standard to be applied
undetermined. What constitutes high quality remains undefined. It would not improve
matters much even if one qualified ‘high quality’ and required it to be ‘high aesthetic
quality’. The task of specifying high aesthetic quality would still remain and would allow
interests, tastes, goals, and ideologies considerable influence.
However, the interests, tastes, ideologies, and goals of a group of people will also
influence the choice of standard to be adopted. A standard that demands “a vital capacity
for experience, a kind for reverent openness before life, and a marked moral intensity”
(Leavis 1960, 9) would pick out works that were different from a standard based on the
view that, “No artist has ethical sympathies. An ethical sympathy in an artist is an
unpardonable mannerism of style” (Wilde 2000, 3, “Preface”). Tolstoy in What Is Art?
applied the standard that only such works should be included in the catalogue of works
worth reading as “transmit religious feelings urging towards the union and brotherhood
of man” or “feelings that may unite everyone without exception” (Tolstoy 1930, 265).
He was left with very few authors and works on his list.
An even more important reason for the elasticity of the canon as “a sanctioned or
accepted group or body of related works” is that the choice of the standard to which the
authors on the canonical list must conform, will be determined by the immediate prac-
tical purposes that the catalogue is aimed at serving. “In Antiquity”, says Curtius,
referring back to the passage in Aulus Gellius where the term ‘a classic’ was first intro-
duced: “the concept of the model author was oriented upon a grammatical criterion, the
criterion of correct speech” (Curtius 1953, 250).
On the other hand, the Middle Ages, again according to Curtius, sought in their
‘auctores’ technical information, “worldly wisdom and general philosophy” compressed
into sententia, and descriptions of human excellence and weakness, exempla. The two
standards invoked in the two different periods necessarily produced very different canons.
As was pointed out above, the authority of any group attempting to define a canon in
the sense of “a sanctioned or accepted group or body of related works” is a weak form of
authority with all that implies of endless squabbles and quarrels about what the canon is
and what works are to be accepted into the canon. Consequently, this concept of canon
is a contested concept. However, contested concepts like this present no real theoretical
or philosophical problem. In analytic philosophy a usual way to deal with this kind of
concept has been conceptual clarification. Contested concepts are contested because
they are radically confused. Conceptual clarification will reveal the source of confusion
and of disagreement between contesting parties or it may reveal that there is no such
disagreement. In either case the contest ceases and the parties can continue to use the
concept but with the awareness that the other party is simply using a different concept.
If the contest nevertheless continues, it has its roots not in the concept but in a “head-on
conflict of interests or tastes or attitudes, which no amount of discussion can possibly
dispel” (Gallie 1964, 157). This latter seems to be the case with this concept of canon. It
presents no theoretical problem, only a problem of clarification. Once this clarification
is undertaken, the elements of the concept are easily isolated embodying as it does a
reference to authority as well as to a standard of value. It is also elastic, enabling a theorist
to see why there are debates about the canon and why these debates are unlikely to be
settled: the contest over the concept continues because there are conflicting aims, pur-
poses, and objectives, as well as interests and ideologies, governing the selection of the
works that are accepted into the canon, and because there can be many standards for
inclusion in the canon, the adoption of a standard depending on the purpose it serves.

Stein Haugom Olsen

However, this concept of canon is not what Gallie has called an “essentially
contested concept”. Essentially contested concepts belong “in academic terms . . . to
aesthetics, to political and social philosophy and the philosophy of religion” (Gallie,
157). As examples of this kind of concept Gallie discusses “the concepts of a religion,
of art, of science, of democracy and of social justice” (ibid.: 168). The disputes
focused on essentially contested concepts continue even after conceptual clarifica-
tion has taken place and are thus far similar to those disputes involving “head-on
conflict of interests or tastes or attitudes” which are focused on radically confused
concepts. However, disputes centred on essentially contested concepts do not involve
conflicts of interests or tastes or attitudes. These disputes are “perfectly genuine” and
“although not resolvable by argument of any kind, are nevertheless sustained by
perfectly respectable arguments and evidence” (ibid.: 158). Arguments in favour of
one use of an essentially contested concept, though it may fail to convince those who
interpret the concept differently to give up their use of the concept, will nevertheless
have a logical force recognised by all sides in the dispute. Though inconclusive, disputes
about these concepts are rational. This, however, is not the case with the concept of
canon under discussion. The disputes about the canon in this sense are inconclusive
because they are due to irreducible differences in interest, ideology, aims, purposes,
and objectives. Nevertheless, the concept of the literary canon as “a sanctioned or
accepted group or body of related works” has a certain, limited usefulness as a critical
instrument for certain pedagogical purposes, i.e. that of recommending a list of
authors as canonical and therefore worth reading, given a certain purpose and a
certain target audience.

One can finally distinguish a concept of literary canon which has a universal intension
(referred to in the following as the universalist concept of literary canon): it is equivalent
to the concept of literature itself. Just as the concept of literature, it has a descriptive as
well as an evaluative use. In its descriptive use it refers to “the entire written corpus”. In
its evaluative use it refers to everything worthy to be read:

The literary canon in the broadest sense comprises the entire written corpus,
together with all surviving oral literature. But much of this po­tential canon
remains in practice inaccessible for a variety of reasons, such as the rarity of its
records, which may be sequestered in large libraries. Hence the more limited
accessible canon. Accessible literature is very much narrower than the New
Cambridge Bibliography of English Literature might suggest.
(Fowler 1982, 214–5)
If we follow this argument wither it leads, we ac­cept as being literary not only
sociological or linguistic or historical or political inquiry, we also accept a very
broad conception of the canon of literature—one closer to the conception that
prevailed before the late nineteenth century. Under that older conception,
literature comprises every­thing worthy to be read, preferably the best tho­ughts
expressed in the best manner, but above all the best thoughts.
(Hirsch 1976, 140)


In its descriptive use the concept loses all connection with the concept of a canon as “a
sanctioned or accepted group or body of related works” as well as with the notion of a
scriptural canon. There is no reference to authority, no reference to authenticity and
there is no criterion for constructing a catalogue of authors “worthy to be read”. In its
evaluative use it retains the criterion for exclusion, “worthy to be read”, but the criterion
of worth employed does not make any reference to aesthetic value.
This concept of ‘literary canon’ has no theoretical or practical point. ‘Canon’,
in this use, is merely an alternative expression for ‘literature’ and its introduction
into the vocabulary of literary studies does not add anything to the theorist’s or the
educator’s classificatory scheme. When Eagleton in the following passage tries to
give a theoretical characterization of literature he uses the two expressions

Literary theorists, critics and teachers, then, are not so much purveyors of doc-
trine as custodians of a discourse. Their task is to preserve this discourse, extend
and elaborate it as necessary, defend it from other forms of discourse, initiate
newcomers into it and determine whether or not they have successfully mas-
tered it. The discourse itself has no definite signified, which is not to say that it
embodies no assumptions: it is rather a network of signifiers able to envelop a
sole field of meanings, objects and practices. Certain pieces of writing are selec­
ted as being more amenable to this discourse than others, and these are what is
known as “literature” or the “literary canon”.
(Eagleton 1983, 201)

The problems, theoretical and practical, raised by this concept of ‘literary canon’ are the
same as those raised by the concept of ‘literature’. The question “Does this work belong
to the literary canon?” has to be answered in the same way as the question “Is this a
(good) work of literature?”
Related to the universalist notion of a literary canon, there is, however, a more lim-
ited notion equating ‘the literary canon’ with what Leavis with a memorable phrase
called “the great tradition”. The literary canon comprises those works that are paradig-
matically great literature, i.e. those works that by example set the standard for what great
literature is:

Above all, the reason why friend and foe are both able to do as they wish, and
to recognize that in the end their discourses are of the same sort, is that
Hamlet is unshak­ably canonical—the dislodgement of such a work from
canonical status would certainly involve the dislodgement of those discourses,
of whatever party.
(Kermode 1985, 62)

The role of a canonical text like Hamlet is to define the concept of literature, and if this
type of text was removed from the canon, it would destroy the concept of literature itself
and therefore any meaningful discourse about literature. However, again it is true that
this notion of literary canon adds nothing to the classificatory scheme of the theorist or
the educator. The problem of which works constitute the literary canon in this sense is
simply the problem of what is paradigmatically great literature.

Stein Haugom Olsen

The universalist notion of ‘literary canon’, then, has no function as a theoretical
instrument. At the same time, because of its roots in the notion of a scriptural canon,
employing it opens up considerable opportunity for confusion and also for manipulation.
In an argument where he makes use of a concept ‘canon of art’, a concept which in every
respect behaves like the universalist concept ‘canon of literature’, Frank Kermode puts
forward the view that Botti­celli, after a long period of neglect, was admitted to the
artistic canon in the late nineteenth century, not by becoming appreciated and
understood, but simply by becoming accepted by the ignorant leaders of taste:

Now firmly established in this new setting, Botti­celli was accorded a position
of eminence from which he was unlikely ever to be completely dislodged. He
owed his promotion not to scholars but to artists and other persons of modern
sensibility, whose ideas of history were more passionate than accurate, and
whose connoisseurship was, as I have said, far from exact. At this stage exact
knowledge had no part to play. Opinion, to some extent informed, required, at
this modern moment, a certain kind of early Renais­sance art; Botticelli, along
with some contempo­raries “though first among them” provided it. Enthusiasm
counted for more than research, opinion for more than knowledge.
(Ibid.: 6)

And Kermode concludes his argument:

Botticelli became canonical not through scholarly effort but by chance, or

rather by opinion.
(Ibid.: 30)

In this argument, for Kermode, “to become canonical” is simply to be ac­cepted by the
leaders of taste. It does not involve any es­tablishment of merit. According to Kermode,
it was only after Botticelli became available to scholars through being acciden­tally dis-
covered by the leaders of taste, that he got the attention due to canonical works and
became properly appre­ciated. Without considering its appropriateness Kermode here
applies a notion of ‘canon’ which is dependent upon autho­rity in the same way as is the
notion of ‘canon of scripture’. Kermode does not raise the centrally important question
whe­ther it is at all appropriate to say that Botticelli became a part of the canon (in the
universalist sense of ‘artistic canon’) before he came to be genuinely appreciated. This
uncritical use of the concept of artistic canon produces a contradiction in Kermode’s
argument in Forms of Attention. In the passage quoted at the end of the last section
Kermode states that the role of great works of art is to set a standard for what it is to be
canonical. What he says about the works of Botticelli is that his works became canonical
not by conforming to or setting a standard, but by chance. The contradiction arises
because the concept of canon which in most of its uses has as an element an implicit
reference to authoritative fiat leads Kermode to reason in a certain way about Botticelli.
Though the universalist concept of a literary canon is identical with the notion of ‘great
literature’ the associations of the expression ‘canon of literature’ invite the critic to take
up the view that the paradigmatic, great works of literature become paradigmatic by
authoritative fiat expressed in the opinions of the leaders of taste.


Employing the universalist notion of ‘literary canon’ not only opens up considerable
opportunity for confusion but also for manipulation.When the word ‘canon’ came into
currency in the last quarter of the twentieth century and became central in the discus-
sion of literary value, replacing expressions like ‘the great tradition’, ‘the great
monuments of literature’, ‘great literary works’ etc., it imported into this discussion
certain assumptions about the role of authority in constituting the notion of literary
value. The notion of a canon was imported into this discussion by those critics who
wanted to focus atten­tion on other works than those that are the agreed master­pieces
of the culture, and who did not want to do this by showing that these other works are
valuable in the way in which the masterpieces are valuable, but instead wanted to chal­
lenge the whole agreement about what these masterpieces should be. Note in the
following passage how the agreement about great and good works is set up for attack by
taking it for granted that this agreement constitutes a literary canon which rests on

The great author is great because (occasionally even she) has managed to convey
an authentic vision of life; and the role of the reader or critic is to listen respect-
fully to the voice of the author as it is expressed in the text. The literary canon
of “great literature” ensures that it is this “repre­sentative experience” (one
selected by male bourgeois critics) that is transmitted to future generations,
rather than those deviant, unrepresentative expe­riences discoverable in much
female, ethnic and working-class writing. Anglo-American feminist criticism
has waged war on this self-sufficient cano­nization of middle-class male values.
(Moi 1985, 78)

According to this kind of argument, a work survives initially for some accidental reason
that has got nothing to do with quality but rather with hegemonic power, and then
when it has been assigned a niche by that hegemonic power, the work itself ensures its
own canonicity by defining the values by which it will be judged:

[the work] will also begin to perform certain characteristic cultural function by
virtue of the very fact that it has endured—that is, the functions of a canonical work
as such—and be valued and preserved accordingly: as a witness to lost innocence,
former glory, and/or apparently persistent communal interests and values and thus
a banner of communal identity; as a reservoir of images, archetypes, and topoi—
characters and episodes, passages and verbal tags—repeatedly invoked and
recurrently applied to new situations and circumstances. In these ways, the canon-
ical work begins increasingly not merely to survive within but to shape and create the
culture in which its value is produced and transmitted, and, for that very reason, to
perpetuate the conditions of its own flourishing. Nothing endures like endurance.
(Smith 1988, 50)

Critics such as as Moi and Herrnstein Smith need the con­cept of literary canon because
it enables them to introduce the necessary first step in their argument: that a work
becomes part of the tradition of great literature through the exercise of arbitrary author-
ity ‘canonizing’ that work. This in its turn provides the necessary foundation for
attacking the ‘repressive’ authority (“male bourgeois cri­tics”) that dictates the canon.
Any discussion of the standards which a work has to conform to in order to be admitted

Stein Haugom Olsen

into the canon as well as of the validity of these standards becomes unnecessary because
literary value is seen as “radically contingent”. One also avoids discussion of the nature
of the value that literature confers on the culture of which it is a part. That is, one is cut
off from posing and discussing the crucial question of whether there are any reasons why
works of art do not endure and, if so, what are these reasons?

The fact that the concept of the canon lends itself to these kinds of misunderstanding
and manipulation makes it singularly unsuitable as a critical instrument in literary studies.
The point can be reinforced by a brief look at the concept of ‘tradition’, which the concept
of ‘canon’ to a large extent replaced. Again the Oxford English Dictionary is helpful:

A long established and generally accepted custom or method of procedure,

having almost the force of a law; an immemorial usage; the body (or any one)
of the experiences and usages of any branch or school of art or literature, handed
down by predecessors and generally followed.
(Oxford English Dictionary Online, “Tradition”, entry 4a)

The notion of ‘literary tradition’ has four important elements that are absent from the
concept of ‘canon’. The notion of tradition is tied to the notion of practice, to the notion
of a way of doing things, a way of writing, a way of painting, a way of reasoning that has
built into it a set of standards and a notion of skill. The great works of a tradition are great
not because they are pronounced to be great, but because they display to a high degree
the required skill and meet the requirements of the tradition in an exemplary manner.
They have a high degree of merit as defined in and through the tradition.
Second, a tradition has continuity: it is handed down. The notion of tradition captures
the continuity as well as the development that is constituted not only through the
similarities and differences between literary works since Homer, but also through the act
of authors of all periods placing themselves self-consciously in relations of opposition
and/or discipleship to earlier writers as well as to writers contemporary with themselves.
Since Homer, authors have made reference either in their prac­tice or in their reflections
on their practice, back to what previous authors have done. One does not have to go
down “the anxiety of influence” route that Harold Bloom once took (Bloom 1997) to
recognize how poorly a concept like ‘intertextuality’ captures the phenomenon of literary
influence. Already for Roman authors, with their sense of belatedness, the notion of
‘imitation’ acquired a second meaning so well captured by Pope:

When first young Maro in his boundless mind 

A work t’ outlast immortal Rome design’d,
Perhaps he seem’d above the critic’s law,
And but from Nature’s fountains scorn’d to draw;
But when t’ examine ev’ry part he came,
Nature and Homer were, he found, the same.
. . . 
Learn hence for ancient rules a just esteem;
To copy Nature is to copy them.
(Pope 1903, An Essay on Criticism, Part I, ll. 130–40)


Third, a tradition is anonymous, an ‘immemorial usage’: no named authority is

responsible for or can create a tradition. A tradition develops. It may change and if it
changes beyond recognition it will have disappeared. However, it cannot be changed
by authoritative fiat as can a ‘canon’. A cultural practice like the literary tradition has
its own life and cannot be steered through arbitrary authority. As Raymond Williams
once remarked,

We have to plan what can be planned according to our common decision. But
the emphasis of the idea of culture is right when it reminds us that a culture,
essentially, is unplanable. We have to ensure the means of life, and the means of
community. But what will then, by these means, be lived, we cannot know or say.
(Williams 1963, 320)

When a cultural tradition, such as the literary tradition, is changed through authoritative
fiat, and this has been attempted, it is the authority of those making the attempt that is
going to be discredited, as were the French Classicist critics who constructed a canon
that could find no room for Shakespeare or for new literary forms like the novel. Or
closer to home: the failure of the Marxist attempt fundamentally to change the category
of literature:

It is significant that “Marxist criticism” and “Marxist literary studies” have been
most successful, in ordinary terms, when they have worked within the received
category of “literature”, which they may have extended or even revalued, but
never radically questioned or opposed. By contrast, what looked like fundamental
theoretical revaluation, in the attempted assimilation to “ideology”, was a
disastrous failure, and fundamentally compromised, in this whole area, the
status of Marxism itself.
(Williams 1977, 53)

Finally, and this is an element which is not captured by the OED definition but which
again appears in the Merriam-Webster Dictionary, the notion of tradition is linked to the
notion of culture: a tradition is a “cultural continuity in social attitudes, customs, and
institutions” (Merriam-Webster Online Dictionary, “Tradition”, entry 3) Traditions are
culturally embedded and are by their nature local and culture specific. Different cultures
and different language communities have different literary traditions closely linked to
what is perceived as the identity of a culture and of a language community. That is why
literature sometimes has played such an important role in the definition of national
identity at times of struggle for political and cultural independence. It is also one of the
reasons for the continuity and stability of the literary tradition: it is one of the identity
markers of a culture. However, different traditions can be culture specific and neverthe-
less be the same kind of tradition involving the same types of skill and the same standards.
To what extent they actually do, is an empirical question, and if traditions become too
different in the demands they make on their practitioners, they will no longer be the
same kind of tradition.
Employing the concept of tradition does not settle the problem of what is a good
or a great literary work. Nor does employing it settle the question of whether literary
value is radically contingent. Nor does its settle any question concerning the value of
literary practice as such. But employing it does open up for a discussion of all these

Stein Haugom Olsen

topics and does not foreclose avenues of inquiry in the way that employing the
concept of a literary canon does.

Bloom, Harold. 1995. The Western Canon. The Books and School of the Ages. London: Macmillan. Original
edition, 1994.
Bloom, Harold. 1997. The Anxiety of Influence. A Theory of Poetry. 2nd ed. Oxford: Oxford University Press.
Original edition, 1973.
Curtius, Ernst Robert. 1953. European Literature and the Latin Middle Ages. Translated by Willard R. Trask.
London: Routledge and Kegan Paul.
Eagleton, Terry. 1983. Literary Theory. An Introduction. Oxford: Blackwell.
Fowler, Alastair. 1982. Kinds of Literature. An Introduction to the Theory of Genres and Modes. Oxford:
Clarendon Press.
Gallie, W.B. 1964. Philosophy and the Historical Understanding. London: Chatto and Windus.
Graff, Gerald. 1987. Professing Literature. An Institutional History. Chicago: University of Chicago Press.
Hirsch, E.D. 1976. The Aims of Interpretation. Chicago: University of Chicago Press.
Kermode, Frank. 1985. Forms of Attention. Chicago: University of Chicago Press.
Leavis, F.R. 1960. The Great Tradition. New ed. London: Chatto & Windus. Original edition, 1948.
Merriam-Webster Online Dictionary, available at: http://www.merriam-webster.com (accessed 10 September 2015).
Moi, Toril. 1985. Sexual/Textual Politics. Feminist Literary Theory. London: Methuen.
The National Curriculum Online, English, Key Stage 3, En2 Reading: http://www.nc.uk.net/webdav/servlet/
id=6209&POS[@stateId_eq_note]/@id=6209 (accessed 4 March 2005).
Oxford English Dictionary Online, available at: http://www.oed.com (accessed 10 September 2015).
Pope, Alexander. The Complete Poetical Works. Ed. by Henry W. Boynton. Boston and New York: Houghton,
Mifflin & Co., 1903; Bartleby.com, 2011. www.bartleby.com/203/27.html, #130–40 (accessed 27 April 2015).
Sampson, George. 1961. The Concise Cambridge History of English Literature. 2nd ed. Cambridge: Cambridge
University Press.
Smith, Barbara Herrnstein. 1988. Contingencies of Value: Alternative Perspectives for Critical Theory.
Cambridge, MA: Harvard University Press.
Tolstoy, Leo. 1930. What Is Art? and Essays on Art. Translated by Aylmer Maude (1900). Oxford: Oxford
University Press. Original edition, 1896.
Wilde, Oscar. 2000. The Picture of Dorian Grey. London: Penguin Books. Original edition, 1891.
Williams, Raymond. 1963. Culture and Society, 1780–1950. Harmondsworth: Penguin Books. Original
edition, 1958.
Williams, Raymond. 1977. Marxism and Literature. Oxford: Oxford University Press.

Further Reading
Alter, Robert. 2000. Canon and Creativity. Modern Writing and the Authority of Scripture. New Haven: Yale
University Press.
Hallberg, Robert von, ed. 1984. Canons. Chicago: University of Chicago Press.
Harris, Wendell V. 1991. “Canonicity”. PMLA no. 106 (1): 110–21.
Morrissey, Lee. 2005. Debating the Canon: A Reader from Addison to Nafisi. Houndmills: Palgrave Macmillan.
Olsen, Stein Haugom. 2007. “Why Hugh MacColl Is not, and Will never Be, Part of any Literary Canon”.
In The Quality of Literature, edited by Willie van Peer, 31–51. Amsterdam: John Benjamins.
Peer, Willie van. 1996. “Canon Formation: Ideology or Aesthetic Quality?”, British Journal of Aesthetics
no. 36 (2): 97–108.

Part III

Matthew Kieran

Characterizing Literary Creativity

What is literary creativity? Philosophical and literary tradition holds that the following
two conditions must be met: novelty and value. There is, unsurprisingly, a large amount
of controversy about just how both conditions should be spelt out. Nonetheless, there
is a surprising amount of agreement that both conditions are required (Attridge 2004,
Boden 2004, Carroll 2003, Gaut 2003, O’Quin and Besemer 1999, Stokes 2008).
Novelty is necessarily a relational term, being a matter of newness with respect to
something that has gone before, whilst (literary) value is at least partly relational,
concerning how we appreciate and value literature.
It might seem as if the conjunction of the novelty and value conditions yields what
it is for a work to be original. Yet we should be careful not to define creativity in terms
of originality, given that writers can be creative without being original. Whilst originality
may be a high-end creative achievement, literary creativity does not as such entail
originality. Many novels or poems might not be particularly original – consider much
genre fiction – and writers sometimes plagiarize or pastiche for literary effect. It is helpful
here to distinguish between psychological and historical creativity (Boden 2004).
Psychological creativity, Boden argues, is a matter of coming up with something valuable
that is ‘new to the person who comes up with it’ (Boden 2010: 30). Hence we can judge
a writing student’s poem to be psychologically creative without thereby being committed
to any claim about originality. Historical creativity, Boden suggests, is psychological
creativity that stands in a special relation to what else has been done: it is a matter of
coming up with something that is valuable, new to the person who comes up with it and
has never been thought or done by anyone before, i.e. ‘has arisen for the first time in
human history’ (Boden 2010: 30).
The distinction between psychological and historical creativity is stark and perhaps
overly demanding in the latter case. We might more profitably conceive of Boden’s
distinction as two ends of a spectrum along which judgements of creativity exhibit a
huge amount of context sensitivity and degree (Meskin unpublished manuscript). What
is judged to be creative when comparing first-year creative writing students is rather
different from what is required to be creative, inventive or even original for an already
established literary author. Even in the case of established authors what standards and
background comparators are appealed to in ascriptions of creativity will depend upon
the relevant context. In a crime fiction review it might plausibly be claimed that James
Ellroy is one of the most original contemporary writers – assuming the comparison class
is that of crime fiction – and yet elsewhere it may be right not to put him in the same
class as writers such as Philip Roth, Margaret Atwood, or J. M. Coetzee.
Matthew Kieran

More controversially the presumption that novelty is required for a work to be

creative might be setting the bar too high, at least where novelty is supposed to be a
significant, substantial, non-trivial condition. To create in the most minimal sense just
is to bring something into existence and we can be more or less creative with respect to
how we do so. Imagine clicking the mouse on a computer that runs a software program
(let us call it ‘Author Author’) resulting in the printer churning out a poem. Even if the
poem is a decent one, it is not a particularly creative way to write one (at least you are
not specially creative in doing so given you only clicked the mouse and ran the program).
By contrast an author may sit down and skillfully bring about different variations on the
same kind of imaginative poem or romance novel time and time again. Writers can be
creative in a minimal sense just in virtue of the skillful, imaginative ways in which they
write what they do, even where this does not involve doing anything significantly or
saliently new. Angela Thirkell, to take one example, set many of her novels in Anthony
Trollope’s fictional county of Barsetshire and lifted names, characters, plots, and devices
from other writers such as Gaskell, Dickens and Galsworthy. In some respects Thirkell’s
novels are not hugely dissimilar to contemporary fan fiction, where literary enthusiasts
write stories in the same fictional universe created by their favoured authors – Tolkien’s
Middle Earth for example – or using characters from works they admire – Darcy from
Austen’s Pride and Prejudice. More conventionally works in standard genre fiction such
as romance, chick lit, horror, science fiction, crime, or fantasy are often not particularly
novel in any interesting sense. This is part of what is taken to characterize something as
mere genre fiction as opposed to literary fiction. Nonetheless this does not show that
genre writing is uncreative tout court. A work of standard genre fiction may well not be
as creative as other more literary works (though this is not necessarily the case) and
we may often value the uniqueness, novelty, and originality in literary fiction much
more than we value the skillful, imaginative realization of genre fiction. Yet we should
be careful to avoid constitutively defining creativity as such in terms of the kind of
substantial novelty or originality we most value in literary fiction.
Literary creativity is not just a matter of end product or outcome success. As we have
just seen, creativity is also a matter of how something came about. Running the
computer program ‘Author Author’ is one thing, sitting down to write a story yourself
or with others is another. The person or group involved must be responsible for the
relevantly surprising, novel, or valuable features of the work in the right kind of ways.
Consider a thought experiment inspired by Borel’s infinite monkey theorem. Borel’s
hypothesis holds that a monkey hitting typewriter keys for an infinite length of time
would almost surely produce – amongst an infinite range of rubbish and other works –
the complete works of Shakespeare. Presumably the monkey would also produce novel,
valuable works, the like of which had not been read before. Yet we would not thereby
hold that the monkey is creative in any deep or interesting sense. The monkey has made
something that is historically novel and valuable in literary terms (along with mostly
rubbish nonsense). Nonetheless the relation between what the monkey is doing,
randomly hitting typewriter keys, and why, maybe hitting typewriter keys is fun, is
entirely accidental to the production of literary masterpieces. In literary terms the
monkey has no idea what it is doing. Attributing creativity to an agent presumes that
there is a non-accidental relation between what someone is doing, the intentional
description under which the act is performed, why, and the nature of the end result
(Stokes 2008, Gaut 2009, Kieran 2014a). Creativity thus involves something like


a relevant purpose (in not being purely accidental), some degree of understanding
(not using purely mechanical search procedures), a degree of judgement (in
how to apply a rule, if a rule is involved) and an evaluative ability directed to
the task at hand.
(Gaut 2010: 1040)

This is not to claim that creativity requires a writer to know or plan absolutely everything
beforehand. Writers sometimes only have a pretty vague aim in view or start from the
most minimal images, associations, or phrases in committing pen to paper. What matters
is that the author’s agency is exercised in making the relevant literary choices and giving
form to the work. This is consistent with the recognition that writers often experiment
by introducing accidental or non-conscious elements into the creative process (Gaut
2010). Whilst cases vary, what renders the use of accidental or non-conscious elements
praiseworthy aspects of a writer’s creative process – as opposed to mere serendipity –
will depend upon a variety of factors including (a) the author’s willingness to use or
experiment with such elements for literary reasons, and/or (b) doing so in a manner
that draws upon automated processes related to the author’s literary expertise, and/or
(c) requires the author’s expertise in appraising the results and making use of them in
interesting and valuable ways.

Types of Creativity, Canonicity and Traditions

Literary creativity is a function of the agential processes that bring the text about, the
literary values realized in or through the text and the relations in which the resultant
work stands to other literary works. Adopting Boden’s (2004; 2010) tripartite taxonomy
of kinds of creativity to literature would give us the following:

1. Combinatorial creativity. This kind of creativity involves the unfamiliar recombination

of familiar ideas and devices. George Orwell’s Animal Farm, for instance, draws on
a host of familiar concepts about politics, human beings and animal behaviour to
recombine them in unfamiliar ways in the service of standard allegorical ends.
2. Exploratory creativity. This amounts to exploring the conceptual spaces of the
structured style of thought being worked within. In literary terms this would
include developing some of the possibilities foreshadowed by other works or
authors or better realizing the literary potential and effects implicit in a given
literary style, genre, form, or structure. Moliere’s comedies, for example, refined
pre-existing conventions in comedies of manners to explore dramatic characteriza-
tion for satirical ends.
3. Transformational creativity. The transformation of conceptual space and literary
devices such that someone ends up writing a work that could not have been written
prior to such a transformation. Thus, for example, the introduction of the epistolary
novel in the seventeenth century or stream-of-consciousness writing in the early
twentieth century gave rise to radically new literary possibilities that subsequent
novels went on to exploit. Rather than explore the potentiality of previous forms
and conventions, the transformative creativity of writers such as Samuel Richardson
or James Joyce, respectively (amongst others), gave rise to new literary forms.

Matthew Kieran

Boden’s basic categorization is open to challenge or refinement. Novitz (1999), for

example, argues that transformations of conceptual space need not amount to radical
creativity. Alternatively, if creativity need not require significant novelty, we may need
a category that is more modest than either combinatorial or exploratory creativity to
capture the ways in which works can be modestly creative just in virtue of the skillful,
imaginative realization of the same kind of work over and over again (as is sometimes
found in mere genre fiction). Challenges such as these may point toward the need for a
richer basic taxonomy. Carroll (2003), for example, outlines fives basic types of creativity
which may be articulated in relation to literature as follows: (1) repetition with vari-
ation, either with respect to literary structures or themes found in previous traditions;
(2) hybridization, where a literary work yokes together two or more elements from
distinct styles, genres, or traditions; (3) interanimation, which involves bringing devices,
strategies, or values from one art form into another; (4) amplification, which is a matter
of enlarging the nature and resources of a tradition through developing new solutions to
enduring problems or projects within that tradition; and (5) revolutionary creativity,
which rejects fundamental aspects of literary practice, devices, style, movements, values,
or tradition in order to reconfigure literature (though usually this is achieved in part by
foregrounding relations to more distant or seemingly foreign traditions).
Whatever the most useful basic taxonomy is, and indeed by which criteria particular
works fall under one category rather than another, may partly depend on the uses to
which we want to put such categorizations. Nonetheless one striking feature of such
taxonomies is the role that relations to other literary works and traditions play. In order
to identify whether a work or literary movement is combinatorial, amplificatory, trans-
formative, or revolutionary, we have to know how the relevant work(s) stand in relation
to other works and movements.
One way of cashing this out is by appeal to the notions of canonicity and tradition.
A ‘kanon’ in Ancient Greece was a measuring rod and a literary canon (or canons) can
be thought of analogously. Literary canons can be thought of as being constituted by the
literary classics that pass the test of time (Hume 1993) and afford the touchstones in
light of which we appreciate and evaluate literary creativity. Characterizing literary
creativity (or the lack of it) partly in terms of the literary canon has a strong explanatory
appeal. Authors or critics often characterize the nature and value of even radically
transformative literary works in terms that commonly refer to the nature and elements
of literary classics. Yet the idea of authoritatively prescribed (and proscribed) works may
be problematic. As Olsen (2009) argues, the idea of canonical lists makes sense in theo-
logical or legalistic contexts, but by what authority can someone prescribe particular
literary works to be central touchstones (or peripheral ones come to that) for literature?
There are all kinds of writers and readers from all over the globe and it is far from obvious
that particular kinds of readers hold authoritative sway over others. Thus it may be more
appropriate to think in terms of canons informed by value and distinct literary traditions.
Indeed it may be better, as the poet T. S. Eliot would have us believe (1919), to think
that novelty and literary value only make sense in relation to tradition. What matters
then, presumably, would be (a) how authors are cultivated into particular traditions in
terms of how to write and the devices, genres and forms used; (b) the relations in which
newly created works stand to others within and between literary traditions; (c) how
traditions develop including reconfigurations and transformations brought about by the
new inter-relations brought about by new works amongst the old. Eliot, we might say,
conceived of the shock of the new as necessarily the renewal of living literary tradition.


None of this is to claim that literary tradition must be preserved in aspic. Tradition
is the starting framework into which writers are typically cultivated – through learning
and engaging with the devices, conventions, and structures that work – and which
thereby set the initial terms and horizons of creative possibilities. Healthy literary traditions
continually remake themselves through engaging with new creative challenges and
possibilities in the ways indicated above.
Here we can usefully distinguish between a weaker and a stronger claim with
respect to the role of tradition. The strong claim would be that literary creativity
requires some kind of relation to tradition (even in radical cases) in order to make
sense. This view renders tradition as essential to literary creativity and value. A more
modest claim holds that whilst tradition standardly sets the literary terms and framework
from which writers work, working within or from some tradition is not as such required
for someone to produce a literary work. The weaker thesis has the advantage of making
sense of the important role that tradition plays, whilst also recognizing that outsiders
can nonetheless produce highly creative literature. The weaker view may have an
added advantage in being able to account for how literary traditions themselves must
once have got going in terms of the emergence of literary creativity from other non-
literary traditions. The stronger view, by contrast, must argue for a theoretical mutual
inter-dependence between the notions of literary creativity and tradition (where it
would be pointless to ask which ‘got going’ first). Nonetheless both the strong and
weak claims seem to be in principle consistent with Virginia Woolf’s bracingly open
attitude towards literary creativity:

Let us bear in mind a piece of advice that an eminent Victorian who was once
also an eminent pedestrian once gave to walkers: ‘Whenever you see a board up
with “Trespassers will be prosecuted”, trespass at once.’
Let us trespass at once. Literature is no one’s private ground; literature is
common ground. It is not cut up into nations; there are no wars there. Let us
trespass freely and fearlessly and find our own way for ourselves. It is thus that
English literature will survive this war and cross the gulf – if commoners and
outsiders like ourselves make that country our own country, if we teach ourselves
how to read and how to write, how to preserve and how to create.
(Woolf 1940: 181)

Literary Genius and the Sleep of Reason

Perhaps the single most famous origination story of literary creativity is to be found in
Coleridge’s preface to Kubla Khan. Alone and ill Coleridge describes himself taking a
prescribed ‘anodyne’ (a medicinal draft that was most likely opium) causing him to fall
asleep whilst reading the following passage from Purchase’s Pilgrimage:

Here the Khan Kubla commanded a palace to be built, and a stately garden
thereunto. And thus ten miles of fertile ground were inclosed with a wall.

In a deep slumber Coleridge experiences a fantastical dream ‘in which all the images rose
up before him as things, with a parallel production of the correspondent expressions,
without any sensation or consciousness of effort’ (Coleridge 1816: 52).

Matthew Kieran

Coleridge even says that whilst in the dream he has the sense of composing at least
two to three hundred lines and, upon awakening, busily sets to transcribing the poem
down. Fatefully, in the midst of writing, someone from Porlock, a nearby town, comes
knocking to interrupt him on business. After the visit, Coleridge returns to write only
to find that ‘with the exception of some eight or ten scattered lines and images, all the
rest had passed away like the images on the surface of a stream in which a stone has been
cast’ (Coleridge 1816: 53).
The end result is the incomplete fragment of fifty-four lines that constitute Coleridge’s
most famous poem, opening, as it does, with the lines:

In Xanadu did Kubla Khan

A stately pleasure-dome decree:
Where ALPH, the sacred river, ran
Through caverns measureless to man
(Coleridge 1816)

Here, in one episode, we have the encapsulation of the Romantic conception of literary
genius; one that can be disaggregated into a number of constituent parts with distinct
philosophical precedents.
The poem putatively originates from a dream prompted by Coleridge’s reading as he
drifted off into sleep, with the unconscious forces of his mind working themselves out
from who knows where and partly fuelled by opium. The idea that poetic labours are the
result of unconscious irrational or non-rational inspiration reaches as far back as Plato’s
Ion. Early on in Plato’s dialogue, Socrates compares poets to prophets thus:

For all good poets, epic as well as lyric, compose their beautiful poems not by
art, but because they are inspired and possessed. . . . For the poet is a light and
winged and holy thing, and he has no ability to create until he has been inspired
and is out of his senses, and reason is no longer in him (for absolutely no man,
while he retains that faculty, can make poetry or prophesy).
(Plato [380 BC] 2001; 10–11)

Divine muses inspire and possess the poets who are thus mere vessels for the ideas visited
upon them. The emphasis on irrational or non-rational aspects of literary creativity,
albeit typically in less supernatural form, runs throughout various Western intellectual
traditions and can be found one way or another in the work of Nietzsche (1872), Freud
(1907), and Koestler (1964) to name but a few. A recurring claim in such authors
(though put rather differently) is the idea that unconscious desire, pattern association,
or recognition requires the rational, self-conscious editing aspect of the mind to be in
abeyance in order to bypass certain mechanisms whether that be repression or self-
conscious sense making (see Gaut 2012). Writers sometimes seem to have ideas or
images pop up unbidden into the mind’s eye or feel as if the characters in a fiction sud-
denly start to ‘write themselves’. Moreover, writers sometimes seek out ways of damping
down or bypassing rational, conscious thought in order to facilitate inspiration (ranging
from processes such as automatic writing, distraction, or ‘sleeping on it’ to using drink
and drugs).
There is also a significant body of work that suggests that literary writers (and in
particular poets) are more prone to mental illness. Aristotle (or one of his followers,


Theophrastus) observed that the preeminent in poetry and the arts are especially subject
to melancholia and contemporary psychology suggests something similar. Jamison
(1993) studied forty-seven contemporary writers and found that 38 per cent had under-
gone significant treatment for mood disorders (i.e. depression). Her analysis of thirty-six
major historical poets from 1705 to 1805 revealed a broadly consonant pattern, with
thirteen retrospectively diagnosed as probably being bipolar I and six probably bipolar
II or cyclothymic. Ludwig’s (1995) study of over 1,000 leading cultural figures, including
writers, found that 87 per cent of poets, 77 per cent of fiction writers, 51 per cent of
social scientists and 28 per cent of natural scientists suffered from some kind of mental
disorder. There is much that is controversial about the state of the evidence and, even
granting empirical claims, what we should take it to show (Gaut 2012; Kieran 2014b).
In virtue of what specific states or mechanisms do such conditions enhance someone’s
creative capacity? Various possibilities have been suggested ranging from idea generation
and ruminative evaluation to mood enhancement effects. And, we might also ask, in
virtue of what do such conditions tend to diminish creativity? Depression and anxiety
can be crippling conditions for writers as opposed to enhancements.
Whether prompted by depression, sleep, or opium, Coleridge’s non-rational inspira-
tion is a fantastical vision which is in turn expressed in literary terms on the page. It is
neither particularly clear what the poem means, nor is the poem written in the then
standard rhyming iambic pentameter (it is, rather, an admixture of tetrameter and pen-
tameter). The poem’s form may not strike us as particularly unconventional today,
especially by comparison with free verse, but in Coleridge’s time regularity of rhyme and
metre were deemed essential to good poetry. It is no surprise then that Kubla Khan met
with much critical hostility at the time, William Hazlitt, for one, judging Kubla Khan to
show only that ‘Mr. Coleridge can write better nonsense verses than any man in
England. . . . it is not a poem, but a musical composition’ (2002 [1816]: 208). Here we
have the image of Coleridge as the literary genius, ahead of his time and critical con-
temporaries, experimenting with literary form in the service of a bewitching vision that
means we know not quite what.
According to Kant genius is ‘the innate mental disposition through which nature gives
the rule to (beautiful) art’ (Kant: S46). A genius is not just someone who produces some-
thing that is valuable as literature or art. Rather the literary genius is one who has a
natural talent to go beyond that which can be merely imitated or taught, i.e. beyond mere
craft in producing distinctively original works. Genius is not rule bound. Literature and
art more generally, unlike science, arise from the harmonious interplay of the writer’s
imagination and understanding in giving literary form to a work. In science, by contrast,
we can, in principle, fully grasp the concepts and attendant chains of reasoning which
culminate in discoveries about the natural world. In literature and art more generally,
genius creates new associations or concepts but is not determined by them (Kant: S49).

Creativity and the Craft of Literature

A contrasting conception conceives of literary creativity as being akin to a rational
craft-like activity. Writing literature, wherever inspiration may come from, is, like many
other artistic activities, a process of problem solving. What kind of interest does the
writer want to evoke in the central character? Why, if at all, should we care about the
central protagonist(s)? What is the best way to explore the literary theme? In some ways
Aristotle’s Poetics can be seen as a kind of recipe book for what writers should aim for in

Matthew Kieran

writing the best kind of tragedy. We may be skeptical of the details of Aristotle’s account
of ideal tragic structure or more grandiose claims such as Booker’s (2004) characteriza-
tion of the seven most basic plots. Nonetheless, creative writing courses, practising the
styles of admired writers or playing with literary conventions of form and genre, have all
helped many writers in the course of their creative development.
There may be a tendency to overestimate the role that natural talent plays in literary
creativity as opposed to immersion and motivated, deliberate hard work. It is one thing
to have an inspired idea for a work, it is quite another to acquire the expertise required
to flesh it out into decent literary form. There are several points worth noting here. First,
inspiration tends to come to those who have worked intensively in the relevant domain.
In other words, writers who work at developing their own literary style or strive to work
at writing particular kinds of literature will tend to see certain problems where others do
not, pick out what renders something distinctive or generic and set themselves their own
particular literary problems. Hence, even unconsciously, those immersed in a domain
will tend to be subject to much greater inspiration than those who are not. Second,
there is an increasing amount of literature on expert performance across a host of
domains which suggests that what matters is how people practise and train in the rele-
vant domains (Bloom 1985; Ericsson 2006). It is not just hard work that is required but
also, crucially, a matter of how you work. Striving for literary creativity requires working
out where weaknesses lie, what must be improved upon, how automaticity leads to bad
habits and so on. Lastly, what matters may not just be a matter of being strongly driven
but why. Authors write poems or stories for all sorts of reasons. Yet how and why moti-
vating reasons figure as they do may have a significant effect on an author’s creativity.
In one of her early studies Amabile (1985) divided seventy-two creative subjects into
three groups. The control group was tasked with writing a snow-themed poem, followed
by reading a short story and finally composing a laughter-themed poem. Subjects in the
second group were given the same tasks, except that after the short story reading they
were also required to rank order intrinsic reasons for writing (such as expression or the
joys of word play). The third group was asked to do the same as the second group except
that the list of reasons to be ordered were extrinsic ones (such as money, social status,
and graduate prospects). The group primed with extrinsic motivating reasons produced
the least creative work of the three groups. Indeed, the work the subjects in this third
group produced was judged to be significantly worse than that which they had produced
for the creative writing course prior to the experiment (as judged by twelve indepen-
dently successful writers). This is not to claim that extrinsic motivations such as money
and status necessarily corrupt literary creativity. However, it may suggest that where
such considerations pull apart from literary goals, extrinsic motivation can inhibit or
undermine the creative writing process.
No doubt inter-relations between motives and capacities are complex. It could be
that creativity is a matter of exercising certain capacities or skills in craft-like ways
(Gaut 2009; 2014). Alternatively it might be stressed that the empirical work is conso-
nant with the view that creative excellence consists in possessing certain traits or
virtues. The writers we tend to admire as writers are curious, open to experimentation,
or willing to take risks and fail. Such traits might be thought as exemplifying or mani-
festing creative virtues. We also admire and praise where their creativity is driven by
intrinsic motivations despite the lack of recognition or commercial reward. On this kind
of account literary creativity may be a craft, but to pursue the craft well involves creative
virtues (Kieran 2014a; 2014b).


How might such an approach be squared with the evidence concerning the connections
adduced above between literary creativity and mental illness? Perhaps there is some
reason to be wary about the evidence. It could be that a higher proportion of people who
go into the arts have a greater predisposition toward mental illness than normal people.
Mental illness and suffering more generally may often be alleviated or made sense of
through literary expression. In line with some of the traits mentioned above, literary
writers may also be open to riskier experiences or subject to greater frustrations because
of what they are trying to achieve. Alternatively, a higher incidence of mental illness
amongst creative writers could partly be explained in terms of greater exposure to the
kind of frustrating life conditions that trigger or precipitate mental illness. A standard
poet will likely typically struggle to get institutional paid employment to fund what they
desire to do, whilst a graduating natural scientist will typically go on to be employed by
a university or company which provides huge structural and socio-economic support.
Those who have structured lives with strong socio-economic support and goods tend to
suffer far less from mental illness than those who live in comparative poverty and lack
institutional support.
Nonetheless, despite some skepticism, there is significant evidence for interesting
links between literary creativity and mental illness. Is this then compatible with the idea
that creative excellence is partly a matter of virtue? Or does it fit more neatly with the
idea that creativity is a matter of capacity or skill or non-rational? It is worth considering
that such claims might be mutually consistent (at least once the claims are refined in
appropriate ways). Literary creativity constitutively includes the skills and capacities to
produce works which are interesting, surprising, new, or valuable. Literary or creative
virtues explain how and why writers often do so in the face of derision, poverty, and
indifference (and why we admire and praise them for it).
Creative people tend to score highly for traits such as curiosity, novelty seeking,
sensation seeking, challenge seeking, imaginativeness, openness, and unconventionality.
Such traits seem at least consonant with a virtue conception of literary creativity.
However, there is also much work to suggest that creativity is associated with a host of
negative personality or character traits. In general creative people in the arts have been
found to be more emotionally unstable, colder, stronger on rejecting group norms, and
less conscientious than scientists (Feist 1998). It might thus be claimed that creative
excellence in literature is bound up with traits that are morally suspect. Writers can be
an ambitious, domineering bunch and such characteristics may help spur them on to
ever-greater achievement or renown. Yet all this may show is either (1) that the virtues
required for creative excellence are not identical with the virtues required for moral
excellence (though there may be some overlap) or (2) the tension may be more apparent
than real (Kieran 2014b). Over-inflated self-belief is one thing, well-founded self-assur-
ance quite another. The former may get you some way but only the latter constitutes
true creative excellence. It is one thing to be a self-absorbed teenager carried away by
flashes of inspiration, it is quite another to work away like Milton.

Amabile, Teresa M. (1985), ‘Motivation and Creativity: Effects of Motivational Orientation on Creative
Writers’, Journal of Personality and Social Psychology, 48 (2): 393–99.
Aristotle (1986), The Poetics of Aristotle, trans. S. Halliwell, London: Duckworth.
Attridge, Derek (2004), The Singularity of Literature, London: Routledge.
Bloom, Benjamin (1985), Developing Talent in Young People, New York: Ballantine Books.

Matthew Kieran

Boden, Margaret (2004), The Creative Mind: Myths and Mechanisms, 2nd ed., London: Routledge.
Boden, Margaret (2010), ‘Creativity in a Nutshell’, in Creativity and Art, Oxford: Oxford University Press.
Booker, Christopher (2004), Seven Basic Plots: Why We Tell Stories, London: Continuum.
Carroll, Noël (2003), ‘Art, Creativity and Tradition’, in Berys Gaut and Paisley Livingston (eds) The Creation
of Art, Cambridge: Cambridge University Press.
Coleridge, Samuel Taylor (1816), Christabel; Kubla Khan; A Vision; The Pains of Sleep, 3rd ed., London:
John Murray.
Eliot, T. S. ([1919] 1950), ‘Tradition and the Individual Talent’, in his The Sacred Wood: Essays On Poetry
and Criticism, London: Methuen.
Ericsson, K. A. (2006), ‘The Influence of Experience and Deliberate Practice on the Development of
Superior Expert Performance’, in K. A. Ericsson, N. Charness, P. J. Feltovitch, and R. R. Hoffman (eds)
The Cambridge Handbook of Expertise and Expert Performance, Cambridge: Cambridge University Press.
Feist, Gregory J. (1998), ‘A Meta-Analysis of Personality and Scientific and Artistic Creativity’, Personality
and Social Psychology Review, 2 (4) 290–309.
Freud, Sigmund ([1907] 1995), ‘Creative Writers and Day Dreaming’, in The Standard Edition of the Complete
Psychological Works of Sigmund Freud, Vol. IX, Delusions and Dreams in Jensen’s Gradiva and Other Works,
London: Institute of Psychoanalysis and Hogarth Press.
Gaut, Berys (2003), ‘Creativity and Imagination’, in Berys Gaut and Paisley Livingston (eds) The Creation
of Art, Cambridge: Cambridge University Press.
Gaut (2009), ‘Creativity and Skill’, in Michael Krausz, Denis Dutton, and Karen Bardsley (eds) The Idea of
Creativity, Leiden: Brill, 83–103.
Gaut, Berys (2010), ‘The Philosophy of Creativity’, Philosophy Compass, 5 (12): 1034–1046.
Gaut, Berys (2012), ‘Creativity and Rationality’, Journal of Aesthetics and Art Criticism, 70 (3): 259–270.
Gaut, Berys (2014), ‘Mixed Motivations: Creativity as a Virtue’, in G. Currie, M. Kieran, A. Meskin, and
M. Moore (eds) Philosophical Aesthetics and the Sciences of Art: Royal Institute of Philosophy Supplement,
vol. 75, Cambridge: Cambridge University Press.
Hazlitt, William ([1816] 2002), Review in The Examiner, as reprinted in Samuel Taylor Coleridge: The Critical
Heritage, vol. 1, J. R. de J. Jackson (ed.), 2002, 205–208.
Hume, David ([1757] 1993), ‘Of the Standard of Taste’, in his Selected Essays, Oxford: Oxford University Press.
Jamison, Kay Redfield (1993), Touched with Fire: Manic Depressive Illness and the Artistic Temperament,
New York: Free Press.
Kant, Immanuel ([1790]1928), The Critique of Judgement, trans. J. C. Meredith, Oxford: Oxford University
Kieran, Matthew (2014a), ‘Creativity as a Virtue of Character’, in Scott Barry Kaufman and Elliot Samuel
Paul (eds) The Philosophy of Creativity, New York: Oxford University Press.
Kieran, Matthew (2014b), ‘Creativity, Virtue and the Challenges from Natural Talent, Ill-Being and
Immorality’, in G. Currie, M. Kieran, A. Meskin and M. Moore (eds) Philosophical Aesthetics and the
Sciences of Art: Royal Institute of Philosophy Supplement, vol. 75, Cambridge: Cambridge University Press.
Koestler, Arthur ([1964] 1970), The Act of Creation London: Macmillan.
Ludwig, Arnold M. Ludwig (1995), The Price of Greatness: Resolving the Creativity and Madness Controversy,
New York: Guildford Press.
Meskin, Aaron (unpublished manuscript), ‘Maximum, Standard, Absolute, Gradable Adjectives’.
Nietzsche, Friedrich ([1872] 1999), ‘Birth of Tragedy’, in The Birth of Tragedy and Other Writings, trans.
R. Spiers, eds. R. Geuss and R. Speirs, Cambridge: Cambridge University Press.
Novitz, David (1999), ‘Creativity and Constraint’, Australasian Journal of Philosophy, 77 (1): 67–82.
Olsen, Stein Haugom (2009), ‘Canon’, in S. Davies, K. M. Higgins, R. Hopkins, R. Stecker and D. E. Cooper
(eds) A Companion to Aesthetics, 2nd ed., Oxford: Oxford University Press.
O’Quin, Karen and Susan P. Besemer (1999), ‘Creative Products’, in Mark A. Runco and Steven R. Pritzker
(eds) Encyclopedia of Creativity, vol. 1, A–H, 413–422.
Plato ([380 BC] 2001), Ion, in Selected Dialogues of Plato, trans. B. Jowett and H. Pelliccia, New York:
Modern Library.
Stokes, Dustin (2008), ‘A Metaphysics of Creativity’, in K. Stock and K. Thomson-Jone (eds) New Waves
in Aesthetics, London: Palgrave Macmillan.
Woolf, Virginia (1940), ‘The Leaning Tower’, Collected Essays, vol. 2, New York: Harcourt, 1967, 162–181.

Paisley Livingston

What is authorship? How are answers to that question related to ideas about the
understanding, interpretation, or appreciation of literary and other works? In what
follows I provide a selective survey of the voluminous literature on these divisive
questions, offer criticisms of some influential theories, and present an alternative.

Two Conceptions of Authorship

It is often thought that creating or making a literary or some other kind of work is both
necessary and sufficient to being that work’s author. Authorship, then, amounts to
performing certain kinds of actions, such as composing a song, writing the text of a poem
or novel, and deciding when the work has been completed. It is generally acknowledged
that such work-constitutive actions can be performed either by an individual or by two
or more collaborating persons.
In many nations these basic ideas about authorship have been codified in legislation
designed to protect not only intellectual property but the “moral rights” of authors, such
as the right to control the conditions under which one’s work is made public. German
law, for example, rules that “Urheber ist der Schöpfer des Werkes” [The author is the
creator of the work] (Adeney 2006, p. 230). Many nations have similar legislation,
including clauses recognizing co-authorship (for informative surveys, see Davies and
Garnett 2010, and Rajan 2011).
Various philosophers and literary theorists have, however, contended that this notion
of authorship is inadequate. They claim that to read a text as authored by someone, or
to identify and think of someone as an author, is to accept—usually unwittingly—various
ideological assumptions. The ideology of authorship, they claim, blinds people to the
fact that different social formations have different conceptions and practices related to
discourse. The modern European system of authorship does not discover or refer to an
essence, but is a contingent social construction. As one philosopher puts this prevalent
thesis, “all authorship is constructed, assigned, and developed; there is no such thing as
a given or natural, non-constructed author” (Morgan 1988, p. 354). Authorship is said,
more specifically, to involve the attribution of authorship by readers or other representatives
of the literary institution or system. Authorship is therefore not equivalent to simply
writing, composing, or creating a work.
For shorthand we may refer to the two contrasting approaches evoked above as the
‘causal’ and ‘attributionist’ conceptions of authorship. According to the causal con-
ception, authorship is reducible to the actions that proximately cause a work to be
Paisley Livingston

created. According to the attributionist conception, the writer’s or speaker’s contribu-

tions are insufficient to constitute authorship. Instead, something more—something
on the side of the work’s reception—is required, beginning with a system of authorial

Foucault’s Attributionist Conception

The single most influential example of the attributionist approach is Michel Foucault’s
oft-cited (1969) lecture, “Qu’est-ce qu’un auteur?” A thorough analysis of the published
version of this lecture cannot be provided here, but aspects of Foucault’s position will be
discussed in some detail since they are crucial to an understanding of the contemporary
literature on authorship (for additional remarks, see Merquior 1985, Hendricks 2002,
Livingston 2005).
Foucault explicitly draws a distinction between the writer [le rédacteur] and the
author or “author-function.” He claims that in some discourses (his examples being
a personal letter, graffiti on a wall, and a legal document), there is a writer but not
an author, whereas various literary, philosophical, and other discourses have both
a writer and an author. The motivation for this distinction becomes clear in the
following passage:

Third characteristic of this author-function. We no doubt try to give a realistic

status to this figment of our minds [être de raison]. This would be something
within the individual, a “deep” instance, a “creative” power, a “project,” the
originary locus of writing. But in fact, those aspects of the individual that are
designated as author (or which make an individual an author) are only our
projection, in more or less psychological terms, of the treatment to which we
subject texts, the comparisons we draw, the traits that we establish as pertinent,
the continuities that we recognize, the exclusions that we practice. All these
operations vary according to periods, and to types of discourse.
(1994 [1969], p. 801)

The published English translations of Foucault’s essay get one part of this passage quite
wrong. Foucault’s “être de raison” is mistranslated as “a rational being,” “a rational
entity,” and even as a “being of reason,” whereas in fact the expression in French means
“a figment of thought,” or more colloquially, a figment of the imagination.
In the transcription of the discussion that took place after Foucault’s talk, Foucault is
reported as twice denying that he had asserted that the author does not exist (ibid., p. 817).
He presumably meant that he allowed that the author exists qua writer, or again that the
author-function exists qua mode of reading and attribution in a given discursive
formation. What does not exist, according to Foucault, is an author tout court. Foucault
sometimes misleadingly said or wrote ‘auteur’ when he meant ‘la fonction auteur’ (e.g.
when he said a personal letter has no author), but in the discussion he clearly asserted
that his goal was to “analyse the function within which something like an author could
exist” [“J’analysais la fonction à l’intérieur de laquelle quelque chose comme un auteur
pouvait exister”] (ibid., p. 818).
In the passage cited above, Foucault explicitly espouses a strong historicist thesis
about authorship, declaring that the author-constitutive operations vary with regard to


periods and kinds of discourse. He somewhat puzzlingly adds: “Yet one can find through
time a certain invariance in the rules of the construction of the author” [“Pourtant, on
peut retrouver à travers le temps un certain invariant dans les règles de construction de
l’auteur”] (ibid., p. 801). In an attempt to flesh out this claim, Foucault turns to Jerome’s
De Viris Illustribus [On Illustrious Men] (329–3 CE) and attributes to Jerome the following
four criteria of authorial attribution:

1. a constant level of value;

2. conceptual or theoretical coherence;
3. stylistic unity;
4. a single historical location.

In the literature on Foucault and authorship, these comments about Jerome are either
endorsed (e.g. During 1992, p. 122), or left unmentioned. Foucault’s claims in this
passage are, however, highly dubious. De Viris Illustribus is a fairly brief bibliographical
catalog of works by 135 early Christian authors. It contains no explicit generalizations
about the conditions on authorial attribution. Moreover, some of Jerome’s specific
attributions flatly contradict the general criteria Foucault has attributed to Jerome. For
example, Jerome attributes the Epistle to the Hebrews to Paul while remarking that its
style is quite different from that of other letters that are to be attributed to Paul. This
overtly contradicts criterion (3). Jerome allows that works by Plato and Philo exhibit a
very great similarity of both style and substance, but he cites a Greek proverb to the
effect that either Philo platonized or Plato philonized. Since Plato wrote long before
Philo did, only the former option could be correct. Here Jerome’s remarks could be taken
to imply that a causal condition on authorship can trump Foucault’s criteria (1–3).
Jerome’s actual attributions suggest that he recognized that the works firmly attributed
to a single author can be written at different times and places and manifest strikingly
different styles, attitudes, and levels of literary or other value. Were one to reconstruct
a theory of attribution based on Jerome’s particular judgements, the result would at best
be that the four “criteria” listed by Foucault may be included alongside many other
fallible indicators of authorship.

The History of Authorship

Many literary theorists and scholars have relied on Marxist assumptions in framing their
claims about the historical emergence of the author-function. Economic factors are in
the driver’s seat, followed up by legal constructions and the other rationalizations and
devices of bourgeois ideology. In some of the influential accounts that crop up in film
and literary theory, a large population of critics and readers—the victims of bourgeois
ideology—are said to have had astoundingly implausible beliefs about the Author. A key
source here is the straw man operation provocatively undertaken by Roland Barthes in
his influential (1968) essay, “La mort de l’auteur.” Barthes conjures up the specter of an
essentially solitary and sovereign figure, a masterful and self-conscious “Author-God,”
whose intention unilaterally determines the meaning of a unique and profoundly origi-
nal œuvre. To escape from the shackles of absolute intentionalism, the reader must sever
the ideological bond between text and Author, the assumption being that it is somehow
impossible for a reader operating within the modern author-functional regime to explore

Paisley Livingston

the unintended meanings and significance of texts while recognizing that they were
written by fallible, more or less skillful human beings.
Some literary theorists continue to applaud Barthes’ critique of the Author-God. For
example, Andrew Bennett writes that the Author-God conception is the apt target of
“the most powerful explanatory discourses of our, of contemporary, culture,” namely,
Marxism and psychoanalysis, which are to be credited with having revealed the human-
all-too-human author to have an unconscious and to be determined by capitalist
conditions (2005, pp. 7–8). Other literary theorists (Burke 1992, Gallop 2011) have
noted that Barthes himself announced a kind of “friendly” return of the author some
three years after the publication of the original French text of “The Death of the
Foucault notoriously linked the author-function’s emergence to that of a legal system
of ownership that was supposedly established “towards the end of the eighteenth and
beginning of the nineteenth century” (ibid., p. 799). This is a strikingly misleading
claim about the complex legal history pertaining to authorship (for relevant evidence,
see Ardeney 2006). Although the view has many defenders (e.g. Rose 1993,
Woodmansee 1984), it is highly controversial to yoke the emergence of “the” hege-
monic author-function to intellectual property legislation, such as the 1710 Statute of
Anne in England.
That Foucault’s historical conjectures are highly inaccurate has been argued by many
commentators, including Lamarque (1990), Saunders (1992), Chartier (1992, 2003),
Abrams (1995), and Kimmelman (1996). By far the most extensive source on this topic
is Vickers (2002). Vickers amasses evidence in support of the conclusion that the “author
emerged as a professional writer in the sixth century BC, and many of the attributes that
we associate with authorship—a sense of individual identity, in style, attitude, literary
structure; a hatred of plagiarism; a respected role in society—were already found in
abundance in Greco-Roman antiquity” (2002, p. 527; for a feeble attempt to brush aside
Vickers’ argument, see Maley (2010, pp. 34–35), who implausibly contends that Vickers
should accept Foucault’s views because Vickers holds that Shakespeare co-authored
some plays).
The idea that a radically distinct author function appeared in Europe sometime
between the seventeenth and the twentieth century discounts longstanding and recur-
rent aspects of authorship. Very basic practices of authorial attribution, starting with the
identification, praising, and blaming of the artifex or maker, had already emerged in
earlier periods and so should not be identified as the unique product of a modern
European economic, legal, and ideological formation. Such evidence includes the sillyboi
identifying the names of authors and titles on ancient scrolls; records of Greek literary
competitions identifying the names of the competing authors; authorial self-identifica-
tions figuring within ancient texts from a wide variety of ancient cultures, including
Egyptian wisdom texts; Aristotle-inspired, author-centered models of literary explana-
tion in the medieval period (Minnis 1988); and multiple complaints about plagiarism
written in the absence of intellectual property legislation (Ziegler 1950).
Perhaps Foucault was wrong about the historical specifics but right more generally in
promoting a historicist and attributionist approach to authorship. Could such an account
of authorship in ancient contexts be developed? An apparent example is Alexander
Beecroft’s (2010) comparative study of authorship in ancient Greece and China. As far
as Beecroft’s explicit definition of ‘authorship’ is concerned, he would appear to be a
thoroughgoing attributionist:


Authorship is a property ascribed to a literary text. It reflects an attempt to

ground and contextualize that text by assigning its composition and/or performance
to a specific individual, real or hypothetical, and the narrative representation
of that composition and/or performance constitutes a major category of evidence
concerning authorship.
(2010, p. 16)

Beecroft’s attributionist definition of ‘authorship’ finds a good part of its motivation in

the poverty of evidence we have about the actual origins of the ancient texts he
discusses. Attributions of authorship, especially those presented in texts, can be studied
in cases where we lack other evidence about the actual writing or composition of the
texts. It would appear, however, that Beecroft also works with a causal conception of
authorship distinct from the attributionist one cited above. For example, in his discussion
of Liji [The Record of Rites] he comments that “We cannot be certain of its dating or
authorship” (2010, p. 42). This remark makes good sense as a prudent expression of
uncertainty with regard to the identity of the writer and the time of writing, but cannot
be charitably read as a confession of ignorance about the long history of conjectures
regarding the authorship of this important source for the Confucian tradition.
I turn now to some additional reasons why a coherent and sufficiently comprehensive
attributionist account of authorship is not to be had.

Objections to Attributionist Conceptions of Authorship

At first glance, some of the influential attributionist accounts would appear to correspond
to the following basic schema, where the description on the right-hand side of the
equation is supposed to provide an explication or clarification of the left-hand side:

Authorship of work W = the authorship of work W is attributed to S

As the term “authorship” appears on both the right- and left-hand sides, the explication
suffers from a Molièresque circularity: one must already know what “authorship” means
if one is to make sense of what is said on the right-hand side. One step towards a solution
of this problem is to identify two different senses of “authorship” figuring on the two
sides of the equation, as in:

Authorship of work Wsense one = the authorshipsense two of work W is attributed to S

In literary theory, the sense of authorship figuring on the right-hand side of the equation
is usually the wrong-headed ideology of authorship that the author-function theorist
attributes to other attributors, as in Foucault’s potted account of Jerome’s attributional
principles, or the Barthesian evocation of Romantic ideas about the Author-God. So we
are invited to understand ‘authorship’ as a term that refers to the range of cases where a
wrong-headed idea of authorship is applied. Such an explication purports to provide a
socio-historical debunking of the author function or ideology.
A more appealing version of the attributionist scheme introduces a causal notion of
authorship into the right-hand side of the equation: authorship, then, is explicated as
the attribution of work creation or production. This would appear to make the explication

Paisley Livingston

compatible with the tempting idea that making or creating a work does after all have
something to do with its authorship. Yet anyone who is inclined to think of authorship
as the making of works may still find this explication highly counterintuitive: does
authorship require attribution? Are there no created works that remain unpublished? Are
there no works the authors of which were never identified? How did the system of
authorship ever get started?
It is especially hard to see how an attributionist conception of authorship can account
for an important category of cases, namely, those where an attribution of authorship does
not identify all and only those who composed or wrote the work (where writing a work
can be a joint action performed by two or more persons or something done by a single
agent). More specifically, we ought to be on the lookout for:

•• Ghost authorship: cases where a work has been created by someone to whom
authorship is not attributed; and
•• Gift authorship and forgeries: cases where a work is attributed to someone who did
not actually create it.

It is important to espouse an account of authorship that allows us to say that such cases
involve incorrect attributions and that these attributions are incorrect because they do
not successfully track the action of authoring performed in the making of a given item. In
a type of case that is often complained about in discussions of the social dimensions of
contemporary science (e.g. Lawrence 2003, Kwok 2005), the prestigious and powerful
head of a lab does not do any of the intellectual work and does not write up the results,
but puts his name on the list of authors. Although he did not perform the requisite
actions, he wrongly claims credit for being one of the authors. In literary contexts rel-
evant cases include plagiarism, forgeries, and the exploitation of unacknowledged ghost
writers (for a presentation of a number of cases, see Love 2002 and Ziegler 1950; the
latter documents ancient Greek and Roman complaints about plagiarism as well as
seventeenth-century treatises on the topic, such as Jacobus Thomasius Subaci’s 1673
Dissertatio philosophica de plagio literario; for a contemporary complaint about plagiarism,
see Weber 2007).
A hard-core attributionist can bite the bullet and insist that there are no cases of
ghost or gift authorship. It is hard to see why such a position should be accepted, how-
ever, especially when the attributionist recognizes the actions performed by the readers,
editors, and theorists who engage in the making of attributions.

Fictionalist Conceptions of Authorship

Assume that it is granted that phrases such as ‘Dickens authored Great Expectations’ are
true by virtue of the writer’s actions. It does not follow logically from this fact that the
question of how the reader should interpret the text of this novel is thereby settled.
More specifically, the reader need not search the text for evidence of the actual writer’s
attitudes and intentions. Instead, the reader might imagine an Author for this text along
entirely different lines, developing this attribution in a fictional or “as if” way of
thinking (Nehamas 1981, 1986, 1987; Morgan 1988). The reader’s concoction of a
make-believe Author is based on the features of the text and does not have to corre-
spond to what is believed about the actual writer of this text and the historical context
in which the text was written. The reader may find the text more interesting when ideas


that were unknown to the actual writer are brought to bear on its interpretation. Perhaps
the shift to a fictionalist approach to authorship allows the proliferation of meanings
longed for by some poststructuralist theorists.
There are, however, objections to this fictionalist proposal. One is that it amounts to
a massive and insufficiently motivated revision of the way people generally read
literature. Attributions of authorship routinely express heartfelt belief: persons believed
to have authored some work are admired or criticized for having done so. Having read
and admired work W by author A, the reader wants to find other works by the same
author, where “the same author” does not refer to a figment of the reader’s imagination.
The fictionalist approach would appear to give the reader a desirable freedom, but may
do so at the cost of preventing the reader from realizing the goal of discovering, or at
least of trying to discover, the value and meaning of the actual writer’s works.
This objection can by supported with reference to recent work by the psychologist
Eefje Claassen (2012). The results of her experiments with readers indicate that readers’
understandings of texts are influenced by their beliefs about the persons responsible for
writing those texts. Readers develop a mental representation that includes the author’s
characteristics, communicative intentions, and moral positions or attitudes. When
they are given what they take to be information about the actual author’s attitudes and
background, readers’ understandings of the text take this information into account.
Claassen’s conclusion is that “the theoretical claim that the author is irrelevant for the
interpretation of literary texts is untenable” (2012, p. 219).
The fictionalist approach recommends that an adequate response to literary works
can be had by thinking only about the attitudes of the implied authors, which are the
personae manifested in the texts alone. Such a recommendation is misguided if there
are cases where features crucial to apt appreciation arise from divergences between the
attitudes of the actual author and those expressed by the implicit authorial persona.
Autobiographies are cases that would appear to belong to this category. While it would
be naive to think that someone’s autobiography conveys only truths about that person,
it would be absurd to read it as only making assertions about an implicit author to be
imagined by the reader. For example, when Jean-Paul Sartre writes in Les mots (1963)
that he had no superego because he had his mother all to himself, this is a claim about
the actual Sartre, not about some figment of the reader’s imagination. If in writing about
my own life I deviate from what I believe, my reports are lies, not fictions or invitations
to make-believe. It can also be argued convincingly, as Alex Neill (1999) has done, that
a poet’s insincerity can be critically relevant, even in a case where this insincerity is not
manifest in the text of the poem and would not be discernible to a reader interested only
in the attitudes of the implied author.

A Causal Conception of Authorship

Causal conceptions of authorship fill in the following schema:

S authors some work, W, just in case S intentionally performs actions A1–An.

Performing an intentional action entails exercising sufficient control over one’s

behavior: an involuntary sneeze is not an action (Mele and Moser 1994). This
requirement is compatible with the intentional use of some random process in the
generation of features of a work, since such a procedure can be intentionally selected

Paisley Livingston

as a means of production. Yet someone who exercises no control over what is included
in a text is not the author of that work. The sufficient control requirement on authorship
pertains to both the internal and external conditions under which actions and
choices take place. Authorship is vitiated by coercion. For example, when terrorists
coerce someone into writing and signing a declaration, the hostage is not the author
of the document. Even milder sorts of coercion are deemed relevant: some critics
who claim that Mary Shelley was not really free to reject Percy Shelley’s revisions
of her text contend that her authorship of Frankenstein was thereby diminished or
converted into a kind of collaboration; Byron, on the other hand, freely allowed
Mary Shelley to introduce revisions into the drafts she rewrote for him (for back-
ground, see Leader 1996).
Which kinds of intentional actions are required by authorship? Livingston’s (1997,
2005) proposal is that authorship requires expressive actions, and more specifically, the
making of an utterance. And what is an utterance? According to Grice’s influential
proposal, “utterance” refers to anything that is a (plausible) candidate for non-natural
meaning, which means anything that is the result of a certain complex kind of
communicative intention (as discussed in Grice 1989). As Grice’s strictures about
communicative intentions are highly problematic, it is preferable to work with the
neo-Gricean account set forth by Wayne C. Davis (2002). Here the key, utterance-
constitutive intention is an expressive one aimed at indicating or manifesting the
utterer’s attitudes. According to this proposal, expression need not be sincere, veridical,
or original. That an action is indicative of some attitude does not mean this attitude is
actual (cf. “The clouds indicated rain, but it didn’t rain”).
Given these remarks about the sufficient-control and expressive intention conditions
entailed by the making of an utterance, we arrive at the following causal account of

S authors x just in case x is S’s utterance; and

S1–Sn co-author x just in case x is their joint utterance

If this seems far too broad, remember that “John was the author of the boring and poorly
written email” is perfectly good English, as is “Jacques’ missive was a tissue of clichés.”
To say that authorship requires the making of an utterance in a very broad sense leaves
it open whether one wishes to make additional claims about sub-categories of authorship.
One might, for example, develop some distinction between everyday utterances and
works. The authorship of works, then, would be a subset of authorship more generally.
This sort of thing is stipulated in many legal codes, such as the French legislation, where
the problem of saying what does and does not count as une œuvre de l’esprit [a work of
the mind] is handled by listing “statutory examples” (Adeney 2006, pp. 174–175).
The definitions of ‘author’ figuring within legal codes often include a novelty or
originality condition: to be an author one must not only create a work, but the work has
to be original. The rationale behind this sort of definition is that only authorial achieve-
ments satisfying these conditions can earn someone the authorial rights the code was
designed to protect. Such legal usage is contradicted, however, by prevalent talk that
allows that someone can be identified as the author of a plagiarism. The plagiarist copies
all or part of the text of a work previously authored by someone else; the plagiarist then
deceptively categorizes or presents this copied text as the product of his or her own devising,
and not as something that was merely copied. We can allow that this is a species of


authorship in our broadest sense and coherently go on to criticize the plagiarist’s

deceptive action. Authorship is one thing, honest and valuable authorship is something
more. The most viable version of a causal theory of the authorship of utterances and
works is a non-honorific, value-neutral one.
To come up with a conception of literary authorship, it would be necessary to apply
some favored distinction between literary and non-literary utterances or works, a topic
that cannot be surveyed here (see Livingston 2003 for a survey). Also, different genres,
such as the Petrarchan sonnet or the philosophical paper, clearly weigh additional
success conditions on authorial intentions. Finally, a detailed account of joint authorship
must identify assumptions about the conditions under which a given action can be
jointly performed by two or more persons, another complex topic that cannot be taken
up here.

Abrams, Martin H. (1995) “What is Humanistic Criticism?” In The Emperor Redressed: Critiquing Critical
Theory, ed. Dwight Eddins. Tuscaloosa: The University of Alabama Press, pp. 13–44.
Adeney, Elizabeth (2006) The Moral Rights of Authors and Performers: An International and Comparative
Analysis. Oxford: Oxford University Press.
Barthes, Roland (1968) “La mort de l’auteur.” Manteia, 5, 12–17; reprinted in Œuvres complètes, ed. Eric
Marty. Paris: Seuil, 1994, vol. 2, 491–495; trans. “The Death of the Author,” in Image-Music-Text, ed.
Stephen Heath. London: Fontana, 1977, 142–148.
Beecroft, Alexander (2010) Authorship and Cultural Identity in Early Greece and China: Patterns of Literary
Circulation. Cambridge: Cambridge University Press.
Bennett, Andrew (2005) The Author. London: Routledge.
Buranen, Lise and Alice M. Roy, eds. (1999) Perspectives on Plagiarism and Intellectual Property in a Postmodern
World. Albany: State University of New York Press.
Burke, Seán (1992) The Death and Return of the Author: Criticism and Subjectivity in Barthes, Foucault, and
Derrida. Edinburgh: Edinburgh University Press, 1992.
Burke, Seán, ed. (1995) Authorship: From Plato to Postmodern: A Reader. Edinburgh: Edinburgh University
Chartier, Roger (1992) L’Ordre des livres. Aix-en-Provence: Alinéa.
Chartier, Roger (2003) “Foucault’s Chiasmus: Authorship between Science and Literature in the Seventeenth
and Eighteenth Centuries.” In Mario Biagioli and Peter Galison, eds., Scientific Authorship: Credit and
Intellectual Property in Science. New York and London: Routledge, pp. 13–32.
Claassen, Eefje (2012) Author Representations in Literary Reading. Amsterdam and Philadelphia: John Benjamins.
Davies, Gillian and Kevin Garnett (2010) Moral Rights. London: Sweet & Maxwell.
Davis, Wayne C. (2002) Meaning, Expression, and Thought. Cambridge: Cambridge University Press.
Donovan, Stephen, Fjellestad, Danuta, and Rolf Lunden, eds. (2008) Authority Matters: Rethinking the Theory
and Practice of Authorship. Amsterdam: Rodopi.
During, Simon (1992) Foucault and Literature : Towards a Genealogy of Writing. London: Routledge.
Foucault, Michel (1969) ‘Qu’est-ce qu’un auteur?’, Bulletin de la Société Française de la Philosophie, 63
(July–September 1969), 73–104. Reprinted in Dits et écrits 1954–1988, ed. Daniel Defert and François
Ewald (Paris: Gallimard, 1994), i.789–821; Trans. Josué V. Harari, ‘What is an Author?’ in Textual
Strategies: Perspectives in Post-Structuralist Criticism, ed. Josué V. Harari. Ithaca, NY: Cornell University
Press, 1979, 141–60; trans. Donald F. Bouchard and Sherry Simon, ‘What is an Author?’ in Language,
Counter-Memory, Practice: Selected Essays and Interviews, ed. Donald F. Bouchard. Ithaca, NY: Cornell
University Press, 1977, 113–38; Trans. Karin Hofer and Anneliese Botont, “Was ist ein Autor?” in Texte
zur Theorie der Autorschaft, ed. Fotis Jannidis, Gerhard Lauer, Matias Martinez, and Simone Winko.
Stuttgart: Reclam, 2000, 198–229.
Gallop, Jane (2011) The Deaths of the Author. Durham, NC: Duke University Press.
Grice, H. Paul (1989) Studies in the Way of Words. Cambridge, MA: Harvard University Press.
Gutbrodt, Fritz (2003) Joint Ventures: Authorship, Translation, Plagiarism. Bern: Peter Lang.
Hadjiafxendi, Kyriaki and Polina Mackay (2007) Authorship in Context: From the Theoretical to the Material.
Houndsmills, Basingstoke: Palgrave Macmillan.

Paisley Livingston

Hendricks, Christina (2002) “The Author[’s] Remains: Foucault and the Demise of the ‘Author-Function’.”
Philosophy Today, 46, 2, 152–163.
Irwin, William, ed. (2002) The Death and Resurrection of the Author? Westport, Conn.: Greenwood Press.
Jerome. De Viris lllustribus. In Nicene and Post-Nicene Fathers, Second Series, Vol. 3. Trans. Ernest Cushing
Richardson. Ed. Philip Schaff and Henry Wace. Buffalo, NY: Christian Literature Publishing Co., 1892.
Kimmelman, Burt (1996) The Poetics of Authorship in the Later Middle Ages: The Emergence of the Modern
Literary Persona. New York: Peter Lang.
Kwok, L. S. (2005) “The White Bull Effect: Abusive Coauthorship and Publication Parasitism.” The Journal
of Medical Ethics, 31, 554­–556.
Lamarque, Peter V. (1990) “The Death of the Author: An Analytic Autopsy.” The British Journal of Aesthetics,
48, 319–331.
Lawrence, P. A. (2003) “The Politics of Publication.” Nature, 422, 259–261.
Leader, Zachary (1996) Revision and Romantic Authorship. Oxford: Clarendon.
Livingston, Paisley (1997) “Cinematic Authorship.” In Film Theory and Philosophy. Ed. Richard Allen and
Murray Smith. Oxford: Oxford University Press, 1997, pp. 132–148. Reprinted in Philosophy of Film and
Motion Pictures, ed. Noël Carroll and Jinhee Choi. Oxford: Blackwell, 2006, pp. 299–309.
Livingston, Paisley (2003) “Literature.” In Oxford Handbook of Aesthetics. Ed. Jerrold Levinson. Oxford:
Oxford University Press, pp. 536–554.
Livingston, Paisley (2005) Art and Intention. Oxford: Clarendon.
Livingston, Paisley (2008) “When a Work is Finished.” The Journal of Aesthetics and Art Criticism, 66, 4,
Livingston, Paisley (2009) Cinema, Philosophy, Bergman. Oxford: Oxford University Press.
Livingston, Paisley (2011) “Discussion: On Authorship and Collaboration.” Journal of Aesthetics and Art
Criticism, 69, 2, 217–220.
Livingston, Paisley and Carol Archer (2010) “Artistic Collaboration and the Completion of Works of Art,”
The British Journal of Aesthetics, 50, 4, 439–456.
Long, Pamela O. (2001) Openness, Secrecy, Authorship: Technical Arts and the Culture of Knowledge from
Antiquity to the Renaissance. Baltimore: The Johns Hopkins University Press.
Love, Harold (2002) Attributing Authorship. Cambridge: Cambridge University Press.
Maley, Willy (2010) “Malfolio: Foul Papers on the Shakespeare Authorship Question.” In William Leahy,
ed. Shakespeare and His Authors: Critical Perspectives on the Authorship Question. London: Continuum,
pp. 23–40.
Martz, Louis L. and Aubrey Williams, eds. (1978) The Author in His Work: Essays on a Problem in Criticism.
New Haven: Yale University Press.
Mele, Alfred R. and Paul K. Moser (1994) “Intentional Action.” Noûs, 28, 39–68.
Merquior, J. G. (1985) Foucault. Berkeley: The University of California Press.
Minnis, A. J. (1988) Medieval Theory of Authorship: Scholastic Literary Attitudes in the Late Middle Ages. 2nd
edn, Aldershot: Scolar Press.
Morgan, Michael L. (1988) “Authorship and the History of Philosophy.” Review of Metaphysics, 42,
Nehamas, Alexander (1981) “The Postulated Author: Critical Monism as a Regulative Ideal.” Critical
Inquiry, 8, 131–149.
Nehamas, Alexander (1986) “What an Author Is.” Journal of Philosophy, 83, 685–691.
Nehamas, Alexander (1987) “Writer, Text, Work, Author.” In Literature and the Question of Philosophy, ed.
Anthony J. Cascardi. Baltimore, Md.: Johns Hopkins University Press, 265–291.
Neill, Alex (1999) “Inauthenticity, insincerity, and poetry.” In Performance and Authenticity in the Arts. Ed.
Salim Kemal and Ivan Gaskell. Cambridge: Cambridge University Press, pp. 197–214.
Rajan, Mira T. Sundara (2011) Moral Rights: Principles, Practice and New Technology. Oxford: Oxford
University Press.
Rose, Mark (1993). Authors and Owners: The Invention of Copyright. Cambridge, MA: Harvard University Press.
Sartre, Jean-Paul (1963) Les mots. Paris: Gallimard.
Saunders, David (1992) Authorship and Copyright. London: Routledge.
Stecker, Robert (1987) “Apparent, Implied, and Postulated Authors.” Philosophy and Literature, 11, 258–271.
Stillinger, Jack (1991) Multiple Authorship and the Myth of Solitary Genius. Oxford: Oxford University Press.
Vickers, Brian (2002) Shakespeare, Co-Author: A Historical Study of Five Collaborative Plays. Oxford: Oxford
University Press.


Weber, Stefan (2007) Das Google-Copy-Paste-Syndrome: Wie Netzplagiate Ausbildung und Wissen gefährden.
Hannover: Heise.
Woodmansee, Martha (1984) “The Genius and the Copyright: Economic and Legal Conditions of the
Emergence of the ‘Author’.” Eighteenth Century Studies, 17 (1984), 425–448.
Woodmansee, Martha and Peter Jaszi, eds. (1994) The Construction of Authorship: Textual Appropriation in
Law and Literature. Durham, NC: Duke University Press.
Ziegler, Konrat (1950) “Plagiat.” In Paulys Realencyclopaedie der classischen Altertumswissenschaft. Ed. Georg
Wissowa, vol. 20:1. Stuttgart: Alfred Druckenmueller, pp. 1956–1997.

Recommendations for further reading

Several of the works cited above merit a careful reading: Love (2002), because it is broad, engaging, and factu-
ally rich, is a good place to start. Vickers (2002) is a crucial vaccination against the theoretic and factual
ignorance that plagues the literature on authorship. Adeney (2006) is a superb compendium on the moral
rights of authors legislation. Stecker (1987) and Lamarque (1990) perform some crucial conceptual spade-
work. Leader (1996) ought not to be missed, even if you’re not a fan of the English Romantics under
For additional background on the literature, three complementary collections are Maurice Birotti and
Nicola Miller, eds. (1993) What is an author? (Manchester: Manchester University Press), Seán Burke, ed.
(1995) Authorship: From Plato to Postmodern: A Reader (Edinburgh: Edinburgh University Press), and William
Irwin, ed. (2002) The Death and Resurrection of the Author? (Westport, CN: Greenwood Press). An interesting
attributionist perspective on the 1960s theories is developed by Carla Benedetti in her (2005) The Empty
Cage: Inquiry into the Mysterious Disappearance of the Author (Ithaca, NY: Cornell University Press). For more
of the current author’s writing on this topic, see Livingston (2009) Cinema, Philosophy, Bergman (Oxford:
Oxford University Press), chapter 3, (2008) “Authorship Redux.” Philosophy and Literature, 32, 191–197,
Kelly Trogdon and Paisley Livingston, ‘The Complete Work’. The Journal of Aesthetics and Art Criticism, 72:3
(2014), 225–233.

Peter Lamarque

In virtue of what does a poem express emotion? Philosophical debate on that question
often reflects the lingering influence of two dominant tendencies in approaches to
literature, broadly labelled “Romantic” and “Modernist”. Where Romanticism is loosely
associated with author-centred views, Modernism is loosely associated with work-
centred views. We shall see how these tendencies play out in the debate, but also where
convergence is in evidence.

In the simplest terms, author-centred or Romantic views of expression hold that when
a poem is said to express an emotion, what is meant is that an actual emotion in the poet
is being expressed by means of the poem. To understand and appreciate the poem is
to understand and appreciate (perhaps even to share) an experience the poet has
undergone. Thus Leo Tolstoy: “one man consciously, by means of certain external signs,
hands on to others feelings he has lived through, and . . . other people are infected by
these feelings, and also experience them” (Tolstoy 1969: 123).
Romantic expression theory is sometimes thought to be epitomized in William
Wordsworth’s much quoted aphorism that “all good poetry is the spontaneous overflow
of powerful feelings” (Wordsworth 2006b: 498). Wordsworth clearly liked the phrase
“spontaneous overflow of powerful feelings”, as he repeated it twice in his Preface to the
Lyrical Ballads. But it is easy to be misled as to his meaning. The impression that poetry
is an uncontrolled outpouring of present emotional states is far from what he meant. On
both occasions Wordsworth immediately qualifies – almost to the point of contradicting –
the natural interpretation that his phrase suggests. On the first occasion he follows it
(in the very next sentence) with this: “Poems to which any value can be attached were
never produced on any variety of subjects but by a man who, being possessed of more
than usual organic sensibility, had also thought long and deeply” (ibid). Thinking “long
and deeply” seems curiously at odds with “spontaneous overflow”.
The second occasion also seems to undermine spontaneity. Again, this time in the
actual sentence in which the phrase occurs, Wordsworth qualifies his catchy aphorism:

it [i.e. poetry] takes its origin from emotion recollected in tranquillity: the
emotion is contemplated till, by a species of reaction, the tranquillity
gradually disappears, and an emotion, kindred to that which was before the

subject of contemplation, is gradually produced, and does itself actually

exist in the mind.
(Wordsworth 2006: 504)

To speak of recollecting an emotion “in tranquillity” seems to weaken all three of the key
terms, “spontaneous”, “overflow”, and “powerful”. Two separate emotions seem to be
invoked: the original (powerful) emotion and a second (initially more tranquil) emotion
stirred by later recollection of the first. It is the second which finds expression in the poem
itself. Many of Wordsworth’s autobiographical poems are based on recollection in tranquillity,
not least “Tintern Abbey”, which mingles a present experience of the Abbey with
recollections of a visit five years previously. The speaker even combines in a single
experience past, present, and future, anticipating the pleasure of looking back in future years:

And now, with gleams of half-extinguished thought,

With many recognitions dim and faint,
And somewhat of a sad perplexity,
The picture of the mind revives again:
While here I stand, not only with the sense
Of present pleasure, but with pleasing thoughts
That in this moment there is life and food
For future years.
(Wordsworth 2006a: 408–9)

It would be wrong, then, to suppose that Romantic expression theory is committed to

poetry as an unmediated pouring forth of a present-moment experience. A poem does
not stand to the emotion it expresses in the way that, say, someone’s shouting and
stamping might stand to the anger that person is experiencing. Percy Bysshe Shelley
describes the difference between a poem and its original inspiration:

the mind in creation is as a fading coal, which some invisible influence . . . 

awakens to transitory brightness. . . . Could this influence be durable in its
original purity and force, it is impossible to predict the greatness of the results—
but when composition begins, inspiration is already on the decline, and the
most glorious poetry that has ever been communicated to the world is probably
a feeble shadow of the original conception of the poet.
(Shelley 2006: 1195)

The implication, also present in Wordsworth, is that composition has a calming effect,
cooling the initial fires of inspiration.
Perhaps the most sophisticated philosophical version of what might be called “classic”
expression theory is that of R. G. Collingwood in The Principles of Art (1938).
Collingwood’s view enlarges on but does not contradict that implied in Wordsworth and
Shelley. It has four distinct features. The first is that an artistic expression of emotion
both articulates and also clarifies the emotion expressed, the nature of which the author
might not know in advance: “Until a man has expressed his emotion, he does not yet
know what emotion it is. The act of expressing it is therefore an exploration of his own
emotions. He is trying to find out what these emotions are” (Collingwood 1938: 111).
It is only in the act of expression that the emotion takes shape and acquires its identity.

Peter Lamarque

The second feature is a distinction between expressing and describing an emotion: “A

genuine poet, in his moments of genuine poetry, never mentions by name the emotions
he is expressing” (Collingwood 1938: 112). Why not? Because the trouble with description
(“joy”, “grief”, “yearning”) is that it “generalizes” and for Collingwood the emotion
expressed in a genuine poem is utterly particular to that poem. A third feature concerns
the difference between expressing an emotion and arousing an emotion:

A person arousing emotion sets out to affect his audience in a way in which he
himself is not necessarily affected. . . . A person expressing emotion, on the
contrary, is treating himself and his audience in the same kind of way; he is making
his emotions clear to his audience, and that is what he is doing to himself.
(Collingwood 1938: 110–11).

The final feature marks a distinction between expressing and “betraying” an emotion.
You might “betray” your fear by “turning pale and stammering” and that indeed is a kind
of expression. But for Collingwood it is not the kind that is important in art. Artistic
expression involves “lucidity or intelligibility” (Collingwood 1938: 122) and that, he
thinks, will be lacking when an emotion is merely “betrayed” and its “precise quality”
not apprehended. The difference might seem not dissimilar to that between Wordsworth’s
“recollected in tranquillity” and his “spontaneous overflow”.
These four features of Collingwood’s theory can be seen as a combined effort to mark
the closest possible relation between the expression (in the poem) and the emotion
(in the poet). If the relation were merely causal then the emotion and its expression
would be distinct, the one identifiable and intelligible without the other. For
Collingwood the relation is more that of constitution than causation: the expression
somehow constitutes the emotion and is not simply its effect. True expression neither
describes the emotion nor betrays it, but rather gives it its unique identity. There is some
tension, then, in the Romantic theory in wanting to emphasize, on the one hand, real
psychological states of authors (and readers), and on the other, properties of works of art
(poems), which embody the expression itself. Collingwood, however, is unambiguous in
keeping actual emotions in play: those of the author, with something personal to express,
and those of the reader with an emotion shared:

When someone reads and understands a poem, he is not merely understanding

the poet’s expression of his, the poet’s, emotions, he is expressing emotions of
his own in the poet’s words, which have thus become his own words.
(Collingwood 1938: 118)

It is this emphasis on the poet’s own personal emotions that became the target for the
Modernists, giving rise to a quite different, work-centred, view of expression.

By the end of the nineteenth century the grip of Romanticism was already beginning to
loosen. Early Modernists, such as the French symbolist poet Stéphane  Mallarmé
(1842–1898), began questioning the ascendancy of the author in favour of an emphasis on
writing itself. Indeed when Roland Barthes postulated the “death of the author” in 1967 he
acknowledged Mallarmé’s influence: “Mallarmé was doubtless the first . . . to foresee in its


full extent the necessity to substitute language itself for the person who until then had been
supposed to be its owner. For him, for us too, it is language which speaks, not the author”
(Barthes 1977: 143). But perhaps most influential of all in promoting what he labelled the
“Impersonal theory of poetry” was T. S. Eliot in his essay “Tradition and the Individual
Talent” from 1919. Eliot sought to “divert interest from the poet to the poetry”, stressing that

the poet has, not a “personality” to express, but a particular medium, which is
only a medium and not a personality. . . . It is not in his personal emotions, the
emotions provoked by particular events in his life, that the poet is in any way
remarkable or interesting. . . . The emotion in his poetry will be a very complex
thing, but not with the complexity of the emotions of people. . . . Poetry is not
a turning loose of emotion, but an escape from emotion: it is not the expression
of personality but an escape from personality.
(Eliot 1972: 75–6)

The critic and novelist C. S. Lewis in turn made the important move of decoupling the
speaker in the poem from the actual poet: “It is . . . impossible that the character repre-
sented in the poem should be identically the same with that of the poet. The character
presented is that of a man in the grip of this or that emotion: the real poet is a man who
has already escaped from that emotion sufficiently to see it objectively . . . and to make
poetry of it” (Tillyard and Lewis 1939: 9). Indeed, he states something stronger still:
“when we read poetry as poetry should be read, we have before us no representation
which claims to be the poet, and frequently no representation of . . . a personality at all”
(Tillyard and Lewis 1939: 4).
The most extended philosophical assault on author-based Romantic theories comes
in Alan Tormey’s book The Concept of Expression (1971). Tormey characterizes what he
takes to be the core of all versions of Romantic expression theory (including those of
Tolstoy and Collingwood) as follows:

If art object O has expressive quality Q, then there was a prior activity C of the
artist A such that in doing C, A expressed his F for X by imparting Q to O
(where F is a feeling state and Q is the qualitative analogue of F).
(Tormey 1971: 103)

The mistake of the Romantic theorists, Tormey argues, is “assuming that the existence
of expressive qualities in a work of art implies a prior act of expression” (Tormey 1971: 104).
They are also mistaken, he thinks, in “the further assumption of a necessary link between
the qualities of the art work and certain states of the artist” (ibid). Instead “statements
about the expressive qualities of an art work remain, irresolutely, statements about the
work, and any revision or rejection of such statements can be supported only by referring
to the work itself”; they are “not subject to falsification through the discovery of any
truths about the inner life of the artist” (Tormey 1971: 105).
Tormey’s work-centred view of expression epitomizes the “autonomy” conception of
the work that characterized New Criticism and it connects naturally with New
Criticism’s two great dogmas, the Intentional Fallacy and the Affective Fallacy. Monroe
Beardsley, the New Critics’ principal philosophical spokesman, followed C. S. Lewis,
and indeed earlier Modernists (including Mallarmé), in insisting on the distinction
between poet and poetic speaker:

Peter Lamarque

Now clearly Conan Doyle’s use of the word “I” in the Sherlock Holmes stories
does not give this pronoun a reference to an actual person. . . . Why then must
we assume that when Keats or Shelley uses the pronoun he is always referring
to himself? If you write a “Happy Birthday” poem and hand it to someone on
his birthday or send it under your own name, you are giving the poem a prag-
matic context. If you write a love poem . . . you are not giving the message a
pragmatic context; it is from nobody to nobody, it is not addressed to, but rather
overheard by the reader: and though you are the writer you are not the speaker.
(Beardsley 1981: 239–40)

Beardsley later drew on speech act theory to clarify his position: “the writing of a poem,
as such, is not an illocutionary act; it is the creation of a fictional character performing
a fictional illocutionary act” (Beardsley 1970: 59).
No doubt the Romantics would simply deny that their first-person autobiographical
poems involve fictional characters. But what they cannot deny is that in indisputable
works of fiction, fictional characters can express emotions. King Lear’s cry of anguish in
holding the dead Cordelia shows this clearly:

Howl, howl, howl, howl! O, you are men of stones:

Had I your tongues and eyes, I’d use them so
That heaven’s vault should crack. She’s gone for ever!
I know when one is dead, and when one lives;
She’s dead as earth.
(King Lear V, iii, 270–4)

The emotion expressed in this passage is powerful and disturbing but there is no
inclination to attribute the emotion to Shakespeare himself; it is the character Lear who
is the speaker. The example is important in showing that language can be expressive
(can express emotion) without the emotion being associated with any real person.
Tormey and Beardsley use that fact to motivate their “autonomy” conception of poetry.
But is there any further pressure to suppose that even lyric poems are autonomous in
this sense? One way to support the idea that poems involve “fictional illocutionary acts”
might be to consider questions as they appear in poems. For example, John Keats’ “Ode
on a Grecian Urn” is full of questions. Here are some:

What men or gods are these? What maidens loth?

What mad pursuit? What struggle to escape?
What pipes and timbrels? What wild ecstasy?
(Keats 2006: 1398)

Who is asking these questions? Is it Keats the poet? After all, he composed the lines
and the sentences he used are in interrogative form. Is that not sufficient for the
sentences to count not just as questions but as his questions? At one level that seems
right – they are questions (they are not commands or assertions) and Keats is solely
responsible for them. Yet at another level (perhaps the level at which J. L. Austin
describes poetry as an “etiolation” of language [Austin 1962: 22]) they must be
recognized as merely rhetorical questions. They are not addressed to anyone and they
do not seek answers. If they are attributed to the speaker in the poem then they are


not questions directly asked by Keats. Whatever other acts Keats might be performing
he is not in any straightforward way performing the illocutionary act of asking questions.
The questions are quite naturally attributable to a speaker who is not identical to
the poet.
It might seem just a short step in combining this observation about the questions in
Keats’ poem with the earlier observation about the expressiveness in the Lear lines to
the conclusion that emotions expressed even in the most favourable (“autobiographical”)
cases should themselves be attributed to a speaker or persona in the work rather than
directly to the poet. Other considerations support that thought. After all, it is the poetic
lines that are the immediate focus of a reader’s attention, not the long-past emotional
state of a perhaps long-dead poet; when we read about “dejection” in Coleridge and
Shelley or “melancholy” in Keats it is the powerful and timeless characterization of those
emotions that draws us in, not, or at least not just, curiosity about an individual’s suffering;
the achievement we admire is a poetic achievement not, or so the Modernists argue,
autobiographical fidelity to the facts.

So are the Romantics finally defeated on the issue of expression? Jenefer Robinson
(Robinson 2005) thinks not and has offered what she calls a “new romantic theory of
expression” which aims to capture, as she sees it, the insights of author-based Romantic
theories while accommodating at least some of the concerns of work-based Modernist
theories. She points out that Tormey has in certain respects mischaracterized the
Romantic theory. Thus it is no part of Collingwood’s theory to postulate a “prior activity”
of the artist which results in the “imparting” of an expressive quality in the work. Rather,
Robinson insists, “it is in writing the poem that the poet expresses his emotions;
expressing is not some activity that occurs before the poem is created” (Robinson 2005: 245,
italics in original). And she thinks that Tormey is wrong to suppose that the Romantic
theorists are confusing “expressions of ϕ with ϕ expressions” (Robinson 2005: 247):
e.g. expressions of sadness with the expressive quality of sadness. Classic expression
theory is simply not about expressive qualities at all. One should readily accept that “a
work of art can be an expression of ϕ in the artist without necessarily having ϕ-ish
expressive qualities, as when a poem expresses the poet’s longing but itself does not long
for anything” (Robinson 2005: 250). And what Collingwood and others are trying to
explain is precisely what being an expression of ϕ is.
This, then, is Robinson’s own “new romantic theory”:

If an artist expresses an emotion in a work of art, then

1. the work is evidence that a persona (which could but need not be the artist) is
experiencing / has experienced this emotion;
2. the artist intentionally puts the evidence in the work and intends it to be
perceived as evidence of the emotion in the persona;
3. the persona’s emotion is perceptible in the character of the work;
4. the work articulates and individuates the persona’s emotion; and
5. through the articulation and elucidation of the emotion in the work, both
artist and audience can become clear about it and bring it to consciousness.
(Robinson 2005: 270)

Peter Lamarque

The theory offers a set of necessary conditions for an artist to express an emotion in a
work. Elements of Collingwood’s account are in evidence, notably in clauses 4 and 5,
which emphasise that the work “articulates and individuates” the emotion and also
helps to clarify it and “bring it to consciousness”. Clause 2 stresses that the expression
of emotion is intentional (a deliberate effort with a deliberate aim), in effect not merely
a “spontaneous overflow”.
The most interesting aspect, though, if somewhat problematic, is Robinson’s
concession to the Modernists of the need to postulate a “persona” in a poem to whom
the emotion is attributed. She allows that the persona could be, but need not be,
identical to the author. One attractive feature of her account is the way it seeks to
preserve agency in expression (expression as an “activity”) even while allowing distance
between poet and persona. She uses Keats’ “Ode to a Nightingale” as an example,
identifying one of the emotions expressed in the poem as a “yearning for a timeless
world of art and beauty”. Her official position, as it were, is that this emotion is attributed
to a persona or dramatic speaker. But traces of more traditional Romanticism still
linger in her account:

we should not be too quick to assert that it doesn’t matter whether it is the
author or his persona who is expressing his emotions in a poem. . . . . [W]e are
often quite interested in the fact that Keats himself probably yearned for a time-
less world of art and beauty, and that it is perfectly appropriate to identify the
poetic voice with that of Keats himself. . . . It is not unreasonable to enjoy the
feeling that one is engaging with the actual emotions of a great poet.
(Robinson 2005: 255)

Robinson’s observation in itself is unexceptional. But a problem emerges in the

appeal to “evidence” in clauses 1 and 2. It might be entirely acceptable (in certain
contexts) to treat a poem as evidence of what an author was experiencing or had
experienced. The poem “Tintern Abbey” might be used as biographical evidence of
what Wordsworth felt on his nostalgic return to the Abbey five years after his earlier
visit. No doubt caution of the usual kind about such inferences would be prudent.
That is another issue. But in this case there is a clear distinction between the poem
and the states of mind of the poet: that is implied by speaking of one as evidence
for the other. But does it make sense to speak of the poem as “evidence” of what
the persona was experiencing or had experienced? That seems doubtful, at least
without equivocation in the idea of evidence, precisely because it does not seem
possible to draw a parallel distinction between the poem and the states of mind of
the persona. The poem just is an expression of the persona’s state of mind so there
is no room to distinguish one as evidence for the other, as if there might be some
other evidence that could confirm or count against the poem’s reliability on that
score. The persona is a construct of the poem and has no other existence; the only
sense in which the poem can provide “evidence” about the persona is that it allows
us to construct a picture of the persona, a character that the poem has brought into
being. In the autobiographical case the evidence points to facts outside the poem,
in the persona case the evidence constructs facts internal to the poem. So Robinson
cannot so easily alternate between poet and persona (as in clause 1) with regard to
the role of evidence.


One objection to the Modernist, work-centred view of expression is that it goes too far
in distancing the emotion expressed in a poem from the poet’s own attitudes or states of
mind. If the emotion expressed is attributed only to a persona then it seems not to
matter, for example, if the emotion is sincere. The emotion might be yearning for love,
grief at death, sorrow at suffering, sympathy for failure. But suppose the poet had no such
feelings in reality, perhaps even feelings in antipathy to these. Is that a fault? The
question is a large one. It can be given focus by asking what role remains for the poet
even when a persona is postulated.
It is common to detect attitudes in a poem towards a persona and it would be natural
to attribute those to the poet. Malcolm Budd states: “The poem will give the impression
of having been written by a person of a certain kind” (Budd 1995: 87). Budd calls this
the “implied author”, but allows that our picture of the implied author might rest on
information we have about the real author. He adds that “a poet’s sincerity is not put in
question by a clash between her own outlook and that of the persona of one of her
poems, but only by a lack of harmony between her own outlook and that of the implicit
poet” (Budd 1995: 89).
Another term to describe the attitudes evinced by a poem towards its persona is
“controlling intelligence” (Lyas 1983). Colin Lyas, whose term it is, uses the example of
Pip in Dickens’ Great Expectations. Pip is the narrator and manifests various personal
qualities including at times immaturity and lack of sensitivity. But these are not qualities
of the work itself which, as Lyas notes, is better described as perceptive, sensitive, and
emotionally mature. It is these positive personal qualities that Lyas attributes to the
controlling intelligence of the work. What he seeks to show is that the qualities of
the controlling intelligence are ultimately qualities of the author. He wants to resist the
thought that the author could be merely pretending to possess these qualities just as a
poet might pretend to share the views or the emotions of the persona. Lyas writes:

with respect to many of the personal qualities that we attribute to the controlling
intelligence of a work it makes little sense to assume that an author could
pretend to possess them when he in fact does not. If the kind of quality of
response shown by the controlling intelligence of a work, and hence by the
work itself, is perceptive, sensitive, emotionally mature and the like, there
seems to be little sense in the supposition that the artist has, by an act of pretence,
embodied these characteristics in a work although he himself was not possessed
of them. The judgment that the work is these things is the judgment that the
author there exhibited those qualities (though he might not otherwise exhibit
them in the responses of his or her non-literary life).
(Lyas 1983: 22)

What Lyas offers in his conception of a controlling intelligence is the notion of an

“author-in-a-work”, a notion that offers a nice bridge between work and author.
Robinson’s account might benefit from such a notion as she offers nothing more than the
persona and the author independently characterized. The distinction between the author-
in-the-work and the persona allows us to retain the thought that the persona in a poem
is expressing an emotion while we also attend to the way the emotion is articulated. The
articulation of the emotion in the Ode is Keats’ own and, being an intentional act, reveals

Peter Lamarque

actual attitudes of the author-in-the-work. Recall, however, that for Collingwood the
articulation defines and constitutes the emotion, so now it seems increasingly difficult to
distinguish the articulation from the emotion itself and to insist that the emotion belongs
to the persona and only the articulation to the poet. The controlling intelligence that
informs the articulation is also evident in the emotion articulated.
The literary critic F. R. Leavis’ much-acclaimed essay “Reality and Sincerity”
(Leavis 1952–1953) subtly addresses these issues with a focus on sincerity in poetry.
At its heart the essay offers a comparison between two poems, Emily Brontë’s “Cold
in the Earth” and Thomas Hardy’s “After a Journey”. Leavis’ considered judgment is
that Hardy’s work is the higher “poetic achievement” and his reason is that Hardy’s
poem has more sincerity, being more grounded in “reality”, than Brontë’s. But Leavis’
point is that the case can be made entirely through analysing the poems themselves.
The sincerity he finds is embedded in the language of the poems, not in the biogra-
phies of the poets.
Brontë’s poem begins like this:

Cold in the earth – and the deep snow piled above thee.
Far, far removed, cold in the dreary grave!
Have I forgot, my only Love, to love thee,
Sever’d at last by Time’s all-severing wave?
Now, when alone, do my thoughts no longer hover
Over the mountains, on that northern shore.
Resting their wings where heath and fern-leaves cover
Thy noble heart for ever, ever more?

There is no mistaking what Leavis calls the “emotional intensity” or “plangency” in the
poem or indeed the potential, given the theme, for sentimental wallowing. But Leavis
also notices something else, an “active principle that informs the poem” – manifest in
lines like “Then did I check the tears of useless passion” – namely a “resolute strength
of will [that]. . . counters the run of emotion”. He writes: “We have it in the movement,
in the tough prose rationality, the stating matter-of-factness of good sense, that seems to
play against the dangerous running swell” (ibid.: 91). This “active principle” is the
controlling intelligence, the implicit attitude, that informs the voice of the persona. For
Leavis it saves the poem from mere self-indulgence.
But Leavis thinks there is still something missing, something he describes as “reality”,
in Brontë’s “dramatizing herself in a tragic role”. There is weakness, in effect, in the
controlling intelligence.

The marks of the imaginative self-projection that is insufficiently informed by

experience are there in the poem, and . . . a duly perceptive reader could
discern and describe them, without knowing the biographical fact. They are
there in the noble . . . declamation, and in the accompanying generality, the
absence of any convincing concreteness of a presented situation that speaks for
itself. Locally we can put a finger on the significance of the declamatory mode
in, for example, the last line of the first stanza:
Sever’d at last by Time’s all-severing wave.


The imagery there, or the suggestion of it, is essentially rhetorical; the noble
declamation, the impressive saying, provides the impressiveness, and when we
consider the impressiveness critically we recognize that to respond to the
declamatory mode is to be unexacting in respect of offered imagery: the unrealized
rhetorical-verbal will receive a deference that cannot be critically justified.
Time’s “wave” is of the order of cliché; prompted, it would seem, by “grave”, it
makes as rime – and the closing rime of the stanza – a claim to strength that it
certainly cannot sustain.
(Ibid.: 94)

The focus is on the rhetoric of the poem, the declamatory mode, the generality, “the
absence of any convincing concreteness”. To the extent that there is a lack of integrity
here it shows itself, if Leavis is right, in the articulation of the emotion expressed and
the attitude taken to it. The articulation and attitude belong to the poet, at least the
By way of contrast Leavis finds in the Hardy poem an integrity lacking in the Brontë.
Again he offers fine detail of analysis. Here are some comments on the final four lines:

Trust me, I mind not, though Life lours.

The bringing me here; nay, bring me here again!
I am just the same as when
Our days were a joy, and our paths through flowers.
Not to take the significance of that “Trust me, I mind not” is to have failed to
respond to the complexity of the total attitude, and to have failed to realize the
rare kind of integrity the poem achieves. It is to miss the suggestion of
paradoxical insistence, the intensity of directed feeling and will, in “Nay, bring
me here again”. For what in the bringing him here, he may be supposed to mind
is not the arduousness, for an old man, of the long journey and the ramble by
night. “To bring me here”, says Hardy, “is to make me experience to the full the
desolation and the pang – to give a sharp edge to the fact of Time’s derision. But
I don’t mind . . .’
The rare integrity appears in the way in which the two aspects, the affirmation
and the void, affect us as equal presences in the poem. Vacancy is evoked as an
intensity of absence . . . [but] “After a Journey” closes on the affirmation. But if
“affirmation” is the word . . . it mustn’t suggest anything rhetorical; the affirmation
is dramatic in a quite other sense.
(Ibid.: 96–7)

The integrity or sincerity that Leavis finds in the poem makes no reference to indepen-
dently discovered facts about Hardy. It is an emergent feature of the language in the
poem, a feature that informs the articulation of the emotion expressed, a feature at least
partially attributable to the controlling intelligence manifested in the work. Leavis offers
a further comment which might seem to undermine the kind of analysis pursued: “It is
a poem that we recognize to have come directly out of life; it could, that is, have been
written only by a man who had the experience of a life to remember back through”
(ibid.: 98). But that point, true or not, adds nothing new to the analysis. It is comparable
to Robinson’s observation that we might be quite interested in the fact that Keats

Peter Lamarque

himself probably yearned for a timeless world of art and beauty. What is of greater interest
in both cases is the precise way that the poets chose to articulate the emotions expressed
and the attitudes that informed that expression.
Perhaps, then, there is evidence of convergence, retaining insights from both the
Romantic and Modernist accounts of expression. Robinson and Leavis, in their different
ways, show how this might be achieved. We can shift, as the Modernists demand, from
saying that the poet expresses an emotion in the poem to saying that the poet produces an
expression of an emotion in the poem. But much of the agency that the Romantics demand
in expression is left intact. Producing an expression is an action and authorial agency
informs all aspects of the expression. Collingwood’s view that expression involves the
elucidation and articulation of an emotion is easily refocused in this manner. The
achievement of the Romantic poets lies in their acts of articulation, capturing timelessly
and memorably emotions that could be expressed in no other way, but these acts are not
independent of attitudes and even feelings in the poets themselves.

Austin, J. L. (1962) How To Do Things With Words, Oxford: Oxford University Press.
Barthes, R. (1977) “The Death of the Author”, in R. Barthes, Image-Music-Text, Essays Selected and Translated
by Stephen Heath, London: Fontana/Collins.
Beardsley, M. C. (1970) The Possibility of Criticism, Detroit, MI: Wayne State University Press.
Beardsley, M. C. (1981) Aesthetics: Problems in the Philosophy of Criticism, 2nd edn, Indianapolis: Hackett.
Budd, M. (1995) Values of Art: Pictures, Poetry and Music, Harmondsworth: Penguin Press.
Collingwood, R. G. (1938) The Principles of Art, Oxford: Clarendon Press.
Eliot, T. S. (1972) [1919] “Tradition and the Individual Talent”, in D. Lodge (ed.) 20th Century Literary
Criticism, London: Longman.
Keats, J. (2006) [1820] “Ode on a Grecian Urn”, in D. Wu (ed.) Romanticism: an Anthology, 3rd edn, Oxford:
Blackwell Publishing.
Leavis, F. R. (1952–1953) “Reality and Sincerity”, Scrutiny, 29: 90–8.
Lyas, C. (1983) “The Relevance of the Author’s Sincerity”, in Peter Lamarque (ed.) Philosophy and Fiction:
Essays in Literary Aesthetics, Aberdeen: Aberdeen University Press.
Robinson, J. (2005) Deeper Than Reason: Emotion and Its Role in Literature, Music, and Art, Oxford: Oxford
University Press.
Shelley, P. B. (2006) [1840] A Defence of Poetry, in D. Wu (ed.) Romanticism: an Anthology, 3rd edn, Oxford:
Blackwell Publishing.
Tillyard, E. M. W. and Lewis, C. S. (1939) The Personal Heresy: A Controversy, London: Oxford University
Tormey, A. (1971) The Concept of Expression, Princeton, NJ: Princeton University Press.
Tolstoy, L. (1969) [1897] What is Art?, Oxford: Oxford University Press.
Wordsworth, W. (2006a) [1798] “Lines Written a Few Miles above Tintern Abbey, On Revisiting the Banks
of the Wye during a Tour, 13 July 1798”, in D. Wu (ed.) Romanticism: an Anthology, 3rd edn, Oxford:
Blackwell Publishing.
Wordsworth, W. (2006b) [1800] Preface to Lyrical Ballads, in D. Wu (ed.) Romanticism: an Anthology, 3rd edn,
Oxford: Blackwell Publishing.

Wolfgang Huemer

I Dimensions of “Style”
It should come as a surprise that the notion of style is elusive: after all, style is often
taken to pertain to the surface (rather than the substance) of a (literary) work of art.
The elements of style should, thus, lie open before our eyes and be easy to discern.
Moreover, we all seem to have a clear intuitive understanding of style: we know that it
is essential to the aesthetic dimension of a work and makes it accessible for aesthetic
appreciation, we are familiar with the idea that it allows us to attribute a work to an
author, a genre, a school, or a period, and might even think that it reveals the artistic
personality of the author. Yet we find that there is no universally accepted definition of
“style.” Moreover, the numerous definitions that have been proposed in the past
highlight very different aspects that often stand in contrast to one another, which shows
that style has many faces, is rich of different dimensions and performs a great variety of
functions: it has been characterized as a “dress of thought” that adorns a pre-existing
content; as a choice between alternative expressions; as a set of recurrent, individual, or
collective characteristics; as signature; as an expression of the author’s personality; as a
way of writing dictated by rules or an acquired disposition to act on a set of rules; but
also as a systematic violation of rules or deviation from a norm; it is taken to be an
expression of originality, to manifest a perspective or point of view, or to foreground the
possible uses of a medium and so to draw our attention to the workings of language.
Etymologically the word “style” refers to manners of linguistic expression. It is derived
from the Latin “stilus” that stands for “a stake or pale, pointed instrument for writing,
oral or written style” (The Oxford English Dictionary 1989), but its meaning has broadened
considerably. Nowadays the term is used for characteristic features of groups, schools,
genres, or periods in the history of art, literature, music, or architecture, etc., for ways
of producing works of art or other artifacts, as in writing style or painting style, for styles
of reasoning, ways of performing an action, as in dance style or swimming style, or style of
chess playing, for phenomena related to fashion, personal outfit and lifestyle, as in
hairstyle or dress style, for furniture or design, for ways of conducting interpersonal
relations as in leadership style, for specific formats, for manners, procedures, or skilled
production processes that are typical for an individual, an institution, or a company,
etc.; sometimes it is used without any further qualification—when we say that someone
or something has style.
This short list shows that style is attributed to entities of very different ontological
categories—to persons, schools, periods, ways of doing something, complex patterns of
behavior, works of art and other artifacts, or social entities like companies and institutions.2
Wolfgang Huemer

All these entities are in some way or another related to the performance of actions,
which allows us to say that the bearers of style are ways or manners of performing actions,
the person or the group of persons who perform the actions, or the objects created by them.
In the case of literary works of art, the kind of action that counts is not the actual process
of writing—it is typically not relevant whether a poem was written with a pencil, a pen,
or a typewriter. It is rather “the nature of the choices or decisions the author apparently
made about how the work was to be” (Walton 1987, 84) that determines the style. These
choices, if relevant for the style, manifest themselves in the text. In consequence,
some philosophers suggest that we ought to focus primarily on the style of writing,
while others argue that it is the style of the writing, i.e. the stylistic features of the text,
that matters.
Despite the etymological origins of the word, most philosophical contributions on
style are not primarily focused on literary style—typically they are more likely to discuss
pictorial style. Moreover, the notion of style does not seem to play a central role in the
contemporary debate in the philosophy of literature, which is often more concerned with
problems related to the notions of truth and reference of propositions contained in
fictional, literary texts, i.e. with problems related to the philosophy of language, metaphysics,
or epistemology, rather than with the role of the poetic or aesthetic dimension of
language. In the following pages I will focus on aspects that have been—or should be—of
particular interest for a philosophical discussion of literary style. I am convinced that the
philosophy of literature can only benefit from paying more attention to the stylistic
dimension of literature, for this would allow us not only to gain a new perspective on the
problems that are widely discussed, but also to pay due attention to aspects that are
essential to literature, but are nonetheless often overlooked in the current debate.

II Style as Choice
Style has been defined as choice. Where there is more than one way of performing an
action or achieving a certain goal, style consists in choosing to do so consistently in a
specific manner. These choices are particularly salient when someone opts for doing
something in a manner that deviates from the way it is usually done. In the case of
literary style, linguistically oriented approaches often focus on choices that manifest
themselves in recurrent patterns of linguistic expressions on a lexical, syntactic, phono-
logical, or morphological level (cf. Havránek 1964, 4). According to this view, specific
linguistic patterns of expression can count as stylistic elements of a work or a body of
works when their frequency distributions in the respective works deviate, in a statisti-
cally relevant sense, from that in language as a whole, which implies that style can be
measured empirically—an assumption that has led to the development of quantitative
stylistics. When it comes to the analysis of literary style, however, quantitative methods
can easily seem insufficient or reductive, for the style of literary works of art consists in
choices not only at the level of the linguistic material, but also in choices that determine
“the way of deploying narrative structures, portraying characters, and articulating points
of view” (Eco 2005, 163).
This conception of style embraces two aspects that seem to stand in tension to one
another: originality and recurrence. A linguistic pattern or feature of a text can be indi-
viduated as stylistic element only if it is the result of a deliberate choice to deviate from
the ordinary way of expressing or developing a certain point; style consists, in other
words, in adopting new or unusual ways of expression that stand out against the standard


way of putting it. These ways of expression, however, have to be recurrent features of
the text or the body of texts, otherwise we would not be able to identify them as stylistic
elements that distinguish the work of a particular author or period.
The definition of style as choice presupposes that one and the same goal can, in fact, be
achieved in more than one way. In the case of literary works of art this presupposes that the
medium, language, does actually offer a choice between different formulations that convey
the same literary meaning. A skilled author, so the idea goes, who sufficiently masters her
medium, can deliberately choose one from a number of different, synonymous ways of
expressing the point she wants to make—and she will choose the one that best fits her
aesthetic conceptions, that best displays a specific point of view, or that best expresses her
artistic personality. “The point is,” as Arthur Danto has put it, “that the same substance may
be variously stylistically embodied, and synonymous vehicles may have marked stylistic
differences” (Danto 1981, 198). Accordingly, it has been argued that “[s]ynonymy, in the
widest sense of the term, lies at the root of the whole problem of style” (Ullmann 1957, 6).
This conception rests on the idea, however, that we can separate the content of a literary
work of art, what is said, from how it is said and that “[we] may . . . reserve the term style for
this how, as what remains of a representation when we subtract its content” (Danto
1981, 197). It, thus, invites a dichotomy between form and content of a work and to the
view that style pertains exclusively to the former, a perspective that is explicitly endorsed
by Cleanth Brooks and Robert Penn Warren in their characterization of the term style:

This term is usually used with reference to the poet’s manner of choosing,
ordering, and arranging his words. But, of course, when one asks on what
grounds certain words are chosen and ordered, one is raising the whole problem
of form. Style, in its larger sense, is essentially the same thing as form.
(Brooks and Warren 1950, 694)

This conception, as well as the implied distinction between form and content, is not
without problems, though. Particularly in the case of literary texts there seems to be a
close connection between form and content and it is all but clear that an author could
have expressed the same literary meaning with an alternative formulation that, in every-
day standards, would pass as synonymous. Style is more than a superficial ornament that
could be replaced without altering the essence of the work, it rather contributes to the
meaning of a text, literary or not. Hemingway’s preference for short sentences, for
example, is more than an arbitrary choice and it is hard to imagine that he could have
told the same stories in complex and long-winded sentences. This shows that “stylistic
features,” as Monroe Beardsley has suggested,

stylistic features and hence style