693 Pages • 351,641 Words • PDF • 8.2 MB
+ film+ companion+ Philosophy+ Routledge
Uploaded at 2021-08-15 15:00
Report DMCA
PREVIEW PDF
THE ROUTLEDGE COMPANION TO PHILOSOPHY AND FILM The Routledge Companion to Philosophy and Film is the first comprehensive volume to explore the main themes, topics, thinkers, and issues in philosophy and film. The Companion features sixty specially commissioned chapters from international scholars which are organized into four clear parts: • Issues and concepts • Authors and trends • Genres and other types • Film as philosophy Part I is a comprehensive section examining key concepts, including chapters on acting, censorship, empathy, depiction, ethics, genre, interpretation, narrative, spectatorship, and style. The second part covers authors and scholars of film and significant theories. Part III examines genres such as documentary, experimental cinema, horror, comedy and tragedy. The final part includes chapters on key directors such as Tarkovsky, Bergman and Terrence Malick, and on particular films including Gattaca and Memento.
The Routledge Companion to Philosophy and Film is essential reading for anyone interested in philosophy of film, aesthetics and film, and cinema studies. Paisley Livingston is Professor of Philosophy at Lingnan University, Hong Kong. He is the author of Art and Intention: A Philosophical Study, and coeditor (with Berys Gaut) of The Creation of Art: New Essays in Philosophical Aesthetics. Carl Plantinga is Professor of Film Studies in the Communication Arts and Sciences Department at Calvin College, USA. He is the author of Rhetoric and Representation in Nonfiction Film, and coeditor (with Greg M. Smith) of Passionate Views: Film, Cognition and Emotion.
TF17557.indb 1
16/9/08 15:47:13
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, Second Edition Edited by Berys Gaut and Dominic Lopes The Routledge Companion to Philosophy of Religion Edited by Chad Meister and Paul Copan The Routledge Companion to the Philosophy of Science Edited by Stathis Psillos and Martin Curd 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 Forthcoming: The Routledge Companion to Nineteenth Century Philosophy Edited by Dean Moyar The Routledge Companion to Philosophy of Psychology Edited by John Symons and Paco Calvo The Routledge Companion to Ethics Edited by John Skorupski The Routledge Companion to Metaphysics Edited by Robin Le Poidevin, Peter Simons, Andrew McGonigal, and Ross Cameron The Routledge Companion to Epistemology Edited by Sven Bernecker and Duncan Pritchard The Routledge Companion to Seventeenth Century Philosophy Edited by Dan Kaufman The Routledge Companion to Eighteenth Century Philosophy Edited by Aaron Garrett The Routledge Companion to Philosophy and Music Edited by Andrew Kania and Theodore Gracyk
TF17557.indb 2
16/9/08 15:47:13
THE ROUTLEDGE COMPANION TO PHILOSOPHY AND FILM
Edited by Paisley Livingston and Carl Plantinga
TF17557.indb 3
16/9/08 15:47:14
First published 2009 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 This edition published in the Taylor & Francis e-Library, 2008. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.”
Routledge is an imprint of the Taylor & Francis Group, an informa business © 2009 Paisley Livingston and Carl Plantinga for selection and editorial matter; individual contributors for their contributions All rights reserved. No part of this book may be reprinted or reproduced or utilized 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. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data The Routledge companion to philosophy and film / edited by Paisley Livingston and Carl Plantinga. p. cm. -- (Routledge philosophy companions) Includes bibliographical references and index. 1. Motion pictures. 2. Motion pictures--Philosophy. I. Livingston, Paisley, 1951- II. Plantinga, Carl R. PN1994.R574 2008 791.43--dc22 2008017372 ISBN 0-203-87932-5 Master e-book ISBN
ISBN10: 0-415-77166-8 (hbk) ISBN13: 978-0-415-77166-5 (hbk)
TF17557.indb 4
16/9/08 15:47:14
C ONT ENT S Notes on contributors Preface Acknowledgments
xii xx xxii
PART I Issues and concepts
1
1 Acting JOHANNES RIIS
3
2 Authorship AARON MESKIN
12
3 Censorship SUSAN DWYER
29
4 Consciousness MURRAY SMITH
39
5 Definition of “cinema” TREVOR PONECH
52
6 Depiction ROBERT HOPKINS
64
7 Digital cinema BERYS GAUT
75
8 Emotion and affect CARL PLANTINGA
86
9 Empathy and character engagement AMY COPLAN
97
10 Ethics FOLKE TERSMAN
TF17557.indb 5
111
16/9/08 15:47:14
contents
11 Film as art ROBERT STECKER
121
12 Formalism KATHERINE THOMSON-JONES
131
13 Gender ANGELA CURRAN AND CAROL DONELAN
142
14 Genre BRIAN LAETZ AND DOMINIC MCIVER LOPES
152
15 Interpretation GEORGE WILSON
162
16 Medium KEVIN W. SWEENEY
173
17 Music JEFF SMITH
184
18 Narration NOËL CARROLL
196
19 Narrative closure NOËL CARROLL
207
20 Ontology DAVID DAVIES
217
21 Race DAN FLORY
227
22 Realism ANDREW KANIA
237
23 Spectatorship CARL PLANTINGA
249
24 Sound GIORGIO BIANCOROSSO
260
25 Style NOËL CARROLL
268
vi
TF17557.indb 6
16/9/08 15:47:14
C ON T E N T S
26 Violence STEPHEN PRINCE
279
PART II Authors and trends
289
27 Rudolph Arnheim JINHEE CHOI
291
28 Walter Benjamin STÉPHANE SYMONS
301
29 David Bordwell PATRICK COLM HOGAN
313
30 Bertolt Brecht ANGELA CURRAN
323
31 Noël Carroll JONATHAN FROME
334
32 Stanley Cavell WILLIAM ROTHMAN
344
33 Cognitive theory DAVID BORDWELL
356
34 Gilles Deleuze RONALD BOGUE
368
35 Sergei Eisenstein DAVID BORDWELL
378
36 Christian Metz FRANCESCO CASETTI
387
37 Jean Mitry BRIAN LEWIS
397
38 Edgar Morin DUDLEY ANDREW
408
39 Hugo Münsterberg DON FREDERICKSEN
422
vii
TF17557.indb 7
16/9/08 15:47:14
contents
40 Phenomenology VIVIAN SOBCHACK
435
41 Psychoanalysis RICHARD ALLEN
446
42 Semiotics and semiology JOSEPH G. KICKASOLA
457
43 Wittgenstein MALCOLM TURVEY
470
PART III Genres and other types
481
44 Dogme 95 METTE HJORT
483
45 Documentary CARL PLANTINGA
494
46 Horror AARON SMUTS
505
47 Pornography SUSAN DWYER
515
48 Avant-garde film MAUREEN TURIM
527
49 Tragedy and comedy DEBORAH KNIGHT
536
PART IV Film as philosophy
547
50 Film as philosophy THOMAS E. WARTENBERG
549
51 Ingmar Bergman PAISLEY LIVINGSTON
560
52 Terrence Malick DAVID DAVIES
569
viii
TF17557.indb 8
16/9/08 15:47:14
C ON T E N T S
53 Andrei Tarkovsky ANDRÁS BÁLINT KOVÁCS
581
54 Why be moral? CHRIS FALZON
591
55 Skepticism RICHARD FUMERTON
601
56 Personal identity DEBORAH KNIGHT
611
57 Practical wisdom and the good ground of Gettysburg JOSEPH KUPFER
620
58 The Five Obstructions METTE HJORT
631
59 Gattaca NEVEN SESARDIC
641
60 Memento ANDREW KANIA
650
Index
661
ix
TF17557.indb 9
16/9/08 15:47:14
CONT R IBUT O R S Richard Allen is Professor and Chair of Cinema Studies at New York University. He is author of Projecting Illusion (1995) and Hitchcock’s Romantic Irony (2007) and coeditor of Film Theory and Philosophy (1997) and Wittgenstein, Theory and the Arts (2002) among other books. Dudley Andrew is Professor of Film and Comparative Literature at Yale University, where he directs the graduate film program. Specializing in French film and culture and issues of world cinema, he has written books on the history of film theory and is currently at work on the specificity of the cinematic image and reflections around the writing of André Bazin. Giorgio Biancorosso was a Mellon Fellow at the Society of Fellows in the Humanities at Columbia University in 2001–3 and is now an Assistant Professor in Music and Film Theory, Music Department, the University of Hong Kong. Biancorosso is completing a book, under contract with Oxford University Press, called Musical Aesthetics through Cinema. He has written for the journals Music and Letters, ECHO, and Future Anterior and the collections Bad Music (2004), Wagner and Cinema (forthcoming), and Popular Music and the Postmodern Auteur (forthcoming). Ronald Bogue is Distinguished Research Professor of Comparative Literature at the University of Georgia. His books include Deleuze and Guattari (1989), Deleuze on Cinema (2003), and Deleuze’s Way: Essays in Transverse Ethics and Aesthetics (2007). David Bordwell is Jacques Ledoux Professor Emeritus of Film Studies at the University of Wisconsin, Madison. He has written many books on film aesthetics, including The Way Hollywood Tells It (2006) and Poetics of Cinema (2007). Noël Carroll is a Distinguished Professor of Philosophy at the Graduate Center of the City University of New York. His most recent books are The Philosophy of Motion Pictures and On Criticism. He is presently working on A Short Introduction to Humor. Francesco Casetti, born in Trento, on 2 April 1947, is Full Professor at Università Cattolica of Milano, where he also serves as Chair of the Department of Media and Performing Arts. He is the author of Inside the Gaze: The Fiction Film and Its Spectator (1999), Theories of Cinema, 1945–1995 (1999), and Eye of the Century: Film, Experience, Modernity (2008). Coeditor (with Roger Odin) of the special issue
TF17557.indb 10
16/9/08 15:47:14
C ON T R IBU TOR S
of Communications, 51 (1990) (“Télévisions/Mutations”), he has been a visiting professor at many universities, including, most recently, Yale, and is President of the Association of Film and Television Teachers in the Italian Universities. Jinhee Choi is Lecturer in Film Studies at the University of Kent, UK. She is the coeditor of Philosophy of Film and Motion Pictures (2005) with Noël Carroll and is completing a monograph on contemporary South Korean cinema. Amy Coplan is Associate Professor of Philosophy at California State University, Fullerton. Her research interests include philosophy of emotion, philosophy of film, feminism, and ancient Greek philosophy. She has published articles on various types of character engagement, including empathy, sympathy, and emotional contagion, as well as on Terrence Malick’s The Thin Red Line, the popular television show House, and horror films. She is currently editing a collection on the film Blade Runner for the Routledge series, Philosophers on Film, and co-editing an interdisciplinary collection titled Empathy: Philosophical and Psychological Perspectives (forthcoming). Angela Curran is Assistant Professor of Philosophy at Carleton College, Minnesota. Her work in aesthetics has investigated topics in Aristotle’s aesthetics and the philosophy of film, and her current project is on Aristotle’s account of our emotional engagement with art. She is the coeditor of Philosophy of Film: Introductory Text and Readings (2005). David Davies is Associate Professor in the Department of Philosophy at McGill University, Montréal. He is the author of Art as Performance (2004) and Aesthetics and Literature (2007) and editor of The Thin Red Line (2008) in the series Philosophers on Film. He has published articles on philosophical issues in the philosophies of film, photography, literature, and the visual arts and on topics in metaphysics, philosophy of mind, philosophy of language, and philosophy of science. He is currently working on a book on the philosophical foundations of the performing arts. Carol Donelan is an Associate Professor in the Department of Cinema and Media Studies at Carleton College, Minnesota, where she teaches courses in film history and theory, film modes and genres, directors, national cinemas, and television studies. She has published articles on the politics of gender in cinema and on teaching film theory in a post-film, post-theory era. Her research interests include melodrama and the history of moviegoing and film exhibition in small towns and rural areas. Susan Dwyer is Chair and Associate Professor of Philosophy at the University of Maryland, Baltimore County. Previously she taught at McGill University, Montréal. She specializes in ethics and public policy, moral psychology, and feminist theory. xi
TF17557.indb 11
16/9/08 15:47:14
C ON T R IBU TOR S
Chris Falzon teaches philosophy at the University of Newcastle, Australia. He is the author of Foucault and Social Dialogue (1998) and Philosophy Goes to the Movies (2002, 2nd ed. 2006). Dan Flory is Associate Professor of Philosophy at Montana State University, Bozeman. His publications include several essays on philosophy, film, and critical race theory. He has also guest-edited two volumes of the journal Film and Philosophy. Currently President of the Society for the Philosophic Study of the Contemporary Visual Arts, his most recent publication is Philosophy, Black Film, Film Noir (2008). Don Fredericksen is Professor of Film and Director of Undergraduate Studies in Film at Cornell University and author of a recent book on Bergman’s Persona (2005), coauthor of a book on Wajda’s Kanal (2007). He also practices as a psychotherapist who is convinced, contra Münsterberg, that the unconscious exists. Jonathan Frome is an Assistant Professor of Film and Media Studies at the University of Texas at Dallas. His research focuses on how media generate emotion and why various media tend to generate different emotions. He is also interested in documentary film and animation. His website is www.jonathanfrome.net Richard Fumerton received his PhD from Brown in 1974 and is currently the F. Wendell Miller Professor of Philosophy at the University of Iowa. His research has focused mainly on epistemology, but he has also published in metaphysics, philosophy of mind, philosophy of science, value theory, and philosophy of law. He is the author of Metaphysical and Epistemological Problems of Perception (1985), Reason and Morality: A Defense of the Egocentric Perspective (1990), Metaepistemology and Skepticism (1996), Realism and the Correspondence Theory of Truth (2002), and Epistemology (2006) and the coeditor (with Diane Jeske) of Philosophy through Film (forthcoming). Berys Gaut is Reader in Philosophy at the University of St Andrews. He holds a PhD from Princeton University and is author of Art, Emotion and Ethics (2007) and A Philosophy of Cinematic Art (forthcoming). He is coeditor of Ethics and Practical Reason (1997), The Routledge Companion to Aesthetics (2001, 2nd ed. 2005), and The Creation of Art (2003). He has also published numerous articles on aesthetics, the philosophy of film, ethics, political philosophy, and Kant’s moral philosophy. Mette Hjort is Professor and Program Director of Visual Studies at Lingnan University, Hong Kong. She has published Stanley Kwan’s “Center Stage” (2006), Small Nation, Global Cinema (2005), and The Strategy of Letters (1993). She has also edited and coedited numerous volumes on film. Patrick Colm Hogan is Professor in the Department of English, the Program in Comparative Literature and Cultural Studies, and the Program in Cognitive xii
TF17557.indb 12
16/9/08 15:47:14
C ON T R IBU TOR S
Science at the University of Connecticut. He is the author of ten books, including Cognitive Science, Literature, and the Arts: A Guide for Humanists (2003), The Mind and Its Stories: Narrative Universals and Human Emotion (2003), and Understanding Indian Movies: Culture, Cognition, and Cinematic Imagination (2008). He is also the editor of The Cambridge Encyclopedia of the Language Sciences (forthcoming). Robert Hopkins is a member of the Department of Philosophy in the University of Sheffield. His research interests lie in various topics in aesthetics, epistemology, and the philosophy of mind. As well as work on film, he has published on pictorial representation, the senses, sensory imagining, the epistemology and objectivity of aesthetic judgment, and the aesthetics of painting and sculpture. Andrew Kania’s principal research is in the philosophy of music, film, and literature. He is the editor of Philosophers on Memento, and, with Theodore Gracyk, The Routledge Companion to Philosophy and Music (both forthcoming). He recently won the inaugural Essay Prize of the British Society of Aesthetics. Joseph G. Kickasola is Associate Professor and Director of the Baylor Communication in New York program, Baylor University. He is the author of The Films of Krzysztof Kieslowski: The Liminal Image (2004). His publications include articles in Quarterly Review of Film and Video, Journal of Moving Image Studies, and various film studies anthologies. He lives in New York City. Deborah Knight is Associate Professor of Philosophy at Queen’s University at Kingston, Canada. Her research focuses on philosophy of art, philosophy of film, and philosophy of mind. She has recently published on The Matrix, The Age of Innocence, Dark City, and Blade Runner. András Bálint Kovács is Head of the Film Department at Eötvös Loránd University, Budapest, Director of the National Audiovisual Archive, and Artistic Advisor for Béla Tarr’s production company, T&T Filmmuhely. He has taught at several universities, including the University of Paris (la Nouvelle Sorbonne) and the University of Stockholm. He translated Deleuze’s Cinéma I–II into Hungarian. His books include Screening Modernism: The European Art Cinema 1950–1980 (2007), Les mondes d’Andrei Tarkovsky (1987), Metropolis, Paris (On German Expressionism and the French New Wave) (1992), Tarkovszkij (1997), Film and Narration (1997), Collection of Essays (2002), and Trends in Modern Cinema (2005). Joseph Kupfer is Professor of Philosophy at Iowa State University, teaching and publishing in ethics, social-political philosophy, and aesthetics. Recent work includes aesthetics of nature, virtue theory, articles and book on philosophy in film, Visions of Virtue in Popular Film (1999), as well as a book on virtue and vice in personal life, Prostitutes, Musicians, and Self-Respect (2007). xiii
TF17557.indb 13
16/9/08 15:47:14
C ON T R IBU TOR S
Brian Laetz is a doctoral student at the University of British Columbia. He works in aesthetics and epistemology, and he is writing a dissertation on art and evolution. He also has an interest in genres and has written on horror and fantasy. Brian Lewis is Dean of the Faculty of Applied Sciences, Professor of Communication, and founder of the Leonardo Institute at Simon Fraser University. He teaches film production, documentary media, and film theory and history. His current research looks at emerging communication technologies, together with their public and policy implications. Paisley Livingston is Chair Professor of Philosophy and Dean of Arts at Lingnan University, Hong Kong. He previously taught at McGill University and at Aarhus University and the University of Copenhagen. His books include Ingmar Bergman and the Rituals of Art (1982), Literary Knowledge (1988), Literature and Rationality (1991), Models of Desire (1992), and Art and Intention (2005). He coedited The Creation of Art (2003) with Berys Gaut. Professor Livingston’s Cinema, Philosophy, Bergman: On Film as Philosophy is forthcoming. Dominic McIver Lopes is Distinguished Professor of Philosophy at the University of British Columbia. He is author of Understanding Pictures (1996) and Sight and Sensibility: Evaluating Pictures (2005) and is now completing a book on computer art. Aaron Meskin is Senior Lecturer in Philosophy at University of Leeds. Before moving to Leeds in 2005, he taught at Texas Tech University. He has published articles on aesthetics and the philosophy of art in The Journal of Aesthetics and Art Criticism, The British Journal of Aesthetics, and Philosophy and Phenomenological Research, as well as in a number of collections. He is the editor of the aesthetics/philosophy of art section for the online journal Philosophy Compass, and he recently coedited Aesthetics: A Comprehensive Anthology (2007). He also serves on the Executive Committee of the British Society of Aesthetics and is on the Board of Trustees of the American Society for Aesthetics. Carl Plantinga is Professor of Film Studies at Calvin College, USA. His research interests extend from the history and theory of documentary film, to the nature of film spectatorship, to film theory and the philosophy of film generally. He is the author of Moving Viewers: American Film and the Spectator’s Experience (2009) and Rhetoric and Representation in Nonfiction Film (1997) and coeditor of Passionate Views: Film, Cognition, and Emotion (1999). He serves on the Board of Directors of the Society for Cognitive Studies of the Moving Image and is Associate Editor for Projections: The Journal for Movies and Mind. Trevor Ponech is Associate Professor of English at McGill University, Montréal, and author of What Is Non-Fiction Cinema? (1999). His recent work on the ontology of xiv
TF17557.indb 14
16/9/08 15:47:14
C ON T R IBU TOR S
cinema appears in The Journal of Aesthetics and Art Criticism and The British Journal of Aesthetics. Stephen Prince is Professor of Communication at Virginia Tech. He is the author of numerous books on film, which include Classical Film Violence (2003) and Savage Cinema (1998). His new book project looks at American film in the age of terrorism. Johannes Riis is Associate Professor of Film Studies at University of Copenhagen, Denmark. His publications have centered on film acting and psychologically based film theories. He is currently working on an historical study of acting styles that rely on realist techniques. William Rothman received his PhD in philosophy from Harvard University, where he taught film studies for many years. He is currently Professor of Motion Pictures and Director of the Graduate Program in Film Studies at the University of Miami. His books include Hitchcock: The Murderous Gaze (1982), The “I” of the Camera (1988; expanded edition 2004), Documentary Film Classics (1997) and Reading Cavell’s The World Viewed: A Philosophical Perspective on Film (2000). He is the editor of Cavell on Film (2005). Neven Sesardic is Professor of Philosophy at Lingnan University in Hong Kong. His main interests are philosophy of science and philosophy of biology. His book Making Sense of Heritability was published in 2005, and his articles have appeared in leading philosophical journals, such as The British Journal for the Philosophy of Science, Ethics, and Philosophy of Science. Jeff Smith is a Professor in the Department of Communication Arts at the University of Wisconsin, Madison, and the author of The Sounds of Commerce: Marketing Popular Film Music (1998). His scholarship on film music has appeared in several edited anthologies, including Post-Theory: Reconstructing Film Studies (1996), edited by David Bordwell and Noël Carroll; Passionate Views: Film, Cognition, and Emotion (1999), edited by Carl Plantinga and Greg M. Smith; Music and Cinema (2000), edited by James Buhler, Caryl Flinn, and David Neumeyer; and Soundtrack Available: Essays on Film and Popular Music (2002), edited by Arthur Knight and Pamela Robertson Wojcik. Murray Smith is Professor of Film Studies at the University of Kent, UK. He is the author of Engaging Characters: Fiction, Emotion, and the Cinema (1995) and Trainspotting (2002) and the coeditor of three anthologies: Thinking through Cinema: Film as Philosophy (2006), Film Theory and Philosophy (1997), and Contemporary Hollywood Cinema (1998). He has published widely on the relationship between ethics, emotion, and films, including essays in The Journal of Aesthetics and Art Criticism and Cinema Journal. xv
TF17557.indb 15
16/9/08 15:47:14
C ON T R IBU TOR S
Aaron Smuts earned his PhD in philosophy at the University of Wisconsin, Madison, where he also studied film. He works in a variety of areas in the philosophy of art and ethics, widely construed. Aaron is particularly interested in horror, humor, philosophy of film, analytic existentialism, and well-being. Recently he has written for American Philosophical Quarterly, Asian Cinema Journal, Contemporary Aesthetics, Kinoeye, Film and Philosophy, Film-Philosophy, The Journal of Aesthetic Education, The Journal of Aesthetics and Art Criticism, Philosophy Compass, Philosophy and Literature, and Stanford Encyclopedia of Philosophy. He is currently a Visiting Assistant Professor in the Department of Philosophy at Temple University. Vivian Sobchack is Professor in the UCLA School of Theater, Film and Television. She has published widely in film studies and is author of Screening Space: The American Science Fiction Film (1997), The Address of the Eye: A Phenomenology of Film Experience (1992), and Carnal Thoughts: Embodiment and Moving Image Culture (2004). Robert Stecker is Professor of Philosophy at Central Michigan University. He is the author of Artworks: Definition, Meaning, Value (1997), Interpretation and Construction: Art, Speech and the Law (2003), and Aesthetics and the Philosophy of Art: An Introduction (2005). He has written numerous papers in the areas of aesthetics, the philosophy of mind and language, and the history of modern philosophy. Kevin W. Sweeney is an Associate Professor of Philosophy at the University of Tampa. He is an associate editor of Film and Philosophy. He has published articles on film theory, silent comedy, and horror. Recently he edited an anthology of interviews with Buster Keaton, Buster Keaton: Interviews (2007). Stéphane Symons studied philosophy, liberal studies, and international politics at the University of Leuven, Belgium and the New School for Social Research, New York. He recently finished his doctoral thesis on the philosophy of Walter Benjamin. He is currently employed by the Institute of Philosophy at the University of Leuven and the Research Fund Flanders. His main academic interests are twentieth-century German and French philosophy and aesthetics. Folke Tersman is Chair Professor of Practical Philosophy at Uppsala University, Sweden and has previously taught at the University of Stockholm and University of Auckland, New Zealand. His main interest lies in moral philosophy, and in particular metaethics. His recent publications include Moral Disagreement (2006). Katherine Thomson-Jones is Assistant Professor in the Department of Philosophy at Oberlin College. As well as having published several articles in aesthetics, she is the author of Aesthetics and Film (forthcoming) and the coeditor of New Waves in Aesthetics (2008). xvi
TF17557.indb 16
16/9/08 15:47:14
C ON T R IBU TOR S
Maureen Turim is author of Abstraction in Avant-garde Films (1985), Flashbacks in Film: Memory and History (1989), and The Films of Oshima Nagisa: Images of a Japanese Iconoclast (1998). She has published more than eighty essays in anthologies and journals on a wide range of theoretical, historical, and aesthetic issues in cinema and video, art, cultural studies, feminist and psychoanalytic theory, and comparative literature. Her current book project, entitled Desire and Its Ends: The Driving Forces of Recent Cinema, Literature, and Art, will look at the different ways desire structures narratives and images in various cultural traditions, and the way our very notion of desire may be shaped by these representations. Malcolm Turvey is an Associate Professor of Film Studies at Sarah Lawrence College and an editor of October. He has published widely on film theory and avant-garde film. A book on film theory, Doubting Vision: Film and the Revelationist Tradition, is forthcoming. He is currently at work on a book about European avant-garde film of the 1920s, titled The Filming of Modern Life. Thomas E. Wartenberg is a Professor of Philosophy at Mount Holyoke College, where he also teaches in the Film Studies program. He is the author of, among other works, Unlikely Couples: Movie Romance as Social Criticism (1999) and Thinking on Screen: Film as Philosophy (2007) and the editor of a number of anthologies on the philosophy of film, most recently, Thinking through Cinema: Film as Philosophy (2006), with Murray Smith. His most recent book is Existentialism: A Beginner’s Guide (2008). George Wilson is Professor of Philosophy and Cinematic Arts at the University of Southern California. He has written a book on point of view in film (Narration in Light, 1986) and a book on the philosophy of action (The Intentionality of Human Action, 1989). He has published articles in the philosophy of language, Wittgenstein on rule-following, and film theory and interpretation.
xvii
TF17557.indb 17
16/9/08 15:47:14
P reface The recent surge of interest in film in academia has seen the emergence of film studies as a legitimate academic discipline, and a concomitant rise in film scholarship and courses in film offered at colleges and universities. Yet film has also become a medium of interest in other disciplines, especially in philosophy, which is currently witnessing a “boom” in the general area of philosophy and film. One reason for this trend pertains to the expansion of the topics typically covered within the field of philosophical aesthetics: whereas film was largely ignored in the philosophy of art even twenty years ago, today leading aestheticians have made extensive philosophical investigations of film a central part of their work. Courses are now offered in the philosophy of film; film is regularly discussed in courses in aesthetics; and books examining film from the standpoint of philosophical aesthetics have begun to multiply. Another reason why philosophers are contributing to the burgeoning literature in the area of philosophy of cinema is that many publications in the area of “film theory” emerging from film studies have constituted an implicit or explicit challenge to central philosophical assumptions and methods. To mention two central examples, film semioticians’ statements about the conditions of possibility of cinematic meaning have prodded some philosophers to assess these broad claims in light of arguments and findings from philosophy of language, cognitive science, and linguistics, just as the tenets of psychoanalytic film theory have generated responses from philosophers informed about the philosophy of mind and psychology. Philosophy of cinema has in part emerged, then, as a critical engagement and dialogue with work in film theory, and vice versa. Another significant recent trend has been the emergence of a sizeable literature, authored by professional philosophers, exploring the philosophical significance of individual films and genres. Although the precise sense in which a film can “be” philosophical or contribute to philosophical knowledge is a matter of lively ongoing debate, it is uncontroversial to observe that a range of films, including popular ones, resonate in fruitful ways with traditional and contemporary philosophical issues. Reference to films in the context of the undergraduate philosophy curriculum has the dual payoff of vividly illustrating philosophical issues and generating excitement and discussion about them, both for philosophy majors and nonmajors. The Routledge Companion to Philosophy and Film has been designed to function as a textbook and a reference work for undergraduates, graduate students, and faculty who wish to study cinema and philosophy. The conjunction of film and philosophy in this companion’s title should be understood as embracing both the philosophical study of cinema and the investigations of films’ philosophical dimensions, implications, and pedagogical value. Given the open-ended and interdisciplinary nature of this emerging field of inquiry, our target readership is not comprised exclusively or even primarily of philosophers.
TF17557.indb 18
16/9/08 15:47:14
PR EFAC E
Our ambition has been instead to build a bridge between philosophers working in this area and film students and scholars with theoretical or philosophical inclinations. The fact that philosophical scholarship on film often goes ignored in film and media studies is an unfortunate consequence of the artificial boundaries often drawn between the disciplines. This volume has been designed to bridge this gap by aiming the entries at students and faculty in both disciplines, and at those from other disciplines who are simply interested in the topics presented. For that reason the volume’s entries cover topics and thinkers familiar to philosophers and to film and media scholars, with the hope that further mutual interests will develop. These general ambitions orient the volume’s division into four parts. The first part, “Issues and concepts,” provides detailed coverage of the most central questions and concepts in the area. Philosophers and film theorists writing these entries offer a systematic, argumentative perspective on these key topics, well-grounded in knowledge of the history of the literature, yet not organized or presented in the form of historical narratives or exegetical discussion. The second part, “Authors and trends,” provides crucial information in the form of intellectual histories and author studies. This part was designed to provide incisive and crisp surveys of the central contentions of the most influential philosophers and theorists who have contributed significant generalizations about the cinema. Part III, “Genres and other types,” deals with some of the categories of films that have been salient to filmmakers, audiences, and critics, as well as to philosophical commentators. While there could be no attempt to be comprehensive here, the entries cover some of the most salient of the genres and types that have elicited philosophical interest. The fourth part of the Companion, “Film as philosophy,” provides a number of selected “case studies” focusing on individual films, filmmakers, and philosophical themes in films. While this section is obviously not designed to offer detailed coverage of all of the philosophically significant moments in the complex history of world cinema, it does provide examples of the ways in which films have philosophical significance, and these entries also provide the reader with examples of philosophically oriented interpretations of films. With eleven entries, the aim is not comprehensiveness, but rather the provision of significant case studies demonstrating the usefulness of film as a source of and stimulus to philosophical analysis. It is our hope that this part may be of special pedagogical use in the undergraduate curriculum. The editors of this Companion have no illusions about having covered all of the relevant topics falling beneath the broad rubric of philosophy and film. The topics actually covered were usually the result of choices rooted in principle but also depended on the availability of suitable authors willing (and able) to work within our timeframe, and on other practical considerations. It is our hope that our contributors have provided a great deal of insight into the subjects covered, as well as a point of entry to many other topics in this wide and rapidly expanding field.
xix
TF17557.indb 19
16/9/08 15:47:14
Acknowledgments Paisley Livingston is very thankful to Carl Plantinga for inviting him to join him in this project. Work on the book was supported by two different grants from the Research Grants Council of the Hong Kong Special Administrative Region, China: Paisley’s research on Ingmar Bergman and philosophy was supported by a CERG (Competitive Earmarked Research Grant) (Project No. LU3401/06H), and work on the collection more generally was funded by an Academic Programme Research Grant (DA07B7). Paisley is very grateful for this support. Carl Plantinga is thankful to Calvin College for supporting this book in the form of a Calvin Research Fellowship and for a leave of absence for the fall of 2007. He also thanks Lingnan University, and especially Mette Hjort, for the opportunity to become a visiting scholar at Lingnan during that time. We are both grateful to Solly Leung for his dutiful assistance. Thanks as well to James Thomas for careful copy-editing work.
TF17557.indb 20
16/9/08 15:47:15
Part I
Issues and Concepts
TF17557.indb 1
16/9/08 15:47:15
TF17557.indb 2
16/9/08 15:47:15
1
Acting Johannes Riis The centrality of acting to film narratives raises several issues, some related to the ontology of actors and characters, others to realism and the role of acting styles, and others pertaining to the performance’s contribution to a film’s artistic merit. This entry surveys some of the most influential and pertinent ideas and issues related to these topics.
Acting in images In film, the value of acting depends on the abilities of others, such as cinematographers and editors; to probe deeper into the nature of acting’s contribution we need to see performances relative to images and the sound track. First, our experience of a performance as a part of the film implies that acting causally affects the image and sound track. Computer-generated images, which are based on motion or performance capturing, illustrate even more pointedly the relation of acting to images. Andy Serkis’ performance as Gollum in the trilogy of The Lord of the Rings is a case in point. Even though we look at a fantasy creature with an outward appearance created from scratch at a computer, not unlike the object of a painting, Gollum’s quirky movements and gestures and his uncanny postures all rely on Serkis’ acting technique. The idea that photographic pictures are fundamentally different from other pictures has been highly contested, and it is clear that moving pictures share with other pictures many aesthetic properties. A painter and a film director who wish to depict a certain posture will face similar decisions concerning, for example, how to frame, compose, and light the figure, and the painter may instruct a model to hold a certain pose, not unlike the director who asks the actor to behave in a certain manner. We are looking at images in both cases, and they both rely on the techniques of a painter and a director; yet one of them also relies on acting technique as the actor moves within the picture frame. Kendall L. Walton’s concept of transparency may shed light on the kind of causal relation that is possibly at work in images of a performance (Walton 1984). Elaborating upon an idea first expressed by André Bazin, Walton points out that photographic pictures will show what happened in front of the camera and thereby manifest a
TF17557.indb 3
16/9/08 15:47:15
JOH A N N E S R IIS
counterfactual dependency between the content of the picture and the objects in front of the camera, whereas a painting depends on what the painter believes he or she has observed and what he or she intends to be the pictorial content. Second, acting implies that someone plays a role as part of a fictional narrative. It may be defined, as James Naremore has put it, as “a special type of theatrical performance in which the persons held up for show have become agents in a narrative” (Naremore 1988: 23). Other kinds of performances, such as singing on stage, might also entail a narrative function, but acting entails that the person “held up for show” tries to enact the role of agent in a narrative. What acting shares with other kinds of performances is the evaluation by the audience: the question is whether the claim to center stage, as it were, is justified by the performer’s abilities. Third, the performance is carried out for the purpose of a film audience. Public service television is in some countries used to broadcast successful stage performances at large theaters, but this is an example of performance as an object of distribution in an audio-visual medium. In film performances, we have distinct expectations concerning the relation of acting to film technology. Comparative studies of scenes which are based on the same play and use cinematography in similar ways show that a performance can be more or less “calibrated” to framing and camera movements (Jacobs 1998), and small nuances of performance can alter a remake (of an earlier film) in thematically important ways (McDonald 2004: 27–32). Acting is an integral and distinct part of the work in question. Historically, a prevalent use of film technology has allowed for counterfactual dependency on acting, but only empirical study can reveal the extent to which acting accounts for valuable properties such as character expressiveness. Nevertheless, the pictorially inclined may point out that our experience of a performance is affected, for example, by the use of editing, and that acting therefore is a kind of raw material for the editor, subject analogously to the so-called Kuleshov effect. Although the premise is true, it does not follow that editing constructs the expressive content of acting, and the Kuleshov effect should not be trusted as proof in this regard. Lev Kuleshov, an editor and director in post-Revolutionary Russia, did an experiment in which a close-up of the actor Ivan Mozhukin was intercut with various objects. The reactions of the spectators who viewed the intercut shots ostensibly showed that it was the editing, rather than Mozhukin’s acting, that created meaning. The footage used in the Mozhukin experiment has been lost, but two similar experiments of Kuleshov indicate that he used this preexisting footage for pedagogical purposes, to illustrate cutting on eye-lines, rather than to prove a theoretical point (Tsivian et al. 1996: 359). We should be skeptical of any references to the Kuleshov experiment with regard to acting and expressiveness. First, it is often assumed that Mozhukin took part in what could have been a semi-scientific experiment, when in fact, by the time the “experiment” was conducted, he had fled the country, due to the Revolution (see Albéra 1995: 76). Second, it is often assumed that Mozhukin was inexpressive and that it therefore was the editing that created the illusion that he was emoting. Yet it is rather implausible that Kuleshov would have picked an inexpressive performance by Mozhukin, even if he could find one. An inexpressive glance in a close-up does not cut easily with shots 4
TF17557.indb 4
16/9/08 15:47:15
AC T I NG
of other objects, since the spectator has no reason to ask for an offscreen cause of an expression. We are better off with a more recent model of how acting and editing may work together, supplied by Noël Carroll in his theory of point-of-view editing. Carroll implies that facial expressions are somewhat ambiguous. To communicate emotions in a precise manner in cinema, the film structure known as point-of-view editing has been developed (Carroll 1993).
Dualism of actor and character Acting offers spectators the pleasure of letting us see and hear the artistic agent as part of the film, but it remains unclear whether we can claim to watch an actor and a character at the same time. According to most accounts, our awareness of the performer need not interfere with the spectator’s comprehension of or immersion in the narrative. Naremore provides a striking example from The King of Comedy (1983), in which the extras on location (or, alternatively, the fictional bystanders) stop to watch celebrity Robert De Niro and Sandra Bernhard (or their fictional characters). Thus, filmmakers may play upon our ability to notice nuances of role-playing to dramatize the theme of celebrity in yet another form, leaving the spectator to linger over the ambiguities of identity and role-playing (Naremore 1988: 285). This kind of experience presupposes a distinction between what George M. Wilson calls photographic and what he calls dramatic representation. According to Wilson, “there is a complex, dynamic interaction between these types of representation which makes it impossible, in analyzing a film, to unfuse the interaction, to treat them as discernibly separate and distinct” (Wilson 1986: 140). Established practices in the film industry, such as casting according to type and the promotion of certain actors to stardom, exemplify how the interaction often works. For instance, we may analyze the role of star image in a performance by looking for a selective use, a problematic or a perfect fit (Dyer 1979: 143–9). One model for understanding type casting and star images is to assume that we form a composite or extract that finds its basis in particularly powerful roles. Thus, in a discussion of Humphrey Bogart, Stanley Cavell suggests that Bogart’s screen personality, as established across a range of roles, became so powerful as to render character names subordinate; “ ‘Bogart’ means ‘the figure created in a given set of films’ ” (Cavell 1979: 28). There may be other reasons than good acting for attaining stardom; numerous star studies have examined individual stars and their historical and industrial context (see, for example, Studlar 1996; Basinger 2007). Underlying the actor and character dualism is a necessary distinction between two sets of actions and attitudes. Acting, as noted initially, means that a performance is carried out for the benefit of an audience, and that the performer becomes an agent in a narrative. This means that we cannot make the mistake of attributing a fictional murder to an actual actor; character beliefs and desires are kept distinct and separate. Conversely, it means that we may admire an actor’s performance, even though we do not approve of morally repulsive actions his or her character performs. 5
TF17557.indb 5
16/9/08 15:47:15
JOH A N N E S R IIS
Similarly, actor and character do not share an emotional experience, even when the actor uses realist techniques. For instance, the American actor Lindsay Crouse contends that good acting occurs when “what is going on in the scene dovetails exactly with something that you [the actor] have to do . . . it is your life in that moment” (quoted in Zucker 1995: 20). Thus, a fictional event in a scene comes to stand for an actual event that she wishes to perform. However, that does not mean that actor and character share an emotional experience. An emotion which has the concern for role and audience as part of its content – the actor’s emotion – is most plausibly viewed as different from one that does not. An actor’s emotional experience is bound to reflect that he or she is playing a role, since what the role requires (or offers, demands, etc.) is the origin of the need for applying a realist technique in the first place. The question is how we may acknowledge two distinctive sets of actions, beliefs, and desires, as well as a constant interaction between photographic and dramatic representation. I suggest we distinguish between two ways of understanding the actor and character dualism and, for lack of better terms, I will call them the duck/rabbit and the realist model. According to the duck/rabbit model, actor and character are distinct and separate because they result from mutually exclusive perspectives, and we can only see one at a time, not unlike the famous trick-drawing of a duck and a rabbit. That is, we see in parts of the picture either an actor or a character, two distinct agents: pictorial information is anchored to one agent from the very outset. The distinction between two different pictorial contents leaves unexplained how an actor’s previous roles can make themselves felt, but one might hypothesize that properties drift from one agent to the other because actor and character are identical in outward appearance in the images. For instance, Stephen Heath argues that “expression” or “figure of acting” is the result of circulation between several levels: agency of action; the actor, who supplies the body; the actor’s image; the character as a specific individual (Heath 1981: 179–82). The question, however, is whether Heath makes too much of the fact that we look at a picture, since Heath’s levels could be applied to any agent. For instance, I may describe myself as an agent; I have a body, I can pretend to be someone else, and other people form an image of me. In this case, we would not necessarily claim that meaning circulates as Heath describes it. According to a realist model the spectator stores information watched in the film with the photographed actor. Instead of being seen as belonging to a fictional entity, an actor’s character beliefs and desires can be viewed as the actor’s pretense. In other words, acting (or dramatic representation) might be viewed as a certain mode in which actions can be carried out. Not unlike everyday life’s playful actions or use of irony, it is a benign and nondeceptive way of carrying out actions. We are unlikely to attribute what we might call pretense-actions to a nonexisting agent, for the same reason that we keep track of the agent who uses irony or does an imitation of us in everyday life. Whereas the duck/rabbit dualism implies a kind of perceptual mistake when we are affected by previous roles or star promotional material, the realist model can view previous roles as ways of setting up expectations of the actor, given a new role and new 6
TF17557.indb 6
16/9/08 15:47:15
AC T I NG
set of circumstances. However, the realist model may face difficulties when encountering roles played by multiple actors. In, for instance, Palindromes (2004), we assume continuity of character beliefs and desires while recognizing eight distinct actresses; arguably, we assume that each actor has an intention to play the protagonist from where the previous one left off.
Technique When trying to assess the value of actors’ contributions, we are probably better off looking for performance technique rather than, for instance, asking questions about authorship. The script and the director’s choices set up a framework within which the actor works, not unlike the constraints facing, for example, musicians and ballet dancers. Acting technique could be viewed, simply, as that which allows the actor to perform a given role, or parts of it. Admittedly, such an inclusive definition does not distinguish between roles that are demanding and those that appear easily performed. Technique, however, can be taken to mean many things. A role can showcase skills that we would not normally take to be part of a contemporary acting school’s core curriculum but are valued by spectators; this is clear, for instance, in musicals and martial arts films. A role can also entail the kind of actions that we are used to performing in our everyday life, for instance, driving a car or opening a door, and such actions can appear anything but artful. To account for the way in which mundane actions can sometimes be considered aesthetically valuable properties of a performance, we are probably best served with an inclusive and nonnormative view of acting technique. Film performances need not be technically demanding to serve the role and the film; nonprofessional acting is a case in point. The French director Robert Bresson aimed for the portrayal of character actions as automatic and habitual – an acting style that often contrasted strikingly with what was at stake in his narratives. Bresson therefore preferred amateurs, whom he would ask to repeat an action over and over, until their relation to objects and characters “were right because they will not be thought” (Bresson 1977: 12). When a director works with child actors, Vsevolod Pudovkin recommends (drawing on his experience in cinema in the 1920s and early 1930s) that they be given a counterinstruction in the middle of a take in order to deliver a genuine surprise or confusion, thereby creating what he calls real-life conditions (Pudovkin 1954: 119). Although these kinds of performances feed the idea that technique is altogether unnecessary in film acting, they can be taken to demonstrate that a limited technical repertory is sometimes sufficient to perform the role in a style appropriate to a particular scene or film. Technically demanding roles may require that the actor combine deliberate control of gestures and speech patterns with an emotional experience. To communicate the role’s state of mind in a clear and characteristic way, the actor may develop a certain manner of speaking or gesturing; an emotional experience can then serve to remove signs of volition and suggest that emotions underlie the gesture or line. The highly influential Konstantin Stanislavsky has written extensively on actors’ means for 7
TF17557.indb 7
16/9/08 15:47:15
JOH A N N E S R IIS
fostering emotions and conveying character, based on his groundbreaking work as a stage director in the early twentieth century (Stanislavsky 1987, 1989). In his chapter on scenic speech in Creating a Role, he advises the actor to distinguish between logical and psychological pauses. In a logical pause, the verbal meaning of a line is rendered lucid by demarcating a thought, whereas in a psychological pause the actor brings a subtext to the following line or word by using emotion-provoking imagery. In the same chapter, Stanislavsky also gives advice on how to deliver a soliloquy in a manner that will give the impression of increasing force without sacrificing clarity, by means of upward steps in tone of voice. This technique, along with psychological and logical pauses, is recognizable in, for example, Laurence Olivier’s delivery of soliloquies in Hamlet (1948). Other acting theorists and pedagogues have downplayed the need for deliberate control and careful execution of a plan, thus giving the actor more room to let emotions affect their performance. One of the most prominent is Lee Strasberg, who claimed that in a performance the actor should be “really thinking something that is real to him at that particular moment” (Strasberg 1991: 46). Whereas Stanislavsky felt that getting too personally involved would jeopardize the buildup toward a climax, as well as the capacity for repetition (Stanislavsky 1989), Strasberg’s approach is possibly more suited for cinema (Carnicke 1999, see also Carnicke 1998). One of Strasberg’s coteachers and an important twentieth-century director, Elia Kazan, developed strategies for maintaining control of structure and climaxes, while getting performances that have captivated audiences in large part because the actors, most notably Marlon Brando and James Dean, did not concentrate on the execution of a deliberate plan. Through improvisations and by relating “the events in the material to their own life,” Kazan ensured that actors had an intense emotional experience (Kazan 1999: 81, 137). Stretching the emotional moment by cutting back and forth between close-ups, the climax of the scene is controlled at the editing table rather than in the performance alone (Kazan 1999: 70–1). By attending to what actors, directors, and acting instructors prefer in terms of methods, exercises, and training, we gain a heuristic for analyzing a given performance. Unless one reads what Sergei Eisenstein has to say about expressive movement, one is unlikely to describe the performance by Nicolai Cherkasov, in the final scene of Ivan the Terrible I (1944), as a series of bodily conflicts, in which the actor works against gravity and the inertia of a prior movement (Eisenstein 1988: 52–6). Once we recognize the technique, however, Eisenstein’s description goes a long way toward explaining what lends the performance a sense of both artificiality and forcefulness. The conflict of movement taxes Cherkasov’s physique to an extent that it would be impossible to simultaneously employ imagery with psychological pauses, and the scene is a fine example of a technique first developed by Vsevolod Meyerhold in a “nonempathetic” approach to acting that was later pursued also by Bertolt Brecht (see also Leach 1993). An actor can use different techniques for different roles, even different scenes, and two actors can come up with similar techniques, even though only one of them has been trained for it. In other words, we also have to rely on traditional film analysis 8
TF17557.indb 8
16/9/08 15:47:15
AC T I NG
and a critical eye. A case in point is André Bazin’s comparison of the acting style that Humphrey Bogart developed in the early 1940s, with what Bazin calls the Kazan school of acting. If we suppose that actors face common constraints, some of them due to the psychological mechanisms at work in emotions, we should also expect different performances to draw on similar techniques. In this perspective, the technique possibly at work in the performances of Bogart and the Kazan school serves to convey what Bazin saw as a shared sense of interiority (Bazin 1985: 100). By attending to acting techniques, we may also become better at understanding styles. Techniques are applicable to different roles by different performers, and when this occurs, there will be interesting differences and similarities that will lead us to speak of, say, a group style or period style. More importantly, we need to discern when different performances reflect differences of style rather than, for instance, differences of role or narrative. Stylistic terms, as pointed out by Rudolf Arnheim, can be arrived at in two ways: either by means of a bottom-up approach in which we identify characteristic and recurrent patterns in a predefined empirical body, or by means of a hypothesis-driven approach in which we assume the existence of certain constraints which artistic choices help overcome (Arnheim 1986: 264). For example, we may look for uses of logical and psychological pauses in trying to identify an actor’s characteristic style; to explain a certain use of camera and editing, we may hypothesize that realist acting techniques are difficult to use because especially intense emotions easily distort the structure of the whole. By looking for techniques, we need not assume that all acting styles are historically contingent codes for representing characters in a world similar to ours. Not all acting styles aim at realism in this sense, and some that do not may still successfully engage the audience relative to an agent in a narrative. Prior to and well into the nineteenth century, a stage actor was expected to play a role as though he or she were a speaker, someone able to move the audience by use of eloquent, oratory techniques, for instance, illustrating words in gesture (Christiansen 1975: 196). Good acting in such styles relied on “how the actor negotiates the intermediate stages of the gesture to reach its limits, or how the actor makes effective transitions from one gesture to another,” as well as the ability of the actor to adjust to musical accompaniment (Mayer 1999: 18). Early film acting bore traces of performance techniques that first served to engage a theater audience. In the 1910s and 1920s, actors would use broad gestures and extended postures in order to stretch a narrative climax, as though the spectator were momentarily looking at a still picture, in what has been coined a pictorialist style (Brewster and Jacobs 1997). In the early 1930s, predefined, rather mask-like expressions – which were originally used to empower a stage actor’s eloquent speech by making its emotional content unequivocal – could be applied to film roles that exemplified high moral standards (Riis 2004). In studying acting, it is important that we do not take for granted what we might call a behaviorist thesis, according to which the actor just behaves as called for in the script and as suggested by the director. This assumption will soon have us focusing too much on the techniques of casting actors, framing, and editing, at the expense of acting. Perhaps adequate to technically undemanding roles, a behaviorist 9
TF17557.indb 9
16/9/08 15:47:15
JOH A N N E S R IIS
assumption does not serve the appreciation of most acting in film. It is through the actor’s balancing of deliberate control and emotional experience, a balancing which can take a variety of forms, that the ability to convey the emotional life of characters is achieved. This ability is at the heart of good acting. See also Rudolph Arnheim (Chapter 27), Bertolt Brecht (Chapter 30), Noël Carroll (Chapter 31), Digital cinema (Chapter 7), Emotion and affect (Chapter 8), Sergei Eisenstein (Chapter 35), Realism (Chapter 22), and Style (Chapter 25).
References Albéra, F. (1995) Albatros: des Russes à Paris, 1919–1929, Milano: Mazzotta; and Paris: Cinémathèque francaise. Arnheim, R. (1986) “Style as a Gestalt Problem,” in New Essays on the Psychology of Art, Berkeley: University of California Press. Basinger, J. (2007) The Star Machine, New York: A. A. Knopf. Bazin, A. (1985) “The Death of Humphrey Bogart,” in J. Hillier (ed.) Cahiers du cinéma: The 1950s, London and New York: Routledge and Kegan Paul. Bresson, R. (1977) Notes on Cinematography, trans. J. Griffin, New York: Urizen Books. Brewster, B., and Jacobs, L. (1997) Theatre to Cinema: Stage Pictorialism and Early Film, Oxford and New York: Oxford University Press. Carnicke, S. M. (1998) Stanislavsky in Focus, Amsterdam: Harwood Academic Publishers. —— (1999) “Lee Strasberg’s Paradox of the Actor,” in A. Lovell and P. Krämer (eds.) Screen Acting, London and New York: Routledge. Carroll, N. (1993) “Toward a Theory of Point-of-View Editing: Communication, Emotion, and the Movies,” Poetics 14: 123–41. Cavell, S. (1979) The World Viewed: Reflections on the Ontology of Film, Cambridge, MA: Harvard University Press. Christiansen, S. (1975) Klassisk skuespilkunst. Stabile konventioner i skuespilkunsten 1700–1900, Copenhagen: Akademisk Forlag. Dyer, R. (1979) Stars, London: British Film Institute. Eisenstein, S. M. (1988) “The Montage of Film Attractions,” in R. Taylor (ed.) S.M. Eisenstein: Selected Works, vol. 1: Writings, 1922–34, Bloomington: Indiana University Press. Heath, S. (1981) Questions of Cinema, Bloomington: Indiana University Press. Jacobs, L. (1998) “Keeping Up with Hawks,” Style 32: 402–26. Kazan, E. (1999) Kazan on Kazan, ed. J. Young, London: Faber and Faber. Leach, R. (1993) “Eisenstein’s Theatre Work,” in I. Christie and R. Taylor (eds.) Eisenstein Rediscovered, London and New York: Routledge. Mayer, D. (1999) “Acting in the Silent Film: Which Legacy of the Theatre?,” in A. Lovell and P. Krämer (eds.) Screen Acting, London and New York: Routledge. McDonald, P. (2004) “Why Study Film Acting: Some opening Reflections,” in C. Baron, D. Carson, and F. P. Tomasulo (eds.) More Than a Method: Trends and Traditions in Contemporary Film Performance, Detroit: Wayne State University Press. Naremore, J. (1988) Acting in the Cinema, Berkeley: University of California Press. Pudovkin, V. I. (1954) Film Technique and Film Acting, trans. I. Montagu, London: Vision Press. Riis, J. (2004) “Naturalist and Classical Styles in Early Sound Film Acting,” Cinema Journal 43: 3–17. Stanislavsky, K. (1987) Creating a Role, trans. E. R. Hapgood; ed. H. I. Popper, New York: Theatre Arts Books. —— (1989) An Actor Prepares, trans. E. R. Hapgood, New York: Routledge. Strasberg, L. (1991) “A Dream of Passion: The Development of the Method,” in J. Butler (ed.) Star Texts: Image and Performance in Film and Television, Detroit: Wayne State University Press.
10
TF17557.indb 10
16/9/08 15:47:15
AC T I NG
Studlar, G. (1996) This Mad Masquerade: Stardom and Masculinity in the Jazz Age, New York: Columbia University Press. Tsivian, Y., Thompson, K., and Tsivian, E. (1996) “The Rediscovery of a Kuleshov Experiment: A Dossier,” Film History 8: 357–64. Walton, K. L. (1984) “Transparent Pictures,” Critical Inquiry 11: 246–77. Wilson, G. M. (1986) Narration in Light: Studies in Cinematic Point of View, Baltimore, MD: Johns Hopkins University Press. Zucker, C. (1995) Figures of Light: Actors and Directors Illuminate the Art of Film Acting, New York: Plenum Press.
11
TF17557.indb 11
16/9/08 15:47:15
2
Authorship Aaron Meskin Directors play a significant role in ordinary thought and talk about film. We often choose to watch films on the basis of who directed them (“I’ll see anything Tim Burton directs”), and we commonly evaluate and interpret them in light of other works in the director’s oeuvre (“It’s the best he’s directed since Edward Scissorhands [1990], and it shows that his films have not, in fact, become more sentimental”). We also compare one director’s oeuvre with another’s (e.g., Hawks’ films with Ford’s and Lucas’ movies with those of Spielberg). Perhaps even more basically, we regularly identify films by reference to their director (“Have you seen the new Scorsese film?”). It is also a commonplace that the director is typically the most important figure involved in the making of a film. As Bordwell and Thompson put it, “Within most film industries, the director is considered the single person most responsible for the look and sound of the finished film” (Bordwell and Thompson 1993: 13). And it is natural to think that it is precisely because the director is typically so central to the making of films that it makes sense to talk and think about them in the ways sketched above. These considerations might tempt one to think that film directors are enough like literary authors that we may gain significant insight into the workings of film by (1) borrowing our best theories of authorship from the literary domain and applying them to cinema; and (2) identifying film directors as typically standing in the authorship role. Perhaps, as Ernest Hemingway stands to the novel To Have and Have Not (1937) so Howard Hawks stands to the 1944 film of the same name. Moreover, there is a wellestablished film-critical tradition that does just this. From the defense of la politique des auteurs by figures such as François Truffaut to Andrew Sarris’ articulation of the auteur theory, a wide range of critics and theorists have explicitly treated many of the major directors of mainstream and “art” cinema as authors (Truffaut 2000; Sarris 2000, 2003). But although this might eventually prove to be a fruitful strategy, a number of considerations should give us pause before pursuing it. First, although it is certainly true that directors figure heavily in ordinary discourse about film, they are not the only ones who play a significant role in our talk and thought about cinema. The category of the “Julia Roberts movie” seems just as robust and important to our thought as that of the “Altman film.” And Rudolf Arnheim suggested that in certain cases, such as Anna Christie (1930), “we have indirect
TF17557.indb 12
16/9/08 15:47:15
AU T HOR SH IP
experimental proof that the leading actress was the main author of the film,” because without Garbo there would have been “a totally different film” (Arnheim 1997: 68). (See Dyer [1979] for an influential work that highlights the importance of film stars.) In certain other cases, screenwriters seem to be especially important. Being John Malkovich (1999) and Eternal Sunshine of the Spotless Mind (2004) are as well known as Charlie Kaufman films as they are as by their directors Spike Jonze and Michel Gondry, respectively. Moreover, screenwriters literally produce written texts, so their claims to be the authors of films seem – at least in one way – on firmer ground than those of directors. (For an early defense of the view that the screenwriter “is the primary creative source,” see Koch [2000].) In still other cases – especially but not only during the heyday of the Hollywood studio system – producers such as David O. Selznick, as well as film companies such as Hammer Film Productions, were especially significant. (See Schatz [1988] for an key work emphasizing the importance of studios and producers.) The Academy Award for Best Motion Picture is, after all, given to the producer or producers of the film! So it is not at all clear that directors really are the only – or even best – candidates for being cinematic authors. Second, film is simply a very different sort of thing than literature. Even though texts go into the making of most films, films themselves are not linguistic texts. There is no language of cinema in any literal sense of “language” (Currie 1993, 1995a). And films are typically (though not essentially) made collaboratively – in most cases by very large groups. Literature – on the other hand – is at least primarily a matter of texts (oral literature may be the exception) and is typically (though not essentially) produced by individuals. (See Stillinger [1991] and Inge [2001] for dissenting views about literature.) Pace certain semiotic approaches then, literature seems a poor model for understanding the cinema, and adapting theories of the literary author to the cinematic case may seem especially misguided. In addition it is true that the term “author” is not usually applied to the makers of films in colloquial English. Perhaps, then, films have no authors at all or, at least, no authors in any literal sense of the word. Skepticism about the very idea of film authorship might also be engendered by a few more considerations. There is the broadly genealogical worry that the very idea of cinematic authorship and the director-as-author thesis largely appear to be fairly recent products of the aforementioned film-critical movements. A reminder that directors didn’t always seem to be as important as they do now, and that early advocates of auteurism sought to elevate the status of certain directors in order to advance the causes of film and film criticism, may undercut the apparent naturalness of the director-as-author position. There may also be worries that the very notion of authorship or, at least, the traditional notion of authorship is only really applicable in the sphere of the high or fine arts and, hence, is somehow inappropriately applied to film, which is often characterized as a popular or mass art. For example, in a discussion of television authorship, Rosalind Coward suggests that “the study of the author seems a peculiarly limiting way of approaching the mass media” (Coward 1987: 79). Pauline Kael famously excoriated auteur theorists for following a rigid and misguided approach to film criticism, criticizing them for “their truly astonishing inability to 13
TF17557.indb 13
16/9/08 15:47:15
A A RON M E SK I N
exercise taste and judgment within their area of preference” (Kael 2005: 109). Any approach to film that leads to such misguided critical evaluation looks questionable. More importantly, there is a significant current of thought that is decidedly skeptical about authorship in any domain. Roland Barthes famously proclaimed “the death of the author” (Barthes 2002), and another famous French theorist, Michel Foucault, optimistically predicted a time when even “the author function will disappear” (Foucault 2002: 22). Although these two authors (!) were primarily concerned with the case of literature, the relevance of their arguments to the question of cinematic authorship is obvious and their influence has been huge. So it appears there is a deep tension in our thinking about film and authorship. On the one hand, treating some if not all directors as authors is tempting, and there is a long and distinguished history of making just such an identification. Furthermore, it must be said that despite Kael’s skepticism, the director-as-author thesis has paid off critically. The directors and films championed by Truffaut (e.g., Renoir, Bresson, Cocteau, Ophuls and their movies) and Sarris (e.g., Flaherty, Ford, Hawks, Hitchcock, Keaton and theirs) have to a very large extent survived the test of time. The recognition of the very significant artistic successes of American moviemaking during the height of the Hollywood studio system seems due in large part to the writings of the auteur theorists. And this must weigh in favor of the director-as-author view. But, on the other hand, there is something deeply unintuitive about treating directors as authors. The analogy between authors of novels and directors of full-length films seems to obscure difference rather than illuminate similarity. And the very notion of authorship seems as if it may be an outdated and perhaps limiting romantic concept. How then to proceed? One traditional philosophical approach would be to start by arguing for a definition of the author concept. Armed with such a definition, one would then seek to apply it to the case of film. Does any person or group of persons involved in the making of films meet the criteria for authorship laid out in the definition? If so, films have authors, and the person or persons who fit the criteria are such. If not, they don’t. Although I am suspicious of such a strategy – philosophy doesn’t have the greatest record when it comes to definition, and the history of failure may have a psychological basis (Ramsey 1998) – I admit that such an approach may ultimately be useful. (Livingston [1997, 2005] offers the most compelling advertisements for such an approach that I know. See below for brief discussion.) Nevertheless, I believe that we will gain more insight into the debate about cinematic authorship (and get a firmer grasp on what cinematic authorship is) by first considering why the issue of authorship matters. But before we get on with that, let me address some of the aforementioned skeptical views about cinematic authorship. For if there are good arguments against the existence of film authors, we might save ourselves a lot of trouble.
Skepticism about (cinematic) authorship As should be clear from the discussion above, much of the skepticism that auteurism and related views faced focused on specific theses about actual authors (i.e., the director-as-author view and the related claim that solo authorship is common in 14
TF17557.indb 14
16/9/08 15:47:15
AU T HOR SH IP
film). Such forms of skepticism do not seek to do away with the notion of authorship altogether – far from it. Defenders of multiple-authorship views, for example, do not doubt that films have authors – they are simply critical of common views of how many persons typically fill that role (Gaut 1997; Sellors 2007). Other critics of the directoras-author view may simply be in favor of identifying a different figure as the standard author of a film or, perhaps, holding that various figures may fill that role in different circumstances. Even someone sympathetic to auteurism, such as Ian Cameron, is sympathetic to the latter view: “Given a weak director the effective author of a film can be its photographer . . . composer . . . producer . . . or star” (Cameron 2000: 54–5; see also Arnheim 1997: 68). Some more robustly skeptical views about cinematic authorship can be dismissed rather quickly. Is there any inconsistency in applying the idea of authorship to works of mass or popular culture? In short the answer is no. Stephen King and Danielle Steel work in the sphere of mass culture, and they are clearly authors in at least some important sense. If commercial films were produced by some process akin to that involved in making cars or computers, then it might be reasonable to deny authorship to them. But even at the height of the studio system, this is not how films were made (Bordwell and Thompson 1993: 10). And, of course, not all film falls into the categories of mass or popular culture. The genealogical worry may be swiftly dismissed too. Cinema is a relatively young art form. It should not be surprising that the appropriate theoretical apparatus for understanding it took some time to discover. And the purposes for which the idea of cinematic authorship was initially used are simply irrelevant to the validity of the concept. In response to the suggestion that colloquial English doesn’t typically involve the application of the term “author” to films, it is worth noting that it is not the case that the only relevant literal usages of the term apply to the makers of texts. A quick perusal of the OED unearths a wide range of literal uses. Moreover, the issues that underlie the debate about cinematic authorship are live ones even if films do not literally have authors: the question may be about the film artist (as Gaut argues), or about the value of treating films as if they had authors, or about whether there is anything significantly analogous to literary authorship in the case of filmmaking. Stephen Heath’s influential criticism of the idea of the cinematic authorship focuses on the putative way in which “the notion of the author . . . avoids . . . the thinking of the film text in relation to ideology” (Heath 1986: 217), and he suggests that the notion of the film author is inconsistent with the much needed development of “a theory of the subject” (217), by which he seems to mean a theory that would address both viewers and filmmakers in relation to ideology. Heath gestures at a significant problem with auteur theory – it could, in certain modes, be rather one-dimensional in its approach. A narrow critical focus on directors and their oeuvres might involve a concomitant failure to attend to the role of the film industry and the social conditions in which films are produced, as well as the ways in which viewers receive films. But this is not reason to reject the idea that films have authors. It merely provides a good reason to reject the idea that authorship is the only thing worth studying when it comes to film. For example, in the article about which Heath’s essay comments, 15
TF17557.indb 15
16/9/08 15:47:15
A A RON M E SK I N
Edward Buscombe suggests augmenting auteur theory with an attention to “the effects of the cinema on society . . . the effect of society on the cinema . . . the effects of films on other films” (Buscombe 1986: 32). Heath’s suggestion that simply adding these approaches to the auteur theory is impossible rests on the assumption that “the notion of the author itself is a major ideological construction” (Heath 1986: 217), but this is tendentious, since it is not at all clear that the author concept is inherently epistemically defective, nor that it has an essential link to social domination. (See Carroll [1998] for the characterization of ideology that underwrites my skepticism about Heath’s claim.) Nor is there an inconsistency between a belief in cinematic authors and the practice of ideological criticism. Careful attention to the ways in which films reflect, express, and instill ideology may go hand in hand with the view that they are authored. What is true is that many of the most plausible accounts of ideology in cinema are not explicitly concerned with authorship (e.g., Carroll 1996). But this does not imply inconsistency – at most it suggests that we may not have a unified “Grand Theory” of cinema. So Heath’s concerns are misplaced. It should be noted, too, that Heath himself suggests that “the author . . . may return as a fiction” (Heath 1986: 220), so his skepticism is itself ambivalent. But what about the aforementioned “death of the author”? If the author is dead then the cinematic author is dead as well and debates about the nature and number of film authors are moot. As mentioned above, Barthes and Foucault have certainly been influential, but it is not even clear whether they were arguing that authors do not exist, will not exist, or should not exist. (See Lamarque [1990] for relevant discussion.) Barthes, for example, claimed that “the sway of the Author remains powerful” (2002: 4). But if the author and/or author function currently exists, then theorists interested in developing accounts of cinematic authorship still have a legitimate area of inquiry. Perhaps more importantly, their arguments for the death-of-the-author conclusion – insofar as those can be made out – are unpersuasive. I shall ignore the purely predictive reading of the death thesis, since it is hard to know what possible arguments could support such crystal ball pronouncements. With respect to arguments for the extant nonexistence of the author the case does not seem much better. For example, Foucault’s discussion of authorship and the author function emphasizes the social, legal, and moral status associated with it: “the author function is linked to the juridical and institutional system that encompasses, determines, and articulates the universe of discourses” (Foucault 2002: 17). Authorship, on his account, is then a constructed and therefore contingent social role: “Texts, books, and discourses really began to have authors . . . to the extent that authors became subject to punishment, that is, to the extent that discourses could be transgressive” (14). But if we focus on the social role of authorship and, in particular, the moral and legal status of those people who produce works of literature and film, then it is obvious that the social role of the author still exists (Lamarque 1990). In many contexts filmmakers are still subject to moral, legal, and religious censure for what they produce. Films may still transgress. The juridical and institutional systems that allegedly underwrite the author function are still in place. 16
TF17557.indb 16
16/9/08 15:47:16
AU T HOR SH IP
This leaves us with the most interesting of the claims suggested by Barthes and Foucault – the normative view that we would be better off in some sense if authorship disappeared or came to an end. There might, in fact, be some advantages to such a development (although pretty clearly there would be some serious drawbacks to doing away with copyright law), but Barthes and Foucault do not make the case for this. Rather they argue that the elimination of the author would be hermeneutically liberatory and “truly revolutionary” (Barthes 2002: 6). The author, Foucault states, “is a certain functional principle by which, in our culture, one limits, excludes, and chooses” (Foucault 2002: 21), and here he is talking about meaning or signification which – on his view – would be better left to “proliferate” to some significant extent (21). But, in the first place, the degree to which authorship limits and excludes is often overstated. Even a cursory glance at the contemporary philosophical work on interpretation should make evident the degree to which a commitment to the importance of authorship is consistent with interpretive pluralism (Stecker 2006). And, in the second place, it is not at all obvious how valuable the proliferation of meaning would really be. In many practical contexts (e.g., with traffic signs and shouts of warning or distress) it is crucial that meaning doesn’t proliferate. Similarly in scientific contexts – a little too much proliferation of meaning and the cooperative pursuit of knowledge is jeopardized. Limitation and exclusion of meaning are often good things. And as Foucault recognizes, “it would be pure romanticism . . . to imagine a culture in which the fictive would operate in an absolutely free state” (Foucault 2002: 22). The upshot is that there are significant reasons for being skeptical of the claim that we would be better without the social role of the author and that, hence, all versions of the death of the author thesis are suspect. Friends of auteurism have nothing to worry about from Foucault and Barthes. Skepticism about the very idea of film authorship can be put safely aside.
Why does (cinematic) authorship matter? So what is at stake in the debate about film authorship? Once we know how different sorts of films are made, what difference does it make whether we identify an individual or group as author or deny that films have authors altogether? We may gain some insight into why the issue matters by considering the place of authorship in literature. For authorship – though an almost irredeemably contested notion – seems to play a central role in ordinary and critical appreciation of literature. In particular, some notion of authorship seems to play an important role in our engagement with literature as literature – authorship matters (or, at least, seems to matter) in the contexts of literary evaluation, interpretation, and stylistic attribution. (It also matters legally and morally – more on this below.) And it is natural to think this relevance carries over to the cinematic realm and that film authorship is just as central to ordinary and critical engagement with cinema; that is, that it plays an important role in film evaluation, interpretation, and stylistic attribution. Examination of the debate on film authorship in light of these activities helps clarify what is at stake in the debate about cinematic authorship. 17
TF17557.indb 17
16/9/08 15:47:16
A A RON M E SK I N
Evaluation Although this is tendentious, many philosophers would agree that to evaluate a work of art as art is, at least in part, to evaluate the achievement of the artist who made it (Dutton 1979; Currie 1989). The artists who make literary works are their authors, so if the general principle is true then authorship matters in the context of literary evaluation. This is not the whole story about literary evaluation of course, but it does point to an important connection between authorship and evaluation – a connection that is manifest in much of the discussion of cinematic authorship. Evaluative concerns were central to the early development of the auteur theory. Truffaut’s influential distinction between mere metteurs en scène and auteurs was primarily a matter of defending and explaining certain critical evaluations; namely, his negative assessment of the dominant mode of postwar French cinema and his valorization of the films of the directors mentioned above. Andrew Sarris’ “premises” of the auteur theory – “the technical competence of a director as a criterion of value,” “the distinguishable personality of the director as a criterion of value,” and “interior meaning, the ultimate glory of the cinema as an art” (Sarris 2000: 69) – are clear expressions of the evaluative concerns underlying his commitment to the director-as-author thesis, as is his suggestion that although directors “do not always run true to form,” it is “almost always” the case that bad directors make bad films (68). And Kael’s critical remarks, mentioned above, are an expression of her negative assessment of the evaluative dimension of the auteur theory. Moreover, authorship isn’t just relevant to the evaluation of individual cinematic works or specific bodies of work like Hollywood films of the studio era. It is plausible that the very idea of applying the concept of authorship to cinema was centrally tied up with the establishment of film – especially commercial film – as on an artistic par with the other arts. So the debate about authorship centrally has evaluative significance. Interpretation The dominant view among philosophers who have addressed themselves to interpretation in the literary context is that authorship of some sort or other (i.e., actual, implied, postulated) plays a role in interpretation. Only the most extreme forms of conventionalism or anti-intentionalism about interpretation reject any appeal to authors whatsoever, and it is noticeable that the most vigorous recent philosophical defender of an anti-intentionalist approach to literary and artwork interpretation argues that “an author’s intentions seem to be essential to her work’s identity and thereby central to the identification of the appropriate object of interpretation” (Davies 2006: 233). This should also reinforce the point that thinking authorship has something to do with interpretation is not thereby to think that the actual author’s semantic intentions fully determine acceptable interpretation. These reflections on the literary case suggest that the question of cinematic authorship may matter for interpretive or hermeneutic reasons. And, in fact, the interpretive significance of authorship seems central to much of the writings of 18
TF17557.indb 18
16/9/08 15:47:16
AU T HOR SH IP
auteur theorists, as well as their critics. A concern for the interpretive significance of authorship is clearly manifest in the work of Peter Wollen, whose auteur structuralism is centered around discerning structures implicit in the bodies of films produced by particular directors – structures whose varying manifestations are key to “decoding” or deciphering their films. So, for example, Wollen states that, “the auteur theory enables us to reveal a complex of meanings in films such as Donovan’s Reef” (Wollen 1986: 142). Robin Wood’s version of auteurism, as expressed in his writings on Ford and Hawks, shows a significant concern for the role that directors’ intentions and values may play in the interpretation and understanding of their films (Wood 1986, 2000). The philosopher Gregory Currie appeals to the “implied author” of certain films (i.e., the author as he/she/they seem from the evidence of the work) to explain our ability to grasp certain sorts of cinematic unreliability (Currie 1995b). Even some critics of traditional author theory have argued that “the ‘fiction’ of the author enables us to locate an author of the fiction . . . who in ‘his’ notional coherence provides the means for us to grasp the text in the moments of its production before us” (Nowell-Smith 1986: 223). Style Style and stylistic attribution are central to our engagement with literature and the other arts. So, for example, we interpret, evaluate, and identify literary works in virtue of our recognition of the various styles that they manifest. The same is true of cinema, and one key role that authorship plays in film criticism and theory is to underwrite certain sorts of judgments about film style. Now it is not the case that all forms of film style (e.g., universal film styles such as the realist style) have much to do with authorship. But our concern for the individual styles that are manifested in works of the cinema (i.e., the styles of film artists such as Godard and Hitchcock) seems to mirror our concern for the individual styles of authors found in great works of literature. Moreover, one very important – arguably dominant – view about individual literary style connects it to the personality or character of the author or implied author of the work of literature (Robinson 1985). What is more, a concern for style and the connection between individual style and personality is a key aspect of the auteurist view. Sarris writes that “Over a group of films, a director must exhibit certain recurring characteristics of style which serve as his signature” (Sarris 2000: 69). Bazin distinguishes between the auteur (e.g., Hitchcock) and the metteur en scène (Huston) who lacks a “truly personal style” (Bazin, quoted in Buscombe 1986: 23). And Perkins claims that the “most telling argument for a critical belief in the ‘director’s cinema’ is that it has provided the richest base for useful analyses of the styles and meanings of particular films” (Perkins 1993: 185). Legal and moral aspects of authorship I have to this point focused on the broadly artistic aspects of cinematic authorship. This is reflective of the literature on the topic. But authorship is not purely an aesthetic category – it also has legal and moral significance. (See Salokannel [2003] 19
TF17557.indb 19
16/9/08 15:47:16
A A RON M E SK I N
for a useful discussion of some of the legal aspects of cinematic authorship.) So, for example, authors are typically – at least in the first instance – treated as the legal owners of the works they produce. And in virtue of ownership and the possession of copyright, authors are treated as having certain legal rights. Copying and adaptation of copyrighted works typically require the consent of the author. Moreover, authors often have the right to be identified as authors, and they have certain rights to object to the way in which their works are treated. These rights, of course, can be transferred; for example, it is common in the academic world for all or some of these rights to be assigned to a publisher. But it is the author who initially holds copyright and the rights that go with it. Even more significantly for the case of film, in some legal jurisdictions works produced as part of employment may be treated as owned and authored by the employer. US copyright law counts the typical commercially made film as a “work made for hire” and states that in “the case of a work made for hire, the employer or other person for whom the work was prepared is considered the author for purposes of this title” (17 USC §201 [b]). Laws vary of course, as do their application to film. In the European Union the principal director is typically treated as the (or a) film’s author – there is some variation within the European Union about whether or not other makers count as authors (Salokannel 2003: 167). The key points to take from this are that authorship is not purely an artistic or aesthetic category – it is of legal significance too. And of moral significance. Plagiarism and film piracy are not simply legal affronts: they seem to be moral affronts to the author as well. More centrally, it is widely (although not uncontroversially) held that the intimate relationship in which authors stand to their works underwrites certain sorts of author-focused moral evaluations. It is natural to treat the author of a racist work of literature or film as to some extent morally responsible for that racism. And the author of a work that is designed to foment hatred and violence or to honor something that is despicable is commonly morally censured. The fascist sympathies of Triumph of the Will (1935) have seemed to many to provide reason to morally criticize Leni Riefenstahl. There are, of course, substantive debates about the ethical evaluation of art and literature (and, in particular, whether it is real authors or manifest and/or implied authors who are appropriately ethically evaluated on the basis of works); my only point is that the idea of authorship plays a role in these debates. One potential upshot of considering the legal and moral aspects of authorship is the recognition that authorship may be fragmented (i.e., distinct notions of authorship or distinct aspects of the authorship role may pull apart in certain circumstances). In fact, I think there is a significant potential for this sort of fragmentation in the case of film, and I shall return to this below.
Two issues in film authorship Now that we have a sense of what is at stake in the debate about cinematic authorship, let us briefly turn to a consideration of two issues about film authors: (1) How many authors do films typically have? And (2) are film authors real or constructed? 20
TF17557.indb 20
16/9/08 15:47:16
AU T HOR SH IP
Individual or collaborative authorship? Let us assume that it makes sense to say that films have authors. It would be a mistake to ask whether films in general are singly or multiply authored, for it is plausible that such a question would be ill-formed. There are many kinds of films and many kinds of filmmaking practices. Surely the only reasonable position is that some films have individual authors and some are authored by groups. In addition, it is plausible that there are (or could be) entirely unauthored films. Cases to consider here are accidentally produced films (see Gaut 1997: 169n14) or “traffic jam” films that, as Livingston puts it, are “the unintended result of disparate and unintentional activities” (Livingston 1997: 138). (Perhaps such films could not be [good] works of art.) That some films have individual authors seems hard to deny. Some homemade digital films are entirely made by one person. The same is true of certain experimental and avant-garde films (e.g., Brakhage’s 1963 Mothlight). Other films are almost entirely made solo (i.e., at the most only certain technical processes are carried out by others), and it does not seem that, for example, sending Super 8 stock out to be developed undercuts one’s status as the sole author of a film. Louis Lumière is often treated as a solitary filmmaker, and other early pioneers such as Le Prince also seem to have worked largely alone. On the other hand, there are cases where it seems difficult to deny multiple authorship. The close working relationships between the Coen brothers and the Brothers Quay suggests that their films are multiply and, in fact, collaboratively authored. Other cases of multiple, although perhaps noncollaborative, authorship can be found in certain omnibus films such as New York Stories (1989), which is composed of three shorts made by three distinct filmmaking teams. And some films have been produced by collectives such as Newsreel. But although it should be fairly uncontroversial that there are both singly and multiply authored films, the question of what is typical or standard is tendentious. The view that most popular films are individually authored (if authored at all) is dominant, but there is a significant minority who argue that multiple authorship is typical for commercial “mainstream” film (e.g., Gaut 1997; Sellors 2007). The sheer number of people involved in the making of a standard studio film seems to weigh in this direction. How could one person (even the director) possibly count as the author of a standard Hollywood film? One might worry that the popularity of the single-author hypothesis (the view that a great many popular films are authored by a single individual) rests on confusion or a failure to take seriously the ways in which mainstream films are made. But this would be a mistake. The defenders of the single-author hypothesis are not blind to the essentially collaborative nature of mainstream filmmaking (see, for example, Perkins 1993: 158ff.). And some sorts of collaborative or at least cooperative production do seem consistent with solo authorship or artistry. The editors of academic journals in the humanities often make (or, at least, suggest) changes to articles before they are published. But the idea that such a journal article is, therefore, authored by a group rather than an individual seems perverse. The situation is the same in the literary context. Although some editors may make such significant contributions 21
TF17557.indb 21
16/9/08 15:47:16
A A RON M E SK I N
that they may deserve coauthorship status (Gordon Lish’s contributions to Raymond Carver’s stories and Maxwell Perkin’s editing of Thomas Wolfe are notable cases), it does not seem plausible that every novel that has been edited is thereby coauthored. Individual authorship does not require that a text or artwork be produced solely by a single person. As Livingston writes: “some items can be collectively produced . . . without having been collaboratively or jointly authored” (Livingston 2005: 75). So how can the collective or collaborative production that is standard in film be made consistent with solo authorship? A variety of suggestions are made in the literature (notably, these are typically in service of a director-as-author thesis as well). V. F. Perkins claims that directors are “in charge of relationships, of synthesis” and are, therefore, “in charge of what makes a film a film” (Perkins 1993: 184), moreover, that the “director’s authority is a matter not of total creation but of sufficient control” (184). Bordwell and Thompson suggest that “usually it is through the director’s control of the shooting and assembly phases that the film’s form and style crystallize” (Bordwell and Thompson 1993: 16). Peter Wollen argues that we may focus on the structures found in groups of films produced by a single director and treat everything else as irrelevant, i.e., mere “noise” contributed by producers, camera operators, and actors (Wollen 1986: 143–4). Livingston, who has argued that some studio films are singly authored, points to the “high degree of control” and “huge measure of authority” that some directors have (Livingston 1997: 144) and the fact that some films express a director’s “sensibility and attitudes so vividly” (Livingston 2005: 85). Although one may tease apart at least two different broad strategies here (Gaut distinguishes the “sufficient control” strategy from the “restriction” strategy [Gaut 1997: 155–8]), perhaps the most plausible argument for solo authorship combines the two, appealing to the director’s significant (and, hence, sufficient) control over the factors that determine the artistically or cinematically relevant aspects of the film. Can such a strategy be successful in shoring up a solo authorship view of film? There are two broad ways one might challenge this sort of argument for solo authorship – one could dispute the empirical claim about the typical degree and extent of directors’ control or one could dispute the broadly conceptual claim about what suffices for authorship. Gaut offers both sorts of challenges. Although he admits that some architects might be properly seen to be solo authors/artists in virtue of their sufficient control over the artistically relevant features of the buildings they design, the case of film is simply different: “the director’s degree of control over the work is not that of the architect’s” (Gaut 1997: 158) and, hence, is not properly seen as a solo author/ artist. (Presumably no other person involved in the making of a film has the requisite degree of control either.) Gaut also suggests that sufficient control may not be enough to establish solo authorship – the fact that the director is the person with the greatest degree of responsibility for the look and sound of the finished film does not imply that he or she is the only author for “if there are others who make a significant artistic difference to the work, then it is only fair to acknowledge them as artistic collaborators” (157). It remains to be seen whether defenders of the single-author view can successfully meet these challenges. 22
TF17557.indb 22
16/9/08 15:47:16
AU T HOR SH IP
Defenders of certain versions of the multiple-authorship view face their own challenges. Gaut goes so far as to argue that the “actors of films are among its co-creators” (158); hence, they count as coauthors. He bases his claim on the fact that film performances are “internal to the film” (163) and that screen actors partially determine the nature of the characters they play. But at least two considerations weigh against the actor-as-author view. In the first place, Gaut’s reasoning would seem to be extendable to cases of photographic portraiture, and it is counterintuitive that subjects of photographic portraits typically count as coauthors. More significantly, actors’ performances are “internal” to films in virtue of their being represented cinematographically. But it is simply hard to see how this could provide a basis for a legitimate authorship claim – the grounds of performance internality simply don’t seem the right ones to underwrite a claim of authorship. The topic is a rich one for future research. Cinematic authors and/or author constructs? I have already mentioned Foucault’s invocation of the “author function” rather than the author itself. Foucault’s usage reflects an important trend in recent thinking about both literary and cinematic authorship – a shift from thinking of authorship as a role filled by actual individuals or collectives to thinking of them as constructs or fictions or something of the sort. Peter Wollen distinguishes between the real individuals Fuller, Hawks, and Hitchcock, on the one hand, and “ ‘Fuller’ or ‘Hawks’ or ‘Hitchcock’, the structures named after them” (Wollen 1986: 147), and this distinction is taken up by Geoffrey Nowell-Smith, among others, who states that “What the play with inverted commas aims to do is to distinguish the author as empirical origin (John Ford, the man with the eye-patch who isn’t Raoul Walsh) from the author as effect of the text (‘John Ford’)” (Nowell-Smith 1986: 222). Seymour Chatman explicitly borrows the idea of the “implied author” from literary theory to make sense of the way in which a collaboratively produced film may seem as if it is the unified product of a single controlling intelligence (Chatman 2005: 191–2). The issue of the ontological status of authors is a huge one (see Livingston [1997] and Gaut [1997] for discussion in the context of film), but we need not see it as one of forced choice. That is, we can and should allow that there are both real empirical authors and authors as they are seen in their works (i.e., author constructs or implied authors). In some but not all works of fiction, there are additionally full-fledged fictional authors; that is, it is true-in-the-fiction that they are authored by some agent or agents. The character David Holzman (played by L. M. Kit Carson) is portrayed as the sole maker and author of David Holzman’s Diary (1967), even though it was directed by Jim McBride. But this limited pluralism about authorship does not settle the disputes about evaluation, interpretation, and stylistic attribution that underlie the authorship debates. So, for example, does film interpretation properly concern itself with the intentions of the actual cinematic author(s) or, rather, the intentions of the author construct (i.e., the author as manifest in the work)? Is the attribution of individual cinematic style (e.g., Kurosawa’s style) concerned with the actual director’s personality or character 23
TF17557.indb 23
16/9/08 15:47:16
A A RON M E SK I N
or, rather, the personality of the implied author (Kurosawa as he appears from the evidence of the film or body of films)? It is, of course, also possible that both concerns are legitimate. How are we to decide? I have already mentioned one argument in favor of the significance of the implied cinematic author – the suggestion that recognition of and appeal to the author construct provides the best explanation of the apparent unity and coherence that many films present us with. This, I believe is not persuasive. The assumption seems to be that unity is only attributable as an effect of an individual maker, but this assumption is flawed. There are a range of ways that groups may organize themselves to produce unified and coherent outputs – the use of a project manager (or lead author!) is a common strategy in the academic context. Moreover, it is not obvious that the implied author of a film is best thought of as an individual. Gaut argues that “were we to construct just one author for each film, it would be so different from people as we know them that its use . . . would systematically distort our responses to cinema” (Gaut 1997: 160). So it does not seem that appeal to an implied author can comfortably explain the apparent unity of many films. And, as Gaut points out, appeal to collective authorship may help us explain why some films give the appearance of disunity and internal contradiction (165–6). Let me offer two other reasons to be skeptical of the appeal to implied authors or author constructs. The first is that one of key motivations for the shift away from actual author to author construct looks to be questionable. Fans of author constructs often point to the frequent apparent mismatch between the character or personality of the real person(s) who made a work and the character or personality expressed in the work. For example, D. W. Griffith made films that appear to express viciously racist ideas, but some have suggested that he is not properly seen as a racist (Curtis 2003). The possibility of mismatch suggests to some that we are not and should not be concerned with the attitudes of the actual person(s) who made the film, when it comes to interpretation and stylistic attribution. Perhaps the right thing to say is that Griffith’s films express the attitudes of Griffith-as-he-is-seen-from-his-films (i.e., the views of the Griffith construct or “Griffith”), rather than those of Griffith himself. (See Robinson [1985] for a classic statement of the general form of argument in the case of literature.) But this mismatch argument seems to rely on the idea that character traits are broad or “global,” that is, that they are consistently manifested across a wide range of relevant domains. Recent work on character has called this assumption into question (Doris 2002). And if character traits are narrow or “local,” then a mismatch between a trait exhibited in fiction-writing or filmmaking and a trait exhibited in other contexts should not trouble us. If character traits are more like courage in a particular context than courage simpliciter (62), then there should be nothing surprising about the fact that the style of a director’s cinematic oeuvre may express different character traits than are expressed by him or her in other contexts and activities. After all, we all know people who are nice in one sort of situation and nasty in another. The second reason for being skeptical of the appeal to implied authors or author constructs rests on the claim that the concept of authorship seems to be a broadly explanatory notion. Authors make works of literature. The psychological states of 24
TF17557.indb 24
16/9/08 15:47:16
AU T HOR SH IP
authors play a role in explaining the individual styles that the works they produce exhibit (cf. Wollheim 1987: 26–36). It is because authorship is, at least in part, a causalexplanatory notion that authors seem appropriately blamed and praised for certain moral aspects of the works they produce. But author constructs, fictive authors, and implied authors cannot play an appropriate explanatory role. It is Hawks who has causal relations with real-world entities, not “the fiction of Hawks,” nor Hawks-as-he-seems-tobe-on-the-basis-of-his-films (Gaut 1997; see also Stecker 1987). Films may have implied authors, but these will do little work in explaining what needs explaining.
Trends and new directions In an important paper on the topic, Livingston offered a definition (or “partial analysis”) of cinematic authorship: Cinematic author = the agent or agent(s) who intentionally make(s) . . . an action the intended function of which is to make manifest or communicate some attitude(s) by means of the production of apparently moving images projected on a screen or other surface. (Livingston 1997: 141) I do not have the space to address this account in detail (for objections, see Sellors 2007; and Gaut 1999: 171). But it is worth noting that Livingston’s account is part of a very recent trend that seeks to apply the tools of the theories of action, agency, and communication to the question of cinematic authorship. Additional work by Livingston on authorship more generally, as well as recent work by C. Paul Sellors and Trevor Ponech, is part of this promising turn (Livingston 2005; Sellors 2007; Ponech 1999). Further clarification about the nature and possibility of authorship under the massively collaborative conditions of ordinary filmmaking requires careful development and deployment of theories of collective and complex action and intention. Let me suggest two other directions for further exploration. First, theorists interested in cinematic authorship may find it useful to examine the nature of authorship in certain nonliterary contexts where collaboration is standard, such as scientific publishing. Recent articles testify to significant growth in the mean number of authors in scientific papers, as well as the stunningly large numbers of authors that certain papers may have (King 2007; Greene 2007). I do not claim that attention to scientific publishing will straightforwardly answer questions about cinematic authorship. It is certainly possible that conceptual confusion about authorship is as – or more – endemic in that domain than in the domain of film theory. Nonetheless, there is an important model here of how authorship is theorized and attributed in a collaborative setting. One particular payoff from attending to this domain might be a richer account of the notions of lead and primary authorship (and, along with it, secondary or additional authorship). For it is very tempting to think that appeal to such notions would enrich the discussion of cinematic authorship. Second, I mentioned above the possibility that authorship seems to have the potential to exhibit a sort of fragmentation in the sphere of filmmaking. What do I 25
TF17557.indb 25
16/9/08 15:47:16
A A RON M E SK I N
mean? Compare the paradigmatic case of literary authorship. In the literary context, the author role is a complex and unifying functional role that is typically filled by a single individual. That is, the interpretive, evaluative, moral, legal, and even individuating functions of literary authorship are typically all located in a single individual. But we have already seen that in the film case legal authorship may pull apart from other authorship roles. I suggest that in very ordinary cases of filmmaking other sorts of fragmentation of the authorship role may occur. An agent (e.g., composer) who is partly artistically responsible for a film (i.e., who deserves some credit for its artistic merit) may be in no way morally responsible for it (perhaps because the relevant ethical content was generated in the editing process after the score’s composition was completed). It is easy to think of many other such cases. In this way authorship in film can exhibit a kind of disunity that is not typical in literature. It is not simply that no one person involved in the making of a film stands in the authorship role, it is that no single group of people might be properly seen as standing in the authorship role (although different subgroups might be seen as performing one or more aspects of the authorship role). There may, of course, be much less extravagant ways of putting this point. But my suggestion is that this dimension of filmmaking (the way in which various aspects of the authorship role that are unified in the literary case may come apart in the film case) is one to which theorists of cinematic authorship should attend.
Acknowledgments Thanks are due to the editors, Stephen Meskin, and Andrew Kania for their comments on previous versions of this chapter. See also Ethics (Chapter 10), Interpretation (Chapter 15), Narrative closure (Chapter 19), Style (Chapter 25), and Ingmar Bergman (Chapter 51).
References Arnheim, R. (1997) Film Essays and Criticism, trans. B. Benthien, Madison: University of Wisconsin Press. Barthes, R. (2002) “The Death of the Author,” trans. S. Heath, in W. Irwin (ed.) The Death and Resurrection of the Author?, Westport, CT: Greenwood Press. Bordwell, D., and Thompson, K. (1993) Film Art: An Introduction, 4th ed., New York: McGraw-Hill. Buscombe, E. (1981) “Ideas of Authorship,” in J. Caughie (ed.) Theories of Authorship, London and New York: Routledge. Cameron, I. (2000) “Films, Directors and Critics,” in J. Hollows, P. Hutchings, and M. Jancovich (eds.) The Film Studies Reader, London: Arnold. Carroll, N. (1996) “Film, Rhetoric and Ideology,” in Theorizing the Moving Image, New York: Cambridge University Press. —— (1998) A Philosophy of Mass Art, Oxford: Oxford University Press. Chatman, S. (2005) “The Cinematic Narrator,” in T. Wartenberg and A. Curran (eds.) The Philosophy of Film: Introductory Text and Readings, Malden, MA: Blackwell. Coward, R. (1987) “Dennis Potter and the Question of the Television Audience,” Critical Quarterly 29: 79–87.
26
TF17557.indb 26
16/9/08 15:47:16
AU T HOR SH IP
Currie, G. (1989) An Ontology of Art, London: Macmillan. —— (1993) “The Long Goodbye: The Imaginary Language of Film,” British Journal of Aesthetics 33: 207–19. —— (1995a) Image and Mind: Film, Philosophy and Cognitive Science, New York: Cambridge University Press. —— (1995b) “Unreliability Refigured: Narrative in Literature and Film,” Journal of Aesthetics and Art Criticism 53: 19–29. Curtis, B. (2003) “D.W. Griffith in Black and White,” Slate Magazine. Available at http://www.slate.com/ id/2076307 (accessed 7 March 2008). Davies, S. (2006) “Author’s Intentions, Literary Interpretation, and Literary Value,” British Journal of Aesthetics 46: 223–47. Doris, J. (2002) Lack of Character: Personality and Moral Behavior, Cambridge: Cambridge University Press. Dutton, D. (1979) “Artistic Crimes: The Problem of Forgery in the Arts,” British Journal of Aesthetics 19: 304–14. Dyer, R. (1979) Stars, London: British Film Institute. Foucault, M. (2002) “What is an Author?,” in W. Irwin (ed.) The Death and Resurrection of the Author?, Westport, CT: Greenwood Press. Gaut, B. (1997) “Film Authorship and Collaboration,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Clarendon Press. Greene, M. (2007) “The Demise of the Lone Author,” Nature 450, 1165 (20 December 2007). Heath, S. (1981) “Comment on ‘The Idea of Authorship’,” in J. Caughie (ed.) Theories of Authorship, London and New York: Routledge. Inge, M. T. (2001) “Collaboration and Concepts of Authorship,” PMLA 116: 623–30. Kael, P. (2005) “The Idea of Film Criticism,” in T. Wartenberg and A. Curran (eds.) The Philosophy of Film: Introductory Text and Readings, Malden, MA: Blackwell. King, C. (2007) “Multiauthor Papers Redux: A New Peek at New Peeks,” Science Watch 18. Available at http://archive.sciencewatch.com/nov-dec2007/sw_nov-dec2007_page1.htm (accessed on 6 March 2008). Koch, H. (2000) “A Playwright Looks at the ‘Filmwright’,” in J. Hollows, P. Hutchings, and M. Jancovich (eds.) The Film Studies Reader, London: Arnold. Lamarque, P. (1990) “The Death of the Author: An Analytic Autopsy,” British Journal of Aesthetics 30: 319–31. Livingston, P. (1997) “Cinematic Authorship,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Clarendon Press. —— (2005) Art and Intention: A Philosophical Study, Oxford: Oxford University Press. Nowell-Smith, G. (1981) “Six Authors in Pursuit of The Searchers,” in J. Caughie (ed.) Theories of Authorship, London and New York: Routledge. Perkins, V. F. (1993) Film as Film: Understanding and Judging Movies, London: De Capo Press. Ponech, T. (1999) “Authorship and Authorial Autonomy: The Personal Factor in the Cinematic Work of Art,” Canadian Aesthetics Journal 4. Available at http://www.uqtr.ca/AE/vol_4/trevor.htm (accessed 6 March 2008). Ramsey, W. (1998) “Prototypes and Conceptual Analysis,” in M. R. DePaul and W. Ramsey (eds.) Rethinking Intuition, Lanham, MD: Rowman and Littlefield. Robinson, J. (1985) “Style and Personality in the Literary Work,” Philosophical Review 94: 227–47. Salokannel, M. (2003) “Cinema in Search of Its Authors: On the Notion of Film Authorship in Legal Discourse,” in V. W. Wexman (ed.) Film and Authorship, Piscataway, NJ: Rutgers University Press. Sarris, Andrew (2000) “Notes on the Auteur Theory in 1962,” in J. Hollows, P. Hutchings, and M. Jancovich (eds.) The Film Studies Reader, London: Arnold. —— (2003) “The Auteur Theory Revisited,” in V. Wexman (ed.) Film and Authorship, New Brunswick, NJ: Rutgers University Press. Schatz, T. (1988) The Genuis of the System: Hollywood Filmmaking in the Studio Era, New York: Pantheon. Sellors, C. P. (2007) “Collective Authorship in Film,” Journal of Aesthetics and Art Criticism 65: 263–71. Stecker, R. (1987) “Apparent, Implied, and Postulated Authors,” Philosophy and Literature 11: 258–71.
27
TF17557.indb 27
16/9/08 15:47:16
A A RON M E SK I N
—— (2006) “Interpretation and the Problem of the Relevant Intention,” in M. Kieran (ed.) Contemporary Debates in Aesthetics and the Philosophy of Art, Malden, MA: Blackwell. Stillinger, J. (1991) Multiple Authorship and the Myth of Solitary Genius, New York: Oxford University Press. Truffaut, F. (2000) “A Certain Tendency of the French Cinema,” in J. Hollows, P. Hutchings, and M. Jancovich (eds.) The Film Studies Reader, London: Arnold. Wollheim, R. (1987) Painting as an Art, Princeton, NJ: Princeton University Press. Wollen, P. (1981) “The Auteur Theory,” in J. Caughie (ed.) Theories of Authorship, London and New York: Routledge. Wood, R. (1981) “Shall We Gather at the River? The Late Films of John Ford,” in J. Caughie (ed.) Theories of Authorship, London and New York: Routledge. —— (2000) “Hawks de-Wollenized,” in J. Hollows, P. Hutchings, and M. Jancovich (eds.) The Film Studies Reader, London: Arnold.
Further reading See John Caughie (ed.) Theories of Authorship (London and New York: Routledge, 1981); and V. W. Wexman (ed.) Film and Authorship (Piscataway, NJ: Rutgers University Press, 2003). A recent and useful collection on the topic is B. K. Grant (ed.) Auteurs and Authorship: A Film Reader (Malden, MA: Wiley-Blackwell, 2008).
28
TF17557.indb 28
16/9/08 15:47:16
3
CE NS O R SHIP Susan Dwyer For individuals at all points on the political spectrum, and especially for those engaged in any form of expressive enterprise – from comic book illustrators, to film directors, to performance artists – censorship typically carries very negative connotations. Indeed, for many, censorship is the very antithesis of freedom and creativity. However, we can and should conceive of censorship more neutrally – simply as the imposition of constraints. On such a construal, censorship is not obviously always a Bad Thing. This point is crucial for any effective argument against particular instances of censorship. For, if we simply define censorship as something always to be rejected, then we have merely asserted what potential censors deny, and we have provided no reasons to reject censorship. Moreover, a neutral construal of censorship better allows us to grasp that there are different agents of censorship, that censorship can and does have a range of targets and motivations, and that (apparently paradoxically) constraints of some kind may well be part of the very creative enterprise itself. Here I shall be concerned only with censorship as it has affected and continues to affect film. Obviously, there is much else to be said about the censorship of the press, art, and literature (see “Further reading” in this chapter and “Pornography”).
Agents of censorship The agent of censorship is not always, and perhaps not typically, the state. To be sure, across the globe and through time, governmental bodies are and have been engaged in the imposition of constraints on film, ranging from outright prohibition to what, in United States jurisprudence, are called time, place, and manner restrictions. But nonstate actors, such as film industry associations, advertisers, and advocacy groups, can be just as and sometimes even more powerful controllers of what films get made, distributed, and exhibited. State censorship When the state undertakes to prohibit or restrict public access to expressive materials, it can run afoul of a variety of constitutional protections. Most democratic constitutions embody protections for speech, for the press, and for expression more generally.
TF17557.indb 29
16/9/08 15:47:16
SUSA N DW Y E R
In the United States, the government must demonstrate that a proposed instance of censorship is motivated by a “compelling state interest” (increasingly cast in terms of fighting the “war on terror”) and that this particular act of censorship represents the least intrusive means to protect that interest. In addition, the government may not practice “prior restraint” – that is, prohibit some form of expression before it is published. Material is legally actionable only when it is in the public domain. The history of film is interesting in this regard. For it was not until 1952 that the United States Supreme Court conceded that motion pictures fell under the protection of the First Amendment. The controlling case prior to this was Mutual Film Corporation v. Industrial Commission of Ohio, 236 US 230 (1915), which concerned Ohio’s attempts to suppress D. W. Griffith’s The Birth of a Nation (1915). In this case, the Supreme Court contended that motion pictures are “a business pure and simple, originated and conducted for profit . . . [and] not to be regarded . . . as part of the press of the country or as organs of public opinion” (245). And so the Court upheld Ohio’s law that stated, in part, that “only such films as are in the judgment and discretion of the board of censors of a moral, educational or amusing and harmless character shall be passed and approved.” It was several decades later, in Joseph Burstyn, Inc. v. Wilson, 343 US 495 (1951) that the Supreme Court reversed itself, opining that motion pictures – in this instance, Rossellini’s The Miracle (1948) – could not be constitutionally denied an exhibition license, on the grounds that they were sacrilegious. Most importantly, in Burstyn, the Court recognized that cinema is “a significant medium for the communication of ideas . . . [and that] the importance of motion pictures as an organ of public opinion is not lessened by the fact that they are designed to entertain as well as inform” (502). (However, prior restraint on movies, manifested by the power of local and state censorship boards to prohibit screenings without first going to court, was not invalidated until 1965 [Freedman v. Maryland, 380 US 51].) In the wake of the Second World War, the United States government targeted Hollywood, in particular screenwriters and directors, during the infamous operations of the House on Un-American Activities Committee (HUAC). First in 1947, and then again in 1951, HUAC investigated individuals and examined films for the presence of what it deemed subversive (that is, pro-communist) messages. The HUAC hearings were extremely controversial; some members of the movie industry refused to cooperate, resulting in their imprisonment, while others testified against their peers. Careers and lives were literally ruined during these years (Gladchuk 2007). Studio heads were called before the Committee. Here, in a passage worth contrasting with the contemporary context, is Jack L. Warner, then the head of Warner Bros: Our company is keenly aware of its responsibility to keep its product free from subversive poisons. With all the vision at my command, I scrutinize the planning and production of our motion pictures. It is my firm belief that there is not a Warner Bros. picture that can fairly be judged to be hostile to our country, or communistic in tone or purpose. (Congress, HUAC, 1947: 11) 30
TF17557.indb 30
16/9/08 15:47:17
C E NS OR SH IP
While state censorship of the movies has been typically quite direct, the Family Movie Act of 2005 (part of the Family Entertainment and Copyright Act of 2005) is a very recent example of indirect state action. The Family Movie Act indemnifies the manufacturers of DVDs for home consumption against copyright violation charges if they make “imperceptible” (read, delete) certain scenes from films without the copyright holder’s permission, as long as the alteration is clearly noted. The ethical issues here are quite interesting. Arguably, the defensible motivation behind the Act is to support parents and caretakers in protecting children from images and language that is unsuitable for them. However, the Act would also seem to undermine the moral rights of filmmakers. In contrast to several European nations, the United States (as a common law nation) does not recognize the moral rights of artists as legal rights (Holland 2006). Copyright law, and thus the protection of a creator’s product, is largely justified in economic terms – no one may profit from my labor, pretending it is theirs. Copyright protections are not legally understood in terms of the artists’ moral interest in maintaining the integrity of their work and in not having their name associated with mutilated or distorted versions of that work. Industry associations Ironically, the most influential nonstate agent of censorship in the domain of film is the Motion Picture Association of America (MPAA). Initially called the Motion Picture Producers and Distributors of America (MPPDA), this industry body supported the establishment, in 1930, of the so-called Hays Code. Code “seals” were denied to any film that contained action or dialogue in violation of the thirty-some rules articulated in the Code, with the effect that such films could not be distributed or shown by any member company of the MPPDA. The Hays Code articulated the following general principles, here paraphrased: 1 No picture shall be produced that will lower the moral standards of those who see it. Hence the sympathy of the audience should never be thrown to the side of crime, wrongdoing, evil, or sin. 2 Correct standards of life, subject only to the requirements of drama and entertainment, shall be presented. 3 Law, natural or human, shall not be ridiculed, nor shall sympathy be created for its violation, which are supplemented by quite specific rules governing the presentation of crime, including murder, sex, nudity, dress, vulgarity, and national feelings. (MPPDA 1982 [1930]: 324) In 1968, the MPAA replaced the, by then, dated Hays Code with a ratings system. Again, this move by the industry itself was intended to preempt any state control or censorship of movies.
31
TF17557.indb 31
16/9/08 15:47:17
SUSA N DW Y E R
Ratings systems Ratings systems are operative in the United States, the United Kingdom, and India, among other moviemaking nations. In the United Kingdom, the 1912 British Board of Censors was renamed the British Board of Film Classification in 1985. It currently uses eight classifications for film, primarily designed to provide parents with guidelines about what material might be suitable for their children to see, and to clearly isolate sexually explicit material as available only to persons over eighteen years of age. India – which has the largest movie business in the world, producing almost eight hundred films a year – also has a Central Board of Film Classification (CBFC), which imposes ratings. But prior restraint is alive and well in India. For example, the CBFC requires that no mouth-to-mouth kissing scenes, and certain words survive editing cuts. Proposals made in 2004 to liberalize The Cinematograph Act of 1952 have not yet been enacted. In the United States, movies can receive any of the following ratings: G (Suggested for General Audiences); PG13 (Parental guidance suggested); R (Restricted, persons seventeen years of age or younger not admitted unless with parent or guardian); NC-17 (persons seventeen years of age or younger not admitted). The last rating in this list was not introduced until 1990 – Henry and June (1990) was the first film to receive it – when the MPAA decided to cease using X as a rating for movies with significant sexual content. The reason for this change was that the X rating was not trademarked by the MPAA, meaning that any filmmaker could use it. During the 1970s and 1980s, producers of pornography adopted the X rating for their films and so in the public’s view the rating became synonymous with pornography. Hence, while the MPAA had applied the X rating to films such as A Clockwork Orange (1971) and Taxi Driver (1976), it chose to adopt the trademarked NC-17 rating to dissociate any of its sanctioned films from pornography. As evidence of the power that the MPAA wields, the public is largely unaware of how the ratings system works in the United States. This Film Is Not Yet Rated (2006) exposes what its director, Kirby Dick, takes to be the secrecy and arbitrariness surrounding the MPAA ratings procedure. The names of the ten to thirteen members of the Ratings Board remain confidential, as do Board deliberations. Critics say this shields the Board and the MPAA from being accountable to filmmakers and the public for its ratings decisions. The MPAA, however, claims that anonymity is required to prevent Board members from being pressured by filmmakers and others to give movies particular ratings. Critics like Kirby Dick also complain that Board members are not especially qualified to judge movies. Indeed, on its website, which in response to Kirby Dick’s film now contains information about the ratings scheme, the MPAA says, “There are no special qualifications for Board membership, except that the members must have shared parenthood experience, must be possessed of an intelligent maturity, and most of all, have the capacity to put themselves in the role of most American parents” (MPAA 2008a: online). By itself, this might seem unproblematic. However, it is worth noting two things. 32
TF17557.indb 32
16/9/08 15:47:17
C E NS OR SH IP
First, the rating a movie receives determines whether and how it will be distributed and exhibited. Exhibition and distribution companies have substantial influence. Just fifteen companies control all the movie theaters in the United States, and the top five companies (Regal Entertainment Group, AMC Entertainment, Carmike Cinema, Cinemark USA, Loews Cineplex Entertainment) own 45 percent of the screens and 25 percent of exhibition sites (Eliasburg 2005). And these companies are loath to pick up movies that are rated NC-17. As a result, filmmakers have cut material from their movies in order for the Board to grant a more economically viable rating. In the face of criticism of the power of the ratings system, the MPAA emphatically points out that moviemakers are not obliged to submit their films for rating and that films may be released unrated. However, such films are likely to have difficulty finding a distributor. Second, it would appear that the Board employs a double standard in assessing the respective suitability of representations of violence and representations of sexuality. It is prepared to give an R rating to films that contain extreme violence – like The Departed (2006), Hostel (2005), and Saw (2004) – but it gives the NC-17 rating to films that depict pubic hair or male frontal nudity. For example, both The Cooler (2003) and Eyes Wide Shut (1999) were cut or digitally altered in order to receive a more congenial rating. Many critics claim that the MPAA rating system is at base a form of protectionism, ensuring a market for United States products from the major production companies while significantly restricting the market for independent, art-house, and foreign films (Lewis 2000: 189). It is important not to forget the pragmatic and economic motivations for censorship. The average Hollywood movie now costs approximately $100 million to produce, promote, and advertise (MPAA 2008b). Hence production companies take very big financial risks. It would be naïve not to grant that script revisions, cuts and so on may be, for the most part, economic and not artistic or ideological decisions at all. Lewis (2000) argues that “film censorship only incidentally and superficially regards specific film content . . . The MPAA supervises the regulation of film content solely to protect studio products in the marketplace” (2–3). Advocacy groups Though hardly financial behemoths akin to the corporations that control the production, distribution, and exhibition of motion pictures across the world, particular interest or pressure groups are also effective agents of censorship. Organized, grassroots opposition to certain films can be quite powerful, since it typically arises suddenly and relatively unpredictably. Notable examples have been Queer Nation’s protests of Basic Instinct (1992) (for its portrayal of gays and lesbians), Women Against Violence Against Women’s protests of Dressed to Kill (1980) (as encouraging violence against women), and Greek Orthodox opposition to The Da Vinci Code (2006) (as an insult to Christianity), and Johnson (2005) reports an anonymous source who claims that Paramount gave into pressure from the Council on American Islamic Relations, which objected to the fact that a terrorist in The Sum of All Fears (2002) was depicted as a Palestinian. (The Palestinian character was replaced by a neo-Nazi.) 33
TF17557.indb 33
16/9/08 15:47:17
SUSA N DW Y E R
Public controversy can, of course, be good publicity. Viewers will see movies precisely because they want to know what the all the fuss is about. Still, it is worth noting that local pressure has kept a variety of films (and other expressive works) out of libraries and schools (Sova 2001). Targets here have been The Tin Drum (1979), 1900 (1976), and Schindler’s List (1993).
Targets of censorship As the discussion above reveals, representations of sex and sexuality, race, ethnicity, violence, and religion have been the primary targets of state and nonstate efforts at censorship. However, it is important to note that the discussion has primarily focused on film in developed Western nations. In these regions, censorship is typically explicitly motivated by the desire to protect children from images that may be harmful to them and tacitly motivated by financial and moralistic concerns. Moreover, in these locations censorship is more likely to be the work of the film industry itself, rather than the state, since most forms of expression enjoy strong protection from governmental interference. The same cannot be said for other nations, China in particular, where political views and certain historical perspectives are more often than not the targets of state censorship. Films banned recently in China include Seven Years in Tibet (1997), for treatment of a free Tibet; Memoirs of a Geisha (2005), for fear of provoking anti-Japanese sentiment; Brokeback Mountain (2005), for depictions of homosexuality; and The Departed (2006), for the suggestion that China may use nuclear weapons against Taiwan. Chinese director Ye Lou’s Summer Palace (Yihe yuan) (2006) has also been banned by the Chinese government, since it deals in part with the 1989 student uprising at Tiananmen Square, and Ye has been banned from making films in China for five years. Somewhat ironically, China’s censorship committee, unlike the MPAA’s ratings board, includes “filmmakers, such as playwrights and directors, and . . . scholars on film study, film critics, and experts on film technology” (Liu 2007: online). Still, until 2006, filmmakers were obliged to submit full screenplays to the committee prior to beginning shooting. Allegedly motivated by the desire not to hold up production and to support the growing Chinese film industry, the government now requires only a short plot summary to be submitted for approval, unless the “film story is related to significant revolutionary or historical subjects, religion, diplomacy, minors, or judicial system, a complete script still must be submitted to the state bureau” (ibid.).
Arguments against censorship We have been working with a neutral construal of censorship, according to which censorship is just the imposition of some constraints or other – on content, on exhibition venues and so on. Still, it is true that censorship is typically thought to be a bad thing. To understand why censorship is problematic, when it is, we need to understand what it is that censorship threatens or damages, or alternatively, what values protections of expression protect. 34
TF17557.indb 34
16/9/08 15:47:17
C E NS OR SH IP
Why is freedom of expression valuable? Freedom of expression has been variously defended in terms of the necessity of maintaining a “marketplace of ideas” for the discovery and/or testing of the truth (Mill 1859), the bad consequences thought likely to ensue upon the suppression of expression, the role of free expression in democratic self-governance (Meiklejohn 1948), and the connection, thought to exist, between a person’s ability to express herself freely (and to hear others do so) and her dignity and moral autonomy (Dwyer 2001; Dworkin 1992). It is not obvious that these justifications are or need be mutually exclusive. But one point of difference among them deserves mention. The last idea – namely, that freedom of expression is connected with human dignity – has it that freedom of expression is a good, in and of itself. However, all the other justifications just mentioned are consequentialist – that is, they appeal to the likely outcomes of allowing or not allowing free expression. According to these views, then, the value of free expression is a function of the value of the end to which it is thought to be a (necessary) means; for example, the discovery of truth, the avoidance of tyranny, or the possibility of democracy. No matter how freedom of expression is defended, it is rarely defended as an absolute value, as one that may never be permissibly infringed. What counts as a sufficiently compelling reason to limit expression in any particular instance will depend on how freedom of expression is defended in the first place. Suppose, for example, that freedom of expression is defended on the grounds that it is essential to the pursuit of truth. Then, one might argue for the restriction of some form of expression by showing that it plays no role in furthering that goal. When it comes to defending the freedom of artistic expression, especially that at work in popular movies, we need to be particularly careful. For it is not at all clear that the typical products of Hollywood, or Bollywood, for that matter, contribute to the search for truth, are essential for a marketplace of ideas, or directly support democracy. It would seem that the firmest ground on which to defend the liberty of filmmakers is that which connects human artistic expression to human dignity and autonomy itself. This may appear to invest Hollywood movies with more gravitas than they deserve. However, it is important to think about the influence that movies have on individuals and the popular imagination more generally. What responsibilities does the movie industry have? In the face of complaints about violence, say, in contemporary movies, many people reply, “But it is just a movie! It is just entertainment!” Our question now is whether this is a plausible position for the film industry to adopt, such that filmmakers need not be concerned with the content of their creations (over and above their potential to entertain) or with the ways in which their creations may be interpreted. Is it reasonable for filmmakers to disregard the effects that the portrayals of women and ethnic and racial minorities may have on viewers’ attitudes toward members of these groups? 35
TF17557.indb 35
16/9/08 15:47:17
SUSA N DW Y E R
As Valenti (2000) notes, it is paradoxical of filmmakers and industry executives to say both (1) that their product is just for fun and (2) that their product is a serious instance of meaningful expression that falls under First Amendment protection. Films are one thing or the other. It is in the interests of filmmakers to cleave to the second claim. However, if they do, then (arguably) they must assume some responsibility for the likely effects and influences of their creations. Ironically, this was recognized in the Preamble to the previously mentioned Hays Code. It is worth quoting at length. Motion picture producers recognize the high trust and confidence which have been placed in them by the people of the world and which have made motion pictures a universal form of entertainment. They recognize their responsibility to the public because of this trust and because entertainment and art are important influences in the life of a nation. Hence, though regarding motion pictures primarily as entertainment without any explicit purpose of teaching or propaganda, they know that the motion picture within its own field of entertainment may be directly responsible for spiritual or moral progress, for higher types of social life, and for much correct thinking. Further: Hence the MORAL IMPORTANCE of entertainment is something which has been universally recognized. It enters immediately into the lives of men and women and affects them closely; it occupies their minds and affections during leisure hours; and ultimately touches the whole of their lives. The motion pictures, which are the most popular of modern arts for the masses, have their moral quality from the intention of the minds which produce them and from their effects on the moral lives and reactions of their audiences. This gives them a most important moral quality. Hence, the larger moral responsibilities of the motion pictures. Naturally, the film industry and individual filmmakers cannot be held individually responsible for the ways a particular person may interpret or react to a movie. Still, it would appear to be somewhat cavalier of filmmakers to claim that their creations have no influence whatsoever and that therefore they need not be concerned with these matters. Human social attitudes are formed under all manner of environmental influences, and it would seem to behoove the motion picture industry occasionally to examine what contribution it is making to this pool of influence.
36
TF17557.indb 36
16/9/08 15:47:17
C E NS OR SH IP
Self-censorship and creativity The last point brings us finally to the role that constraints in the form of selfcensorship might play in the creative enterprise itself. Presumably, expressive artists want their creation to succeed in some sense or other. There may be a message to convey, a set of emotions to evoke, a motivation to get into gear. Consider films such as Philadelphia (1993) (raising awareness of the stigma attached to AIDS), Norma Rae (1979) (concerning labor rights), and documentaries like An Inconvenient Truth (2006). Additionally or alternatively, a filmmaker’s primary purpose might be to amuse, horrify, or titillate. And, finally, the intentions of the creative artist might simply be to concretely express him- or herself. Succeeding in realizing any of these intentions and desires requires doing or saying one set of things rather than another. Quite trivially, this means that the artist is always working with a set of self-imposed constraints. If anything goes, then there can be no meaning. In addition to this perfectly straightforward point, it might be argued that constraints play a special role both in the manifestation of creativity and in the aesthetic value of a completed artwork. Elster (2000) importantly distinguishes between different types of self-imposed constraints; they may be invented by the artist him- or herself (as when novelist Georges Perec produced La Disparition, which contained not one instance of the letter e), or they may be chosen by the artist from the prevailing parameters of various genres (as when a director decides to make a romantic comedy or horror movie). Elster also emphasizes that a chosen set of constraints does not fully determine the final creative product, for an artist may continue to make choices within those constraints. Elster contends that artists make such choices in order to improve the aesthetic value of their creations. This is no doubt true in some cases, however, we may question, as Levinson (2003) does, whether Elster’s picture of artistic creativity entails an overly rational and self-conscious view of the artist at work. A slightly different example of how working with constraints is related to creativity is provided by Hjort’s discussion of Lars von Trier’s film The Five Obstructions (2003) (see “Further reading” in this chapter, “Dogme 95,” and “The Five Obstructions”). This film shows how Danish director, Jorgen Leth, remakes his short film The Perfect Human (Det perfekte menneske) (1967) in accordance with a number of technical and conceptual obstructions provided by von Trier. The obstructions pose significant challenges to Leth, and yet, in every instance, he produces a new version of The Perfect Human, each one a manifestation of profound creativity that would not have been possible without those very constraints. See also Authorship (Chapter 2), Pornography (Chapter 47), and Violence (Chapter 26).
37
TF17557.indb 37
16/9/08 15:47:17
SUSA N DW Y E R
References Congress, HUAC (1947) Hearings Regarding the Communist Infiltration of the Motion Picture Industry, 8th Congress, 1st session, 20 October. Dworkin, R. (1992) “The Coming Battles over Free Speech,” The New York Review of Books 11 (June): 55–8, 61–4. Dwyer, S. (2001) “Free Speech,” Sats: The Nordic Journal of Philosophy 2: 1–18. Eliasburg, J. (2005) “The Film Exhibition Business: Critical Issues, Practice, and Research,” in C. C. Moul (ed.) Concise Handbook of Movie Industry Economics, Cambridge: Cambridge University Press. Elster, J. (2000) Ulysses Unbound: Studies in Rationality, Precommitment, and Constraints, Cambridge and New York: Cambridge University Press. Gladchuk, J. J. (2007) Hollywood and Anti-Communism: HUAC and the Evolution of the Red Menace, New York: Routledge. Holland, B. L. (2006) “Note: Moral Right Protection in the United States and the Effect of the Family Entertainment and Copyright Act of 2005 on U.S. International Obligations,” Vanderbilt Journal of Transnational Law 39: 217n. Johnson, B. (2005) “In the Fray: Hollywood’s Last Taboo,” The Wall Street Journal (Eastern edition) (13 July): D10. Levinson, J. (2003) “Elster on Creativity,” in B. Gaut and P. Livingston (eds.) The Creation of Art: New Essays in Philosophical Aesthetics, New York: Cambridge University Press. Lewis, J. (2000) Hollywood v. Hard Core: How the Struggle Over Censorship Saved the Modern Film Industry, New York: New York University. Liu, W. (2007) “Censoring Movies Done according to Script,” ChinaDaily.com.cn. Available at www. chinadaily.com.cn/cndy/2007-08/31/content_6070081.htm. Meiklejohn, A. (1948) Free Speech and Its Relation to Self-Government, New York: Harper. Mill, J. (1859) On Liberty, London: J. W. Parker and Son. MPAA (Motion Picture Association of America) (2008a) Film Ratings. MPAA website. Available at www. mpaa.org/FilmRatings.asp. —— (2008b) Research and Statistics. MPAA website. Available at www.mpaa.org/researchStatistics.asp. MPPDA (Motion Picture Producers and Distributors of America) (1982 [1930]) “The Motion Picture Production Code of 1930,” in G. Mast (ed.) The Movies in Our Midst, Chicago and London: University of Chicago Press. Sova, D. B. (2001) Forbidden Films: Censorship Histories of 125 Motion Pictures, New York: Checkmark Books. Valenti, F. M. (2000) More Than a Movie: Ethics in Entertainment, ed. L. Brown and L. Trotta, Boulder, CO: Westview Press.
Further reading J. Green and N. J. Karolides (eds.) Encyclopedia of Censorship, rev. ed. (New York: Facts on File, 2005) contains exhaustive account of censorship of all kinds across the world. Lee Grieveson, Policing Cinema: Movies and Censorship in Early-Twentieth-Century America (Berkeley: University of California Press, 2004) gives a subtle theoretical analysis of early censorship in the United States. See also R. J. Haberski, Freedom to Offend: How New York Remade Movie Culture (Lexington: University Press of Kentucky, 2007); and Mette Hjort (ed.) Dekalog 1: On The Five Obstructions (London: Wallflower Press, 2008). Index on Censorship is a magazine devoted to the basic human right of the freedom of expression. See also C. Lyons, The New Censors: Movies and the Culture Wars (Philadelphia: Temple University Press, 1997).
38
TF17557.indb 38
16/9/08 15:47:17
4
CONSCIO U S N E S S Murray Smith [T]here is some deceiver both very powerful and very cunning, who constantly uses all his wiles to deceive me. There is therefore no doubt that I exist, if he deceives me; and let him deceive me as much as he likes, he can never cause me to be nothing, so long as I think I am something . . . this knowledge . . . I maintain to be more certain and evident than all I have had hitherto. (Descartes 1968: 103) Notwithstanding Descartes’ confidence in his own consciousness, and James’ influential characterization of conscious experience as a flowing “stream” (James 1950: 239), for much of the twentieth century, consciousness was put on the back burner of scholarly debate (though debate was hardly extinguished – consider Broad [1925], Ryle [1949], as well as the tradition of phenomenology – discussed elsewhere in this volume). Numerous forces conspired to put it there, but three factors might be singled out as particularly important. First, Freud. Notions of an “unconscious” dimension to the mind and behavior certainly existed before psychoanalysis, but there is little doubt that Freud and his followers put the notion at center stage, making it the fulcrum of the psychoanalytic theory of mind. Within the tradition of psychological behaviorism, consciousness perhaps fared even worse. While psychoanalytic theorists may routinely doubt the reliability of conscious avowals and states of minds, behaviorism denies, or at least remains agnostic about, their very existence, restricting itself to observable behavior, the billiard balls of stimulus and response. Even the advent of cognitive psychology – the paradigm-shifting force which overturned behaviorism and in certain ways began to rehabilitate the study of consciousness – continued (and continues) in its own way to keep consciousness at arm’s length. The dominant framework within cognitive psychology conceives of mental activity as computation, and the mind as an information processor; and to the extent that our chief model of an information-processing device – a computer – lacks consciousness, so the analogy does not make conscious computation very salient. A great deal of cognitive theory has concerned itself with information processing which is always or normally nonconscious – like the brain computations allowing you to see this page and to make sense of the words you are reading. Of these traditions, the first and the third – psychoanalysis and cognitivism – have exerted considerable influence on the study of film. Perhaps it
TF17557.indb 39
16/9/08 15:47:17
M U R R AY SM IT H
is not surprising, then, that the relationship between consciousness and film remains largely uncharted territory. The goal of this essay is to provide a preliminary map of the terrain, with some suggestions concerning lines of future inquiry. Despite the situation I describe above, the study of consciousness has been all the rage for a decade or more, to the extent that “consciousness studies” has now established itself, with all the paraphernalia of an academic discipline – journals, symposia, conferences, dedicated departmental centers and graduate-level courses, a steady flow of book publications, along with internal debates and controversies, which sometimes spill into the public domain (for overviews, see Flanagan 1992 and Carter 2002). What explains this resurgence? The emergence of cognitive theory made reference to internal states – thoughts, intentions, beliefs, feelings – respectable once more, while also engendering a new confidence that the nature of mind, so conceived, could become the subject of scientific investigation. Moreover, the initial emphasis on nonconscious states – or computations for which consciousness did not appear to be an important feature – eventually led back to consciousness, in the form of the question: if our minds can crunch so much information nonconsciously, why would we have evolved consciousness at all? In short, what are the functions of consciousness? One way of approaching this question is to consider the types of cognition that involve consciousness (Baars 1997). Our thinking becomes highly conscious when we face challenging or novel tasks. Learning to type, or drive a car, initially requires great conscious deliberation over the sequence of movements and steps involved. Mastery of such tasks is characterized in part by greater speed and fluidity in their execution, which goes hand in hand with an attenuation of conscious attention. As I type this sentence, I don’t consciously think about the location of individual letter keys – that aspect of the task has become automatic for me, freeing up processing capacity for the more demanding job of coming up with an overview of consciousness and its relationship with film. And once a skill like typing or driving has been mastered, finding ourselves becoming conscious of it usually impedes performance. But some tasks always seem to require conscious attention, as V. S. Ramachandran notes: Imagine you are driving your car and having an animated conversation with your friend sitting next to you. Your attention is entirely on the conversation, it’s what you’re conscious of. But in parallel you are negotiating traffic, avoiding the pavement, avoiding pedestrians, obeying red lights and performing all these very complex elaborate computations without being really conscious of any of it unless something strange happens, such as a leopard crossing the road . . . Intriguingly it is impossible to imagine the converse scenario: paying conscious attention to driving and negotiating traffic while unconsciously having a creative conversation with your friend. (Ramachandran 2003: 35–6) Ramachandran characterizes the conversation which requires conscious attention as “creative” in character, but any use of language – beyond the smallest of small talk and the most vacant of affirmatives – depends upon such attentiveness. 40
TF17557.indb 40
16/9/08 15:47:17
C ONS CIOUS N E S S
Consciousness also appears to be closely associated with detailed, fine-grained perception and cognition. Thus parts of the visual field that we fixate foveally, in the course of
consciously exploring it, will either be immediately available in consciousness, or poised for access in working memory, while those parts of the field which hover unscanned in peripheral vision are registered much more roughly. So it is that we can be blind to very significant elements in a visual array before us if we have devoted little or no conscious attention to them, a phenomenon captured in the related concepts of “inattentional blindness” and “change blindness” (Mack and Rock 1998; Levin and Simons 1997; Rensink et al. 1997). (Inattentional blindness and change blindness refer to the failure to perceive, respectively, stable features of a visual array, and apparently striking changes to a visual array.) These forms of perceptual “blindness” play an important role in relation to principles of image composition in film, as well as principles of continuity editing. The conventions of continuity editing keep our attention fixed on elements carefully controlled for continuity; other details across a sequence may be less consistent, but perceptual blindness makes it unlikely that we will pick them up. (The phenomenon of “blindsight” – registration of the information by the visual system with absolutely no conscious awareness of vision – might appear to undermine this relationship between consciousness and perceptual detail, but in fact it supports it. Blindsight demonstrates that perception can occur without consciousness, but the “resolution” of such unconscious perception is coarse, compared with that delivered by conscious perception.) Broadly, then, it seems that consciousness attends certain types of cognition in novel situations, where detail and vividness of perception and flexibility of mind and action are at stake, providing an adaptive advantage over creatures lacking such attributes. Given the centrality of novelty and vividness to artistic practice and experience, we should not be surprised to find consciousness playing a crucial role in relation to art – a point I return to in the final section. The question of the function(s) of the conscious mind has come to be known as one of the “easy problems” in debates around consciousness. The hard problem arises from the very fact that consciousness exists (Chalmers 1996; Clark 2001). A brain is a complicated thing, but it is made of nothing more than flesh and blood. How does it give rise to a conscious mind? How can meaning emerge from meat, subjectivity from synaptic firing? There are a variety of responses to the hard problem, each generating new problems. Substance dualism, which owes its modern philosophical origin to Descartes, holds that consciousness doesn’t arise from physical matter; it somehow coincides with it in human beings, residing as the “Ghost in the Machine” (Ryle 1949). Physicalists generally hold that consciousness supervenes on material brain states, prompting the search for the “neural correlates of consciousness,” that is, the specific brain states which correlate with – and thus are plausible candidates for the causation of – conscious states of mind. But as many skeptics would maintain – physicalists among them – we still don’t have a clue as to how it could be that a physical event, no matter how complex, could give rise to glimmers of consciousness. As David Chalmers puts it, “How could a physical system such as a brain also be an experiencer? . . . We do not just lack a detailed theory; we are entirely in the dark about 41
TF17557.indb 41
16/9/08 15:47:17
M U R R AY SM IT H
how consciousness fits into the natural order’ (Chalmers 1996: xi). Chalmers himself adopts property dualism: there may be no separate nonmaterial substance constituting consciousness, but it is irreducible to any currently conceivable physical explanation. A more extreme skepticism – the “new mysterian” – holds that an understanding of the relationship between matter and mind is permanently out of our reach, in the same way that calculus is beyond the ken of guinea pigs. Smart as we are relative to other species, there are going to be some problems which are beyond our comprehension, just because we don’t have the right neural equipment (McGinn 2006). Still another response to the hard problem goes by the name of “panpsychism,” arguing that the solution to the hard problem lies in not restricting consciousness to ourselves, or even to animal life in general; if we meat machines possess consciousness, perhaps consciousness is much more widely distributed in the physical universe than your average physicalist is wont to admit (Chalmers 1996; Strawson 2006). One other colorful member of the troupe warrants mention here: the eliminativist thesis that consciousness – at least as traditionally understood, as intrinsically private and ineffable subjective experience, the unique “what it is like” to be a certain individual in a state and situation – does not exist (Dennett 1991). None of these arguments is easy to swallow: the interest in the hard problem of consciousness seems inversely proportional to the plausibility of available solutions to it.
Consciousness analyzed Although reference to “consciousness” in general is standard in the contemporary debate on consciousness, implicit in that description are numerous types and levels of conscious experience. Breaking the overarching concept of consciousness down into various subtypes marks an important step toward its understanding. The first distinction we might draw lies between perceptual or “primary” consciousness (Edelman 1992), on the one hand, and self-consciousness, on the other. For an entity to have perceptual consciousness, it must have conscious awareness of some object or phenomenon – the increasing intensity of light, let us say, as the sun rises. Such consciousness might be momentary and episodic, or more continuous. One step on from this, an entity which possesses a conscious perceptual sensitivity to the state of its body we might say possesses “sentience.” Another elaboration involves the integration of perception with memory, setting the stage for personal identity, conceived in the Lockean tradition as an at least partly continuous, temporally extended conscious experience. But fully fledged self-consciousness involves a level of awareness well beyond, even if continuous with, these basic levels of consciousness. To have self-consciousness is to have knowledge of, and to attend to, the nature of one’s being in the world. Classically, the emphasis has lain on our awareness of ourselves as thinking, free-willing agents, transcending our status as physical, animal beings. A more contemporary picture would lay emphasis on the embodied nature of human existence, acknowledging our place in the physical and biological realms, while emphasizing the character and limitations of such embodiment. Such self-consciousness does not eradicate or transcend more basic forms of consciousness in human experience, however. Much of the time we 42
TF17557.indb 42
16/9/08 15:47:17
C ONS CIOUS N E S S
consciously attend to the task at hand, whether that be a relatively humdrum exercise (fixing a nail to a wall), or a more demanding and notionally “exalted” activity, such as focusing intently on an aesthetic object (a film or a piece of music, say). At the very least, for much of the time self-consciousness drops into the background as we extend ourselves into the world; indeed, constantly sustained and intense self-consciousness is probably pathological and maladaptive. Consciousness varies around other axes too. Consciousness comes in different grades; we can be conscious of things to different degrees. Watching a film, I am most conscious of those elements that grab my attention from moment to moment, like the flow of conversation and play of facial expressions between two characters. But at the same time I have some awareness of a large array of secondary and peripheral factors, such as the score and sound design of the film, the quality of the light in the depicted space, and objects in the space behind the characters. Theorists of film music often mark just this distinction in contrasting “hearing” with “listening,” where the latter describes focused attention on the soundtrack, the former a still conscious but more peripheral, and “lower grade,” form of attention (Kalinak 1992: 3). The dynamic and flexible nature of the stream of perceptual consciousness ensures that at any given moment, however, the items occupying the spotlight of conscious attention may change – a change in ambient sound breaks through into conscious awareness, and I ask myself, why do things sound different? Has a door been opened off screen? Psychologists of emotion have also marked a distinction between “declarative” and “representational” memory which hinges on the degree to which the memory can be brought to consciousness. A declarative memory is one which we can describe and discuss in language, a memory, of which – whatever its veracity – we feel we have a more or less complete understanding. I can recall vividly the occasion of my learning about the sudden and untimely death of a colleague. I know how it made me feel and I know why, and I can describe both the qualia of the state and its causes. Representational memories, by comparison, are elusive but potent states. Visiting a house I was familiar with in an earlier phase of life, I experience a definite unease, but I can neither recall nor articulate the cause of the unease. The sights and sounds of the house trigger certain memory associations; I consciously experience a certain kind of emotion, but cannot bring to consciousness the earlier experiences causing the emotion. Greg Smith (2003) and Karen Renner (2006) have both argued that films may exploit this form of memory through “emotional markers” – filmic motifs, like musical cues or lighting schemes, which redundantly mark a character or an action with a particular emotional tone without drawing great attention to themselves. The upshot of this strategy is that spectators have an intense experience of the emotion in question, with a very incomplete conscious knowledge of the devices in the film which have given rise to this experience. Although the model of mind here is not psychoanalytic, and in particular does not posit a mechanism of repression whereby certain experiences are actively “buried” beneath conscious attention, the concept of representational memory does acknowledge the existence of unconscious sources of conscious states of mind – mental states which are not merely currently non- or only
43
TF17557.indb 43
16/9/08 15:47:17
M U R R AY SM IT H
peripherally conscious, like the objects of secondary awareness as I watch a scene in a film, but largely or wholly unavailable to consciousness. To complete our sketch of the varieties of conscious experience, we need to acknowledge the place of dreaming as a form of consciousness. Dreaming as a form of consciousness?! Well, yes – insofar as we experience things in dreaming, it is a form of consciousness. The claim will only seem odd because of the legacy of psychoanalysis, which explains the functions of dreaming by recourse to its particular conception of the unconscious. Contemporary psychological research into dreaming, however, paints a very different picture, identifying the functions of dreaming in terms of memory and learning – dimensions of mind which often enter into fully conscious, waking experience. So there are two conceptual revisions at stake here: first, simply appreciating that while there is obviously an important difference between dreaming and being awake, when we dream we do not lack consciousness – as we do during other phases of sleep. And second, emerging evidence suggests that dreaming has evolved to enhance aspects of the conscious mind, not to provide a playground for experiences repressed by the conscious mind. The idea that film viewing is akin to dreaming goes back pretty well to the origin of cinema, and psychoanalytic film theory has erected elaborate theories on the basis of what was originally an evocative metaphor. This revised “cognitive” conception of the dreaming mind, brought within the orbit of conscious cognition, raises an intriguing question. We certainly don’t literally fall into a dream state when we watch films; but is it possible that we do enter a neural-cum-mental state distinct in certain ways from the mental state we possess when we engage with the world? A state so systematically in contrast with our waking navigation of the world that it warrants identification in the way that we identify dreaming as a distinct type of mental experience? Or to pose a parallel question, is there a distinctive mental state which constitutes the “aesthetic attitude” (Beardsley 1981) – a form of consciousness systematically distinct from ordinary, “interested” consciousness, characteristically prompted by films and other aesthetic works?
Consciousness in the history of film A phenomenon as central to human existence and identity as consciousness is bound to be manifest in the history of an art form in innumerable ways. We can, nevertheless, pick out a number of ways in which the idea or the fact of consciousness has not merely been present in film but is an explicit object of fascination or inquiry. The conscious mind – or at least the particular types of conscious state that comprise it – is integral to the earliest extended work of film theory in English, Hugo Münsterberg’s The Photoplay (2002 [1916]). Münsterberg focused on the way in which the development of technique and form over the first two decades of cinema sought (on his view) to mimic the mental mechanisms of attention, memory, and emotion. Few subsequent theorists put such an emphasis on the role of these basic conscious mental processes in shaping cinematic form; and, as we have seen, psychoanalytic theory played down their relevance, stressing instead the idea that a film, like a human subject understood in psychoanalytic terms, should be understood in terms of the dynamics of the 44
TF17557.indb 44
16/9/08 15:47:17
C ONS CIOUS N E S S
unconscious. Some recent theory has shown a renewed appetite for conscious mental processes, in terms of their significance for both the design of films and the experience of film viewing, but little work shows much interest in the fact that these mental processes are conscious. Such research has focused on the functional roles of various types of perception, memory, and emotion, rather than their qualitative character as conscious experiences. The real successors to Münsterberg’s early ruminations are not to be found in any body of film theory, but in particular traditions of filmmaking. The most obvious stamping ground for the representation and exploration of the conscious mind has been the tradition of art cinema (Kawin 1978; Bordwell 1985). One of the defining features of such cinema, and one that can be traced back to its roots in the 1920s movements of German expressionism and French impressionism (Abel 1984), is an emphasis on the rendering of subjective experience. Filmmaking in this mode confirmed the thrust of Münsterberg’s thesis, but expanded the range of mental states given expression to include fantasy and dream states (which, to reiterate a point above, we may regard as an important form of conscious experience). Broadly speaking, art cinema also expands the techniques through which subjectivity can be rendered, and often makes them highly salient. It is not that such techniques are wholly absent in the classical tradition of filmmaking; indeed, in certain periods and genres, they have been highly visible. But by and large, it is within art cinema that subjectivity itself has been dramatized. One director who confronted a particular conception of human consciousness in the classic, postwar era of European art cinema was Michelangelo Antonioni. Antonioni’s L’Eclisse (1962), the third of his famous trilogy, bears the marks of the widespread influence of Sartrean existentialism during this period. The film is restrained in its overt rendering of the subjectivities of its central characters, instead directing our attention to their outward behavior and the surface of the physical world. But the diffuse discontent and unfocused anxiety of the characters make themselves felt, hanging like a cloud over the entire film. András Kovács has argued that the characters suffer from the angst characteristic of those living in the disenchanted state described by Sartre – the absence of any traditional metaphysical guarantee leaves them facing “Nothingness,” all too conscious of the contingency of the world, their fragile and insignificant place within it, and the weight of their own decisions in shaping their destinies (Kovács 2006). Consciousness has also been explored in the still more rarefied atmosphere of avant-garde filmmaking. The movement known as structural filmmaking – essentially the manifestation of sixties art world minimalism in the context of avant-garde film – was given an early and influential interpretation as a “metaphor for consciousness.” The stripped down form of a film like Wavelength (1967) was held to represent, and invite the spectator to contemplate, nothing less than human “apperception” (selfconsciousness) itself. The gradual, “stammering” zoom across the loft space that forms the backbone of Wavelength represents, on this argument, not merely certain objects and the occasional event but, at a deeper level, the conscious subjectivity itself which might apprehend particular objects and events. Only by drastically minimizing what we normally perceive and cognize through consciousness can consciousness itself 45
TF17557.indb 45
16/9/08 15:47:18
M U R R AY SM IT H
become the focus of consciousness (precisely, self-consciousness), via the agency of the film-as-metaphor. Only by eliminating or attenuating the action available to be viewed through a window are we likely to attend to the window itself. In proposing this interpretation of Snow’s work, Annette Michelson (1971; see also Peterson 1994) was in part inspired by the repeated use of cinema as an analogy for conscious experience in philosophical and psychological writings – an image found in the writings of William James, Henri Bergson, and Edmund Husserl and still active today in the work of Oliver Sacks (2004). Citing Gérard Granel, Michelson likens the method of phenomenology to “an attempt to film, in slow motion, that which has been, owing to the manner in which it is seen in natural speed, not absolutely unseen, but missed, subject to oversight” (Granel 1968: 108, quoted and translated by Michelson 1971: 30; Michelson, p. 37, also describes Münsterberg’s The Photoplay as “an early and remarkable attempt at a phenomenological analysis of the cinematic experience”). Reciprocally, structural filmmaking offers back to philosophy this metaphor, now incarnate in film itself. (See also the comparable analysis of Derek Jarman’s Blue presented by Vivian Sobchack in her overview of phenomenology and film, this volume, p. 437.) Such filmic reflections on consciousness are not, however, the unique possession of such ostentatiously serious traditions of cinema as art and avant-garde cinema. Consciousness has been fertile material for Hollywood as well. Many contemporary science-fiction films thematize aspects of consciousness: in different ways, 2001: A Space Odyssey (1968) and I Robot (2004) both dramatize the possibility that a computer might develop consciousness and assert independent agency on the basis of it. On one reading, Blade Runner (1982) takes as its subject the centrality of memory, and awareness of death, in human consciousness (a theme central to the philosophy of Heidegger, in the form of “being-toward-death” – Heidegger 1962; Mulhall 2002). Scott’s film about the hunting down of a group of advanced, renegade androids explores the impact of doubt about the reliability of memory, uncertainty as to the duration of a human life, and the ever-present threat of death (and the demise of consciousness). We also find within contemporary Hollywood the “comedy of consciousness,” in films which exploit some of the absurdities implicit in our confused and often paradoxical assumptions about consciousness. Films like All of Me (1984), The Man with Two Brains (1983), and Being John Malkovich (1999) create humor from the dualist idea that the conscious mind can be decoupled from its physical seat and relocated in other bodies, and even objects (a joke that resonates with the doctrine of panpyschism – that consciousness, in different grades, is pervasive in the universe). In Being John Malkovich, Malkovich’s personality makes its way through several individual bodies, finally arriving, in rather sinister fashion, in the body of a child; while in All of Me, the soul of the character played by Lili Tomlin cohabits with Steve Martin in “his” body, before taking up residence in the body of Victoria Tennant (with a brief stay in a bucket along the way) (M. Smith 2006).
46
TF17557.indb 46
16/9/08 15:47:18
C ONS CIOUS N E S S
Qualia and (film) art In 1982 Frank Jackson introduced the world to Mary the color scientist (Jackson 1982; Ludlow et al. 2004). Mary has been raised in an artificially contrived blackand-white world, but as an expert color scientist she has nevertheless acquired a complete understanding of all there is to be known about the physics of color. One day Mary is exposed to colored objects: does she now learn anything new – that is, does she obtain new knowledge? Jackson (and many others) held that she does, and that no amount of propositional knowledge concerning the physical underpinning of color experience will deliver knowledge of (for example) what it is like to see red, the distinctive subjective “raw feels” that constitute conscious experience. The character of conscious experience as given by qualia is as such sui generis and, according to Jackson, exposes the falsity of physicalism as a metaphysical doctrine (the thesis that reality is constituted wholly by physical elements, their properties, and their relations). How does Jackson’s thought experiment, and the “knowledge argument” that it supports – that knowledge of the qualia that constitute conscious experience is a distinctive, irreducible form of knowledge – bear upon the making and appreciation of films? Like all art forms, film trades in the domain of phenomenal experience. The visual arts furnish us with novel visual experiences; the sonic arts provide us with new aural experiences; and the literary arts, while not directly rendering novel perceptual experiences, instead evoke phenomenal experiences through the imaginative positing of actual and fictional agents, spaces, and situations. Films characteristically work on all of these fronts. This emphasis on the conscious, qualitative character of our experience of art may strike many contemporary literary and art theorists as peculiar, or worse, naïve and even pernicious. But remove conscious qualia in the domain of art and you truly have, to adapt Kant’s well-known phrase, a system without purpose. The knowledge argument matters for a theory of art because it impinges on what we understand art to provide us with, and how we value it. If art can provide us with knowledge, and if it is uniquely or especially well-placed to furnish us with particular kinds of knowledge, then phenomenal knowledge – knowing what it is like to perceive and experience from a particular point of view – must be a frontrunner. If the knowledge argument fails, then its failure critically undercuts one of the grounds for attributing to art a special and unique epistemic role in human existence. What, then, can we say about the kinds of qualia created by different sorts of art? And how do the arts go about creating such qualia? Yasujiro Ozu’s The Flavour of Green Tea Over Rice (Ochazuke no aji) (1952), provides a good starting point. The title of the film refers to the traditional and simple meal of rice with green tea. The meal – which is shared in harmony by a married couple, who have been rowing over the course of the film – is symbolically laden, representing the virtues of a quiet, unpretentious, and relaxed attitude to life. But what is key here is the way that whatever symbolic value the meal has is embodied in the taste of the meal, or rather, for viewers of the film, in the way that the taste of the meal is evoked by the film. What this brings to 47
TF17557.indb 47
16/9/08 15:47:18
M U R R AY SM IT H
mind is another, rather more famous instance of the evocation of a specific taste in a work of narrative: Proust’s description of the madeleine in the opening pages of A la recherche du temps perdu (Remembrance of Things Past) (Proust 1981 [1913–27]). What both examples remind us of is the extent and depth of connectedness between the subjective experience of something as apparently simple as a taste, and the network of memory, cultural association and symbolism it can become integrated with: the way a whole life can be conjured up by a taste or a smell, and the way “thick” percepts and sensations like this – qualia laden with, or better, constituted by, memory, association, beliefs, desires, values – can be created and evoked by works of art. Other films which similarly evoke a dense network of associations through sensory imagery include Ozu’s An Autumn Afternoon, the Japanese title of which is Samma no aji – literally, “The Taste of Samma,” evoking “the brief season in late summer when samma, a variety of mackerel, is at its most savory” (Bordwell 1988: 370–1); Sergei Parajanov’s The Colour of Pomegranates (Sayat Nova) (1968); and Trân Anh Hùng’s The Scent of Green Papaya (Mùi ðu ðu? xanh) (1993). The work of Stan Brakhage takes us into the matter of art and qualia from a different angle. Brakhage was an experimental filmmaker whose films can most straightforwardly be likened with abstract expressionism. Some of Brakhage’s films are purely abstract; many contain representational elements, but even where they do so, the films retain a strongly abstract quality, in the sense that they put before us fields of textured light whose primary role is not to depict objects, characters, or events. In a famous passage from his essay “Metaphors on Vision,” Brakhage asks us to: Imagine an eye unruled by man-made laws of perspective, an eye unprejudiced by compositional logic, an eye which does not respond to the name of everything but which must know each object encountered in life through an adventure of perception. How many colors are there in a field of grass to the crawling baby unaware of “Green”? How many rainbows can light create for the untutored eye? (Brakhage 1963: 25) Brakhage’s films work against our natural inclination to find objects in the visual arrays we are presented with, instead presenting us with colors, forms, and textures which refuse to “attach” themselves to stable depicted objects. It is in this sense that the films create an “adventure of perception,” in which we are confronted with strikingly novel visual experiences. Comparing Brakhage’s art with the example of The Flavour of Green Tea Over Rice, we can mark a distinction between Brakhage’s ambition to create radically new qualia – ones which cannot be adequately described through familiar concepts or predicates like “green” – and Ozu’s ambition to evoke for us a familiar quale, even as he renews our awareness of it and bestows special significance upon it. One way in which this conception of the distinctive knowledge furnished by art has been articulated is through the notion of defamiliarization, first propounded by Viktor Shklovsky in 1919. Ordinary cognition, Shklovksy argues, works to “automatize” the perception of objects and execution of tasks with which we have become familiar. 48
TF17557.indb 48
16/9/08 15:47:18
C ONS CIOUS N E S S
And so . . . life fades into nothingness. Automatization eats away at things, at clothes, at furniture, at our wives, and at our fear of war. If the conscious life of many people takes place entirely on the level of the unconscious, then it’s as if this life had never been. And so, in order to return sensation to our limbs, in order to make us feel objects, to make a stone feel stony, man has been given the tool of art. The purpose of art, then, is to lead us to a knowledge of a thing. (Shklovsky 1991: 5–6) Shklovsky’s account of ordinary cognition, then, prefigures in outline one of the central ideas in cognitive neuroscience – that we become most conscious of an activity when it is novel. The vocation of art to defamiliarize or “make strange” ranges across all aspects of experience, from simple object perception to emotionally charged personal relationships and social events. The contrast between Brakhage and Ozu can be illuminated in these terms as well: where Brakhage works at (or even before) the level of object perception, Ozu focuses upon the experiences of family and community life. The contrast should not be oversimplified, however: many of Brakhage’s films represent existential themes like birth and death, and as we have seen, Ozu shows the interpenetration of perception, emotion, memory, and personal life. Each filmmaker is concerned with “life” in the broadest sense but tackles the task of representation through a different point of entry. The role of art in creating new qualia, and renewing our awareness of familiar qualia, does not end with the art object itself and the perceiver’s engagement with it. On this view of the function of art, one of the major roles of criticism is to evoke the qualia characteristic of particular works of art. Criticism in this mode sustains the focus on and renewal of the qualia of experience achieved by art, by describing and evoking what it feels like to engage with particular works of (film) art. The aim of such criticism is to describe, and thereby enable us to imagine, the set of “phenomenal information properties” (Dennett 1991) that, taken together, uniquely characterize a given work of art. And here the target of such description will be both the qualia represented by the work (the taste of green tea over rice), as well as the unique qualia furnished by the work itself (The Flavour of Green Tea Over Rice). The role of criticism in this respect is perhaps most obvious in the form of impressionistic passages, where critics attempt to capture the experience of viewing a film through metaphor and adjectival description. But, in this respect at least, such impressionistic criticism is continuous with more technical analysis. For those with the relevant knowledge, describing a cue within a film score as a “major seventh with a sharp attack and equally rapid decay,” or a film sequence as an example of “metrical montage at a tempo of 10 frames a second,” will serve to bring to mind the characteristic feel of the sequences in question. Again, it is important not to misunderstand the contrast here between impressionistic and more technical analysis: technical analysis enables precise description where it is possible, but there may be aspects of perceptual experience beyond the reach of such analysis (Raffman 1993), and this is where more indirect, especially metaphorical, description may play an essential role. Not all impressionistic analysis is mere belle 49
TF17557.indb 49
16/9/08 15:47:18
M U R R AY SM IT H
lettrisme. The vocation of art is the creation of novel and highly nuanced adventures in perception and cognition. The work of artists like Brakhage and Ozu is the call to which evocative criticism is the response, and the savoring of the qualia of conscious experience is central to both enterprises. See also Hugo Münsterberg (Chapter 39), Avant-garde film (Chapter 48), Phenomenology (Chapter 40), and Edgar Morin (Chapter 38).
References Abel, R. (1984) French Cinema: The First Wave, 1915–29, Princeton, NJ: Princeton University Press. Baars, B. J. (1997) In the Theater of Consciousness: The Workspace of the Mind, Oxford: Oxford University Press. Beardsley, M. C. (1981) Aesthetics: Problems in the Philosophy of Criticism, second edition, Inianapolis: Hackett Publishing Company. Bordwell, D. (1985) Narration in the Fiction Film, Madison: University of Wisconsin Press. —— (1988) Ozu and the Poetics of Cinema, Princeton, NJ: Princeton University Press. Brakhage, S. “Metaphors on Vision,” Film Culture 30 (special issue). Broad, C. D. (1925) The Mind and Its Place in Nature, London: Routledge and Kegan Paul. Carter, R. (2002) Exploring Consciousness, Berkeley: University of California Press. Chalmers, D. J. (1996) The Conscious Mind: In Search of a Fundamental Theory, New York: Oxford University Press. Clark, A. (2001) Mindware: An Introduction to the Philosophy of Cognitive Science, Oxford: Oxford University Press. Dennett, D. (1991) Consciousness Explained, Boston: Little, Brown, and Company. Descartes, R. (1968) Discourse on Method and The Meditations, trans. F. E. Sutcliffe, Harmondsworth: Penguin. Edelman, G. (1992) Bright Air, Brilliant Fire: On the Matter of the Mind, Harmondsworth: Penguin. Flanagan, O. (1992) Consciousness Reconsidered, Cambridge, MA: MIT Press. Granel, G. (1968) Le Sens du temps et de la perception chez Husserl, Paris: Gallimard. Kalinak, K. (1992) Settling the Score: Music and the Classical Hollywood Film, Madison: University of Wisconsin Press. Heidegger, M. (1962) Being and Time, trans. J. Macquarrie and E. Robinson, Oxford: Blackwell. Jackson, F. (1982) “Epiphenomenal Qualia,” Philosophical Quarterly 32: 127–36. James, W. (1950) The Principles of Psychology, vol. 1, New York: Dover. Kawin, B. F. (1978) Mindscreen: Bergman, Godard and First-Person Film, Princeton, NJ: Princeton University Press. Kovács, A. B. (2006) “Sartre, the Philosophy of Nothingness, and the Modern Melodrama,” in M. Smith and T. E. Wartenberg (eds.) Thinking through Cinema: Film as Philosophy, Malden, MA: Blackwell. Levin, D. T., and Simons, D. J. (1997) “Failure to Detect Changes to Attended Objects in Motion Pictures,” Psychonomic Bulletin and Review 4: 501–6. Ludlow, P., Nagasawa, Y., and Stoljar, D. (eds.) (2004) There’s Something about Mary: Essays on Phenomenal Consciousness and Frank Jackson’s Knowledge Argument, Cambridge, MA: MIT Press. Mack, A., and Rock, I. (1998) Inattentional Blindness, Cambridge, MA: MIT Press. McGinn, C. (2006) Consciousness and Its Objects, Oxford: Clarendon. Michelson, A. (1971) “Toward Snow,” Artforum 9: 30–7. Mulhall, S. (2002) On Film, London: Routledge. Münsterberg, H. (2002 [1916]) The Photoplay: A Psychological Study, in A. Langdale (ed.) Hugo Münsterberg on Film: The Photoplay: A Psychological Study and Other Writings, New York: Routledge. Peterson, J. (1994) Dreams of Chaos, Visions of Order: Understanding the American Avant-garde Cinema, Detroit: Wayne State University Press.
50
TF17557.indb 50
16/9/08 15:47:18
C ONS CIOUS N E S S
Proust, M. (1981 [1913–27]) Remembrance of Things Past, trans. C. K. Scott, M. Kilmartin, and T. Kilmartin, London: Chatto and Windus. Raffman, D. (1993) Language, Music, and Mind, Cambridge, MA: MIT Press. Ramachandran, V. (2003) The Emerging Mind: The Reith Lectures 2003, London: Profile Books. Renner, K. (2006) “Repeat Viewings Revisited: Emotion, Memory, and Memento,” Film Studies: An International Review 8: 106–15. (Special issue, “Film, Cognition, and Emotion,” edited by Daniel Barratt and Jonathan Frome.) Rensink, R. A., O’Regan, K. J., and Clark, J. J. (1997) “To See or Not to See: The Need for Attention to Perceive Changes in Scenes,” Psychological Science 8: 368–73. Ryle, G. (1949) The Concept of Mind, London: Hutchinson’s University Library. Sacks, O. (2004) “In the River of Consciousness,” New York Review of Books 51 (January 15). Shklovsky, V. (1991) Theory of Prose, trans. B. Sher, Elmwood Park: Dalkey Archive Press. Smith, G. M. (2003) Film Structure and the Emotion System, New York: Cambridge University Press. Smith, M. (2006) “Film Art, Argument, and Ambiguity,” in M. Smith and T. E. Wartenberg (eds.) Thinking through Cinema: Film as Philosophy, Malden, MA: Blackwell. Strawson, G. (2006) Consciousness and Its Place in Nature: Does Physicalism Entail Panpsychism?, Exeter, UK: Imprint Academic Press.
51
TF17557.indb 51
16/9/08 15:47:18
5
DEFINI T I O N O F “CIN EM A” Trevor Ponech Definitions and essences Philosophical and scientific requests for definitions often arise as “what is” questions. Here, noûs consists in proposing noteworthy features in virtue of which a thing or kind of thing, K, is to be identified. Proposals of that sort are apposite when users’ ideas about what it is a K-designating term picks out are sketchy, muddled, conflicting, or contraindicated by available evidence. The primary goal is to enhance theoretical understanding of K; if possible, to discover what is so special about it. The method is to learn what makes it true that something is K or an instance of K. Such a definition might further support legislating what a term ought to mean to competent users. Consider the International Astronomical Union’s 2006 redefinition of “planet,” informed by recent scientific research. Heretoforth, “planet” refers to a celestial body which, among other characteristics, has cleared the neighborhood around its orbit of other bodies (IAU 2007: 13–14). Certain items people believed or desired to be planets – notably, Pluto – now belong to a new category, “dwarf planet,” the members of which haven’t cleared their neighborhoods. The IAU has fixed the extension of a pair of terms on the basis of what a planet evidently is, after all. An ambitious answer to a “what is” question might take aim at a target’s essence. Essentialism is multiform. One important type of essentialist definition tries to list individually necessary and jointly sufficient conditions or properties. To be a K, an item must satisfy or have each of these; together, they are enough to identify that item as a K. A bachelor is an adult, male, unmarried person. Knowledge is justified true belief. Formulating a statement of necessity and joint sufficiency is a powerful tool for achieving conceptual precision and logical rigor; it clarifies and renders coherent the very idea of K. But strongly realist, explicitly naturalistic definitions suppose that certain concepts and their associated words have a less abstract, nonmental anchor in empirical reality. Of examples of nature’s kinds we might ask, what is it about this physical thing or sample of stuff that makes it a K (brain, tree, sample of gold, planet)? The realist’s response is that empirical investigation is conducive to knowledge of essences, that is, packages of precisely similar inner features possessed uniformly across
TF17557.indb 52
16/9/08 15:47:18
DEF I NIT ION OF “CI N EM A”
particular objects or individual samples. All samples of gold have the atomic number 79, melt at 1,073°C, and resist corrosion by chemical agents other than aqua regia. This property cluster is a candidate for gold’s real essence. A real essence is a mindindependent package of properties whereby an item is what it is. Disappearance of just one of these properties would end that item’s existence (Elder 2005). Granted, we always risk misidentifying things’ inner natures or encountering tough to classify cases. However, with Locke, one might be skeptical about our ability to discover essences without being antirealist about essences themselves (1979: III.vi.9).
What is cinema? Aesthetic philosophers and cinema scholars disagree over the conceptual content of the term “cinema.” Not that experts’ minds utterly fail to meet. They routinely make correct, intelligible use of this word, relative to prevailing beliefs, habits, practices, and conventions. Imagine I tell a colleague during projection of an immaculate 35-mm VistaVision print of The Searchers (John Ford, 1956), “Now that’s cinema!” She will likely grasp that I am expressing how impressive I find the motion picture image’s visible magnitudes – its size, luminosity, vivacity, and depth on the big screen. We make sense of one another’s cinema-talk by drawing upon our shared resources. These include generally recognized paradigm cases, mutual tacit assumptions about what it is like to have various sorts of viewing experiences, plus common knowledge of the intermeshing technological, economic, and artistic developments commonly associated with cinema history. So what’s the problem? The definition of “cinema” is controversial because people dispute both the nature of cinema and how, if at all, a definition beyond lexicology or social-history, in which ontology does the heavy lifting, could illuminate and unambiguously determine once and forever the reference of “cinema.” Whether cinema has an essence is the debate’s crux. CINEMA is distinct from the word “cinema.” Speakers of different languages can share my CINEMA concept, yet associate it with other lexical units (cinéma, Kino, cinematografo, eiga, diàn ying). It is a public concept insofar as different people can share it, in whole or in part. My CINEMA concept is similar, for instance, to Noël Carroll’s concept. Another philosopher might have a CINEMA concept so similar to Carroll’s that it makes sense to say they possess the same concept. We tell one CINEMA concept from the next not principally by whose brains they are lodged in but by their content. Roughly, content comprises whatever a concept is about or refers to beyond itself, along with the constraints on its applicability. CINEMA, as I understand it, refers to the aforementioned screening of The Searchers, among other things; it does not refer to moving imagery produced by flipbooks, because flipbook displays are not made of light. CINEMA is a mental item which sorts external objects, states of affairs, or events. Someone equipped therewith has a tool for differentiating between cinematic and noncinematic things and has ideas, however rudimentary, about what properties or conditions make it true that something is a candidate for cinemahood. Opinions differ 53
TF17557.indb 53
16/9/08 15:47:18
T R E VOR P ON E C H
regarding how to establish CINEMA’s content. Let’s turn to some of these differences, deferring awhile discussion of whether it is plausible that cinema could have a real essence to which CINEMA might refer.
Some essentialist conceptions Classic attempts to pinpoint cinema’s unique, distinguishing features often start from a quintet of precepts, adoption of which is understandable, given cinema’s most salient and familiar forms during the late nineteenth through mid-twentieth century. Theorists throughout much of the twentieth century tended to construct CINEMA concepts atop empirical assumptions to the effect that cinematic items paradigmatically if not necessarily are [a] pictorial images [b] made photographically with a camera, said images being [c] recorded, stored, and exhibited using flexible film strips, [d] their public exhibition normally employing a projector casting light through film strips and onto a screen so as [e] to produce the impression of movement. These presuppositions – plus the culturally overwhelming popularity of pictorial imagery and the longevity of early twentieth-century cinema’s dominant technologies – help explain why “film,” “movie,” “motion picture,” and “cinema” are still used synonymously. One grand old definition nominates as cinema’s essence photogénie (Delluc 1920; Epstein 2004; Morin 2005). When its proponents talk about “cinema,” they have in mind things matching [a] through [e]. They do not, however, conceptualize the empirical quintet itself as marking the cinematic. Their idea is that something which is [a], [b], [c], [d], and [e] is apt to have a certain other, definitive property. Photogénie eludes precise analysis but seems generally thought a capacity to trigger archaic affective responses. First published in 1956, Edgar Morin’s anthropologically inspired reflection on the nature of cinema treats photogénie as a sort of partnership between rationality and superstition. People of modern, industrialized societies harbor a residual propensity toward magical, supernatural thinking; imaginatively and emotionally, they remain in the thrall of phantoms, doubles, divinities, and occult forces. Thanks to the thaumaturgical technologies they employ, movies, in their most popular form, possess an “inner power” (Morin 2005: 7) to arouse the homo demens in us. This power derives from their being at once realistic and ethereal. Moving, photopictorial imagery of persons and things makes viewers feel that the image’s objects are both absent and present, as if these objects’ mystical doubles or lifelike shadows had broken free of the bodies that cast them. Projected across the screen, the image is, moreover, “dematerialized, impalpable, fleeting,” its ghostly charm magnified (34–5). In declaring photogénie cinema’s essence (7), Morin primarily means “essence” teleologically, as that anthropocentric purpose explaining cinema’s existence. His essentialism is thus only as defensible as his sweeping, a priori teleological speculation that cinema’s invention is the expression and fulfillment of “the imaginary man,” that part of us which lives in and through myths, fantasies, and dreams. Were this social function a necessary, nonaccidental condition for the emergence and persistence of cinematic phenomena, it still wouldn’t suffice to divide cinema from noncinema. Morin himself believes numerous cultural artifacts, including photography, share 54
TF17557.indb 54
16/9/08 15:47:18
DEF I NIT ION OF “CI N EM A”
cinema’s raison d’être and power to elicit archaic, magical thinking. The implied solution to this problem relies on the empirical quintet. Movies’ “inner power” to trigger archaic thinking depends on – is perhaps reducible to – their being moving pictorial images and so on. Let these jointly distinguish cinema from noncinema. But now, instead of the essence, photogénie is at most part of a package of reputed defining features, albeit a specially valued one which might explain cinema’s development and nearly universal appeal. Elucidating human nature is Morin’s ulterior motive for pondering cinema’s nature. Other theorists are driven principally by normative aspirations. As movie cameras, film stocks, optical systems, and engineering strategies for integrating them evolved, so did arguments, not yet concluded (Scruton 2002; Gaut 2002), about whether these burgeoning technologies’ confluence could produce art versus mechanical records and reproductions. During the late 1920s and the 1930s – the silent era’s tipping point and denouement – Rudolf Arnheim (1957) published several defenses of cinematic art. These generally rely upon various premises regarding cinema’s Material, its medium. Arnheim believes cinema’s medium is necessarily the moving or animated image projected from a film strip. Significantly, he does not consider [b] photographic imagery a necessary feature of the medium. Furthermore it is possible that he would permit modifying [a] to include nonpictorial imagery, as suggested by his express approval of abstract movies (5). However, [b] troubles Arnheim. Though recognizing it as a key artistic resource, he’d rather cinema abandon photographic representation to become “a pure work of man” by taking the form of “animated cartoon or painting” (213). Why? As Carroll (1988a: 27) explains, Arnheim shares his opponents’ premise that mechanically recording reality indeed is antithetical to creativity and expressivity, hence to art. His mission is to show that cinema is not art’s mechanical antithesis. He proceeds by citing the medium’s alleged failures and limitations: the absence of sound and color, the fragmentation of the spatial-temporal continuum, the lack of three-dimensionality, lessening of depth perception, and the loss of constancy of size (Arnheim 1957: 127–30). Arnheim thinks these deficiencies differentiate cinema from other media and, crucially, that they prove that film neither perfectly reproduces reality nor replicates human perception. Instead, the medium, imperfectly, idiosyncratically molds, transforms the world according to its own principles. If one skillfully seizes upon these as working principles, one can decisively turn cinema to expressive ends and thereby leverage it into art. But Arnheim’s CINEMA concept does not postulate [a]–[e] as cinema’s real essence. His argument is not that the quintet describes necessary and sufficient conditions for membership in a certain category of physical object or substance. Cinema’s “true meaning” as art form is what he claims to have ascertained (210–11). That there exists a medium actually, not essentially, comprising the quintet is an empirical given. And given its inherent properties and constraints, we can figure out how to exploit that medium so to realize a unique, authentic art form. Following Arnheim, this medium dictates that film art must reject extraneous properties like sound, recorded dialogue, and color photography, should embrace montage, and ought to avail itself to photographic and nonphotographic techniques that promise to bring cinema closer to being animated painting. 55
TF17557.indb 55
16/9/08 15:47:18
T R E VOR P ON E C H
Arnheim’s normative interest in establishing the content of CINEMATIC ART qualifies him as a “medium-essentialist.” Medium-essentialism posits an art form’s essence – in the sense of its telos, i.e., its specific aims, effects, and style – on the basis of empirical observations regarding whatever is taken to be the art form’s distinctive physical medium (Carroll 1996: 50). André Bazin, whose writings of the 1940s and ’50s influenced Morin, similarly proposes a medium-essentialist CINEMATIC ART concept. Unlike Arnheim, Bazin makes a strong claim regarding the nature of cinema’s medium. Journalist, rather than academic, Bazin does not argue about necessary-sufficient conditions and properties. Nonetheless, he apparently believes he has isolated three necessary features accounting for a fundamental difference between cinema and the material bases of other art forms. To wit, cinema is [a] pictorial imagery, [b] produced by photographic means and displayed so as [e] to produce the impression of movement. The linchpin of Bazin’s CINEMA concept is [b]. Photography endows imagery with “objectivity,” by which he intends an automatic, natural, mind-independent ontological connection between the image and its object. Photography by the same token endows film with a unique relation to motion and change: “[C]inema is objectivity in time . . . the image of things is likewise the image of their duration” (Bazin 1967: 14–15). Bazin goes so far as to invite the thought that, owing to its objectivity, a photograph does not just resemble its model, it is the model, having snatched something of the model’s existence or identity from time’s corrupting effects (14). Unequivocally, though, he concludes that cinema’s telos is realism. This realism is partly a psychological phenomenon, because it has the power to satisfy the desire to preserve existence against time. It is also aesthetic, cinema’s objectivity having the power to fulfill the perennial figurative artistic ambition to make ever more lifelike representations of reality, untrammeled by convention and artifice. Authentically cinematic art is that which embraces technologies, styles, and content – long takes, deep focus, camera movement, panchromatic film, and sound recording; characters, settings, or problems drawn from ordinary life – consistent with fulfilling this realist telos versus, say, Arnheim’s expressivist medium-essentialism. Not every essentialist definition advocates a teleology. Gregory Currie offers one he considers no more prescriptive “than the claim that water is essentially H2O” (1995: 1). That analogy is telling, because Currie holds cinema to be “something like a ‘natural kind’ ” (Currie 1998: 357). He concedes that, relative to our contemporary, pluralistic attitudes to cinematic technology and art, his CINEMA concept is stipulative: He would reserve “cinema” for a naturally identified category comprising just some of the many sorts of items intelligibly collected under this term. His stereotypical examples of cinema are projected from film prints. But to fall under Currie’s CINEMA concept, it is necessary and sufficient that something be a photographically produced pictorial representation delivered onto a surface so as to produce, or have the power to produce, an apparently moving image (1995: 2, 4). Identifying [a]-[b]-[e] of the quintet as cinema’s essence, Currie’s CINEMA concept reprises the core tenets of Bazin’s definition – although extravagant metaphysical ideas about the image’s identity with its object play no role here. Currie is also adamant that his definition embodies no aesthetic recommendations for maintaining cinema’s 56
TF17557.indb 56
16/9/08 15:47:18
DEF I NIT ION OF “CI N EM A”
artistic purity. He means only to restrict the extension of “cinema” to a familiar quasinatural kind. Though artifacts, instances of cinema are like instances of gold because “cinema,” like “gold,” can be used to pick out something possessing a certain cluster of mind-independent properties. Owing to natural regularities governing reflected light and its interactions with photosensitive receptors, photographically generated imagery is automatically, all things being equal, pictorial imagery of some real object. The image’s existence and visible condition naturally depend on, and naturally afford information about, its object’s condition. The specifically cinematic image can likewise preserve transtemporal information (1995: 69). If its object moves, the image changes correspondingly, such that viewers have perceptual experiences as of the object’s movement taking just so long to occur and as of one stage of its trajectory following another.
Some nonessentialist conceptions Nonessentialists generally charge that, considered as art form, material basis, and sociohistorical phenomenon, cinema is richer and more complex than essentialists fathom. It has become commonplace since the late 1960s to observe that cinema, from an historical perspective, derives from so many technological, cultural, and artistic sources, and is so open to technological and stylistic change that it is implausible to ascribe to it a single, immutable essence. Alluding to its multifarious origins – including prephotographic optical devices like the magic lantern and phenakistoscope, painting, cartoon strips, theatrical melodrama, pulp novels, and magic shows – Peter Wollen (1969: 153) denies any one dimension of cinema can be nonarbitrarily validated as grounding or mandating a specifically cinematic aesthetics. His own CINEMA concept holds cinema a “strikingly mixed and impure” (153) semiotic system. Its impurity, consistent with its mixed origins, is reflected in a movie’s being a “text” constructed of iconic, indexical, and symbolic signs. Medium-essentialisms err by unilaterally anchoring CINEMATIC ART in one kind of sign at the others’ expense. Bazin’s realism skews toward indexical signs, the significance of which depends on their causal or existential bonds to what they signify; Arnheim’s expressivism skews toward symbolic signs, the significance of which depends on human agency and conventions. In fact, says Wollen, only by considering the interactions of a triad of signs can we begin to understand cinema’s aesthetics (141). Wollen’s CINEMA concept is not about medium-specifying material parts and properties. It pertains to a type of signification, a form of production and exchange of meaningful signs. Conceived as cinematic text, a movie is a locus of signifying properties, functions, and processes. Cine-semioticians’ paradigmatic texts frequently match the empirical quintet. Wollen himself confines discussion to movies fitting the stereotypical empirical profile insofar as his focus is projected film texts of a predominantly indexical-iconic character, those largely comprising signs standing in an existential relationship to their objects as well as resembling them (165). But officially, semiotics doesn’t identify CINEMA with any cluster of necessary empirical properties. 57
TF17557.indb 57
16/9/08 15:47:18
T R E VOR P ON E C H
Indeed semiotics is generally presumed “absolutely non-essentialist” for regarding cinema as sociohistorically instituted (Nowell-Smith 1976: 40). Cinema is described as a specific discursive form delimited by the conjunction of two sorts of determinants. On one hand are socially constructed “codes” – like those allegedly governing meaning production and comprehension through continuity editing – prevailing at particular historical moments. These arbitrary signifying practices, conventions, and rules operate within yet transcend individual movies. On the other hand are the technical and material determinants serving to “inscribe” movie texts and the codes guaranteeing their meanings. These, too, change historically. For example, safety film and digital editing supersede nitrate stock and cutting-and-splicing. Though widely embraced by theorists, premises to the effect that movies are language-like texts, and cinema, a signifying system, attract intense philosophical criticism (Carroll 1988b; Currie 1993; Harman 1999). Furthermore, cine-semiotics is hardly free of essentialist tendencies. Its vanguard theorist proposes that cinema’s unique, identifying feature is the “imaginary signifier” (Metz 1982). Sarah Bernhardt, live on stage, could be, but need not be, a signifier. She might represent Phèdre, or simply greet the audience as herself. Regardless, she and her public are co-present at a proximal spatiotemporal location. Bernhardt photopictorially recorded in a movie is necessarily a signifier, namely, the-image-of-Bernhardt. It is unclear how Bernhardt, the image’s referent, is a real part or property of the signifier itself. That the-imageof-Bernhardt is an imaginary signifier evokes the putative fact that it is both absent, since Bernhardt is not there in the room; and present to audiences, as a manifestly real screen image consisting of her “phantom,” “double,” or “replica” (45). This indexical-mimetic phantom moves about, even makes sounds. Thanks to this vivacity, only cinematic representation contains signifiers that are absent, yet present to much the same degree that on-stage Bernhardt is present. Explicitly psychoanalytic, with intimations of photogénie and Bazinian realism, the further claim is that cinematic imagery, fascinating spectators with absent presence, is an especially powerful trigger of pleasurable psychic responses. Noël Carroll offers an alternative nonessentialist CINEMA concept as a class of artifacts conceived and constructed by human agents with sociohistorically varying practical, communicative, and artistic aims in mind. Membership involves satisfying five necessary but not jointly sufficient conditions. The venerable empirical quintet is virtually banished from this definition. No particular medium – no specific material basis, physical process, implement, technology, or congeries thereof – figures in the definition. No particular entrenched social function – representation, art-making, gratification of primordial human desires – is associated with that which identifies cinemahood. Carroll also disavows interest in grounding a CINEMATIC ART concept, noting that his five conditions have no implications regarding which if any formal and stylistic features befit uniquely, authentically cinematic artworks (1996: 72). Let “cinema” designate any image that is [1] two-dimensional, [2] presented in a detached display, [3] mechanically generated from a [4] template, and [5] produced by such means that the impression of movement is technically possible. Condition [1] 58
TF17557.indb 58
16/9/08 15:47:18
DEF I NIT ION OF “CI N EM A”
differentiates cinematic artifacts from sculptures, music boxes, and the like. Display detachment concerns the way in which cinematic like other imagery involves spatial dislocation. Unlike windows, mirrors, and magnifying lenses, movie pictures do not connect us spatially with their depicta. Merely by looking at movie imagery of it, one can’t orient one’s body to Mt Aconcagua’s direction. A template, [4], is any mass-producible storage, recording, exhibition, or transmission format, including but not limited to DVDs, videocassettes, MPEG files, broadcast signals, and film prints. Coupling templates with appropriate optical systems generates two-dimensional images in detached displays. Though a sort of “performance,” this coupling process differs from live theatrical performance: its achievement consists of the technology operating, or being operated by somebody, according to its mechanical specifications. Finally, [5] accommodates commonsense empirical as well as philosophical (Sparshott 1979: 321) intuitions that movement is a defining feature of cinema. Trouble is, some movies contain no movement. Hollis Frampton’s Poetic Justice (1971–2) is a film of a shooting script sitting on a tabletop; Michael Snow’s One Second in Montréal (1969) consists exclusively of still photos. Yet these movies share with orthodox ones the property of having been produced using technologies making a certain result, moving imagery presented in a detached display, a possibility. Their makers could have employed their chosen technology to generate moving images, had they preferred. While categorically impossible for drawings, paintings, or photographic stills to be moving imagery, “movement in a film image is an artistic choice which is always technically available” (Carroll 1996: 64). This definition is strikingly inclusive and not at all parochial. No style of imagery is deemed more cinematic or more artistically valuable than another. No restrictions are imposed on the means by which conditions [1] through [5] are satisfied. Cinema might be instantiated photographically or digitally, by cell animation, or by relatively heteroclite methods like joining strips of clear and opaque leader. Apropos, Carroll’s definition dissociates cinema from depiction. Indeed he prefers the less freighted term “moving image” to “cinema” for its distance from connotations of an art form, but also for its applicability to pictorial items (photographically or otherwise made), as well as purely abstract, nonpictorial ones. He would regroup under CINEMA both The Searchers and Ernie Gehr’s History (1970), made without a camera by exposing raw stock to dim light and consisting of scintillating beads of illumination distributed within a depthless black space. Moreover, Carroll’s definition regards cinema as protean: novel technologies of moving-image production, dissemination, and spectatorship constantly emerge, accompanied by new formal possibilities; imagemaking practices, and imagemakers’ communicative and artistic aims and strategies undergo continuous historical mutation. Conditions [1]–[5] individuate a concept capable of distinguishing instances of cinema from painting, sculpture, photography, and live theater in all its manifestations. Carroll believes [1]–[5] necessary if CINEMA is to be a distinct, coherent concept, reliably enabling us to identify instances of cinema, but denies they are sufficient to differentiate cinema from noncinema. The five jointly grant cinemahood to arguably precinematic items like mass-produced flipbooks and series of still 59
TF17557.indb 59
16/9/08 15:47:19
T R E VOR P ON E C H
photographs animated by zoetrope devices. And they leave open whether certain esoteric items satisfying the five conditions – e.g., a radically minimalist pitch-black “movie” made with underexposed film – are cinematic. These classification problems are endemic. Cinema, being a protean class of artifacts, cannot be identified with reference to closed, preexisting boundaries awaiting discovery and conceptualization. Hence an essential definition is unavailable, meaning decisions about how to classify hard cases must be outsourced. Invoking a tactic he advocates while discussing definitions of “art” (2001), Carroll recommends considering history and context. Hence, were it “part of an intelligible, ongoing filmworld conversation” (1998: 329), we would be warranted in taking aboard the radically minimalist candidate, its berth secured by its relevance and contribution to practices, concerns, and interests generative of more clear-cut cases of cinema.
The nature of cinema? Nonessentialism conceives cinema’s identity as mutable insofar as its constitutive technologies, practices, and conventions are variable. Relative to gold, it is hard to believe cinema has a real essence. Essentialist definitions seem naïve and illiberal for treating some selection of historically contingent properties, like the empirical quintet or part thereof, or adventitious functions, practices, and values, like one or another strain of realism, as real essences. Perhaps cinema is ultimately a projection of our CINEMA concepts. “[C]inema has not yet been invented,” muses Bazin (1967: 21), alluding to a Platonic CINEMA eidos, a “myth of total cinema,” awaiting full instantiation. More mundanely, CINEMA’s content is up to us. What counts as cinema is up to us, too. Objects so counted are of our conception and making. Someone might concede cinema’s inexpugnable conceptual and historical dimensions yet still use “cinema” to pick out items having a possible real essence. Principally, these are artifacts. Some philosophers hold artifacts in low ontological esteem, reasoning that, even if they do genuinely exist, they cannot have essences because an artifact’s being the kind of thing it is perforce derives from our attitudes and doings, not from nature. Critics find this argument tendentious, given that we are part of nature (Baker 2004; Elder 1995) – as is every physical part, particle, and process composing the artifact. Moreover, there are methods of categorizing artifacts other than with reference to their anthropocentric functions, histories, and meanings. One way to categorize artifacts is with reference to their intrinsic physical properties. Thus an artifactual kind, like a natural kind, can be identified with a collection of objects each of which possesses a precisely similar package of inner properties. Because essentially intention-involving, the causes of an artifact’s nature, the clustering of its identifying properties, and the multiplication of similar artifacts differ from those underlying instances of gold and tigers. We needn’t posit or know these intentions in order to collect members of an artifactual kind. Hypothesizing the aforementioned property clusters does the job, too. Cinema’s identifying property cluster might be that of a stroboscopic-luminescent field or visual display (Ponech 2006, 2007). Described synchronically, the strobolumi60
TF17557.indb 60
16/9/08 15:47:19
DEF I NIT ION OF “CI N EM A”
nescent display (SLD) is a delimited, spatially contiguous field of resolving elements comprising points of light – “pixels” – plus any nonilluminated areas and intervals separating them. These pixels have intensity, brightness, and color. Diachronically, an SLD manifests for the duration of its existence a kind of dynamic or movement: it has the property of “stroboscopy,” consisting of the continuous, high-frequency redistribution of illumination across a reflective or light-emitting surface. During this cycle of phase changes, usually sixty or more per second, pixels refresh, that is, they independently vary their brightness, intensity, and color. “Cinema” might be defined as any such stroboluminescent field or display, whatever its provenance, engineering specifications, or function. Ordinarily, SLDs are vehicles for achieving or presenting expressive and communicative acts, including artworks. In the preponderance of ordinary instances, the active play of light comprising the display is informed by some template, to borrow Carroll’s term, which itself has been given a certain structure or content in an effort to adapt it to serve expressive purposes in conjunction with an SLD. But from a naturalistic perspective, being an SLD, rather than being an SLD with a certain psychohistory or one coupled with a template, is what counts decisively toward cinemahood. Perhaps cinema is only contingently an artifact. It is not inconceivable that an SLD could spring into existence in a mind-independent, arational way, the result of a mindless cosmic accident. Absent sentient beings to make, exploit, and view it, the SLD would still have the parts, properties, dispositions, and causal powers by virtue of which it is an SLD, versus a tree, tiger, gold sample, or airplane. In contrast, concepts like ART, LITERATURE, MUSIC, and MOVIE presumably cannot be grounded in strictly internal property packages uniformly observable across individual physical things. A material artifact’s, α’s, also fitting the MOVIE (ART, etc.) concept, depends upon a network of complex psychohistorical events and relations obtaining between α and the relevant thoughts and actions of some cultural agent(s). Psychohistorical facts – facts about why and by whom they were made – are (partially) constitutive of α’s being a movie (art, literature, etc.). The concept CINEMA need make no essential reference to agency. To be sure, we employ what amounts to an anthropic principle when identifying candidate SLDs. That there could exist cosmologically large or microphysically tiny SLDs seems a given. But the ones of interest to us are those we can pick out on the basis of their effects on us. That anthropic principle does not turn the SLD into something subjective or lacking properties that do not depend on us, on our consciousness of and responses to it. It is a mind-independent object, possessing just the complex qualities that it has. One such property is its power or disposition to appear to percipients such as ourselves as (part of) a spatiotemporally contiguous, delimited stroboscopic field. Essentialism – the belief that certain traits are indispensable to an object’s or sort of object’s existence – is philosophically contentious. It is difficult to establish that things have the essences attributed to them and to demonstrate that our apparent knowledge of real natures is not ultimately a projection of our capacities and conventions of individuation, identification, and categorization. These framework debates are not the only ones troubling stroboluminescence’s candidacy for cinema’s real 61
TF17557.indb 61
16/9/08 15:47:19
T R E VOR P ON E C H
essence. Whether the foregoing description of this property cluster is true awaits empirical verification. Which if any of the items we label “cinema” are SLDs is also a live empirical question. Another question is why we should accept stroboluminescence as a constraint on what we mean by “cinema.” Cinema-talk is mostly about works, canons, institutions, practices, traditions, revolutions, representations, experiences. A reductive, naturalistic CINEMA concept ignores cinema as a human phenomenon. A nuclear definition of “cinema” does not reduce cinema as we know it to some putative essence. Stroboluminescence is another in a line of proposed restrictions on identifying and individuating one basic unit of analysis in the humanistic study of that totality of works, canons, institutions, etc. This restriction is nonetheless latitudinarian. It eschews medium-essentialism and dissociates cinemahood from any particular technology, causal chain, or cognoscenti “conversation.” Yet a naturalistic definition contains the resources for adjudicating hard cases, thereby excluding flipbooks, zoetropes, and, pace Carroll, lightless cinematic displays. It also clarifies that which is so special about cinematic artifacts. They are dynamic. Movement – change, if you prefer – is deep within their nature. Cinematic motion is most apparent in the redistribition of patterns of illumination such that reidentifiable items seemingly undergo displacement within the display space. Stroboluminescence, rather than unspecified means, underlies this identifying technical possibility. See also Medium (Chapter 16), Ontology (Chapter 20), Film as art (Chapter 11), Realism (Chapter 22), Semiotics and semiology (Chapter 42), Rudolph Arnheim (Chapter 27), and Edgar Morin (Chapter 38).
References Arnheim, R. (1957) Film as Art, Berkeley: University of California Press. Baker, L. R. (2004) “The Ontology of Artifacts,” Philosophical Explorations 7: 99–112. Bazin, A. (1967) What Is Cinema?, trans. H. Gray, Berkeley: University of California Press. Carroll, N. (1988a) Philosophical Problems of Classical Film Theory, Princeton, NJ: Princeton University Press. —— (1988b) Mystifying Movies: Fads and Fallacies in Contemporary Film Theory, New York: Columbia University Press. —— (1996) “Defining the Moving Image,” in N. Carroll (ed.) Theorizing the Moving Image, Cambridge and New York: Cambridge University Press, 49–74. —— (1998) “The Essence of Cinema?,” Philosophical Studies 89: 323–30. —— (2001) “Identifying Art,” in N. Carroll, Beyond Aesthetics: Philosophical Essays, Cambridge and New York: Cambridge University Press, 75–100. Currie, G. (1993) “The Long Goodbye: The Imaginary Language of Film,” British Journal of Aesthetics 33: 207–19. —— (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge: Cambridge University Press. —— (1998) “Reply to My Critics,” Philosophical Studies 89: 355–66. Delluc, L. (1920) Photogénie, Paris: De Brunhoff. Elder, C. L. (1995) “A Different Kind of Natural Kind,” Australasian Journal of Philosophy 73: 516–31. —— (2005) Real Natures and Familiar Objects, Cambridge, MA: MIT Press. Epstein, J. (2004) “On Certain Characteristics of Photogénie,” in P. Simpson, A. Utterson, and K. J.
62
TF17557.indb 62
16/9/08 15:47:19
DEF I NIT ION OF “CI N EM A”
Shepherdson (eds.) Film Theory: Critical Concepts in Media and Cultural Studies, vol. 1, London and New York: Routledge, 52–6. Gaut, B. (2002) “Cinematic Art,” Journal of Aesthetics and Art Criticism 60: 299–312. Harman, G. (1999) “Semiotics and the Cinema: Metz and Wollen,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 5th ed., New York: Oxford University Press, 90–8. IAU (International Astronomical Union) (2007) Information Bulletin 99. Available at http://www.iau.org/ fileadmin/content/Ibs/ib99.pdf (accessed 11 May 2007). Locke, J. (1979) An Essay Concerning Human Understanding, ed. P. H. Nidditch, Oxford: Clarendon; and New York: Oxford University Press. Metz, C. (1982) The Imaginary Signifier: Psychoanalysis and the Cinema, trans. C. Britton, A. Williams, B. Brewster, and A. Guzzetti, Bloomington: Indiana University Press. Morin, E. (2005) The Cinema, or the Imaginary Man, trans. L. Mortimer, Minneapolis: University of Minnesota Press. Nowell-Smith, G. (1976) “Moving on from Metz,” Jump Cut: A Review of Contemporary Media 12–13: 39–41. Ponech, T. (2006) “The Substance of Cinema,” Journal of Aesthetics and Art Criticism 64: 187–98. —— (2007) “Cinema Again: A Reply to Walley,” Journal of Aesthetics and Art Criticism 65(4): 412–16. Scruton, R. (2002) “Photography and Representation,” in A. Neill and A. Ridley (eds.) Arguing about Art: Contemporary Philosophical Debates, 2nd ed., London and New York: Routledge, 195–214. Sparshott, F. E. (1979) “Basic Film Aesthetics,” in G. Mast and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 2nd ed., New York: Oxford University Press, 321–44. Wollen, P. (1969) Signs and Meaning in the Cinema, London: Martin Secker and Warburg.
63
TF17557.indb 63
16/9/08 15:47:19
6
D E P IC T ION Robert Hopkins Depiction is the kind of representation that pictures – ordinary, “still” pictures – display. A picture depicts whatever it is a picture of. There are many kinds of representation: along with listing pictures, we might include words, theatrical performances, hand gestures, or road signs. Not all of these represent in the same way. Depiction is the way of representing that distinguishes pictures from other representations. That’s not to say that every picture depicts. Some purely abstract pictures don’t represent anything at all. Nor is it to say that a picture depicts everything it represents. A painting might depict a woman who in turn symbolizes the sufferings of Russia. The picture in some way represents Russia’s suffering, but it does not depict it. Nonetheless, every representational picture depicts at least some of what it represents. That’s what it is to be a picture of something. We describe cinema films as “moving pictures,” and rightly so. Films represent, and films involve pictures. It’s likely, then, that they represent depictively, that they depict at least some of what they are about. But what is depiction? Can we give an account that leaves room for the idea that cinema depicts? Can we say what is special about depiction in film, as opposed to still pictures? Can a single account apply equally to all film – traditional fiction film, documentary, old-style animation, and computergenerated imagery (CGI) – or do the diverse ways in which movies are made prevent that? Finally, how much of what a film represents is depicted?
Moving and still depictions Freeze a film at any moment, and the result is a picture of whatever that sequence in the film represents. True, if the sequence involves considerable movement, the picture might be blurred. True too, not everything the sequence represents might appear in the picture – what the flung fist is aimed at, or why, might be lost. But that we can generate pictures by freezing cinema images, and that the resulting pictures represent much of what the unfrozen sequences do, strongly suggests that films are themselves pictures, that they depict at least some of what they represent. However, before we can be certain of this, we need to say something about the obvious difference between ordinary pictures and film – the former are “still,” the latter “move.” Can we give an account of this difference that allows both to depict?
TF17557.indb 64
16/9/08 15:47:19
DEPIC T ION
We might think that what distinguishes moving from still pictures is their content: only the former represent time. That would be a mistake. A painting of an execution might show the moment before the guillotine drops, and an illustration on a safety card might show the process by which the plane is to be evacuated. Since moments are points in time, and processes occupy periods of it, both pictures represent time. Yet neither counts as moving. A second suggestion would be that, while moving pictures are themselves spread across time, still pictures are not. Only in the latter are all the parts of the picture present at once. Only for the former does it make sense to talk of the picture having a beginning, middle, and end. However, this proposal is also flawed. Most pictures don’t change interestingly over time. But consider a picture that does: a neon sign that lights up bit by bit, to reveal a cowboy on a horse. There’s a perfectly good sense in which this picture is not all present at the beginning of the process, and we can sensibly talk of the beginning, middle, and end of its appearing. Nonetheless, all we have here is a still picture, revealed bit by bit. True, that the process of revealing the picture has a beginning, middle, and end does not show that the picture itself does, but then we could raise that quibble for the case of film. True too, the natural way to take the example does involve the whole picture being present at once at the end of the process, if not before. However, we can imagine the example tweaked so that this is not so. Perhaps earlier bits switch off as later ones light up. Rather than turning on whether they have temporal content, or whether they extend over time, the distinction between still and moving pictures is instead a matter of the relation between properties of the representation and its content. Any representation will have properties that determine what it represents. In the case of pictures, those properties include the distribution of paint on a canvas or that of light projected onto a screen. These properties can change over time, or remain stable. The paint can fade, but may well not do so. The pattern projected might alter, but it might not (think of a long shot of an unchanging scene). The difference between still and moving pictures is that only in the latter does development (stability or change) in properties of the representation determine development (stability or change) in properties represented. How the picture develops over time dictates how its objects are represented as developing. Thus the neon sign does not count as a moving picture because, although it changes over time, that does not dictate that the cowboy does. In contrast, even an exceptionally stable movie, such as Warhol’s film of John Giorno asleep, does count as moving because, if nothing on the screen changes for ten minutes, nothing in the scene represented changes for ten minutes either. This is not to say that a neon sign couldn’t count as a moving picture. We can easily alter the example so it does. If two differently positioned neon tubes represent the cowboy’s rope, one lighting up as the other fades, the result might be a moving picture of the rope rising and falling. Nor is it to say that moving pictures always involve the straightforward relations between representation and content that we find in the Warhol. Flashbacks, flashforwards and the like show that the time of the image and that of the world in the film can be much more obliquely related than that. In these cases, how the images evolve still determines how the world represented does, just not in the most straightforward way. 65
TF17557.indb 65
16/9/08 15:47:19
ROBE RT HOPK I NS
If this is the difference between moving and still pictures, we can indeed allow that both depict. The difference turns on something purely structural: the way certain properties of the representation relate to certain properties represented. Why should depiction not provide the way to fill out that structure? So film usually depicts. But must it? Can there be cinematic representations that fail to represent in the way still pictures do? There are two possible sources of nondepictive cinema. One source is film that doesn’t represent at all. In a famous sequence toward the end of 2001: A Space Odyssey (1968) the screen is filled for long minutes with dazzling but indecipherable patterns of colored light, accompanied by a soundtrack of pure noise. Perhaps these patterns depict the strange scenes the remaining astronaut witnesses in his last journey. Even if so, they might have been used otherwise, to create a sort of symphony in light. Some filmmakers have explored such extreme abstraction. The other source is film that represents without pictures, such as Derek Jarman’s Blue (1993), for the entire duration of which the screen is a radiant blue. Although Jarman perhaps intended the undifferentiated image to reflect his own experience as he lost his sight, it is not clear that it depicts the world as experienced by someone going blind. Despite these examples, it is clear that almost all film depicts. Cases such as Blue raise the question where cinema stops and other forms of representation, such as Radio Theater, begin. Purely abstract cinema, if there is such a thing, lies outside the realm of representation altogether. Where film does represent, and does so in ways that are clearly cinematic, it seems always to do so at least in part by depicting things. And, while we might wonder whether a work without pictures could count as film, we’d have no doubts over a work composed entirely of images. The earliest movies were just that.
What is depiction? Although confined almost exclusively to the discussion of still pictures, there has been considerable debate over the nature of depiction. Here there is space to describe only some of the available positions (for more, see Hopkins 1998: chapters 1 and 2). Some (Goodman 1969; see also Kulvicki 2006) say that depiction is symbolic. It is just as much a matter of convention what a picture depicts as what a word means. In consequence, there can be different systems of picturing, just as there can be different languages. The obvious difference between systems that count as pictorial and those that count as linguistic reduces to formal properties of the systems involved. Our task as theorists of depiction is then to identify these formal differences. This view is likely to have appeal for those who take seriously the idea of a language of cinema. Such thinkers need not locate that language in cinema’s capacity to depict. Perhaps cinema represents in various ways, and perhaps it is the nondepictive ways that are language-like (Metz 1985). But since depiction has some claim to be fundamental to cinema, if it were conventional, all cinema would bear important similarities to language (Eco 1985). The central challenge for the symbolic view is to explain the sense in which cinema and the other kinds of picturing count as visual. This cannot reduce to the fact that 66
TF17557.indb 66
16/9/08 15:47:19
DEPIC T ION
we take in pictures by looking at them: the same is true of written language. Nor can it lie in whatever formal features are supposed to distinguish depictive symbols from those of other kinds. There seem to be important parallels between the way things are presented in pictures and the way they are presented in vision. Not only can the same sorts of property be represented in each, but space is given in pictures in ways strikingly similar to the way it is given by sight (Hopkins 1998, 2004). Yet there is nothing conventional (or language-like) about the way vision represents the world. Some will deny this, claiming that the culture one is raised in determines the nature of one’s visual experience. But even if this is true, it is not enough to render vision conventional. For a practice to be conventional is (i) for it to be one of two or more equally good solutions to a problem, all of which are available to a given community; and (ii) which is adopted by each member of that community in part because it is adopted by the others, it being common knowledge that this is so (Lewis 1969). Even if (i) is true of a particular way of seeing the world, (ii) is not. Thus picturing bears striking parallels to seeing, and seeing is not conventional. This strongly suggests that depiction is not conventional, at least not all the way to the core. What, then, is the relation between picturing and seeing that supports these parallels? One answer (Schier 1986; Lopes 1996) appeals to the perceptual processing that pictures trigger. The same neuropsychological mechanisms that are involved in visually recognizing, say, a dog, are involved in understanding a picture of one. Since the parallel claim is not plausible for the (written) word “dog,” or for a dog description, pictures are visual in a sense in which words are not. Understanding them requires that we deploy the same mechanisms responsible for face-to-face recognition of the things they represent. This could not be the whole story. We don’t, except in very unusual circumstances, mistake pictures of dogs for dogs. So there must be more going on when we see dog pictures and grasp what they depict than simply the engagement of the mechanisms by which, in general, we recognize dogs. The recognition theorist will say that pictures not only engage our capacity to recognize objects such as dogs, they also engage our capacity to recognize flat surfaces. Thus our perceptual system delivers conflicting results about what is before us – that it’s a dog, and that it’s a flat surface. The conflict is overcome by judging that what we’re seeing is not a dog, but a picture of one. How, though, does this reconciliation figure in conscious experience? It’s not as if we’re presented with something contradictory, something that both looks to be a dog and looks to be a flat surface, and then just form the belief that it’s a picture. Dog pictures don’t look contradictory any more than do other representations, such as dog descriptions. True, in the former case there are, in a sense, two aspects to what we see before us. True, too, that is not so in the same way when we read the description of a dog. In the picture case, a dog is somehow seen; it is present to us in experience, as it is not in the case of words. But though our experience of pictures involves these two aspects, it integrates them in a harmonious way. How does it do so? What is our experience of pictures like? If we can answer, we might hope to define depiction as representation that gives rise to that experience. We would then not need the claims about perceptual processing that constitute the recognition theory. 67
TF17557.indb 67
16/9/08 15:47:19
ROBE RT HOPK I NS
Although various thinkers agree that experience holds the key to depiction, they disagree about how that experience should be described. Following Richard Wollheim (1987) let’s call that experience “seeing-in.” We see things in variegated surfaces, and it is in virtue of this that those surfaces depict those things. Disagreements aside, there are two claims all those who back an experiential theory should accept. First, seeing-in cannot alone define depiction. We see things in surfaces that don’t depict at all, such as clouds and flames. And we sometimes see in pictures things they do not depict, as when we see the face of a friend in a portrait painted long before she was born. At most, then, depiction is seeing-in plus something else. The extra is generally agreed to lie in the surface’s history. Perhaps someone marked the surface intending that person or thing to be visible in it. Perhaps the surface was the product of a mechanism, such as a camera, designed to produce surfaces in which whatever provides the initial input can be seen. Either way, a “standard of correctness” has been set – something making it right to see one thing, rather than another, in the surface. Depiction is seeing-in governed by some such standard of correctness (Wollheim 1987: chapter 2). Second, seeing-in is a special form of seeing. If we don’t, in a perfectly ordinary sense, see the marks before us, then we’re not seeing things in them, but hallucinating. Equally, however, seeing-in can’t simply reduce to seeing the marks. For we see the marks even when seeing them only as blotches on a surface, and that is precisely to fail to see something in them. So, while seeing-in involves seeing the marks, it must also involve something else. The questions are what this extra is, and how it combines with seeing the marks to form a coherent experience. Answers to these questions amount to a characterization of seeing-in. Wollheim’s own characterization hardly adds to the sketch I used to introduce the idea. According to him, just as there are two aspects to the pictures we see, so there are two aspects to our experience of them. One aspect somehow presents us with a marked surface, the other with something else, such as a dog. To see a dog in the marks just is to have an experience with these two aspects. Nothing more informative about it can be said. Dissatisfied with Wollheim’s caution, others have tried to say more. Some have claimed that seeing-in involves the interpenetration of ordinary seeing by a special kind of imagining (Walton 1990). Others say it is an experience of resemblance. When we see a dog in the marks, we see the marks as resembling a dog in some respect. There is then disagreement about what that respect might be (Peacocke 1987; Budd 1993; Hopkins 1998). Any of the various accounts of depiction in still pictures will transfer, without further strain, to film. Anyone wanting to define depiction in cinema will therefore have to engage with the debate about depiction in general. Let us not pursue further which view is to be preferred. I am simply going to assume that some form of experiential account is correct. Doing so offers the easiest, and in some cases the only, way to pursue some further issues concerning depiction in film.
68
TF17557.indb 68
16/9/08 15:47:19
DEPIC T ION
Seeing film as pictures One striking kind of picture is trompe l’œil, pictures so lifelike that we mistake them for the objects they depict. We fail to see these images as pictures. Usually this occurs only under special viewing conditions. A flat ceiling might be painted so as to appear made of ornately formed plasterwork. The fact that it is impossible to get closer to the detail helps sustain the illusion. Many realistic pictures are seen in conditions that make it impossible for them to fool the eye in this way. We see the craquelure, the ink strokes, or the light glinting off the photographic paper. Many other pictures could hardly sustain illusion under any conditions – think of stick figures, cubist painting, and caricature. Nonetheless, while among still pictures it is rare for the eye to be fooled, it does sometimes occur. Some pictures, then, are illusionistic: the experience we have before them (under the right conditions) matches that we would have before what they depict. What of cinema pictures? Some (Allen [1995] is a recent, sophisticated example) claim that they too sustain illusion. Of course, almost any film viewer knows not to take what is before her as real. She doesn’t believe she’s seeing the events shown. But, according to this view, she nonetheless has illusory experience of the film. Her visual experience when watching matches that she would have before those events. There is illusion in experience here, if not in belief. No doubt cinematic pictures could be shown in conditions which would enable them to fool the eye. But in the conditions under which they are usually viewed, we do not have illusory experience of them. As we sit in the cinema, we are fully aware, visually as well as in belief, that what is before us is an illuminated screen. In part, this is a matter of context: we see the dark room that contains the screen. But in part it is simply a matter of the appearance of the projected image itself. For all that cinema pictures lack the obvious textures of canvas, paper, crayon, or paint, they are composed of lights that vary in intensity far less than does the light reflecting off most objects viewed in the flesh. Often the scale of the images does not match that of things seen in real life. And changes in apparent size, as the camera moves, are not usually accompanied, as they would be in reality, by any sense of our own movement. In these and other ways, cinematic images make plain that we are looking at pictures, not things. If moving pictures differ from still ones in this respect, it is only by degree. With still pictures, we are often aware both of the features, such as what colored paint lies where, that fix what they represent and of those that don’t, such as the grain of the canvas (Lopes 2005: chapter 1). Perhaps cinema images differ in that we find it hard to see the content-fixing features. We are frequently aware of them only by seeing in them the content they fix, and see what is before us as a picture only by seeing other features, such as the shape and size of the screen. Or perhaps we can after all see what content-fixing features there are, but a far narrower range of features play that role, little more than what color is projected where. However that may be, that we do see film images as pictures seems beyond doubt.
69
TF17557.indb 69
16/9/08 15:47:19
ROBE RT HOPK I NS
What do we see in film? So we experience films as pictures. But what, exactly, do we see in those pictures? In particular, does what we see in a film depend on how it was made? There are various means by which films are made. We might, as in documentary, make a photographic record of real events and use the film to inform the audience about them. We might, as in traditional animation, photograph hand-drawn pictures. We might avoid photography altogether, using CGI techniques to create pictures directly. (Even where CGI does begin with photography, many aspects of what the finished picture shows were not present in the original scene.) But historically the vast majority of films have been made another way. Actors, with the aid of sets, props, and perhaps live special effects, act out the events the film seeks to narrate. (Call those events the story told.) The complex events involving the actors, sets, props and the like are themselves recorded photographically. (Call those the events filmed.) Since photographic recording and acting (etc.) are both forms of representation, the film, or each of the sequences of which it is made, thus begins life as the representation of a representation. The second kind of representation here clearly bears strong points of contact with that found in theater. So call it theatrical representation. (That is not to assume that there aren’t important differences between the two.) Call movies made in this way two-tier films. What do we see in two-tier film? Often our experience reflects the film’s making. We see in the images before us the events filmed, representing the story told. For instance, in The Searchers we see John Wayne on a horse, representing a cavalryman, Ethan Edwards, searching for his niece. That we sometimes see films in this tiered way is shown by our ability to focus on the quality of the acting. Unless we see the actor in the picture before us, we can have no sense of his contribution. But unless we see him as representing such and such a person, undergoing certain travails, we can have no idea of what his contribution is intended to achieve. We often see films in this tiered way, but do we always do so? The alternative is that our experience does not acknowledge the different tiers of representation involved in the making of the film. The level of events filmed drops out, and all we see in the film is the story told. Do we, in fact, ever undergo this “collapsed” form of seeing-in? Perhaps whether we do in part depends on who “we” are. The film critic may spend her time pondering how the camera has been used, and the quality of the acting. She is, perhaps, professionally obliged to see films in the tiered way. The question, however, is whether any of us ever do anything else. (And note that the question concerns our actual experience of film. It is obvious that some subject could see in two-tier film nothing more than the story told. Just consider showing a two-tier film to someone brought up in a culture where all film is very realistic CGI.) Now, not only is much film two tier, but the lower tier, the theatrical representation, is often illusionistic. The story told is represented by the events filmed in an illusionistic way: someone watching the events filmed from the camera’s viewpoint would have visual experiences matching those she would have before the events in the story themselves. Perhaps illusion, so defined, is more aspiration than reality for many 70
TF17557.indb 70
16/9/08 15:47:19
DEPIC T ION
films. Certainly it is not always even aspiration: some films refuse to attempt illusion at this level. But that it is often at least aspired to is the only way to make sense of the tremendous care usually taken to get the sets and costumes right, to act convincingly, and to ensure that items of studio equipment, such as sound booms, are not visible in the filmed scene. Illusionistic representations, be they pictorial or theatrical, are hard to see for what they are. Looking at them, we seem to see what they represent. We don’t, therefore, seem to see a representation of that thing. Moreover, this fact about illusionistic representations carries over to pictures of them. Consider again our trompe l’œil painting of ornate plasterwork. Suppose we create an accurate picture of this illusionistic picture, e.g., by photographing it. What will we see? There’s no reason to think the photograph is illusionistic, so presumably we will see it as a picture in which we see something else. But will we see in the photograph another picture, in which in turn we see plasterwork? If we approach the photograph knowing what it shows, perhaps we will. However, our seeing-in might just as easily collapse: we might simply see ornate plasterwork in the photograph. So, where picture 1 accurately depicts picture 2, and picture 2 is an illusionistic depiction of object O, not only do we seem to see O (rather than a picture of one) when looking at picture 2, but we seem to see a picture of O (and not a picture of a picture of O) on looking at picture 1. Nothing here turns on the fact that the illusion is pictorial. We would expect just the same in the case of an accurate depiction of an illusionistic theatrical representation. But two-tier film is (often) just that. It is made by creating a photographic record of the events filmed. Photographic pictures have a strong tendency to accuracy (the camera does not lie). But what is recorded in two-tier film is (often) a theatrical representation that is illusionistic. So we should expect to see in two-tier film only the story it tells. Our experience of two-tier film does indeed collapse. Note two things. First, the claim is not that we cannot see two-tier films any other way. Of course we can, and often do. All we need do is bring to bear our knowledge of how such films are made. We attend to what is before us as the photographic representation of a series of events that themselves represent. For instance, where a moment before we saw only Ethan’s remorse, now we see how Wayne uses his eyes to convey that emotion. We can slip between tiered and collapsed seeing-in, depending on where we direct our attention. Second, I have not claimed that the images composing two-tier film are illusionistic. The illusionism is at the level of the theatrical telling of the story, not that of the depiction of that telling. So nothing here contradicts the claim of the last section, that we do not have illusionistic experience of film. Indeed, there is a sense in which illusion in two-tier film is not a feature of our experience of it at all. When seeing-in collapses, we are blind to a feature of the film’s making, i.e., that it was made by photographing a theatrical representation of the story told. But our experience does not misrepresent the facts. We don’t see the picture as made in some way other than it was. Collapsed seeing-in need not suggest anything about how the movie was made. Thus, while illusion in the film’s making leads seeing-in to collapse, collapsed seeing-in is not itself illusory. 71
TF17557.indb 71
16/9/08 15:47:20
ROBE RT HOPK I NS
The unity of cinematic representation Why does it matter whether seeing-in collapses? It matters because it bears on one final issue: how far we can hope to tell a unified story about representation in film. (In fact, I think it matters in other ways too. See Hopkins 2008.) The unity of cinematic representation might fail in two ways. It could be that single films represent in more than one way. Or it could be that different films, whether they represent in one way or many, represent in different ways from each other. The fact that some, but only some, films are two tier makes both threats real. The natural thing to say about two-tier films is that they display two kinds of representation. They depict, since they are a photographic record of the events filmed. But the events filmed represent the story told in another way, one we dubbed “theatrical.” As a result, the film itself can only, it seems, represent the story told in dog-legged fashion, by depicting events that themselves theatrically represent that story. But if this is true of some film, it is hardly true of all. Traditional cartoons are made by photographing pictures, not theatrical representations; documentaries are made by photographing the very events the film narrates; and CGI offers a way to make pictures without photographing anything at all. If the making of a film dictates how it represents, films made differently must represent differently. If seeing-in collapses, however, both threats can be avoided. The idea is that even in two-tier film all we see, at least often, is the story told. If so, we experience these films in the same way we experience others. In traditional cartoons all we see is the story they tell: the sufferings of various comic characters. The same is true of CGI. We see, say, massed armies, and do not even attempt to get our experience to reflect the process by which the image was generated. Finally, consider documentary. Here, at least sometimes, we take what we see in the image really to have occurred. But, despite that difference from the other cases, we still see in the film only the events it narrates. We see, for instance in the famous sequence at the start of Triumph of the Will (1935), Hitler’s plane landing in Nuremberg. Riefenstahl filmed Hitler’s plane, not a representation of it, and that is what we see in her movie. Where experience is unified across different types of film, we can hope that representation will be. Just as depiction by still pictures is a matter of generating a certain experience, seeing-in, so is depiction by film. Just as still pictures depict what we see in them, so do films. And what is seen in all films, regardless of how they are made, is simply the story they tell – be it pure fiction, fact told by means of actors, or fact told by recording the events themselves. Cinematic representation is thus unified, and, for all we’ve said so far, is nothing but depiction. I close with four comments on this proposal. First, cinematic representation may just be depiction, but that is not to say that there are no differences between it and depiction by still pictures. We’ve already seen one, in our discussion of the distinction between moving and still images. We are now close to another. The proposal is that cinematic representation is something like telling a story in pictures. It is a way of narrating a story by showing the viewer pictures in which she sees (nothing but) the events that compose the tale. Individual still pictures sometimes tell a story, but 72
TF17557.indb 72
16/9/08 15:47:20
DEPIC T ION
it’s a different challenge to do so in moving images. Depiction in cinema might be distinctive in the narrative structures it uses to accomplish this. Examples would be linear and other forms of storytelling, flashbacks, jump cuts, and reverse angles. Second, I don’t deny that we can construe two-tier film as involving two kinds of representation, and therefore as involving different forms of representation from other film. The facts about how two-tier film is made, and indeed our ability to see it in the light of those facts (i.e., to undergo “tiered” seeing-in), make that only too plausible. What I’ve done is to articulate a theoretical perspective on cinema which makes sense of it as displaying just one kind of representation. From another perspective, it displays several. We should not think that only one of these perspectives offers us the truth. Both are right. Third, in developing that perspective, I relied on the idea that much two-tier film is illusionistic (at the theatrical tier). But I conceded that not all is. Some film refuses to aspire to illusion. Think of films where the acting is deliberately mannered, or where there are constant references to the fact that this is merely acting, and the whole is merely a film (e.g., the Austin Powers series). Some film falls short of illusion through incompetence – through bad acting, poor sets, or accidentally leaving sound booms in view. And some falls short of it, however illusionistic it otherwise is, by using actors so familiar that we struggle not to see them as the celebrities they are, rather than the characters they convey (see Cavell [1979] on the “persona”). In all these cases, lack of illusion in theatrical representation of the story told makes it hard to see that story in the picture, rather than an acting-out of it. If the theoretical perspective I’ve offered is to apply to all cinema, something needs to be said about these cases. Fourth, and finally, I am not saying categorically that all representation by cinema can be brought under one heading. There are other sources of cinematic content that I have not considered. Language is the most obvious, whether it be spoken by the protagonists, added by a narrator, or simply used in the titles that sometimes introduce scenes and convey background. But the narrative structures mentioned above might also complicate matters. Even the use of music in film might sometimes bear on what it represents. Considering these sources of cinematic meaning might lead us to conclude that cinema does indeed involve various kinds of representation. My point here is limited to that part of cinematic meaning that derives from pictures. That those pictures are made in various ways does not, in itself, show that cinematic representation has several forms, rather than one. See also Digital cinema (Chapter 7), Music (Chapter 17), Narrative closure (Chapter 19), and Documentary (Chapter 45).
References Allen, R. (1995) Projecting Illusion: Film Spectatorship and the Impression of Reality, Cambridge and New York: Cambridge University Press. Budd, M. (1993) “How Pictures Look,” in D. Knowles and J. Skorupski (eds.) Virtue and Taste: Essays on Politics, Ethics and Aesthetics, Oxford, UK; and Cambridge, MA: Blackwell.
73
TF17557.indb 73
16/9/08 15:47:20
ROBE RT HOPK I NS
Cavell, S. (1979) The World Viewed: Reflections on the Ontology of Film, exp. ed., Cambridge, MA: Harvard University Press. Eco, U. (1985) “Articulations of the Cinematic Code,” in B. Nichols (ed.) Movies and Methods: An Anthology, vol. 1, Berkeley and Los Angeles: University of California Press. Goodman, N. (1969) Languages of Art, 2nd ed., Oxford: Oxford University Press. Hopkins, R. (1998) Picture, Image and Experience: A Philosophical Inquiry, Cambridge and New York: Cambridge University Press. —— (2004) “Painting, Sculpture, Sight and Touch,” British Journal of Aesthetics 44(2): 149–66. —— (2008) “What Do We See in Film?,” Journal of Aesthetics and Art Criticism 66(2): 159–69. Kulvicki, J. (2006) On Images: Their Structure and Content, Oxford and New York: Clarendon. Lewis, D. (1969) Convention: A Philosophical Study, Oxford: Basil Blackwell. Lopes, D. (1996) Understanding Pictures, Oxford: Clarendon; New York: Oxford University Press. —— (2005) Sight and Sensibility: Evaluating Pictures, Oxford and New York: Oxford University Press. Metz, C. (1985) “On the Notion of a Cinematographic Language,” in B. Nichols (ed.) Movies and Methods: An Anthology, vol. 1, Berkeley and Los Angeles: University of California Press. Peacocke, C. (1987) “Depiction,” Philosophical Review 96: 383–410. Schier, F. (1986) Deeper into Pictures, Cambridge and New York: Cambridge University Press. Walton, K. (1990) Mimesis as Make-Believe: On the Foundations of the Representational Arts, Cambridge, MA: Harvard University Press. Wollheim, R. (1987) Painting as an Art, London: Thames & Hudson.
74
TF17557.indb 74
16/9/08 15:47:20
7
D IGITAL C INEM A Berys Gaut A brief history We should distinguish between full and partial digital cinema. The cinematic process, from planning to screening, can be divided into five phases: pre-production, production, post-production, distribution, and exhibition. Partial digital cinema is digital at one or more of these phases, but not at all of them; full digital cinema is digital at all of these phases. Digital cinema began in the post-production arena, where films were digitally processed and manipulated, usually as a way of creating special effects or animation. One of the earliest partially digital films was TRON (1982), which employed digital sequences to simulate a kind of computer game. Jurassic Park (1993) and Forrest Gump (1994) brought digital special effects to popular awareness, and George Lucas’ Industrial Light and Magic studio created the effects in these and many other pictures. Digital color grading (allowing manipulation of film color by digital means) was widely employed from the late 1990s, in preference to chemical grading, and digital editing gradually became standard. Toy Story (1995) was the first digital-animated feature film, created by Pixar Studios (originally Lucasfilm Computer Development Division). In the production process, some of the first widely distributed films to be shot with digital cameras were the two earliest Dogme 95 films, The Idiots (1995) and The Celebration (1995), and many independent filmmakers adopted digital shooting thereafter. In distribution and exhibition, 1999 saw both four widely publicized digital screenings of Lucas’ The Phantom Menace (though the film was shot on conventional film, and later digitized), and the Sundance Film Festival’s acquisition of a digital projector. (Before this point, digitally produced films were transferred by laser recording back onto conventional celluloid for commercial screenings.) And in pre-production, animatics (animated storyboards) were increasingly employed in the 1990s. By the late 1990s there were experimental fully digital films, and the first commercial fully digital films appeared in the new century: Mike Figgis’ Timecode (2000) was one of the earliest, and George Lucas’ Attack of the Clones (2002), shot on digital video, brought fully digital cinema to popular attention. So fully digital cinema, involving all parts of the filmmaking and distribution processes, is less than ten years old. But it is likely to become the dominant form
TF17557.indb 75
16/9/08 15:47:20
BE RYS GAU T
of cinema within the next two decades, given its exclusive adoption by influential filmmakers such as George Lucas and Robert Rodriguez, and given the fact that it is cheaper to produce and distribute than conventional film. It is also likely to accelerate the dispersion of filmmaking practice away from Hollywood, and to lead to an increasing democratization of the filmmaking and film distribution process, particularly through the Internet.
What is digital cinema? So what precisely is this new type of cinema? Cinema is the medium of moving images. Moving images have been constructed by several methods over the course of history. The oldest method employs objects to cast shadows onto a backdrop, as in traditional Javanese shadow plays. Handmade images have also been used, as in Émile Reynaud’s films, regularly screened from 1892 in Paris; these were composed of hand-painted images, projected onto a screen. Most familiar are moving photochemical images, traditional photographs, pioneered by several figures in the late 1880s and 1890s. More recently, video images have been used, constituting electronic cinema. Video can employ analogue or digital images. An analogue image is one that is completely specifiable only by continuously varying values. Object-generated, handmade, photochemical images and predigital video images are all analogue. In contrast, a digital image is one that is composed of discrete values, typically integers (whole numbers). The digital image, when it is a mechanical recording of some object, is produced by digitizing a continuously varying value, such as light waves emanating from objects: the waves are sampled thousands of times a second, and the amplitude of the wave at each sample point is recorded as an integer. These integers are stored as a bitmap. A bitmap is composed of a grid of picture elements (pixels), and each point of the grid has an integer assigned to it that is generated by sampling the light emanating from the part of the object that the pixel represents. These integers are stored as binary digits (bits), i.e., as integers to base 2. At each point of the pixel grid, one out of millions of integers, representing colors, is stored: a common digital standard is 24-bit color; i.e., any one of 224 (more than 16.7 million) colors can be represented at any point in the pixel grid. Digital cinema is the medium of moving images generated by bitmaps. It follows from this account that digital images can be directly subject to computational manipulation: computers are designed to process binary digits and hence can process bitmaps. Analogue images, in contrast, can be subject to computational manipulation only after they have been digitized. It follows from their computability that digital images can be easily and infinitely manipulated without degradation: for any numerical array, there is an algorithm which can transform it into any other numerical array, and information stored as numbers suffers no degradation in being reproduced. In contrast, analogue images, such as traditional photographs, are relatively difficult and cumbersome to manipulate, and their reproduction, involving an analogue copy of an analogue original, leads to generational loss: a photograph of a photograph has less information than does the original photograph. And whereas there is a fixed amount 76
TF17557.indb 76
16/9/08 15:47:20
DIGITA L CI N EM A
of information in a digital image (given by its bitmap), this is not true for the analogue image, since there is a potentially infinite amount of information in a continuously varying value. Finally, given the possibility of direct computer processing in real time, digital cinema can be interactive, whereas traditional photochemical film cannot. (For useful accounts of some of the technical aspects of digital cinema and imagery, see McKernan [2005] and Monaco [2000: chapter 7]; and for digital animation, see Kerlow [2004].) So, digital cinema is radically distinct from traditional analogue, photographic cinema. A number of philosophical and theoretical questions can be raised about these differences. I will consider four: whether digital cinema constitutes a new medium; the nature and implications of interactivity; the kind of realism available in digital cinema; and the possibility of single authorship in digital cinema.
A new medium? Is digital cinema a new medium, distinct from traditional cinema? Is it even a medium? Timothy Binkley argues that it is not a medium at all, since the digital image is an abstract object, a bitmap, and media are physical – the medium of painting is paint, for instance. In contrast, “The computer is not a medium” (Binkley 1997: 114). However, Binkley’s assumption about media always being physical (what we might call their material) is mistaken: literature is a medium, but it is composed of semantic structures, not physical ones. Rather, a medium is a set of practices for using material, whether symbolic or physical, so digital imagery can constitute a medium (Lopes 2004: 110). One might suppose that digital cinema cannot be a new medium, since cinema is the medium, and digital cinema is only one kind of cinema; and media cannot contain other media. But that assumption is false: we can properly talk of the medium of printing, but also of the media of woodcuts, engraving, etching, lithography, and so on, which are all types of printing. So, one medium can contain other media: media can, as I will say, nest. And given the distinctiveness of digital imagery, which we noted in the previous section, there is good reason to believe that it constitutes a new medium, but one which nests in the wider medium of cinema, along with other media such as shadow plays, handmade cinema (e.g., Reynaud’s cinema), traditional photochemical films, and analogue electronic cinema. How distinct is this medium? According to W. J. T. Mitchell, the digital image is a radically distinct type of image. He holds that talk of digital photographs is only metaphorical: “Although a digital image may look just like a photograph when it is published in a newspaper, it actually differs as profoundly from a traditional photograph as does a photograph from a painting” (Mitchell 1992: 4). To support this claim, he appeals to various features of the digital image that we noted earlier, such as the existence of a fixed amount of information and the ease and speed with which it can be manipulated. He also notes that “A digital image may be part scanned photograph, part computer-synthesized shaded perspective, and part electronic ‘painting’ – all smoothly melded into an apparently coherent whole” (Mitchell 1992: 7). 77
TF17557.indb 77
16/9/08 15:47:20
BE RYS GAU T
Mitchell’s supporting reasons are true and important. One can generate digital images by three main methods. First, one may employ mechanical capture techniques: these include not just (what we ordinarily call) photography, but also two and threedimensional scanning techniques, and techniques such as motion capture, which record a performer’s body and/or facial motions. Second, one can handcraft the image, using a graphics editing tool, for instance; one can digitally “paint” an image, using Adobe Photoshop or Corel Painter. Third, one can have a computer synthesize an image by running a set of algorithms: for instance, rendering is the process of a computer generating an image by running algorithms specifying perspectival and shading relations on a 3D computational model of an object (and that model can in turn be produced by capture, hand construction, or synthesizing techniques). And all three techniques can, as Mitchell maintains, be combined seamlessly. For instance, the character of Gollum in The Lord of the Rings trilogy (2001, 2002, 2003) was generated by motion-capturing the body movements of Andy Serkis and editing them; animators then used traditional key-frame animation techniques to “paint” an image on top of these recorded movements, and a computer rendered the three-dimensional virtual model into a series of two-dimensional views, which were themselves then further manually adjusted (Jackson 2004). These techniques were pushed further in King Kong (2005), where motion capture was used on Serkis’ face as well as his body (Jackson 2006). The resulting images do, indeed, come across largely as seamless, in the sense that the viewer can see no trace of the various techniques that were used in their making. So the digital image can be produced by three kinds of techniques (capture, computer-synthesis, and painting) in any proportion, and with a seamlessness to the image that does not betray its origins. Given this mixture of techniques, we can refer to the digital image as a mélange (or blended) image – that is, it can be produced by any of three distinct techniques, and each technique may vary in its importance in the making of a particular image. The fact that the digital image is a mélange image is an important one: it shows that digital images are very different from traditional photographs and supports the claim that digital cinema is a radically new medium. However, Mitchell is incorrect in saying that it follows from these facts that digital photographs do not, strictly speaking, exist. For one option with a mélange image is to use only one of these generation techniques, rather than using all three, and this one might be the capture technique, that is, in one of its varieties, very similar to that used by a traditional photograph. The crucial difference between a digital camera and a traditional camera is the replacement of a photochemical film with an electronic sensor (a chargecoupled device); the lenses, optical systems, shutter mechanisms, and so on, can be identical. So we can still allow that digital photographs exist, while holding that the digital image is very different from the analogue image, in part because it can be generated by nonphotographic means. Finally, one might suppose that, though digital cinema is a new medium, it does not follow that it is a new artistic medium. The telephone was a new medium of communication, but it was not an artistic medium. What more is required for a medium to be an artistic one, that is, to form the basis for an art form? Plausibly this – that one can 78
TF17557.indb 78
16/9/08 15:47:20
DIGITA L CI N EM A
create artistic effects in the medium that are either impossible or prohibitively difficult in other media. On this criterion, digital cinema counts as a new artistic medium. For consider the ideal of photoreal animation. Photoreal animation is animation where the animation image of something is visually indiscernible from how a photograph of that thing would look, if it existed with the properties that the animation image ascribes to it. For instance, the animation image of King Kong looks visually indiscernible from how a photograph of King Kong would look, if he existed with the properties that the animation image ascribes to him. In contrast, a traditionally drawn cartoon of King Kong can easily be visually discriminated from a photograph of what he would look like, if he existed. Much digital animation is now driven by the photoreal ideal, and it has been successfully realized several times. Photoreal animation is an artistic achievement, affecting how one responds to the image in artistically relevant ways (for instance, the greater realism of the animation helps promote empathy with Kong and engagement with his plight). And photoreal animation is all but impossible in traditional animation: imagine the superhuman difficulty of drawing frames of Kong so that they look like photographs of him. So there is at least one aesthetically relevant effect that is distinctive of the medium of digital cinema. There are many more, most strikingly those arising from interactivity; but they occur also in the realm of special effects, where these do not depend on animation. It follows, then, that digital cinema is not only a new medium, but also a new artistic medium.
Interactivity Since computers can process information in real time, digital cinema can be interactive – and much of it is. The latter claim may seem surprising, but recall that cinema is the medium of moving images; so videogames, which are composed of digital images, are a part of digital cinema on this definition, as are massively multiplayer online role-playing games, such as World of Warcraft. There are also virtual worlds that support moving images, such as Second Life; live-action interactive movies, such as I’m Your Man (1992); and digital interactive dramas, such as Façade (2005). The history of digital cinema recounted in the first section concerned only noninteractive digital cinema; interestingly, the history of interactive cinema is older: the first videogame was produced in 1962 or arguably 1958 (Poole 2004: 15), though the quality of early videogame graphics was far inferior to that of noninteractive digital cinema. Little work of philosophical distinction has been produced on interactivity; but of that which has been undertaken, one central topic concerns the issue of what is meant by “interactivity,” and another concerns the implications of interactivity for our engagement with works. One mooted distinction is between weakly interactive and strongly interactive works. In the case of a DVD with a chapter selection function, the viewer can choose which parts of the film to access and in which order. This gives her more control over the order in which she views the film than the viewer in a movie theater has. However, the structure of the work still exists independently of the viewer’s activities: such works are weakly interactive. But this kind of interactivity is the same as that provided by a table of contents and the index of a book, so it does 79
TF17557.indb 79
16/9/08 15:47:20
BE RYS GAU T
not represent a radical departure from traditional works (though Rafferty [2003] sees even this form of interactivity as a threat to the integrity of film, since it shifts control away from the director to the viewer). In contrast, in strongly interactive works, the structural properties of the work are partly determined by the viewer’s own choices: in the case of interactive cinema, the images and sounds shown are partly determined by the viewer’s decisions, and there need be no fixed canonical order to their showing. This kind of interactivity is a radically new departure, unlike the weaker sort. It depends on the fact that interactive works can be specified by a set of rules with a certain provenance: these rules, in the case of computer programs, are algorithms, which determine, given a certain input and the interactor’s choices, what the output will be (Lopes 2001; see Saltz 1997, for a different account of interactivity). Audiences interact with strongly interactive works in a very different way than they do with traditional works. In a literal sense, constructivism is true for interactive works: the images on the screen partly depend on viewer’s choices, so the viewer is part author of the particular images she sees (though the work itself, determined by a set of algorithms, is still independent of her choices). And, whereas the traditional audience of a work is forced to a more contemplative attitude toward what is being enacted before it, interactors typically have a pragmatic attitude, since they must determine how they are (make-believedly) to act. In particular, this means that they can feel a wider range of emotions than is appropriate in noninteractive works: it makes sense to feel guilty, for instance, about what one may make-believedly have done in a videogame (Tavinor 2005). And, though Lopes argues against the point, interactors are plausibly like performers of musical works and plays: just as, within the permissible boundaries specified by the work, performers have to make certain choices that partly determine the properties of the performance they give, so interactors make certain choices, within the rules specified by the work, that partly determine the properties of their interactions with the works (what they see and hear).
Realism Is digital cinema different in respect of realism from traditional film? The question cannot be answered simply, since the notion of realism is one that has many aspects. One sense of “realism” is that of transparency: some philosophers, most prominently Kendall L. Walton, have argued that photographs are transparent in the sense that, when we look at a photograph, we literally see the object photographed. Look at a photograph of your ancestors and you literally see them. The claim is general: it would apply to all photographs, both traditional and digital. However, crucial to Walton’s defense of photographic transparency is that features of the photograph are counterfactually dependent on features of the subject in a way that is not mediated by the photographer’s beliefs. Nonphotographic pictures, in contrast, are mediated by the artist’s beliefs about her subject: for this reason, Walton holds that (with the exception of a few mechanically executed paintings and drawings), handmade pictures are not transparent (Walton 1984). So, on this view, nonphotographic digital images are opaque, i.e., not transparent. Given the prevalence of mélange images in 80
TF17557.indb 80
16/9/08 15:47:20
DIGITA L CI N EM A
digital cinema, Walton’s view has the consequence that digital cinema is sometimes transparent, sometimes opaque, and that certain aspects of a single image may be transparent and other aspects opaque. And digital images may seem transparent, even though they are opaque. Lopes, in contrast, has argued that not just photographs, but also handmade pictures of actual objects, can be transparent (Lopes 1996: chapter 9). This has the advantage of preserving a unitary account of transparency for all digital images, whether photographic or not, of actual objects, but it inherits the problems of Walton’s account and then adds several counterintuitive consequences of its own. An alternative account is also unitary, but holds that all pictures, whether photographic or nonphotographic, are opaque; this has the advantages of preserving a unitary account of pictures in respect of transparency, while avoiding the counterintuitive consequences of Lopes’ account (Gaut forthcoming: chapter 2). Another sense of realism is ontological realism: from the existence of a photograph, it follows that its subject existed. Photographs have a causal relation to their subjects; handmade pictures have an intentional relation to them. From the obtaining of a causal relation, it follows that the subject of the photograph existed at the time the photograph was made: one cannot photograph something that does not exist, such as King Kong. From the obtaining of an intentional relation, it does not follow that the subject existed at the time that the picture was made: I can make a painting of Kong. So photographs, whether traditional or digital, can only be of what is actual. (A traditional animated film is a photograph of an actual animation drawing or painting; live-action photographic fiction films are photographs of people, such as actors, who are make-believedly someone else; and in both cases audiences can also make-believe that the photographs are images of fictional things.) Nonphotographic digital images or aspects of digital images that are not photographic, in contrast, can be of what is not actual. So traditional film is ontologically realistic, but digital film is not in all cases. (And note that these features show only that the subject existed, not that the subject possessed the properties it seems to possess: there is plenty of scope for manipulation in both traditional and digital film.) The traditional film image is, in Peirce’s terms, an index (has a causal relation to its referent), whereas many digital images are merely icons (resembling their subjects but lacking a causal relation to them) (Prince 1996). A third sense of realism, perceptual realism, is that in which a realistic depiction looks like the object it depicts. How to make that claim precise is notoriously difficult, but a plausible view is that the picture and its object resemble each other in respect of their possession of those visual properties that trigger the capacity to recognize the object depicted. A horse picture resembles a horse in that both trigger the visual horse-recognition capacity. Pictures have greater realism, the greater is the number of properties that they depict in this fashion (Currie 1996: chapter 3). This view grounds the claim that certain filmmaking styles are more realistic than others. Renoir’s deepfocus, long-take style is more realistic than Eisenstein’s montage style, since looking at a deep-focus-style film is more like looking at the subjects depicted than looking at a montage-style film is like looking at the subjects it depicts. On this construal of realism, digital cinema has a greater capacity for realism than does traditional film, 81
TF17557.indb 81
16/9/08 15:47:20
BE RYS GAU T
since it can, for instance, make takes of much greater length than can traditional film, which is limited to a film length of about fifteen minutes, due to the bulkiness of the film roll in the camera. Digital films have been composed of one take that is the length of the entire film, over one and a half hours: for instance, Figgis’ Timecode and Alexander Sokurov’s Russian Ark (2002). A fourth sense of realism is that of illusionism. It is sometimes claimed that the cinema spectator is under an illusion, understood as a false belief, of one sort or another, such as that she is really in the presence of the events depicted. Most of these claims are straightforwardly false: if, for instance, the spectator believed that she were really in the presence of the ravenous monster in a horror film, she would be fleeing the movie theater, not contentedly munching on her popcorn. However, the viewer of digital cinema may well be under illusions in respects that the viewer of traditional film is standardly not. For instance, the viewer may assume that she is seeing an image of an actor moving in a real location, but the set has in fact been created purely digitally. Or she may assume that she is seeing an image of actors moving, whereas she is seeing purely digitally created characters: in The Lord of the Rings and King Kong, for instance, “digital doubles” are used at several points, where placing actors in the scene would have been too dangerous. Even close inspection may fail to reveal whether a digital double is used, especially in long shots. Digital creation and manipulation of characters and sets, then, give a much greater scope for illusionism in digital cinema than in traditional film. A fifth sense of “realism,” already adverted to, is photorealism, and it can be achieved comparatively easily in digital cinema, unlike traditional film. An animation image is photoreal when it looks like a photograph of a real object. Digital animation employs features such as motion blur and lens flare so that its images look like photographs of real objects: here the standard of realism is not that the image looks like the object would in real life, since motion blur and lens flare are artifacts of the camera and are not something that we see when observing a scene through our own eyes. Rather, the notion of photorealism is a derivative one: it takes the photograph as the standard of reality and tries to make the animated image look like a photograph. Traditional handmade animation can pursue the photoreal ideal too, but as noted earlier, it is one that is all but impossible for it to achieve. Photoreal animation is a central means for achieving the realism of fantasy and special effects digital movies. Finally, there is what we can call epistemic realism: an image is epistemically real just in case it provides strong (though defeasible) evidence that what it apparently depicts really happened. Traditionally, photographs have been accorded great evidential authority: we accept a photograph as providing much stronger evidence for something’s having occurred than we would a drawing or painting, however meticulous, of that scene. A traditional film can be viewed as constituting good evidence that certain actors were at certain locations in the past. But a digital photograph can be subject to so much manipulation that it can be systematically misleading about what it is a photograph of, as in the case of digital manipulation of models to make them look more glamorous. As Barbara Savedoff has noted, the rise of the digital photograph threatens the general “destruction of photographic credibility” (Savedoff 82
TF17557.indb 82
16/9/08 15:47:20
DIGITA L CI N EM A
1997: 212). And obviously the wider category of the digital image may provide no evidence whatsoever for something’s having occurred (such as King Kong’s fall from the Empire State Building). The problem is made worse by the achievability of the photoreal ideal: we may think we are looking at a photograph of something, though it is in fact a mere animation frame. So, digital cinema has greater powers to achieve realistic-looking images than traditional film. The irony, though, is that when viewers come to know of these powers, they have every reason to be suspicious about whether what they seem to have evidence for happening really did happen. In this sense, digital cinema’s pursuit of realism is a self-defeating project: the informed viewer increasingly has reason for a suspension of judgment about the evidential authority of the digital image.
Authorship and control Is it possible to have a single author of a digital film? More weakly, does digital cinema extend the degree of directorial control over films? Some directors have thought so: Robert Rodriguez, for instance, has said of digital cinema that “It’s the speed and power of almost having a paintbrush in your hand” (Waters 2005: 16). Technological achievements appear to support this view. The mélange image, as we have seen, increases the power of filmmakers to manipulate the image. With Gollum and King Kong, a combination of motion capture, motion editing, and key-frame animation allowed increased control, relative to the actor, over a character’s actions, since only aspects of the actor’s performance were incorporated into the character’s actions. However, though the filmmakers collectively have more control over the image, it is not true that the director’s control is increased relative to that of all other collaborators. On the contrary, even a brief glance at the credits of most digital films shows a large increase in the number of people credited with working on a film. This is hardly surprising, given the technically and artistically demanding nature of many of the tasks associated with digital filmmaking. Digital films are currently more, not less, collaborative than traditional films. And, as I have argued elsewhere, traditional films are, if they involve at least one actor distinct from the director, essentially collaborative, and the single authorship view is false of them. One reason is that the actor is an expressive agent, and hence the director is not in the same position as, say, the author of a novel, who creates characters without having to rely on the recording of others’ expressive actions (Gaut 1997; for the best defense of the single authorship view, see Livingston 1997). One might suppose that increased collaboration in digital cinema is merely a feature of the current state of technology and that technical progress will one day result in all of these digital tasks being performable by the director. This may be so, since it is impossible to forecast the path of technical development. So suppose that such developments result in a director being able to modify on his own all features of the digital image, and that he did choose to modify every single feature. Then, if the actor’s contribution is still visible in the finished product, it would still not be a 83
TF17557.indb 83
16/9/08 15:47:20
BE RYS GAU T
single authored film, since there would be at least two expressive agents involved in the final film. Or, alternatively, the modification might be so extensive that nothing was left of the actor’s contribution. In that case the film would be the equivalent of an animated film, and there is no doubt that animated films can be the products of a single individual. So such digital films would have a single author, but only for the same reason that animated films can have a single author, due to the absence of actors’ expressive contribution. Digital cinema creates new possibilities for collaboration in performance. One can now unbundle an actor’s role: Serkis’ voice and bodily and facial movements were incorporated into the Kong character, but other aspects, such as his particular physical look, were discarded. And the role of Kong was in effect collectively acted. The motion-capture footage was edited to alter aspects of Serkis’ performance – a motion editor talks of “performance change” and of “keeping the tension in the eyes” of Kong (Jackson 2006). Key-frame animators added further nuances to the performance, particularly around the face, where the motion-capture technology was insufficiently precise to record emotional subtleties and in scenes where it was impossible for the real actor to do what Kong was doing. So, here again, digital cinema has saliently different features from traditional film, but it is not in virtue of its enhancing single authorship, but in expanding the role of collaboration in film. See also Realism (Chapter 22), Authorship (Chapter 2), Definition of “cinema” (Chapter 5), Film as art (Chapter 11), Interpretation (Chapter 15), and Medium (Chapter 16).
References Binkley, T. (1997) “The Vitality of Digital Creation,” Journal of Aesthetics and Art Criticism 55: 107–16. Gaut, B. (1997) “Film Authorship and Collaboration,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Oxford University Press. —— (forthcoming) A Philosophy of Cinematic Art, Cambridge: Cambridge University Press. Jackson, P. (dir.) (2004) The Lord of the Rings (extended DVD version). (See esp., “The Taming of Smeagol,” appendix 3; “Visual Effects” on MASSIVE, appendix 6.) —— (dir.) (2006) King Kong, Disc 2, 2006: Post-Production Diaries, esp., “Bringing Kong to Life: I & II.” Kerlow, I. (2004) The Art of 3D Computer Animation and Effects, 3rd ed., Hoboken, NJ: John Wiley. Livingston, P. (1997) “Cinematic Authorship,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Oxford University Press. Lopes, D. (1996) Understanding Pictures, Oxford: Oxford University Press. —— (2001) “The Ontology of Interactive Art,” Journal of Aesthetic Education 35: 65–81. —— (2004) “Digital Art,” in L. Floridi (ed.) The Blackwell Guide to the Philosophy of Computing and Information, Oxford: Blackwell. McKernan, B. (2005) Digital Cinema: The Revolution in Cinematography, Postproduction and Distribution, New York: McGraw-Hill. Mitchell, W. (1992) The Reconfigured Eye: Visual Truth in the Post-Photographic Era, Cambridge, MA: MIT Press. Monaco, J. (2000) How to Read a Film, 3rd ed., New York: Oxford University Press. Poole, S. (2004) Trigger Happy: Videogames and the Entertainment Revolution, New York: Arcade Publishing. Prince, S. (1996) “True Lies: Perceptual Realism, Digital Images, and Film Theory,” Film Quarterly 49:
84
TF17557.indb 84
16/9/08 15:47:20
DIGITA L CI N EM A
27–38. Reprinted in L. Braudy and M. Cohen (eds.) Film Theory and Criticism, 6th ed. (New York: Oxford University Press, 2004). Rafferty, T. (2003) “Everybody Gets a Cut: DVDs Give Viewers Dozens of Choices – And That’s the Problem,” New York Times Magazine (4 May): 58, 60–1. Reprinted in N. Carroll and J. Choi (eds.) Philosophy of Film and Motion Pictures (Oxford: Blackwell, 2006). Saltz, David. (1997) “The Art of Interaction: Interactivity, Performativity, and Computers,” Journal of Aesthetics and Art Criticism 55: 117–27. Savedoff, B. (1997) “Escaping Reality: Digital Imagery and the Resources of Photography,” Journal of Aesthetics and Art Criticism 55: 201–14. Tavinor, G. (2005) “Videogames and Interactive Fiction,” Philosophy and Literature 29: 24–40. Walton, K. (1984) “Transparent Pictures,” Critical Inquiry 11: 246–77. Waters, R. (2005) “Hollywood Sees Power Shift from Film-set to Desk-top,” Financial Times (20 June): 16.
85
TF17557.indb 85
16/9/08 15:47:20
8
EMOT ION AND AF F EC T Carl Plantinga Why should anyone be concerned about the affective experiences of film viewers? We could list at least the following reasons: (1) a pleasurable affective experience is one of the primary motivations for film viewing; (2) emotions provide narrative and character information and are necessary to fully understand a narrative film; (3) emotions and affects, whether pleasurable or not, are a central element of the phenomenological experience of the cinema; (4) emotions are intimately tied to cognitions, and for this reason affective experience, meaning, and interpretation are firmly intertwined; and (5) emotions as experienced in films have powerful rhetorical functions and contribute to a film’s ideological effects. In the 1970s and 1980s film theorists approached questions of cinema and affect nearly exclusively from the standpoint of psychoanalysis. Psychoanalytic film theory certainly dealt with human affect but focused on desires and pleasures, rather than emotions (Plantinga and Smith 1999; Plantinga forthcoming). Of typical emotions – such as fear, pity, admiration, disgust, and compassion – psychoanalytic theory has little to say. Neither could psychoanalytic film theory shed light on affective phenomena such as mimicry and emotional contagion. This essay focuses on cinematic affect from what might be termed a cognitive or naturalistic perspective, derived from contemporary developments in psychology, aesthetics, and film and media theory.
Emotions and affects A human emotion is a complex phenomenon, but is probably best thought of as a mental state accompanied by physiological and autonomic nervous system changes, subjective feelings, action tendencies (that is, tendencies toward certain behaviors in response to emotions), and outward bodily behaviors (facial expressions, body postures, gestures, vocalizations, etc.). Emotions are intentional in the sense that they are directed toward some “object” (which may or may not be real or correctly understood by the person having the emotion). In some cases, such as phobias or delusions, the emotion is powerful and real, but its object is not. Robert C. Roberts calls an emotion a “concern based construal,” that is, it is a person’s perception or judgment of a state of affairs rooted in that person’s concerns (Roberts 2003: 65–83). To take a simple example, Ann becomes fearful when she construes that her cabbie is driving
TF17557.indb 86
16/9/08 15:47:20
EMOT ION A ND A F F E C T
dangerously fast; her concern is for her safety. Jane experiences guilt when she mulls on the fact that she has stolen a cookie from the cookie jar; her construal is that she is blameworthy and her concern is that she not be a “bad” person. This may seem to make an emotion a simple phenomenon, but such is not the case; emotions are complex, episodic, dynamic, and structured (Goldie 2000: 12–16). There is also considerable debate about whether emotions, as mental states, should be considered to be judgments or beliefs, rather than something broader like generalized “perceptions” or “impressions.” Roberts, for example, writes that the construals that in part make up emotions may be unconscious, and often have an immediacy reminiscent of sense perception: “They are impressions, way things appear to the subject; they are experiences and not just judgments or thoughts or beliefs” (Roberts 2003: 75, 84–7, 89–92). Derek Matravers identifies the view that an emotion must necessarily be a belief “narrow cognitivism” (Matravers 1998: 14); Jenefer Robinson takes a similar position (Robinson 2005). Both favor a broader view in which the mental state characteristic of an emotion need not imply belief. “Affect” is a broader category than “emotion.” Affects are any felt bodily state, including a wide range of phenomena, including emotions, moods, reflex actions, autonomic responses, mirror reflexes, desires, pleasures, etc. Emotions are a special case of affect because they are intentional, that is, they “take” an “object” and represent a relationship between a person and her or his environment. For example, the object of my fear might be a charging rhinoceros. The object of my jealously is that man or that woman chatting up my spouse. My fear and jealously are intentional because they are about something and are directed at something (Goldie 2000: 16–28). Affects, on the other hand, lack this intentionality or aboutness. My indigestion and feelings of nausea are caused by eating too many tacos, perhaps, but my indigestion is not about those tacos in the way that my jealously is about that threatening lothario. The indigestion is merely a matter of a stimulus and response and is not an intentional relationship. Noël Carroll holds that affects are more structurally primitive than emotions and reserves the term “emotion” for affects that have a complex structure that integrates computation (or what Roberts would call a construal) with feelings (Carroll 2008). Moods, another type of affect, are thought to be longer lasting and more diffuse than emotions proper. Thus a mood may last for hours or days, and it may take no object whatsoever, when, for example, one is in a good or a bad mood for no apparent reason. The claim that moods take no object is not without problems, as Peter Goldie points out (Goldie 2000). Sometimes bad moods follow negative emotions, and good moods follow positive emotions. There is a sense, then, in which the object of the emotion is also the object of the mood. It is also worth noting that moods and emotions, however conceived, impact each other. A person in a giddy mood is primed to experience certain emotions because moods influence construals. A giddy person is more likely to construe events or situations in ways that giddiness makes salient – as amusing or exciting or benign. Likewise, emotional experience can also alter subsequent moods.
87
TF17557.indb 87
16/9/08 15:47:21
CA R L PL A N T I NGA
Emotions at the cinema What kinds of emotions do spectators have at the movies? One question we might ask is whether “filmic emotions” are different in kind from emotions had outside the movie theater. Many film and communication scholars hold that filmic emotions are similar in many ways to garden-variety emotions. Some maintain that spectator responses to fictional events closely approximate responses to actual events, as though the spectator entertained the illusion that the events were actually happening to her (Tan 1996: 82) or as though the spectator were a “side participant” in actual events (Gerrig and Prentice 1996). Tan claims that the emotions spectators have at movies are “witness emotions,” comparable to those we might have in witnessing actual events, and that the spectator imagines that she or he is “in the scene.” Richard Gerrig and Deborah Prentice, similarly, write that spectator emotions approximate those that an actual “side participant” might have. These claims, however, neglect important differences between spectator emotions and “extrafilmic” witness emotions. Emotional responses, I have argued, are concernbased construals, in which the emoting party responds to elements of her environment in relation to her concerns. This construal is often automatic and immediate, in part the result of commonplace “assumptions” that are largely unconscious. We can call these background assumptions the emoter’s “mental set.” In the case of the movie spectator, what elements of the spectator’s mental set inflect emotional response away from “real-world” emotional responses? There are at least two elements of the spectator’s mental set that significantly alter response: (1) the realization that the spectator cannot affect the events in any way; and/or (2) (in the case of narrative fiction) the understanding that what she sees is in fact fictional and not actual. To discuss the first element, one should understand that emotions are often accompanied by “action tendencies” that are suppressed or disabled in a film viewing environment. Emotions evolved in part as adaptive mechanisms that enable us to quickly detect threats and/or affordances, or in other words, opportunities in the environment, and to mobilize action. Thus the threat of a tiger leads to actions designed to enhance the possibility of survival, such as flight. Disgust initiates avoidance tendencies; and pity, the desire to assuage the hurt in the other. The spectator of any traditional film understands that no action will influence the events depicted on the screen. This gives the emotions a different phenomenological feel and makes the negative emotions, at least, less laden with anxiety, since emotional response is freed from any immediate need or responsibility to act. Second, the spectator’s mental set (for fictions) also includes the knowledge that what she sees is fictional and not actual. Film viewing, whether in the movie theater, at home, in a commercial airliner, or on an iPod or cell phone, rarely ever involves the illusion that what is seen is actually occurring at the time of the viewing, and this fact vitiates the analogy to being an actual witness to an event. Again, this is precisely what allows films to be emotionally powerful without threatening the spectator’s safety and comfort. The assumption of fictionality allows spectators the enjoyment of scenes involving rampaging dinosaurs, raging battles, and scenes in which characters undergo 88
TF17557.indb 88
16/9/08 15:47:21
EMOT ION A ND A F F E C T
extreme emotional distress. Such scenes would be unpleasant and unbearable for most in actual life, but with the potential to provide pleasure in the realm of fiction film spectatorship. The institutions of fiction, of which the film and media industries are important components, prompt the spectator to approach a film in the context of social convention and of play (M. Smith 1995b). If it is true, in fact, that narrative fiction films and the institutions that provide them invite the audience to relate to the film as a work of fiction, then this raises an interesting question: Why do spectators have any emotional responses at all? If we know that serial killer Anton Chigurh in No Country for Old Men (2007) does not actually exist, why is he such a frightening figure? Why have compassion for Dorothy in The Wizard of Oz (1939) if we understand that she is a mere fiction? This so-called “paradox of fiction” has intrigued philosophers since Colin Radford argued that if emotions do depend fundamentally on belief, then perhaps our emotional responses to fiction are irrational and undesirable, since we do not believe the fictional events or characters to be real (Radford 1975). Two prominent “solutions” to the paradox are the pretense theory and the thought theory. Pretense theory contends that spectators and readers of fiction have emotions, or perhaps “quasi-emotions,” within the context of a game of make-believe (Walton 1990). Thought theory contends that spectators or readers can have genuine emotional responses to propositions or imaginations they do not necessarily believe to be true (Scruton 1974; Lamarque 1981; Carroll 1990: 59–88). But one might also point out that the paradox is most difficult to resolve for “narrow cognitivists,” and less difficult for those with a broader perspective on the emotions. If emotions sometimes result from impressions or perceptions that do not rise to the level of belief, then the paradox begins to dissolve because it depends on the faulty premise that emotions are necessarily dependent on belief. We respond to fictions with actual emotions because the human mind/brain is modular, with various systems operating more or less independently. Fictions engage parts of the brain that generate automatic affective and emotional responses, while eliciting high-order cognitive processing that precludes viewers from responding as though the fictions were actual events. If human responses were solely the result of rationalistic deliberations and separate from automatic and unconscious “built in” psychic and bodily mechanisms, perhaps the paradox of fiction would seem more troublesome. To return to the question of how cinematic emotions differ from extracinematic emotions, we might say that cinematic emotions differ in at least the two ways described above. It should also be remembered that the emotions had at the cinema do not all take as their object a fictional character or narrative events. Spectators may also have artifact emotions that take as their object the film itself, and metaemotions that take as their object the spectator’s own prior response, or that of an audience. A fascination with the construction of a film narrative or production design, admiration for the skill with which a scene is constructed, or outrage at racist or homophobic stereotypes, all are artifact emotions. Metaemotions are important as well, as a spectator can have pride or satisfaction for having responded compassionately to a character’s plight (Feagin 1983), for example, or can become disgusted with herself for responding to the same scene if it is later deemed to be sentimental. 89
TF17557.indb 89
16/9/08 15:47:21
CA R L PL A N T I NGA
Fiction emotions can be either direct or sympathetic/antipathetic. The direct emotions take as their object the unfolding narrative itself. Examples are anticipation, curiosity, surprise, suspense, and fascination. Tan proposes that “interest” is not only the primary direct emotion at the cinema, but is a global and long-lasting emotion that functions to rivet the spectator’s attention to the screen (Tan 1996: 85–120). Sympathetic and antipathetic emotions take as their object the character, behavior, or situation of a film character, ranging from “pro-emotions” like pity, compassion, and admiration to “con-emotions” such as anger, disgust, and contempt. There are good reasons to believe that fiction, artifact, and metaemotions in response to films have strong similarities to extrafilmic emotions, but can be altered or inflected by various filmic strategies. Thus, for example, audience contempt for and disgust at Dr Chilton in The Silence of the Lambs (1992) is elicited by his behavior, which arguably constitutes sexual harassment of a sympathetic protagonist Clarice (Jodie Foster) and demonstrates both sexist patronization and an untempered drive to retain power. Such patterns for the elicitation of disgust and contempt exist for the spectator both inside and outside the movie theater. Yet in many ways the filmmakers amp up the disgust and contempt by the way Dr Chilton is framed in relation to Clarice, in tight two-shots that emphasize an unwelcome intimacy that is forced on Clarice, and in directing the audience’s attention to unpleasant features of Chilton’s appearance, facial expressions, and vocal intonations. Movies are typically more “hyperrealist” – exaggerated, clear, focused, and simplified – than realist (Carroll 1996b: 86–7). (It should also be noted that films have the capacity not only to clarify and intensify emotional response, but may also complicate responses in intriguing ways by employing techniques that elicit contradictory or discordant affective charges.) The study of how particular film genres elicit emotion also promises to shed light on the types of emotions spectators have at the movies. The most sustained study in this regard, from a cognitive/analytic perspective, has been Carroll’s The Philosophy of Horror. There Carroll posits that “art horror” combines disgust with fear, both directed at an “impure being,” or monster, that threatens the safety of sympathetic characters. Carroll also proposes a solution to the “paradox of horror,” which is related to the “paradox of tragedy” (Hume 1965; Feagin 1983) and like the latter, a subspecies of what might be termed the “paradox of negative emotion.” Why do audiences, in their consumptions of fictions, willingly subject themselves to negative emotions such as fear and disgust, given that these emotions are typically among those that bring displeasure, discomfort, and distress? Do film audiences actually enjoy fear and disgust as “art emotions,” or, as Carroll suggests, are audiences willing to abide these unpleasant emotions because their curiosity about this monster – this categorically impure being, together with a narrative of gradual disclosure, brings a pleasure that is strong enough to outweigh the displeasures of fear and disgust? Carroll’s view has generated intense debate (e.g., Feagin 1992; Gaut 1995; Yanal 1999), and the paradox of negative emotion remains a formidable issue.
90
TF17557.indb 90
16/9/08 15:47:21
EMOT ION A ND A F F E C T
Narrative and character Film and television are typically experienced through audio-visual displays that appeal to the senses in ways that differ from literature. This has implications for affective response, a topic further discussed in the following section. But it makes sense first to examine the ways in which narrative films employ narrative and character, common to both film and literature, in the elicitation of spectator emotion. Most theorists of filmic emotions see narrative and character engagement as central to the spectator’s emotional response to a film. Emotions occur in time and, like narratives themselves, constantly evolve as situations change and as various bodily affects run their course. Thus the spectator’s concern for a character in relation to evolving narrative situations is typically seen as the backbone of spectator emotion. Emotional response is typically rooted in the spectator’s appraisal of the narrative in conjunction with character goals and desires. We name emotions in accordance with particular kinds of “concern-based construals,” and those construals, in turn, accord with conventional paradigm scenarios. Thus certain narrative paradigms, often corresponding with particular genres, tend to elicit pity, fear, elation, compassion, etc. Sympathy or antipathy for various characters provides a compass for the spectator in light of which she construes the narrative situation. Thus at the end of The Silence of the Lambs, when Hannibal tells Clarice that he plans to dine on Dr Chilton’s liver, the spectator may experience a rather grim satisfaction, rather than abject horror, because the spectator has been led to detest Dr Chilton. Had Hannibal informed Clarice that he planned to dine on her, spectator response would have been far different (and manifestly unpleasant), because the spectator has been led to sympathize with her. One issue of contention has been the degree to which spectator emotions are identical with the presumed emotions of a sympathetic protagonist. Spectator identification or engagement, sympathy, and empathy all are essential in any discussion of film affect. Since they are discussed at length in another essay in this volume (see “Empathy and character engagement” in this volume), only brief mention will be made here. The film theorist who most strongly promotes the “identification hypothesis,” that is, the view that spectator and protagonist emotions are largely identical, is Torben Grodal, whose “PECMA flow model” of emotion elicitation in film (the acronym “PECMA” standing for perception, emotion, cognition, and motor action) combines cognitive theory with neuroscience. For Grodal, spectator emotional response depends fundamentally on the viewer’s identification with characters. In viewing a film, the spectator adopts the character’s goals, mentally simulates the character’s situation, and responds to the film based on that simulation. Grodal’s model is a “flow” model in that it compares the temporal experience of the viewer to that of one flowing down a narrative river that leads to the film’s end. The protagonist becomes the viewer’s mental raft, so to speak. The viewer assesses the protagonist’s goals, progress, blockages, and obstacles, just as the passenger might assess the progress of a raft on which she rides. The viewer responds accordingly as the viewer and protagonist float down the narrative river together. The viewer projects him- or herself into the protagonist, responding much as the protagonist would (Grodal 1997, 2006). 91
TF17557.indb 91
16/9/08 15:47:21
CA R L PL A N T I NGA
One objection to this view is based on counterexamples to the thesis that the viewer’s mental responses match the fictional feelings of the film’s protagonist. There are many instances in which the spectator’s goals and desires conflict with those of even the most sympathetic protagonist. One thinks of Ethan Edwards’ (John Wayne) obsessive and racist desire to kill his own niece, Debbie (Natalie Wood) in The Searchers (1956) because she has become the mate of the hated Indian chief Scar. Or one thinks of the spectator’s desire that Rick (Humphrey Bogart) in Casablanca (1942) overcome his cynical narcissism, forgive Ilsa (Ingrid Bergman), and help the war hero Laslow escape Casablanca. In both cases, the spectator’s goals and desires do not match those of the protagonist, at least until the film’s ending. Or consider films with flawed protagonists such as Five Easy Pieces (1970), Unforgiven (1992), or Erin Brockovich (2000), warped protagonists in films such as The Godfather: Part II (1974), conflicted, confused, and morally ambiguous protagonists such as Deckard in Blade Runner (1982) or Johnny Cash in Walk the Line (2005). The spectator can understand these protagonists as flawed, warped, confused, conflicted, and morally ambiguous only because the spectator responds to them from an egocentric perspective and does not wholly take on the protagonists’ goals and desires, as if from the inside. The view that spectators invariably respond to characters, whether sympathetic or not, from an exterior perspective we might call “ineliminable egocentrism,” or what Carroll calls “assimilation” (Carroll 1990: 88–96) or, in relation to sympathetic characters, “solidarity” (Carroll 2008). It is plausible that the film’s narration constructs a preferred or intended trajectory of emotional responses that is congruent with the responses of a sympathetic protagonist, but never identical (Plantinga forthcoming). Spectator responses, as elicited by the film, often have a similar valence to those of a sympathetic protagonist, but also differ, and sometimes markedly. For example, while many spectators will have a great deal of sympathy for Erin (Julia Roberts) as she battles corporate polluters in Erin Brockovich her tendency for impulsive anger and foul language are shown to detract from her effectiveness. While Erin is neither amused by nor disdainful of herself during her angry outbursts, the spectator is led to judge these outbursts as mild yet amusing faults that need correction. In these cases, Erin is merely angry, while the spectator may be slightly disdainful, disappointed, and/ or bemused. The viewer’s “relationship” with a favored character is one of sympathy, together with an assimilation of the character’s situation. The spectator’s psychic “relationship” with a sympathetic protagonist is both internal and external. The internal part consists of the viewer’s simulation of the character’s thoughts, together with automatic processes such as mimicry and emotional contagion (M. Smith 1995a; Plantinga 1999; Carroll 2008). The external part involves an assessment of the character’s broader situation, incorporating information sometimes unavailable to the character, and incorporating the character’s assessment of, and response to his situation. Given this simultaneously internal and external engagement with a character, then, neither the thoughts nor the feelings of the spectator can possibly be identical with those of the character, since the response necessarily involves that external portion of character engagement (M. Smith 1995a). 92
TF17557.indb 92
16/9/08 15:47:21
EMOT ION A ND A F F E C T
The film theorist Greg M. Smith takes an alternative approach to a focus on narrative and character as primary elicitors of emotion. Smith has built his theory of affect elicitation at the cinema on the importance of moods in film. Smith argues that the “primary emotive effect of film is to create mood” (G. Smith 2003: 42). Where emotions are intense, brief, and intermittent, moods have longer duration. Emotions depend on moods as “orienting states” that prepare the viewer for the specific emotional responses that are cued through various stylistic techniques. Smith rejects the centrality of narrative and character in emotion elicitation, claiming that the advantage of the mood-cue approach is in its focus on the cues offered by film style, rather than primarily by narrative and character. Stylistic elements cue emotions by way of associations, rather than by way of the appraisals associated with narrative and character, and it is affective experience via association and other less prototypical means that most interests Smith. Thus Smith turns, first, to the stylistic cues designed to elicit emotion, and thus differs from the majority of theorists/philosophers, who would put narrative and character at the center of emotion elicitation, and who would argue that moods and stylistic cues in films gain their salience in relation to overarching character goals and narrative situation.
Affect and style A cinematic work is viewed in the form of some kind of audio-visual display, and as such, film viewing is sensual in a way that reading literature is not. Any account of affect elicitation in the movies must take in account the sensual nature of the audiovisual display, or the means by which films appeal to the senses of sight and hearing. Research into this aspect of filmic experience is in a preliminary stage. The perceptual realism (Prince 1996) that the medium often employs allows a film to take advantage of various perceptual processes with strong affective impacts. Films draw their power in part from the automatic bodily processes that they invoke in the spectator. Linda Williams’ essay, “Film Bodies: Gender, Genre, and Excess,” explores the excesses of violence, sexual desire, and weeping elicited by horror, pornography, and melodrama, respectively (Williams 1999). Williams calls these “body genres,” but this would seem to imply that some genres appeal simply to the mind. All films appeal to the corporeality of the viewer. This has been acknowledged recently in film studies with books such as Vivian Sobchack’s Carnal Thoughts: Embodiment and Moving Image Culture (2004) and Carl Plantinga’s Moving Viewers: American Film and the Spectator’s Experience (forthcoming) and in philosophy, for example, in Carroll’s The Philosophy of Motion Pictures (2008) and work by Amy Coplan (2006). Coplan ably describes the affective importance of perceptually realistic images of human bodies – and especially of the human face – in her entry in this volume (“Empathy and character engagement”). But there are other ways in which films, through their visual and aural appeals, have a strong affective charge. Spectator responses to movements, sounds, colors, textures, and spatial configurations – not to mention certain kinds of pictorial content – are in large part automatic and prereflective. The human brain did not evolve to interact with the visual media, or 93
TF17557.indb 93
16/9/08 15:47:21
CA R L PL A N T I NGA
indeed, with representations of any sort, but adapted itself to the more immediate environmental data to which humans must daily respond for their survival and flourishing. Thus the spectator’s responses to the audio-visual media are rooted, in part, in those natural perceptual responses that have developed over long periods of human history, and that in some cases have been conventionalized through decades of filmmaking. The filmmaker can affect the spectator through all of the various parameters of film style, from shot composition, to movement, to editing, to color, to sound and music. A few examples will suffice to demonstrate the various effects and devices through which films provide affect. Music resonates with or otherwise impacts our very physiologies through rhythm, dynamics, tempo, and pitch. It affects spectators in physical ways that researchers are only beginning to understand. Auditory entrainment, or what is also sometimes called the “Frequency Following Effect,” for example, is an auditory “mirror” effect. Physiological functions of the body, such as the heartbeat or brainwaves, tend to synchronize with the rhythmic patterns of audio, whether it is a complex musical score or a lone heartbeat (Vickers 2004). A mere heartbeat on the soundtrack, slowly increasing in volume and/or tempo, can elevate suspense and anxiety in response to a film scene, in part by elevating the spectator’s heartbeat (Stern et al. 1972). Auditory entrainment is one of the ways in which sound, and especially music, can have direct bodily effects on the spectator. Other examples would include the startle response and other forms of perceptual surprise and disorientation (Baird 2000; Choi 2003); erotic attraction and other visceral responses to various sorts of content; affective mimicry or mirror responses in relation to the human face, voice, posture, and gesture (Plantinga 1999; Carroll 2008); perceptual and physiological effects of color and shading; kinesthetic or movement effects; and proxemic patterns (closeness and distance responses). Although many of these responses are automatic and immediate, they relate to more complex emotions in a variety of ways. Affects may enable or intensify emotions, and may color experience in subtle ways by complicating, nuancing, or deepening the affective experience of a scene (Plantinga forthcoming).
Rhetoric and ideology The cognitive or naturalistic approach to affect in film studies is in its infancy, and it has not yet approached questions of the rhetoric and ideology of emotion in film sufficiently. How does affect play into the ability of movies to confirm or alter values, beliefs, and/or ideologies? Psychoanalytic film theory, conjoined with Marxist thought, sought out the spectator’s deepest motivations for viewing films and related spectator pleasure and desire to broad ideological concerns. If such theories are judged by some to be unsatisfactory in their determinism and conceptual flaws, they nonetheless made these issues a central feature of inquiry. Cognitive and analytic theorists and philosophers, on the other hand, have so far been preoccupied with developing basic theories of emotional response to films, a preoccupation that is justified if one considers such fundamental theories to be prolegomena to considerations that run further afield. 94
TF17557.indb 94
16/9/08 15:47:21
EMOT ION A ND A F F E C T
Cognitive theorists have made preliminary forays into this territory, however, suggesting avenues from which questions of rhetoric and ideology in relation to affect might be examined from a broadly cognitive perspective. Carroll, for example, has defended the “images of women” approach to feminist film criticism, suggesting that the repetition of sexist stereotypes and paradigm scenarios (such as the “fatal woman” scenario of Fatal Attraction [1987] or the “spider woman” scenario of many films noir) may “warp the emotions” of viewers by negatively influencing the conventional emotional responses of men to women (Carroll 1996b: 270). Murray Smith (1996) and Plantinga (1997) have both criticized what they see as the misplaced suspicion of emotion in Brechtian criticism (see “Bertolt Brecht,” in this volume). Carroll is no doubt correct that stereotypes and paradigm scenarios have the capacity to alter conventional emotional responses on a broad scale; Plantinga offers a theory of the means by which fiction films make those stereotypes both salient and emotionally powerful (forthcoming). He focuses on the “gatekeeper” emotion of sociomoral disgust, and then generalizes to a consideration of the means by which various emotions can be used to elicit judgments of persons rooted in stereotypes and to make salient and pleasurable conventional ways of responding to narrative scenarios. See also Consciousness (Chapter 4), Empathy and character engagement (Chapter 9), Spectatorship (Chapter 23), Bertolt Brecht (Chapter 30), Noël Carroll (Chapter 31), Cognitive theory (Chapter 33), Phenomenology (Chapter 40), Psychoanalysis (Chapter 41), and Horror (Chapter 46).
References Baird, R. (2000) “The Startle Effect: Implications for Spectator Cognition and Media Theory,” Film Quarterly 53: 12–24. Carroll, N. (1990) The Philosophy of Horror, New York and London: Routledge. —— (1996a) “The Power of Movies,” in Theorizing the Moving Image, Cambridge: Cambridge University Press. —— (1996b) “The Image of Women in Film: A Defense of a Paradigm,” in Theorizing the Moving Image, Cambridge: Cambridge University Press. —— (2008) The Philosophy of Motion Pictures, Maldon, MA, and Oxford: Blackwell. Choi, J. (2003) “Fits and Startles: Cognitivism Revisited,” Journal of Aesthetics and Art Criticism 61: 149–57. Coplan, A. (2006) “Catching Character’s Emotions: Emotional Contagion Responses to Narrative Fiction Film,” Film Studies: An International Review 8: 26–38. Feagin, S. L. (1983) “The Pleasures of Tragedy,” American Philosophical Quarterly 20: 95–104. —— (1992) “Monsters, Disgust and Fascination,” Philosophical Studies 65: 75–84. Gaut, B. (1995) “The Enjoyment of Horror: A Response to Carroll,” British Journal of Aesthetics 35: 284–9. Gerrig, R. J., and Prentice, D. A. (1996) “Notes on Audience Response,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Goldie, P. (2000) The Emotions: A Philosophical Exploration, Oxford: Clarendon Press. Grodal, T. (1997) Moving Pictures: A New Theory of Genres, Feelings, and Cognition, Oxford: Oxford University Press. —— (2006) “The PECMA Flow: A General Model of Visual Aesthetics,” Film Studies: An International Review 8: 1–11.
95
TF17557.indb 95
16/9/08 15:47:21
CA R L PL A N T I NGA
Hume, D. (1965) “Of Tragedy,” in J. V. Lenz (ed.) Of the Standard of Taste and Other Essays, Indianapolis and New York: Bobbs-Merrill. Lamarque, P. (1981) “How Can We Fear and Pity Fictions?,” British Journal of Aesthetics 21: 291–304. Matravers, D. (1998) Art and Emotion, Oxford: Clarendon Press. Plantinga, C. (1997) “Notes on Spectator Emotion and Ideological Film Criticism,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Oxford University Press. —— (1999) “The Scene of Empathy and the Human Face on Film,” in C. Plantinga and G. M. Smith (eds.) Passionate Views: Film, Cognition, and Emotion, Baltimore, MD: Johns Hopkins University Press. —— (2006) “Disgusted at the Movies,” Film Studies: An International Review 8: 81–92. —— (forthcoming) Moving Viewers: American Film and the Spectator’s Experience, Berkeley: University of California Press. Plantinga, C., and Smith, G. M. (1999) “Introduction,” in C. Plantinga and G. M. Smith (eds.) Passionate Views: Film, Cognition, and Emotion, Baltimore, MD: Johns Hopkins University Press. Prince, S. (1996) “True Lies: Perceptual Realism, Digital Images, and Film Theory,” Film Quarterly 49: 27–37. Radford, C. (1975) “How Can We Be Moved by the Fate of Anna Karenina?,” Proceedings of the Aristotelian Society (suppl. vol.) 49: 67–80. Roberts, R. C. (2003) Emotions: An Essay in Aid of Moral Psychology, Cambridge: Cambridge University Press. Robinson, J. (2005) Deeper Than Reason: Emotion and Its Role in Literature, Music, and Art, Oxford: Oxford University Press. Scruton, R. (1974) Art and Imagination, London: Routledge and Kegan Paul. Smith, G. M. (2003) Film Structure and the Emotion System, Cambridge: Cambridge University Press. Smith, M. (1995a) Engaging Characters: Fiction, Emotion, and the Cinema, Oxford: Oxford University Press. —— (1995b) “Film Spectatorship and the Institution of Fiction,” Journal of Aesthetics and Art Criticism 53: 113–27. —— (1996) “The Logic and Legacy of Brechtianism,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Sobchack, V. (2004) Carnal Thoughts: Embodiment and Moving Image Culture, Berkeley: University of California Press. Stern, R. M., Botto, R. W., and Herrick, C. D. (1972) “Behavioral and Physiological Effects of False Heart-Rate Feedback: A Replication and Extension,” Psychophysiology 9: 21–9. Tan, E. S. (1996) Emotion and the Structure of Narrative Film, Mahwah, NJ: Lawrence Erlbaum. Vickers, E. (2004) Music and Consciousness. Available at www.sfxmachine.com/docs/musicandconsciousness.html (accessed 19 May 2008). (Paper on the conscious effects of music on listeners.) Walton, K. (1990) Mimesis as Make-Believe: On the Foundations of the Representational Arts, Cambridge, MA: Harvard University Press. Williams, L. (1999) “Film Bodies: Gender, Genre, and Excess,” in L. Braudy and M. Cohen (eds.) Film Theory and Aesthetics: Introductory Readings, 5th ed., Oxford: Oxford University Press. Yanal, R. J. (1999) Paradoxes of Emotion and Fiction, University Park, PA: Penn State University Press.
Further reading See chapter 6, “Affect and the Moving Image,” in N. Carroll, The Philosophy of Motion Pictures (Malden, MA: Blackwell, 2008). Mette Hjort and Sue Laver (eds.) Emotion and the Arts (Oxford: Oxford University Press, 1997) is a solid collection of philosophical essays. C. Plantinga, Moving Viewers: American Film and the Spectator’s Experience (Berkeley: University of California Press, forthcoming) offers a general theory of affect in cinema and its implications, concentrating on the theatrical experience of mainstream American films. C. Plantinga and G. M. Smith (eds.) Passionate Views: Film, Cognition, and Emotion (Baltimore, MD: Johns Hopkins University Press, 1999) is a collection of philosophical and theoretical essays on film and emotion by film scholars and philosophers. Although not dealing much with film, J. Robinson, Deeper than Reason: Emotion and its Role in Literature, Music, and Art (Oxford: Oxford University Press, 2005) offers an excellent analysis of emotion in relation to the arts.
96
TF17557.indb 96
16/9/08 15:47:21
9
EMPATH Y A N D C HAR AC T ER E NGAGEM ENT Amy Coplan Character engagement is a central topic in the philosophy of film that has implications for several other philosophical issues. Although discussions of character engagement in film studies and philosophy tell us a lot, they are often confusing for at least two reasons. First, a variety of explanatory concepts gets used in these discussions, and few of these are used uniformly. Second, the experience of character engagement itself is highly complex. It refers to all and any elements of spectatorial experience that are based on or relate to characters. This encompasses several distinct types of relationship, a number of additional types of response, and connections among all of the different experiences. There are two major theoretical approaches to questions about spectatorship: psychoanalytic film theory and cognitive film theory. Psychoanalytic film theory, as the name suggests, draws extensively on psychoanalysis, particularly the work of Jacques Lacan, as well as on Marxist theory, Saussurean linguistics, critical theory, and the work of several recent postmodern and postcolonial thinkers. Cognitive film theory is based largely on research in empirical sciences related to the mind, such as cognitive psychology, developmental psychology, social psychology, and neuroscience. Most cognitive film theorists also share the assumption that our cognitive and perceptual experience when watching film is more or less like our cognitive and perceptual experience of events in ordinary life. In this essay I will survey some of the leading explanations of character engagement developed in psychoanalytic and cognitive film theory. These explanations vary greatly in their operating assumptions, their theories of human psychology, and their sources of evidentiary support. What unifies them is that all are attempts to answer the question of how viewers engage with fictional characters in film.
TF17557.indb 97
16/9/08 15:47:21
A M Y C OPL A N
Psychoanalytic film theory During the 1970s, psychoanalytic film theory emerged as the leading framework in film studies for understanding film and our engagement with it. A central concern for theorists in this tradition is how film, the technology of film, and filmmaking conventions make film viewing a pleasurable experience that exploits unconscious desires, and reproduces the dominant ideology. Many of these theorists base their understanding of spectator psychology on a Lacanian model of human development and the formation of subjectivity. Apparatus theory – Jean-Louis Baudry and Christian Metz The apparatus theory of Jean-Louis Baudry and Christian Metz represents one of the most influential movements in psychoanalytic film theory, particularly on the issue of spectatorship (Baudry 1986a, b; Metz 1986). Apparatus theory asserts that in order to understand spectatorial response, we must first understand the nature of the cinematic apparatus, that is, the ways in which the organization and presentation of film interacts with human psychology, desire, and understanding. Baudry claims that the cinematic apparatus transforms the spectator into a transcendental subject “whose place is taken by the camera which constitutes and rules the objects in this ‘world’ ” (Baudry 1986a: 295). The stream of on-screen images creates an impression of reality that locates the ideal viewer (or spectatorial position) at the center of vision as the creator of meaning, providing a sense of unity and control. It is from the privileged position of the camera that the spectator identifies with the on-screen characters. Baudry describes two different processes of identification. First, there is the spectator’s identification with the camera. This, in turn, makes possible the spectator’s identification with the characters. On this view, then, just as during the mirror phase the child identifies with idealized images of himself, so the film viewer, from the perspective of the camera, identifies with idealized characters on screen. Following Lacan, Baudry argues that we all possess the desire to return to early stages of our development, during which we experienced moments of unified completeness, though we are not consciously aware of this desire. Nevertheless, this desire explains the natural appeal of the film viewing experience. Watching films does more than entertain us; it fulfills an unconscious wish by triggering an artificial regression back to a state of complete narcissistic fusion, which Baudry says is characterized by “a lack of differentiation between the subject and his environment” (Baudry 1986b: 313). Identification, as characterized by Baudry, is a process that eliminates the boundaries between self and other, creating a type of all-encompassing fusion. The basis of this identification is an unconscious desire to regress that has far less to do with specific characters in a given film or features of an individual spectator’s psychology than with the ability of the cinematic apparatus to fulfill this preexisting desire. This is one of the features of film form that Baudry points to in order to show how film is ideological. 98
TF17557.indb 98
16/9/08 15:47:21
EM PAT H Y A ND C H A R AC T E R E NGAGEM E N T
Like Baudry, Christian Metz focuses on identification with the camera and likens this process to the mirror phase. Metz, however, emphasizes a key difference between the child’s relationship to the mirror and the spectator’s relationship to the screen. When the child gazes into the mirror, he sees himself reflected back. This does not occur in cinematic identification. The spectator is not part of what is perceived on screen. He is the perceiver but not the perceived. But he’s more than this. He is an all-perceiving subject such that the world on screen appears to be there for him, as created by him: At the cinema, it is always the other who is the screen; as for me, I am there to look at him. I take no part in the perceived; on the contrary, I am all-perceiving. All perceiving as one says all-powerful (this the famous gift of “ubiquity” the film makes its spectator); all-perceiving, too, because I am entirely on the side of the perceiving instance: absent from the screen, but certainly present in the auditorium, a great eye and ear without which the perceived would have no one to perceive it, the instance, in other words, which constitutes the cinema signifier (it is I who make the film). (Metz 1986: 252) The construction of the spectator as transcendental subject occurs through identification with the camera, which the spectator experiences as type of identification with the self as a pure act of perception. Where does this leave identification with characters? Films give us opportunities to identify with characters, on Metz’s view, but character identification is less central to his analysis. He writes that, “As for identifications with characters, with their own different levels (out-of-frame character, etc.), they are secondary, tertiary cinematic identifications, etc.; taken as a whole in opposition to the identification of the spectator with his own look, they constitute secondary cinematic identification” (Metz 1986: 259). One of the aims of Baudry’s and Metz’s respective discussions of primary cinematic identification with the camera and its gaze is to shed light on the ideological structure of the cinematic apparatus. The all-perceiving, meaning-making subject position is revealed as an illusion. The spectator is not in fact creating the images on screen nor is he beyond the limits of space, time, and the body, as the created perspective suggests. It seems this way, as a result of the inherently deceptive nature of the cinematic apparatus, which, as Baudry explains, “uses the system but also conceals it” (Baudry 1986a: 289). Feminist analysis – Laura Mulvey Feminist theorist Laura Mulvey examines the source of our pleasure in watching films and analyzes its implications. She complicates Baudry’s and Metz’s account of spectatorial identification by using psychoanalysis to show how film fosters processes of identification that perpetuate oppressive patriarchal ideals. 99
TF17557.indb 99
16/9/08 15:47:21
A M Y C OPL A N
Mulvey identifies a pattern in the cinematic conventions of standard Hollywood films. This pattern includes (1) the assumption of a male spectator; (2) the establishment and coding of a male protagonist as an active subject who controls the meaning of the narrative and of the female characters by being the “bearer of the look”; and (3) the establishment and coding of female characters as passive, weak objects of erotic desire, there to be “looked at” by the male characters and the male spectators (Mulvey 1986). This pattern exemplifies the “male gaze,” a way of seeing, thinking about, and acting in the world that takes women as passive, weak objects to be looked at by male subjects (Devereaux 2002). When a male spectator watches a standard Hollywood film, he identifies with the male protagonist who represents his ego ideal. Through this identification with the active controlling agent on screen, the spectator controls the unfolding of the narrative events and takes pleasure in looking voyeuristically at the female characters or fetishizing some parts of their bodies. This gives him the feeling of omnipotence and the pleasure associated with looking. Mulvey’s analysis aims to show that identification with male protagonists exploits unconscious desires and capitalizes on institutionalized sexism to generate pleasure. If Mulvey is right, then certain forms of character engagement, perhaps those that are most common, play an essential role in making film an instrument of oppressive ideological values. Mulvey’s original account of spectatorial identification has been challenged from all sides: feminist theory, psychoanalytic film theory, queer theory, and cognitive film theory. Feminist and queer theorists have charged that she exhibits covert heterosexism through her claim that the assumed spectator is a straight male and that she fails to address the female spectator, the role of race in representation, and the position of transgendered spectators (e.g., Creed 1998; White 1998; Friedberg 1990; Doane 1990). Many have also taken issue with Mulvey’s characterization of the relationship between spectators and characters as fixed and thus not subject to resistance or to subversive viewing practices (e.g., Friedberg 1990; Smith 1995; White 1998). Mulvey is also criticized for failing to explore alternative cinematic forms and effects. However one evaluates Mulvey’s particular account of character identification, it cannot be denied that she brought the topic of gender and oppressive practices from the periphery of film studies to its center. Prior to Mulvey’s discussion, little attention was paid to gendered aspects of filmmaking, film, and character engagement. This is no longer the case.
Cognitive film theory Cognitive film theory takes a very different approach to the question of character engagement. It appeals to research in empirical science and philosophy of mind to understand how engagement works, and concentrates not on the ideological influence of engagement but on how engagement explains viewers’ emotional and affective responses to film. Moreover, most film theorists and philosophers in this tradition assume that we can use what we know about how psychological processes work in regular life to understand how they will work when we view films. 100
TF17557.indb 100
16/9/08 15:47:21
EM PAT H Y A ND C H A R AC T E R E NGAGEM E N T
Identification Although theorists in both the psychoanalytic and cognitivist traditions of film theory talk about identification with characters, they understand the concept very differently. Within cognitive film theory alone, a variety of psychological processes get referred to as identificatory, including empathy, simulation, sympathy, and mirroring processes. Some theorists have argued that, due to its vagueness and ambiguity, the term “identification” should no longer be used to characterize spectatorial response (Allen 2003). Along the same lines, some favor replacing identification with a more precise concept or set of concepts (Smith 1995). This is not because these theorists don’t think something like identification occurs between viewers and characters but because they think that whatever this process is, it can be better explained some other way. Noël Carroll, by contrast, believes that identification, construed in any number of ways, is not a major part of character engagement and does little to explain spectatorial response (e.g., Carroll 2007, 2008: 161–77). Despite these concerns, Berys Gaut considers identification a worthwhile concept for explaining our relationships to characters. He defends its use on two separate grounds (Gaut 1999). First, ordinary film viewers use the term more than any other to describe their experience. If they like a film, it is because they identify with one or more characters. If they don’t like it, it is because they could not identify with any of the characters. For most spectators, films succeed or fail based on whether or not and to what extent they foster identification with characters (Gaut 1999: 200). Second, Gaut insists that the concept of identification is less confused than theorists suggest; it simply needs refining. To do this, Gaut explains that when we talk about identifying with a character, we don’t mean that we imagine being the character. Rather, we mean that we imagine being in the character’s situation. This is the general meaning of identification, but there are many different aspects to any situation, and thus there are many different ways in which we can identify with a character. In short, identification is aspectual. It involves neither a total replication of another’s experience, nor a loss of identity. Gaut is careful to point out that both the everyday use of identification by filmgoers and his use have little in common with the concept of identification employed by psychoanalytic film theorists. Gaut enumerates four aspects of identification: perceptual, affective, motivational, and epistemic. Thus we can imagine seeing what the character sees, feeling what the character feels, wanting what the character wants, or believing what the character believes (Gaut 1999: 205). Identification with one aspect of a character’s situation neither entails nor rules out identification with any other aspect; however, in some cases, identification of one kind encourages identification of another. Gaut defends the concept of identification by clarifying, not replacing or redefining, its meaning. He regards as important the fact that, unlike most of the explanatory concepts offered to describe character engagement, identification is a term used by ordinary filmgoers. His analysis suggests that the ordinary filmgoer may not be as confused as theorists have believed. 101
TF17557.indb 101
16/9/08 15:47:21
A M Y C OPL A N
The structure of sympathy – plural engagement Murray Smith critiques standard accounts of character identification for their failure to capture the complexity of spectatorial responses to characters. He argues that the notion of identification does little to clarify the relationship between spectators and characters, since identification and related terms get used in so many different ways. Identification accounts based on Lacanian psychoanalysis make the additional mistake of focusing too much on film’s visual appeal, at the expense of its capacity to involve spectators in imaginative and epistemological activities (Smith 1995). Smith’s alternative account of spectator engagement allows for multiple modes of character-spectator experience. Smith utilizes Richard Wollheim’s distinction between central and acentral imagining. Central imagining refers to imagining a situation “from the inside,” which is to say from a particular point of view. Acentral imagination, in contrast, refers to simply imagining that some situation is occurring without doing so from within the situation. According to Smith, acentral imagination plays a more important role in spectator engagement than central imagination. This goes against the standard identification accounts. Smith develops a model of spectatorial experience that comprises three distinct but related levels of engagement, a system he labels the “structure of sympathy.” Empathic phenomena make up a separate feature of Smith’s account that can work within or against the structure of sympathy. The three levels of engagement in the structure of sympathy, which are all associated with acentral imagination, are recognition, alignment, and allegiance. Recognition is Smith’s term for spectators’ construction of characters as individuated and continuous human agents. Alignment refers to the way a film narrative gives viewers access to characters’ thoughts, feelings, and actions; it is primarily about the communication of information. Allegiance describes the process by which film creates sympathies for and against characters. Smith’s model reveals how and why viewers experience plural identification, that is, how they can experience one level of engagement with a character but not another level, or different levels of engagement with a character during the course of a film, or different levels of engagement with different characters. Empathy Another relatively recent popular approach to character engagement focuses on empathy. Since empathy is believed to be one of the primary ways we emotionally engage with one another’s experience in ordinary life, and most everyone agrees that we emotionally engage with film characters’ experiences, theorists have reasoned that at least some of this engagement can be characterized as empathetic (see, e.g., Feagin 1996; Neill 1996; Currie 2004; Coplan 2004). One advantage empathy has over identification as an explanatory concept is that it refers to a psychological process that has been studied by empirical scientists (e.g., Decety and Sommerville 2003; Decety and Jackson 2004; Eisenberg and Strayer 1987; Hoffman 2000; Eisenberg 2000; 102
TF17557.indb 102
16/9/08 15:47:22
EM PAT H Y A ND C H A R AC T E R E NGAGEM E N T
Batson et al. 2007), and yet is also a concept ordinary film viewers frequently employ to describe their reactions to characters. Despite this advantage, accounts of spectatorial response that highlight empathy face many of the same difficulties that plague identification accounts, namely, multiple competing conceptualizations of empathy that refer to distinct psychological processes that vary, sometimes widely, in their function, phenomenology, mechanisms, and effects (see, e.g., Eisenberg 2000; Battaly forthcoming; Coplan forthcoming). Only if we employ a more precise conceptualization of empathy will we make progress in our attempts to understand it. I have proposed such a conceptualization, based on current psychological and neuroscientific research (Coplan forthcoming; see also Coplan 2004). Under my proposed conceptualization, empathizing with a character is a complex imaginative process through which a spectator simulates the character’s situated psychological states, including the character’s beliefs, emotions, and desires, by imaginatively experiencing the character’s experiences from the character’s point of view, while simultaneously maintaining clear self/other differentiation. If understood in this more precise way, empathy does not risk dissolving the boundaries between self and other, nor does it require that spectators have only the emotions of the characters with whom they empathize, though it does require that they have at least these (Coplan 2004). It also fits with the most recent empirical findings (e.g., Decety and Sommerville 2003; Decety and Jackson 2004; Eisenberg 2000). To empathize with a character, a spectator must accurately represent the character’s relevant psychological states to a greater or lesser degree, but she may also experience additional states as part of her own separate response. A separate response is made possible by her clear self/other differentiation. Empathy allows spectators to connect to characters while remaining separate from them. Spectators’ involvement in characters’ experiences in this case is deep, but it does not come at the expense of a separate identity, which means that the spectator can continue to have a wide range of psychological experiences that do not match those of the character (Coplan 2004: 147–9). Characterizations of spectatorial response as empathetic are common. We often talk about films allowing us to vicariously experience others’ lives through an empathic process, and we point to various techniques filmmakers use to enable and encourage this. In addition, empirical research on text processing and narrative comprehension provides some evidence, though it is tentative, for thinking that empathy is a regular part of how we relate to characters (Coplan 2004; Harris 2000). Simulation In the past fifteen years, many of the discussions of empathy in philosophical aesthetics and philosophy of film have been tied to discussions of simulation theory (e.g., Smith 1995; Feagin 1996; Neill 1996; Currie 1995, 2004; Harold 2000). Simulation theory emerged in the 1980s as an alternative to “theory theory,” the dominant view in philosophy and psychology of how we how we come to understand and predict others’ mental states, or “mindread.” According to theory theory, we do this appealing to our 103
TF17557.indb 103
16/9/08 15:47:22
A M Y C OPL A N
theories of folk psychology, which are made up of a set of law-like generalizations that connect various mental states with other mental states, with external circumstances, and with overt behavior. Simulation theorists explain things very differently. On their view, we attempt to determine what others think, feel, and desire by simulating their mental states, that is, attempting to adopt their perspective, using our own mind to model theirs under certain conditions (for recent formulations and defenses of simulation theory, see Currie and Ravenscroft 2002: 49–70; Currie 2004; and Goldman 2006; for a recent critique of simulation theory, see Stich and Nichols 2003: 131–42). To perform a successful simulation, it is not enough for us to experience the same emotions and thoughts as those of a target individual; we must come to have these emotions and thoughts through similar processes (Feagin 1996: 87–8). Soon after the emergence of simulation theory in philosophy of mind, philosophers and film theorists used it to explain our affective responses to art, including our affective responses to fictional characters. How is simulation related to empathy? In both philosophy of mind and aesthetics, discussions of empathy and simulation are closely connected. In fact more often than not, either the concepts are used interchangeably or simulation is offered as an explanation of empathy and how it works. Despite their connections, it may be better at this point to keep these two models of character engagement separate. Neither the term empathy nor the term simulation is used uniformly, and the relations between them are rarely entirely clear. If we use simulation to define empathy or empathy to understand simulation, we risk confusing rather than clarifying what’s happening during character engagement (for further discussion of the difficulty identifying simulation and empathy, see Goldie 1999; Nichols et al. 1996: 59–67; Stich and Nichols 1997: 299). Mirroring processes As the previous subsections show, cognitive film theorists and philosophers of film have focused primarily on modes of character engagement that involve high-level processing. This is in part a reaction to what they regard as an overemphasis in psychoanalytic film theory on unconscious processes and is in part a consequence of their theoretical frameworks. This pattern is beginning to shift, as a small number of cognitive film theorists and philosophers have begun to take seriously less sophisticated ways of responding to characters, such as motor mimicry, affective mimicry, low-level simulation, mirror reflexes, and emotional contagion (Smith 1995; Plantinga 1999, forthcoming; Choi 2003; Coplan 2006; Carroll 2007, 2008). Since each of these is an automatic process that causes mimicry or mirroring, it will be useful to collect them under the broad category of mirroring processes. Mirroring processes are a set of innate mechanisms and feedback systems that under certain conditions cause us to have experiences that are the same in kind as those of a person whom we observe. We therefore end up “mirroring” that person. These processes are largely automatic and involuntary. They occur through direct perception and involve neither the imagination nor high-level cognitive processing. As such, 104
TF17557.indb 104
16/9/08 15:47:22
EM PAT H Y A ND C H A R AC T E R E NGAGEM E N T
they represent a much more primitive mode of character engagement than the types normally discussed in cognitive film theory. Emotional contagion is one of the mirroring processes we know the most about. It’s also one that is especially important for character engagement. Psychologists Elaine Hatfield, John Cacioppo, and Richard Rapson define emotional contagion as the “tendency to automatically mimic and synchronize expressions, vocalizations, postures, and movements with those of another person, and, consequently, to converge emotionally” (Hatfield et al. 1992: 153–4). As Stephen Davies explains, this “involves the transmission from A to B of a given affect such that B’s affect is the same as A’s but does not take A’s state or any other thing as its emotional object” (Davies forthcoming). Due to our hardwired ability to mirror one another, our emotional experiences can be directly and immediately affected by others’ emotions and need not depend on any conscious evaluation or interpretation of the other’s emotions or external events. Given the nature of emotional contagion, film is especially well suited to produce it. Carl Plantinga explains this in his discussion of the “scene of empathy,” which refers to a scene in which the narrative slows down and a character’s emotional experience becomes the locus of attention. Such scenes, which are typically shot in close-up and focus on a character’s face, contain several eliciting conditions of emotional contagion. Most important is attention, since contagion responses require that a spectator’s attention be fixed on a character’s facial expressions. Filmmakers achieve this through the use of extreme close-ups, shallow focus, and various pointof-view structures and by using progressively closer shots of a character’s face and expressions (Plantinga 1999: 239–55). Although mirroring responses result from spectators’ perception of characters, on their own they cannot provide deep understanding of characters, nor can they generate empathy or sympathy. Observation of a character’s experience triggers the mirroring response, but the average film viewer will not be aware of this having happened. Nevertheless, these processes increase spectators’ level of bodily excitement and can intensify their emotional responses (Carroll 2007, 2008; Plantinga 1999, forthcoming). Since contagion responses can be triggered by direct sensory stimulation alone, they require neither involvement in a narrative nor investment in a character. This is one of the reasons why film often elicits emotional responses more easily than literature. Another interesting feature of emotional contagion is its ability to cause affective responses that are incongruent with a spectator’s overall emotional experience of a film (Smith 1995). Significantly, mirroring or mimicry responses are unique to our experience of film. Engagement with nondramatic literary fictions can produce empathy, sympathy, and emotions that result from cognitive evaluations. It cannot produce mirroring, since it does not appeal directly to readers’ perception or senses. This makes mirroring responses all the more important, since they are part of what distinguishes film in general and our responses to film characters in particular from literature and our responses to (nontheatrical) literary characters (Carroll 2008: 185–91; Plantinga forthcoming: especially chapter 5; Coplan 2006). 105
TF17557.indb 105
16/9/08 15:47:22
A M Y C OPL A N
Sympathy and criterial prefocusing The most serious challenge to identification accounts of character engagement, including those that emphasize the roles of empathy and simulation, has come from Noël Carroll, who has repeatedly charged that such accounts fail to explain the majority of our reactions to fictional film characters (Carroll 2007, 2008: 177–84). Carroll doesn’t deny that something akin to simulation, empathy, or identification can sometimes occur during the film viewing experiences, but he argues that when spectators’ emotions match those of the characters, it is typically due to criterial prefocusing (which I’ll explain below), not a process of identification or simulation (2008: 149–91). Carroll offers several arguments to support his position. First, he says that spectators’ emotions have different objects from those of the characters, and thus the emotions cannot be the same. This, he claims, is because we observe characters’ situations from outside the narrative. Second, spectators often have different information or more information than the characters, and so they have a different experience of the narrative events. Finally, spectators often have different desires or preferences from those of the characters and ones that can be in conflict with those of the characters. If empathy, simulation, or some other form of identification were a major part of our interaction with characters, there would be greater symmetry than occurs between the characters’ mental states and those of the spectators. On Carroll’s alternative view of character engagement, the majority of spectators’ emotional responses to film are caused by one of two processes: (1) criterial prefocusing of the narrative; or (2) sympathy with film characters. Criterial prefocusing is Carroll’s term for the process by which filmmakers foreground certain events and actions in the presentation of the narrative so that audience members will recognize them as fitting into familiar schemas that are likely to elicit an emotional reaction. Carroll accepts a version of the judgment theory of emotion for standard emotions. On this view, emotion requires a judgment or belief, which causes some feeling state. For example, if I believe that X has wronged me or mine, then I feel anger toward X. The relevant judgment in this case is that X wronged me or mine, which corresponds to the emotion anger (for an overview and critique of the judgment, or cognitive, theory of emotion, see Robinson 2005, forthcoming). Filmmakers attempt to ensure that viewers will respond emotionally by highlighting or focusing on actions, events, and character traits that fit the standard criteria for specific emotions. This predisposes us to respond emotionally, yet is not sufficient to evoke a response. The viewer must also care about the relevant characters or the outcome of the story. Without this, viewers may recognize how the film’s narrative events fit certain criteria but experience them dispassionately. Sympathy, then, is essential for responding to characters, but it is even more than this. According to Carroll, sympathy is the “major emotional bond between audiences and leading fictional characters” (Carroll 2007: 103). Carroll uses the term “sympathy” in roughly the same way it is used by contemporary psychologists, namely, to refer to an other-oriented emotional process involving feelings of care and concern 106
TF17557.indb 106
16/9/08 15:47:22
EM PAT H Y A ND C H A R AC T E R E NGAGEM E N T
for another and the other’s well-being (Carroll 2007: 101; Eisenberg 2000; Wispe 1991). Sympathy is not a synonym for empathy, nor is it a type of identification; when we sympathize with another, we feel for him or her, not with him or her. Carroll’s discussion of sympathy emphasizes that there is both a close relationship and an asymmetry between the feelings of the spectator and those of the character. We may occasionally experience the same feelings as a character, but when this happens, Carroll says, it is typically the result of criterial prefocusing, not a process of identification. What reasons do we have for thinking sympathy is the primary mode of character engagement? Carroll claims that we respond sympathetically to the film’s protagonists more than in any other way, and sympathy provides the “emotive optic” through which we survey the narrative events. In other words, we evaluate and respond to various characters, actions, and narrative events on the basis of our sympathy for a film’s protagonists. As a result, the mistreatment of the protagonist will arouse our anger but only because of our sympathy. If we didn’t care about the protagonist, we wouldn’t care about his or her mistreatment. Filmmakers establish sympathetic bonds, according to Carroll, by depicting protagonists as representative of values shared by the audience. The narrative reveals that the protagonists share some interest, project, or loyalty with the audience members. Another determining variable is moral character. Provided that a character, even an antihero, is morally praiseworthy in at least one significant respect, that character can arouse our sympathy (Carroll 2008: 181–2).
Effects of character engagement I have now discussed some of the leading explanations of character engagement. The final section of this essay will briefly consider some of its effects. Psychoanalytic film theorists, such as Baudry and Metz, who characterize spectatorial response in terms of identification, have tended to focus on the harmful effects of engagement. Identification draws spectators into an illusion in which they feel more powerful and important than they actually are. The social and political implications go beyond this. If Mulvey is right, then identification with male protagonists in standard Hollywood films leads to the adoption of the male gaze and the pleasurable experience of males as subjects and females as objects. Even though female characters are coded as weak and passive, they nevertheless represent a threat to male characters and male viewers. To neutralize this threat, standard Hollywood narratives foster either a sadistic voyeurism or the fetishizing of the female form, neither one of which recognizes women as subjects (Mulvey 1986). Standard Hollywood narratives punish female sexuality so that female characters who fail to fit stereotypical norms cannot survive and must be brought back into the dominant patriarchal structure. Alternatively, the threat female characters pose can be eliminated by reducing them to some body part and then fetishizing it. Most cognitive film theorists and analytic philosophers of mind have explored the beneficial effects of character engagement. By increasing the stakes of the narrative 107
TF17557.indb 107
16/9/08 15:47:22
A M Y C OPL A N
events, character engagement heightens dramatic conflict. Certain forms of character engagement may do more than enhance aesthetic appreciation and enjoyment. Empathy and simulation can improve and expand our understanding of ourselves and others, including others whose experiences are very different from our own (Smith 1995, 1997; Neill 1996). The beneficial effects of these sorts of character engagement are significant, but we need to be careful not to overstate their transformative power. Empirical research in social and developmental psychology shows that people are both more willing and better able to empathize with those whom they perceive to be like them in some significant respect; the more similar they in fact are, the more likely the attempt to empathize will succeed (Hoffman 2000). Dan Flory has carefully examined aspects of the relationship between race and film viewing and has shown that white viewers are often incapable of successfully empathizing with African American characters. Due to their ignorance of the life circumstances of nonwhite characters, they fail to appreciate those characters’ situations and fail to deeply engage with them (Flory 2006). Beyond this, the assumption that empathy will necessarily lead to altruistic or even prosocial behavior is flawed (Goldie 2000; Battaly 2008; Coplan forthcoming), and contrary to popular understanding there is no causal connection between empathy and sympathy (Eisenberg and Miller 1987). Heather Battaly’s recent discussion of empathy clarifies this by classifying empathy as a skill rather than a virtue (Battaly forthcoming). Undoubtedly it is a skill with the potential to greatly impact the outcome of moral deliberation and action, but it cannot and should not be assumed that this is the only way or even the most common way it will be employed. Keeping all of this in mind, engagement with characters still has the potential to benefit viewers in many ways. For example, through empathy or simulation, when it succeeds, we can gain a unique kind of experiential understanding of characters. While this is not tantamount to prosocial behavior, it’s something much more than working out the features of the plot. It provides a representation, however partial, of another person’s subjective experience – one that is not a representation of causes and effects but rather is a representation of what it is like to be another person. See also Acting (Chapter 1), Emotion and affect (Chapter 8), Spectatorship (Chapter 23), Bertolt Brecht (Chapter 30), Noël Carroll (Chapter 31), Cognitive theory (Chapter 33), Gender (Chapter 13), Narrative closure (Chapter 19), Race (Chapter 21), Christian Metz (Chapter 36), and Psychoanalysis (Chapter 41).
References Allen, R. (2003) “Identification in the Cinema: A Conceptual Investigation,” unpublished manuscript, New York: New York University. Batson, C. D., Håkansson, E., Chermok, V. L., Hoyt, J. L., and Ortiz, B. G. (2007) “An Additional Antecedent of Empathic Concern: Valuing the Welfare of the Person in Need,” Journal of Personality and Social Psychology 93: 65–74. Battaly, H. (forthcoming) “Empathy: Virtue or Skill?,” in A. Coplan and P. Goldie (eds.) Empathy: Philosophical and Psychological Approaches, Oxford: Oxford University Press.
108
TF17557.indb 108
16/9/08 15:47:22
EM PAT H Y A ND C H A R AC T E R E NGAGEM E N T
Baudry, J.-L. (1986a) “Ideological Effects of the Basic Cinematographic Apparatus,” in P. Rosen (ed.) Narrative, Apparatus, Ideology, New York: Columbia University Press. —— (1986b) “The Apparatus: Metaphysical Approaches to Ideology,” in P. Rosen (ed.) Narrative, Apparatus, Ideology, New York: Columbia University Press. Carroll, N. (2007) “On the Ties That Bind: Characters, the Emotions, and Popular Fictions,” in W. Irwin and J. Garcia (eds.) Philosophy and the Interpretation of Popular Culture, Lanham, MD: Rowman & Littlefield. —— (2008) The Philosophy of Motion Pictures, Malden, MA: Wiley-Blackwell. Choi, J. (2003) “Fits and Startles: Cognitivism Revisited,” Journal of Aesthetics and Art Criticism 61: 149–57. Coplan, A. (2004) “Empathic Engagement with Narrative Fictions,” Journal of Aesthetics and Art Criticism 62: 141–52. —— (2006) “Catching Characters’ Emotions,” Film Studies: An International Review 8: 26–38. —— (forthcoming) “Understanding Empathy: Its Features and Effects,” in A. Coplan and P. Goldie (eds.) Empathy: Philosophical and Psychological Approaches, Oxford: Oxford University Press. Creed, B. (1998) “Film and Psychoanalysis,” in J. Hill and P. Church Gibson (eds.) The Oxford Guide to Film Studies, New York: Oxford University Press. Currie, G. (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge: Cambridge University Press. —— (2004) Arts and Minds, New York: Oxford University Press. Currie, G., and Ravenscroft, I. (2002) Recreative Minds: Imagination in Philosophy and Psychology, Oxford: Clarendon Press. Davies, S. (forthcoming) “Music Listener Emotional Contagion,” in A. Coplan and P. Goldie (eds.) Empathy: Philosophical and Psychological Approaches, Oxford: Oxford University Press. Decety, J., and Jackson, P. (2004) “The Functional Architecture of Human Empathy,” Behavioral and Cognitive Neuroscience Reviews 3: 71–100. Decety, J., and Sommerville, J. (2003) “Shared Representations between Self and Other: A Social Cognitive Neuroscience View,” Trends in Cognitive Science 7: 527–33. Devereaux, M. (2002) “Oppressive Texts, Resisting Readers, and the Gendered Spectator: The ‘New’ Aesthetics,” in A. Neill and A. Ridley (eds.) Arguing about Art, 2nd ed., New York: Routledge. Doane, M. (1990) “Remembering Women: Psychical and Historical Constructions in Film Theory,” in E. A. Kaplan (ed.) Psychoanalysis and the Cinema, New York: Routledge. Eisenberg, N. (2000) “Empathy and Sympathy,” in M. Lewis and J. Haviland-Jones (eds.) Handbook of Emotions, New York: Guilford. Eisenberg, N., and Miller, P. (1987) “Empathy, Sympathy, and Altruism: Empirical and Conceptual Links,” in N. Eisenberg and J. Stayer (eds.) Empathy and Its Development, Cambridge: Cambridge University Press. Eisenberg, N., and Strayer, J. (eds.) (1987) Empathy and Its Development. Cambridge: Cambridge University Press. Feagin, S. (1996) Reading with Feeling: The Aesthetics of Appreciation, Ithaca, NY: Cornell University Press. Flory, D. (2006) “Spike Lee and the Sympathetic Racist,” Journal of Aesthetics and Art Criticism 64: 67–79. Friedberg, B. (1990) “A Denial of Difference: Theories of Cinematic Identification,” in E. A. Kaplan (ed.) Psychoanalysis and the Cinema, New York: Routledge. Gaut, B. (1999) “Identification and Emotion in Narrative Film,” in C. Plantinga and G. M. Smith (eds.) Passionate Views: Film, Cognition, and Emotion, Baltimore, MD: Johns Hopkins University Press. Goldie, P. (1999) “How We Think of Others’ Emotions,” Mind and Language 14: 394–423. —— (2000) The Emotions: A Philosophical Exploration, Oxford: Oxford University Press. Goldman, A. (2006) Simulating Minds: The Philosophy, Psychology, and Neuroscience of Mindreading, New York: Oxford University Press. Harold, J. (2000) “Empathy with Fictions,” British Journal of Aesthetics 40: 340–55. Harris, P. (2000) The Work of the Imagination, Oxford: Blackwell Publishers. Hatfield, E., Cacioppo, J. T., and Rapson, R. L. (1992) Emotional Contagion, Cambridge: Cambridge University Press.
109
TF17557.indb 109
16/9/08 15:47:22
A M Y C OPL A N
Hoffman, M. (2000) Empathy and Moral Development, New York: Cambridge University Press. Metz, C. (1986) “The Imaginary Signifier” (excerpts) in P. Rosen (ed.) Narrative, Apparatus, Ideology, New York: Columbia University Press. Mulvey, L. (1986) “Visual Pleasure and Narrative Cinema,” in P. Rosen (ed.) Narrative, Apparatus, Ideology, New York: Columbia University Press. Neill, A. (1996) “Empathy and (Film) Fiction,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Nichols, S., Stich, S., Leslie, A., and Klein, D. (1996) “Varieties of Off-line Simulation,” in P. Carruthers and P. Smith (eds.) Theories of Theories of Mind, Cambridge: Cambridge University Press, 39–74. Nichols, S., and Stich, S. (2003) Mindreading: An Integrated Account of Pretense, Self-Awareness, and Understanding Other Minds, Oxford: Oxford University Press. Plantinga, C. (1999) “The Scene of Empathy and the Human Face in Film,” in C. Plantinga and G. M. Smith (eds.) Passionate Views: Film, Cognition, and Emotion, Baltimore, MD: Johns Hopkins University Press. —— (forthcoming) Moving Viewers: American Film and the Spectator’s Experience, Berkeley: University of California Press. Robinson, J. (2005) Deeper than Reason: Emotion and its Role in Music, Literature, and Art, Oxford: Oxford University Press. —— (forthcoming) “Emotion,” in J. Prinz (ed.) The Handbook of Philosophy of Psychology, Oxford: Oxford University Press. Smith, M. (1995) Engaging Characters: Fiction, Emotion, and the Cinema, New York: Oxford University Press. —— (1997) “Imagining from the Inside,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, New York: Oxford University Press. Stich, S., and Nichols, S. (1997) “Cognitive Penetrability, Rationality, and Restricted Simulation,” Mind and Language 12: 297–326. White, P. (1998) “Feminism and Film,” in J. Hill and P. Church Gibson (eds.) The Oxford Guide to Film Studies, New York: Oxford University Press. Wispe, L. (1991) The Psychology of Sympathy, New York: Plenum Press.
Further reading Barbara Creed, The Monstrous Feminine: Film, Feminism, Psychoanalysis (London: Routledge, 1993) offers a book-length treatment of the issue of how women are portrayed in horror films, based on psychoanalytic feminist theory. An edited collection on the topic of simulation theory is M. Davies and T. Stone (eds.) Mental Simulation (Oxford: Blackwell, 1995). Berys Gaut, “Film,” in J. Levinson (ed.) The Oxford Handbook of Aesthetics (Oxford: Oxford University Press, 2003) is a survey article of key issues in philosophy of film, including spectatorial response. Jerrrold Levinson, “Emotion in Response to Art: A Survey of the Terrain,” in Mette Hjort and Sue Laver (eds.) Emotion and the Arts (New York: Oxford University Press, 1997) is an overview article covering recent philosophical issues on the relationship between emotion and art. Aaron Meskin and Jonathan Weinberg, “Emotions, Fiction, and Cognitive Architecture,” British Journal of Aesthetics 43 (2003): 18–34, offers a critique of simulation accounts of responding to fiction. Carl Plantinga, “Disgusted at the Movies,” Film Studies: An International Review 8 (2006): 81–92, explores the topic of film-elicited disgust as a direct emotion, not an emotion triggered by spectators’ relationship and response to characters.
110
TF17557.indb 110
16/9/08 15:47:22
10
E TH IC S Folke Tersman I know no great work of art in all of world culture that would not be linked to an ethical ideal, that is based on some other motives such as on the dark aspects of life. There are some talented works of such a nature, but no masterpieces. Andrei Tarkovsky (quoted in Lipkov 1988; translated by Robert Bird) Many films address ethical questions and morally loaded situations, and this is part of the reason why we find them interesting. Films occasionally also prompt moral criticism. Some of such criticism focuses on events in the creation of the film. Consider Herzog’s masterpiece Fitzcarraldo (1982). The main character – the Irishman Fitzgerald (who changed his name to Fitzcarraldo, when he reached South America toward the end of the nineteenth century) – wanted to build an opera house in the jungle town of Iquitos and planned to finance this endeavor by making a fortune from rubber. To transport the rubber, he decided to move a ship from one part of the Amazon River to another by crossing a large hill in between. The film is based on a true story, and the real Fitzcarraldo had the ship dismantled and carried over the hill in pieces. Herzog, however, decided to reenact the event by pulling a giant, three-story boat over a large hill, using an ancient pulley system and the local Indians, and the image of these Indians straining to pull a gigantic boat over the jungle hill became the central one in the film. This project involved huge risks, as was pointed out to Herzog many times in advance. And what people feared did in fact happen. A cable snapped, sending the huge boat tumbling down back to the bottom of the hill. Miraculously, none of the poorly paid Indians died, although three were injured. However, the mere fact that the crew was put to such immense risks raises, of course, ethical questions. From a moral point of view, there is nothing special about actions and decisions that lead to an artwork. Like other, more domestic, activities, they can be evaluated morally, and the considerations that are relevant to such evaluations are the same as in other cases: the consequences of the decisions, the underlying intentions, whether they represent breaches of important moral rules, and so forth. Of course, some people might want to say that artists should be allowed a certain freedom that is not acceptable for ordinary folks (quod licet lovi, non licet bovi), at least when they act qua artists. But, if this is reasonable (which I doubt), it can be accommodated by pointing
TF17557.indb 111
16/9/08 15:47:22
F OLK E T E R SM A N
to the special circumstances involved, or to the special role that these people have. Just as teachers may have special obligations as a result of their role and responsibilities as teachers, there may be special rules applying to artists. This is compatible with thinking that, at a more general level, the same rules and principles apply to all agents, in the sense that it holds for everyone that if she were a teacher, or an artist, the special rules associated with those roles would apply to her as well. In what follows, however, I shall largely ignore the kind of questions raised by filmmaking. Instead, I shall focus on questions regarding the outcomes of such activities, the films themselves. Can they too be evaluated from a moral point of view? The answer to this question may seem to depend on our view on the nature of a film when conceived as an artwork. On one view, what we are referring to is simply an “audio-visual display,” detached from any connection with the manner in which it was produced and the agents who produced it. This does not rule out moral evaluation, but perhaps it limits the types of evaluations that can be made. Thus, we can presumably ask whether it is, from a moral point of view, a good thing that it exists, and whether we should encourage people to see it. But there are other moral properties (such as rightness or blameworthiness) that it makes less sense to ascribe. Only persons and actions can have certain moral properties. For example, consider a film that propagates racism and invites people to share that attitude. This is perhaps a reason to censor it, but not to consider it (as contrasted with the filmmaker or distributor) blameworthy. However, on another view about the nature of an artwork, a film is not merely a bare, detached audio-visual display. Rather, it should be understood in relation to facts about the context of production, such as facts about the underlying intentions. Indeed, such considerations are constitutive of the very difference between a work and a type of display of lights and sounds. On this, more “contextualist,” view, a work of art is an accomplishment, something somebody does (Levinson 1990; and Wollheim 1980; the view is also elaborated in Livingston 2005). This conception allows for more complex ethical evaluations of the film, evaluations informed also by facts about the process that led to the film, which slightly blurs the distinction between ethical criticism of films and of filmmaking. In what follows, I shall consider various types of ethical criticism and praise that a film can be subjected to. This is not to say that the idea that it is appropriate to assess artworks from an ethical perspective is uncontroversial. On the contrary, it is often held that ethical criticism of art involves a kind of radical error, which is illustrated by a famous quote about literature by Oscar Wilde: “There is no such thing as a moral or an immoral book. Books are well written, or badly written. That is all.” The plausibility of this attitude, however, is easier to assess once one has considered what ethical criticism could consist in. If one grants that a film can have moral qualities, further questions arise. How do the supposed moral flaws or merits of a film interact with its artistic and aesthetic value? Can a moral flaw make a film worse, artistically or aesthetically speaking? If so, does a moral flaw always detract from the artistic value of a film, or could a moral flaw sometimes even enhance it? A number of different positions addressing such issues 112
TF17557.indb 112
16/9/08 15:47:22
ET H IC S
have emerged. I shall survey some of those positions, as well as some of the arguments used for or against them. However, I shall also argue that what view we take on this issue does not necessarily settle the question of whether it may be appropriate for critics or scholars to engage in ethical criticism of film. Successful interpretation of a narrative such as a film often presupposes recognition of the moral virtues and vices of the characters in the film, and of the attitudes projected in the work toward these characters. It is probably difficult for a spectator to do this without engaging in a dialogue with the values thus projected, and to respond by forming ethical judgments. Whether the moral qualities of the film that may thus be discerned affect its aesthetic value or not seems to have little bearing on whether it might be worthwhile to communicate those judgments. At the end of the article, I shall also make some brief comments about why moral philosophers and others who engage in moral thinking might find films a useful tool in their explorations.
Can films be immoral? Ethical criticism of art is as old as art itself, and in the history of philosophy it goes back at least to Plato. However, more recently, this attitude has come into disrepute. Indeed, I suspect that many think that talk of “moral qualities” of films has a slightly ridiculous ring to it. In the 1960s and 1970s, left-wing people in Sweden criticized literature and other cultural expressions as being reactionary, and as consolidating the allegedly unjust capitalist system of the day. One of the people who thus came under fire was the children’s writer Astrid Lindgren. Lindgren is well known for the books and films about the girl Pippi Longstocking (“the strongest girl in the world”), who lived by herself in a big house (her mother had died and her father was far away), entirely self-sufficient and independent because of her strength, cleverness, and a huge chest of gold coins. According to the critics, these stories endorsed a fascist ideology that praises the strong and independent and loathes the weak. Today, this story is cited as an example of how incredibly silly and hysterical the radicals of the generation of 1968 could be. Of course, there are more worrying examples of ethical criticism. These are cases where authoritarian regimes have banned books and films because they propagate the “wrong” values and subversive ideologies. It is probably association with those types of cases and a commitment to liberal values, such as free speech and freedom of expression, that underlies much of the skepticism toward the idea of ethical criticism of artworks. However, it is important to notice that by allowing for the appropriateness of such criticism one commits oneself neither to any particular moral views (such as the upper- and middle-class Victorian values that Wilde scorned or the authoritarian views that Plato is associated with) nor to the view that artworks should be censored or banned. One may well believe that there are good reasons for allowing freedom of expression and still think that some artworks have moral defects, just as one may hold that adultery is often morally problematic but that it should not be considered a legal offense. As John Stuart Mill has famously argued, there are good consequentialist 113
TF17557.indb 113
16/9/08 15:47:22
F OLK E T E R SM A N
arguments for allowing freedom of speech that do not presuppose any assumptions about the moral status of the speech thus being allowed. If one does think that it may be appropriate to evaluate artworks morally, what could provide grounds for such evaluations? There are many possibilities. If a work records real acts of torture and violence, it might be argued that the immorality of these acts is inherited by the work itself. A work can also be held to be morally defective by the simple fact that it upsets people, such as when it mocks certain religious beliefs. A criticism that needs more complex argumentation is when it is claimed that the work will encourage certain undesirable and antisocial types of behavior. Today, this type of criticism is probably more often launched against films than other cultural expressions, given the impact they currently have in society. The idea is that violence in films will lead to violence in the streets, if it is depicted in certain (glorifying) ways, and that similar things can be said about other types of undesirable phenomena (male dominance, racism, and so on). Arguments of this kind are in a wide sense consequentialist, but they do not rely on the assumptions about what makes an outcome better or desirable, usually associated with consequentialism. For example, by advocating such arguments, one does not commit oneself to the utilitarian view that the desirability of an outcome depends on how much “happiness” it contains. However, an obvious problem with these arguments is that the truth of the empirical claims on which they rely – the claim that films will indeed affect people’s attitudes and behavior in the way assumed – is very hard to determine. Indeed, their truth is even difficult to research, as people’s attitudes and behavior are the complex result of many different factors. What reason do we have for thinking that it is the films that provide the (or a) crucial factor? The genre that has been the object of most research of this kind is probably pornography, as there are worries that consumption of pornography will lead to sexual violence and consolidate gender inequality. But the evidence that has been gathered through this research seems far from conclusive (for a survey of some research, see Donnerstein et al. 1987). However, there is yet another type of ethical criticism, a type that does not rely on any far-reaching predictions about how films might affect general behavior. On this view, in order for a film to be morally defective it is not necessary for it to actually succeed in implanting morally flawed attitudes or perspectives in its audiences, at least not in any extensive way. It is enough if it endorses those attitudes. To use the word “endorses” in this context might seem awkward, as we usually believe that only agents or thinkers can endorse. However, given the wider concept of a film, mentioned above, the one that sees it as an accomplishment rather than a “bare audio-visual display,” this objection has less force. The idea is that, given facts about the film and the making of it (the genre it belongs to, the creator’s intentions, etc.), it can be seen as prescribing certain responses for the audience, such as feelings of sympathy or antipathy for certain characters. If these responses are morally problematic, that could be seen as a flaw in the film (such a view is defended in, e.g., Carroll 2000; and Stecker 2007). At least, this is so, it seems to me, if it is designed so as to secure or merit those responses – i.e., so that the formation of those responses is in a sense explicable or reasonable, given features of the film. If it does not have such a design, it might be 114
TF17557.indb 114
16/9/08 15:47:22
ET H IC S
more reasonable to think that the flaw lies (solely) with the filmmaker and not with the film. Although this type of criticism does not rely on problematic empirical claims of the kind that the first type does, it raises other methodological issues, regarding interpretation. How can we determine that a film endorses a certain perspective? After all, we must distinguish between endorsing and merely depicting or representing or perhaps exploring a certain attitude (while remaining neutral toward it). The latter need not be a ground for thinking that the film is morally flawed. On the contrary, it might be considered a merit, as it increases an audience’s understanding of the moral issues and attitudes addressed. That is, a film may have features that superficially seem to indicate that it invites audiences to share a certain flawed perspective. However, a closer look that takes more factors into account may undermine our first impression. For example, consider the action comedy Ocean’s Eleven (2001), in which the character Danny Ocean seeks revenge on an enemy, who previously has put him in jail and is courting his girlfriend, by robbing his casino. Given the sympathetic portrayal of the hero and his collaborators, the film might seem to endorse criminality. However, one must also take into account the special conventions that govern films belonging to the pertinent genre. In action films, people are hit in their heads and get thrown through glass windows without getting seriously hurt. This indicates that some of the worries that apply to the real world should not be applied to the fictional world that is portrayed. Indeed, in these films, we often see very little of the suffering and misery of the hero’s victims. Those characters remain two-dimensional cut-outs, which sends the message that the spectator is expected to suspend some of her moral principles when judging what happens to them. It seems to me that judgments to the effect that a film endorses immoral perspectives are, for such reasons, often premature (see Davies 2006, for a discussion of the relevance of conventions of the kind discussed here). If encouraging or endorsing ethically problematic attitudes and perspectives can be a moral flaw in a film, encouraging or endorsing admirable ones can presumably be a merit. Such judgments obviously face the same problems as those pertaining to their “negative” counterparts. However, more recently, much discussion has focused on another possible ground for valuing films and other artworks from a moral (or at least moral philosophical) point of view. And although the type of value that is thus ascribed to films is instrumental, so that the films are valued for the help they provide to some external aim (moral development), it does not rely on the kind of ambitious empirical claims that the other consequentialist arguments rely on. These suggestions are often made in connection with the relatively recent renaissance of virtue-ethical and Aristotelian approaches to moral philosophy (see, e.g., Hursthouse 1999; and Slote 1992). Such approaches are suspicious toward attempts to codify moral obligations in general principles (such as utilitarianism) and stress that the basic moral questions concern agents and their characters rather than actions. More specifically, they stress the importance of developing and fostering certain stable character traits – virtues. These are traits that allow the agent to register and acknowledge the importance of ethically salient features of a situation and to act 115
TF17557.indb 115
16/9/08 15:47:22
F OLK E T E R SM A N
accordingly. That is, to acquire these traits is to develop a certain sensitivity, which allows one to perceive aspects of the situation that are morally relevant. It has been held that the appreciation of narrative art, such as literature and film, may have a crucial role for the development of such a sensitivity, in that they allow for representations of the complexity of the situations that raise moral questions and invite audiences to see these situations from different perspectives, by allowing them to “simulate” how it would be to be in such positions (see Currie 1995; Murdoch 1970; and Nussbaum 1990). It should be noted, however, that although this type of reasoning is usually used by virtue theorists, other theorists can adopt it as well. All viable moral theories stress that responsible action requires something like empathy and the ability to put oneself in the shoes of others to gather relevant information.
Moral and artistic value If we grant that it is meaningful to say that films have moral qualities of the kind exemplified above, we might want to know how these qualities interact with their artistic value. A number of different positions addressing that issue have emerged. These positions are often stated in a way that presupposes a distinction between artistic and aesthetic value. The aesthetic value of an artwork focuses on just some of its features – its gracefulness, beauty, unity, expressive subtlety, and on – whereas assessments of artistic value are all-things-considered judgments that take other evaluative dimensions into account, such as, perhaps, its cognitive value (what we can learn from it), or art-historic value. That is, artistic value is conceived as a “composite” value that includes, in addition to aesthetic value, other types of values as well (Dean 2002; Stecker 2005, 2007). Of course, it is possible to use “aesthetic value” in a wider sense, so that it includes the other dimensions as well, but I shall in what follows accept this terminology. How do the moral flaws and merits of an artwork affect its artistic value? This is the question that I take the positions to address. The idea is that although affecting the aesthetic value of a work would be one of the ways a moral flaw or virtue may alter its artistic value, it is at least conceivable there are other ways. There are many possible views about the relationship between an artwork’s moral qualities and its artistic value. The two extremes are “ethicism” and “autonomism.” Ethicism is the view that moral defects or virtues always have an artistic significance, so that an ethical defect is always an artistic flaw (Gaut 1998). Autonomism, by contrast, denies that they have any artistic significance whatsoever (Isenberg 1973). Both these positions are incompatible with the “anti-theoretical” one that denies that there is a systematic relationship of any kind between artistic and ethical value. On that view, an ethical flaw can sometimes be an artistic defect, sometimes be insignificant, and can sometimes even be an artistic virtue (for some recent proponents, see Jacobson 1997; and John 2006). The position that allows for the latter possibility – that an ethical flaw may be an artistic virtue – is sometimes called “immoralism” (Kieran 2002). Another position that has recently received some attention is “moderate moralism” (Carroll 2000). Moderate moralism holds only that some ethical flaws are artistic flaws, and that some ethical virtues are artistic virtues. As it stands, 116
TF17557.indb 116
16/9/08 15:47:22
ET H IC S
it is possible to hold this view and also to think that ethical flaws can sometimes be artistic virtues. So, moderate moralism is compatible with the anti-theoretical view. The assessment of these positions will depend on how we understand the concept of “artistic value.” On one view, the idea that assessments of artistic value are all-thingsconsidered judgments is to be taken literally. On this view, all possible grounds for evaluation should be taken into account. So, if artworks allow for ethical evaluation, it becomes trivial to say that moral flaws or virtues can affect artistic value (see, e.g., Dean 2002: 2). However, it is possible to adopt a narrower view of the concept of artistic value and to restrict the grounds and considerations that are relevant. For example, some insist that to make an assessment of the artistic value of an artwork is to assess it from a certain perspective, i.e., to assess it as art and not relative to some other role or function it may have, such as when it is viewed as an educational tool (John 2006). On such a view, whether a moral flaw can be an artistic flaw remains an open question, although one might wonder, of course, what it means to say that an artwork is evaluated as art. Anyway, the cases that have prompted most controversy are those where ethical flaws are supposed to affect the artistic value of a work, not directly (as ethical value is supposed to be built into the very concept of artistic value), but by affecting something else, such as its aesthetic value. It is on such cases I shall focus. To adjudicate between ethicism and these other positions we also need to say something about what it means for a moral flaw to “affect,” “enhance,” or “detract” from the artistic value of a work. Thus, suppose that the filmmaker, in order to secure a response from an audience, makes certain moves that detract from a film’s artistic value, for example by simplifying the story and its characters. Then, if the response is questionable from a moral point of view, this seems to illustrate how a moral flaw can affect the artistic value of a film. However, if the response could have been secured without taking such measures, the relationship between the moral flaw and the artistic one is merely contingent. Presumably, that moral flaws could affect artistic value in this sense is not anything controversial. So, if we want to find a meaningful topic for the debate, we must assume that the alleged relationship between moral and artistic value is stronger than that. How could one try to establish that there is a more direct connection? One argument for ethicism takes as its point of departure the idea that if a film is not designed so as to mandate or justify the responses it prescribes, this represents an artistic flaw. For example, if a comedy is not funny, or a thriller is not thrilling, then the film has failed on its own terms and is thereby artistically flawed. Now, if a film is immoral in virtue of endorsing an immoral attitude, it prescribes that we should come to share that attitude. However, as the attitude is immoral, it can never be justified. Therefore, a moral flaw is also (always) an artistic flaw. This is called “the merited response argument” (Gaut 1998: 195; Carroll 2000: 375). An obvious problem with this argument is that it exploits an ambiguity in the term “justified.” One sense in which something could be justified is that it is morally acceptable (as when we say that violence in self-defense is justified). However, another sense is that it makes sense from a rational point of view (as when we say that 117
TF17557.indb 117
16/9/08 15:47:22
F OLK E T E R SM A N
a belief is justified by the available evidence). It might be argued that it is only when the responses fail to be warranted in the second sense that we should conclude that it has failed on its own terms and thus that the failure is an aesthetic flaw. So, if the fact that a response is not morally justified does not exclude its being justified in the second sense, the argument fails. Another, but related, argument has been proposed by Noël Carroll (although he takes his argument to support moderate moralism, rather than ethicism). Carroll imagines a novel that invites audiences to respond by admiring a sadistic colonizer who “tortures every Indian he encounters” as he conceives of them as “vermin,” an attitude with which the novel concurs. This makes the work morally defective. However, even if such a novel is skillfully designed, audiences will normally be unable to form the feelings of admiration toward the colonizer invited by the work, simply because we are disinclined to have admiration for such characters. In other words, the character fails or even contravenes “the morally relevant criteria for admiration.” This makes the moral flaw an artistic one as well, as it represents “a failure to design the character in accordance with the warranting conditions for the emotional response prescribed.” Or as Carroll puts it, “[T]he reason the work is aesthetically defective is that it is morally defective. Though the work prescribes admiration as part of its aesthetic agenda, it fails to secure it, precisely because it is evil” (Carroll 2000: 377–8). Like the argument for ethicism, this argument relies on the idea that a film is artistically flawed if it is not designed so as to mandate or secure the responses it prescribes. In this case, however, the question is not whether the responses are morally justified, but whether we can be justified in thinking that spectators will actually form those responses. The reason why the book about the colonizer does not satisfy this condition is simply that it prescribes responses (e.g., admiration for racist cruelty) that people with normal moral sensibilities are unable to form. And what explains why they are unable to form these responses is, exactly, that they are immoral. Notice that I have now construed the argument as relying on claims about what responses audiences are in fact able to form. This means that, by accepting the argument, we commit ourselves to the idea that if people were to deteriorate morally (so that they would be able to admire cruelty), the artistic value of the work would, other things being equal, increase. This implication may seem a bit odd. One way to avoid it is to take the question to be whether spectators would be able to form the responses insofar as they are morally sensitive. But to count failure to merit prescribed responses in this sense as an artistic failure is to beg the question against the autonomist. Another possible objection to Carroll’s argument relies on the suggestion made earlier, that when an artwork fails to secure or mandate the responses it prescribes the morally problematic character of those responses is more reasonably seen as a flaw of the artist behind the work, rather than of the work itself. Can ethical flaws sometimes enhance the artistic value of a film? The fact that a film depicts a certain immoral perspective in a way that makes it intelligible and psychologically credible can presumably enhance the artistic value, for example by increasing an audience’s understanding of the perspective. However, that it depicts such a perspective (while remaining neutral toward it) is not enough to make the film 118
TF17557.indb 118
16/9/08 15:47:23
ET H IC S
morally flawed. It must also endorse it. How could such endorsement contribute to the artistic value of the film? The aesthetic literature does not abound with examples of films whose immorality is supposed to contribute to their artistic value. But one case that is often discussed is Leni Riefenstahl’s Triumph of the Will (1936). This is a documentary where Hitler is represented as a visionary hero who is loved by the German people, to whom he in turn is supposed to bring fulfillment and self-respect. The film is often praised for its cinematic mastery and for its stunning scenes and powerful images and symbols. It is also condemned for its endorsement of Hitler and his ideas. That the film both has a high artistic quality and moral flaws is not enough, however, to prove the immoralist’s thesis. It must also be shown that the artistic value somehow depends on (rather than simply outweighs) the moral flaw.
What difference does it make? One may well wonder about the substance of the debate between autonomists, ethicists and so on. As the above discussion illustrates, given a sufficiently wide understanding of the concept of artistic value, autonomism becomes less promising, whereas, given a more narrow one, it is easier to defend. How are we to choose which conception of artistic value to adopt? It seems to me that there is no non–question-begging argument here, but that the discussion has to be decided by stipulation. After all, this question about how to delimit artistic value from others is largely a taxonomical question, and such questions are best answered by pondering the purpose the taxonomy is supposed to serve. Accordingly, it seems to me that the outcome of the debate between autonomism, ethicism and so on will have only a limited significance. Suppose that autonomism comes out the winner. This does not exclude ethical criticism or evaluation being a worthwhile activity for a critic or scholar. After all, adopting such a critical perspective may be a crucial part in the process of trying to understand the film, by trying to discern what moral points of view, if any, it projects or endorses. Nor would the victory of autonomism undermine the interest moral philosophers and others who engage in moral thinking take (and have reason to take) in films and other narratives. Artworks are uniquely able to represent human interactions, situations, and relationships in a way that can capture much of their complexity, or at least far more so than the simplistic stories used in the thought experiments by which moral philosophers “test” and explore their principles and moral theories. They can therefore be useful for allowing us to give substance to the concepts our moral theories invoke, as well as allowing us to perceive where distinctions need to be made, even if the theories posit none. Having said that, however, it seems to me that there is rather little overlap between the films that prompt moral criticism and those that moral philosophers have most reason to take an interest in. In any narrative, the advocacy of a certain point of view is usually accompanied by a degree of oversimplification. This may manifest itself in the fact that some considerations that speak against the point of view being advocated 119
TF17557.indb 119
16/9/08 15:47:23
F OLK E T E R SM A N
are overlooked and that some of the complexity of the issues is ignored. So, if the main ground for morally criticizing (or praising) a film is that it endorses a certain moral perspective, the films that are most prone for such criticism tend also to be less interesting for a moral philosopher. We have enough oversimplification in moral philosophy as it is. See also Censorship (Chapter 3), Emotion and affect (Chapter 8), Empathy and character engagement (Chapter 9), Race (Chapter 21), Stanley Cavell (Chapter 32), and Pornography (Chapter 47).
References Carroll, N. (2000) “Art and Ethical Criticism: An Overview of Recent Directions of Research,” Ethics 110: 350–87. Currie, G. (1995) “The Moral Psychology of Fiction,” Australasian Journal of Philosophy 73: 250–9. Davies, S. (2006) The Philosophy of Art, Malden, MA; and Oxford: Blackwell. Dean, J. (2002) “Aesthetics and Ethics: The State of the Art,” American Society for Aesthetics Newsletter 22: 1–4. Donnerstein, E., Linz, D., and Penrod, S. (1987) The Question of Pornography: Research Findings and Policy Implications, New York: Free Press. Gaut, B. (1998) “The Ethical Criticism of Art,” in J. Levinson (ed.) Aesthetics and Ethics, Cambridge and New York: Cambridge University Press. Hursthouse, R. (1999) On Virtue Ethics, Oxford and New York: Oxford University Press. Isenberg, A. (1973) Aesthetics and the Theory of Criticism, Chicago: Chicago University Press. Jacobson, D. (1997) “In Praise of Immoral Art,” Philosophical Topics 54: 155–99. John, E. (2006) “Artistic Value and Opportunistic Moralism,” in M. Kieran (ed.) Contemporary Debates in Aesthetics and the Philosophy of Art, Malden, MA: Blackwell, 331–41. Kieran, M. (2002) “Forbidden Knowledge: The Challenge of Immoralism,” in J. Bermudez and S. Gardiner (eds.) Art and Morality, London and New York: Routledge. Levinson, J. (1990) Music, Art and Metaphysics, Ithaca, NY: Cornell University Press. Lipkov, A. (1988) “Strasti po Andreiu” [The Passion according to Andrei], Literaturnoe obozrenie 9: 74–80. Livingston, P. (2005) Art and Intention, Oxford: Clarendon Press. Murdoch, I. (1970) The Sovereignty of Good, London: Routledge and Kegan Paul. Nussbaum, M. (1990) Love’s Knowledge: Essays on Philosophy and Literature, New York: Oxford University Press. Slote, M. (1992) From Morality to Virtue, New York: Oxford University Press. Stecker, R. (2005) “The Interaction of Ethical and Aesthetic Value,” British Journal of Aesthetics 45: 138–50 —— (2007) “Interaction of Artistic and Ethical Value: Immoralism and the Anti-Theoretical View” (unpublished manuscript). Wollheim, R. (1980) Art and Its Objects, Cambridge and New York: Cambridge University Press.
120
TF17557.indb 120
16/9/08 15:47:23
11
FI L M A S ART Robert Stecker Question: Is film art? The question is ambiguous. It might be asking whether films are artworks, or, more cautiously, whether there are films that are artworks. Or, it might be asking whether film is an art form, or more cautiously, whether a kind of film medium is or contains an art form. A natural follow-up to the first question is, What makes an individual film an artwork? A natural follow-up to the second questions is, What makes a film medium (contain) an art form? That there are distinct questions here may not be obvious. But the distinction is important and neglected. So is the philosophical investigation of art forms, which has always taken a backseat to artworks when people ask the question “what is art?” The reason the distinction is important is this. Consider one film medium: still photography. There are lots of still photos, such as most family snapshots, that have no pretension to being artworks. There are others that almost certainly are artworks. But given that there are a vast number of nonart photos, someone might question whether still photography is an art form. Or, if they don’t question this, they might still wonder what is responsible for the existence of an art form in a sea of nonart photographs. The main focus of this essay is film in the sense of moving pictures as seen in the cinema, on television and on video or DVDs, though I will continue to pull in still photographs when it is helpful to do so. The majority view today is that film is an art form and, hence, many cinematic works are artworks. On this view, the main issues concern the nature of the art form, the nature of the film artwork, and the boundary between film art and film nonart. A further question, but one that we have space only to touch on, is whether there are important differences in the way we appreciate artworks and nonartworks that share the same medium? Why should anyone care whether some films are artworks and whether there are one or more film art forms? The short answer concerning works is that, while asserting that a film is an artwork does not entail that it has any special value, it does entail it is the sort of thing capable of having great value. Regarding forms, the short answer is that asserting that there is a film art form entails that an institution exists for creating, distributing, interpreting, and evaluating a class of items, all of which are films, capable of having such value. Fuller answers will emerge from what is said below.
TF17557.indb 121
16/9/08 15:47:23
ROBE RT S T E C K E R
Skepticism about film as art Before going further, we should face up to the fact that there is skepticism about film as an art form. As Noël Carroll (1988) and Berys Gaut (2002, 2003) have noted, much of classical film theory is an attempt to argue against such skepticism and in favor of a cinematic art form. The focus here will be on a more recent version of these skeptical arguments, forcefully advanced by Roger Scruton (1983). Scruton’s main conclusion is that photography and film are not representational arts, and by “arts” it is better to interpret him to mean art forms rather than artworks. In fact, Scruton does not deny that some photographs and some films are representational artworks but thinks that when they are, this is so in virtue of their using means borrowed from genuine art forms. In the case of photographs, the relevant art form is painting. Photographs become representational artworks, according to Scruton, only when they are “polluted” with painterly techniques. The relevant art form in the case of film is not painting but drama. There are films that are dramatic artworks, but Scruton seems to think that their existence in a film medium is merely contingent, rather than essential, to their being artworks. Scruton also does not deny that photographs and films can have aesthetic value in virtue of their form, though he doesn’t go so far as to claim (or explicitly deny) that they can be artworks in virtue of their formal interest. I suspect he would not be sympathetic to avant-garde filmmaking that attempts to get at the essence of film by emphasizing its formal dimensions – such as its basis in the projection of light or its creation of an impression of movement. Nor would he applaud other avant-garde works that emphasize film as a recording mechanism, though that is what Scruton thinks film essentially is. He just might not believe that the use of the film medium to tediously illustrate this “truth” makes the illustration an artwork. Scruton believes that film is not a representational art form, because he thinks that film (or photography) does not create representations. The bare-bones argument for this conclusion is 1 A is a representation of B only if A expresses a thought about B. 2 Photographs (films) result from B causally interacting with a photographic mechanism in a way that does not require thought or intentionality. 3 Since A results from such a process, it does not express a thought about B. 4 Therefore, A is not a representation of B. As it stands, this argument is not persuasive. First, what justifies the partial definition of representation in premise 1? If one looks at the literature on pictorial representation, one will not find in it any definition that has a condition as strong as Scruton’s. Nor do all “handmade” pictures satisfy this condition. Suppose you ask me to make a picture of a man – any man – and I draw a stick figure. Have I expressed a thought about a man or men? It’s not clear that I have. Have I depicted a man? Yes, I have, per instruction. Second, premise 3 looks to be false. Just because A results from a process that does not require thought, this does not imply that the photographic process is 122
TF17557.indb 122
16/9/08 15:47:23
F I LM AS A RT
never expressive of thought. It only implies that it needn’t be (Gaut 2003: 631). But as we have seen, that is also true of handmade pictures. Scruton might reply to the objection to premise 1, by claiming that we have misinterpreted it. All “expressing a thought” comes to is that there is an intentional relation, not just a causal relation, between the representation and its object. To put it another way, a genuine representation must always have an intentional object, which may or may not exist, and if it does exist, may or may not capture the object’s true appearance. My stick figure meets this condition, and there are definitions of depiction in the literature that do so as well. Scruton replies to the objection to premise 3 by claiming that he is talking about ideal photographs (films), for which the premise is true. They can only have real objects that causally interact with the photographic process, and the result must at least roughly capture the objects’ true appearance. However, this is still not convincing. For one thing, we can imagine genres of painting which would seem not to count as representational on this account but surely are. I will call one such genre “ideal portraiture.” To be an “ideal portrait,” the following conventions must be met. First, like all actual portraits, there must be an actual person represented in the picture. There is no possibility that the portrait is not of something that exists. Second, A is an ideal portrait of B only if A at least roughly captures a true appearance of B. A painting that fails in this respect is not a bad ideal portrait; it is not an ideal portrait at all. This convention may not be true of all actual portraits, but it is a perfectly possible convention. Third, for A to be an ideal portrait of B, B must causally interact in the right way with the process that brings A into existence. Ideal portraits are like ideal photographs in the ways that purportedly disqualify the latter from representationality. Its intentional object is always identical to a real object, the true appearance of which is invariably captured in the painting, as a result of causal interaction between the objects and the production process. Ideal portraits clearly are capable of expressing thoughts about their subjects, even though their subjects must exist and have an appearance not too different than the one represented. What about ideal photographs (films)? That depends on what Scruton means by this coinage. He sometimes seems to mean something brought about by a causal process devoid of intentionality. If so, ideal photographs (films) cannot express thought, but Scruton is then just making this true by definition. What he should mean by an ideal photograph (film) is one produced using only photographic or filmmaking techniques, and not ones imported from painting or drama. However, if this is what is meant, then ideal photographs (films) are not barred from expressing thoughts about their subjects and certainly do so. To give just one obvious example, consider the expressive potential of the close-up. This is not a technique available to theater, it only uses photographic resources, but it can be used in film for expressive purposes. For example, in Woody Allen’s Hannah and Her Sisters (1986), at least two scenes end with Hannah (Mia Farrow) filling nearly the entire screen in close-up. This expresses various thoughts about the character – the central role she plays in her extended family, her conception of this role as the lynchpin of the family, and her Christ-like willingness to take on the burdens of all the other family members (represented tongue-in-cheek). (See Gaut [2002], King [1992], and Warburton [1988], 123
TF17557.indb 123
16/9/08 15:47:23
ROBE RT S T E C K E R
for other examples of photographic and cinematic techniques that express thoughts about a subject.) This same example illustrates what is wrong with another claim Scruton makes, namely, that our interest in a photograph (film) is in the subject shown in it, rather than in the representation in its own right. This is precisely not the case in the example just given. Our interest is not in what Mia Farrow looked like when these scenes were being shot, but in the meaning expressed in the way the scenes were shot. There is one final charge that Scruton makes about photography and film that we should address. It is the claim that both photography and film are “fictionally incompetent,” i.e., incapable in their own right of fictional representation. If this were true, it would be especially problematic for cinema, since so much of it is understood to be a type of fiction. (Fictions, roughly, are works a primary function of which is to authorize the imagining of various states of affairs. Whether the states of affairs do or do not exist in reality is a secondary matter, though it is also a mark of a fiction that it is capable of getting us to imagine what doesn’t exist.) Scruton’s main argument for this claim is no more convincing than the previous one. Suppose I take a photograph of a denizen of skid row and entitle it “Silenus.” Why wouldn’t this be a fictional representation of the mythical old man? Scruton asks us to imagine a parallel case. I see an individual on skid row, and pointing to him utter “Silenus.” Is my pointing a representation? Scruton rightly says no: my gesture at best makes the tramp into a representation, rather than being one itself, and he thinks the same is true of a photograph. However, there is a ready reply. First, Scruton does not deny that the pointing expresses a thought, and hence, so does the photograph, since it is after all “a pointing.” So Scruton can’t assert that the photograph lacks “intentionality.” Second, you can’t point to something that doesn’t exist, so if the thought is “about” Silenus there must be more involved than pointing. Finally, the main reason the pointing is not a representation is that it is an act that results in no independent representational object. That is precisely what a representation is: an object that represents something else. But a photograph, which, ex hypothesi, expresses a thought intended by its maker, is such an object. Even if we see the tramp (or a surrogate) in the photograph as Scruton thinks, that is no bar to the photograph fictionally representing Silenus. Underlying this argument is a simpler thought: that a photograph is not a fictional representation but a record of a preexistent fictional representation. Unfortunately, this is not true in the Silenus example (nothing is a representation of Silenus until the photo is taken), and it is even less plausible in the case of a movie. A film recording of a stage production approximates to what Scruton has in mind. But a movie, made by filming and editing many scene fragments, made significant by the use of many photographic techniques, is hardly the recording of a preexistent fictional representation.
124
TF17557.indb 124
16/9/08 15:47:23
F I LM AS A RT
The cinematic artwork If films can represent actual people and things as well as fictions, they can do so in the rich and complex ways characteristic of artworks. In fact, even skeptics about the representational capacities of film, like Scruton, do not deny that there are artworks that happen to be films. Such works don’t contingently exist in the medium in which we view them but are essentially cinematic works of art. This is so in two respects. First, the nature of the representational content of a film typically depends on the use of essentially photographic and other techniques peculiar to filmmaking and not available to live stage drama. Second, the modes of presentation and reception for film are distinctive of the medium, although it is interesting that there is not a unique format here. The same film can be viewed by projecting it onto a screen or by viewing it in VCR, DVD, or still other formats. When is a film an artwork? The answer will depend on the conception of art one brings to the table. What is now known as classical film theory was concerned with this question and approached it in a way characteristic of the period in which it developed. It looked for established artistic functions that the film medium fulfills in virtue of resources peculiar to the medium – special ways of realizing standard functions. Two functions received the bulk of attention: expression and realistic representation. In part, the expressionist view is motivated by a desire to refute an objection to film art, much like Scruton’s. Film cannot create artworks, because it is a mechanical recording device that only reproduces the reality in front of camera. “Expressionists” (Eisenstein 1988; Arnheim 1957; Münsterberg 1970) attempt to refute this picture by emphasizing the ability of film to manipulate and rearrange reality to manifest an attitude about its subject. André Bazin (1967, 1997) is the most distinguished proponents of film as the art of realistic representation. However, it’s worth noting that he is perfectly aware of the expressive and antirealist potential of film and is lucid in analyzing the two chief ways of realizing this potential: “the plastics of the image . . . and the resources of montage, which . . . is . . . the ordering of the image in time” by means of editing (1997: 60). Nevertheless, Bazin views the history of filmmaking as an evolution to an aesthetically superior realism. This is realized by avoiding the most manipulative forms of montage, such the juxtaposition of distinct events by alternating shots, which Bazin described as “chopping the world into little bits” (1997: 70). Alternative techniques such as the deep-focus shot create an artistically superior spatial realism. Furthermore, Bazin seems to agree with Scruton and Kendall Walton (1984) that in a photograph, we actually see the objects photographed. However, it is not so clear what logical role this plays in Bazin’s defense of realism in film. Like Walton and unlike Scruton, Bazin does not mistakenly infer from this belief about the presentational capacities of the photograph that it is not a representational medium, or that it is fictionally “incompetent.” Nor does he infer that the purpose of film is just to present the reality in front of the camera. On the contrary he thinks of film as a medium for realistic fictional representation. The techniques allowing for greater realism also enhance the filmmaker’s ability to manipulate reality (1997: 71), to create ambiguity, and it requires more 125
TF17557.indb 125
16/9/08 15:47:23
ROBE RT S T E C K E R
active participation from its audience (1997: 68). It makes the filmmaker the “equal” not of the chronicler but of the “novelist” (1997: 71). Theorists like Bazin and even to a greater extent Arnheim are reminiscent of philosophers who attempt to define art in terms of a single function or valuable property. Like Bazin’s account of film realism, such philosophical accounts can be complex and highly nuanced, but the dominant view today is that simple functionalism will always fall to counterexamples of works in the relevant class that lack the function or valuable property in question. In recognizing masterpieces of montage, Bazin provides the counterexamples himself to a general theory of film art, based on realism. The rejection of simple functionalism leads some philosophers to offer accounts of arthood or art status, in self-consciously nonfunctional terms. However, it is not strictly necessary to eschew functions in explaining the nature of film as art as long as one is a pluralist about relevant functions. So we should recognize multiple functions that films can realize and be artworks in virtue of these. This includes evocative, expressive, and representational functions realized in both narrative fictional movies and documentaries. It also includes functions specific to the visual and pictorial nature of most films. Further it includes the function fulfilled by films that explore the nature of the film media themselves. The list should be left open ended. Functions evolve as a result of the evolution within a medium, of the larger art world, and of the culture at large.
The cinematic art form Not only are there cinematic artworks, but it is plausible that there is a cinematic art form. We will discuss what makes this claim plausible, but before doing this, let us briefly consider a counting issue, namely, assuming there is at least one, how many cinematic art forms are there? (This issue is not trivial, because different art forms need to be appreciated in different ways and knowledge of form guides one’s appreciative efforts.) Compare film with literature, which seems like a roughly analogous category within the arts. Poetry, the novel, and the short story are three different literary forms. Within each of these categories, there are different subforms or genres. So while the feature film has lots of genres perhaps it should be regarded as a cinematic art form, and similarly for the documentary and the short film. Cutting across this kind of classification are others. Animation seems like a different form than the live-action film, but both can be feature films or shorts. Silent film might be regarded as a different form than talkies. The technology used in creating a moving image also can have a much bigger impact on a final product than it does in literature, where it matters little to the reader whether the writer uses a pencil or a word processor. But whether one uses photography, video, digital technology, computer graphics, motion capture, or animation makes a difference to what one sees and the way one sees it. Sometimes this will just make possible modifications in already existing forms, but should we regard an entire film made using motion-capture technology such as A Scanner Darkly (2006) as a new form? While it seems plausible that within the broad category of cinematic art there are several proper forms, how they are to be individuated is an open question. 126
TF17557.indb 126
16/9/08 15:47:23
F I LM AS A RT
So is cinema an art form (in a broad sense in which literature is one too)? We have mentioned two grounds for doubting this. We have considered and rejected skepticism about film as a representational art. At the beginning of this essay we also mentioned the fact that many films (and photos) just aren’t artworks. The underlying thought here is that if the typical product of a given medium is not an artwork, there is no art form, even if some works in the medium rise to the level of art. However, it turns out that the situation in film (and photography) is not unusual. Consider architecture. There are architectural artworks but a sea of buildings, many designed by architects, that have no pretension to art status. Again the question might be raised whether there is an architectural art form while admitting there are architectural artworks (Davies 1995). No one doubts that painting is an art form. However, there are lots of paintings – made for illustration or various commercial purposes that also exist outside the art form. One can say the same thing about musical and literary forms. These examples show that large numbers of nonartworks in a given medium is not evidence of the nonexistence of an art form Given all this, we have no reason to deny that “architecture,” “painting,” “music,” “poetry,” and “film” can refer to art forms. It is true, though, that these terms can also refer to a class of items that include artworks and nonartworks, which I have been referring to as a medium. Perhaps all buildings belong to the medium of architecture, but not to the art form architecture. The same goes for films (and photos). It is plausible, then, to see art forms as existing within wider media that get used for both artistic and nonartistic purposes. An art form typically encompasses a subclass of items in a medium, items made with certain intentions, or which achieve certain aims, which are capable (and possibly worthy) of receiving the attention of critics and audiences and being placed in certain institutional settings. Crucial to making something an artwork are the intentions with which it is made and the aims achieved. Crucial to the creation of an art form is the existence of institutional settings to organize the availability of such works to interested audiences, to interpret and evaluate them, and so on. Such institutions have developed in the film world. There are settings for the presentation or distribution of films in general: movie theaters, Cineplex, video stores, but there are additional venues that can be used to distinguish a set of films as artworks: film festivals, retrospectives of the work of certain directors, and film series. There is film criticism that exists in a number of forms ranging from five-minute television reviews to academic criticism. As we have noted already, there is film theory that has obvious similarities to literary and art theory. There is by now a canon of great movies, and there is a film avant-garde. All these items are indications that film is an art form. It might be objected that these indications are insufficient. This is so for two reasons. First, there seems to be similar institutional arrangements for such things as fashion and food, for example, but this does not show that fashion designing or food preparation are art forms. Second, the film institutions I have mentioned do not do a very good job at segregating film artworks from film nonartworks. Just about any film can be grist for film criticism or film theory. Film festivals may try to show films that have artistic aims, but they also have a commercial rationale: to hook films up with 127
TF17557.indb 127
16/9/08 15:47:23
ROBE RT S T E C K E R
distributors. This means that a film’s presence at a festival could be a sign of its art status or its commercial potential. There is no space to discuss the multiple issues raised by the first of these reasons. The second reason makes an important point about film as an art form. With respect to the distinction between the art form and the medium, where the line is to be drawn in the case of film is even more controversial than usual. Most home movies, and various “products” which use the film medium, such as film previews and advertising are presumably neither artworks nor in the art form. Beyond these, there is the large and diverse class of commercial films, and then avant-garde cinema. The latter clearly aspires to be art. The controversial question is how we sort out commercial cinema. Is it all part of the art form? Should we distinguish, within commercial cinema, entertainment from serious films, only the latter qualifying as artworks? Or do none qualify? The best way to explore the film art form is to turn to these questions.
Mass art One category to which mass market, commercial movies might belong is that of popular art. A distinction between fine art and popular art is long-standing. But the point of the distinction varies among those who use it. For some, popular art is just a kind of art, to be studied alongside others that are put in the fine-art category. For others, to call something popular art is to deny it is really art (henceforth, Art) even if it resembles Art in some respects. On this usage, if it were Art, it would be fine art. This makes some sense because, among the many senses of “art” in English and other languages, it is the “fine-art” sense that is often assumed to be the one defining the subject matter of philosophy of art. However, because of the variable import of the distinction, the status of popular art as Art is perennially somewhat blurry. Recently, Noël Carroll (1998) has introduced another concept related to popular art but intended to be sharper in its import: mass art. Mass art is Art. That is, it is an appropriate subject for the philosophy of art, for art criticism, art history, etc. It is to be contrasted with various categories of esoteric art, including avant-garde art, and with “middlebrow art” rather than fine art, a term that is absent from Carroll’s discussion. What is distinctive of mass art is that it is a type of art produced and distributed by mass technology and is designed to be easily accessible to “the largest number of . . . relatively untutored audiences” (1998: 196). Although mass art makes far fewer demands on its audience, its reception and appreciation is continuous with these other species of Art. Carroll’s proposal is attractive for two reasons. First, Carroll is right that there is some sort of continuity between the crassest commercial product such as the movie Scooby Doo (2002) and, say, La Dolce Vita (1960). Both attempt to fulfill representational, expressive, and other functions traditionally associated with narrative art, by making use of the potentialities of the film medium. Second, accepting the proposal appears to eliminate the need to make what inevitably will be a messy and uncertain distinction between commercial film Art and non-Art. All of it is film Art. 128
TF17557.indb 128
16/9/08 15:47:23
F I LM AS A RT
There are also reasons to question the proposal. One issue is whether it picks out mass art in the right way. With respect to cinema, since every work in the medium is made using a “mass technology,” the whole burden of this falls to the condition that says that mass art is designed to be accessible to the largest relatively untutored audience. But it’s not clear that this is so. Slasher films aim at a much smaller niche audience than blockbuster films like Spider-man (2002). It still should count as mass art, and Carroll would agree. He would claim that slasher films are accessible to, but just not enjoyed by, everyone in the Spider-man market. What probably is true is that most in that market could follow the narrative (such as it is) of a slasher film. However, with respect to appreciation – being able to say what makes one slasher film better than another – aficionados of the genre can make fine-grained distinctions quite unavailable to the larger public. Carroll’s definition needs adjusting to account for the fact that different mass artworks aim at distinct audiences and are often not fully accessible to outsiders. Perhaps the task is not so hard. Mass art aims at a large, if not the largest possible, audience (Novitz 1992). Finally, even if we accept it, it doesn’t actually follow from Carroll’s definition that all commercial films are Artworks. For we all know from the definition, Scooby Doo is not mass art. Remember, it is part of the definition of mass art that it is a type of Art. Since “art” appears on both sides of the definition, it won’t help us decide which objects fall under it unless we already know which ones are Art. One might wonder whether there are some films where commercial intentions so overwhelm artistic ones that it is implausible to think of them as mass art. They may have a narrative form, but representational, expressive, and other properties of the production are just not guided by an artistic rationale. After all, advertisements can have a narrative form too, but not every commercial with such a form is an artwork. Just when one sort of intention does not merely constrain but overwhelms the other is hard to say, but it evidently sometimes happens. If so, the question that began this section simply gets transformed into the question, Which films are Art (mass, middlebrow, esoteric) and which are non-Art?
Conclusion We have argued that there are film artworks and a film art form. Skepticism here should be rejected. We have been unable though to fix the boundary of the form. We leave it an open question. Let’s conclude by pointing to an approach that may or may not help, but raises a question important in its own right. All film can provide some sort of aesthetic experience (good, bad, indifferent). They can be aesthetically appreciated and evaluated. Whether a film is art or nonart, we will look for many of the same features in seeking out an appreciative experience. We evaluate the fictional world it creates, the way it tells its story, the skillfulness of the camera work, the emotions it elicits, and the questions it raises, the integration of dialogue, visual elements, music, and so on. Is there something distinctive about the aesthetic appreciation of films that are art? Do only they, perhaps, have a style, and does appreciation require viewing the film with an understanding of its style? This is another issue that must be left for resolution on another occasion. 129
TF17557.indb 129
16/9/08 15:47:23
ROBE RT S T E C K E R
See also Authorship (Chapter 2), Medium (Chapter 16), Realism (Chapter 22), Style (Chapter 25), Rudolph Arnheim (Chapter 27), Sergei Eisenstein (Chapter 35), and Hugo Münsterberg (Chapter 39).
References Arnheim, R. (1957) Film as Art, Berkeley: University of California Press. Bazin, A. (1967) What Is Cinema?, trans. H. Gray, vol. 1, Berkeley: University of California Press. —— (1997) “The Evolution of the Language of Cinema,” trans. Hugh Gray, in P. Lehman (ed.) Defining Cinema, New Brunswick, NJ: Rutgers University Press. Carroll, N. (1988) Philosophical Problems of Classical Film Theory, Princeton, NJ: Princeton University Press; New York: Oxford University Press. —— (1998) A Philosophy of Mass Art, Oxford: Clarendon Press. Davies, S. (1994) “Is Architecture an Art?,” in M. Mitias (ed.) Philosophy and Architecture, Amsterdam: Editions Rodopi. Eisenstein, S. (1988) Selected Works, ed. and trans. R. Taylor, London: British Film Institute; Bloomington: Indiana University Press. Gaut, B. (2002) “Cinematic Art,” Journal of Aesthetics and Art Criticism 60: 299–312. —— (2003) “Film,” in J. Jevinson (ed.) The Oxford Handbook of Aesthetics, Oxford and New York: Oxford University Press. King, W. (1992) “Scruton and the Reasons for Looking at a Photograph,” British Journal of Aesthetics 32: 258–65. Münsterberg, H. (1970) The Film: A Psychological Study, New York: Dover. Novitz, D. (1992) “Noël Carroll’s Theory of Mass Art,” Philosophic Exchange 23: 39–50. Scruton, R. (1983) “Photography and Representation,” in The Aesthetic Understanding, London and New York: Methuen. Walton, K. (1984) “Transparent Pictures: On the Nature of Photographic Realism,” Critical Inquiry 11: 246–77. Warburton, N. (1988) “Seeing through ‘Seeing Through Photographs’,” Ratio 1: 64–74.
Further reading Perhaps the most useful work for the project of linking film art and style is David Bordwell, On the History of Film Style (Cambridge, MA: Harvard University Press, 1997). Ian Jarvie, Philosophy of the Film: Epistemology, Ontology, Aesthetics (New York: Routledge and Kegan Paul, 1987), argues that film is art, and responds to arguments skeptical of this point of view. Another defense of the realist film aesthetic is Siegfried Kracauer, Theory of Film: The Redemption of Physical Reality (New York: Oxford University Press, 1960). A. Sesonske, “Aesthetics of Film, or a Funny Thing Happened on the Way to the Movies,” Journal of Aesthetics and Art Criticism 33 (1974): 51–7, is a seminal philosophical work on the aesthetics of film, as is Francis Sparshott, “Basic Film Aesthetics,” in G. Mast, M. Cohen, and L. Braudy (eds.) Film Theory and Criticism: Introductory Readings, 4th ed. (New York: Oxford University Press, 1992).
130
TF17557.indb 130
16/9/08 15:47:23
12
FO RMALISM Katherine Thomson-Jones Film formalism is an unusual kind of formalism, for at least four reasons. The first reason is that, whereas in relation to other art forms – including painting, music, and dance – formalism has been a fairly dominant, mainstream view, in relation to film, formalism is a minority view overshadowed by a strong interpretive tradition. Moreover, it is not a minority view because it has gone largely unnoticed; on the contrary, formalism in its contemporary guise is seen by many film scholars as a threatening force to be vigorously resisted. The second reason film formalism is unusual is that it does not uphold many of the familiar formalist tenets. Most notably, film formalists deny the aesthetic significance of a sharp distinction between form and content, the existence of a distinctly aesthetic response to formal beauty, and the appropriateness of purely “immanent” and ahistorical criticism. The third reason that film formalism is unusual is that it is associated with a constructivist view of meaning, whereas classical formalism, in its commitment to the autonomy of art, would claim that the meaning of a work is determined by its intrinsic formal features. Finally, the fourth reason that film formalism is unusual is that it is derived from literary theory and yet finely attuned to the historical and technical possibilities of the film medium. As we shall see, it is precisely this attunement that explains the unusual features of film formalism. Yet film formalism remains part of a broader formalist tradition insofar as it is, in essence, a program for the structural analysis and appreciation of films as aesthetic objects. The first formalist approach to film emerged in the 1920s, when the medium was still trying to legitimate itself as an art form. This approach was an attempt made by the Russian formalists to extend their literary poetics to cinema. In this attempt, they influenced the Soviet montage filmmakers, particularly Sergei Eisenstein. Both groups shared an interest in language and the technical construction of films. The influence of Russian formalism in the English-speaking world of film theory is comparatively recent, since the first translation of Russian formalist essays did not appear until 1965 (Lemon and Reis). In the mid- to late-1980s, David Bordwell and Kristin Thompson, frustrated with the established direction of film study, revived the methodology and key concepts of Russian formalism as part of an updated poetics of film. Thompson refers to this approach as “neoformalism,” but Bordwell, while acknowledging his and Thompson’s shared goals and interests, does not in fact
TF17557.indb 131
16/9/08 15:47:23
K AT H E R I N E T HOMS ON-JON E S
consider himself a neoformalist. Nevertheless, the Russian formalists are arguably the greatest influence on Bordwell’s work, and critics of neoformalism regularly take Bordwell as their primary target. For these reasons, I will treat the work of Bordwell and Thompson as belonging to the same broadly formalist tradition. Neoformalism does not rely on the Russian formalists’ own attempt to apply their ideas about literature to film, because the neoformalists are suspicious of the Russian formalists’ reliance on an analogy between film and poetic language. Based on a much more extensive knowledge of film history and technique, the neoformalists attempt their own application of Russian formalist literary theory to film. In the process, they draw on other literary and film theorists (including Eisenstein, André Bazin, TzvetanTodorov, Gérard Genette, Noël Burch, the 1966–70 Barthes, and contemporary Israeli poeticians like Meir Sternberg) and add some important corollaries, most notably, a constructivist view of meaning. Bordwell and Thompson continue to practice neoformalist (or formalist-influenced) poetics today, as indicated by the most recent edition of Film Art (Bordwell and Thompson 2006), which continues to instruct the student of film, not to “read” the film as a symptomatic text but to analyze it as a complex structure.
Form and content The defining characteristic of any formalist approach is its theoretical and critical emphasis on form. In relation to film, the neoformalists have a unique understanding of form, which is sensitive to historical context while revealing a methodological commitment to close analysis of the work. Whereas formalism is traditionally understood to uphold a strict distinction between form and content and to deny the aesthetic relevance of content, the neoformalists follow the Russian formalists in simply denying the aesthetic relevance of the distinction itself. They do this because they worry that insisting on the distinction encourages “extrinsic” criticism – assessing the work biographically, sociologically, psychologically, or politically but never on its own terms, qua art. One way that the distinction between form and content is understood is in terms of form being the container for the work’s content, which can thus be extracted for independent analysis. Rather than focusing on the conceptual problems with the container account, the neo- and Russian formalists focus on the way that such an account seems to prioritize content in analysis, as though the function of art is to convey a certain content to its audience. To discourage this disjunctive privileging of content, Bordwell and Thompson, in the first edition of Film Art (1979), refer to a film as an integrated “system.” As we shall see, however, the neo- and Russian formalists ultimately reject the container account because they assign a different function to art. By doing so, Thompson thinks that they have “eliminated the need for the formcontent split” (Thompson 1981: 11). Another way that the distinction between form and content is understood is in terms of content being the work’s subject matter and form being the mode of presentation for that subject matter. On this understanding, however, the Russian formalists 132
TF17557.indb 132
16/9/08 15:47:23
F OR M A LISM
claim that content is merely raw material for an artistic treatment, which results in a unified structural whole with a distinctly aesthetic function. Consequently, the Russian formalists substitute another distinction for the form/content distinction adopted for film by the neoformalists. This is the distinction between materials and devices. The reason that the materials/devices distinction is supposed to be different from the form/content distinction is that, unlike content, materials are not alleged to be part of the artwork at all. This is because the same materials can be used to make both art and non-art objects. Whereas the Russian formalists identified words (and sometimes also ideas and emotions) as the materials of literature, Thompson identifies mise en scène, sound, camera framing, editing, and optical effects as the five kinds of cinematic material (Thompson 1981: 25–6). Further on, she also identifies as materials meanings or ideas from real life and artistic conventions (Thompson 1981: 51). According to the Russian formalists, words are the common material for both “practical” and poetic discourse, and this is the case even if particular ways of using words – say figuratively or literally, are more commonly associated with one form of discourse than the other. What distinguishes the work of literature, therefore, is not that it contains certain words with certain connotations but that those words are used in a certain way to serve a particular purpose. Similarly, framing, sound, editing, and so on are the common materials for both nonaesthetic films – for example, informational or promotional films – and aesthetic films, including narrative fiction, documentary, and experimental films. What distinguishes the work of film art is not that it exhibits technical features of the medium but that these features are used in a certain way for a certain purpose. Insofar as a device is any medium-specific technique for manipulating, transforming, and structuring materials, the neo- and Russian formalists understand the work of art simply as a set of devices. The use of the term, “device,” reflects the formalists’ understanding of artworks as the products of craftsmanship, rather than as the product of inspiration or as vehicles for expression and communication. As Thompson points out, the Russian formalists reliably use a variety of craft metaphors in their analysis of the literary process – for example, metaphors of weaving and sewing. They also conceive of the work as a reassuringly solid and material construction – as seen, for example, with Viktor Shklovsky’s description, embraced by the neoformalists, of narrative structure as a staircase. Essentially, the Russian formalists were interested in how a work works, both in terms of the technical possibilities and artistic conventions that shape the artists’ constructional choices, and in terms of the purpose of all the various components in the work and the work as a whole. As we shall see further ahead, there is also a sense in which, for the neoformalists, a work works by working on us, or by achieving certain effects. Thus Bordwell’s interest in the way a work cues our comprehension and interpretation is continuous with the neoformalist project, as is a shared interest in the way that the structures of a work can trigger affect. Thompson points out, however, that this interest in affect and structure does not mean that neoformalism posits a distinctly aesthetic emotion. 133
TF17557.indb 133
16/9/08 15:47:24
K AT H E R I N E T HOMS ON-JON E S
The Russian formalists assume that every component of a work has a purpose and thus it is a legitimate critical activity to consider the justification – or, in their own terms, the “motivation” – of any and all components of a work. The neoformalists adopt the Russian formalists’ taxonomy of three kinds of motivation – compositional, realistic, and artistic – and then add a fourth kind of their own, transtextual motivation. As David Bordwell nicely illustrates, a single component of a film can be motivated in several ways: in answer to the question, Why does Marlene Dietrich sing a cabaret song at just this point in the film?, we could say that her character meets the hero at this point (story relevance), she plays a cabaret singer (realism), and Dietrich often sings these kinds of songs in her films (transtextual reference) (Bordwell 1985: 36). According to Thompson, every component of a film has at least an artistic motivation insofar as it contributes to “the work’s abstract, overall shape,” but we tend only to notice this motivation when the other kinds of motivation are lacking (Thompson 1988: 19). Given that a work can be broken down into its components in a number of ways, the analyst faces a practical difficulty in knowing where to focus his or her attention. To solve this problem, the Russian formalists introduce the concept of the “dominant,” the structural component which organizes all other components and grounds the integrity of the work. To identify the dominant in a work, the analyst must know the overall function of the work. According to the Russian formalists, all art has the same primary function. The neoformalists’ attraction to this view explains, in turn, how they conceive of the history of art and the relation between a work and its context.
The history and function of (film) art The terms introduced by Shklovsky to characterize the function of art have become familiar catchwords for students and scholars of the arts: ostranenie is translated as either “defamiliarization” or “making strange”; and zatrudnenie is translated as “making difficult.” By making its materials strange and more difficult to apprehend through the process of aesthetic transformation, the work of art renews and sharpens perception, breaking down “automatized” recognition and bringing us back to “the sensation of life” – for example, the “stoniness” of the stone (Shklovsky 1965: 11). This does not mean that art only has instrumental value, as a means for exercising our senses. According to Shklovsky, the way that art achieves its perceptual function is just by furnishing a perceptual experience that is fully absorbing and is had for its own sake. This explains how, despite the fact that Thompson explicitly distinguishes the neoformalists’ historical conception of form from the narrower conception associated with the “art for art’s sake” doctrine (Thompson 1981: 10–11), the traditional formalist idea that art should be created and consumed solely for its aesthetic value is preserved by neoformalism. The historically contingent aspect of defamiliarization is also emphasized by Bordwell, but in this case, such an emphasis may betray the influence of the early Czech structuralist, Jan Mukarovský, who recognized that even strangeness can become automatized within a particular cultural context (Burbank and Steiner 1977). 134
TF17557.indb 134
16/9/08 15:47:24
F OR M A LISM
At first glance, there may appear to be a tension between conceiving of the work as the product of craftsmanship and claiming that the function of art is achieved by defamiliarization. Craftsmanship, after all, involves following technical rules to achieve a predetermined result whereas defamiliarization, as we shall see, is at least partly about challenging artistic conventions. The emphasis on craftsmanship may, however, serve simply to focus attention on the material structure of a work and the use of techniques to achieve defamiliarization. The interesting question, therefore, is how exactly defamiliarization is to be achieved by a work. In Thompson’s application of Russian formalism, the options for defamiliarization become richly varied and context dependent. This is due in part to the variety of materials that can be manipulated or “deformed” by the filmmaker. But it is also due to when, where, and by whom a film is watched. Any film is viewed in a particular context – or as the neo- and Russian formalists say, against a particular “background.” The context of a work has several aspects to it, which the neo- and Russian formalists analyze in terms of three kinds of background: artistic conventions, real life, and the nonaesthetic uses of film. It is through our tacit awareness of these backgrounds that we can both make sense of works of art and appreciate the ways in which they achieve defamiliarization. A film like Alain Resnais’ Last Year at Marienbad (1962) can be appreciated as subversive and challenging against the background of mainstream Hollywood-style cinema. Alternatively, a film like Jean Renoir’s Rules of the Game (1939) may seem highly realistic against the background of German expressionist and Soviet montage cinema, but highly mannered against the background of Italian neorealism. It is worth noting that this emphasis on “backgrounds” has prompted a general critique of neoformalism as relatively ahistorical. The argument is that, since history is mere “background” for the neoformalists, they fail to appreciate its active role in determining the range of stylistic options available to a particular filmmaker at a particular time. As Robert Stam explains, the neoformalist ignores “the historicity of forms themselves, i.e., forms as themselves historical events which both refract and shape a multi-faceted history at once artistic and trans-artistic” (Stam 2000: 197–8). This kind of criticism is widespread but rarely supported by close textual analysis. There is, after all, no logical barrier to the neoformalist adopting a more dynamic conception of the relation between form and history. In addition, the emphasis on backgrounds does not imply that only experimental or highly original films can achieve the function of art. Thompson gives three reasons why any film, however generic or mainstream, can renew perception. The first reason is implied by the contextualization of film viewing. A generic film can be taken out of its familiar context and appreciated anew. This may depend on the critic scrutinizing the work and revealing its complex structures (Thompson 1988: 11), or simply on the temporal or cultural distance of a new audience from the work’s generative context. As a result of this switch in backgrounds, the work can no longer be watched with mindless ease; we must pay attention, watching and listening carefully and thereby exercising and extending our perceptual capacities. The second reason why generic films can achieve the function of art is that they have “no immediate 135
TF17557.indb 135
16/9/08 15:47:24
K AT H E R I N E T HOMS ON-JON E S
practical implications for us” (Thompson 1988: 8–9). It is not entirely clear why the nonpractical nature of film viewing would effect defamiliarization. But perhaps the idea is that since the depicted space in any film is separate from the space we occupy, any film is in some sense strange to our everyday, practical selves. If this is the case, it is noteworthy that art contributes to our lives – through defamiliarization and the renewal of perception – precisely by being removed from our lives. The third reason why generic films can achieve the function of art is that to the extent that these films purposefully order depicted narrative events or simply have a structure, they activate our perceptual skills differently from unstructured everyday experience. Thompson then goes on to say that those films we consider highly original either defamiliarize reality more strongly, defamiliarize artistic conventions, or defamiliarize both reality and artistic conventions (Thompson 1988: 11). Thus defamiliarization seems to come in degrees, which suggests that works can achieve the function of art to a greater or lesser extent. Does this mean that the more difficult and strange a work, the better it is? Perhaps not, since Thompson also recognizes that it is important for a critic to aid in the understanding of a work by providing key background information – for example, in the case of Ozu’s films shown to a Western audience, background information about marital and familial practices in postwar Japan (Thompson 1988: 22–3). As well, we regularly criticize works for being unintelligible and praise others for their clarity, efficiency, or precision. These considerations suggest the need for clarification as to whether there is a limit to defamiliarization, either in terms of its efficacy for perceptual renewal or in some other terms of appreciation. There are also two further and more general questions about the function of art that the neoformalists leave unanswered: Can the perceptual function of art be achieved by means other than defamiliarization? And, does art really have a single (primary) function? Although these are important questions, the neoformalists may not feel the need to answer them, since they are not defending a theory of art but merely a functional mode of film analysis. As well as having provocative implications for our understanding of aesthetic appreciation, the notion of defamiliarization has implications for our understanding of the history of art. One view of art history traditionally associated with formalist criticism is that each art form progressively purifies itself until it reveals its own unique essence. The Russian formalists do not hold this view, however. For them, the history of art is a constant overturning of artistic conventions for the sake of defamiliarization. This explains the Russian formalists’ fascination with the avant-garde, since on their view the avant-garde is the mechanism of historical change. The neoformalists, however, are as much concerned with established traditions of filmmaking as with groundbreaking work. Nevertheless, they use the notion of defamiliarization to explain the historical contingency of a certain style being considered perspicuous, realistic, expressive, or properly cinematic. This suggests that there is something relative about the function of art, since a work may achieve its perceptual function with some people at a certain time but not with other people at other times, depending on their familiarity with the techniques used by the work. The neoformalists would not shy away from this suggestion, however, as they are deeply interested in the effects of a film, 136
TF17557.indb 136
16/9/08 15:47:24
F OR M A LISM
given its structure and context, on the audience. This is all part of their interest in how films work. It is crucial to consider the effects of a work on an audience because, according to the neoformalist, the aesthetically significant features of a work are projected onto it by the audience in response to the work’s devices. As Thompson explains, “all those qualities that are of interest to the analyst – its unity; its repetitions and variations; its representation of action, space, and time; its meanings – result from the interaction between the work’s formal structures and the mental operations we perform in response to them” (Thompson 1988: 25–6). This brings us to the corollary that distinguishes neoformalism most sharply from other formalist approaches: a constructivist account of the activity of film viewing, or how viewers, both from a psychological and from a social perspective, comprehend and interpret films. This account is principally developed by David Bordwell in relation to narrative fiction film. According to Bordwell, the viewer constructs the literal meaning of a film through the activity of comprehension, and the more abstract meaning of a film through the activity of interpretation. Bordwell’s “narrative science” is regularly criticized by film theorists for ignoring “subjectivity” – specifically, by conceiving of the viewer as a “hypothetical entity,” rather than as an embodied, culturally defined individual (see, for example, Nichols 1992). In addition, Bordwell’s account can be seen as a case study for the debate among philosophers about the intelligibility of constructivism: how it is that meaning can be given to an artwork which does not already have meaning (see, for example, Stecker 1997).
Neoformalism and constructivism On Bordwell’s account, there are at least four levels of construction in film viewing. First, the viewer must construct everything perceived – the sounds, shapes, and colors of the film, but also the two-dimensional images projected on the screen. Second, the viewer must construct the events and characters that he or she takes the two-dimensional images to represent. Third, the viewer must construct the unified causal history, not entirely shown on screen, of which the depicted events and characters are a part. And fourth, if the viewer is also an interpreter, he or she must construct the more abstract, thematic, or symptomatic meaning of the film as a whole. At every level, the viewer’s constructive activity must have a degree of discretion; if the viewer constructs entirely according to the work’s plan, he or she cannot be said to be the one making the work’s meaning. This suggests that in order to determine whether constructivism is the correct theory of the viewer’s activity, we will need to determine whether the discretionary requirement for viewer construction is compatible with the notion that construction is a structural effect of the work. First, however, we need to examine Bordwell’s account more closely. At the most basic level, the very act of perceiving a film is constructive. Bordwell embraces the dominant view in cognitive psychology that perceiving an object does not just require passively receiving visual, aural, or tactile data but also making an inference from the data in order to reach a perceptual judgment. Bordwell refers to such 137
TF17557.indb 137
16/9/08 15:47:24
K AT H E R I N E T HOMS ON-JON E S
judgments as “hypotheses” to suggest that perception is always open to revision in the face of new data or a new application of relevant background knowledge. Even though our perceiving an object is an instantaneous and automatic process, it still involves the application of concepts in the making of inferences. Interestingly, film depends on our making the wrong inferences as a result of two physiological deficiencies in our visual systems. The raw perceptual data of a film consists in rapidly flashing light and a rapid display of static images, and yet what we see, based on inferences from this data, is a sequence of continuously lit, moving images. It is only because our eyes cannot keep up with the rapidity of changes, both in light intensity and in the image display, that a film is what it is experienced as, namely, a motion picture. When Bordwell applies the general psychological theory of constructivism to aesthetic experience, he reintroduces the Russian formalists’ notion of defamiliarization. By delaying, confusing, and complicating the making of perceptual judgments, works of art defamiliarize and thus bring us to a new awareness of the way in which perception is constructive. As Bordwell explains, in the context of aesthetic experience, what is nonconscious in everyday mental life becomes consciously attended to. Our schemata get shaped, stretched, and transgressed; a delay in hypothesisconfirmation can be prolonged for its own sake. And like all psychological activities, aesthetic activity has long-range effects. Art may reinforce, or modify, or even assault our normal perceptual-cognitive repertoire. (Bordwell 1985: 32) Thompson corroborates this attempt to reconcile Russian formalist theory and constructivist psychology by distinguishing those cognitive processes in film viewing that are conscious, as opposed to preconscious or unconscious, and then suggesting that “[i]n a sense, for the neoformalist, the aim of original art is to put any or all of our thought processes onto this conscious level” (Thompson 1988: 27). For most viewers familiar with the relevant representational, narrative, and cinematic conventions, once they have constructively perceived the two-dimensional images projected on screen, it is just as automatic for them to construct a three-dimensional, fictional world from those images. This is not to say, however, that the second level of construction is straightforward, since the viewer must draw on considerable background knowledge to determine the right configuration of three-dimensional objects captured in what is an inherently ambiguous two-dimensional image and to determine the fictional state of affairs represented by the image. Part of constructing the fictional state of affairs is constructing its spatial and temporal parameters, which extend beyond the image. Thus, for example, when an actor walks out of the frame of the shot, we do not think, without special reason, that her character has ceased to exist; rather we think that the character has moved into part of the fictional world that is not presently shown to us. Insofar as the viewer is constructing fictional events and characters according to the order of images, he or she is constructing what the Russian formalists, and hence 138
TF17557.indb 138
16/9/08 15:47:24
F OR M A LISM
Bordwell, call the syuzhet, or “plot.” The syuzhet is the narrative just as it is shown on screen – as an incomplete and often out-of-order sequence of events. By filling in missing narrative information and reordering the depicted events into a causal sequence, the viewer then constructs the fabula, or “story,” from the syuzhet. Once the viewer has figured out what is going on in a film by constructing the fabula, he or she may wish to continue generating more abstract thematic and symptomatic meanings for the film. The process by which academic critics do this has, according to Bordwell, become highly regulated such that it is governed by institutional norms. Nevertheless, the interpretive process is constructive just as perception is, namely, by being actively inferential. On Bordwell’s analysis, the critic maps concepts, which are structured as “semantic fields,” onto those cues in a film that the critical tradition considers effective in viewers’ comprehension and capable of bearing meaning. The mapping process is aided by “socially implanted hypotheses about how texts mean,” in particular the hypotheses that the text is unified and related to an external world (Bordwell 1989: 133). The mapping process is achieved by the employment of heuristics, standard rules of thumb which have proved useful in generating novel and plausible interpretations. One popular heuristic is the punning heuristic, which involves, for example, taking the depiction of passageways in a film to suggest that the film is about the “passage” of life or certain “rites of passage,” or taking the framing of certain shots to indicate that the film is about the “framing” of innocents. Once the critic has settled on and refined her semantic fields, and organized their application to suggest an overall meaning for the film, she must write up her interpretation in a way that is recognized as suitably persuasive by the interpretive institution. Not surprisingly perhaps, Bordwell’s analysis of interpretation has been very unpopular among academic critics who do not like to think of themselves as something akin to assembly-line workers in “Interpretation, Inc.” However, a deeper criticism of Bordwell’s analysis can be found by examining his prior commitment to constructivism. Although, as Berys Gaut (1995) points out, there may be more than one version of constructivism at work in Bordwell’s account, the basic idea is that the film viewer’s activities are constructive insofar as they are inferential and involve the application of concepts. Since perception, narrative comprehension, and interpretation are all constructive processes, Bordwell assumes that their objects are constructs. Hence his decisive and oft-repeated statement, “Meanings are not found but made” (Bordwell 1989: 3). Unfortunately, however, Bordwell is making a false assumption here. The fact that perception is an active, inferential process does not mean that the objects of perception – namely, the things we perceive around us – are constructed. After all, one can be a realist about the external world and a constructivist about the mind’s activities. Furthermore, a critic who thinks, contra Bordwell, that meanings are not made but found in the text can still allow that the process of finding them is active and inferential. The same critic can also allow that interpretation is governed by institutional norms, since many forms of detection, including criminal and scientific investigation, occur within institutions and yet retain independent objects of inquiry. Further 139
TF17557.indb 139
16/9/08 15:47:24
K AT H E R I N E T HOMS ON-JON E S
evidence against interpretive constructivism is suggested by the fact that interpretation is constrained, not by the norms of the interpretive institution, but by the norms of the film and its generative context. A critic who fails to understand how certain techniques were standardly used in a particular filmmaking tradition is likely to misinterpret their significance. If the critic were the one constructing the interpretation, then there would be no possibility of his being wrong about what a film means. And yet critics are regularly shown to be wrong in light of the social and artistic norms governing a certain film. Interestingly, these are just the kinds of norms to which the Russian formalists are so sensitive. Moreover, as we have seen, Thompson shares the Russian formalists’ view that the backgrounds of a work (and not the critic’s background) determine its formal significance. This suggests a deep tension between Bordwell’s commitment to interpretive constructivism and neoformalism. Perhaps, however, Bordwell is not endorsing interpretive constructivism. Perhaps he is simply describing current interpretive procedure and could readily admit that this procedure often generates false interpretations (see Bordwell 1993). But if this is the case, we have no reason to count constructivism as related to neoformalism. Without the constructivist corollary, what remains of neoformalism is a structural, technical, phenomenological, and historical mode of film analysis, which can be applied to any kind of film with diverse and highly particularized results. Film scholars have tended to spend most of their time figuring out what films mean. For the neoformalist, however, this is just part of a much broader task of figuring out how films work. Films work according to their formal structure, or how they are put together, but they also work differently in different contexts. The neoformalists’ realization of this explains how they can breathe new life into formalism for the sake of redirecting film studies. See also David Bordwell (Chapter 29), Film as art (Chapter 11), Interpretation (Chapter 15), Medium (Chapter 16), Style (Chapter 25), and Spectatorship (Chapter 23).
References Bordwell, D. (1985) Narration in the Fiction Film, Madison: University of Wisconsin Press. —— (1989) Making Meaning: Inference and Rhetoric in the Interpretation of Cinema, Cambridge, MA: Harvard University Press. —— (1993) “Film Interpretation Revisited,” Film Criticism 17: 93–119. Bordwell, D., and Thompson, K. (1979) Film Art: An Introduction, Reading, MA: Addison-Wesley. —— (2006) Film Art: An Introduction, 8th ed., New York: McGraw-Hill. Burbank, J., and Steiner, P. (eds. and trans.) (1977) The Word and Verbal Art: Selected Essays by Jan Mukarˇovský, New Haven, CT: Yale University Press. Gaut, B. (1995) “Making Sense of Films: Neoformalism and its Limits,” Forum for Modern Language Studies 31: 8–23. Nichols, B. (1992) “Form Wars: The Political Unconscious of Formalist Theory,” in J. Gaines (ed.) Classical Hollywood Narrative: The Paradigm Wars, Durham, NC: Duke University Press.
140
TF17557.indb 140
16/9/08 15:47:24
F OR M A LISM
Shklovsky, V. (1965) “Art as Technique,” in L. T. Lemon and M. J. Reis (trans.) Russian Formalist Criticism: Four Essays, Lincoln: University of Nebraska Press. Stam, R. (2000) Film Theory: An Introduction, Malden, MA: Blackwell. Stecker, R. (1997) “The Constructivist’s Dilemma,” Journal of Aesthetics and Art Criticism 55: 43–52. Thompson, K. (1981) Eisenstein’s Ivan the Terrible: A Neo-Formalist Analysis, Princeton, NJ: Princeton University Press. —— (1988) Breaking the Glass Armor: Neoformalist Film Analysis, Princeton, NJ: Princeton University Press.
141
TF17557.indb 141
16/9/08 15:47:24
13
GE ND ER Angela Curran and Carol Donelan “Gender” is a term that refers to the behavioral, social and psychological traits typically associated with being male or female. Since the early 1970s, film critics, theorists, and philosophers have taken up the study of gender in film, engaging theories rooted in psychoanalysis, cultural studies, feminist historiography, and cognitivism. Scholars in film studies and philosophy have been particularly interested in thinking about the relation of viewers and films and whether gender is a central factor in understanding this relation. In what follows, we survey various approaches to the study of gender in film, noting points of convergence and divergence between them.
Surveying the terrain of feminist film theory Image studies The feminist study of gender in film was launched in the early 1970s, with publications such as Marjorie Rosen’s Popcorn Venus and Molly Haskell’s From Reverence to Rape (Rosen 1973; Haskell 1974). These works chronicled the changing image of women in Hollywood and European art film, highlighting how the female characters related to the history of each era, how they were stereotyped – virgin or sex goddess – how active or passive they were, how much screen time they were allotted, and whether they served as positive or negative models for women in the audience. While acknowledged as groundbreaking, “image studies” such as these were also criticized as theoretically unsophisticated. Rosen and Haskell were accused of naïvely assuming that film mirrored or reflected reality without taking into account how film representations are constructed through the conventions of narrative, genre, camerawork, editing, lighting, and so on. Influenced by a heightened political awareness brought on by worker and student revolts in France in 1968, and drawing on theories of structuralism and Marxism, critics associated with influential film journals such as Les Cahiers du cinéma in France and Screen in Britain, in particular, argued that mainstream film, like language and the bourgeois family, shaped the consciousness and beliefs of individuals. These complex theories suggested a more comprehensive analysis of how film works to support gender hierarchies. Feminist film critics drew on these analyses to provide a deeper account of
TF17557.indb 142
16/9/08 15:47:24
GE NDE R
how the formal conventions of film position viewers to accept the values of patriarchal society. Mulvey on visual pleasure and the classical Hollywood cinema One of the most influential essays to emerge out of this discussion was Laura Mulvey’s “Visual Pleasure and Narrative Cinema,” first published in 1975 (Mulvey 2004). In this essay, Mulvey advances a theory of gendered representation and the “masculinization” of spectatorship in the classical Hollywood cinema. Her theory was informed by the ideas of Marxist philosopher Louis Althusser and psychoanalysts Sigmund Freud and Jacques Lacan. Althusser argued that ideology, which asserts itself through cultural and social institutions and practices, “hails” or “interpellates” us. Individuals do not exist prior to or apart from systems of ideology – they are produced as effects of it. Mulvey adapted Althusser’s ideas to argue that the formal conventions of classic Hollywood cinema position the viewer to accept the patriarchal ideology articulated by films. Mulvey also adopted the psychoanalytic theories of Freud and Lacan to argue that the classical Hollywood film is structured according to the logic of the “patriarchal unconscious” and privileges male desire. Men look and women are looked at. The central male protagonist advances the plot through his gaze and actions. The male spectator engages in a narcissistic identification with the male hero, a process that repeats the discovery of an image of oneself in the “mirror phase” postulated by Lacan. The female character is presented as a passive erotic object for the visual pleasure of both the male hero and spectator. Gazing upon the female form – or “scopophilia,” to borrow Freud’s term – is posited in Hollywood films as pleasurable, but in psychoanalytic theory the sight of the female form for the male is also a potential source of displeasure. The female body represents sexual difference and provokes castration anxiety in the male unconscious. Mulvey argues that Hollywood narratives are thus structured to displace or mask the threat posed by sexual difference as exemplified in the female body, either by subjecting that body to voyeuristic punishment (Scottie’s interrogation of Madeleine in Hitchcock’s Vertigo [1958]) or fetishistic idealization (the stylization of Marlene Dietrich in Sternberg’s films). Mulvey ultimately calls for the “destruction of pleasure” associated with the realism of the classical Hollywood cinema. The aesthetic of realism is “illusionistic,” performing the work of ideology in “interpellating” subjects to freely accept their subjectivity and subjection. Mulvey calls on feminists to create an alternative, antirealist cinema that does not demean women.
Feminist film theory post-Mulvey Critical responses to Mulvey’s theory set the agenda for feminist film theory throughout the 1980s. Responses to Mulvey grappled with three problems with her theory. First, central to Mulvey’s theory is her claim that the spectator is constructed 143
TF17557.indb 143
16/9/08 15:47:24
A NGEL A C U R R A N A ND CA ROL D ON EL A N
as an effect of the text. This did not explain how it was possible for actual viewers to engage critically with a film and question its ideological viewpoint, something that feminist film critics clearly did in their responses to mainstream cinema. Second, Mulvey argues that it was not possible for viewers to take pleasure in the narrative structures of classic Hollywood cinema without participating in the gender hierarchies of patriarchy. Feminists argued that Mulvey’s theory of the pleasure of Hollywood films was too general and did not account for the way that certain mainstream movies may subvert rather than support social structures of power. Third, in arguing that narrative Hollywood cinema constructs a normative male viewer, Mulvey’s theory was universalizing and exclusionary. She did not account for how viewers who are excluded from the male norm of spectatorship respond to mainstream cinema. In attempting to analyze how Hollywood film functions as a tool of patriarchy, Mulvey’s theory seemed to support the system of gender inequalities that she hoped to dismantle with her analysis. Some feminist theorists attempted to modify some aspect of the psychoanalytic framework in order to address these problems. As we will see, a challenge for this line of reply is how to accept Mulvey’s central claim – that mainstream cinema constructs the spectator as an effect of the text – yet still account for a range of responses that actual viewers can have to these films. Other responses to Mulvey came from cultural studies, critical race theory, and queer studies. These analyses argued that the relationship between a viewer and a text must be reconfigured if we are to address the problems with Mulvey’s theory. In the following sections, we examine these responses to Mulvey’s theory and analyze their success in proposing alternative accounts of the relationship between film, gender, and the viewer. Psychoanalytic accounts of spectatorship In an afterword to her “Visual Pleasure” essay, Mulvey proposed that the “woman in the audience” could either identify with the passive woman on screen or engage in “trans-sex” identification with the active male hero, neither of which seemed satisfactory as an explanation of women’s cinematic pleasure (Mulvey 1989). Mary Anne Doane argued that certain genres, such as the “woman’s film,” posit a female spectator (Doane 1987a, b). The pleasurable tears of the woman seated in the audience signal her complicity with the film’s organization of vision and desire. At the same time, Doane wants to argue that the feminist viewer, at least, is capable of a “critical moment” in which she steps back and questions the representation of gender in the film (Doane 2000). Other critics working within a psychoanalytic framework challenged Mulvey’s insistence that men cannot be the object of the “gaze.” Steve Neale maintained that the homoeroticism implicit in the male gaze directed at the male body is repressed and disavowed, manifesting in the symptomatic violence accompanying so many representations of the male body in the Hollywood cinema (Neale 1993). In her analysis of the films of Rudolf Valentino, Miriam Hansen revealed how the star’s persona oscillates between active and passive, sadism and masochism, opening up possibilities for 144
TF17557.indb 144
16/9/08 15:47:24
GE NDE R
alternative conceptions of visual pleasure while also posing a challenge to the myth of masculinity in American culture (Hansen 2004). More recent accounts by scholars such as David Gerstner and Peter Lehman examine and critique the way films characterize and enforce ideologies of masculinity (Gerstner 2006; Lehman 2007). Some feminist critics argued that the positioning of the film viewer by the classical Hollywood film is not as monolithic as Mulvey supposed. Drawing upon the psychoanalytic theories of Nancy Chodorow, Linda Williams argued that female identity is formed through a process of double identification, first with her mother, her primary love object, and then also with her father (Williams 1990). A woman’s film such as Stella Dallas (1937) mobilizes the female spectator’s ability to take up multiple identifications, prompting her to empathize with a variety of conflicting points of view. The female viewer of Stella Dallas is not locked into accepting the viewpoint of a single character; she can use the viewpoint of one character to criticize that of another. Elizabeth Cowie turned to Freud’s analysis of fantasy arguing, contra Mulvey, that identification is not determined according to gender and that films such as Now, Voyager (1942) and Reckless Moment (1949) offer multiple points of identification – father, mother, child, lover, wife, or husband – for the viewing subject (Cowie 1997). Assessing psychoanalytic accounts Throughout the 1980s, feminist film theorists used psychoanalytic theories to address a central problem in Mulvey’s theory: her failure to account for the pleasures female spectators take in mainstream Hollywood film. At the same time, critical race theorists such as Jane Gaines and bell hooks argued that psychoanalytic feminist film theory had lost touch with the actual viewers’ lived experience of film (Gaines 1990; hooks 2000). Psychoanalytic accounts primarily focused on the spectator “implied” by the film and did not consider the impact of race, ethnicity, and sexual orientation in an actual viewer’s experience of film. Critics charged that in theorizing about an ahistorical, abstracted “female spectator,” psychoanalytic accounts ironically helped perpetuate the same social arrangements of power that their analyses attempted to dismantle. This neglect of the actual viewer’s response to film has roots in the theoretical commitments of most psychoanalytic feminist film theories. Following Althusser, Mulvey posits that film assumes or “interpellates” a male spectator, producing him as an effect of the film. Feminist film theorists have challenged this account on multiple levels, but without critiquing the assumption that the viewing subject is an effect of representation and is caught up “within the language of patriarchy” (Mulvey 2004: 838). Indeed, as Jennifer Hammett argues, if women are “constituted as subjects by patriarchal representations,” they “do not have the epistemic resources necessary to escape patriarchy” (Hammett 1997: 245). They are not able to adopt the critical standpoints identified by Doane and Williams. In her analysis of Stella Dallas Williams seems to want to concede the point that Ann Kaplan makes in her reply to Williams’ account: that as viewing subjects, women viewers are constituted as subjects within the discourses of representation, including 145
TF17557.indb 145
16/9/08 15:47:24
A NGEL A C U R R A N A ND CA ROL D ON EL A N
those of film (Kaplan 1985). But she appeals to Chodorow to argue that patriarchy constructs women differently than men, as “fragmented” viewers who have a more complex and contradictory stance toward the world. It is not clear, however, how women viewers can develop a critical response to the representations they see on screen if Williams concedes the underlying assumption – made by Mulvey, Doane, Kaplan, and other psychoanalytic accounts – that the viewer is a product or construct of patriarchal “discourse.” Cultural studies, critical race theory, queer theory Dissatisfaction with the psychoanalytic account of an ahistorical and abstracted spectator prompted feminist and other film theorists to turn to cultural studies in the late 1980s and 1990s. Central to cultural studies is the idea that cultural institutions, such as cinema, and cultural practices, such as textual interpretation, are sites of political struggle. Drawing on the work of Stuart Hall, among others, critics argued that meaning is not inherent in the film text but is produced in the interaction with viewers. Viewers are themselves constituted as subjects within a complex system of “discourses” circulating in society, including the representations in the film. Rejecting the ideas of Althusser in favor of those of the Italian Marxist Antonio Gramsci, critics argued that viewers are capable of resisting the dominant ideology presented in mainstream films. This is not so much because the viewer is a “free agent” who is not influenced by cultural and social forces, but for the reason that systems of representation like popular Hollywood films are not homogeneous in their meaning but are capable of multiple interpretations. An attentive viewer can “negotiate” with the film text and “read against the grain” to bring out strands or places in the film where the dominant viewpoints break down. Manthia Diawara, for example, developed a theory of black male “resisting” spectatorship, based on his own experience as a black man watching mainstream films (Diawara 2004). bell hooks made a case for the black female spectator’s “oppositional” gaze as a basis for critical analysis and active resistance against racist representations in film (hooks 2000). Alexander Doty identified queerness as a viewing or reception position that could be taken up by queers and nonqueers alike in relation to homophobic texts (Doty 1993). Criticisms informed by cultural studies, critical race, and queer theory were attempts to direct feminist film theorists to take account of viewing experiences excluded from Mulvey’s analysis. In pointing to the wide range of response available to viewers of mainstream films, cultural studies theorists argue that a film can subvert gender norms if the viewer “negotiates” with the text in the right way. But not all films challenge social arrangements of power or do this in the same way. Feminists should be interested in analyzing which films and formal structures support social arrangements of power and which do not. The philosophers we examine in the third section of the paper argue that another analysis of the relationship between the viewer and a film is needed. Such an account would be able to show how formal features in a film can cue viewers to adopt an ideology implicit in the film, but also explain how a viewer is capable of an active, critical response to such a film. 146
TF17557.indb 146
16/9/08 15:47:24
GE NDE R
The advent of feminist historiography A “historical turn” in film studies in the 1990s prompted a new generation of feminist film scholars to investigate the historically specific relationships between viewers and films. The availability of digitized historical materials, especially those related to the era preceding the classical Hollywood cinema, have prompted feminist film scholars to rethink their assumptions about early cinema and women’s participation in it (Rabinovitz 1998; Stamp 2000; Bean and Negra 2002). This work has also investigated the foundational position that women had in all roles in the beginning of the film industry, challenging conventional film histories. While feminist historiography breaks with the psychoanalytic account of an abstracted spectator it also continues to be “catalyzed by questions of spectatorship, ideological coding, and cultural interpolations that persist from earlier conversations” (Bean and Negra 2002: 4). Film philosophers argue that a more direct examination of the basic theoretical assumptions of psychoanalysis, as well as of cultural studies, is needed in order to open up the analysis of gender and film to new directions of study. In the final section of the paper, we examine these criticisms as well as some other ideas of philosophers who have analyzed the topic of gender and film.
Philosophical approaches As film studies came of age in the 1980s, philosophers and allies in film studies called “cognitivists” began to engage critically with the view that films mobilize unconscious processes in the viewer. Philosophers Thomas Wartenberg, Michael Ryan, and Douglas Kellner also interrogated the dominant view in film studies that the formal conventions of classic Hollywood films perpetuate social arrangements of power (Kellner and Ryan 1988; Wartenberg 1999). Although his analysis was not explicitly focused on gender, Stanley Cavell examined certain screwball comedies and women’s melodramas from the 1930s and 1940s (Cavell 1981, 1996). His argument – that these films represent women in the pursuit of their desires – challenged Mulvey-influenced accounts that the conventions of classic Hollywood films do not allow for the representation of female desire (Scheman 1995). In this section we examine several of the salient criticisms and issues that have emerged out of these discussions. Link of filmic conventions with ideology Mulvey argued that the “gaze” of the camera in classic Hollywood film commonly aligns with the “look” of the central male protagonist so that the viewer adopts the desires of this character. The male characters in Hollywood film desire the woman as a passive recipient of male desire, making the viewer participate in the structures of patriarchy encoded in the film narrative. Mulvey’s argument about the gaze could be interpreted in two ways: that the male gaze is inherent to the formal structures of Hollywood film or that there is a historical use of these structures to position women as passive objects of male desire. The latter seems to be a more plausible construal of her argument. Mulvey seems to conclude 147
TF17557.indb 147
16/9/08 15:47:24
A NGEL A C U R R A N A ND CA ROL D ON EL A N
that it is not possible to separate the historical use of Hollywood filmic conventions from the intrinsic features of film as an artistic medium. Mulvey argues that feminist filmmakers must develop an avant-garde cinema that eschews narrative realism, emotional engagement with central protagonists, and so on. But it is not clear why the formal conventions of Hollywood cinema might not be able to prompt a pleasure in viewers that is not based in structures of domination. Noël Carroll has argued this in a discussion of women characters from Hollywood screwball comedies such as Bringing Up Baby (1938) and His Girl Friday (1940), as has Kathleen Rowe in her analysis of the “unruly woman” of the genres of comedy (Carroll 1995; Rowe 1995). Berys Gaut proposes that the sexism prevalent in mainstream film is explained by the “sexism of the surrounding society, the scarcity of women directors, [and] the almost total absence of female producers and financiers,” not by “the structure of narrative, ‘illusionistic’ film per se” (Gaut 1994). Gender and the viewer’s emotional response to film Some feminist philosophers of art have argued that Mulvey’s account is useful because she exposes the way in which a dominant art is not epistemologically neutral but represents the social vantage point of the dominant (masculine) social class (Hein 1995). But others have been highly critical of Mulvey-influenced feminist film theorizing. Cynthia Freeland criticizes feminist film theorists’ use of psychoanalysis (Freeland 1998; see also Shrage 1993). Like Noël Carroll, she argues that feminists interested in analyzing films’ relation to a patriarchal society have appealed to psychoanalysis without arguing that this is superior to other theoretical frameworks (Carroll 1995). She also thinks an alternative is needed that enables us to understand how viewers are capable of active, critical responses to the ideological viewpoints in film. Freeland proposes that new work being done in the philosophy of film can offer new, nonpsychoanalytic explanations for the pleasures of mainstream film. This work is grounded in the methodology of cognitivism, a dominant thread in the philosophy of film over the last fifteen years or so. Cognitivists propose to use current studies of perception and cognitive science to understand how films engage cognitive processes such as inference making, hypothesis testing, and perceptual and information processing. They hold, for the most part, that these theories offer a much better explanation of the way actual viewers engage and find pleasure in films than the theories offered by psychoanalysis. Cognitivists tend to argue that emotions are “cognitive appraisals” of situations and as such they need not be irrational; rather they are cognitive processes that we use to make sense of our everyday encounters with the world. We use these same cognitive processes in our response to film (Plantinga and Smith 1999). Cognitivists have drawn on analyses of emotional response to explain some of the same issues with which psychoanalytic feminist film theorists engage. In an updated version of the “image studies” approach Noël Carroll has argued that emotions are forms of multifaceted behavior acquired from “paradigm scenarios” and that movies can be a source of these scenarios (Carroll 1995). From Fatal Attraction, for example, a 148
TF17557.indb 148
16/9/08 15:47:24
GE NDE R
male viewer might learn that the “scenario” for the appropriate emotional response to an ex-lover’s demand for fair treatment is to distance her as irrational and not deserving of his attention. In her analysis of gender and the genre of the horror film, Freeland argues that horror films prompt the emotions of fear, sympathy, dread, anxiety, and disgust and in doing so provoke reflection about patriarchy and its institutions as well as the nature of evil (Freeland 2000). Looking at the subgenre of melodrama known as the woman’s film, Flo Leibowitz analyzes how the response of pity and admiration that male and female viewers feel toward the characters can be part of a coherent assessment of the women protagonists’ values and priorities (Leibowitz 1996). These analyses are useful for several reasons. First, they point to an area that has for the most part been overlooked by feminist film theorists: the way that films mobilize emotions other than pleasure or desire (Plantinga and Smith 1999). Feminist film theorists have tended to think that all emotional engagement with characters in film must implicate viewers in submitting to structures of domination (as with Doane’s analysis of women’s melodrama). But emotional engagement with characters in film need not make us acquiesce to a questionable system of social arrangements. When emotions are engaged as part of a critical reflection on a movie’s characters and their predicaments, they can lead to a rational reassessment of such an ideology (Curran 2005). Second, cultural studies theorists attempt to account for critical viewing, but they often neglect how the formal features of a film can prompt a viewer’s critical engagement with what she or he sees on screen. A cognitive model of our engagement with film promises to help us understand how the formal features of film can either prompt viewers to accept the ideology implicit in the film or encourage the viewer to engage in a critical response to the characters and their situations. Future work on these issues of emotional engagement and critical viewing promises to open up discussions on several important but neglected aspects of feminist film theorizing.
Conclusion There are no doubt sharp points of difference between feminist historiography and film philosophy. Most notably, there is a dominant trend in film philosophy that rejects the use of psychoanalysis as an overly general “grand theory (Bordwell and Carroll 1996).” On the other hand, arguably there is agreement between feminist historiographers and cognitivists that it is best to eschew generalizations about abstracted spectators and focus on theorizing that is more local and applicable to specific contexts of film viewing. A charge often made in film studies is that cognitive theorists are, in effect, hoisted by their own petard. Cognitivists criticize the psychoanalytic accounts of abstracted “spectators” that bear little relation to active, cognizing viewers, but they also invoke the raceless, classless, and genderless “schemas” of an ahistorical mind as they theorize about viewer’s engagement with film (Stam 2000: 241). It might then seem that cognitivist approaches to gender and film are at odds with feminist historiographers’ emphasis on more historically and socially situated methods of investigation. Coming from two different disciplinary perspectives, philosophy and film studies, we think there is much we can learn from one another. Cognitivists might 149
TF17557.indb 149
16/9/08 15:47:25
A NGEL A C U R R A N A ND CA ROL D ON EL A N
take into consideration the emphasis in feminist historiography on a socially and culturally specific viewer. Feminist historiographers might study cognitivism to learn how recent theories of the mind and cognition can open up new avenues of investigation as we continue to explore the complex relationships between film, the viewer, and social arrangements of power. See also Empathy and character engagement (Chapter 9), Psychoanalysis (Chapter 41), Race (Chapter 21), Stanley Cavell (Chapter 32), Spectatorship (Chapter 23), and Pornography (Chapter 47).
References Bean, J. N., and Negra, D. (eds.) (2002) A Feminist Reader in Early Cinema, Durham, NC: Duke University Press. Bordwell, D., and Carroll, N. (eds.) (1996) Post Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Carroll, N. (1995) “The Image of Women in Film: A Defense of a Paradigm,” in P. Brand and C. Korsmeyer (eds.) Feminism and Tradition in Aesthetics, University Park, PA: Pennsylvania State University Press. Cavell, S. (1981) The Pursuits of Happiness: The Hollywood Comedy of Remarriage, Cambridge, MA: Harvard University Press. —— (1996) Contesting Tears: The Hollywood Comedy of Remarriage, Chicago: University of Chicago Press. Cowie, E. (1997) Representing the Woman: Cinema and Psychoanalysis, Minneapolis: University of Minnesota Press. Curran, A. (2005) “Stella at the Movies: Class Critical Spectatorship and Melodrama in Stella Dallas,” in T. Wartenberg and A. Curran (eds.) The Philosophy of Film: Introductory Text and Readings, Malden, MA: Blackwell. Diawara, M. (2004) “Black Spectatorship: Problems of Identification and Resistance,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism, 6th ed., New York: Oxford University Press. Doane, M. A. (1987a) “The Woman’s Film: Possession and Address,” in C. Gledhill (ed.) Home Is Where the Heart Is: Studies in Melodrama and the Woman’s Film, London: British Film Institute. —— (1987b) The Desire to Desire: The Woman’s Film of the 1940s, Bloomington: Indiana University Press. —— (2000) “Film and the Masquerade: Theorizing the Female Spectator,” in R. Stam and T. Miller (eds.) Film and Theory: An Anthology, Malden, MA: Blackwell. Doty, A. (1993) Making Things Perfectly Queer: Interpreting Mass Culture, Minneapolis: University of Minnesota Press. Freeland, C. (1998) “Feminist Film Theory,” in M. Kelly (ed.) The Encyclopedia of Aesthetics, New York: Oxford University Press. —— (2000) The Naked and the Undead: Evil and the Appeal of Horror, Boulder, CO: Westview Press. Gaines, J. (1990) “White Privilege and Looking Relations: Race and Gender in Feminist Film Theory,” in P. Erens (ed.) Issues in Feminist Film Criticism, Bloomington: Indiana University Press. Gaut, B. (1994) “On Cinema and Perversion,” Film and Philosophy 1: 3–17. Gerstner, D. (2006) Manly Arts: Masculinity and Nation in Early American Cinema, Durham, NC: Duke University Press. Hammett, J. (1997) “The Ideological Impediment: Epistemology, Feminism and Film Theory,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Clarendon; and New York: Oxford University Press. Hansen, M. (2004) “Pleasure, Ambivalence, Identification: Valentino and Female Spectatorship,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism, 6th ed., New York: Oxford University Press. Haskell, M. (1974) From Reverence to Rape: The Treatment of Women in the Movies, New York: Holt, Rinehart, and Winston.
150
TF17557.indb 150
16/9/08 15:47:25
GE NDE R
Hein, H. (1995) “The Role of Feminist Aesthetics in Feminist Theory,” in P. Brand and C. Korsmeyer (eds.) Feminism and Tradition in Aesthetics, University Park, PA: Pennsylvania State University Press. hooks, b. (2000) “The Oppositional Gaze: Black Female Spectators,” in R. Stam and T. Miller (eds.) Film and Theory: An Anthology, Malden, MA: Blackwell. Kaplan, A. (1985) “Dialogue: E. Ann Kaplan Replies to Linda Williams’ ‘Something Else besides a Mother’: Stella Dallas and the Maternal Melodrama,” Cinema Journal 24: 40–3. Kellner, D., and Ryan, M. (eds.) (1988) Camera Politica: The Politics and Ideology of Contemporary Hollywood Film, Bloomington: Indiana University Press. Lehman, P. (2007) Running Scared: Masculinity and the Representation of the Male Body, new ed., Detroit: Wayne State University Press. Leibowitz, F. (1996) “Apt Feelings or Why ‘Women’s Films’ Aren’t Trivial,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Mulvey, L. (1989 [1981]) “Afterthoughts on ‘Visual Pleasure and Narrative Cinema’ inspired by King Vidor’s Duel in the Sun,” in Visual and Other Pleasures, Bloomington: Indiana University Press. —— (2004) “Visual Pleasure and Narrative Cinema,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism, 6th ed., New York: Oxford University Press. Neale, S. (1993) “Masculinity as Spectacle: Reflections on Men and Mainstream Cinema,” in S. Cohan and I. R. Hark (eds.) Screening the Male: Exploring Masculinities in Hollywood Cinema, London and New York: Routledge. Plantinga, C., and Smith, G. M. (eds.) (1999) Passionate Views: Film, Cognition, and Emotion, Baltimore, MD: Johns Hopkins University Press. Rabinovitz, L. (1998) For the Love of Pleasure: Women, Movies, and Culture in Turn-of-the-Century Chicago, New Brunswick, NJ: Rutgers University Press. Rosen, M. (1973) Popcorn Venus: Women, Movies and the American Dream, New York: Coward, McCann, and Geoghegan. Rowe, K. (1995) The Unruly Woman: Gender and the Genre of Laughter, Austin: University of Texas Press. Scheman, N. (1995) “Missing Mothers and Desiring Daughters: Framing the Sight of Women,” in C. Freeland and T. Wartenberg (eds.) Philosophy and Film, New York: Routledge. Shrage, L. (1993) “Feminist Film Aesthetics: A Contextual Approach,” in H. Hein and C. Korsmeyer (eds.) Aesthetics in Feminist Perspective, Bloomington: Indiana University Press. Stam, R. (2000) Film Theory: An Introduction, Malden, MA: Blackwell. Stamp, S. (2000) Movie-Struck Girls: Women and Motion Picture Culture after the Nickelodeon, Princeton, NJ: Princeton University Press. Wartenberg, T. (1999) Unlikely Couples: Movie Romance as Social Criticism, Boulder, CO: Westview Press. Williams, L. (1990) “ ‘Something Else besides a Mother’: Stella Dallas and the Maternal Melodrama,” in P. Erens (ed.) Issues in Feminist Film Criticism, Bloomington: Indiana University Press.
151
TF17557.indb 151
16/9/08 15:47:25
14
GE NR E Brian Laetz and Dominic McIver Lopes Genres like film noir, fantasy, and action are readily and reliably used by movie audiences and film scholars. Renting a DVD means navigating the genres, and genres belong in the tool kit of many film scholars. Genre criticism is a major mode of film studies, and it has spawned a rich literature that touches on theoretical questions of what movie genres are, how they function, and how they should best be studied (e.g., Grant 2003; see also Duff 2000). But although philosophers have written about specific movie genres like horror (Carroll 1990) and melodrama (Cavell 1996), they have paid little attention to the nature of film genre – or indeed of any art genre (the exception is Currie 2004). It is time to attempt a philosophy of movie genres that takes advantage of and potentially enriches related work in film studies.
Toward a philosophy of movie genre The first step is to get a fix on what a philosophy of genre is intended to explain. That means understanding roughly how genre categories function, because a rough idea of the work that genre does in our thinking about movies can guide us as we work toward a philosophy of genre that explains genre as doing the job it does (see also Tudor 2003). Two tasks are obvious: genres help with interpretation and with appreciation. Movie audiences deploy genre concepts as they interpret and appreciate, and filmmakers also work with an eye to genre as long as they aim to make movies to be interpreted and appreciated. Film scholars have documented in some detail how specific genres set audience expectations and thereby drive interpretation (e.g., Altman 2003; Sobchack 2003; and Neale 2003). A useful way to represent these expectations builds on David Lewis’ (1978) account of truth in fiction, especially as it was developed by Kendall Walton (1990). Parts of the story a movie tells are not normally told explicitly, but are implied instead. When it shows her using a telephone, Dial M for Murder (1954) implies that Margot is speaking to someone. The implication is sustained by a Reality principle: a story represents that q if it explicitly represents that p1 . . . pn and it would be the case that q, were it the case that p1 . . . pn.
TF17557.indb 152
16/9/08 15:47:25
GE N R E
In reality, people using telephones are talking to someone, so Margot is talking to someone. What a movie explicitly represents may join with a fact about reality to fill in the story. Not always, though. In some of her movies, Katharine Hepburn plays an independent woman and it is implied that her prospects of marriage are poor. (Exploiting this implication, the movies get the audience to worry for her.) Of course, in reality, independent women do not have poor marriage prospects. Another principle generates the needed implication: Mutual-belief principle: a story represents that q if it explicitly represents that p1 . . . pn and it is mutually believed among members of the story’s original target audience that it would be the case that q, were it the case that p1 . . . pn. For much of the last century it was mutually believed by American movie viewers that an independent woman would probably not marry. This belief of the target audience joins with what the movies represent explicitly to imply that Hepburn’s characters face a poor chance of marriage. However, a third principle is also needed, and it brings in genre. Imagine a movie, Dragon Story, which represents a dragon, Snuffles, who is rather timid and never works up much anger. Snuffles breathes fire, although he is never explicitly shown engaged in this activity. What warrants the implication that he breathes fire? Not the reality principle, since it is not true in reality that a creature would breathe fire were it a large, flying serpent. Not the mutual-belief principle either, because movie audiences do not believe that a creature would breathe fire were it a large, flying serpent. What audiences do know is that the movie belongs to the fantasy genre and in this genre dragons breathe fire (see McArthur 1972: 23, for a detailed example). The third principle could be put roughly like this: Genre principle: a story belonging to genre K represents that q if it explicitly represents that p1 . . . pn and it is a feature of K that it would be the case that q, were it the case that p1 . . . pn. This formulation leaves it open what the feature in question is. Many film scholars follow Robert Warshow’s (1962) view that the feature in question is a convention or a formula, though other views are available (e.g., Neale 2003). Any formulation must explain how we can miss the story a movie tells when we get its genre wrong. Put another way, the genre principle explains how a movie can be interpreted as telling different stories when it is viewed in different genres. Genres also figure in movie appreciation. Part of appreciating a movie is attributing aesthetic properties to it. Walton (1970) showed in a famous thought experiment that attributions of aesthetic properties are category relative, and his argument can be adapted to genres. F.W. Murnau’s Nosferatu (1922) is a classic horror movie. Imagine a genre –call it the “nosferatu” genre – which is made up of movies each edited together from the same shots as Nosferatu. By definition, Nosferatu is a nosferatu, but the 153
TF17557.indb 153
16/9/08 15:47:25
BR I A N L A ET Z A ND D OM I NIC MCI V E R LOPE S
aesthetic properties we attribute to Nosferatu depend on whether we view it as horror or as a nosferatu. Viewed as horror, it is provocatively, deliciously creepy. Viewed as a nosferatu, it is tediously linear and predictable. So the aesthetic properties we attribute to any movie depend on the genre to which we assign it. Viewed as a buddy/road movie, Thelma & Louise (1991) is moving, though it stagnates when viewed as an action movie. As Walton points out, this means that we can rarely get a movie’s aesthetic properties wrong. West Side Story (1961) is hilarious: true or false? True unless it is viewed as a musical, in which it is conventional for characters to burst into song. Otherwise, it might come across as silly that characters sing and dance in public. So we cannot be wrong in saying that West Side Story is hilarious –only in saying that it is hilarious when viewed as a musical. Walton solves this problem by identifying the aesthetic properties a work actually has with those it has when viewed in the correct genre. Thus appreciating a movie involves honing in on its genre and seeing what aesthetic properties it has in that genre. Genre is not merely a cataloging scheme. It is a cataloging scheme that performs some useful tasks: recognizing a movie’s genre helps the audience to fill in its story and it provides a background against which to view its aesthetic properties. In constructing a philosophy of genre, we are not in the difficult position of having to model inchoate, possibly irreconcilable intuitions generated from the folk concept of genre. A good philosophy of genre will explain the work that genre does in interpretation and appreciation. Like a philosophy of art more generally, a philosophy of genre features three main elements: a theory, an ontology, and a value theory.
Theory: which categories are movie genres? Some movies are animations based on computer-generated imagery (CGI), some are realistic, some are film noir, some are released in 1982, and some are Toho Company productions. Each of these categories belongs to a different metacategory. One metacategory is the art forms, whose subcategories are motion pictures, literary fiction, music, and the like. Another is art media like film, video, language, pastel, the chromatic scale. Going more fine grained, we get oeuvres like those of Eastwood and Fellini. Film and some other arts have traditions such as Hollywood and Bollywood; and some arts have styles like realism or postmodern mannerism. Standing apart from all of these is the metacategory of genre, which includes film noir, epic, satire, and many other genres. So, then, which categories belong to the metacategory of genres? Why is film noir a genre and the Toho backlist not? Why is satire a genre but not video? An answer to these questions is a theory of movie genre, which completes the schema: Category K is a movie genre if and only if . . . A completion that gives individually necessary and jointly sufficient conditions is a definition. Anything else is a nondefinitional theory of movie genre. 154
TF17557.indb 154
16/9/08 15:47:25
GE N R E
A theory of movie genre is not the same as theories of the individual movie genres. What makes a movie a satire? A musical? A gangster movie? Answers to these and similar questions complete the following schema: Movie x belongs to movie genre K if and only if . . . As we will see below, theories of the movie genres must be worked out one at a time. A theory of movie genre does not deliver up a readymade theory of each individual genre. An example of a nondefinitional theory of movie genre is a disjunctive theory, which states that something is a movie genre if and only if it is horror or suspense or romance or action, etc. This theory is a last resort, since it simply lists categories that are genres and refuses to say what makes them genres. This problem is especially serious if the list includes uninstantiated genres. No detective movies are set in ancient Egypt, but is the category of detective movies set in ancient Egypt a genre? If the answer is “yes” then the genre goes on the list. No detective movies feature murders in Iceland on a Tuesday, but is this category also a genre? If not, why not? It is hardly satisfying to be told nothing about why one category makes it onto the list of genres and the other does not. Perhaps, in the end, a disjunctive theory is the best we can do –some writers suspect as much (e.g., Buscombe 2003). Nevertheless, the first step is to seek something more ambitious –a definition (or a cluster theory modeled on Gaut 2000). A definition of movie genre specifies conditions that are individually necessary and jointly sufficient for a category to be a movie genre. It specifies what horror, suspense, romance, and the other genres share in common and what distinguishes them from other categories. Specifying these features would explain what makes some categories movie genres. To begin with, every movie genre has more than one member. This is not true of every movie category. For example, films that are identical to Midnight Cowboy (1969) is a category that has only one member. Indeed, any work belongs to a category of which it is the only member. We may not think of these as categories, but if not, let us be explicit about this. Likewise, movies in a genre are made by more than one artist. This is not true of all movie categories. For example, movies directed by Alfred Hitchcock is a category but not a genre – it is an oeuvre. Complicating matters, movie artists include directors, producers, actors, cinematographers, and so forth (Gaut 1997), and movies can be categorized according to who occupies each of these artist roles. Rear Window (1954) belongs to the categories of movies-that-star-Jimmy-Stewart and movies-directedby-Alfred-Hitchcock. However, movies-that-star-Jimmy-Stewart is not a genre, and neither is a complex category like movies-directed-by-Alfred-Hitchcock-starringJimmy-Stewart. So, a category is a movie genre only if more than one artist occupies any given role in making movies in that genre. Artists from any background make movies in a genre. In principle, anyone can make a comedy or action film, but this is not true of every movie category. Only 155
TF17557.indb 155
16/9/08 15:47:25
BR I A N L A ET Z A ND D OM I NIC MCI V E R LOPE S
people who in some sense belong to Bollywood can make a Bollywood film. This difference between movie genres and traditions is sharply revealed by two explanatory asymmetries. Consider the genre of suspense and the tradition of Hollywood films. What makes The Sixth Sense (1999) suspense is not that M. Night Shyamalan made it. Rather, he is a suspense filmmaker because he makes suspense films. By contrast, what makes The Sixth Sense a Hollywood film is simply that Shyamalan made it under the auspices of Hollywood, and his doing so is what makes him a Hollywood filmmaker. Quite unlike traditions, movie genres are not defined in terms of social facts about their producers. Admittedly, there are tough questions regarding the nature of social facts and traditions, but since our focus is defining movie genre, we need not answer them here. From here, things get tricky. A cursory glance at movie genres reveals much diversity. Some are partly defined by setting (western), some by subject (war), others by affect (comedy), some by format (musicals), and still others by style (film noir). Yet some settings (rural New Jersey), subjects (lawn mowers), affects (calmness), and styles (surrealism) do not help define any movie genre. What distinguishes these features from the features that do help define movie genres? The answer is that the features of movie genres that help define them are ones that factor into their role in appreciation and interpretation. These features include ones that impact a large group of audience members’ decisions about whether they want to view a film. Many people sometimes want to see movies with the historical setting that helps to define the western, but knowing that a film is set in New Jersey has little if any impact on most viewing decisions. Moreover, movie genres set audience expectations for purposes of appreciation and interpretation. War movies are partly defined by their subject matter because their subject matter factors in our appreciation of them. Knowing that Saving Private Ryan (1998) is a war movie, we do not judge its violence to be a flaw in it, as we would if we thought it was a screwball comedy. Since the Three Stooges’ Disorder in the Court (1936) is a screwball comedy (and not a courtroom drama), you anticipate that justice will prevail, acquitting the innocent Gail Tempest, no matter how much mayhem occurs in the witness box. Comedies are partly defined by their happy affect because that affect helps set our interpretive expectations. All this can be plugged into the schema for a theory of movie genre: Category K is a movie genre if and only if K has multiple members, which are made by more than one artist (for any given artist role), from any background, and K has features in virtue of which K figures into the appreciations or interpretations of K’s audience. This formulation does not imply that every movie belongs to no more than one movie genre – it allows for genre hybrids like action comedies and war film noir (Staiger 2003). It also allows for complex movie categories combining movie genres and other categories, such as thrillers directed by Hitchcock, Hong Kong action cinema, and classic comedies. Finally, it allows that a movie can migrate from one genre to another, if movie genres change over time. 156
TF17557.indb 156
16/9/08 15:47:25
GE N R E
This is a proposal for a definition of movie genre. It should be thoroughly tested to see if there are counterexamples. Movies that are three hours long, R-rated movies, movies with happy endings, and movies with foreign dialogue meet all the conditions laid out in the proposed schema, but it is a stretch to think of them as genres. One option is to bite the bullet and count them as genres, revising any contrary intuitions, so that the proposal states necessary and sufficient conditions. Another option is to side with intuitions, and admit that the definition only states necessary conditions and hence that more work is needed to identify sufficient conditions. This is not bad news so long as addressing counterexamples will refine and possibly complete the definition. In evaluating the proposed definition, it is also helpful to note which questions it raises and which it rules out, since a good theory should raise the right questions. Thus this proposal quashes the question, “why do movie genres impact our aesthetic decisions or judgments?” After all, it just defines movie genres as categories that impact our aesthetic decisions and judgments about movies. At the same time, the proposal raises questions worth raising. For example, it accommodates variation in the kinds of features that help define movie genres – format helps define musicals, affect helps define horror, style helps define film noir, and setting helps define westerns. But why does the setting of the western impact aesthetic decisions and judgments? What is the interpretive and appreciative role of affect in horror movies? Of course, these are the very kinds of questions that interest film scholars and philosophers writing about the individual movie genres (e.g., Kitses 1969; Carroll 1990).
Ontology: Are there any movie genres? What sort of entity is a movie genre? Is it a material object, a property, an event, or an abstractum, like a set or a type? Or maybe it is none of these? A theory of movie genre does not supply the answer. Suppose that movie genres are sets of movies. That does not tell us which sets of movies are genres –movies released in 1962 form a set which is not a movie genre. And suppose movie genres are defined in part by characteristic emotional effects. That says nothing about whether movie genres are sets or types or individuals – any of which can be defined by emotional effects. To find out what sort of entity is farce or horror we need an ontology of movie genre. Ontologies of movie genre are readily adapted from the rich literature on the ontology of multiple-instance artworks, especially musical works (Rohrbaugh 2005). Philosophers have identified musical works with sets, types, and historical individuals; and some have even denied that there are any musical works. At the same time, a consensus has developed about how to choose among the options: a good ontology of art is implicit in sound appreciative practices (Thomasson 2005). A similar principle can help choose the correct ontology of movie genre: a conception of what movie genres are should reflect their role in interpretation and evaluation. Ontologies of movie genre divide into realist ontologies, according to which movie genres exist, and nominalist ontologies denying that they exist (after Goodman 1976). According to nominalists, there is no satire or film noir. Of course, there are movies that we classify as satire and film, but “satire” and “film noir” are labels that we apply 157
TF17557.indb 157
16/9/08 15:47:25
BR I A N L A ET Z A ND D OM I NIC MCI V E R LOPE S
to some movies, so the movies exist and the labels exist but the genres do not. By denying that movie genres exist, this view deftly sidesteps the tricky problem of saying what kind of entity a movie genre is; but the viability of this view depends on its making sense of the role movie genre plays in interpretation and appreciation, and the prospects do not look good. A label like “satire” does not invoke conventions which carry us from what is represented explicitly to what is implied, and it does not delineate a comparison class as a background against which to attribute aesthetic properties. According to a simple realist ontology, movie genres are sets. The trouble is that the members of a set make it the set that it is. Take the set of all satirical movies released before 1962. Add a new satire and you get a new set. Satire pre-1962 is not the same movie genre as satire post-1962. Again, this does not make sense of the role of movie genre in interpretation and appreciation. The comparison class for appreciating The Great Dictator (1940) is the same as the comparison class for Dr. Strangelove (1964). Indeed, the comparison class includes possible movies as well as movies that have been made. Part of appreciating The Great Dictator is seeing how bad similar movies might have been. An improved realist proposal is that movie genres are types. Each type has a typeconstitutive feature, a feature that items must have in order to be tokens of the type. Thus Richard Wollheim (1980) argues that musical works are types constituted by a sonic structure which any performance must have in order to be a performance of the work. Perhaps, then, movie genres are types whose tokens are movies having whatever feature constitutes the type. (What feature is that? A theory of each movie genre supplies the answer.) The advantage of this proposal is that the identity of a genre is secured not by its tokens but by a feature that they have. Movies actually made in a movie genre are not essential to it. Thus the comparison class in appreciating a movie in a genre may go beyond movies actually made in the genre. These three proposals demonstrate how to go about developing an ontology of genre and how to argue for it by appealing to facts about the role of genre in interpretation and appreciation. They are merely starting points, however. Alternative proposals should be formulated and assessed. The process will net a better understanding of the role of genre in practices of film spectatorship and criticism, which may prove instrumental in reaching the right ontology of genre.
Value: what’s worth got to do with it? A theory of genre and value centers on at least three questions. First, some critics take the fact that a movie is a “genre movie” to be a mark against its quality. The best movies, they say, are not genre movies. Is this a sound critical principle? Second, some people rank movie genres: they typically rank action low and drama higher, comedy as lower brow than satire, and tragedy as higher brow than the tearjerker. Are such rankings justifiable? Finally, Charlie Chaplin’s The Great Dictator is a superb instance of the genre of satire, whereas Jonathan Demme’s 2004 remake of The Manchurian Candidate (1962) fails as a satire. This raises the question, what is it for a movie to be good or bad as a K, where K is a genre? 158
TF17557.indb 158
16/9/08 15:47:25
GE N R E
Sometimes auteur films, like those of Woody Allen, are contrasted with genre movies and the former are sometimes deemed better than the latter, because they are more creative and less formulaic. This argument is suspect on multiple grounds. Originality is just one meritorious feature of a film, so a formulaic movie may surpass more original fare on other grounds. Indeed, originality can mar a movie –a homage, for instance. Moreover, it is questionable whether genre movies are automatically less creative. The first movies in any genre can be just as innovative as any movie that allegedly lacks a genre, and creativity can thrive even when subject to constraints (Boden 2004). Adding to these general points, some film scholars have defended the merits of particular movie genres and of particular genre movies (e.g., Cawelti 1976; Braudy 1999; Sobchack 2003). Finally, and more fundamentally, it is not clear that any movie is in fact genreless. Annie Hall (1977) is still a comedy, albeit an exceptional one. Indeed, it is particularly exceptional when compared with other comedies. So perhaps every movie belongs to some genre, with drama currently acting as a sort of catchall for everything outside more fine-grained movie genres, like horror or action. People commonly rank musical genres (saying that punk is better than disco, for example), and some also rank movie genres. Right off the bat, there is the problem of what sample of movies in a genre determines its ranking. We might look at the total number of good movies in each movie genre or the proportion of good to bad movies in the genre. Defenders of a movie genre will object that it is a matter of luck whether the true potential of a movie genre is exploited – maybe some genres attract less accomplished filmmakers. Alternative sampling methods use ideal or typical movies in each movie genre. But assuming that the sampling problem can be solved, is drama really better than action and comedy? The answer to this question depends on how movie genres are evaluated. Obviously movie genres differ in instrumental value. Horror movies are better at frightening audiences than comedies, tearjerkers are better at saddening viewers than martial arts epics, and historical dramas are better at educating about the past than fantasies. Each genre wins when judged by its own ends. One route to a genuine ranking evaluates these ends. If educating about the past is better than making people sad, then historical dramas are better than tearjerkers. However, both these ends are valuable in some circumstances and not others. Another route to a genuine ranking judges movie genres by a common end. It judges each genre by the ability of sample movies to deliver an aesthetic experience. However, the danger is that our judgments simply echo our genre preferences. An action buff will claim that sample action movies get higher scores on the aesthetic standard than sample historical dramas, and a member of the highbrow crowd will give the nod to the historical dramas over action movies on the same standard. The only way to adjudicate incompatible rankings is to come up with a substantive theory of the aesthetic which all rankers should accept and which falsifies some of the incompatible rankings. Notoriously, no theory of the aesthetic meets both these conditions, and it is natural to suspect that theories of the aesthetic that do imply genre rankings smuggle in the theorist’s own genre preferences (e.g., Horkheimer and Adorno 1972). Given this impasse, the best we can do is clarify debates about a hierarchy among movie genres. 159
TF17557.indb 159
16/9/08 15:47:25
BR I A N L A ET Z A ND D OM I NIC MCI V E R LOPE S
The final question for a theory of value and genre concerns what it is to evaluate a movie as belonging to a genre. Unlike the 2004 remake of The Manchurian Candidate, The Great Dictator is a superb instance of satire. Viewed as film noir, however, Chaplin’s masterpiece would come off rather poorly. So ideas about genre play into movie evaluations. The question is how genres play into these evaluations. One possibility is that evaluating a movie as a satire requires a belief that it is a satire. However, this cannot be the whole story. I may come to believe that a movie is a jidaigeki by reading the DVD case, but this belief alone cannot make a difference to my evaluation of it. To evaluate the movie as a jidaigeki, I must surely have some beliefs about the genre. The precise nature of these beliefs is an open question –one proposal is that I should believe that it is a samurai movie, governed by certain conventions (e.g., the villains have mussed-up hair). In sum, where K is a genre, S evaluates a movie as a K only if (1) S has some beliefs about K, and (2) the content of the evaluation counterfactually depends on S’s beliefs about K. Applied to the jidaigeki example, this says that I evaluate a movie as a jidaigeki only if I have some beliefs about the genre. In addition, these beliefs must make a difference to my evaluation –if I had different beliefs, I would evaluate the movie differently. Once again, though, the above principle is only a start. Here is something stronger: S evaluates a movie as a K only if (1) S has some true beliefs about K, and (2) the content of the evaluation counterfactually depends on S’s true beliefs about K. Considering the case for this stronger principle and for alternatives to it promises to shed light on the role of genre in evaluation (Lopes 2007). That nobody has worked out a comprehensive philosophy of movie genre is not a sufficient reason to attempt one now. Some assurance is needed that genuine progress toward a philosophy of movie genre is possible. This chapter has sought to highlight the importance of genres in mediating our engagement with movies – something implicitly acknowledged in recent studies of individual movie genres. It illustrates how resources developed in other areas of the philosophy of art can be harnessed to make sense of the work done by genre in our engagement with movies. And it connects some proposals for philosophical accounts of movie genres to work in film studies. More should and can be said about the theory of movie genre, its ontology, and questions of value. See also Dogme 95 (Chapter 44), Documentary (Chapter 45), Horror (Chapter 46), Pornography (Chapter 47), Avant-garde film (Chapter 48), Tragedy and comedy (Chapter 49), Noël Carroll (Chapter 31), and Authorship (Chapter 2).
160
TF17557.indb 160
16/9/08 15:47:25
GE N R E
References Altman, R. (2003) “Genre,” in B. K. Grant (ed.) Film Genre Reader III, Austin: University of Texas Press. Boden, M. (2004) The Creative Mind: Myths and Mechanisms, London and New York: Routledge. Braudy, L. (1999) “Genre: The Conventions of Connection,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, New York: Oxford University Press. Buscombe, E. (2003) “The Idea of Genre in the American Cinema,” in B. K. Grant (ed.) Film Genre Reader III, Austin: University of Texas Press. Carroll, N. (1990) The Philosophy of Horror, or Paradoxes of the Heart, New York: Routledge. Cavell, S. (1996) Contesting Tears: The Hollywood Melodrama of the Unknown Woman, Chicago: University of Chicago Press. Cawelti, J. (1976) Adventure, Mystery, and Romance: Formula Stories as Art and Popular Culture, Chicago: University of Chicago Press. Currie, G. (2004) “Genre,” in Arts and Minds, Oxford: Clarendon Press; and New York: Oxford University Press. Duff, D. (ed.) (2000) Modern Genre Theory, Harlow, UK; New York: Longman. Gaut, B. (1997) “Film Authorship and Collaboration,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Clarendon Press; and New York: Oxford University Press. —— (2000) “ ‘Art’ as a Cluster Concept,” in N. Carroll (ed.) Theories of Art Today, Madison: University of Wisconsin Press. Goodman, N. (1976) Languages of Art, 2nd ed., Indianapolis: Hackett. Grant, B. K. (ed.) (2003) Film Genre Reader III, Austin: University of Texas Press. Horkheimer, M., and Adorno, T. (1972) “The Culture Industry,” in J. Cumming (trans.) Dialectic of Enlightenment, New York: Herder and Herder. Kitses, J. (1969) Horizons West, Bloomington: Indiana University Press. Lewis, D. (1978) “Truth in Fiction,” American Philosophical Quarterly 15: 37–46. Lopes, D. M. (2007) “True Appreciation,” in S. Walden (ed.) Photography and Philosophy: New Essays on the Pencil of Nature, Malden, MA: Blackwell. McArthur, C. (1972) Underworld U.S.A., New York: Viking. Neale, S. (2003) “Questions of Genre,” in B. K. Grant (ed.) Film Genre Reader III, Austin: University of Texas Press. Rohrbaugh, G. (2005) “Ontology of Art,” in B. Gaut and D. M. Lopes (eds.) Routledge Companion to Aesthetics, 2nd ed., London and New York: Routledge. Sobchack, T. (2003) “Genre Film: A Classical Experience,” in B. K. Grant (ed.) Film Genre Reader III, Austin: University of Texas Press. Staiger, J. (2003) “Hybrid or Inbred: The Purity Hypothesis and Hollywood Genre History,” in B. K. Grant (ed.) Film Genre Reader III, Austin: University of Texas Press. Thomasson, A. (2005) “The Ontology of Art and Knowledge in Aesthetics,” Journal of Aesthetics and Art Criticism 63: 221–9. Tudor, A. (2003) “Genre,” in B. K. Grant (ed.) Film Genre Reader III, Austin: University of Texas Press. Walton, K. (1970) “Categories of Art,” Philosophical Review 79: 334–67. —— (1990) Mimesis as Make-Believe, Cambridge, MA: Harvard University Press. Warshow, R. (1962) The Immediate Experience: Movies, Comics, Theatre, and Other Aspects of Popular Culture, Garden City, NY: Doubleday. Wollheim, R. (1980) Art and Its Objects, 2nd ed., Cambridge and New York: Cambridge University Press.
161
TF17557.indb 161
16/9/08 15:47:25
15
INT E RP R ETAT ION George Wilson In her famous 1966 essay, “Against Interpretation,” Susan Sontag asserts, “What the overemphasis on content entails is the perennial, never consummated project of interpretation. And conversely, it is the habit of approaching works of art in order to interpret them that sustains the fancy that there really is such a thing as the content of a work of art” (Sontag 1966 [1964]: 5). However, Sontag immediately grants that it is not just any concept of “interpretation” that she views as pernicious. The conception to which she objects is one in which the interpreter selects a set of elements from the targeted work and constructs an interpretation by claiming of each element that it really means such and such or stands for so and so. As she says herself, interpretation, so conceived, is virtually a task of translation – translating a part or aspect of the work into some proprietary content it purportedly expresses. However, Sontag’s brief against interpretation is confusing. Most of the leading practitioners of close interpretation in the early sixties, the New Critics for instance, had themselves rejected the idea that interpretation consisted in the systematic translation or allegorization of the work to be interpreted. Interpretation grounded on the “translation idea” commits “the heresy of paraphrase,” in Cleanth Brooks’ well-known phrase (Brooks 1947). From the point of view of these critics, Sontag is really not objecting to the proper enterprise of interpreting works of art, but to an insidious misconception of what practical criticism amounts to. For these critics, the “project of interpretation” is a tenable one, and it is important. What need to be repudiated are simply confusions about the proper aims and methodology of the project. In the fifty years since Sontag’s essay was published, the same confusing dialectic has recurred, and it recurs significantly in the study of film. On the one hand, film viewers, professional and otherwise, persist in prolifically producing interpretations of a great variety of films. On the other hand, Sontag’s broad skepticism about “the project of interpretation” may well be the dominant view within the discipline (King 1998). Interpretation is commonly regarded as having at best marginal importance within film studies. And yet, as we’ll see, various attempts to formulate the grounds for such skepticism often rest on tendentious notions of what is involved or presupposed in the actual interpretation of movies. For instance, generalizations and extensions of “the translation idea” still abound. On the other hand, when a less loaded conception is adopted, it is hard to see how interpretation could be avoided and why it cannot
TF17557.indb 162
16/9/08 15:47:25
I N T E R PR E TAT ION
play a significant role in our overall understanding of movies. Without a better sorting out of a range of critical issues, there is the danger that the continuing debate will remain inconclusive, with each side arguing at cross-purposes with the other. Sontag speaks as if there was some one identifiable enterprise that can legitimately be called “the project of interpretation,” but it is unlikely that the messy mélange of interpretative works share any substantial common agenda, however broadly defined. Hence, questions about the nature and aims of interpretation are probably best relativized to one or another more specific interpretative project (Gaut 1993). Similarly, it is widely assumed that interpretation always purports to specify a meaning or a constellation of meanings for a work of art. First, however, interpretative work on film is frequently focused on understanding, e.g., aspects of style, tone, and point of view, and the detailed analysis of these matters need not issue in something that one would naturally refer to as “meaning” (Pye 2007). Second, the word “meaning” is used in English to cover such a range of diverse phenomena that it may not help much to insist on associating interpretation too closely with the retrieval or construction of some variety of “meaning.” For example, we speak of the linguistic meaning of a word or sentence, the meaning a speaker intended to convey by an utterance, the symptomatic meaning of red spots on the skin, and the meaning of an important life-episode like the breakup of a friendship. Some or all of these concepts of meaning may have some bearing on some particular interpretative projects, but the concepts are quite different in import, and it assigns no higher unity to distinct interpretative projects to subsume them all under one or another of these senses of “meaning.” In fact, the temptation to link interpretation with the exegesis of meaning probably has done a lot to reinforce the unfortunate conception of interpretation as some sort of decoding of the targeted text into an antecedently favored configuration of concepts and themes. If one wishes to insist that the interpretation of an element E in a work W always involves a claim that E, in the context of W, means so and so, then the thesis tells us very little about the character of the interpretation in question in the absence of an explication of the notion of “meaning” that one intends to invoke. In his book, Making Meaning, David Bordwell also expresses a distinctly skeptical attitude toward interpretation, but at least he makes an attempt to draw some of the needed discriminations between different purported modes of understanding movies. He draws a distinction between what he calls, somewhat stipulatively, issues of “comprehension,” “explicative interpretation,” and “symptomatic interpretation.” It is notable that he pairs each of these categories with a corresponding sort of “meaning.” Comprehension is concerned with “referential” and “explicit” meanings, where these are the overt facts about story or theme that are directly presented as such within the film. Bordwell regards referential and explicit meanings, taken together, as the analogue of literal meaning in language. Explicative interpretation has “implicit meanings” as its object, and symptomatic interpretation aims at uncovering “repressed (symptomatic) meaning.” Very roughly, symptomatic interpretation attempts to explain a cinematic phenomenon as the manifestation (the symptom) in film viewers or in the film itself of some way in which social/cultural forces or unconscious psychological forces in the 163
TF17557.indb 163
16/9/08 15:47:26
GE ORGE W I L S ON
individual subject impinge significantly upon the construction of the movie or upon standard ways of watching and responding to it. This category of interpretation is meant to cover a great range of the interpretative strategies that were favored especially in the period of Grand Theory during the seventies and eighties. For example, in her widely read and highly influential essay, “Visual Pleasure and Narrative Cinema,” Laura Mulvey (1990) argues that certain fundamental features of narrative and its presentation in classic Hollywood films are muted manifestations of unconscious and pathological male fascination and anxiety engendered in contemplating the erotically charged female body. However, in the present entry, I will focus on issues concerning explicative interpretation, beginning with Bordwell’s conception of the enterprise and the sort of implicit meaning he thinks that it purports to specify. According to Bordwell, implicit meanings satisfy two conditions. 1 An implicit meaning is one that is not overtly expressed or represented in the film. It is a content that the film merely implies, suggests, or otherwise communicates indirectly. 2 It is a general, abstract, or thematic content that is not included in referential meaning, although an implicit meaning is standardly conveyed, in part, by means of the exposition of narrative materials. Thus, Bordwell suggests hypothetically that the proposition “Madness is indistinguishable from sanity” could be a part of the implicit meaning of Hitchcock’s Psycho (Bordwell 1989: 9). The second condition, however vague, is important because, as Bordwell allows, referential meanings (aspects of the story) can very well be imparted in the movie only in an oblique manner as well. But the conception of explicative interpretation registered in 1) and 2) together is in danger of making explication the attempt to retrieve some general message or moral that is covert but somehow contained in the film. If this danger is not avoided, then Bordwell, like Sontag, describes explicative interpretation in a manner that is arguably untrue to the upshot of the best practical criticism of film and at odds with what the pertinent practical critics have taken themselves to accomplish. On the whole, Making Meaning tends to be dubious about, and even dismissive of, the products of Interpretation Inc., but Bordwell frames the basic issues about interpretation in such a way that some of his arguments are in danger of being aimed largely at a straw man. This discussion of Bordwell points to two broad and basic questions about the nature and upshot of explicative interpretation. First, how is one to understand what it is for a film itself to mean, imply, express, or otherwise convey, whether directly or indirectly, certain content? Call this the question of the grounding of meaning or implication relations. Philosophers writing on film have discussed this question quite extensively. However, in doing so, they have commonly presupposed that (explicative) interpretation is to be construed as a kind of exegesis of the cinematic text. The exegetical model of film interpretation, as I will understand it, is the conjunction of 1) and 2) together with a realist or anticonstructivist view of meaning or implication relations in film. That is, the model assumes that 164
TF17557.indb 164
16/9/08 15:47:26
I N T E R PR E TAT ION
3 There are objective facts about what films or parts of a film mean, express, or imply, where those facts are determinately established in the making of the film and are not constructions projected by the viewer or the critic on the work. This conception yields the second question. Is this the proper model, in general, of the practice and goals of explication? The idea is certainly not inevitable. Thus, consider thesis 2). When one attempts to make sense of some incident in a person’s life, the outcome of the attempt will be a suitable pattern of explanation – an illuminating story – about the incident, concerning both its antecedents and consequences. (For a good overview of the connections between making sense of an event and constructing a narrative about it, see Williams [2002].) It is not as if there were some propositional content that the incident is supposed to have expressed, implied, or somehow come to signify. Nothing analogous to 2) holds in this case. It is a plausible conjecture that certain kinds of interpretation of films aim at patterns of explanation of this or a related variety. At any rate, we will also examine basic challenges to the exegetical model of explication. Let us begin with the question of meaning relations. It is natural to believe that a film or a part or aspect of a film (X) can mean or imply a content only in a derivative sense. That is, it is natural to insist that X means or implies the content C only in virtue of the fact that the relevant filmmakers constructed X with the intention of expressing or invoking C for an appropriate audience. It is the actual intentions of the filmmakers in question that crucially determine whatever meaning or content that X may have. This is a form of actual intentionalism about meaning relations in connection with movies, a position that is familiar from debates about meaning relations in works of literature (Livingston 2005). The position models the meanings conveyed in literary works upon the meanings that speakers try to convey or communicate by the utterances they produce. There are various familiar objections to actual intentionalism as a general account of artistic meaning. First, it has commonly been argued that the critic is often in no position to know what the artist actually intended concerning interpretatively vital matters, and, partly for that reason, recovering intentions is not what the critic seems to be after. Second, it is possible for an agent to intend to communicate or express something, but fail utterly to do so. Whatever one thinks about these common objections to actual intentionalism about meaning in literary criticism, the doctrine faces an immediate and obvious challenge in the case of film. Most movies are the product of collaborative work, and the fact of elaborate collaboration raises the question of whose intentions are supposed to be determinative of the film’s content. The intentionalist may pick a single individual (the director, for example) as the putative source of the meaning-making intentions for the film, but any such choice seems wildly implausible in the general case. No auteur theorist ever went so far as to claim that the meaning or significance of any film was determined in detail by the intentions of its director. Of course, there are also shared or joint intentions. Two or more people may have the intention of robbing a bank together, and, more pertinently, they can coauthor a book or article. However, it is unlikely that the intentionalist about film can plausibly frame the account in terms 165
TF17557.indb 165
16/9/08 15:47:26
GE ORGE W I L S ON
of intentions about content that are shared among the various collaborators. There are simply too many cases in which an apparently coherent film was made despite the fact that the director, the screenwriters, and the actors had very different conceptions about what they were attempting to convey. (Actors sometimes hated working with Fritz Lang because he could be so demanding in his direction without being at all forthcoming about what it was that he was seeking to achieve.) Perhaps there is a way for the actual intentionalist to find a well-motivated multilayered version of the theory that allows properly for the complexities engendered by cinematic collaboration, but the problems faced by such an account are challenging (Gaut 1997). The actual intentionalist likens the “meaning” of a movie shot, scene, or sequence to what speakers mean in producing their utterances, verbal or otherwise. However, when one considers the typical character of the sort of collaboration involved in fiction film, the implausibility of this comparison is especially striking. A verbal utterance is characteristically the product of a single agent, and it is the intentions of that agent that ground the particular content that the utterance conveys. A movie scene, by contrast, is normally the product of various agents operating at various levels of production, and all of these agents may well have had intentions, sometimes conflicting, about what the completed scene is supposed to achieve. Consider, for example, just the speech of a character in a fiction film. Although it will be the actor who literally performs the utterance, the intentions of various collaborators will have shaped and guided its content-relevant nature. For instance, the director, with his or her relevant intentions, may have done a lot to shape an actor’s delivery of the lines, even in instances when the actor has not grasped the aims behind the director’s instructions. And naturally, the actor’s performance is significantly defined and structured from the outset by the words and instructions that the screenwriters have supplied. Where in this complicated intermeshing of artistic intentions in relation to the on-screen behavior of the character is the critical flash point that yields anything like an analogue of the relations that individual speakers bear to their own utterances? The situation becomes patently more complicated when all the diverse contributions that played a role in the construction of the completed scene are considered. In fact, philosophers have paid too little attention to the specific question of the ways in which screen performances concretely embody a detailed and nuanced dramatic significance (Livingston 1996; Klevan 2005). Correlatively, they have usually said little about the specific ways in which staging, photography, editing, and other procedures of postproduction contribute individually to the significance that is finally presented to the audience on screen. Treating a shot or sequence as an analogue of a speaker’s utterance tends to discourage analysis of the distinct but interanimating dimensions involved in the construction of cinematic significance. A number of philosophers writing on film have opted for some version of hypothetical intentionalism (Currie 1995; Levinson 1996). Here is one way of characterizing the view. In watching a fiction film, viewers know that they are seeing an audio-visual construction that is the product of a (collaborative) intelligence, personality, and sensibility. Knowing this to be so, viewers are disposed to imagine the characteristics of intelligence, personality, and sensibility that they find manifest in the detailed 166
TF17557.indb 166
16/9/08 15:47:26
I N T E R PR E TAT ION
crafting of the film. They imagine a kind of “implied filmmaker” – an implied version of the filmmaker(s) expressed in the movie’s detailed articulation. According to the hypothetical intentionalist, a shot or sequence in a movie expresses a content C just in case the film as a whole makes it maximally plausible to imagine that the implied filmmaker(s) intended that the designated shot or sequence was to convey or communicate C to a suitably attentive and responsive audience. What ground the ascription of meaning to a film segment, on this approach, are not facts about the intentions that some artist or group of artists actually had. They are putative facts about what the film’s internal relationships prescribe concerning the content-relevant intentions justifiably imagined as part of a viewer’s engagement with the film. As this formulation indicates, hypothetical intentionalism is a complex view, and its viability depends on how certain relations to which it appeals are ultimately to be spelled out. For instance, does the hypothetical intentionalist about movies mean to appeal to the imagined intentions of an implied filmmaker (in the singular), and if so, is it an implied “director” to whom his theory refers? Or, perhaps, it would be better to appeal to the implied filmmakers (plural), imagining that they have effectively collaborated in the making of the film. However, this maneuver is likely to face objections in cases where there is plain evidence to the contrary. In addition, a great deal turns on the conception put forward about what justifies a viewer in imagining that a part or aspect of a movie was intended to be construed in such and such a way. For instance, if one knows as a matter of historical fact that no pertinent film artist had the intentions that the critic has been prompted to imagine, does that preclude the critic being justified in imagining otherwise? Or again, what is supposed to constitute a suitably attentive and responsive audience? In answering these questions, there is a danger of circularity. It may be that an intention about meaning that one is justified in ascribing imaginatively is grounded upon the meaning that one supposes is manifested in the work, where “manifest meaning” is a factual matter, independent of anyone’s actual or hypothetical intentions. This is the position of the anti-intentionalist about meaning relations. Many writers of a structuralist or semiotic persuasion seem to have supposed that meaning relations in a “semiotic system” are grounded in their quasi-conventional roles in social practices of signification and do not rely on individual communicative intentions. If this is their view, then these writers are anti-intentionalist in the sense delineated above. In any case, the hypothetical intentionalist and the anti-intentionalist differ about whether the imagined intentions of an implied filmmaker can serve as the source of the meaning of the work. Bordwell holds that (implicit) meaning relations in the cinema are not objectively embodied in the work and are not discovered there by interpreters. Rather, they are something that interpreters construct by assigning contents of favored types on the basis of a rich assortment of audio-visual cues that the film supplies. In other words, Bordwell seems to reject thesis 3) of the exegetical model of interpretation. More specifically, for him, an explicative interpretation is projected onto the film by mapping a range of the salient cues it offers onto the elements of what he calls a “semantic field” – roughly, a conceptual framework that the critic has chosen as particularly suitable for an illuminating exercise in cinematic analysis. It is not that Bordwell contends that an 167
TF17557.indb 167
16/9/08 15:47:26
GE ORGE W I L S ON
interpreter’s choice of cues, semantic fields, and the interpretative mappings between them is unconstrained. He maintains that the constraints that bear legitimately on a prospective interpretation are the interpretative practices and conventions endorsed by the critical community to which the interpreter belongs. To repeat, Bordwell’s account of the critic’s methodology repudiates the exegetical model of explicative interpretation, but it is striking that he accepts, with Sontag, a fairly strong version of the “translation” conception of interpretation. It is difficult on such an approach to avoid a counterintuitive relativism about acceptable interpretation. The problematic character of the translation idea is highlighted if we consider an alternative to the exegetical model of interpretation. What is the sort of explanation that explicative interpretation aims primarily to produce? Given the vast range of explicative interpretations on the market, it is preposterous to suppose that there will be a single substantial answer to this question. Nevertheless, if we restrict the question to the understanding of narrative fiction film, we can narrow its scope to some degree. In Film as Film (1972), Victor Perkins asserts, “Our understanding and judgment of a movie . . . will depend largely on the attempt to comprehend the nature and assess the quality of its created relationships” (118). Among the relationships to which Perkins is here referring are the dramatized narrative relationships between the characters. In trying to understand a movie story, we normally expect to achieve some overall understanding of the chief actions and reactions of its fictional agents. Working within the frameworks of genre, narrative exposition, and audio-visual articulation that inform the movie, viewers want to know why the characters have acted, thought, and felt in the ways depicted in the film. Similarly viewers also want to grasp a range of the consequences of the characters’ actions, thoughts, and feelings. In other words, viewers seek to limn the relations of explanatory coherence and value assessment that the movie progressively reveals. So, one upshot of an interpretation can be a perspicuous representation of these purported patterns of explanation and evaluation in the narrative (Wilson 1997; Carroll 2002). Moreover, there is no reason to suppose that such a configuration must carry with it some general theme or thesis encapsulating, so to speak, a principle or lesson that the story covertly implies. In fact, the relevant patterns of causal and teleological explanations involve something more like an explicit elaboration of the details of the plot and the implied connections between them. The explication will be an especially cogent and specific reconstruction of the implicit fictional history presented in the film. On this line of thought, the challenge to the exegetical model turns on a rejection of thesis 2) above, and it contrasts sharply with the constructivist’s rejection of 3). To render the proposal here relatively concrete, it is worth working through an actual example, at least in outline. Toward the end of Shadow of a Doubt (Alfred Hitchcock, 1946), young Charlie (Teresa Wright) discovers that her uncle – also called Charlie (Joseph Cotton) – has in the past married and murdered a number of rich widows for their money. Although young Charlie has been thoroughly enthralled by the romantic mystery that she associates with her uncle, she comes eventually to learn that he is a profoundly evil man. She plans to let him know that she can prove that he committed the murders, because she possesses a ring that he has given her – a 168
TF17557.indb 168
16/9/08 15:47:26
I N T E R PR E TAT ION
ring, she now realizes, that belonged to one of his victims. To convey this warning to him, she puts the ring on and displays it conspicuously when she enters a gathering of her relatives and friends at her home. Uncle Charlie sees the ring, grasps the message his niece is sending, and promptly announces that he has made a decision to leave town the next day. Young Charlie’s mother is utterly distressed at his announcement, and she speaks movingly about her love for her brother and the value she places on her family. In the face of her mother’s grief, young Charlie covers the ring and does not press the confrontation further. Since she has apparently effected her uncle’s imminent expulsion from the town and because she is plainly dismayed by her mother’s unhappiness, it may seem that her actions in this scene are perfectly in order. But, there is a complication. The little gathering includes an attractive widow, Mrs Potter, and it is clear to everyone – to the people at the party, including young Charlie, and to the viewer – that if her uncle carries out the intentions he has just announced, then he will be headed out of Santa Rosa with still another widow who is thoroughly smitten with him. Therefore, young Charlie has every reason to know that by keeping silent about Uncle Charlie she is, at a minimum, placing Mrs Potter in mortal danger. So why does young Charlie suppress the damning information she possesses? Does she suppose that her mother’s grief outweighs the dire consequences that the infatuated widow might face? Surely, that would be a questionable presumption! Earlier in the film, there have been strong indications that young Charlie and her uncle share some powerful but mysterious affinity. The nature of their affinity is never fully explained, although we learn early on that young Charlie is restless – bored and dissatisfied with the conventional outlook of small-town bourgeois life. Naturally, she comes to detest the utter nihilism about middle-class values that her uncle endorses, but is it possible that, even as the film moves toward its conclusion, she remains complicit with some facets of his distorted and disturbing outlook? An answer to these questions would yield an interpretation of Charlie’s action in the designated scene and an interpretation of other significant aspects of her actions elsewhere in the story. Such an interpretation would take the form of an imagined explanation of her fictional behavior – an explanation conjoined perhaps with some evaluation of the motives that are thereby deemed to have prompted her key decisions. One can rightly say that such an interpretation assigns a meaning or significance to Charlie’s behavior, but this is the sort of meaning we are after whenever we attempt to make sense of an action or incident in a person’s life by setting it in a plausible explanatory/evaluative framework. Call this “narrative meaning.” The framework we aim at is an enlightening narrative about the action or incident in question. Hence, the narrative meaning of Charlie’s behavior here is not something in addition to and more abstract and general than an account of the concluding scenes of Shadow of a Doubt. It is the product of the interpreter’s attempt to give a more fine-grained specification of what he or she takes that story to involve when Charlie conceals the incriminating ring. The interpreter’s version of the elaborated story may well suggest to some minds a comprehensive moral or abstract thesis that this version appears to them to illustrate. (“Everyone is corrupted by the temptation to disvalue the lives that one views as parasitical and useless.”) Nevertheless, any such a generalizing 169
TF17557.indb 169
16/9/08 15:47:26
GE ORGE W I L S ON
gloss is a dubious add-on to the interpretation of narrative meaning, strictly construed. The gloss is not constitutive of the interpretation itself. These reflections demonstrate why it is implausible to separate, as Bordwell tries to do, the interpretation (“comprehension”) of diegetic and narrative questions from the explicative interpretation that aims at what he calls “implicit meaning.” In trying to understand young Charlie’s somewhat surprising silence at the end of Shadow, one is trying to grasp the significance of her action, a meaning that is merely implicit in the context of this particular fictional narrative. But this is almost certainly not the sort of “implicit meaning” that Bordwell has in mind in his taxonomy of interpretation. And yet it is an important kind of meaning and is often the object of the puzzlement when viewers find themselves puzzled by the “meaning” of a scene. Narrative meaning, as I have explained the notion, fails to satisfy the precepts of strong versions of actual intentionalism. That is, a fiction film or some segment thereof can have a narrative meaning that neither the director, nor the screenwriters, nor the actors intended to convey. Let us imagine (as I believe) that it is possible to argue convincingly and in detail that young Charlie, in failing to reveal that her uncle is the Merry Widow Murderer, is still somewhat tainted by her earlier complicity with aspects of his moral perspective. Thus, we stipulate, for the sake of argument, that her actions in the late scene make a compelling sense when they are explained along those lines. Suppose in addition, however, that it is also discovered that neither Hitchcock, the screenwriters, nor the principal actors intended that the end of the movie be understood in this way. They did intend, we might suppose, that the audience should notice the irony of the fact that, if Uncle Charlie gets away from Santa Rosa, he will leave in the company of a prospective victim. However, they did not intend, we’ll imagine, that the audience was to focus upon the question of whether young Charlie has apprehended this potential consequence and, if she did, what considerations the knowledge might have provoked in her. Of course, this hypothetical discovery about the filmmaker’s conception of the scene shows that the interpreter’s postulated account does not capture a narrative meaning that the filmmakers intended to communicate, but it does not show that the interpreter has been wrong in holding that the episode naturally manifests, in its setting, the narrative meaning that he or she has proposed. This observation highlights the need to be careful about the nature of the interpretative project that a given interpretation is meant to realize. The project of figuring out what narrative significance that, e.g., the director might have intended to express is a perfectly coherent one, although it is notoriously tricky to carry out the task successfully. Still, the narrative meaning that is manifest in a scene or segment need not coincide with whatever narrative meaning the relevant filmmakers intended to impart. Only confusion will result from running the two together. Moreover, it can be misleading to speak of the narrative meaning of a movie or a part thereof, since there will often be competing ways of constructing the explanations that ground narrative meaning, that tie in terms of overall plausibility and resonance. The example of narrative meaning in film is necessarily a limited one. There are a host of movies (e.g., Blowup [Michelangelo Antonioni, 1961], The Enigma of Casper 170
TF17557.indb 170
16/9/08 15:47:26
I N T E R PR E TAT ION
Hauser [Werner Herzog, 1978], Mulholland Drive [David Lynch, 1999]) in which it is not expected that the audience make detailed sense of the narrative action in conventional psychological terms. This does not mean, of course, that the movies are not apt targets for systematic interpretation. It does mean that at least some of the interpretative questions one poses will have to shift their ground. Critics may ask about the “meaning” of the stylized acting and mise en scène and query the “significance” of the oddities and gaps in the cinematic exposition of events. The proper understanding of “meaning” and “significance” in these contexts will require elucidation on their own. Nevertheless, there is no reason to think in these cases that the results of the interpretative investigations will take the form of some general moral or theoretical doctrine imparted by the movie. An enlightening delineation of systematic patterns of narrative and narration should suffice. Even in the case of narrative meaning, where the interpretation is directed at explaining incidents in the narrative, that explanation is guided and constrained by the host of features that define the manner in which the targeted incident has been portrayed. Hence, interpretative issues about style, tone, point of view, and so on will surely figure in an adequate explication of the narrative development (Pye and Gibbs 2005). In instances in which narrative meaning is only fitfully in focus, then issues of explanation about the forms and strategies of cinematic narration will assume greater prominence. There are many specific questions about the interpretation of various dimensions of narration (stylistic, epistemic, rhetorical) in fiction films that have been widely discussed, but there is not the space in this entry to pursue these topics. One hopes that in future research, there will be a more intimate collaboration between theoretical analysis and the most sophisticated and nuanced products of practical criticism. See also Authorship (Chapter 2), Narration (Chapter 18), and David Bordwell (Chapter 29).
References Bordwell, D. (1989) Making Meaning: Inference and Rhetoric in the Interpretation of Cinema, Cambridge, MA: Harvard University Press. Brooks, C. (1947) “The Heresy of Paraphrase,” in The Well-Wrought Urn and Other Essays, New York: Harcourt Brace and Jovanovich. Carroll, N. (1998) Interpreting the Moving Image, Cambridge: Cambridge University Press. Currie, G. (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge: Cambridge University Press. Gaut, B. (1993) “Interpreting the Arts: The Patchwork Theory,” Journal of Aesthetics and Art Criticism, 51: 597–609. —— (1997) “Film Authorship and Collaboration,” in D. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Oxford University Press. King, N. (1998) “Hermeneutics, Reception Aesthetics, and Film Interpretation,” in J. Hill and P. C. Gibson (eds.) The Oxford Guide to Film Studies, Oxford: Oxford University Press. Klevan, A. (2005) Film Performance: From Achievement to Appreciation, London: Wallflower Press. Levinson, J. (1996) “Intention and Interpretation in Literature,” in The Pleasures of Aesthetics: Philosophical Essays, Ithaca, NY: Cornell University Press.
171
TF17557.indb 171
16/9/08 15:47:26
GE ORGE W I L S ON
Livingston, P. (1996) “Characterization and Fictional Truth in the Cinema,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. —— (2005) Art and Intention: A Philosophical Study, Oxford: Oxford University Press. Mulvey, L. (1990) “Visual Pleasure and Narrative Cinema,” in Visual and Other Pleasures, Bloomington: Indiana University Press. Perkins, V. (1972) Film as Film, New York, Penguin. Pye, D. (2007) “Movies and Tone,” in D. Pye and J. Gibbs (eds.) Close-up 02, London: Wallflower Press. Pye, D., and Gibbs, J. (eds.) (2005) Style and Meaning: Studies in the Detailed Analysis of Film, Manchester, UK: Manchester University Press. Sontag, S. (1966 [1964]) “Against Interpretation,” in Against Interpretation and Other Essays, New York: Farrar, Straus, and Giroux, 3–14. Originally published in Evergreen Review, 34 (December 1964): 76–80, 93. Williams, B. (2002) “Making Sense,” in Truth and Truthfulness, Princeton, NJ: Princeton University Press, Wilson, G. (1997) “On Film Narrative and Narrative Meaning,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy Oxford: Oxford University Press.
Further reading Although not aiming at traditional modes of explication, the essays in Raymond Bellour, The Analysis of Film, ed. Constance Penley (Bloomington: Indiana University Press, 2000), include some striking and instructive frame-by-frame analyses of film segments. Edward Branigan, Narrative Comprehension and Film (London: Routledge, 1992), offers an unusually detailed study of the ways “narrative meaning” comes to be articulated in films. Stanley Cavell, The World Viewed: Reflections on the Ontology of Film, exp. ed. (Cambridge, MA: Harvard University Press, 1979), is a classic of philosophical writing about film – richly suggestive about the nature and importance of interpretation. Gilberto Perez, The Material Ghost: Films and Their Medium (Baltimore, MD: Johns Hopkins University Press, 1998) contains some of the best interpretative writing in recent years, combined with ample reflections on methodological and other theoretical issues. George M. Wilson, Narration in Light: Studies in Cinematic Point of View (Baltimore, MD: Johns Hopkins University Press, 1986) is an idiosyncratic book that joins close analyses of the narrative strategies of five classic Hollywood films with more theoretical considerations about point of view in film.
172
TF17557.indb 172
16/9/08 15:47:26
16
ME D IUM Kevin W. Sweeney Cinema – artistic film or movies – was the great new art form of the twentieth century. In its struggle to establish its artistic credentials, cinema inherited the nineteenthcentury debate about whether photography was an artistic medium. The question now was: could film, involving a process that mechanically recorded whatever was in front of the camera’s lens, be recognized as a medium of artistic expression? Was there a distinctive medium that allowed an artist or artists to create a cinematic work of art? As a developing narrative form, often adapting established theatrical works, film also prompted theorists to clarify the differences between theater and cinema. Exactly how was the film medium different from that of theater? This concern about the distinctive nature of the cinematic medium was not a new theoretical enterprise. Theorists of the arts had been engaging in similar debates about other artistic media for centuries. One of the most famous earlier medium theorists was Gotthold Lessing, who in his book Laocoön (1968 [1766]) proposed distinguishing poetry from painting in terms of their different sign systems. “Since painting,” Lessing claimed, “because its signs or means of imitation can be combined only in space, must relinquish all representations of time, therefore progressive actions, as such, cannot come within its range” (90). Painting, he argued, should only attempt to show bodies or objects in space and not try to show a temporally unfolding series of events. Because it presents its imitative signs sequentially in time, poetry should concentrate on representing actions or events and not the appearance of bodies in space. What is important for later medium theorists about Lessing’s theory is that Lessing connects a description of a medium with a prescription for how artists ought to use that medium. Because of its distinctive material and formal nature, each medium has its own special potential. Artists ought to produce works that recognize the formal restrictions and distinctive representational requirements of the medium; they ought not to employ stylistic strategies more appropriate to another medium. This two-part concern to identify a special functional character of an artistic medium and to prescribe certain projects appropriate to that medium has been referred to by Noël Carroll as “the specificity thesis” (Carroll 1988: 81). This concern to identify the nature of an artistic medium and to prescribe a range of aesthetic features appropriate to that medium was a major formalist interest of modernist artistic theory. Beginning in the middle of the nineteenth century and continuing
TF17557.indb 173
16/9/08 15:47:26
K E V I N W. SW EE N E Y
on into the middle of the twentieth century, modernist theorists questioned the imitative, illusionist aesthetic that had been the main direction of the arts since the Renaissance. In a classic statement of this position, Clement Greenberg in his essay, “Modernist Painting,” claimed that an essential feature of the modernist project was that each particular cultural practice, such as an individual art form, should engage in a self-critical search for its own distinctive nature. Each art form should attempt to discover what is “unique and irreducible” about its nature. He asserts: It quickly emerged that the unique and proper area of competence of each art coincided with all that was unique to the nature of its medium. The task of self-criticism became to eliminate from the effects of each art any and every effect that might conceivably be borrowed from or by the medium of any other art. Therefore each art would be rendered “pure,” and in its “purity” find the guarantee of its standards of quality as well as its independence. (1995: 120) According to Greenberg, the essence of painting – the irreducible medium of painting – was a paint-covered flat surface. Realistic or illusory painting, he says, “had dissembled the medium, using art to conceal art” (ibid.: 120). Modernist painting that conformed to this effort at self-criticism revealed the art form’s true or “pure” medium and proclaimed such works to be an honest expression of that medium, providing an experience for the viewer that no other art form could deliver. Modernism had a profound effect on all twentieth-century art forms, leading theorists of the arts to search for the irreducible essence of each artistic medium so as to validate it as a genuine art form. Artists were encouraged to produce works which reflexively highlighted the medium employed and revealed to perceivers the distinctive nature of that medium. Theorists of modernist poetry, modernist dance, modernist architecture, modernist (Brechtian) theater – and many other art forms – all sought to discover the essential medium of their specific art form. One can see the attempt to search for the “true” medium of film as an endeavor in this same modernist spirit. It was also important to distinguish cinema from theater and to emphasize that artistic films should be true to their medium and not pretend to be works in some other medium, such as filmed plays. An important issue in addressing the question of what is the true medium of film is the question of whether cinema is best thought of as a single medium. Or, because of its history of constant technological change, should film be better thought of as involving a process, where a newer medium constantly replaces an older medium? Or, is film better thought of as a composite medium – such as some might call opera – in which new features are added to established ones? To begin addressing these concerns, one might acknowledge that art forms often have several different, distinctive media. When speaking about painting, for example, one could say that water color is a different medium from oil paint. This way of speaking identifies a medium as a physical object or process with a distinctive set of properties or features. However, by emphasizing the properties of the manipulated physical object, this approach invites 174
TF17557.indb 174
16/9/08 15:47:26
M EDI U M
the question of whether changing the physical object (e.g., the paint) produces a new artistic medium. If because of advances in chemistry, contemporary oil-based paints are different than those used in seventeenth-century Holland, is that reason enough to say that contemporary oil painters and seventeenth-century Dutch genre painters were working in different media? Does each new advance in paint technology produce a new medium? While we might say that water color is a different medium from oil, is acrylic paint, which certainly has some unique properties, really a different medium from oil paint? Or, are they the same medium because they produce generally similar qualitative experiences? If we insist that acrylic is a different medium from oil, important questions are raised about the nature of the cinematic medium, due to the continuous change in film technology. Such changes have often been dramatic (e.g., the introduction of sound), but there have also been minor changes whose accumulative effects have been equally dramatic (e.g., gradual changes in the design of the camera). Are 1920s silent films that used black-and-white nitrate film stock in the same medium as sound Technicolor films from the 1950s or contemporary films using computer-generated imagery? Is the film medium that D. W. Griffith worked with during the 1910s and 1920s the same medium that Steven Spielberg worked with at the end of the century? Has cinema changed from a single distinctive medium into a composite medium or multimedia (Carroll 1996)? Some theorists have tried to find a common element across the myriad changes in film technology to identify as the essence of the medium. Gregory Currie finds the common element in cinema to be moving images (Currie 1995). Gerald Mast has proposed that cinema is essentially “an integrated succession of projected images and (recorded) sounds” (Mast 1977: 111). By emphasizing that cinema essentially involves projection, Mast distinguishes cinema from television, at least the cathode-ray-tube television technology that he was familiar with in 1977. The quality and resolution of the television and film images are quite different. The professional projected image in a theater has a much sharper resolution than that of a TV image in one’s living room. The film image created by light projected through celluloid allows for a true black and white. “On a color television set,” Mast points out, “black-and-white looks more like purple-gray-and-pale-pink, since the color dots still produce faint hues without any stimulus” (94). Of course, lots of films are shown on television; however, Mast claims that one’s experience of a film on a large movie theater screen is markedly different from that of watching even the same film on the little box in one’s living room. Seeing those giant lips say “Rosebud” on a large screen, Mast thinks, is a very different experience from seeing Citizen Kane (1941) on a small television monitor (ibid.: 102). With film we can be overwhelmed by the size of objects in the image, fascinated by the detail of a close-up, startled by loud sounds, held in rapture by soft ones. These characteristics of the screen image, Mast says, “produce almost tactile, sensual effects on the nerves, stomach, even skin . . . on television they remain occasionally interesting” (ibid.: 102–3). Cinema can create an almost “hypnotic” effect on the perceiver; television in one’s living room often competes with many distractions. 175
TF17557.indb 175
16/9/08 15:47:26
K E V I N W. SW EE N E Y
Like some other medium theorists, Mast approaches the problem of identifying the cinematic medium not just from the standpoint of the material features of the process. He closely identifies the nature of the film medium with providing a distinctive experience, one not afforded by other art forms. Even though film technology changed over the course of the twentieth century, Mast would claim that there is still a common store of experience that we would identify as cinematic experience. One might challenge Mast’s distinction between cinema and television by saying that future technological change might not preserve projection as the common, and hence essential, cinematic element. In distinguishing film viewing from watching television, Mast seems to discount the possibility of technological innovation changing TV. An improved high-definition image and larger monitors might narrow the difference between what, for Mast, are two different media. Mast anticipates such a challenge by replying to the following question: “Could television become as kinetically effective as cinema with sharp resolution, stereophonic sound, and an immense wall screen? Perhaps. (But then it would not be television as we know it today.)” (ibid.: 103). Other theorists of the cinematic medium have also tied the conception of the medium to the distinctive experience that film viewers have, a kind of experience not shared by those attending to works in other art forms. French film theorist, André Bazin, denied that cinema as a medium has evolved through a series of different media or that it has become more and more composite with each new technological innovation. In his essay, “The Myth of Total Cinema,” Bazin claimed that prior to any of the inventions of the early film pioneers, the cinematic medium existed as an ideal object in people’s imagination. He asserted, “The cinema is an idealistic phenomenon. The concept men had of it existed so to speak fully armed in their minds, as if in some platonic heaven” (1967a: 17). Inventors of film technology tried to find the means to realize physically this ideal; each invention presumably brought them closer to achieving that goal. “In their imagination,” Bazin claims, “they saw the cinema as a total and complete representation of reality; they saw in a trice the reconstruction of a perfect illusion of the outside world in sound, color, and relief” (ibid.: 20). Bazin thought that achieving this “perfect illusion” is the goal of cinema, the realization of the medium of cinema, and that only cinema has this experience as its goal. Yet, Bazin’s view on the cinematic medium yields a paradox, one that he was certainly aware of. He says, “Every new development added to the cinema must paradoxically take it nearer and nearer to its origins. In short, cinema has not yet been invented” (ibid.: 21). Since the cinematic work has not been produced that is the perfect embodiment of the ideal, paradoxically, true cinema, as a physical process, has not been invented. However, viewed in this way, the concept of a cinematic medium loses considerable value as a critical category useful for understanding and evaluating a director’s achievement. Such a conception becomes a measuring stick that we can use only to measure how close, in an asymptotic convergence, a film comes to achieving an ideal representation of the world. Of course, all films will be found wanting when measured by this perfect standard, although presumably some films will be closer to perfection than others. Bazin thinks that De Sica’s neorealist Bicycle Thieves (1948) 176
TF17557.indb 176
16/9/08 15:47:26
M EDI U M
approaches this ideal. Yet such a conception of the cinematic medium gives little room for appreciating the way in which directors in their films have created new visions of what cinema can be. Bazin conceives of the cinematic medium as being fixed and immutable, as are all ideal objects. There seems little room for aesthetic creativity in inventing, extending, or transforming our conception of film as a medium. Nevertheless, the idea that cinema has an essential connection with the world and that cinema has a unique ability to make us experience either the world (reality) or a fictional world has had a powerful influence on many cinema theorists. A debate developed between the “realists,” who claimed that film did have an essential connection with a world, and “formative” theorists, as Siegfried Kracauer has labeled them (Kracauer 1960), who denied this connection and held that film was a medium of the imagination. This debate could be said to have begun with the noting of the contrasting styles of two nineteenth-century pioneers of the cinema, the Lumière brothers with their films documenting events taking place in the world and the French magician, Georges Méliès, with his fanciful illusionistic films (Kracauer 1960). In the 1895 films such as Workers Leaving the Factory and The Arrival of the Train, the Lumière brothers showed actual events taking place in the world. Watching the films today, audiences – realists would say – see what the world actually looked like in 1895. As a magician, Méliès used stop-action and other photographic tricks to present not actual but imaginary events such as the live disembodied heads of Le Mélomane (1903). Cinema, formative theorists would say, has a distinctive technological capacity for this creative play. The realists would often argue for their position by claiming that the mechanics of the camera and the projector guaranteed that the series of projected images would be a record of the world and that no other artistic medium could connect us to the world in this way. In his classic realist article, “Style and Medium in the Motion Pictures,” Erwin Panofsky holds that the “medium of the movies is physical reality as such” (2004 [1934]: 302). The cinematic medium, he claims, is not an amorphous substance like the paint that a painter uses or the marble that a sculptor forms. Instead, the movies organize material things and persons, not as a neutral medium, into a composition that receives its style, and may even become fantastic or pretervoluntarily symbolic, not so much by an interpretation in the artist’s mind as by the actual manipulation of physical objects and recording machinery. (302) For Panofsky, film as an artistic project requires the filmmaker to shoot scenes in such a way that the camerawork creates the expressive or stylistic qualities of the film. To manipulate the scenery prior to filming, to construct the sets in a distorted way and paint in shadows, as occurs in The Cabinet of Dr Caligari (1920), is to “prestylize” the film. Not to use the camerawork in an expressive way, but instead to attempt to add stylistic features by manipulating the mise en scène prior to filming, violates Panofsky’s realist aesthetic. The use of the camera and film machinery to give style to the work means that cinema organizes “material things and persons” in a different way than does the theater. 177
TF17557.indb 177
16/9/08 15:47:26
K E V I N W. SW EE N E Y
Panofsky distinguishes cinema from theater by pointing out that the film medium has essential capabilities beyond the reach of theater, what he calls the dynamization of space and the spatialization of time. On the stage, space, Panofsky believes, is fixed and static – the theater audience is always the same distance from the stage space. Because of this static relation to the audience, stage space cannot become huge or very small or take on emotional characteristics. (Panofsky seems to discount the space-constructing powers of lighting on the stage.) In the cinema, the viewer’s relation to space is not fixed, and film has the capacity to create a variety of different spaces. Space on the screen, Panofsky believes, is movable; the viewer “is in permanent motion as his eye identifies itself with the lens of the camera which permanently shifts in distance and direction” (ibid.: 292). Time, he believes, can also work differently on stage and screen. On the stage, time can be separated from the space of the main dramatic action. The Shakespearean soliloquy presents an actor speaking lines that are understood by theatrical convention as not occurring in the time of the diegesis, but outside of the drama’s narrative world. The thoughts expressed are “out of time,” so to speak. Panofsky discusses Lawrence Olivier’s soliloquy in the filmed version of Henry V (1944). Instead of presenting Olivier speaking Shakespeare’s soliloquy as it would have been done on the stage, the film shows Olivier not speaking but sitting mutely by the campfire. The speech is given in voice-over. Panofsky suggests that this presentation of a soliloquy in a filmed play is a compromise and that a film, true to the cinematic medium, would connect narrative time with narrative space. What he has in mind is that a film which cuts between two different events can show events taking place at two different times, or can show different events occurring simultaneously. When shown an event on the screen, the viewer can always ask when in the narrative a particular event is taking place. Although Panofsky does discuss the voice-over soliloquy in Henry V, unfortunately he doesn’t draw out the ways in which a voice-over sequence can be extradiegetic – that voice-over narration can be understood as outside the time of the narrative. Another prominent realist theorist of the film medium is Siegfried Kracauer. Film, he says, “is uniquely equipped to record and reveal physical reality” (1960: 28). Yet, it is not the totality of physical reality that Kracauer thinks film records and reveals, only the visible world (Andrew 1976: 111). Nevertheless, because of the recording technology of film, with its variety of lenses, different camera shutter speeds, and numerous other techniques, film can show us events that we would either ordinarily not notice or normally be incapable of seeing. It can show us very large events (the shape and movement of a hurricane) and microscopically small events. It can slow down very fast events and speed up very slow ones (a plant growing). As long as a film adheres to this recording function, Kracauer believes that it is true to the medium: Imagine a film which, in keeping with the basic properties, records interesting aspects of physical reality but does so in a technically imperfect manner; perhaps the lighting is awkward or the editing uninspired. Nevertheless such a film is more specifically a film than one which utilizes brilliantly all the 178
TF17557.indb 178
16/9/08 15:47:27
M EDI U M
cinematic devices and tricks to produce a statement disregarding camerareality. (Ibid.: 30) On Kracauer’s view, the cinematic medium should not be thought of as just having a recording function; it should also reveal the world to viewers. Contemporary life is saturated with theoretical abstractions about the world, and our acquaintance with the world is diminished by these abstract perspectives. Cinema can give the world back to us, so to speak, by revealing to us features of, and events taking place in, concrete reality that might otherwise be inaccessible. Contemporary urban living has disconnected us from the natural world; cinema can show us natural phenomena that are distant or hard to see in all their otherwise obscure detail. Cinema, for Kracauer, is a panacea for the alienation of contemporary living. André Bazin also fits into the realist tradition because he believes that the photographic basis of film produces images that present the viewer with reality. Painters and sculptors can produce representations of the world that are surprisingly accurate; however, the image on the screen allows us to see the object photographed. Bazin claims, “The photographic image is the object itself . . . [I]t shares, by virtue of the very process of its becoming, the being of the model of which it is the reproduction; it is the model” (1967: 14). For Bazin, when watching a film, one does not just see a representational image of an object; one sees the actual model for the image. Using some of C. S. Peirce’s semiotic terms, one could say that the painting is an icon: it resembles the object it represents, and in so doing, Bazin says, “it may actually tell us more about the model” than a photographic image, if, for example, the photographic image is somehow degraded (ibid.: 14). A photographic image, even though it usually has high iconic value, is also an index of what it represents. It is the direct causal effect of a stimulus object, in the same way that a footprint is an index of the foot that made the impression, or an individual’s visual image is an index of an object (caused by light waves bouncing off that object and stimulating the retina). Just as we say that we see the world in virtue of having a visual field, so Bazin would say that film allows us to see the stimulus object whose image is recorded in the camera and projected on the screen. Yet Bazin ties the realist effect produced by the indexical film image to a distinctive experience that the viewer is capable of having with cinematic works. If cinema is identified as the vehicle to achieve the “perfect illusion” of the world, it should also provide viewers with experiences like those they have when they exercise their free choice to interpret and assign meaning to objects and events in the world. Bazin’s reliance on the viewer’s freedom derives from French existentialism, particularly Jean-Paul Sartre’s view that human beings must choose how they act and assign meaning to what they experience in the world (Andrew 1976: 172). The world is an ambiguous place, existentialists believe. Rather than being restricted to a course in life that is preestablished or determined, we must choose our actions, the course of our lives, and ultimately the world in which we live. Bazin felt that cinema should also provide viewers with an ambiguous world in which to choose and to assign meaning. His famous advocacy of a filmmaking style emphasizing deep space and the long take 179
TF17557.indb 179
16/9/08 15:47:27
K E V I N W. SW EE N E Y
was intended to support providing the cinematic occasion for such viewer agency and interpretive initiative. In “The Evolution of the Language of Cinema,” Bazin presents his defense of the deep-space and long-take style. He ties it to a vision of the active viewer facing an ambiguous cinematic world that calls for interpretation. The deep-space and long-take style is prominently associated with the films of Orson Welles and William Wyler, yet Bazin claims that it had been introduced by Jean Renoir and others. It proposes shooting in depth so that objects from the foreground to background of a long to an extreme long shot are in focus. Bazin advocates holding the shot long enough so that viewers have plenty of time to observe on their own terms the objects in the various focal planes. This, Bazin thought, made for a more active viewing experience. He claimed, [D]epth of focus brings the spectator into a relation with the image closer to that which he enjoys with reality . . . [I]t implies . . . both a more active mental attitude on the part of the spectator and a more positive contribution on his part to the action in progress . . . [The viewer] is called upon to exercise at least a minimum of personal choice . . . [D]epth of focus reintroduced ambiguity into the structure of the image . . . The uncertainty in which we find ourselves as to the spiritual key or the interpretation we should put on the film is built into the very design of the image. (1967b: 35–6) Bazin contrasts this style of filmmaking, which mirrors an individual’s active engagement with the world, with a montage style which he thinks promotes a more passive relation to a film on the part of a viewer. Montage, he thinks, predetermines the meaning of a sequence and eliminates viewers’ active involvement. Although very influential, Bazin and his fellow realists were opposed by formative theorists who proposed seeing the medium of film as having, not the capacity to reflect physical reality, but to construct a world. An early film theorist, Hugo Münsterberg, took a Kantian constructivist view of the cinematic medium. In perceiving the world, Kant held that we construct an interpersonal world with our common psychological faculties. Film, Münsterberg claimed, shares with our special psychological faculties a comparable means for world construction. Just as we construct an interpersonal world with a spatial-temporal framework and causal connections, Münsterberg saw film technology as supplying the means for constructing a narrative world with its own spatial, temporal, and causal relations. Film also has techniques comparable to other psychological faculties: memory is shown on the screen with the flashback; attention is indicated by an insert close-up (Münsterberg 1970 [1916]: 57–8). Another formative theorist, equally famous as a filmmaker, was the great Soviet director, Sergei Eisenstein. As a Marxist, Eisenstein believed that the phenomenal world does not present human beings with the truth about history or economic and social reality. If one wants to learn the truth, one has to appreciate the dialectical struggle and class conflicts that go on behind the appearances. To get at the truth, film cannot just record the world as it appears. He is no realist. Long takes of “physical 180
TF17557.indb 180
16/9/08 15:47:27
M EDI U M
reality” do not give the spectator the truth. Only by showing the clash of elements within the frame or the clash of sequences in the editing (montage) does one create the dialectical oppositions necessary for stimulating the viewer in the search for truth. For Eisenstein, film does not reflect or record reality; it creates new cinematic relationships that provoke the spectator into realizing the truth. The most prominent and systematic formative theorist was Rudolf Arnheim, who took pains to refute the realist view that film on the basis of its mechanical nature recorded physical reality. Arnheim points out that the world is three dimensional, but the film image is a projection of that world onto a two-dimensional plane. To suggest the missing depth, Arnheim thinks filmmakers have to make creative choices, as far as camera angles, the kinds of shots, and editing used to ensure that viewers see a sequence as having depth. In presenting a two-dimensional image, film does not automatically preserve the constancies of size and shape: a rectangular table would look trapezoidal, and objects closer to the lens would look much larger than those farther away (Arnheim 1966: 13–14). Again filmmakers have to exercise considerable inventiveness in choosing shots, camera angles, and kinds of editing to create a visual world on the screen that is understandable to the spectator. Arnheim identifies making such creative choices with artistic expression. Arnheim identifies a filmmaker’s making creative choices, so that a particular film sequence exhibits qualities not conveyed by the automatic process of filmmaking, with using the cinematic medium in an artistic way. He places special importance on making these choices within the limitations of current technology. For example, in the silent era, Joseph von Sternberg used a visual sequence to give the impression of a sound. In The Docks of New York (1928), he added a visual insert of the sudden flight of a flock of birds to show the sound of a pistol shot. Arnheim considers the creative choices involved in making that sequence to be an especially expressive use of the cinematic medium (Arnheim 1966: 34). Noël Carroll criticizes Arnheim’s “specificity thesis,” finding in Arnheim’s formative theory a narrow and overly prescriptive view about the medium of cinema (Carroll 1988). Filmmaking has changed over the course of a century, and Carroll holds that one should not insist that cinema conform to a narrow and traditional format excluding new developments as authentic and interesting instances of filmmaking. Carroll labels the view that there is a single essential medium for an art form as “medium-essentialism” (Carroll 1996: 49–55). He rejects “medium-essentialism” and advocates a broader view which would accept a variety of media as equally worthy of cinematic use. With the fading of the modernist urge to find each art form’s “true” medium and adhere to the prescriptions for producing works of art in that unique medium, there has been a greater acceptance of artists bridging artistic disciplines and working in composite or multimedia forms. This postmodernist attitude of accepting a diversity of media suggests that theorists of cinema have moved away from a classical modernist period in which questions about the nature of the film medium led to principles for the evaluation of films. Stanley Cavell’s film theory is representative of this theoretical shift. Instead of claiming that there is a single cinematic medium, he holds that there are “media” of film. He identifies these media with films made in different eras 181
TF17557.indb 181
16/9/08 15:47:27
K E V I N W. SW EE N E Y
and grouped together as genres (e.g., screwball comedies from the 1930s and ’40s). One of the advantages of this approach is that a medium becomes a set of common features determined by a natural historical grouping. There is no abstract essence with this approach: cinema has to be invented – and perhaps invented in different forms – rather than discovered by being matched to some ideal conception. Another advantage is its recognition of the role of technological change in transforming the processes of filmmaking and the invention of new acting styles and genres (Cavell 1979). However, essentialists like Gerald Mast would likely object that if one considers these to be different film media, one begs the question of why they are all film media. Isn’t there some common cinematic element in virtue of which they are all film media rather than media of different art forms? Finally, even if one adopts a multimedia perspective and recognizes different media corresponding to different film periods and genres, advocates of the classical realist and formative approaches (the heirs of the Lumière brothers and Méliès) are still alive and well, although they now pitch their arguments in narrower debates about the nature of individual genres and recent developments in filmmaking. For example, realist arguments abound about the nature of documentary filmmaking: Should documentaries be limited to a cinema-verité approach, which records actual events taking place, or should one allow the “re-creation” of these events in order to better inform the audience? Advocates of the formative approach can be found supporting new uses for computer-generated imagery in film. See also Definition of “cinema” (Chapter 5), Film as art (Chapter 11), Formalism (Chapter 12), Ontology (Chapter 20), Rudolph Arnheim (Chapter 27), Walter Benjamin (Chapter 28), Noël Carroll (Chapter 31), Stanley Cavell (Chapter 32), Sergei Eisenstein (Chapter 35), Jean Mitry (Chapter 37), and Hugo Münsterberg (Chapter 39).
References Andrew, D. (1976) The Major Film Theories, London and New York: Oxford University Press. Arnheim, R. (1966) Film as Art, Berkeley: University of California Press. Bazin, A. (1967a) “The Myth of Total Cinema,” in H. Gray (trans.) What Is Cinema?, vol. 1, Berkeley: University of California Press. —— (1967b) “The Evolution of the Language of Cinema,” in H. Gray (trans.) What Is Cinema?, vol. 1, Berkeley: University of California Press. Carroll, N. (1988) Philosophical Problems of Classical Film Theory, Princeton, NJ: Princeton University Press. —— (1996) Theorizing the Moving Image, Cambridge and New York: Cambridge University Press. Cavell, S. (1979) The World Viewed: Reflections on the Ontology of Film, exp. ed., Cambridge, MA: Harvard University Press. Currie, G. (1995) Image and Mind: Film, Philosophy and Cognitive Science, Cambridge and New York: Cambridge University Press. Greenberg, C. (1995 [1960]) “Modernist Painting,” in A. Neil and A. Ridley (eds.) The Philosophy of Art: Readings Ancient and Modern, New York: McGraw-Hill. Kracauer, S. (1960) Theory of Film, New York: Oxford University Press. Lessing, G. (1968 [1766]) Laocoön: An Essay upon the Limits of Painting and Poetry, New York: Noonday Press.
182
TF17557.indb 182
16/9/08 15:47:27
M EDI U M
Mast, G. (1977) Film/Cinema/Movie: A Theory of Experience, New York: Harper and Row. Münsterberg, H. (1970 [1916]) The Film: A Psychological Study, New York: Dover. Panofsky, E. (2004 [1934]) “Style and Medium in the Motion Pictures,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism, 6th ed., New York: Oxford University Press.
Further reading Dudley Andrew, The Major Film Theories (London and New York: Oxford University Press, 1976) is a survey of classical theories of film, with good analyses of realist and formative medium theories of film. Rudolph Arnheim, Film as Art (Berkeley: University of California Press, 1966) provides a major formative and expressionist theory of film as a medium. André Bazin, “The Myth of Total Cinema,” in H. Gray (trans.) What Is Cinema?, vol. 1 (Berkeley: University of California Press, 1967), presents Bazin’s theory that cinema is an idealist medium; and his “The Evolution of the Language of Cinema,” in the same volume, presents his realist theory of the film medium and his advocacy of the long-take and deep-space style of filmmaking. Noël Carroll, Philosophical Problems of Classical Film Theory (Princeton, NJ: Princeton University Press, 1988) presents extended analyses of the medium theories of Arnheim, Bazin, and Perkins; Carroll also discusses and rejects the “specificity thesis.” His Theorizing the Moving Image (Cambridge and New York: Cambridge University Press, 1996) argues against “medium-essentialism” and advocates an openness to new developments in filmmaking. Stanley Cavell, The World Viewed: Reflections on the Ontology of Film, exp. ed. (Cambridge, MA: Harvard University Press, 1979), rather than urging one cinematic medium, argues for various “media” of film. Greg Currie, Image and Mind: Film, Philosophy and Cognitive Science (Cambridge and New York: Cambridge University Press, 1995) offers a contemporary theory of the cinematic medium, requiring that film have moving images. A classical realist theory of the medium of film is Siegfried Kracauer, Theory of Film (New York: Oxford University Press, 1960). Gerald Mast, Film/Cinema/Movie: A Theory of Experience (New York: Harper and Row, 1977) emphasizes the projected image as an essential aspect of cinema as a medium, and contains a classical discussion of the difference between experiencing film and television. An early formative theory of film as a medium, emphasizing film as mirroring of human psychological faculties, is Hugo Münsterberg, The Film: A Psychological Study (New York: Dover, 1970), originally published in 1916. Erwin Panofsky, “Style and Medium in the Motion Pictures,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism, 6th ed. (New York: Oxford University Press, 2004), originally published in 1934, is a classical realist theory of film as a medium. V. Perkins, Film as Film: Understanding and Judging Movies (Harmondsworth and New York: Penguin, 1972) offers a theory of film as a medium, melding formative and realist elements.
183
TF17557.indb 183
16/9/08 15:47:27
17
MU SIC Jeff Smith During the past two decades, the academic study of film music has emerged as a prolific and productive area of film scholarship. This wave of scholarship was initiated by Claudia Gorbman’s (1987) pathfinding study of music in narrative filmmaking, Unheard Melodies, which inspired many scholars in both cinema studies and musicology to follow in its footsteps. Since then, the field of film-music studies has grown exponentially, with authors such as Caryl Flinn (1992), Royal S. Brown (1994), Kathryn Kalinak (1992), K. J. Donnelly (2001), Annahid Kassabian (2001), and Daniel Goldmark (2005) producing important monographs on the subject. In recent years, this intellectual ferment has reached a critical mass, with no fewer than four scholarly journals devoted to the study of music and media, and two series of books in the works at major university presses. Film-music studies has, thus, emerged from the shadows of its more prominent counterparts in cinema and music studies, drawing contributors from the fields of cultural studies, media studies, sound studies, and cognitive psychology. Yet, while the study of film music has reached a sort of maturity in recent years, it remains a very marginal area within philosophy. Perhaps this is because the philosophical study of both film and music remain small subfields within the larger study of art, aesthetics, and communication. To date, only a handful of philosophers have tackled questions and issues related to music in film, and for the most part, even these works tend to focus on more localized queries. Peter Kivy, for example, has written on the aesthetic and filmic functions of film music, which he sees emerging from larger traditions of music and theater, such as opera and melodrama (1997). Noël Carroll proposes a definition of film music as “modifying music,” suggesting that the scores in classical Hollywood films typically function analogously to the role of adverbs and adjectives in language (Carroll 1998: 213–23). For Carroll, music in film helps to characterize the people, places, things, and actions depicted within the diegesis it accompanies, by adding semiotic specificity to their representation. Jerrold Levinson (1996), by contrast, has attempted to define the role of music in its relation to the agencies of cinematic narrator and implied filmmaker entailed in the process of cinematic storytelling. By far the most important work in the philosophical study of film music is Theodor Adorno’s Composing for the Films, which he coauthored with composer Hanns Eisler
TF17557.indb 184
16/9/08 15:47:27
M USIC
(Eisler and Adorno 1994). As a Frankfurt school theorist, Adorno, not surprisingly, offers a neo-Marxist critique of film-music practices as they have developed within the culture industry. Moreover, Adorno and Eisler argue that film music, like other kinds of mass culture, employs conventions and clichés rather than genuine invention. Decrying these “bad habits,” Adorno and Eisler recommend an alternative model of film-music practice that utilizes the formal and stylistic properties represented within the new musical resources provided by Arnold Schoenberg, Igor Stravinsky, Alban Berg, and Anton Webern. Because most of the work on the philosophy of film music has proceeded in a piecemeal fashion, it seems propitious to establish some “first principles” regarding film music’s ontology and its relation to spectators. To that end, I will argue for two specific points in this essay: (1) that film music is best understood through a cluster account of the concept that identifies several possible criteria for its classification; and (2) that a cognitivist account of film-music audition offers the best explanation of the way music contributes to the viewer’s perception and cognition of cinematic representations. While I do not propose to offer definitive answers to these questions, it is hoped that this brief exegesis of the relevant issues might provide a valuable framework for further debate and discussion.
The ontology of film music As Nelson Goodman and others have acknowledged, the ontology of music has been a somewhat slippery concept for philosophers of art (1976). Within the tradition of Western art music, an individual piece, such as Brahms’ First Symphony, exists as a kind of abstract program for performance, with each particular iteration of the piece being slightly different in terms of its interpretation. Although the printed score of the piece fixes its primary musical parameters, such as rhythm, melody, and harmony, secondary parameters, such as dynamics, tempo, and timbre, may vary subtly from performance to performance. The solution to this particular ontological problem has been to treat musical works in terms of a “type-token distinction” that can be traced back to the semiotics of Charles Sanders Peirce. Viewed in this way, a piece of concert music is an abstract sign system that exists as a type, with each performance or recording defined as a token of the type. In contrast to music, film music avoids this ontological problem insofar as it exists as a recorded medium. Although film scores are frequently performed by symphonies in “pops” concerts around the country, the score itself has an independent existence as a component of the film’s soundtrack, which also includes dialogue and sound effects. Soundtrack albums and live performances of film music offer certain advantages to listeners in that they eliminate competing visual and aural stimuli, but they are understood to be derivations of the original score, which exists in a separate physical form as an element in a larger audio-visual totality. Film music, however, poses slightly different problems regarding its definition, having to do with the extent to which the category encompasses a widely divergent group of musical phenomena. Arguing that the study of music in film is marred by 185
TF17557.indb 185
16/9/08 15:47:27
JEF F SM IT H
imprecise terminology, William H. Rosar examines the various ways in which the term “film music” has functioned in film-music scholarship. Noting its origins in earlier studies, such as Kurt London’s landmark Film Music (1936), Rosar argues that the term was originally coined to describe a style or genre of music composition, making the concept analogous to that of incidental music composed for the theater. Over time, however, the concept of film music broadened to include a set of related concepts, such as score, source music, song scores, and compilation scores. For Rosar, the original designation of film music as a term intended to describe a specific approach to music composition has been superceded by academic definitions of film music that focus on the functions of music in films. The result, according to Rosar, is that the emerging field of film-music studies circulates two definitions of the concept that are more or less incommensurate. Rosar suggests that the source of this terminological muddle can be traced to the interdisciplinary study of film music. Indeed, Rosar notes that scholarly inquiry over the last couple of decades has moved apace in the disciplines of both film and music, and further that these scholars have freely borrowed ideas from other domains, such as philosophy, psychology, semiotics, literary criticism, cultural studies, and even marketing research. Because this work has advanced through a “rapidly growing community of writers in various disciplines,” the profusion of scholarship has produced the “potential for a veritable Babel” (2002: 1). Ironically, although scholars refer explicitly to film-music studies, the concept that ostensibly unites the field lacks a common domain of discourse. As a corrective measure, Rosar proposes a conceptual distinction between “film music” and “music in film,” as a means of sorting different types of musical phenomena in the cinema. Rosar suggests that we reserve the term “film music” to refer to the craft and technique of composing original music to accompany filmed scenes and sequences. All other types of musical phenomena, such as source music or interpolated songs, fall into the category of “music in film.” In making this distinction, Rosar explicitly aims to clarify confusion that he believes “already exists in the literature resulting from two incommensurable uses of the term film music” (2002: 15). At the same time, though, Rosar implicitly hopes to place the study of film music on surer footing by providing scholars with a definition that is more specific, more coherent, and more discriminating than the one in current parlance within the field. In advocating a definition of film music that refers to the craft and technique of film scoring, Rosar argues for a lexical designation derived from an understanding of the way the term is already used and accepted. In advancing the case for a customary or reportive definition of film music, Rosar offers an etymological defense of the concept that traces the origins of the term itself through its first usage in German and English publications. For Rosar, the innovation of sound technology had a significant impact on the meaning of film music insofar as it associated the term with scores specially composed for films. This more circumscribed meaning led to an essential definition that highlighted its unique and important characteristics. These characteristics arose out of the silent film’s principle of using musical accompaniment as a form of musical illustration. Thus, for Rosar, film music came to be identified with a compositional 186
TF17557.indb 186
16/9/08 15:47:27
M USIC
technique for fitting music to pictures, rather than an understanding of the music’s function. Perhaps more importantly, the promulgation of scores modeled on late nineteenth-century romanticism yielded a group style, a particular sound that listeners identified as “movie music,” even in compositions not written for the screen. At first blush, Rosar’s essential definition of film music seems to offer both conceptual utility and elegant simplicity. By defining film music as music composed for films, Rosar’s classification highlights its significant characteristics and distinguishes it from other types of music in film, such as “source music” and preexisting classical or popular music (about which I will say more later). The latter items are labeled as “music in film,” a category that would include film music as a particular subset. Upon closer inspection, though, problems emerge from Rosar’s insistence upon divorcing compositional technique from film music’s functions. Rosar argues that this technique arose from the practice of musical illustration, but this simply begs the larger question of what the music is intended to illustrate. This question has no simple answer, but rather must be understood as a range of particular narrative functions that music played in the silent cinema. From a contemporary perspective, musical illustration in the studio era seemed to entail a particular isomorphism between music and image, one in which musical features of film composition are intended to match specific actions depicted in the narrative. Yet this compositional technique, which is sometimes described as “mickey-mousing,” was not part of the repertoire of silent film accompaniment, since it depends on synchronized sound technology for its effect. Instead, the notion of musical illustration promulgated in the silent era was one that correlated musical features with narrative attributes, such as genre, characterization, setting, and mood. Viewed in this way, the compositional technique associated with film scoring in the sound era arose as a consequence of the particular narrative functions that music served in the silent era. Any attempt to define film music in a manner that brackets off those functions seems doomed to failure. Similarly, Rosar’s association of film music with the late romantic idiom is generally accurate as a historical descriptor, but it too fails as an essential definition. The evidence that Rosar cites to support this claim merely shows that the late romantic idiom is sufficient to summon the impression of “movie music,” but hardly a necessary condition for film scoring. The predominance of original music composed in the style of the late romantics was the result of a variety of economic, cultural, and historical factors present during the studio era. These factors were historically contingent, and do nothing to suggest that the romantic idiom has inherent properties that make it more suited to film accompaniment than other types of music. Indeed, during much of the silent era, scores compiled from preexisting concert music or popular songs were easily as common as original compositions for films. By identifying film music with the late romantic idiom, Rosar treats a historical outcome as an essential property of film music, and thereby “freezes” the meaning of the term according to the way its connotations circulated more than sixty years ago. While the problems already evident in Rosar’s ontology of film music might be enough to reject the concept tout court, even more pressing difficulties emerge in his defense of his ontology in relation to competing definitions circulated among 187
TF17557.indb 187
16/9/08 15:47:27
JEF F SM IT H
film-music scholars and historians. Rosar, for example, argues that film music is paradoxically, and perhaps illogically, distinct from film score as a term that refers to nondiegetic, background music. Film scores, unlike film music, need not be written directly for the screen, and may be adapted from preexisting sources. Citing examples, such as The Black Cat (1934) and 2001: A Space Odyssey (1968), Rosar concludes that “score” is a more comprehensive term than film music. Similarly, Rosar also argues against recent attempts to include popular music in the paradigm of film music. In perhaps his most contentious move, Rosar critiques K. J. Donnelly’s claim that the pop music of Performance (1970) constitutes a model of film music that exists as an alternative to the classical Hollywood score. More specifically, Rosar critiques Donnelly’s suggestion that the advent of popular music in films of the 1960s “as a replacement for film music” broke a “film music paradigm that had been weathered but had persisted since the 1930s” (Donnelly 2001: 153). Rosar claims that Donnelly’s definition is caught in a logical contradiction insofar as a film-music paradigm can hardly replace itself. To put it mildly, though, Rosar is rather ungenerous in his reading of Donnelly in suggesting that the latter intends an ontological definition of film music rather than a historical one. Donnelly’s claim about a “replacement for film music” might be clarified by voicing the implicit subordinate clause, “as it was understood historically prior to 1967.” Donnelly’s position advocates a historicized understanding of film music, while Rosar’s paradigm, despite his claims to the contrary, seems fixed, immutable, and unresponsive to historical change. If we accept Rosar’s assertion that the classical style of film music has changed, then his definition of film music needs to specify the reasons why certain types of music were assimilated, while others are consigned to the category of music in film. Rosar’s subsequent discussion of source music versus score and classical versus popular idioms only muddles things further in that he shifts his attention to properties and characteristics of film scores, rather than film music per se. For example, Rosar offers a fairly lengthy discussion of the way in which “As Time Goes By” and “La Marseillaise” appear as performed pieces of music within the diegesis of Casablanca (1942), but also function as musical themes within Max Steiner’s score. Here Rosar’s discussion cogently distinguishes between score and source music – that is, music which is heard by characters within the diegetic space of the film – but it evades a more salient question, namely whether or not Rosar defines Max Steiner’s score as “film music.” If we define film music as original music written directly for the screen, then there seems to be some equivocation about what is meant by original. A cue in Casablanca that features “As Time Goes By” is original in the sense that it did not exist before Steiner wrote it, but the theme itself is certainly not original. Does this commit us to the position that cues featuring themes specifically composed by Steiner are film music, while those cues that feature themes derived from preexisting sources are identified as music in film? If this is the case, then the distinction between film music and music in film seems specious and arbitrary. Scores that emerged as an alternative to Hollywood classicism pose still another problem. As Rosar points out in his discussion of a “movie music” sound, the essential definition of film music derives some of its force from a cinematic character discerned 188
TF17557.indb 188
16/9/08 15:47:27
M USIC
in the music, and further that this cinematic character is heard in works written for the concert stage rather than the soundstage. Yet what are we to make of scores written outside of that late romantic idiom? Composers like Elmer Bernstein and Jerry Goldsmith appear to be worthy inheritors of the traditions disseminated by Steiner, Alfred Newman, Hugo Friedhofer, and Erich Wolfgang Korngold. Yet several of their scores do not employ melody and harmony in ways consistent with the late romantic tradition. Some cues in Goldsmith’s scores, such as Planet of the Apes (1968) and Chinatown (1974), do not sound like “movie music” at all. Does each score’s use of twelve-tone serial technique and jazz, respectively, place it outside the domain of film music? Likewise, does it make sense to identify Bernstein’s score for To Kill a Mockingbird (1962) as film music, but treat his pioneering jazz score for The Man with the Golden Arm (1955) as music in film? Although part of Rosar’s definition emphasizes a cinematic character to film music, perhaps it is not necessary for film music to sound like “movie music.” But this simply opens the door to a range of cases that Rosar would likely reject as “film music.” Can one legitimately include jazz scores like Leith Stevens’ score for The Wild One (1953) or Henry Mancini’s score for Touch of Evil (1958) in the category of film music, but exclude funk scores like Earth, Wind, and Fire’s score for Sweet Sweetback’s Baadasssss Song (1971), Isaac Hayes’ score for Shaft (1971), or Curtis Mayfield’s score for Superfly (1972)? Rosar’s model also encounters problems with films like Out of Sight (1998) and Fight Club (1999), which also appear to be borderline cases for his proposed definition of film music. On the one hand, these scores were specially created for their films and used established techniques to match music to visual action. On the other hand, however, the scores for Out of Sight and Fight Club are hardly composed in the usual manner. Rather DJs David Holmes and the Dust Brothers created these scores by sampling existing recordings to make new pieces of music. Because of its difficulties in dealing with counterexamples and borderline cases, Rosar’s “ordinary language” definition of film music unnecessarily rules out a number of instances that many film-music historians would likely identify as film music. As an alternative, I propose that we treat film music as a “cluster concept,” following the example established by Berys Gaut in his description of art (Gaut 2000, 2005). Drawing inspiration from Wittgenstein’s notion of “family resemblances” as a conceptual basis for categorization, Gaut says that a cluster account is one that cannot be defined in terms of individually necessary nor jointly sufficient conditions. Instead, Gaut suggests that a cluster account employs disjunctively necessary conditions, some of which must be instantiated if the object under investigation is to fall under the concept. That is, no single property is necessary or sufficient for an object to fall under a cluster concept. Rather, a cluster concept employs several criteria, and contains subsets of traits and functions that may be sufficient to count toward an object falling under the concept. In developing his cluster account of art, Gaut offers several reasons why a cluster concept might be preferable to definitions organized around a single essential condition or a set of necessary properties: (1) cluster accounts are better suited for considering borderline cases and counterexamples; (2) cluster accounts are better suited to the 189
TF17557.indb 189
16/9/08 15:47:27
JEF F SM IT H
inductive reasoning protocols employed in considering new objects; and (3) cluster accounts do a better job capturing the way we ordinarily think about the reasons for and against an object’s inclusion in a particular category. Moreover, a cluster account is especially appropriate for a concept of film music, which is part of a creative enterprise that is itself engaged in large-scale processes of aesthetic and historical change. Like the concept art, film music is not rule governed, but is instead shaped by musical conventions and traditions, some of which may be revised or rejected as part of the creative process. A cluster account of film music is, thus, more capable of dealing with borderline cases and counterexamples that come about as a result of invention and innovation, including those that have not yet happened. Following Gaut’s precedent, I also wish to propose a provisional list of criteria to count toward an object’s being film music. This list is not intended to be a definitive taxonomy of film-music properties, but rather should serve as a point of departure in illustrating the advantages of a cluster account of film music. The list contains some overlaps with similar lists offered by Jerrold Levinson and Noël Carroll. My criteria are (1) music specially composed for use as part of a recorded audio-visual medium; (2) music used to accompany cinematic depictions of peoples, places, things, ideas, or events; (3) music used to underline aspects of a film’s setting; (4) music used to communicate a film character’s traits; (5) music used to signify emotion or mood in a filmed scene or sequence; (6) music used to convey a film character’s point of view; (7) music used to accent depicted actions in a filmed scene or sequence; (8) music used to reinforce a film’s formal features, such as its editing; and (9) music that sounds like film music. Armed with a cluster concept of film music, let us revisit some of the examples introduced earlier in the essay to see how they work. Scores composed for classical Hollywood films, for example, exhibit most, if not all, of these various criteria. Consequently, because they exhibit so many of these properties, we can readily see why Rosar tries to define film music in terms of a model exemplified by classical Hollywood composers. Much of the music used to accompany silent films, however, fits criteria 2–8, but not necessarily 1 or 9. This is because much of the music was not specifically written for films, but instead was culled from the literatures of concert music and contemporaneous popular music. The theme song for Goldfinger (1964), which was written specifically for the eponymous James Bond film, would fit criteria 1, 2, 4, 6, and 8. However, a song like “The Sounds of Silence,” which was written before its appearance in The Graduate (1967), omits the criterion for originality. The scores for Superfly, Out of Sight, and Fight Club seem to fit criteria 1–8, but the style of these scores differs from the classical Hollywood paradigm, and thus they do not sound like film music in the sense described by Rosar. Thus, these pop scores would not count as film music under the last criterion insofar as film music has a normative definition that associates it with the romantic idiom of Wagner, Strauss, and Mahler. Finally, a cluster concept of film music can even deal with rather extreme limit cases, such as Alex North’s rejected score for 2001: A Space Odyssey (1967). North’s score fits criteria 1 and 9, but since it was never used in Kubrick’s film, it does not actually accompany images in the film, nor does it perform any of the narrative functions that it was expected to serve. 190
TF17557.indb 190
16/9/08 15:47:27
M USIC
It is also worth noting that this cluster account of film music does not preclude the use of terms like “source music” or “underscore” according to their commonly understood denotations. The concept of source music is logically entailed in criterion 3, regarding music’s role in a film’s setting, but that does not exhaust the ways in which it might count as film music under a cluster account. Source music can, and frequently does, serve other functions, such as mood-setting or conveying a character’s emotions. The same condition also applies to nondiegetic underscore, which routinely fits any of the criteria listed above.
Film music and psychology Recent film-music scholarship has employed two paradigms for explaining the ways in which music in film engages the psychology of film spectators. The first model describes film music’s emotional engagement and semiotic functions in terms of Freudian and Lacanian psychoanalysis. The second explicates film music’s affective dimensions and narrative functions within a cognitive framework. Both paradigms, at least partially, are driven by a question that has bedeviled film-music theorists for ages, namely how does music in film fulfill its narrative functions while operating as something of which viewers are unaware. Film music, thus, seems to have a paradoxical status in terms of its audition; it offers some of the same aesthetic pleasures, formal structures, and emotional signification of other music but is attended to in a way fundamentally different. Several scholars, such as Claudia Gorbman (1987), Caryl Flinn (1992), Annahid Kassabian (2001), and Samuel Chell (1984), employ a psychoanalytic account to explain film music’s unconscious engagement of spectatorial processes. This model suggests that film music plays an especially important role in the constitution of a subject position that individual spectators inhabit when viewing a film. Gorbman analogizes film music’s aural appeal to a hypnotist’s voice that lures us into unconscious obeisance to a film text’s ideological demands. This account is a variant of the “suture” theories that were popularized in film journals like Screen in the 1970s. According to this view, music, like other techniques of cinematic signification, works to “suture” the spectator’s unconscious into that of the film, and music is specially equipped for this task, due to its direct pull on the viewer’s emotions. In contrast to this model, researchers in the subfield of psychomusicology have fleshed out a cognitive account of film-music perception. This paradigm developed through series of experiments that have measured, among other things, music’s effect on our perception of the emotional meaning of depicted actions and characters, music’s ability to enhance the viewer’s memory of filmed events, music’s ability to shape the viewer’s interpretation of diegetic situations and to alter their expectations of future narrative outcomes (Marshall and Cohen 1988; Lipscomb and Kendall 1994; Boltz et al. 1991; Bullerjahn and Güldenring 1994). In summarizing this research tradition, Annabel Cohen proposes a congruenceassociationist framework for understanding film-music communication (2001). For Cohen, cinema is perceived through three parallel channels that are domain specific 191
TF17557.indb 191
16/9/08 15:47:27
JEF F SM IT H
and that are used to process the meaning and structure of speech, music, and visual information communicated through the film’s narration. From moment to moment, the viewer utilizes both bottom-up and top-down perceptual processes to extract the narrative information and emotional meanings necessary to form a coherent story in the diegesis. More importantly, perhaps, the ability to form a coherent story depends upon the viewer’s understanding of crossmodal structural congruence that affects our perceptual groupings in visual and auditory domains in short-term memory. That is, film music works through the viewer’s cognitive processing of perceived correspondences between musical and visual information. For Cohen, although the acoustical aspect of the music is encoded by sensory memory, it fails to be transported to shortterm memory. Instead, because of inferences derived from long-term memory, the acoustical aspect of music falls away while the information about the narrative is stored. Cohen’s associationist-congruence model, thus, explains how music in film communicates the emotional significance of narrative events and arouses low-level affects in viewers. Within each of these paradigms, film music achieves its particular effects by operating at an unconscious level. For Gorbman, film music would cease to have its ideological effect if audiences were aware of its presence; if the subject hears the hypnotist’s voice, the game is over. Likewise, Cohen cites empirical evidence to support the notion that film music is apprehended unconsciously. She references a survey conducted by Archie Levy in which subjects were asked while leaving the theater to evaluate the music’s effectiveness in the film they just saw. Most individuals reported that the film’s score was fine despite the fact that there was no music actually present in the film except for the opening credits (Cohen 2000: 366). Yet it is clear that each paradigm has a different understanding of unconscious listening. For adherents of the psychoanalytic paradigm, the appeal of film music is grounded in processes of unconscious fantasy, more specifically the primordial fantasy of the chora. According to this theory, film music, like all types of music, engages the fantasy of the subject’s unity with the mother’s body insofar as music creates a sonorous envelope that is reminiscent of our experience prior to birth within the womb (Rosalato 1974; Anzieu 1976). In contrast, cognitive theorists typically take unconscious listening to simply mean that sound is apprehended below the threshold of our conscious attention. Cohen, in fact, suggests that film music’s inaudibility is similar to the phenomenon of “inattentional blindness.” In experiments conducted by Dan Levin and Daniel Simons, empirical research has shown that, due to limitations on our attention, subjects are “blind” to much of the information present within a visual field (cited in Cohen 2001: 256). Like the acoustical aspect of music, part of the visual field in these experiments is perceived unconsciously, but the information is not retained in either short- or long-term memory. While the relation between film-music audition and inattentional blindness is intriguing, Cohen suggests that a better parallel might be developed through research on the role of prosody in speech perception, which provides cues to listeners in the form of intensity and intonation of speech patterns, but which subjects generally ignore in favor of the semantic content of a message. 192
TF17557.indb 192
16/9/08 15:47:27
M USIC
While I wholeheartedly endorse the model of film-music psychology elaborated in Cohen’s congruence-associationist framework, I also propose that we might consider film-music audition to be a phenomenon akin to subliminal perception. Philip Merikle et al. define subliminal perception as something that occurs whenever “stimuli presented below the threshold or limen for awareness are found to influence thoughts, feelings, or actions” (Merikle et al. 2001: 123–4). Film music seems to fit both conditions for Merikle et al.’s definition insofar as music is perceived by film viewers, for the most part, without awareness, and clearly affects their perceptions and thoughts about the events depicted on screen. Although she does not describe it as such, Cohen’s review of the psychological literature on film-music perception seems to support these necessary and sufficient conditions for subliminal phenomena. Beyond this prima facie case, though, it seems to me that there is also some additional heuristic value in thinking about film-music audition as a type of subliminal perception. For one thing, it is a more precise descriptor than the notion of unconscious listening. The latter concept is perhaps too broad to be of much value. After all, George Lakoff and Steven Johnson point out that 95 percent of all thought is unconscious thought (Lakoff and Johnson 1999: 13). Treating film-music audition as part of that undifferentiated mass does little to communicate what is unique and specific about the phenomenon. More importantly, though, the notion of “unconscious listening” is tainted, especially in academic film studies, by its association with a psychoanalytic framework for analyzing film’s effects on viewers. For that reason alone, it is perhaps better to reject the term “unconscious listening” outright so as to avoid the implication that Cohen’s “associationist-congruence” framework somehow entails concepts like the Oedipus complex or the Mirror Stage. Second, subliminal perception also better captures film music’s ability to influence cognitive processes. Cognitive psychologists have performed several experiments that demonstrate the ways in which subliminal phenomena “prime” subjects’ cognitive processes by influencing both what stimuli are perceived with awareness and how those stimuli are experienced. Many of these experiments tested the effects of masked stimuli on subjects’ ability to recognize or complete letter strings. For example, in a 1998 study by Mack and Rock (1998), subjects were asked to complete the word stem fla- with the first two English words that came to mind. When the word flake was presented to them as masked stimuli, 36 percent of the subjects selected that as one of their two words. In contrast, among the control subjects, only 4 percent selected flake as one of their two completions. In another study by Moore and Egeth (1997), experimenters used subliminal stimuli to produce the famous Müller-Lyer illusion. For this test, subjects were asked to evaluate the relative length of horizontal lines against a random pattern of black-and-white dots. During the first three trials, the subjects correctly identified which line appeared to be longer. In the fourth trial, however, the lines were equal, but were set against a background pattern in which the black dots formed either inward or outward pointing arrows. Subjects identified the line with the inward pointing angle as longer than that with the outward pointing angle, but were unaware of any pattern in the black dots themselves. 193
TF17557.indb 193
16/9/08 15:47:28
JEF F SM IT H
This kind of priming effect does not seem to be a characteristic of the unattended stimuli in Levin and Simon’s inattentional-blindness studies. In those experiments, the changed conditions that go unperceived do not seem to influence thought or behavior; the stimuli are simply there, but are unattended. Thus, although it is fair to say that viewers experience a kind of “inattentional deafness” with respect to film music, that descriptor does not positively capture the extent to which film music primes our conscious awareness of significant character traits, emotional subtexts, genre conventions, or our expectations of narrative outcomes. Think of the spooky music used to suggest the presence of the killer in the closet or the orchestral swell of romantic music that leads us to anticipate the couple’s kiss. For these reasons, the notion of film music as a kind of subliminae offers the most heuristically useful concept for considering the ways in which film scores influence how viewers consciously perceive diegetic events.
Conclusion: the aesthetics of film music Over the past several decades, the aesthetics of film music has been governed by a tacit concept of appropriateness. By appropriateness, critics and practitioners usually mean that the music matches well with the particular story that it accompanies, that the music effectively and efficiently performs its dramatic functions, and that it does so without distracting the viewer’s attention from more salient narrative information. This provides a good rule of thumb for evaluating the aesthetic qualities of film music, but it is only a loosely articulated principle. Judgments about the worth of particular film scores or specific cues are themselves context dependent. Since every film poses its own problems and challenges, every film score must try to solve those problems and meet those challenges in its own way. That said, even a loosely articulated notion of appropriateness presupposes some understanding of film music’s ontology, as well as of the ways film music engages the psychology of film viewers. To offer conclusions about a particular score’s effectiveness or greatness, one first must have a clear understanding of what film music is and what it is supposed to do. In this essay, I have tried to speak to those issues by sketching a cluster account of film music’s ontology and the role of unconscious listening in film perception. As it turns out, music paradoxically may be most noticeable in its absence, during those moments in film when it is most commonly expected. Underlining its importance in cinematic storytelling, composer Bronislau Kaper averred, “Nothing is as loud in films as silence” (Karlin 1994: 11).
Acknowledgments I’d like to thank Mark Minett for his comments on disc jockeys turned film composers. See also Sound (Chapter 24).
194
TF17557.indb 194
16/9/08 15:47:28
M USIC
References Anzieu, D. (1976) “L’enveloppe sonore du soi,” Nouvelle Revue de Psychoanalyse 13 (spring): 161–79. Boltz, M., Schulkind, M., and Kantra, S. (1991) “Effects of Background Music on Remembering of Filmed Events,” Memory and Cognition 19: 595–606. Brown, R. (1994) Overtones and Undertones: Reading Film Music, Berkeley: University of California Press. Bullerjahn, C., and Güldenring, M. (1994) “An Empirical Investigation of Effects of Film Music Using Qualitative Content Analysis,” Psychomusicology 13: 99–118. Carroll, N. (1998) Mystifying Movies: Fads and Fallacies in Contemporary Film Theory, New York: Columbia University Press. Chell, S. (1984) “Music and Emotion in the Classical Hollywood Film: The Case of The Best Years of Our Lives,” Film Criticism 8(2): 27–38. Cohen, A. (2000) “Film Music: Perspectives from Cognitive Psychology,” in J. Buhler, C. Flinn, and D. Neumeyer (eds.) Music and Cinema, Hanover, NH: University Press of New England, 360–77. —— (2001) “Music as a Source of Emotion in Film,” in P. Justin and J. Sloboda (eds.) Music and Emotion: Theory and Research, Oxford and New York: Oxford University Press, 249–72. Donnelly, K. J. (2001) “Performance and the Composite Film Score,” in K. J. Donnelly (ed.) Film Music: Critical Approaches, New York: Continuum, 152–66. Eisler, H., and Adorno, T. (1994) Composing for the Films, London; and Atlantic Highlands, NJ: Athlone Press. Flinn, C. (1992) Strains of Utopia: Gender, Nostalgia, and Hollywood Film Music, Princeton, NJ: Princeton University Press. Gaut, B. (2000) “ ‘Art’ as a Cluster Concept,” in N. Carroll (ed.) Theories of Art Today, Madison: University of Wisconsin Press, 25–44. —— (2005) “The Cluster Account of Art Defended,” British Journal of Aesthetics 45(3) (July): 273–88. Goldmark, D. (2005) Tunes for “Toons”: Music and the Hollywood Cartoon, Berkeley: University of California Press. Goodman, N. (1976) Languages of Art, Indianapolis: Hackett. Gorbman, C. (1987) Unheard Melodies: Narrative Film Music, London: British Film Institute; and Bloomington: Indiana University Press. Kalinak, K. (1992) Settling the Score: Music and the Classical Hollywood Film, Madison: University of Wisconsin Press. Karlin, F. (1994) Listening to Movies: The Film Lover’s Guide to Film Music, New York: Schirmer Books. Kassabian, A. (2001) Hearing Film: Tracking Identifications in Contemporary Hollywood Film Music, New York: Routledge. Kivy, P. (1997) “Music in the Movies: A Philosophical Enquiry,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Clarendon Press; and New York: Oxford University Press, 308–28. Lakoff, G., and Johnson, S. (1999) Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Thought, New York: Basic Books. Levinson, J. (1996) “Film Music and Narrative Agency,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press, 248–82. Lipscomb, S., and Kendall, R. (1994) “Perceptual Judgment of the Relationship between Musical and Visual Components in Film,” Psychomusicology 13: 60–98. London, K. (1936) Film Music: A Summary of the Characteristic Features of Its History, Aesthetics, Technique, and Possible Developments, trans. E. Bensinger, London: Faber and Faber. Mack, A., and Rock, I. (1998) Inattentional Blindness, Cambridge, MA: MIT Press. Marshall, S., and Cohen, A. (1988) “Effects of Musical Soundtracks on Attitudes toward Geometrical Figures,” Music Perception 6: 95–112. Merikle, P., Smilek, D., and Eastwood, J. (2001) “Perception without Awareness: Perspectives from Cognitive Psychology,” Cognition 79: 115–34. Moore, C. M., and Egeth, H. (1997) “Perception without Attention: Evidence of Grouping under Conditions of Inattention,” Journal of Experimental Psychology 23: 339–52. Rosar, W. (2002) “Film Music – What’s in a Name?,” Journal of Film Music 1: 1–18. Rosolato, G. (1974) “La voix: entre corps et langage,” Revue Française du Psychoanalyse 38 (January): 75–94.
195
TF17557.indb 195
16/9/08 15:47:28
18
NARRAT ION Noël Carroll Most of the motion pictures we encounter – whether nonfiction or fiction – are narrative in nature. Motion pictures, of course, are not the only medium in which narrative figures prominently. Nevertheless, narrative motion pictures are so pervasive that a discussion of cinematic narration is an unavoidable topic for the philosophy of the moving image. Because it seems reasonable to presuppose that narration implies a narrator (Chatman 1990), a question naturally arises about the nature of that narrator. Who or what is the cinematic narrator? The answer might appear to you to require little thought. The narrator is just the motion picture maker or makers responsible for the work at hand. If we are talking about a nonfiction film, like The Fog of War (2003), then Errol Morris is the narrator. If we are speaking of a fiction film, such as No Country for Old Men (2007), then the narrators are the Coen brothers. However, even if the answer seems fairly straightforward when it comes to cases of nonfiction, many philosophers suspect that works of fiction are more complicated. To see why, it is useful first to take a brief detour into literary theory. With respect to the study of literary fictions, distinctions are often made between the actual author, the implied author, and the narrator. The actual author is the real person who wrote the text and is presumably collecting royalties for it. The implied author is the author as he or she manifests him or herself in the text. The implied author may in fact share all her beliefs, desires, attitudes, allegiances, and so forth with the actual author, but, equally, she may not. The actual author – a romantic at heart – may don the persona or mask of a cynic for the purpose of telling her tale in a certain way. The implied author is the agent who is responsible for the way the fiction is written – its tone, structures, ellipses, emphases, etc. qua fiction. Yet in addition to the actual and implied authors of a literary text, there may also be so-called narrators. These are fictional creatures. They are part of the fictional world as it is presented by the text. Indeed, they are the fictional presenters of the text. Often these presenters appear as characters in the story, as in the case of Henry James’ What Maisie Knew. Watson is the explicit narrator with regard to the adventures of Sherlock Holmes. He, as a character, is overtly introduced in the fiction as its narrator; it is fictional, in the stories he inhabits, that he is presenting the story to us. But in addition to overt or explicit narrators, like Watson, it is also argued that
TF17557.indb 196
16/9/08 15:47:28
NA R R AT ION
there may be implicit fictional narrators. For example, in Ring Lardner’s classic short story “The Haircut,” there is an explicit narrator, the barber, but it is not he, with his blinkered understanding of the situation, who lets us know that Jim Kendall was actually murdered. Yet, it is reasonable to suppose that we have been told that this is true in the fictional world of the short story. But if narration requires a narrator, who told us this was so? An implicit fictional narrator. Why must we believe that there are such implicit fictional narrators? The argument goes like this: A fiction is something that we are mandated to imagine. We are mandated to imagine the events in the fiction as true – as obtaining in the world of the fiction. Where an explicit fictional narrator is telling the tale, we typically imagine what he or she reports as true in the world of the narrative (unless we have some reason to suspect the explicit narrator is unreliable or limited in one way or another). However, where there is no explicit narrator in evidence, it is argued that we still have reason to suspect there is an implicit one nearby – that is, a narrative agency that asserts or reports what is true from inside the world of the story. Why? First, recall that it is being presupposed that there is no narration without a narrator. But when it comes to fiction, it is alleged that the narrator cannot be the actual author or the implied author. The actual author cannot report things as true in the story world; what the actual author does is establish that it is fictional that thus and such (rather than asserting that, or reporting that it is true that thus and such). Margaret Mitchell made it fictional that Scarlett O’Hara lived at Tara; Mitchell did not perform the illocutionary act of asserting “Scarlett O’Hara lived at Tara.” Had Mitchell done that, she would have uttered a falsehood. The same problem besets the implied author, since the implied author is responsible for Gone with the Wind qua fiction. Both the actual author and the implied author stand, so to speak, on the wrong side of the fiction operator (the “it is fictional that . . .” operator) with respect to making assertions or reports about what is the case in the fictional world. Implied authors can make assertions about what is fictional, but not about what is true in the narrative. Yet it is a true narrative assertion in the fiction that Scarlett O’Hara lived at Tara. Who made this narrative assertion along with all the others we imagine to be true in the fictional world of Gone with the Wind? If there is no explicit narrator in view, we are urged to postulate an implicit fictional narrator, a narrational agency who (unlike the actual author) believes Scarlett O’Hara, Rhett Butler, Ashley Wilkes, and the rest exist, and who reports their trials and tribulations to us as facts which we then go on to imagine. Moreover, the implicit fictional narrator is, like explicit fictional narrators, a denizen of the fiction – a fictional character – even if he or she is unacknowledged by the other fictional inhabitants of his or her world. The question for the philosophy of the moving image is which of these literary distinctions carry over to motion pictures and to what extent (i.e., if there are implicit fictional narrators, for example, how frequently do they occur: always, sometimes, or never?). Obviously motion pictures have actual authors – as Zhang Yimou is the actual author of Ju Dou (1990) – whose manifestation is implied in the film, whether sincerely or not. Motion pictures may also have explicit narrators – like Lester 197
TF17557.indb 197
16/9/08 15:47:28
NOËL CA R ROLL
Burnham in American Beauty (1999) – who are characters in the story, or explicit narrators in the form of voice-over commentators, as in the case of The Magnificent Ambersons (1942). But do motion picture fictions have implicit fictional narrators, and, if so, how extensive is the phenomenon? There may be some reasons to deny the postulation of implicit fictional narrators across the board. Putatively, we are forced to posit the aforesaid narrator universally or ubiquitously because narration presupposes a narrator. But perhaps the possibility that the actual author is the narrator has been dismissed too quickly. Why suppose that there must be an act of asserting or reporting inside the fictional operator in order for such and such to be true in the fiction? Rather, what makes something true in the fiction is that the actual author has mandated that we imagine (i.e., entertain as unasserted) certain propositional contents – for example: it is true in the fiction Psycho (1960) that Norman Bates is psychotic, just in case Alfred Hitchcock and his team of fictioneers have mandated that we imagine the propositional content “that Norman Bates is psychotic.” In other words, there is no apparent pressure to presume that there are acts of assertion going on within the fiction. That appears to require of the viewer one thought too many or, at least, one thought that is not necessary. Defenders of the implicit-narrator hypothesis argue that it is natural to imagine that there is a continuous activity of reporting going on as a novel unfolds (Wilson 2007). But claims of naturalness are not likely to carry the day here. Not only do I not find myself entertaining the report model, I see no necessity in doing so. When I watch a play, it would seem strained to think that I am watching a report, so why think that I am reading one upon encountering, for instance, a dialogue novel in the fashion of Ivy Compton Bennett? That is, the report model hardly fits every case of narration. It does not fit what Plato called mimesis in books 2 and 3 of the Republic. Hence, claims on behalf of the report model will have to be made on a case-by-case basis, not summarily. And to the extent that the report model does not obtain comprehensively, there is no reason for the blanket postulation of implicit fictional narrators, since their postulation would appear to rest upon the report model. Yet, since the report model occurs at best occasionally, so will the need for introducing implicit fictional narrators. Friends of the report model may find it natural because they endorse the notion that fictions are props in games of make-believe and that the consumption of a fiction involves the reader in making-believe that he or she is reading a report. But again, although some readers may behave this way, it is idiosyncratic enough that asserting that it is natural is immensely controversial. For how many of us are aware of playing this game? So the defender of implicit fictional narrators may be involved in attributing to audiences two more thoughts than are needed – the idea that there is a report before us and the idea that we are make-believing that we are reading one. Another reason to be suspicious about the postulation of ubiquitous, implicit fictional narrators is that they often lead to self-contradiction. For example, readers often learn how things turn out in a story where, at the same time, it is given that no one in the fiction ever learnt what happened (Currie 1995). But if there is an implicit 198
TF17557.indb 198
16/9/08 15:47:28
NA R R AT ION
narrator who dwells in the fictional world, then there is someone or some agency in the fiction who knows what went down and who is reporting it to us. So, in cases like this, we are driven to the unhappy conclusion that no one in the fiction ever learnt the outcome of the story, and yet someone from the fiction is telling us the outcome. Of course, the way to avoid this logical contretemps is simply to abstain from postulating the existence of the implicit fictional narrator in the first instance. Thus far, the arguments against the ubiquity of implicit fictional narrators have applied alike to both literary and cinematic examples. But it might be alleged that there is something about motion pictures that especially warrants hypothesizing them, and which so far our objections have left untouched. Specifically, some may contend that it has to do with the way in which we interact with the visuals in a motion picture. On their account, when we see a house on screen, it is natural for us to suppose that “I imagine seeing a house” (or, that “I am seeing the house imaginarily”). So when presented with an establishing shot of the little house on the prairie, I imagine seeing the little house on the prairie. But for some philosophers, this raises the question of who in the fictional world is responsible for showing us the house; they maintain that reason demands an answer to this question (Levinson 1996). Ostensibly, the actual filmmakers cannot be showing us the house, since the house is fictional, and they are merely showing us an actual house that is being used to represent the fictional house. So, it is argued, we must postulate some fictional presenter at work here. It is this implicit fictional presenter who makes it possible for us to perceive imaginarily the sights and sounds of the fictional world, something the actual filmmaker cannot do, since he/she only has access to the actual world. Nevertheless, insofar as this argument rests upon the notion of seeing imaginarily, it falters. For the notion of seeing imaginarily does not seem very compelling. Watching a gun battle in The 3:10 to Yuma (2007) I find myself curiously unscathed, nor do I imagine myself ducking bullets. Yet if I were imagining that I was seeing a gunfight close up, wouldn’t I have to imagine that my life was endangered? Can I be imagining that I’m seeing a gun battle from a vantage point inside the line of fire and not imagine bullets bursting midair around me? And if I were imagining that, would I continue eating my popcorn so nonchalantly? However, I do eat my popcorn nonchalantly, because I don’t imagine myself amidst a blizzard of flying steel, and if I don’t imagine that I am amidst the fire fight, how can I plausibly imagine that I am seeing it? Similarly, many camera positions are such that were I to imagine myself to be an eyewitness inside the fictional world, I would also have to imagine myself in some very unlikely places. In Casino Royale (2006), a building in Venice sinks into the lagoon. Do I imagine myself submerged? Do I imagine myself wet, when I am dry? Of course, I don’t imagine these things. But how then can it be the case that I am imagining that I am having a close encounter of the third kind within the world of the fiction? And what do I imagine seeing when there is an on-screen dissolve or a wipe? That the world is immaterializing or erasing itself in my presence? Cuts are also a problem. One moment we are shown Paris and the next moment Moscow is before us. If we imagine seeing these two sights in rapid succession, then wouldn’t the question of how 199
TF17557.indb 199
16/9/08 15:47:28
NOËL CA R ROLL
I moved across Europe so quickly arise? Yet it doesn’t, nor do I feel any pressure to imagine how I pulled off this miracle – just because I wasn’t seeing imaginarily in the first place. In other words, if I were seeing imaginarily, I would have to perform a wealth of unlikely supplemental imaginings in order to account for how I came to be able to perform my primary feats of imaginary seeing (Currie 1995). But I do not appear to perform the requisite supplemental imaginings. So the notion of imagining-seeing appears dubious. Instead of seeing imaginarily, it seems that on a more plausible account of my relation to the visual array, I literally see representations of actors on screen, which I use to imagine the fiction. In other words, we may jettison talk of seeing imaginarily. And, if the hypothesis of imagining-seeing is supposed to support the hypothesis of the implicit fictional narrator/presenter, then the two ideas fall together. Another problem, apart from that of improbable supplemental imagining, is that the hypothesis of the fictional presenters who guide our imagining-seeing can lead to self-contradiction. For example, in the TV series Six Feet Under (2001–5) the character, Nathan Fisher, buries his wife in the desert; in the fiction, it is given that no one sees this. Yet, if there is an implicit fictional narrator/presenter, who enables us to see the burial imaginarily, then someone has witnessed the event. But this yields the contradiction that it both is the case that, in the fiction, no one witnessed the burial and that someone witnessed the burial. To avoid this problem, the obvious solution is to eschew the notion of an implicit fictional narrator/presenter (Currie 1995). However, not only does the implicit fictional narrator/presenter lead to untoward consequences, the reasoning in its behalf appears ill-motivated. Supposedly, we are in need of these fictional intermediaries because actual and implied authors lack the right metaphysical relationship to the world of the fiction. They cannot provide us with the proper metaphysical access to the fictional world, because they live on the wrong side of the fiction operator. But if the contents of the fictional world are inaccessible directly to actual authors, implied authors, and their audiences, why would not the same problem apply to the alleged fictional presenters as well as to the named fictional characters in the motion picture? If there is any problem with making contact with the fictional world from outside the fiction operator, that problem, ex hypothesi, would persist with respect to making contact with an implicit fiction narrator/presenter (Kania 2005). That is, if we need a fictional intermediary to secure access to whatever is fictional, and the implicit narrator/presenter is fictional, then to make contact with the first implicit fictional narrator/presenter, we will need to postulate a second fictional narrator/presenter, and then, for the same reason, a third, and so on ad infinitum. The thinking behind the idea of such a narrator threatens to drag us in to an endless regress, such that we could never secure access to the fictional world. But we do have access to fictional worlds. Thus, so much the worse for the implicit fictional narrator/ presenter (Carroll 2006). One way to mute some of the objections to the notion of the implicit fictional narrator/presenter is to think of this narrative agency as a documentary motion picture 200
TF17557.indb 200
16/9/08 15:47:28
NA R R AT ION
(Wilson 2006). Instead of imagining that we are seeing the world of the fiction under the guidance of an implicit narrator who beckons us to look here and then there, we are to imagine that we are viewing a documentary film produced in the world of the fiction. Just as we are supposed to imagine reading the novel A Tale of Two Cities as a report from inside the fiction, when we watch the movie adaptation of the novel, we are to imagine that we are seeing a nonfiction film about happenings that occurred during the French Revolution. This will dodge many of the previous objections involving supplemental imaginings, for if what we imagine seeing is a movie, then we don’t have to explain to ourselves why we weren’t wounded during the gunfight, or dampened during Casino Royale, nor do we have to imagine a story about how we manage to get across Europe in less than a second or be perplexed about what we are to imagine seeing when confronted with visual effects like dissolves and wipes. We can account for all this and more on the supposition that it is a movie, albeit a nonfiction movie, that we imagine we are seeing. Although the documentary version of the report model succeeds quite well with some of the problems of supplemental imagining, it has a number of liabilities. First, it seems plausible that many viewers encountered motion picture fictions before they were exposed to nonfictions. This may be true of children and people in remote areas. These viewers may not yet have the concept of nonfiction film or any idea of how such films are made. Thus, they would not be capable of imagining that they are seeing a nonfiction documentary. But they show no confusion over how they are to process the motion pictures in question. Furthermore, there is the problem of historical films, like Quo Vadis? (1951), Attila the Hun (1954), or Gladiator (2000). Are we to imagine seeing a documentary centuries before moving-picture cameras were invented? And, in addition, the documentary hypothesis does not dispel the problem of motion picture events we are to imagine which were witnessed by no one. Surely we are mandated to imagine at the end of Greed (1924) that there are no onlookers when Marcus and MacTeague die in the desert. But this is incompatible with the documentary hypothesis, which would place a camera crew within spitting distance of the doomed men. Moreover, the documentary hypothesis not only incurs a logical problem here: it would force most viewers to raise ethical questions about the morality of the film, given the callousness of the supposed documentary filmmakers, who idly stand by as these two men perish. One way to blunt these objections is to claim that we imagine that these documentaries are not human artifacts. They are naturally occurring objects – specifically they are naturally iconic representations or images, perhaps like mirrors (like the ones owned by wicked witches) (Wilson 2006). If we imagine that something of this sort is what transmits the images that we imagine seeing, then the problem of motion picture representations of witnessless events disappears, since these natural icons are not humans. Nor are movies whose events antedate the invention of cinema troublesome, so long as we presuppose that these natural iconic representations have been around since the dawn of time. 201
TF17557.indb 201
16/9/08 15:47:28
NOËL CA R ROLL
One might respond to the idea of natural iconic representations by objecting that imagining that we are encountering such entities has implications as unlikely as many of the implications involved in imagining we are confronting things like gunfights face-to-face. How do these naturally occurring iconic representations work? Exactly what are we supposed to imagine – simply that there are such things? But is that any less outlandish – or, at least, any less “magical” – than imagining that somehow, on a single cut, we can move across Europe faster than the speed of light? If the consequences of face-to-face imaginary seeing strain the bounds of sense, how much better off are we with these allegedly natural iconic representations? Moreover, if these naturally occurring iconic representations are supposed to answer the question of who is narrating, then doesn’t this process have to be an agent of some sort, like a human or, at least, an anthropomorphic person? But haven’t natural iconic representations been introduced precisely to avert the problems that arise when we imagine we are dealing with the work of a human documentarian? Of course, it might be stipulated that the natural iconic representations constitute some sort of imagemaker, but if that imagemaker is a person, then we are back to the problem of reports of putatively witnessless events. So, there appears to be a dilemma in the offing here: either the natural iconic representation is not an agent of some sort (violating the demand that narratives have narrators) or the natural iconic representation is an imagemaker (returning us to the problem of witnessless events) (Carroll 2006). One way of attempting to repel this kind of criticism is to maintain that when we are mandated to imagine thus and so regarding the world of the fiction, we are not thereby mandated to imagine everything about how thus-and-so came to be, although we will suppose that it came about somehow. The Son of Kong (1933) features a scion to the one-of-a-kind King; we suppose the son to be legitimate and imagine he inhabits Skull Island, but we need not imagine how he got there with no Queen Kong in evidence. Fictions leave much concerning that which we are mandated to imagine indeterminate and unexplained, but we are not mandated to fill in the gaps, it may be argued by the defenders of the implicit fictional narrators/presenters and/or natural iconic representations. Thus, we are not mandated to imagine how natural iconic representations work, nor whether they are agents. Nor are we prescribed to imagine everything such posits entail. We just imagine that, howsoever these things operate, they are producing the images of goings-on in the fictional world. Indeed, in the old Flash Gordon (1936) serials, there was a viewing device that enabled one to see anywhere in the galaxy with no recording devices in evidence. Ours was not to reason why or how. We were simply to imagine that there were such machines in Flash’s universe. Thus, we may think likewise of the alleged naturally occurring iconic representations. The implicit, fictional presenter/narrating agency has access to or is identical with an equally unexplained and narratively underdeveloped mechanism which allows us to see imaginarily the pictures it produces, although we know not – and cannot even imagine – how. Just as you can imagine that you are in bed with your 202
TF17557.indb 202
16/9/08 15:47:28
NA R R AT ION
favorite celebrity – without imagining how you got there – so we may imagine seeing all sorts of images without imagining how we gained access to them. If such images appear to provoke certain anomalies – such as recordings of putatively unrecorded events – we need not imagine how the events were recorded in such a way that evades this apparent contradiction. We merely imagine that the images are somehow available, sans absurdity, for us to imagine seeing. However, this suggestion will not do, for it is not true that when processing a fiction we are not mandated to imagine a great deal of what is presupposed or implied by what is given in the world of the motion picture. When assimilating a fiction, we constantly need to fill in many of the details that the creator of the story has left out, including things entailed by the narrative, although not stated outright or shown in it. For example, we are mandated to imagine that Philip Marlowe has a heart and that if he is shot in it, he will die. This supplemental imagining is governed by a default assumption – unless otherwise instructed by the fiction, assume that the world of the fiction is like our world and imagine accordingly. Since this is a default assumption, it may be overridden. Some genres, for example, presuppose things at variance with the ways of the actual world – for example, that mummies can be brought back to life. And stories from other cultures and times may come with presuppositions at odds with the way we believe the world works and, in order to process these fictions, we will need to adjust our imaginings to alien views. However, our default response to a fiction is to fill it in in terms of our beliefs about how the world is. This is called the realistic heuristic. Moreover, this heuristic raises problems for the attempt to save the hypothesis of the implicit fictional presenter/narrator by means of the bold assertion that we are not mandated to worry, in our imaginings, whether natural iconic representations provoke contradictions. For, insofar as there is a realistic heuristic, it is wrong to contend that we are not mandated to imagine altogether that which has not been said or shown in the fiction. We are mandated to imagine that Philip Marlowe has a heart, that a bullet can stop it, and that because of this, Marlowe will tread cautiously when a firearm is pointed his way. Similarly, when we deploy the realistic heuristic to the fiction of naturally occurring iconic representations, then, if there are salient implications such a device would have in the world as we know it, then we are entitled to imagine they prevail in the fictional world, unless it is stipulated otherwise by the actual author of the fiction or the presuppositions of the genre to which the fiction belongs. Just as J. K. Rowling establishes that owls can deliver mail to young wizards in the Harry Potter series, so it would fall to Chandler to alert us to Marlowe’s invulnerability, should we be mandated to imagine it. The realistic heuristic can be overridden. If it is stipulated by Chandler that we are to imagine that Philip Marlowe is invulnerable, then we will suspend the realistic heuristic and we will not be nonplussed when he survives repeated blastings. Likewise, we do not wonder about how the viewing device works in the Flash Gordon serials, because we have been told that it works and not to worry our pretty heads about it. But notice that in these cases, the realistic heuristic can be retired because the fiction 203
TF17557.indb 203
16/9/08 15:47:28
NOËL CA R ROLL
has explicitly told us to do so. Flash Gordon’s all-seeing television has been introduced into the story straightforwardly, and we have been overtly reassured that it works in the Flash Gordon universe. But the same is not true of the implicit fictional narrator/ presenter or of the naturally occurring iconic representation. Therefore, if these postulations fall afoul of the realistic heuristic with respect to the supplemental imaginings they enjoin, then the disturbance they represent for our imaginings is genuine (Gaut 2004). In fact, since the implicit fictional narrator is, by definition, implicit, no fiction tells us that it possesses an implicit fictional narrator/presenter or an implicit naturally occurring imagemaker. So, on the one hand, we have no reason to postulate the existence of one on the grounds of the default assumption of the realistic heuristic. On the other hand, if we are told that there are such things by some theorist, then, since within the fiction the realistic heuristic has not been suspended, we can, again by dint of the realistic heuristic, wonder whether the fiction is logically intelligible where it appears to violate the realistic heuristic. That is, if these theoretical postulations concerning the implicit fictional narrator presuppose or entail any absurdity – such as an event given as unrecorded which has been recorded – the contradiction cannot be evaded by appealing to the idea that we need not imagine that which the recommended postulation entails because, without a stipulation to do otherwise, we are to imagine that the kinds of logical, physical, and psychological implications that obtain in the actual world obtain in the world of the fiction. And, of course, if these postulations are implicit, then it has not been stipulated that we should think otherwise. Another consideration against the notion of the naturally iconic imagemaker is that it seems far too complex a posit to attribute to any viewers, save those steeped in analytic metaphysics. As we have seen, the notion of naturally occurring imagemaking emerges from an intense dialectic and is designed to deflect certain counterexamples and conundrums that beset the notion of ubiquitous, implicit, fictional narrators/ presenters. Such a mechanism is extremely unlikely to occur to most viewers. It is not apt to be part of their imaginative processing of fictional motion pictures, since they lack the concept. And yet most do assimilate such movies successfully. Thus, they must be doing it without the benefit of the various metaphysical hypotheses presented so far in the debate (Carroll 2006). Perhaps if normal viewers can do without such things, motion picture philosophers can do so as well. Still another problem with the naturally iconic imagemaker as a solution to the perplexities raised by the implicit narrator/presenter is that even if we could imagine some natural process that might give rise to individual motion picture images, what conceivable natural process could edit them into a coherent story? The natural iconic imagemaker does not merely present single images; those images get organized into narratives. However, there is no Flash Gordon, Dale Arden, or Dr Zharkoff directing the imagemaker where to look and in what order. Even if naturally occurring iconic images were conceivable without too much effort, can entire naturally produced motion picture narratives be imagined as readily? 204
TF17557.indb 204
16/9/08 15:47:28
NA R R AT ION
The friends of ubiquitous, implicit narrators/presenters with respect to fictional movies maintain as their starting point that if we get visual information about the fictional world from the movie, rationality compels us to ask how we get it (Levinson 1996). The implicit fictional narrator is their answer. But why do we stop there in our quest to learn about the provenance of this information? Obviously, postulating the agency of an implicit narrator may lead us to ask what appear to be silly and irrelevant questions – such as, how does the implicit fictional narrator know that x when it is given by said narrator himself that no one within the boundaries of the fiction knows that x? For the sake of forestalling such questions, the defenders of the implicit narrators declare that our questioning about the way in which we learn about the fictional world should stop as soon as we surmise that the fictional narrator has informed us that x. Yet isn’t stopping just here arbitrary? Why has reason suddenly become so easy going? If we really felt driven to learn how we get the relevant information, won’t we want an account of how the implicit narrator gathered it, especially where it conflicts with the assumption of the realistic heuristic? But perhaps if we all agree that the pursuit of such questions is silly, then the best way to stop these questions before they start is by refraining from postulating the existence of an implicit fictional narrator/presenter, since once we get rid of him, these absurdities will disappear. See also Interpretation (Chapter 15), David Bordwell (Chapter 29), and Noël Carroll (Chapter 31).
References Carroll, N. (2006) “Film Narrative/Narration: Introduction,” in N. Carroll and J. Choi (eds.) Philosophy of Film and Motion Pictures, Malden, MA: Blackwell. Chatman, S. (1990) Coming to Terms, Ithaca, NY: Cornell University Press. Currie, G. (1995) Image and Mind, Cambridge and New York: Cambridge University Press. Gaut, B. (2004) “The Philosophy of the Movies: Cinematic Narration,” in P. Kivy (ed.) The Blackwell Guide to Aesthetics, Malden, MA: Blackwell. Kania, A. (2005) “Against the Ubiquity of Fictional Narrators,” Journal of Aesthetics and Art Criticism 63: 47–54. Levinson, J. (1996) “Film Music and Narrative Agency,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Wilson, G. (2006) “Le Grand Imagier Steps Out: On the Primitive Basis of Film Narration,” in N. Carroll and J. Choi (eds.) The Philosophy of Film and Motion Picture, Malden, MA: Blackwell. —— (2007) “Elusive Narrators in Fiction and Film,” Philosophical Studies 135: 73–88.
Further reading The following are further reading on this topic: Peter Alward, “Leave Me Out of It: De Re, But Not De Se: Imaginative Engagement with Fiction,” Journal of Aesthetics and Art Criticism 64(4) (2006): 451–60; David Bordwell, Narration in the Fiction Film (Madison: University of Wisconsin Press, 1985), and “Three Dimensions of Film Narrative,” in Poetics of Cinema (London and New York: Routledge, 2008), 85–134; Paisley Livingston, “Narrative,” in B. Gaut and D. Lopes (eds.) The Routledge Companion to
205
TF17557.indb 205
16/9/08 15:47:28
NOËL CA R ROLL
Aesthetics (London and New York: Routledge, 2001), 359–70; Kendall Walton, Mimesis as Make-Believe (Cambridge, MA: Harvard University Press, 1990); George Wilson, Narration in Light (Baltimore, MD: Johns Hopkins University Press, 1986), “Narrative,” in J. Levinson (ed.) The Oxford Handbook of Aesthetics (Oxford and New York: Oxford University Press, 2003), and “Transparency and Twist in Narrative Fiction Film,” Journal of Aesthetics and Art Criticism 64 (2006): 81–96.
206
TF17557.indb 206
16/9/08 15:47:28
19
N ARRATI V E C LOSUR E Noël Carroll If by narrative we mean the mere recounting or representation of any event and/ or state of affairs through some interval of time, then the overwhelming numbers of motion picture shots, save for freeze frames, are narratives, albeit in many, many cases, minimal ones. However, in most usage, the concept of narrative is reserved for representations of states of affairs and events of a greater degree of complexity and structure. Prototypically, most narratives involve at least two, but generally more, events and/or states of affairs which are related or arranged temporally and causally (where the causation in question may include mental states such as desires, intentions, and motives). “Charlemagne was crowned Holy Roman Emperor in 800 and, in 1945, Japan surrendered unconditionally to the Allies” is not a narrative in the aforementioned sense, although “After being tested in New Mexico, atomic bombs were dropped on Hiroshima and Nagasaki” is, since it comprises three events of which the first temporally preceded and was causally related to the second two. Most motion pictures – including most fictions, nonfictions, and even some abstractions – are narrative in the sense that they involve a number of events and states of affairs standing in temporal and causal relations to each other (Carroll 2001, 2007b). Narratives may be very broadly sorted into two major categories: episodic narratives and unified narratives. Episodic narratives are composed of a string of smaller narratives or episodes, frequently connected by the recurring presence of a central character, but without strong causal linkage between one story and the next. In the ancient world, epic poems recalling the adventures of Hercules would be an example of this. Likewise, the Odyssey is another example of the episodic narrative. It is made up of individual stories of the wanderings of Ulysses, but one of his adventures – say his encounter with the Cyclops – does not figure causally in bringing about his encounter with Circe, or the Sirens, or Scylla and Charybdis. The movie serial Rocketman and the television series Superman are also episodic narratives, as are most sitcoms. So are soap operas; however, instead of rotating around one or a few central figures, their casts usually involve a larger revolving gallery of recurring characters. A unified narrative, as Aristotle put it, has a beginning, a middle, and an end (Aristotle 1996). Although undoubtedly this sounds spectacularly uninformative – can’t anything that is extended through space and/or time be subdivided in this way?
TF17557.indb 207
16/9/08 15:47:28
NOËL CA R ROLL
– these terms should not be understood in their bland, ordinary sense. Rather they are technical terms. By “the end,” Aristotle means an event which secures the feeling of closure in an audience – the almost palpable sensation that the story has finished-up at exactly the right spot. Nothing that needed to be told has been left untold, nor has the story gone on superfluously. At most we learn that the prince and the princess lived happily ever after, but we are not informed of their subsequent adventures. That would be the stuff of other, further stories. The beginning, in turn, is just what the audience needs to know in order to start to follow the story. We are introduced to a locale and a set of characters. We learn of their desires, aspirations, relations, struggles, and so forth. Perhaps we learn about certain conflicts between them, as when we witness the enmity between Achilles and Agamemnon at the opening of Wolfgang Petersen’s 2004 Troy. The beginning typically supplies us with the background that we need to comprehend in order to track with understanding that which happens next. Often the narrative of a motion picture begins with an establishing shot that depicts a state of affairs and tells us where and when the story will unfold. Then we meet a number of characters. Some change intrudes which calls for action, or reaction. Helen absconds with Paris to Troy. This complication generates a response – alliances are forged between various Greek forces in preparation for war. And with this complication, we enter the middle of the narrative, which concludes or ends when the problem that ensues from, or the questions (Who will win? Who will die?) that are provoked by the complications – in this case the Trojan War – are resolved. A narrative of this sort is unified in as much as each part leads smoothly to the next. Given the situation and the people who inhabit it, the kind of change or complication introduced causes certain problems to be raised or questions to be asked; then these queries will gradually be further complicated and eventually answered by the action of the characters, and, precisely when all the presiding questions are settled, the story proper ends (though there may also be a brief coda). Such a narrative is unified insofar as it appears rhetorically to be held together tightly by the logic of questions and answers. Of course, often in Hollywood movies, there is more than one story – there is frequently a problem to be solved (in Casino Royale [2006] James Bond has to entrap the international arms-merchant/gambler), while there is also a romantic subplot about whether 007 and his beautiful co-spy will become a couple (Bordwell et al. 1985). Nevertheless, these two stories usually get intertwined so that the romance often contributes to the solution of the problem. Indeed, as many as 60 percent of classical Hollywood films have been said to exhibit this structure (Bordwell et al. 1985). Aristotle’s theory of the unified narrative has influenced the way in which screenwriters construct narratives. Syd Field (1994) wrote an influential guide to popular screenwriting, in which he divided screenplays into three parts or acts: a beginning (often the first 25 percent of the motion picture), a confrontation (frequently, 50 percent of the movie), and a resolution (25 percent or whatever is left over after the first two acts), with turning points in-between. Paul Joseph Gulino (2004) thinks that 208
TF17557.indb 208
16/9/08 15:47:29
NA R R AT I V E C LO SU R E
the popularity of Field’s book was a major influence on the notion of the Hollywood film possessing a “three-act structure,” although the Danish director Urban Gad recommended that filmmakers construct their works in acts as early as in 1912 (Bordwell 2008). Others have refined the three-part structure, not in a way that refutes Aristotle, but which instead limns with greater clarity the grounds for the interlocking coherence in the kind of story that Aristotle had in mind. For example, Kristin Thompson (1999) sees four acts, where Field says there are three. Hers include the setup, the complication, development, and climax; whereas David Bordwell (1985) describes the classical Hollywood plot as a six-part structure: introduction of setting and characters, explanation of state of affairs, complicating action, ensuing events, outcome, and ending. But despite the varying numbers, it seems fair to suggest that all these authors belong to the school of Aristotle; they do not contradict his Poetics as much as they make more explicit the interconnected elements that render such stories so unified. Motion pictures articulated in the Aristotelian tradition not only are structurally unified: they feel unified or of a piece. That is, as mentioned earlier, when “The End” flashes on screen or, as is becoming more routine nowadays, when the credits start crawling up from the bottom edge of the image, the audience has a strong sense of finality and completeness. Unlike the episodic narrative, which, in principle, seems like it could keep going on and on, the unified narrative appears as though it has to end just where it does, rather like the sonnet that feels as though it concludes on exactly the right word, or the song that closes on just the correct note. Whereas the episodic narrative gives the appearance of an assortment of adventures, of which some might be subtracted with no irreparable loss to the feel of the saga, the unified narrative leaves us with the impression of one indissolubly integrated whole from which nothing can be left out, save at the cost of a tangible experience of perturbation. That is, once again, this sort of narrative engenders a powerful feeling of closure, the sense that every salient element that has been set in motion in the story has been, as they say, wrapped up. This is not to say that closure is altogether an alien factor in episodic narratives. Within episodes in an episodic narrative, closure may obtain. In each episode of the television series, Have Gun Will Travel, for instance, Paladin’s assignment for the week was successfully discharged. Of course, in some episodic narratives, like motion picture serials of the sort referred to as cliff-hangers, the closure of one adventure might be deferred to the beginning of the next installment to ensure that the audience returns in the subsequent week to discover the way the heroine escaped her fate worse than death. But, at the same time, very complicated episodic narratives with large storylines, like soap operas, over time become too intricate to wrap up. If they go off the air, they may have no other option but to leave viewers hanging, as the TV series The Sopranos did with stylish élan at the conclusion of its final season. In one way or another, closure plays a role in our experience of the vast number of the motion picture narratives we encounter, including episodic and unified narratives. As we shall see, closure is not a universal feature of narrative motion pictures, and 209
TF17557.indb 209
16/9/08 15:47:29
NOËL CA R ROLL
moviemakers may have important reasons to foreswear closure. Yet, it is reasonable to say that most motion picture narratives aspire to instill closure in spectators. As already indicated, we signal our feelings of closure by saying of a movie that “it appears to wrap things up” or “it seems like it tied up all the loose ends.” But can we be less metaphorical than this? Can we explain the narrative structures in the movie that account for these impressions? In this regard, one important suggestion comes from David Hume’s essay “Of Tragedy” (1993). Hume writes: Had you any intention to move a person extremely by the narration of any event, the best method of increasing its effect would be artfully to delay informing him of it, and first excite his curiosity and impatience before you let him into the secret. (Hume 1998) That is, Hume recommends that the way in which a narrative can take hold of an audience’s attention is to engender their curiosity about what happens next. He calls that which they want to know “a secret.” Similarly, in his discussion of what he labels “the hermeneutic code,” Roland Barthes dubs it “an enigma” (Barthes 1974: 18–20). For our purposes, we may adopt less inflated terminology and simply say that what Hume has in mind is that a story sustains our attention, often irresistibly, by presenting us with questions that we want answered – questions that the narrative implicitly promises to answer and which we expect will be answered. Hume, of course, was offering advice to playwrights; but it has been accepted even more eagerly by screenwriters. What does this have to do with bringing about the impression of closure? Closure will obtain when all of the saliently posed questions that the narrative has served up are answered. For instance, recall this archetypal plot: boy meets girl and they are attracted to each other; enter some oily Lothario bent on seducing the girl – will the boy be able to unmask his rival and regain the girl’s affections? This is what the audience wants to know. And, finally, the boy gets the girl. The End. Closure. The movie does not go on to tell us that the couple then bought car insurance, because that was not part of the story. That is, whether or not the couple bought an insurance policy is not a question about which the movie has encouraged us to be curious. Were this episode to be added to the movie, the feeling of closure would be diluted; the movie would not strike us as having ended at the right spot. It would seem to ramble. It would have gone on beyond the point where closure would obtain. However, if after establishing the existence of a happy family, their beloved infant is abducted, the audience will want to learn whether the child will be rescued. The complications that unfold contribute to sustaining or answering this question. The movie, then, is over when that question is answered one way or another. Suppose, as is usually the case, we learn that the infant has been saved. The impression of closure will correspond with that revelation. In ordinary suspense movies, we will feel frustration rather than closure if that answer is not forthcoming. And if the movie goes on to show us the child getting a flu shot, that will feel like a narrative non 210
TF17557.indb 210
16/9/08 15:47:29
NA R R AT I V E C LO SU R E
sequitur, since whether or not the baby needed a flu shot is not part of the story – not one of the forcefully advanced questions that have come to preoccupy our attention as spectators. That is, were the scene with the flu shot added, rather than things being wrapped up, a loose end would unravel. It is to be hoped that this discussion of closure in the unified narrative suggests that a major source of the sense of the completeness and coherence evident in this species of narrative – to which, arguably, most motion pictures belong – is what we can call erotetic. “Erotetic logic” is the logic of questions and answers. By extension, there is also erotetic narration. This is narration that proceeds by generating questions that the narration then goes on to answer. Closure obtains when all of the pronounced questions the movie has elected to put emphatically before us have been answered. This hypothesis can be confirmed by turning off the projector as the last reel of the movie is about to wind onto the screen. Irritated, the audience will jump up and demand to know, for instance, whether the baby was rescued. They want their closure. Erotetic narration not only imbues a motion picture with an aura of completeness and completion. The evolving network of questions and answers also holds the story together – renders it coherent – throughout the unified narrative. That is, scenes and sequences are connected to other scenes and sequences by a skein of questions and answers. The motion picture standardly begins by answering the kind of questions we automatically ask whenever we are introduced to a novel situation. We want implicitly to know where the action is set and when, who these people are, what do they want, and why are they acting like this. The opening of the typical motion picture will answer these basic questions at least to the extent that we have enough information to understand the further questions that the subsequent changes in the initial states of affairs and their accompanying complications elicit. Some scenes and sequences simply raise questions to be answered down the line by other scenes and sequences. In early American cinema, two-shot films, involving stories of kidnappings, comprised an opening shot in which a child was kidnapped, thereby raising the question of whether he or she could be recovered; and then the second shot, which delivered the answer, as the child was rescued from the clutches of some stereotypically swarthy Eastern or Southern European immigrant. In other, more complex motion pictures, scenes and sequences may function to prolong, as Hume advised, the delivery of the answers to our questions. This may be the result of a subsequent scene only partially answering an ongoing question. For example, once we learn the child has been kidnapped, part of what we are apt to want to know, if we haven’t witnessed the abduction, is “by whom?” This may be partially answered by learning that the kidnapper is a woman with a limp, but exactly which woman with a limp remains a live question. Likewise, a later scene may sustain an earlier question by keeping our questions aloft. For instance, the escaped convict eludes apprehension in one scene, thereby iterating the question of whether he will be caught in the next scene or a subsequent one. Or, the fearless vampire killers close in on Dracula’s lair, but he transforms himself 211
TF17557.indb 211
16/9/08 15:47:29
NOËL CA R ROLL
into a bat and flies away, leaving us wondering if they will be able to stake him another day. Sometimes when scenes answer one of our questions, they will replace one question with another. When King Kong is subdued on Skull Island, the inquiring minds in the audience want to know what Carl Denham intends to do with him. And, of course, new characters, forces, and situations can be established at any point in the diegesis, bringing in their tow new questions to be answered, as they interact with what has already been given about the story world up to that point. Scenes and sequences in erotetic narratives function to raise and/or answer questions, to answer questions partially, to cause questions to be iterated, and to answer some questions in a way that opens onto others. Because scenes and sequences are bound together by this network of questions and answers, unified narratives give the appearance of coherence – everything seems to belong or to fit, specifically to fit into the network of questions and answers. The questions and answers that hold together the typical motion picture narrative come in hierarchically differentiated orders of magnitude. For convenience sake, we can make the following rough distinctions. First, there are presiding macroquestions. These include the question that dominates the motion picture globally from one end to the other. Will the boy be able to get the girl? Can Goldfinger have his way with Fort Knox? Will the village survive in Seven Samurai (1954)? Of course, a film may have more than one presiding macroquestion. Buster Keaton’s 1927 The General has several, including the following: Will Johnny be able to enlist in the Southern army? Will Johnny be able to recover his locomotive, The General? Will he be able to rescue his love, Annabelle Lee? Will he be able to alert the Confederates of the encroaching Union attack in time? These are the interlocking, indeed, in this case, piggy-backed, macroquestions that keep us riveted to the story in the expectation that they will be answered. When they are answered, The General is effectively finished. We do not ask what Johnny will do after the Civil War, because that is not one of the presiding macroquestions that we have been invited to entertain. When all of the presiding macroquestions in a unified movie narrative are answered, closure is usually secured and we feel that Keaton’s movie has been completed and is complete (Carroll 2007a). However, motion pictures are not merely unified by overarching or presiding macroquestions. There are also more localized questions that call forth or propone answers of a more limited scope. In The General, in one scene, the Union hijackers pile debris on the rail tracks in order to derail Johnny, who is in hot pursuit. These activities raise the question of whether or not Johnny will be derailed in a subsequent scene or sequence. And, of course, subsequent scenes or sequences answer such questions. Johnny doesn’t get derailed. We may call these local erotetic networks microquestions and microanswers. They generally provide the glue that holds the trail of scenes and sequences together on a local basis, while not letting go of our attention. Moreover, these microquestions and answers are typically hierarchically subordinated to the presiding macroquestions that animate the narrative. For example, the question/answer network involving Johnny’s 212
TF17557.indb 212
16/9/08 15:47:29
NA R R AT I V E C LO SU R E
possible derailment in The General provides information in the direction of answering the presiding macroquestions of whether he will save his engine, his love, and the Confederacy and ultimately whether he will win his uniform and enlist. Finally, it should be added that there are erotetic structures that are neither presiding macroquestion/answer networks nor micro ones. These are question/answer complexes that span large parts of a motion picture, but not the entirety of the work. For example, the question of what is going to happen to Marion in Alfred Hitchcock’s 1960 Psycho dominates the opening of the film but is resolved once Norman Bates kills her. The macroquestions about her fate organize a large part of the film, but not the whole of it, since once she is eviscerated in the shower sequence, questions about Marion’s future are replaced by questions about whether her murder and her murderer will be discovered. Since the question about Marion is sustained over many microquestions/answers – like, will the inquisitive-looking cop further investigate her? – it is a macrostructure, but since it is not sustained across the entire film, it is not a presiding macrostructure. It does not provoke closure. Answering presiding macroquestions is what educes closure. Although I have spoken of the narrative organization of motion pictures in terms of its erotetic rhetorical address, narrative exposition is also importantly concerned with the temporal order of the depicted events. Events in films occur before, after, and simultaneously with each other. Nor need the procession of events in a film be that of a progressive linear movement. There can be flashbacks and flashforwards. It is a fact about narration that the time of the telling or recounting of the tale (sometimes called the discourse, or in a related vein, syuzhet) need not follow the order of events as they occurred in the tale world (sometimes called the story or fabula) (Chatman 1978; Bordwell 2008). This is as true of movies as it is of other kinds of narratives. Although the story is told in a progressive fashion, moving forward in unbroken real time from the opening credits to the end credits, the events that comprise the tale in the story world need not be projected consecutively in our world, the world of the audience, outside of the fiction. Mildred Pierce (1945) begins in the present of the story world, with a man being shot, and then moves backward in the time of the fiction – that is, flashes back – to answer the questions of who shot the man and why. Likewise, Easy Rider (1969) flashes forward to an image of a burning motorcycle in order to prompt the question of who is going to die. Or, if there is a flashforward to a shot of Petulia throwing something through a shop window, it primes us to ask “why?” That is, even though the temporal order of the narration is a level of plot structure distinguishable from its erotetic organization, it is not altogether unrelated, since typically the manipulation of time by the discourse of the narration will be in the service of supplying the kinds of questions and answers that contribute to narrative closure. Similarly, breaks out of the forward flow of the narrative and into the temporally suspended fantasies of characters – as in Billy Liar (1963), Lord Jim (1965), and The Pawnbroker (1964) – also serve to answer narrative questions, often concerning the desires of the characters, and thusly they can serve the designs of narrative closure by revealing the deepest wishes of these protagonists in a way that bids us to wonder whether they will be realized or not. 213
TF17557.indb 213
16/9/08 15:47:29
NOËL CA R ROLL
If so far I may have inadvertently suggested that all motion pictures that depict a series of events in time involve closure, then I have misled you, and not only because some motion pictures that strive for closure botch the job. For there are also things like home movies that show the major events in the life of a family – birthdays, vacations, weddings, etc. – in the temporal order in which they occurred. Yet, these – and comparably arranged, chronologically ordered event series – are best thought of as not yet fully prototypical narratives, properly so-called, but only chronicles, and are for that reason bereft of closure (Carroll 2001). And, in addition, even some fullfledged narratives, like the individual installments of cliff-hangers such as The Perils of Pauline, avoid closure at the end of each episode in order to drum up future business. Moreover, there are unified narratives that eschew closure at the level of the depiction of the events and actions of the story world in order to achieve unity at a higher level or plane of organization or signification. For example, at the end of his 1950 masterpiece Rashomon, Akira Kurosawa refrains from announcing, from an omniscient viewpoint, what really happened when the bandit, the husband, and the wife met in the forest. Our questions about what actually went down in the story world go unanswered, but in a way that prompts us to ask about Kurosawa’s artistic motivation or intention in withholding this information from us. That is, the transparently purposive avoidance of closure at the level of the action of the story disposes us to attempt to interpret the significance of this decision on Kurosawa’s part. It encourages what we might name “interpretive ascent.” What is Kurosawa trying to communicate by this obvious subversion of the erotetic model? And once having ascended to that level of interpretation, thoughtful viewers regularly grasped that the point or theme of Rashomon is to suggest a form of epistemological perspectivism – which discovery provokes a kind of interpretive, as opposed to narrative, closure. Likewise in L’Avventura (1960), Michelangelo Antonioni seems simply to drop the question of why the character played by Lea Messari has disappeared, in order to tantalize the viewer into inferring an other-than-narrative significance to the story – such as the insinuation of the existential meaninglessness of contemporary life, an acknowledgment, that is, of the possibility that life does not come – as unified movies do – with answers clearly stamped upon the face of it. There are, of course, other objectives that may lead motion picture makers to forgo closure at the level of the story. It may be done to enhance the expressive unity of a film. That we never learn whether Irena in The Curse of the Cat People (1944) is a ghost or a psychological projection contributes to the eerie, ambiguous, uncanny feeling the film emits. Furthermore, realistic films like Amarcord (1973) often loosen the erotetic grip of the narrative structure in order to appear more “lifelike” – to contrive a rhythm more akin to the everyday flow of often directionless events, as opposed to the propulsive forward pace of erotetic narration. Furthermore, reflexive films, such as Last Year at Marienbad (1961), pose questions they refuse to answer, in order to induce the viewer to reflect apperceptively upon the way in which this strategy discloses the degree to which we normally expect answers to the questions that narratives stimulate in us. Ordinarily those questions are tacit, despite 214
TF17557.indb 214
16/9/08 15:47:29
NA R R AT I V E C LO SU R E
the fact that they structure the way in which we follow movies with understanding. Modernist exercises like Last Year at Marienbad, in contrast, bring this tacit process of question formation to the surface, where the ambitious viewer may use it to flash a searchlight upon her own default motion-picture-processing dispositions and procedures. Thus, not all motion picture makers intend closure, nor do they employ erotetic narration to its fullest natural advantage. Nevertheless a great many do – indeed, most do – including documentaries and narrative television. This, furthermore, is a major factor contributing to the surpassing clarity that most motion pictures possess, especially in contrast to our far more desultory and diffuse mundane lives. In this way, erotetic narration, and the closure it brings to the unified narrative, is a crucial element in accounting for the power of movies. As neurobiologists have begun to prove, the brain has a preference for clarity – for sharp distinctions such as black versus white and good versus evil. So in this regard, the kind of clarity associated with erotetic narratives – which represent the most common type of motion picture – are just the sort of brain food we crave. Moreover, even motion pictures that defer the full effect of erotetic narration by forfeiting closure, nevertheless also frequently exploit it indirectly, since averting closure, where that evasion appears intentional, may inspire the viewer to ask “why?” And this may lead to an enhanced appreciation of the motion picture in terms of, among other things, its expressive and/or aesthetic qualities and/or its thematic and/ or interpretive significance. Although not all narratives are motion pictures and not all motion pictures are narrative, narration and the motion picture have come together so often that theirs would appear to be at least a common-law marriage, and therefore no philosophy of the motion pictures can be truly complete without a consideration of cinematic narrative and its most habitual, recurring structures, such as erotetic narration and closure. See also David Bordwell (Chapter 29), Noël Carroll (Chapter 31), and Narration (Chapter 18).
References Aristotle (1996) Poetics, trans. M. Heath, London and New York: Penguin Books. Barthes, R. (1974) S/Z, trans. R. Howard, New York: Hill and Wang. Bordwell, D. (1985) Narration in the Fiction Film, Madison: University of Wisconsin Press. —— (2008) “Three Dimensions of Narrative Film,” in Poetics of Cinema, London: Routledge. Bordwell, D., Staiger, J., and Thompson, K. (1985) The Classical Hollywood Cinema: Film Style and Mode of Production to 1960, New York: Columbia University Press. Carroll, N. (2001) “The Narrative Connection,” in Beyond Aesthetics, Cambridge and New York: Cambridge University Press. —— (2007a) Comedy Incarnate: Buster Keaton, Physical Humor and Bodily Coping, Malden, MA: Blackwell Publishers. —— (2007b) “Narrative Closure,” Philosophical Studies 135 (August): 1–15. Chatman, S. (1978) Story and Discourse: Narrative Structure in Fiction and Film, Ithaca, NY: Cornell University Press.
215
TF17557.indb 215
16/9/08 15:47:29
NOËL CA R ROLL
Field, S. (1994) Screenplay: The Foundations of Screenwriting, New York: Dell. Gulino, P. J. (2004) Screenwriting: The Sequence Approach, New York and London: Continuum. Hume, D. (1998) “Of Tragedy,” in David Hume: Selected Essays, Oxford: Oxford University Press. Thompson, K. (1999) Storytelling in the New Hollywood: Understanding Classical Narrative Technique, Cambridge, MA: Harvard University Press.
Further reading The following are further reading on this topic: David Bordwell, The Way Hollywood Tells It: Story and Style in Modern Movies (Berkeley: University of California Press, 2006); Noël Carroll, “Toward a Theory of Film Suspense,” in Theorizing the Moving Image (Cambridge and New York: Cambridge University Press, 1996), and The Philosophy of Motion Pictures (Malden, MA: Blackwell Publishers, 2008); Paisley Livingston, “Narrative,” in B. Gaut and D. Lopes (eds.) The Routledge Companion to Aesthetics (London and New York: Routledge, 2001); and Kristin Thompson, Storytelling in Film and Television (Cambridge, MA: Harvard University Press, 2003).
216
TF17557.indb 216
16/9/08 15:47:29
20
ONTO LOGY David Davies What kind of thing is a film? What kind of thing is a photograph? Philosophers who have pondered such matters have generally been moved by the following sorts of considerations. To appreciate an artwork or cultural artifact requires at least an experiential engagement of some kind with an instance of that work or artifact, whereby some or all of the properties bearing upon its appreciation are made manifest to the receiver. In the case of a painting, what is required is an experiential encounter with a particular physical object which is located at any given time in a particular gallery or museum, and this makes it plausible to identify the work itself with that object. In the case of a film or photograph, however, there seem to be many different locations where, at a given time, we might experientially encounter the work or artifact in the manner necessary for its appreciation. You may be watching Citizen Kane, or looking at a photograph by Diane Arbus, in Los Angeles at the same moment that I am watching the same film or viewing the same photograph in London. In this respect, films and photographs seem to resemble musical and literary works. Appreciating a musical work requires hearing it performed, and appreciating a literary work requires reading it, but people in different locations engaging with different objects or events can simultaneously have the necessary experiential encounter with a given musical or literary work. Films and photographs, then, like literary works and works of music, must be the kinds of things that can have multiple instances, and such things are most naturally thought of as types or kinds. Just as the word “dog” is a word type that can have three instances, or tokens, on a given page, and just as the cocker spaniel is a kind of dog, of which my neighbor may have an instance, so, it is assumed, Citizen Kane (1941) is a type or kind whose tokens or instances are the entities with which viewers experientially engage in appreciating the film. If so, then our ontological understanding of film requires that we further clarify what more general type or kind Citizen Kane belongs to. If the cocker spaniel is a type of dog, what type of thing is a film like Citizen Kane? Answering the latter question should help us answer a further ontological question about films: how are films individuated, that is, when do we have two instances of the same film, and when do we have instances of two different films? To ask “of what are films or photographs types or kinds?” is to ask about the distinctive features of their instances. Perhaps the most natural answer to this question
TF17557.indb 217
16/9/08 15:47:29
DAV ID DAV I E S
is that they are types or kinds of images. Instances of photographs and films are still or moving images, and photographs and films themselves are types of still or moving images, it might be said. But this answer presupposes that we understand what, ontologically speaking, an image is, and also, perhaps, what is distinctive of the kinds of still images that are photographs and the kinds of moving images that are films. How do photographs, as images, differ from paintings, for example? Some have argued that the difference between paintings and photographs is not merely one of degree, but rather one of a more fundamental nature. Understanding the ontology of film, then, requires that we address at least the following questions: (1) How is the multiple nature of films and photographs to be understood? (2) What kind of thing are instances of films and photographs? And (3) if filmic and photographic instances are images, do they differ in any fundamental way from the images we encounter in plastic arts, such as painting?
Film and photography as multiple art forms We can think of an instance of a work of art or other artifact as something that has a distinctive place in the proper appreciation of that work or artifact in that it makes directly available to receivers manifest qualities that bear essentially on such appreciation. Then, as noted above, it is natural to distinguish between “singular” and “multiple” art forms (Wollheim 1980; Wolterstorff 1993 [1975]; Carroll 1998; S. Davies 2003). In the former, a work can, as a matter of necessity, have only one instance. Painting and carved sculpture are generally viewed as singular art forms in this sense. Some kinds of photographic processes also seem to result in singular works or artifacts – for example, the processes that produce daguerreotypes and Polaroids (Carroll 1998: 215). But photography and film of the more standard kinds are multiple art forms, as noted above, as are classical music, narrative and dramatic literature, silkscreening, and cast sculpture. There are nonetheless significant differences between multiple art forms, and, of more significance in the present context, between film and photography, on the one hand, and music and literature, on the other. There are at least three ways in which multiple artworks can be brought into existence and made available to members of an artistic community (Wollheim 1980: 78–80; Wolterstorff 1980: 90ff.; S. Davies 2003: 159–63). First, as in the case of a literary work, such as a novel, an instance of the work an artist brings into existence may serve as an exemplar. Further instances are then generated through attempts to emulate the exemplar in those respects determined by relevant artistic conventions in place. Second, as in the case of cast sculpture, an artifact an artist produces may, when employed in prescribed ways, generate instances of the work. Such an artifact may be termed a “production artifact” (Wolterstorff 1980), an “encoding” (S. Davies 2003), or a “template” (Carroll 1998). As in the case of exemplars, conventions in place in the relevant art form determine how this artifact must be used if a properly formed instance is to be produced. Third, as in the case of classical musical works and dramatic works, the instructions an artist provides may, if properly followed by those aware of the relevant conventions and practices, result 218
TF17557.indb 218
16/9/08 15:47:29
ON TOLO GY
in well-formed instances of the work. In such cases, the instructions call for interpretation and the instances of the work count as performances of it. Photographs and films clearly fall into the second of these categories. In the case of an analog photograph, the production artifact or encoding is the negative. In the case of a digital photograph, it is a computer file. In both cases, correct instances are images generated from the production artifact by appropriate means. Films, on the other hand, have as their production artifacts those entities used, in various cinematic media, to generate screenings of those films – film prints, videotapes, digital files, etc., which either are, or stand in a “copy” relation to, master encodings of the film. But what is the ontological status of an artwork in a multiple art form? Or, more generally, what sorts of entities can have other entities as instances? Instantiation is merely one way in which an entity can be “generic” in the sense that it has elements falling under it (Wollheim 1980: 75ff.). The members of a set, for example, can be seen as elements “falling under” the set, but this seems an unpromising way of thinking about the relationship between multiple artworks or artifacts and their instances. Sets are individuated in terms of their members and any difference in membership entails a difference in set, but a given photograph by Cartier-Bresson could surely have had more or fewer instances than it actually has. These reflections suggest that artworks in multiple art forms are types and their instances are tokens of those types. The distinction between types and tokens, first drawn by C. S. Peirce, is familiar in everyday contexts where we distinguish between, say, the letters in the alphabet and different occurrences of those letters. If asked how many letters there are in the word “sheep,” we can correctly answer both five (there are five occurrences of letter-tokens in the word) and four (there are occurrences of four different types of letters). The identity and nature of a type is not changed as the number of its tokens changes. However, if multiple artworks and artifacts stand to their instances in the relationship of types to tokens, they are types of a special sort, for they are generally taken to admit of both correct and incorrect instances. For example, instances of a musical work arguably include not only performances that meet all the requirements for right performance of the work, but also performances containing at least some incorrect notes. Similarly, it would seem, a damaged print of Renoir’s Rules of the Game (1939) can still provide an audience with an instance of the film, albeit an improperly formed one. Types or kinds that admit in this way of correct and incorrect instances can be termed “norm kinds” (Wolterstorff 1980, 1993 [1975]) or “norm types” (Dodd 2007). Whereas descriptive types are individuated by the condition that must be met by their tokens, norm types are individuated by the condition that must be met by their correct or properly formed tokens. While many find attractive the idea that multiple artworks and artifacts are norm types and that their instances are well-formed or improperly formed tokens of those types, this view is not without its problems. Types, we have said, are generic entities that can have elements that “fall under” them, but what is the ontological status of such things? On pretty much every conception of such matters, types are abstract rather than concrete entities and do not have determinate spatial locations. It is therefore questionable whether they can enter into causal relationships with things 219
TF17557.indb 219
16/9/08 15:47:29
DAV ID DAV I E S
that do. This raises concerns about our epistemic access to, and ability to refer to, types if we take knowledge and reference to require some kind of causal engagement with the entity known or referred to. However, it has been argued by some philosophers either that abstract objects can enter into causal relations (Burgess and Rosen 1997: 24ff.; Dodd 2007: 13–14) or that knowledge and reference do not require such relations with their objects (e.g., Brown 1991). A further problem arises if we accept a particular account of the metaphysics of types (Dodd 2007: chapters 2–3), according to which types have a number of features that fit uneasily with our intuitions about films and photographs. Types, as we have seen, are individuated according to the conditions that must be satisfied by their (well-formed) instances. These conditions can be characterized as properties associated with the type – the property of being correctly instanced by strings of the form s-h-e-e-p associated with the word type “sheep,” for example (Wolterstorff 1980; Dodd 2007). But, it can be argued, types exist just in case their property associates exist, and properties cannot be brought into or go out of existence – they exist eternally if they exist at all. If so, then it follows that types, too, cannot be brought into or go out of existence. So, if multiple artworks are types, it seems that they cannot be created by their artists, but only discovered. Furthermore, it can be argued, types, as abstract entities individuated by reference to their associated properties are modally inflexible – which is to say that they could not have had intrinsic (i.e., nonrelational) properties other than the ones that they actually possess. But we seem to think of multiple artworks as entities that could have differed in certain of their constitutive properties and still have been the same works. We certainly have no problem understanding such counterfactual claims about films – for example, the claim that American Beauty (1999) would have been a better film if the original ending had been retained rather than replaced as a result of feedback from focus groups. If films are types and if types are understood in the way just described, then we must take such a claim as pertaining not to American Beauty but to a cinematic counterpart that closely resembles American Beauty. For some (Dodd 2007: chapters 4–5), these implications of identifying multiple artworks with types simply show that we must revise our intuitions about artworks (including films) in light of deeper metaphysical reflection. This line of thought has some plausibility in the case of musical works, where the idea that composers “creatively” discover possible sound sequences is perhaps, on reflection, no more absurd than the idea that mathematicians “creatively” discover novel proofs of theorems. But it seems very counterintuitive to think in this way about films, given both the very complex and collaborative nature of the filmmaking process and the fact that, as seen above, instances of films result from an extended process of manipulating physical media, resulting in the generation of the production artifact used to generate such instances. Some who wish to defend the idea of creatable multiple musical works have appealed to the normative element in such artworks if we construe them as norm types. While types of sound sequence may themselves exist eternally, musical creation, it can be argued, is a matter of making a particular type of sound sequence normative for a work, and this brings into being something that did not exist prior to the composer’s activity. Musical works can then be thought of as “indicated 220
TF17557.indb 220
16/9/08 15:47:29
ON TOLO GY
sound sequences” created by their composers (Levinson 1980; for related proposals, see Wolterstorff 1993 [1975], and [in a more skeptical vein] S. Davies 2003: 170). But it can be responded that, if works are taken to be norm types, they will still be individuated in terms of a condition for correct instantiation that can be expressed as a property: being correct in successfully complying with the performance specifications set out by B in context C, for example. Then it can be argued that this property, like every property, must exist eternally if at all, so that we have not shown multiple works, as norm types, to be creatable or modally flexible (see Dodd 2007: chapter 5). Furthermore, the Levinsonian line of argument seems especially unattractive if applied to films and photographs, for it preserves the counterintuitive idea that the audio-visual structure of a film is itself discovered and becomes art through being made normative for the film. Similar difficulties beset another strategy to which we might appeal in defense of the creatability of photographic and cinematic works in the face of arguments that appeal to the eternal existence of types. Some philosophers have argued that, in order to accommodate the contextually situated nature of artworks, we should view those works not merely as contextualized objects – indicated structures, for example – but as contextualized actions, indicatings of structures or generations of artistic vehicles. Works, then, are action types (Currie 1989) or token performances (D. Davies 2004). At least in the latter case, we can preserve our sense that photographs and films, and indeed all other multiple artworks, are brought into existence by the artist, since the act of generating the artistic vehicle – identified with the artwork – is itself performed by the artist. Whatever the merits of such an account, however, it fails to address the current concern, since it leaves unchallenged the idea that the artistic vehicle is itself discovered, which seems implausible for film. A more radical response is to take the creatability and modal flexibility of multiple artworks such as photographs as given, and argue that, if types are indeed eternally existent and modally inflexible entities, this just shows that such multiple artworks cannot in fact be types (Rohrbaugh 2003). The challenge then is to say how an artwork can have multiple instances if it isn’t a type. Rohrbaugh argues that talk of multiple artworks as types is harmless if it is merely part of a semantic account of the terms used to label such works, since all we are doing in saying that Citizen Kane is a type is distinguishing one use of that label from another use which refers to tokens of the type. Rohrbaugh rejects, however, the further contention that multiple artworks are types in the metaphysically freighted sense outlined above. Rather, he suggests, a multiple artwork like Steiglitz’ The Steerage is a temporal and modally flexible entity that is both created by Steiglitz and may later be destroyed. Such historically situated and extended entities can be termed “continuants.” A continuant is a higher order entity whose existence depends upon those lower order concreta that are its “embodiments.” In the case of a traditional photograph, for example, the embodiments include both the negative and existing prints taken from the negative. While the idea that multiple artworks are continuants whose instances figure among their embodiments preserves some of our intuitions about the creatability and possibly also the modal flexibility of films and photographs, by contrast the nature of 221
TF17557.indb 221
16/9/08 15:47:29
DAV ID DAV I E S
continuants and the relationship they bear to those instances among their “embodiments” is far from clear (Dodd 2007: chapter 6). As a result, the ontological status of films and photographs, as multiple artworks, remains contested. Perhaps, however, we can establish some ground rules for resolving this issue. First, as a number of writers have stressed (Currie 1989: 11–12; S. Davies 2003: 155; Rohrbaugh 2003: 178–9; D. Davies 2004: 16–24), artworks of various kinds are things that have their place in certain human practices, and that have to be understood in terms of the ways in which they are treated in those practices. It is thus ill-advised to try to impose upon our understanding of artworks an a priori ontological framework. Artworks must be things of a kind that can function in the ways artworks function in our practices, and the fact, if fact it be, that we treat photographs, films, and other multiple artworks as creatable, modally flexible, entities gives us the best possible reason to seek a way of characterizing them, ontologically speaking, that explains how they can have such properties. Thus we must countenance, and even anticipate, that an ontology of art adequate to accommodate multiple artworks like films and photographs will be revisionary of our traditional ontological categories. So, even if Rohrbaugh’s “continuants” are found to be unsatisfactory, we have reason to seek another way of reconciling the multiple nature of films and photographs with our sense that such artifacts are created and perhaps also modally flexible, if, as has been argued, no such reconciliation is possible if such entities are classified as norm types.
Filmic and photographic instances Whichever kind of account we favor of the general ontological category to which multiple artworks like films and photographs belong, we must address a further question as to the nature of those instances of films and photographs an engagement with which is necessary for their proper appreciation. Answering this question will also clarify the manner in which films and photographs are individuated, that is, the conditions under which we have two instances of the same film or photograph rather than instances of two different films or photographs. In the case of another multiple art form, classical music, there is general agreement that an instance of a work is a realization, standardly in a performance, of a type of sound sequence or sound structure. Where there is disagreement over (a) whether realizations other than performances count as instances of a work; and (b) whether it is necessary, for a performance, that the sound sequence be realized on particular instruments and/or be consciously guided by a set of instructions for performance specified by a particular individual or in a particular musico-historical context. In the case of films and photographs, we can expect a similar pattern of agreement and disagreement among theorists, although the literature here is much less developed than in the musical case. Take first the case of photographs. A well-formed instance of Stiegltiz’s The Steerage is presumably what one gets if one processes the negative – the production artifact – in a manner determined by the relevant artistic practice. What one gets, in this case, is a print of the photograph which presents a particular image to the viewer. The print itself is a physical object, and the image is a visible aspect of 222
TF17557.indb 222
16/9/08 15:47:29
ON TOLO GY
that object. Given that what bears upon appreciating the photographic work is our engagement with the image, it might seem preferable to think of the image, rather than the object of which it is an aspect, as the instance of the work. Comparison with other artistic images, such as paintings, might suggest otherwise, however. For we standardly identify an instance of a painting with the physical object having as an aspect a particular visible array of pigment on canvas. But in the photographic case, the fact that photographs can be instanced not only in prints but also in screenings of slides and in digitally generated arrays on computer screens may favor a uniform treatment that takes the image, generated from a production artifact in an appropriate way, to be the instance. In any case, it seems open to someone to argue, as in the musical case, either that any realization of the visual design of the image is an instance of the photographic work, or that only those realizations generated by standard processing of a production artifact with a particular history so count. In fact, there seems to be general agreement here on the latter course of action. This may be because our sense of the well-formedness or ill-formedness of a photographic image is tied to its having been generated in this way, as in the case of other multiple artworks whose instances seem tied to a historically situated production artifact, such as works of cast sculpture or engraving. In the case of film, instances are what can be seen (and, in sound films, heard) by appropriately qualified receivers at right screenings (Wolterstorff 1980: 924; Carroll 1998: 212; S. Davies 2003: 161). A right screening is one that uses an authorized production artifact and relevant decoding apparatus in the intended manner, where both the artifact and the apparatus are themselves in good working order. This may involve projection or computer realization, but nothing parallels the presentation of a photographic image through a print. There is therefore no reason to resist identifying an instance of a work with a sequence of images, usually synchronized with a sequence of diegetic and nondiegetic sounds. An instance of a film, then, is a series of two-dimensional moving images and, where relevant, correlated sounds. If, as Wolterstorff proposes, we think of a film as a norm type, the properties that are normative in the work are just those that are manifest to receivers in all right screenings as characterized above. Analogous notions of instance correctness are forthcoming on alternative construals of the ontological status of the cinematic work. If instances of films and photographs are still and moving images respectively, what are the latter? The most obvious answer is that they are representations of whatever is thereby imaged, analogous in this respect to paintings and other representational images, but differing in the manner whereby the representation is produced. A number of writers on film and photography, however, have claimed that differences in the manner of production of photographic and pictorial images have more far-reaching implications as to the epistemological, ontological, and artistic status of those images. Key to these claims is the supposedly “mechanical” manner in which a traditional photographic or cinematic image is produced. What is imaged in such cases, it is claimed, is whatever played the appropriate causal role in a “mechanical” process leading to those features of the production artifact responsible for producing a given 223
TF17557.indb 223
16/9/08 15:47:29
DAV ID DAV I E S
visible manifold in instances of the photograph or film. In the case of paintings, however, the intentions and beliefs of the painter also play a role in determining what is imaged. Some have argued that, in virtue of this difference in the manner of generating the image, photographic and cinematic images are not representations (Walton 1984), or are not representations in which we can take an artistic interest (Scruton 1983). The most radical of these claims, at least on standard readings (Carroll 1988: 125ff.; Currie 1995: 48–50), is to be found in André Bazin’s “The Ontology of the Photographic Image.” Pictorial images represent through depiction. While the nature of pictorial depiction is a matter of dispute, it is generally taken to involve either some kind of intended resemblance between image and subject or our capacity to experience the image in certain ways that resemble how we would experience the things represented. Bazin, however, takes the relation between a photographic image and what it is an image of to be a matter of identity of some sort, rather than of intended or experienced resemblance. Contrasting pictorial and photographic images, Bazin writes that “the photographic image is the object itself . . . It shares, by virtue of the very process of its becoming, the being of the model of which it is the reproduction: it is the model.” Furthermore, this holds, according to Bazin, “no matter how fuzzy, distorted, or discoloured . . . the image may be” (1967: 14). In virtue of this, traditional film for Bazin is by its very nature a “realistic” medium. Given that Bazin’s concern is to explain the distinctive psychological effects of photographs, the startling pronouncement that the photographic image is the imaged object might be seen as a claim about how photographs strike us, phenomenologically, rather than about their ontological status (Friday 2005). However, an ontological reading of Bazin’s claim can be defended by reference to a further analogy that he draws between photographs and death masks. Just as the latter are moldings formed by “a certain automatic process,” so “one might consider photography in this sense as a molding, the taking of an impression, by the manipulation of light” (1967: 12n). A death mask faithfully preserves, in virtue of its manner of production, the shape of its model, independently of the beliefs of the maker. In respect of its shape, then, it is identical with its model. Similarly, the process productive of a photographic or filmic image is “mechanical” or “automatic” in that it doesn’t involve the subjectivity of the maker (ibid.: 12–13), and, as a result, re-presents, rather than merely representing, something of its model. The problem with this claim, as critics have pointed out (e.g., Carroll 1998: 125ff.), is that the death mask analogy is difficult to sustain. Arnheim (1957), for example, argues at length that a photograph cannot re-present in some unique way the pattern of light rays emitted by an imaged object, but only preserve some of the visible information available. Which information is preserved, and how it is encoded, depends upon both the nature of the photographic equipment employed and choices that the photographer must make. But then we are left with the claim that light rays from imaged objects play some causal role in producing photographs – which, even if it might bear upon the psychological impact of photographs, merely reiterates what is distinctive about the process whereby photographic images are generated. 224
TF17557.indb 224
16/9/08 15:47:30
ON TOLO GY
Kendall Walton, however, has argued that Bazin is right in thinking that there is an essentially “realistic” character to photographic images that is lacking in paintings. According to Walton (1984), the essential realism of photographs lies in the fact that photography is not simply a new way of producing pictures, but a new way of seeing. Photographic and filmic images of the traditional sort are transparent, in that we see the world through them. The camera, then, is properly analogized to glasses, mirrors, and telescopes, which are also media for indirect seeing, rather than modes of representation. Paintings, by contrast, are not transparent – in looking at a painting of a subject X, what we see is a representation of X, not X. Walton’s argument presupposes a causal theory of seeing according to which to see something is to have visual experiences caused, in a certain manner, by what is seen. What is distinctive of seeing X is that the manner of causation is mechanical, or nonintentional, whereas in the case of seeing a representation of X our relation to X is mediated by the intentional states of X’s producer. He claims that the manner in which a photograph causes my visual experience is also mechanical or nonintentional. A further necessary condition for visual perception, he maintains, is that, where we obtain information about X via a perceptual medium, our susceptibility to make mistakes reflects similarities among things of type X. Critics have disputed Walton’s account of perception, however, and thus his claim that photographic and filmic images are not representations. On the proposed account, a mode of access to information about X counts as perceptual just in case it is nonintentionally causally – and hence counterfactually – dependent upon X, and it preserves real similarity relations. It has been argued that these conditions are neither necessary nor sufficient for perception (e.g., Currie 1995: 61–9). A further necessary condition, it is argued, is that seeing, direct or indirect, furnishes us with “egocentric information” as to how X is related to us spatially and temporally, and “transtemporal” information that permits us to track X through time. In looking at a photo of X, however, I get neither egocentric nor transtemporal information. With moving images, I get transtemporal information, but not egocentric information. If so, this supports the idea that photographic and filmic images are indeed representations, but differ from paintings in that their content is determined naturally rather than, in part, by the intentional states of the imagemaker. See also Definition of “cinema” (Chapter 5), Medium (Chapter 16), Rudolph Arnheim (Chapter 27), and Realism (Chapter 22).
References Arnheim, R. (1957) Film as Art, Berkeley: University of California Press. Bazin, A. (1967) “The Ontology of the Photographic Image,” in H. Gray (trans.) What Is Cinema? vol. 1, Berkeley: University of California Press. Brown, J. R. (1991) The Laboratory of the Mind: Thought Experiments in the Natural Sciences, London: Routledge. Burgess, J., and Rosen, G. (1997) A Subject with No Object, Oxford: Clarendon Press. Carroll, N. (1988) Philosophical Problems of Classical Film Theory, Princeton, NJ: Princeton University Press.
225
TF17557.indb 225
16/9/08 15:47:30
DAV ID DAV I E S
—— (1998) A Philosophy of Mass Art, Oxford: Clarendon Press, chapter 3. Currie, G. (1989) An Ontology of Art, New York: St Martin’s Press. —— (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge: Cambridge University Press. Davies, D. (2004) Art as Performance, Oxford: Blackwell. Davies, S. (2003) “Ontology of Art,” in J. Levinson (ed.) The Oxford Handbook of Aesthetics, Oxford: Oxford University Press. Dodd, J. (2007) Works of Music, Oxford: Oxford University Press. Friday, J. (2005) “André Bazin’s Ontology of Photographic and Film Imagery,” Journal of Aesthetics and Art Criticism 63: 339–50. Levinson, J. (1980) “What a Musical Work Is,” Journal of Philosophy 77: 5–28. Rohrbaugh, G. (2003) “Artworks as Historical Individuals,” European Journal of Philosophy 11: 177–205. Scruton, R. (1983) “Photography and Representation,” in The Aesthetic Understanding, London: Methuen. Walton, K. (1984) “Transparent Pictures,” Critical Inquiry 11: 246–77. Wollheim, R. (1980) Art and Its Objects, 2nd ed., Cambridge: Cambridge University Press. Wolterstorff, N. (1993 [1975]) “Towards an Ontology of Art Works,” in J. W. Bender and H. G. Blocker (eds.) Contemporary Philosophy of Art, Englewood Cliffs, NJ: Prentice Hall. —— (1980) Works and Worlds of Art, Oxford: Clarendon Press.
226
TF17557.indb 226
16/9/08 15:47:30
21
RA C E Dan Flory Ideas of race have shaped film since the medium’s invention during the late nineteenth century. Some of the first works produced by Thomas Edison, Edwin S. Porter, and Georges Méliès contain demeaning portrayals of African Americans devouring watermelons, stealing chickens, or menacing innocent whites. Other early works show allegedly primitive “natives” living as “natural savages” in exotic settings, far from a presumed white viewership. One of the Edison Company’s earliest major productions was Uncle Tom’s Cabin in 1903, the first of many efforts to mount on film Harriet Beecher Stowe’s flawed narrative concerning the injustices of racialized slavery. Perhaps not too surprisingly, prevailing conceptions of white supremacy, racial hierarchy, manifest destiny, ongoing contemporaneous projects of “race science,” and the American theatrical tradition of blackface minstrelsy fundamentally influenced the depictions of race in these early films. However, similar conceptions or their successors also permeate many later cinematic landmarks, among them Birth of a Nation (1915), Nanook of the North (1922), The Jazz Singer (1927), King Kong (1933), Gone With the Wind (1939), The Searchers (1956), Lawrence of Arabia (1962), and Chinatown (1974). While numerous filmmakers and social critics have examined the impact of race on film, philosophers have been somewhat slower to do so. The reason why is not especially difficult to identify: philosophers have typically thought of race as an empirical problem, resulting from the improper application of moral or political ideals, rather than a conceptual difficulty. Problems of race have not seemed central to the discipline but rather peripheral matters at best, to be dealt with after the theoretical complexities have been overcome. Recent work in the philosophy of race has shown, however, that this standard philosophical stance is nowhere near as clear-cut as many in the field have believed. As Charles Mills has argued, in the modern West “[r]acism and racially structured discrimination have not been deviations from the norm; they have been the norm.” Moreover, philosophers such as Locke, Kant, and Hegel played fundamental roles in theoretically shaping modern ideas of race (Mills 1997: 93, 63–73, 94). The attitude of considering ideas of race and racism as mere unfortunate consequences accruing from misapplications of moral or political concepts has been further thrown into question by the erosion of a sharp distinction between the empirical and the conceptual that has taken place in philosophy during
TF17557.indb 227
16/9/08 15:47:30
DA N F LORY
the last half-century. However, some philosophers have taken the role of race seriously in cinematic imagery, narrative, and audience response, despite the more typical inclination to dismiss race as a nonphilosophical concern. It is also crucial to make clear that while ideas of race themselves have undergone sea changes with regard to their standard meanings during the brief history of film, older, more rigid conceptions remain deeply influential. For example, even though superannuated research projects in biology, psychology, or other scientific disciplines no longer provide support for a hierarchical concept of race based in nature, outmoded senses of this idea as an essence-like “natural kind” continue to structure the social reality of many people’s lives. Moreover, by virtue of being embedded in traditional social practices and ways of thinking, general forms of dealing with human life through the lens of permanent, unchanging, and inherited conceptions of graduated human difference have carried over into many audience members’ viewing habits, making the function of such notions an issue for anyone interested in the role that standard presumptions and expectations play in our understanding of film – particularly their social, moral, political, and affective dimensions, but also at the level of cognition. Lastly, I should note that although critics like James Baldwin (1976) and James Snead (1994) have analyzed race’s role in film with striking perceptiveness, their discussions have typically been aimed as social criticisms, rather than as philosophical ones, by which I mean their analyses have been specific assessments of existing societies or cultural products and not meant as abstract, conceptual points. These discussions, in other words, ordinarily limit their scope to particular flaws in actual social arrangements or films. By contrast, philosophical criticisms aim more at the level of general truths; ideally, such criticisms would be universal, but in any event they would not be so closely tied to historical specifics as those usually offered by such critics. Of course, oftentimes these categories of criticism overlap, meaning that some forms of social criticism may also be philosophical ones, and vice versa, but the two categories remain distinguishable. My reason for noting this distinction is that it is crucial to grasp the differences between these forms of criticism, as well as their commonalities, when discussing race’s operation in film. However, even as such discussions are best understood as not developing philosophical considerations of film per se, they nonetheless frequently contain insights that others have elaborated in ways more directly pertaining to philosophy.
Resisting standard forms of spectatorship One way to begin understanding race’s role in the philosophy of film is by means of the concept of the “oppositional gaze.” The idea of “the gaze” itself is, of course, critically important to film studies, perhaps most famously through Laura Mulvey’s “Visual Pleasure and Narrative Cinema” (1975). The concept, however, has roots in the work of Jean-Paul Sartre (1956) and Maurice Merleau-Ponty (1968). The relevant sense here denotes the way in which human beings often look at others and, rather than acknowledging their full humanity, impose oppressive misconceptions on them. As formulated by bell hooks, the idea of the oppositional gaze harks back to Fanon’s 228
TF17557.indb 228
16/9/08 15:47:30
R AC E
Sartrean-influenced discussion of the “white gaze” and his attempts to defend himself against it (1967: 109ff.). Frequently such attempts at self-defense force one to learn “to look in a certain way so as to resist” this commonplace mode of perceiving human beings and thereby develop a critical perspective that interrogates not only human relations, but visual media as well (hooks 1992: 116–17). Crucially positioned against more determinist theories positing a passive cinematic spectator (such as Mulvey’s), the oppositional gaze argues for the possibility of an active agency that permits viewers to critically analyze representations of human beings, particularly with regard to race. hooks’ concept is thus linked to the idea of the “resisting spectator” such that viewers may oppose racialized depictions that dehumanize and degrade through stereotype (Diawara 1988), as well as the idea that we possess the ability to “read through” filmic texts for alternative, nonracialized meanings (Bobo 1993). However, hooks goes to greater lengths than these critics to indicate that such stances amount to arguing for active critical human capacities that need to be developed, rather than being the essential characteristics of, say, black spectators, who, some have argued, possess such abilities merely because they are oppressed human beings. Whereas conditions of oppression would mean that one needed such capacities much more urgently, their actual development remains an independent, self-determined step that, in terms of general psychological capacities possessed by human beings, anyone could take. Tommy Lott has further elaborated the importance of such resistant perspectives for philosophy of film by explaining how they not only permit one to better understand the concept of “black film” as more heterogeneous than often argued, but also enable us to better grasp the complicated interplay between aesthetics and politics in cinematic representations of racial blackness, such as the ways in which they often speak in complicated, “hybrid” voices that combine both mainstream and independent sensibilities (Lott 1991, 1997, 1998). Crucial to note here is what has been called the “politics of representation,” the contingent, historically determined social meanings attached to racialized images in art forms like film, and how they may influence or reinforce ways viewers think about race without their even realizing it. According to this conception, stereotypes may play themselves out in cinematic narrative or characterization and sway us at a nonconscious level to think, act, or perceive according to rigidified hierarchical rankings of humanity. But these influences may be counteracted, particularly if we realize we may also develop the capacity to think critically and oppose them, even if developing that potential remains a daunting responsibility. Similar ideas have been discussed, although less fully, by Stuart Hall (1989). With their more generalized discussions of such matters, hooks and Lott have provided conceptual tools with which we might grasp not only the pernicious nature of racialized images in film, as well as the presumptions that give them currency, but also how to go about theorizing resistance to them.
Cavellian and cognitivist positions Another place where serious discussion of race’s role in film gains philosophical leverage is through Cavell’s telling observation of more than three decades ago that: 229
TF17557.indb 229
16/9/08 15:47:30
DA N F LORY
Until recently, types of black human beings were not created in film; black people were stereotypes – mammies, shiftless servants, loyal retainers, entertainers. We were not given, and were not in a position to be given, individualities that projected particular ways of inhabiting a social role; we recognized only the role. Occasionally the humanity behind the role would manifest itself; and the result was a revelation not of a human individuality, but of an entire realm of humanity becoming visible. (1979a: 33–4) Through this and associated remarks Cavell argues that race’s function in film is directly linked to the epistemological role of acknowledging the existence of others. Elsewhere Cavell asserts that one’s full-fledged humanity depends fundamentally on its recognition and acknowledgment by other human beings (1979b). When these capacities are transferred over to film viewing, they become critically important to our understanding of certain types of cinematic characterizations, or “individualities,” as Cavell refers to them – the kinds of characters that certain people are, such that we could imagine ourselves as having met or meeting them in other circumstances (1979a: 33, 35). Rather than thinking merely in terms of stereotypes, we may also consider other human beings as “well-rounded” characters like ourselves – for example, as fullfledged equals – provided that we are prepared to extend complete recognition and acknowledgment to their humanity. Cavell thus construes the problem of race in film as analogous to the problem of “other minds” and the attendant skepticisms we might entertain with regard to their existence, while at the same time evoking allusions to Sartre’s, hooks’, and Lott’s discussions of various types of gazes. Although it would no doubt be appropriate to be slightly more reserved about the relative enlightenment of post-Civil Rights era spectators than Cavell is, he opens here a serious possibility for philosophically considering race in film. Cavell’s point about race is also an integral part of his overall argument about film’s appeal. For him, its capacity to present individualities makes the medium attractive and aesthetically interesting in ways that, say, theater is not. Moreover, acknowledgment is “an act of the self” (1979a: 123), which underscores his efforts to integrate our general cognitive responses to cinematic characters with those we have to real human beings. A number of philosophers have followed Cavell’s lead here, including William Rothman, Thomas Wartenberg, and myself. Rothman analyzes specific representations of race in the “unknown woman” genre and other films, noting particularly some of these films’ socially critical dimensions that are at the same time philosophical (2004: 98–109). Similarly taking seriously Cavell’s conception of film as philosophy, Wartenberg outlines a genre he calls “the unlikely couple film,” in which two individuals “transgress a social norm regulating appropriate partnering choice” (1999: xvi). These films may thereby express subversive, philosophical criticism of social hierarchies such as race by raising “questions that go to the very heart of how we imagine a life worth living” (xviii). “Interracial unlikely couple films are noteworthy for their assumption that racial hierarchy is illegitimate” as well as their focus on its eradication (238). Like Cavell, both these philosophers place reflective theorizing 230
TF17557.indb 230
16/9/08 15:47:30
R AC E
about race within particular shots or the narrative content of specific films, as well as in the discussions they generate in viewers. Applying Cavellian methodology in combination with more cognitivist film theory, I have explored some of the asymmetries between audiences, based on their understandings of race. Divergent responses to racial representations, I argue, are significantly due to differing presumptions about the concept and their function in determining what one’s aesthetic reaction will be (Flory 2005, 2006). Understanding such differences allows us to develop greater insights into audience response, concepts of human difference, and “critical cinema” – that is, films that raise explicit criticisms about society – as well as some of the ways in which such criticism may cross over into the philosophical. In particular, I have analyzed at length the use of film noir to influence audience members to reflect on racial injustice. By exploiting this film form’s potential for making viewers think, a number of filmmakers have found ways to urge us to contemplate such matters as the gap between professed endorsements of full universal equality and moral actions that betray otherwise (Flory 2000, 2002, 2007, 2008). By linking our frequently unexamined background assumptions with their cognitive operation in film viewing, I bring to the surface often conflicting attitudes toward race and their effects on our lives as well as on aesthetic experience. Noël Carroll’s work provides a more squarely cognitivist response to cinematic presentations of race. He points, for example, to the frequent associations between racial whiteness, beauty, and morality, on the one hand, and racial “otherness,” ugliness, and immorality, on the other, in cinematic uses of horror and humor (2000). These associations have frequently operated to reinforce racial stereotyping of groups that would otherwise be insupportable. In his analysis Carroll explicitly notes that he is analyzing the politics of representation (38). Similarly, he has examined films like Nothing But a Man and The Cool World for ways in which social contradictions may be made salient to viewers (1998: 203–13), as well as investigating some of the nuances regarding how moving images may or may not influence us with respect to race (2003: 114–20). Like hooks and Lott, the philosophers of film described in this section presume an active spectatorship capable of critically regarding the cinematic representations they see. This presumption opens the possibility that viewers do not necessarily have to passively think or perceive according to racist preconceptions but may vigorously resist them. Clearly, such a presumption would be a crucial element of any philosophically respectable strategy for overcoming problems of race, whether in film or elsewhere. In addition, it is worth noting that the critical positions outlined above broadly conform to the guidelines of a “naturalized” epistemology – the theory that our knowledge of the world is most advantageously conceived as consistent with the best results that the sciences have to offer, such as Joshua Glasgow’s recent assessment that ordinary conceptions of race have no basis in recent findings in biology (2003).
231
TF17557.indb 231
16/9/08 15:47:30
DA N F LORY
Psychoanalytic, Marxist, and existentialist perspectives Other philosophical perspectives have contributed to a better understanding of race in film as well. Although some of these analyses are at times hampered by presuming overly deterministic conceptions of passive spectatorship, they frequently remain helpful investigations of standard racial preconceptions. For example, Kelly Oliver and Benigno Trigo argue that “both the narrative and style of [film noir] are motivated by anxieties over race and sex” (2003: xviii). Using the psychoanalytic concepts of condensation, displacement, and “abjection” (a term for that which calls into question or threatens human identity) to interpret film noir as “a type of Freudian dreamwork” (xv), Oliver and Trigo excavate noir’s “unconscious” for fears of “the other” and their determinations that shape such films. “Implicitly,” they tell us, film noir “is always and everywhere about race . . . racial ambiguity is [its] real anxiety” (4–5). Yet by essentializing this film form as they do, Oliver and Trigo preclude the possibility of resisting its characteristics. Spectatorship is presumed to be passive and without the capacity to critically analyze the racialized dimensions it “always and everywhere” contains. Thus, while their individual analyses of films often contain valuable insights, their overall philosophical perspective excludes the possibilities of transformation that they themselves advocate (e.g., 188, 234–6). Even though these authors seek to avoid having “totalized” their theory, that is precisely what they do. By contrast, Homi Bhabha’s argument that racial stereotypes in film operate as Freudian fetishes largely escapes this criticism by introducing a sense of critical agency (1994: 66–84). Although weakened by its reliance on the nineteenth-century understanding of mind contained in typical Freudian perspectives, Bhabha’s position accommodates the need to have the capacity for critically assessing race’s role in film. Working from a more Marxist stance, Douglas Kellner has used class-sensitive notions such as ideology and alienation to investigate how films may not only instill dominant myths and values, but also question them. Sensitive to the ways in which economic relations affect our sense of identity, both coauthors Michael Ryan and Kellner note that “cultural representations [such as those in film] not only give shape to psychological dispositions, they also play a role in determining how social reality will be constructed” (Ryan and Kellner 1988: 13). Thus control over the production of such representations importantly determines various forms of social power. In focusing on race, they praise works of black radicalism like Sweet Sweetback’s Baadasssss Song (1971) and The Spook Who Sat by the Door (1973) for depicting structural problems of racial oppression (32–3). By the same token, they largely condemn other “blaxploitation”-era films because, despite frequent positive images, these works focused mainly on race as a problem to do with an individual’s particular beliefs, rather than with structures of power and representation (122–6). In turning to an analysis of Spike Lee’s work, Kellner employs Brechtian conceptions to argue that films like Do the Right Thing (1989) and Malcolm X (1992) are “morality tales rather than political learning plays in Brecht’s sense” (1997: 96). While admittedly getting audiences to focus on some forms of structural oppression, Kellner argues that Lee’s work ultimately fails to achieve its aim of making such matters clear by falling victim to “identity 232
TF17557.indb 232
16/9/08 15:47:30
R AC E
politics” and the idea that racial beliefs are predominantly an individual’s responsibility (99–100). I have contested this interpretation of Lee’s work on the grounds that it underappreciates his aim to make viewers think, rather than didactically presenting them with an ideological fait accompli (Flory 2006, 2008). Nevertheless, through his use of Marxist theory, Kellner brings to the surface important issues in analyzing philosophical dimensions of race in film. In connection with the consideration of other philosophical perspectives, worth noting is a controversy surrounding a more recent remark by Cavell about the role of race in film. While analyzing an early sequence in 1953’s The Band Wagon, he asserts that actor Fred Astaire’s “acknowledging his indebtedness for his existence as a dancer – his deepest identity – to the genius of black dancing” is evident (1997: 35). This claim generated a spirited response from Robert Gooding-Williams, as well as a rejoinder by Cavell. Despite the latter’s immediate qualification that “[h]ow fully such an acknowledgment is acceptable is a further question,” Gooding-Williams argues that numerous racial codings and references in the film reveal a “Jim Crow” sensibility that Cavell’s remark overlooks (Gooding-Williams 2006: 54ff.). By closely analyzing the frequent appearances of black waiters and Pullman porters, as well as numerous other references to blackness in The Band Wagon, Gooding-Williams reveals a racialization of the melancholy that Astaire’s character embodies in these early scenes. Even as the film admittedly acknowledges its debt to black dance (61–2), it also presumes racial blackness as a kind of “urbane, sophisticated . . . primitivism” that holds the potential to rejuvenate melancholic white men such as Astaire plays through imparting them with a nonthreatening “blackened masculinity” and “swing” (151n49, 59, 151n53). As in his analyses of Casablanca (1942) and The Lion King (1994) (17–41), Gooding-Williams takes pains to make clear that problems of racial representation remain unresolved in Astaire’s dancing routines. Working from a broadly existentialist perspective, Gooding-Williams concludes that, rather than feeling pleasure in viewing Astaire’s routines in the early sequences of The Band Wagon, as Cavell does, he feels “dismay” – something that he argues others should feel as well, once they put aside their racialized viewing presumptions, because these presumptions remain fundamentally unquestioned in the film’s narrative (GoodingWilliams 2006: 64). In response, Cavell has argued that within the confines of mid-twentieth-century America, Astaire went as far as he could in acknowledging his debt to black dance. Invoking political philosopher John Rawls’ conception of an imperfectly just society that is nevertheless “good enough” to receive one’s assent (1971), Cavell argues that, analogously, Astaire’s dance is “good enough to warrant praise” (2005: 82), even if, from an ideal perspective, it remains tainted by the racial injustices of its era. The dance, then, operates partly as a homage to African American dance, while recognizing the circumstances of injustice that prevent Astaire from dancing as a strict equal for more than a moment with his partner in the second of the sequences, a black shoeshine man played by Leroy Daniels. Thus, asserts Cavell, the dance is ambiguous, even as it works through the circumstances of its own ambiguity, in particular the circumstances of injustice that prevent full equality between Astaire’s and Daniels’ 233
TF17557.indb 233
16/9/08 15:47:30
DA N F LORY
characters (76–9). Still, the dance as “homage has presented itself as a step of change” (79), and with his outstretched arms to the shoeshine man during the finale Astaire declares “his willingness to change” (81), even if far more remains to be done and lone individuals such as Astaire represents could never accomplish these changes on their own. Yet this dancer’s beckoning gesture holds out the as-yet-unfulfilled promise of equality (107–8), even as such a “claim is open to rebuke” and Astaire’s gesture raises further questions of culture that remain difficult to resolve (82). Although I do not propose to settle this controversy here, one thing its extended discussion reveals is the importance of being careful to determine who “we” are when analyzing “our” feelings and what “we” should judge. While Cavell seems accurate to point out the near-impossibility of Astaire expressing his desire for full equality within the confines of his circumstances and thus the difficulty of making the judgment Cavell proposes we make about it, I think Gooding-Williams is right to argue that his opponent’s presumption about the viewing audience is that they are the “privileged ones,” as Cavell describes “us” at one point (ibid.: 78), rather than referencing cinematic spectators in general. A crucial responsibility for philosophers of film that Gooding-Williams emphasizes in this disagreement, then, is the need to thoroughly inspect the foundations of one’s viewing position, particularly for presumptions of racial privilege.
Rethinking the aesthetic A suggestion made by Clyde Taylor directly relates to such a responsibility, namely, his challenge to philosophical theorists of film to rethink its aesthetics (1988, 1989). Synthesizing ideas from many of the positions described above as well as from the “Third Cinema” movement, “black aesthetics,” and elsewhere, Taylor argues that standard Western conceptions of the beautiful and taste must be put aside and our sensibilities reconceptualized in order for us to escape the defects of racialized thinking. Clearly Taylor’s challenge reaches far beyond philosophy of film into aesthetics itself, for he proposes a comprehensive program of research that breaks the prevailing “aesthetic contract,” interrogates current systems of creating as well as discussing art, and elaborates ways in which “art and beauty can be approached with more evenhandedness and less false consciousness” (1998: xiii). Also a project advocated by Sylvia Wynter (1992), such an overhaul of the aesthetic itself emphasizes an antiessentialist, pluralistic stance, the importance of attention to historical contingency, and possesses a striking consistency with naturalized epistemology. As Taylor notes, of special interest is “[t]he interplay between [Western aesthetics] and the representations of people of color in cinema,” particularly “[t]he entrenched precept of aesthetic reasoning that white is equitable with beauty and black with ugliness” (2003: 400), a point that underscores one of Carroll’s charges regarding the racialized politics of representation (2000). In this context, one may see the relevance of Richard Dyer’s White (1997) and Ella Shohat and Robert Stam’s Unthinking Eurocentrism (1994). These works seek to rethink the aesthetics of Western visual media in ways that leave aside harmful conceptions 234
TF17557.indb 234
16/9/08 15:47:30
R AC E
of race and reimagine them from a perspective of full universal humanity. Although both works are somewhat marred by a tendency to essentialize their philosophical opposition and therefore miss subtleties in many conceptions of Western aesthetics as well as art objects, Dyer more successfully meets Taylor’s challenge, perhaps because his goals are more modest and localized, and his method empirically based. This more “piecemeal” approach to theorizing philosophically about race in film has been further carried forward by many whose positions are described above. This project of reimagining cinematic aesthetics so as to understand, as well as eliminate, the negative influences of modern racialized thinking will no doubt continue to be a focus for philosophers of film for some time to come. See also Stanley Cavell (Chapter 32), Noël Carroll (Chapter 31), Cognitive theory (Chapter 33), Film as philosophy (Chapter 50), and Spectatorship (Chapter 23).
References Baldwin, J. (1976) The Devil Finds Work, New York: Dial Press. Bhabha, H. (1994) The Location of Culture, London and New York: Routledge. Bobo, J. (1993) “ ‘Reading through the Text’: Black Women Spectators,” in M. Diawara (ed.) Black American Cinema, New York: Routledge. Carroll, N. (1998) Interpreting the Moving Image, Cambridge and New York: Cambridge University Press. —— (2000) “Ethnicity, Race, and Monstrosity: The Rhetorics of Horror and Humor,” in P. Brand (ed.) Beauty Matters, Bloomington: Indiana University Press. —— (2003) Engaging the Moving Image, New Haven, CT: Yale University Press. Cavell, S. (1979a) The World Viewed: Reflections on the Ontology of Film, exp. ed., Cambridge, MA: Harvard University Press. —— (1979b) The Claim of Reason: Wittgenstein, Skepticism, Morality, and Tragedy, Oxford: Clarendon Press; New York: Oxford University Press. —— (1997) “Something Out of the Ordinary”, Proceedings and Addresses of the American Philosophical Association 71, no. 2: 23–37. —— (2005) Philosophy the Day after Tomorrow, Cambridge, MA: Belknap Press of Harvard University Press. Diawara, M. (1988) “Black Spectatorship: Problems of Identification and Resistance,” Screen 29: 66–76. Dyer, R. (1997) White, London and New York: Routledge. Fanon, F. (1967) Black Skins, White Masks, trans. C. L. Markmann, New York: Grove Press. Flory, D. (2000) “Black on White: Film Noir and the Epistemology of Race in Recent African American Cinema,” Journal of Social Philosophy 31: 82–116. —— (2002) “The Epistemology of Race and Black American Film Noir: Spike Lee’s Summer of Sam as Lynching Parable,” in K. Stoehr (ed.) Film and Knowledge: Essays on the Integration of Images and Ideas, Jefferson, NC: McFarland. —— (2005) “Race, Rationality, and Melodrama: Aesthetic Response and the Case of Oscar Micheaux,” Journal of Aesthetics and Art Criticism 63: 327–38. —— (2006) “Spike Lee and the Sympathetic Racist,” Journal of Aesthetics and Art Criticism 64: 67–79. —— (2007) “Race, Empathy, and Noir in Deep Cover,” Film and Philosophy 11: 67–85. —— (2008) Philosophy, Black Film, Film Noir, University Park, PA: Pennsylvania State University Press. Glasgow, J. (2003) “On the New Biology of Race,” Journal of Philosophy 100: 456–74. Gooding-Williams, R. (2006) Look, A Negro! Philosophical Essays on Race, Culture, and Politics, New York: Routledge. Hall, S. (1989) “Cultural Identity and Cinematic Representation,” Framework 36: 68–1. hooks, b. (1992) Black Looks: Race and Representation, Boston: South End Press.
235
TF17557.indb 235
16/9/08 15:47:30
DA N F LORY
Kellner, D. (1997) “Aesthetic, Ethics, and Politics in the Films of Spike Lee,” in M. Reid (ed.) Spike Lee’s “Do the Right Thing,” Cambridge and New York: Cambridge University Press. Lott, T. (1991) “A No-Theory Theory of Contemporary Black Cinema,” Black American Literature Forum 25: 221–36. —— (1997) “Aesthetics and Politics in Contemporary Black Film Theory,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Clarendon Press; New York: Oxford University Press. —— (1998) “Hollywood and Independent Black Cinema,” in S. Neale and M. Smith (eds.) Contemporary Hollywood Cinema, London and New York: Routledge. Merleau-Ponty, M. (1968) The Visible and the Invisible, ed. C. Lefort; trans. A. Lingis, Evanston, IL: Northwestern University Press. Mills, C. (1997) The Racial Contract, Ithaca, NY: Cornell University Press. Mulvey, L. (1975) “Visual Pleasure and Narrative Cinema,” Screen 16: 6–18. Oliver, K., and Trigo, B. (2003) Noir Anxiety, Minneapolis: University of Minnesota Press. Rawls, J. (1971) A Theory of Justice, Cambridge, MA: Belknap Press of Harvard University Press. Rothman, W. (2004) The “I” of the Camera: Essays in Film Criticism, History, and Aesthetics, 2nd ed., Cambridge and New York: Cambridge University Press. Ryan, M., and Kellner, D. (1988) Camera Politica: The Politics and Ideology of Contemporary Hollywood Film, Bloomington: Indiana University Press. Sartre, J.-P. (1956) Being and Nothingness, trans. H. Barnes, New York: Philosophical Library. Shohat, E., and Stam, R. (1994) Unthinking Eurocentrism: Multiculturalism and the Media, London and New York: Routledge. Snead, J. (1994) White Screens/Black Images: Hollywood from the Dark Side, ed. C. MacCabe and C. West, New York: Routledge. Taylor, C. (1988) “We Don’t Need Another Hero: Anti-Theses on Aesthetics,” in C. Andrade-Watkins and M. Cham (eds.) Blackframes: Critical Perspectives on Black Independent Cinema, Cambridge, MA: MIT Press. —— (1989) “Black Cinema in the Post-aesthetic Era,” in J. Pines and P. Willemen (eds.) Questions of Third Cinema, London: British Film Institute. —— (1998) The Mask of Art: Breaking the Aesthetic Contract – Film and Literature, Bloomington: Indiana University Press. —— (2003) “Black Cinema and Aesthetics,” in T. Lott and J. Pittman (eds.) A Companion to AfricanAmerican Philosophy, Malden, MA: Blackwell. Wartenberg, T. (1999) Unlikely Couples: Movie Romance as Social Criticism, Boulder, CO: Westview Press. Wynter, S. (1992) “Rethinking ‘Aesthetics’: Notes towards a Deciphering Practice,” in M. Cham (ed.) Ex-Iles: Essays on Caribbean Cinema, Trenton, NJ: Africa World Press.
Further reading The classic scholarly history of blacks in American film to the Second World War is Thomas Cripps, Slow Fade to Black: The Negro in American Film, 1900–1942 (New York: Oxford University Press, 1977). The continuation of Cripps’ history is Making Movies Black: The Hollywood Message Movie from World War II to the Civil Rights Era (New York: Oxford University Press, 1993). Jim Pines and Paul Willemen (eds.) Questions of Third Cinema (London: British Film Institute, 1989) examines film-theoretical considerations of “Third Cinema.” Sylvia Wynter, “Black Aesthetic,” in M. Kelly (ed.) Encyclopedia of Aesthetics, vol. 1 (New York: Oxford University Press, 1998), provides a definition, historical survey, and discussion of the black aesthetic movement.
236
TF17557.indb 236
16/9/08 15:47:30
22
Realism Andrew Kania The term “realism” has been put to almost as many uses in film theory as in philosophy. The basic idea in both areas of study is the same: something is realistic if it bears some sort of veridical relation to reality. Thus, in order to specify a particular sense of “realism” one must specify (i) what is being described as realistic; (ii) what one means by “reality”; and (iii) what relation is being posited between them. In film theory and criticism, one major concern has been with whether particular films, or kinds of film (e.g., film noir, neorealist cinema), veridically represent the true nature of the social or political order, or human nature or consciousness, or interpersonal relations. I largely ignore those questions here. Instead, I address more basic questions about the nature of film in general, and whether it can be said to be a realistic medium at some more fundamental level. I also largely ignore the “classical” film theory of such figures as Arnheim, Bazin, and Panofsky, and contemporary film theory that draws on “continental” theories of psychoanalysis, structuralism, Marxism, and so on. Since these alternative approaches are well served by other chapters in this volume, I focus instead on more recent work in “analytic” philosophy of film and “cognitive” film studies, with their roots in AngloAmerican philosophy of the twentieth century and empirical psychology. (On these distinctions, see Currie 1995: xi–xx; Bordwell and Carroll 1996: xiii–xvii; Bordwell 1996; Carroll 1988a, b, 1996c.) There are a number of distinct ways in which film has been said to be realistic, even at a fundamental level. I address three of these claims: (i) that our experience of motion pictures engenders illusions about the reality of what we are seeing (what I will call “motion picture realism”); (ii) that we literally see the objects captured on film (photographic realism); and (iii) that our experience of film is like our experience of the world (perceptual realism). Finally, I briefly discuss the relation between these metaphysical claims and the aesthetics of film.
Motion picture realism Some theorists argue that film is a realistic medium because it engenders an illusion in us that something is real, when in fact it is not. Gregory Currie has usefully divided such theses into two sorts (1995: 28–30). A cognitive-illusionist theory states that film
TF17557.indb 237
16/9/08 15:47:30
A ND R E W K A NI A
engenders a false belief in us, such as that we are literally seeing the fictional events of a film unfold before us. A perceptual illusionist theory states that there is a difference between how film appears to us and how it really is, independently of our beliefs about it. For instance, film images may seem to move, even if we know that such motion is merely apparent and not real. Cognitive-illusionist theories A number of thoroughgoing cognitive illusions have been attributed to film. These claims are not usually explicitly linked to the illusion of movement (though see Baudry 2004a [1970], 2004b [1975]; and Panofsky 2004 [1934–47]), but such a link is perhaps implicit, given that the claims are exclusive to film, rather than being applied to other pictorial, dramatic, or narrative media. (Note, though, that not all films include moving images. See, for example, Derek Jarman’s Blue [1993].) In some cases, cognitive-illusionist claims might plausibly be rejected as harmless hyperbole, but in others they seem central to a theory. In any case, as we shall see, they are all false. Film has been claimed to engender the illusion that the real things we see on screen – Jeff Bridges pouring various nonalcoholic liquids into a glass during the shooting of The Big Lebowski (1998), for instance – are present to the viewer in the theater, or that the viewer is present to these things on the set (e.g., Bazin 1967b [1951]; Metz 1974: 1–15, 43). (These claims should not be confused with the thesis of transparency – that film enables us to see things that are not present – discussed below.) Alternatively, film has been claimed to engender the illusion that when the viewer sees fictional things on the screen – The Dude making a White Russian, for instance – she believes that they are really present to her, or she to them (e.g., Balázs 1970: 48). All of these claims are untenable for the same sorts of reasons. First, they do not cohere with viewers’ behavior. Depending on one’s other beliefs, a person might act in various ways if she thought she was in the presence of Jeff Bridges or The Dude. But no one watching the movie asks for Bridges’ signature, or inquires of The Dude whether there’s enough Kahlúa for two. A familiar response to such objections is to weaken the claim. Is the viewer in a state of partial belief, or uncertainty about whether or not, she is in the presence of Bridges or The Dude? Again, no. Though someone might approach a stranger on the street if she suspected he were a famous actor, not even Bridges’ biggest fan would ask for his signature while watching The Big Lebowski. Similarly, no one even considers attempting to intercede to stop the nihilists from dropping their ferret into The Dude’s bath (Currie 1995: 24–5; Walton 1990: 197–200). Second, these claims do not cohere with viewers’ other beliefs about their abilities to move about in and perceive the world. If you believed you were present in the fictional world, or even at the shooting locations, of a film, then you would have to believe that you were able to jump from one place and time to another almost instantaneously with a cut from one scene to another; you would have to believe that you could perceive things in the manner of a zoom-dolly shot, or as if through different colored filters. 238
TF17557.indb 238
16/9/08 15:47:30
R E A LISM
A slightly weaker form of cognitive illusionism holds that the viewer believes that the fictional things represented in the film are real but is not misled about the film’s status as a representation. That is, the viewer knows he is watching a film but believes, at least while the film is playing, that it is documentary footage. One reason to doubt this hypothesis is that we are no more confused or forgetful about the status of the film we are watching (be it fiction or documentary) than we are about the fact that we are watching a film, as opposed to perceiving reality. Moreover, things are commonly presented on screen that we could not coherently believe to have been filmed in reality, such as the nihilists’ threatening The Dude while he takes a bath. More plausible than any of these cognitive-illusionist claims is the idea that the viewer imagines, rather than believes, what is fictionally represented on the screen (Walton 1990). But such a thesis requires no reference to the viewer undergoing any illusion. Perceptual illusionism Most writers on film argue or assume that the apparent motion of film images is an illusion. However, Gregory Currie, who rejects all forms of cognitive illusionism, also rejects perceptual illusionism; that is, he defends the view that film images really do move (Currie 1995: 34–47, 1996: 334–42). Currie thinks the burden of proof in these debates rests on the illusionist, since we should take things at face value unless we have reason to doubt them. The reason most people think that the motion of film images is illusory is that they understand that film projectors project a succession of still images, separated by moments of darkness, in such rapid succession that we seem to see a continuously projected image that moves. Thus there is no one film image that moves, any more than there is one image that moves in a “flipbook” (Kania 2002: 244, 246n8). Currie responds with an analogy to colors. Colors are generally thought to be “response-dependent” properties, that is, there is no way to specify what “being red” is without reference to the way we experience the world. (A property such as “being square,” by contrast, could be specified in purely geometric terms, without any reference to our experience of that property.) Nonetheless, colors are not illusory, since we are not wrong when we say that blood is red. Hence, we can say that the motion of film images is real, albeit response dependent. This argument rests on a false analogy. Colors are response-dependent properties, but motion is not. Motion consists, at least, in a thing’s being in contiguous spatial locations at contiguous moments in time. No film image meets that condition (Kania 2002: 254–7; Gaut 2003: 634–5). Currie can argue that film images have a distinct property that is related to motion – response-dependent motion – but this merely renames the illusion. A stick in a glass of water could be described as “response dependently bent,” but this does nothing to militate against the fact that the stick is really straight, despite appearing bent, plainly speaking. Currie has a number of ancillary arguments against perceptual illusionism: (1) If film images didn’t move, then all we could see in a cinema would be static images (1995: 34–5). (2) If photography is transparent, and we see people moving in films, 239
TF17557.indb 239
16/9/08 15:47:31
A ND R E W K A NI A
then the film images must be moving (1995: 38). (3) The identity of film images is response dependent, so we should expect their motion to be response dependent, also (1995: 40–2). However, each of these arguments either falsely assumes that there is a continuously existing image on the screen, or fails to acknowledge the vacuity of attributing response-dependent motion to images. It is worth noting that the above arguments apply only to “moving pictures” produced by certain mechanisms (film, video, zoetrope, etc.). It is much more plausible to suppose that images produced by the uninterrupted projection of light can really move. A simple example would be a spotlight sweeping a prison compound. An example from art is the images on the screen of a shadow play. Furthermore, we can imagine a future technology where the projector’s beam is continuously shone through a transparent cell, whose contents move. Nonetheless, it is not indisputable that such images move, since it is not clear what the identity conditions of images and shadows are (Currie 1995: 30–4; Cassati and Varzi 1994: 175–6). Suppose a searchlight is swept across a cloudy night sky. When the beam runs over a gap in the clouds, does the first spot end as the beam hits the gap and a new one begin when it hits the next cloud, or is this the same spot as before? If so, where was it in the meantime?
Photographic realism Most people agree that seeing a photograph of something is quite different from seeing a painting or drawing of the same thing, and this difference is sometimes captured by saying that photographs are more realistic than paintings or drawings. Since most, though by no means all, films are photographic, many theorists have argued that film is a realistic art form on this basis. The characteristic of photography that is appealed to in justifying this sort of realism is that photographs are mechanically produced, and thus the appearance of a photograph is “counterfactually dependent” on the appearance of its subject. That is, if the subject of a photograph had looked different, then the photograph would have looked different, no matter whether the photographer noticed the difference. This is to be contrasted with the case of painting or drawing where the image produced depends on the beliefs or intentions of the artist. That is, a change in the appearance of an object will only affect the appearance of a painting, if at all, by way of affecting the artist’s beliefs about its appearance. (Sound recording is mechanical in the same sense, but has been little discussed, partly due to a tendency to focus on film images at the expense of film sound, and partly since sound synthesis – the sonic equivalent of painting or drawing – is relatively recent and postdates recording technology.) Ontological realism Sometimes the nature of photography is used to defend an extreme realism or illusionism of the kind put aside at the beginning of the previous section – that a photograph is identical with its subject, or that it gives the illusion that it is. Though we have dismissed these claims, the counterfactual dependence of a photograph on its 240
TF17557.indb 240
16/9/08 15:47:31
R E A LISM
subject does justify a related claim that we might call ontological realism, namely that a photograph of something, unlike a painting of it, guarantees the existence of that thing (Gaut 2003: 634; Walton 1984: 250). Of course, we must continue to distinguish between the real and fictional contents of an image. A photograph of Jeff Bridges shows that he existed at some time, but not that The Dude existed. Transparency A more controversial claim is that photographs are transparent, meaning that in looking at a photograph you literally see its subject (Bazin 1967a [1945]; Walton 1984). (If photography is transparent, but film motion is an illusion, then we really see the objects filmed but do not really see them move. If photography is transparent and film motion real, then we really see the filmed objects in motion.) Almost everyone agrees that when you see something out of your window you literally see that thing, rather than seeing an image in the window. But this example sits at the top of a slippery slope. In which of the following examples do you literally see something? Seeing something while wearing glasses, in a mirror, in a distorting mirror, through a periscope, through a telescope, through night-vision goggles, on closed-circuit television, on a live television broadcast, on a delayed television broadcast, on a taped television show, in a mechanically generated drawing, in a person’s drawing, in a mechanically generated description, in a person’s description. Almost everyone denies that we see things through descriptions – even mechanically generated ones. Opinions differ about where to draw the line between the window and the description. What one needs is an argument for a principle according to which photography is ruled in or out of the class of transparent media. Here are some arguments against photographic transparency that we can dismiss. It is irrelevant to the transparency claim that the object photographed may no longer exist. When we look at the night sky we see many stars that no longer exist (Walton 1984: 252). It is also irrelevant to the transparency claim that when we see a photo we may see no object. If you awake in the middle of the night and look around you may see nothing, though there are objects all around you. This doesn’t show that you don’t see things in ordinary vision (Currie 1995: 57). Another argument against transparency contends that photography is like painting after all, since photographs depend on the intentions of the photographer to choose a particular scene, frame it a certain way, expose the film at a certain moment, and so on. None of this is relevant to the key difference between photography and painting, though. A friend might blindfold you and take you to see a certain sight from a certain angle when the light is just right, even holding a frame up to the view so that you appreciate it in a certain way. Nonetheless, you still see the view, so the fact that photographers do equivalents of all these things does not show that we do not see the subjects of their photographs (Scruton 2006 [1983]: 29–30; Walton 1984: 261–2). More recently it has been argued that the reason photographs are not transparent is that they fail to provide “egocentric information.” That is, you do not know where the subject of a photograph is in relation to yourself. But the provision of such information 241
TF17557.indb 241
16/9/08 15:47:31
A ND R E W K A NI A
is a central function of vision. Thus, photographs are not a way of literally seeing things (Currie 1995: 65–9; Carroll 1996d: 61–2). However, the relation between the function of vision and what counts as vision is not clear. The function of vision is to provide us with information in our immediate environment, yet we still literally see the stars. Also, if we see something in a mirror, then it seems we see something in a periscope, even if we don’t know how the periscope is set up and thus have no idea where what we see is in relation to us (Walton 1997: 69–72). Another recent concern in the debate over photographic transparency has been the continuity of the transmission of light. When you look at a star with the naked eye or through an optical telescope, the light that enters your eye comes directly from the star, though it may have been refracted through lenses and reflected off mirrors. Similarly, when you see an object in a mirror or through spectacles, the light that enters your eyes is the very light reflected off the object. When you see something on closed-circuit television or in a photograph, however, the light that enters your eyes may be similar to the light from the object that is responsible for the image, but it is not that very light. Some take this objection to be decisive (Gaut 2003: 637). Others argue that a transducer that acted just like a window, but collected light on one side and instantaneously emitted qualitatively similar light on the other, would be transparent (Currie 1995: 60, 70). It is difficult to see how this dispute could be resolved without begging the question. A further kind of example is that of a three-dimensional model that is counterfactually dependent on some part of the world. Like a photographic film, the visual appearance of such a model is counterfactually dependent on its “subject,” yet, also like a film, that appearance is not conveyed by the continuous transmission of light. Gregory Currie considers a clock that determines the positions of the hands of a second clock (1995: 64–5). Berys Gaut considers a model jungle that is richly counterfactually dependent on a section of real jungle, to which someone sells tickets, advertising the opportunity to “see real gorillas” (2003: 637; see also Noël Carroll’s model rail-yard, 1996d: 61). Gaut rightly moves to the second example, since in the case of the clocks it is not clear that the information preserved from one clock to another is rich enough. (No matter how accurate your porch-light motion sensor is, you do not see the thing that set it off through seeing the light come on.) Gaut is wrong to think that the model case is decisive, however. He argues that you could legitimately ask for your money back on discovering that you were seeing models rather than gorillas. There are three problems here. One is that there are different ways in which the term “see” is used. If someone sold you the opportunity to see Mars up close with your own eyes and then showed you to a telescope, you might legitimately ask for a refund. Nonetheless, it is true that you see Mars through a telescope. Another problem I suspect is a failure of imagination. It is difficult to imagine a model of a section of jungle that looks just like the section of jungle, gorillas and all, as opposed to a clunky animatronic. If a model of such, well, photographic realism could be achieved, it is not clear one could deny its transparency without begging the question. Finally, as Gaut says, our intuitions here are partly driven by “the abiding human desire to be in direct perceptual contact with objects” (Gaut 2003: 637). This 242
TF17557.indb 242
16/9/08 15:47:31
R E A LISM
muddies the waters, since it would be quite reasonable to believe that one was in the presence of a gorilla if one came upon a perfect robotic model of one, unlike the case of being in the presence of a photograph of a gorilla, and hence we are more likely to think of this case in terms of deception. Moreover, again, there are cases of our being in indirect perceptual contact with things that count uncontroversially as seeing, such as the mirror and telescope cases. Given the vagueness or polysemy of “see,” it may be that once we have figured out all the ways in which looking at objects in photographs is like and unlike ordinary vision, any decision to draw the line between literal and nonliteral seeing will be stipulative. Nonetheless, figuring out these similarities and differences is still of value if we want to understand the nature of photographic film. For if we can at least place film viewing with some precision on the spectrum between simply seeing and seeing in a painting (say), we can appeal to this relative immediacy in explaining one way in which photographic film is realistic.
Perceptual realism Most films are (apparent) moving pictures, and pictures are one kind of representation. They differ from other kinds of representation, such as language, in being parasitic on ordinary perception. Unlike our understanding of language, we use the same capacities to recognize what a picture represents as those we use to recognize objects and events in the world (Carroll 1985: 82–8; Currie 1996: 327–30). As a result, pictures have been described as “perceptually realistic.” One way of seeing the difference is that if you can understand a few pictures in a given style, say, you can understand any other picture in that style, while if you can understand a few words or sentences in a given language, it does not follow that you will understand other words or sentences in that language. Perceptual realism has something to do with the fact that we perceive pictures as resembling what they represent (Currie 1996: 328–30; Walton 1984: 270–3). For instance, if you are likely to confuse a rhinoceros with a hippopotamus, then you are likely to confuse a picture of a rhinoceros with a picture of a hippopotamus. But you are unlikely to confuse the word “rhinoceros” with the word “hippopotamus,” since the words do not resemble each other in the way the animals (and pictures of them) do. Since resemblance is a matter of degree, and perceptual realism appeals to resemblance, so is perceptual realism a matter of degree. For instance, the stylized appearance of a cartoon donkey may be less realistic than a photographic film of a donkey, but the cartoon is nonetheless perceptually realistic, as opposed to, say, the word “donkey.” To investigate the nature of pictures further is beyond the scope of this chapter. (See “Depiction” in this volume.) Instead, I will note a few ways in which perceptual realism relates specifically to film. First, not all the visual elements of a film are pictorial. Even in standard narrative fiction films, titles announcing the time and location of scenes are common. These are linguistic rather than pictorial. But motion pictures are perceptually realistic with respect to more properties than still pictures. Since film is a temporal art, in the sense that the parts of a film have specific duration and ordering, it can be perceptually 243
TF17557.indb 243
16/9/08 15:47:31
A ND R E W K A NI A
realistic with respect to the temporal properties of what it represents (Currie 1995: 92–6). Also, whether or not our perception of movement in film is veridical or illusory, it can be perceptually realistic. (It is perhaps worth noting that much “cognitive” film theory has focused on the nature of depiction in general, in reaction to the persistent notion in previous film theory that film is somehow linguistic. For a selection of such linguistic theories, see the first section of Braudy and Cohen [2004]. For arguments against such theories, see Currie [1992] and Prince [1993].) Second, photographs are arguably the most reliably perceptually realistic kind of pictures we have (Cohen and Meskin 2004), and photographic film inherits that realism. Adding the points about film’s perceptually realistic representation of time and movement, we can understand Currie’s claim that “film is a realistic medium, and deep-focus, long-take style is an especially realistic style within that medium” (1996: 328). Deep focus allows us to perceive more easily or directly the spatial relations represented, and a long take allows us to perceive more easily or directly the temporal relations represented. More immersive formats, such as IMAX and 3-D film are presumably still more perceptually realistic. Third, perceptual realism is not restricted to vision. Most films are also highly aurally realistic, at least with respect to diegetic sound (i.e., sound which “comes from within the world of the film”). We recognize the fact that a shot has been fired off screen by using the same aural perceptual capacities that we use to identify real gunshots (though, as with images, these sounds may be more or less stylized. Think, for instance, of the typical sound effect that accompanies an on-screen punch). Some theme-park movie rides represent the rocky progress of a spaceship by shaking the audience’s seats, thus representing the feel of a rocky ride in a perceptually realistic way. There have also been experiments with olfactory realism, such as Smell-O-Vision and “odorama.” Finally, there is a synesthetic perceptual realism that comes from a matching of sounds and images (and movements, smells, and so on). Perceptually realistic images, coupled with perceptually realistic sounds, will result in a markedly less realistic film if the images and sounds do not match than if they do. (Note, however, that this simple idea calls immediately for qualification, since most film music is “nondiegetic,” that is, unlike most sound effects, it does not represent sounds in the fictional world of the film. See Levinson [1996] for a consideration of such music.)
Aesthetic implications The reason realism has been such a hot topic in film theory is twofold. Theorists have argued, first, that film is uniquely, essentially, or particularly realistic; and, second, that this has implications for how films ought to be made. I have investigated three ways in which the first of these claims has been defended, though it must be noted that I have focused on the extent to which film might be considered realistic. Whether what realism we have found is unique or essential to film is a further question, one to be quite skeptical about. For each kind of realism we have had to limit our discussion to certain types of film – those using traditional projection apparatus, photographic film, 244
TF17557.indb 244
16/9/08 15:47:31
R E A LISM
films employing a particular style, and so on. No such restricted realism can be taken as essential to film in general. But even if such a claim could be defended, it is not clear what implications this would have for what we might call the aesthetics of film – the study of what makes a film a good one. Noël Carroll has argued persuasively against “medium-essentialism,” the view that one ought to exploit effects of one’s art form that are either unique to it, or that it achieves better than any other form (1996a [1984–5], 1996b [1985], 1996d). For one thing, it is doubtful that the most interesting effects achievable in film are achievable only in film. Further, it is difficult to know how we could judge whether films or novels, say, are better at telling stories – something at which they both excel. Supposing such a decision were made, it would be strange to decide that the inferior form should no longer attempt to tell stories. So in general we should be suspicious of deriving aesthetic imperatives from claims about the nature of the medium, including claims about its realism. I end by considering some aesthetic issues specific to the types of realism we examined above. Motion picture realism Surprisingly few film theorists have focused on film’s ability to produce (apparently) moving images (though see Arnheim 1957: 161–87; and Kracauer 1965: 41–5). Anyway, the idea that film should concentrate on the depiction of motion is questionable for the general reasons given above. Noël Carroll argues that the motion of film images should affect the aesthetics of film in a different way. He points out that the apparent motion of images is neither exclusive nor essential to film – there are films without motion and moving images in other media – but he argues that this suggests we should reorient the study of these media in an inclusive way. Film, television, video, etc., are all media in which the motion of images is possible, if not necessary. Thus it makes sense to study motion pictures or moving images in a broad sense, rather than focusing parochially on film (Carroll 1996d). (Note that it will make no difference in most cases whether the motion of the images is real or illusory.) Photographic realism The dispute over the transparency of photographs has been hotly debated in part because of its supposed implications for the aesthetics of photography and film. Roger Scruton (2006 [1983]) has argued that since a photograph is transparent, it cannot be a representation in the sense of expressing a thought about its subject. As a result, Scruton claims that a photograph or film cannot in itself be aesthetically valuable, though what we see through it may be aesthetically valuable. If Scruton is wrong about the transparency of photographs, and if this implies that they are representations, then his argument will not go through. However, the implication is questionable, since it may be that the mechanical nature of photography is sufficient to prevent representation, yet insufficient to achieve transparency (Lopes 2003: 441). Nonetheless, Dominic Lopes argues that even given Scruton’s restricted 245
TF17557.indb 245
16/9/08 15:47:31
A ND R E W K A NI A
notion of representation, photographs can be aesthetically valuable despite their transparency, since there is still a difference between seeing an object face-to-face and seeing it through a photograph. We can take an aesthetic interest in a photograph, then, if we take an interest in (i) the object we see through it; (ii) the way in which the photograph enables us to see it; and (iii) the interplay between (i) and (ii) (Lopes 2003: 442–6). Another way to defeat Scruton’s argument is to show that transparency does not preclude a photograph’s being a representation. Walton argues that photographs are representations since they are (usually) props in games of make-believe (1984: 253–4, 1990: 88, 330, 1997: 68). Stephen Davies argues that if photographs are transparent, then so too are some paintings and drawings. Both photographs and handmade pictures can be counterfactually dependent on their subjects, though the dependence is mechanical in one case and intentional in the other. Since paintings and drawings are paradigmatic representations, transparency cannot preclude representation. This does not open the door to all seeing being mediated by representation. What we see through windows, mirrors, spectacles, and so on is not only counterfactually dependent but continuously dependent on the thing seen. It is the spatiotemporal separation between image and object provided by the camera, the canvas, or whatever, that makes pictures representations distinct from the things we see through them (Davies 2006: 185–8). Note that cinematography does not violate this condition, despite the fact that it does not capture its object at one moment in time, like still photography. Whether or not you think that film images move, it is still the case that either each frame or the temporally extended image, once recorded, is no longer sensitive to changes in its subject. Perceptual realism Gregory Currie argues that films with a long-shot, deep-focus style are more perceptually realistic than others, since they represent spatial and temporal relations (between things) by means of spatial and temporal relations (between images). Our experience of such films is more like our experience of the world than is our experience of films with rapid editing, for instance (1996: 327–30). Should more films, then, employ this style? Clearly this is a matter for debate. Some (such as Bazin) will argue for the superiority of such a style on the basis of its ability to involve us in the world of the film, pointing to the success of films in a realist style. Others will argue that this is a dangerous seduction, and that filmmakers should work against it, alienating the viewer in order to force an awareness of the medium upon her, pointing to the success of films such as those of Jean-Luc Godard. It is not clear that one side must win this debate. Film, like every other artistic medium, is capable of employing many different styles for many different purposes (Carroll 1996b [1985]), and we are capable of appreciating many different kinds of films.
246
TF17557.indb 246
16/9/08 15:47:31
R E A LISM
Conclusions Some films are surely better than others, and perhaps some kinds of films are superior to others. However, it seems unlikely that we can discover which ones by measuring the degree to which they are realistic. Nonetheless, the different kinds of realism we have investigated here are relevant to the study of film for other reasons. Most simply, they allow us to describe the nature of film more precisely, in terms of its illusory nature, the extent to which it is transparent, and one dimension along which cinematic styles can be said to be realistic. These descriptions, in turn, may factor into a psychological explanation of the power and popularity of motion pictures in general, or specific films and styles (Carroll 1985). We should remain wary, however, of drawing conclusions about the value of those films, styles, and motion pictures in general, from premises about their power and popularity. See also Consciousness (Chapter 4), Definition of “cinema” (Chapter 5), Depiction (Chapter 6), Film as art (Chapter 11), Formalism (Chapter 12), Medium (Chapter 16), Music (Chapter 17), Ontology (Chapter 20), Spectatorship (Chapter 23), Sound (Chapter 24), Style (Chapter 25), Rudolph Arnheim (Chapter 27), Bertolt Brecht (Chapter 30), Cognitive theory (Chapter 33), Edgar Morin (Chapter 38), and Phenomenology (Chapter 40), Dogme 95 (Chapter 44).
References Arnheim, R. (1957) Film as Art, Berkeley and Los Angeles: University of California Press. Balázs, B. (1970) Theory of the Film: Character and Growth of a New Art, New York: Dover. Baudry, J.-L. (2004a [1970]) “Ideological Effects of the Basic Cinematographic Apparatus,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: Oxford University Press. —— (2004b [1975]) “The Apparatus: Metapsychological Approaches to the Impression of Reality in the Cinema,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: Oxford University Press. Bazin, A. (1967a [1945]) “The Ontology of the Photographic Image,” in H. Gray (trans.) What Is Cinema? vol. 1, Berkeley and Los Angeles: University of California Press. —— (1967b [1951]) “Theater and Cinema: Part Two,” in H. Gray (trans.) What Is Cinema? vol. 1, Berkeley and Los Angeles: University of California Press. Bordwell, D. (1996) “Contemporary Film Studies and the Vicissitudes of Grand Theory,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Bordwell, D., and Carroll, N. (eds.) (1996) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Braudy, L., and Cohen, M. (eds.) (2004) Film Theory and Criticism: Introductory Readings, 6th ed., New York: Oxford University Press. Carroll, N. (1985) “The Power of Movies,” Daedalus 114: 79–103. —— (1988a) Philosophical Problems of Classical Film Theory, Princeton, NJ: Princeton University Press. —— (1988b) Mystifying Movies: Fads and Fallacies in Contemporary Film Theory, New York: Columbia University Press. —— (1996a [1984–5]) “Medium Specificity Arguments and the Self-Consciously Invented Arts: Film, Video, and Photography,” in Theorizing the Moving Image, Cambridge and New York: Cambridge University Press.
247
TF17557.indb 247
16/9/08 15:47:31
A ND R E W K A NI A
—— (1996b [1985]) “The Specificity of Media in the Arts,” in Theorizing the Moving Image, Cambridge and New York: Cambridge University Press. —— (1996c) “Prospects for Film Theory: A Personal Assessment,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. —— (1996d) “Defining the Moving Image,” in Theorizing the Moving Image, Cambridge and New York: Cambridge University Press. Cassati, R., and Varzi, A. (1994) Holes and Other Superficialities, Cambridge, MA: MIT Press. Cohen, J., and Meskin, A. (2004) “On the Epistemic Value of Photographs,” Journal of Aesthetics and Art Criticism 62: 197–210. Currie, G. (1992) “The Long Goodbye: The Imaginary Language of Film,” British Journal of Aesthetics 33: 207–19. —— (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge and New York: Cambridge University Press. —— (1996) “Film, Reality, and Illusion,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Davies, S. (2006) The Philosophy of Art, Malden, MA; Oxford: Blackwell. Gaut, B. (2003) “Film,” in J. Levinson (ed.) The Oxford Handbook of Aesthetics, Oxford and New York: Oxford University Press. Kania, A. (2002) “The Illusion of Realism in Film,” British Journal of Aesthetics 42: 243–58. Kracauer, S. (1965) Theory of Film: The Redemption of Physical Reality, London: Oxford University Press. Levinson, J. (1996) “Film Music and Narrative Agency,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press. Lopes, D. (2003) “The Aesthetics of Photographic Transparency,” Mind 112: 433–48. Metz, C. (1974) Film Language, trans. M. Taylor, New York: Oxford University Press. Panofsky, E. (2004 [1934–47]) “Style and Medium in the Motion Pictures,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: Oxford University Press. Prince, S. (1993) “The Discourse of Pictures: Iconicity and Film Studies,” Film Quarterly 47: 16–28. Scruton, R. (2006 [1983]) “Photography and Representation,” in N. Carroll and J. Choi (eds.) Philosophy of Film and Motion Pictures: An Anthology, Malden, MA: Blackwell. Walton, K. (1984) “Transparent Pictures: On the Nature of Photographic Realism,” Critical Inquiry 11: 246–76. —— (1990) Mimesis as Make-Believe: On the Foundations of the Representational Arts, Cambridge, MA: Harvard University Press. —— (1997) “On Pictures and Photographs: Objections Answered,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Clarendon Press; New York: Oxford University Press.
248
TF17557.indb 248
16/9/08 15:47:31
23
S P E C TAT O R SHIP Carl Plantinga Film spectatorship, as I shall consider it here, is the experience of viewing and hearing fictional feature films, together with the psychological and social contexts in which such viewing/hearing occurs. A consideration of spectatorship leads to several difficult and fascinating philosophical issues about film viewing, including discussions of various models of the hypothetical spectator; the nature of spectators and their interaction with films; models of the relationship between spectators, texts, and contexts; and how to think about differences between spectators in relation to interpretation and response. Given space limitation, the discussion here will be limited to three broad approaches to spectatorship: (1) screen theory, by which I mean the amalgamation of Althusserian Marxism, Lacanian psychoanalysis, and Barthesian semiotics that dominated the field for over fifteen years, and which is still influential; (2) cultural studies, including historical-reception studies; and (3) cognitive film theory. All comprehensive film theories must consider spectatorship, and thus the reader is advised to examine related entries in this volume.
Hypothetical and actual spectators Generalizations about viewer activities and responses, even those based on historical research (see below), require that the theorist construct or assume a model of the spectator. When scholars discuss the audience or the spectator in abstract terms, they invariably presume features of the spectator’s psychology, relation to history and culture, and ways of responding to and interacting with moving images. In making theories about spectators, theorists have tended to characterize the spectator as either hypothetical or real. Of the real spectator I will say more below. The hypothetical spectator is typically a model of all spectators or some subset of spectators. Janet Staiger has noted several conceptions of the hypothetical reader in literary studies, among them “actual, authorial, coherent, competent, ideal, implied, mock, narratee, necessary, programmed, real, resisting, super, virtual, [and] zero-degree” readers (Staiger 1992: 24). Among the most common types of hypothetical spectators are “ideal,” “competent,” and “implied.” The ideal spectator meets a standard of perspicuity set by a filmmaker,
TF17557.indb 249
16/9/08 15:47:31
CA R L PL A N T I NGA
critic, or theorist. As the word “ideal” implies, the standard is typically high, such that the ideal spectator is able to understand the subtleties and ambiguities of a film, or to appreciate its finer points of style and theme. The competent spectator, by contrast, who might also be called the typical or usual spectator, is thought to possess the basic competencies that enable comprehension of a mainstream text. Implied spectators, finally, are the spectators implied in the design and execution of the film. Film theory before 1960, for example, tended to hold some form of the “invisible observer” hypothesis, which claims that narrative films present story events from the perspective of an imaginary and invisible witness (Bordwell 1985: 9–12). V. I. Pudovkin, the Soviet filmmaker and theorist, claimed that a filmmaker, in constructing a scene, should imagine the perspective of an observer perfectly mobile in space and time, such that the finished scene will clarify the event through the selection and framing of shots and their careful editing. For Pudovkin, the filmic representation “strives to force the spectator to transcend the limits of normal human apprehension” (Pudovkin 1949: 90) with scenes of extraordinary clarity, efficiency, and force. It should be noted that the implied spectator is typically also an ideal and/or competent spectator. Both screen theory and cognitive theory assume variations of the hypothetical spectator. David Bordwell, whose work has been singularly influential in the development of cognitive film theory, conceives of the spectator as a hypothetical entity, not an ideal but rather a competent spectator capable of “executing the operations relevant to constructing a story out of the film’s representation.” Bordwell explicitly states that his hypothetical spectator is “real,” “at least in the sense that she or he possesses certain psychological limitations that real spectators also possess” (Bordwell 1985: 30). Judith Mayne, in her book on film spectatorship, classifies the cognitive theory of the spectator as empirical, by which she means that it is about actual viewers rather than implied spectators, or “subjects” (Mayne 1993: 53–62). But this isn’t quite right. We have already seen that Bordwell conceives of the spectator as a hypothetical entity, but one that is conceived to model empirical spectators in some regards. The question of how a theory of spectatorship could avoid claims about empirical spectators is a good one, and leads us to spectatorship as it is conceived in screen theory. Screen theory often uses the term “spectator” or what Mayne calls the “subject,” in a confusing way. It conceives of the spectator not as a hypothetical person but as a “position,” “role,” or “space” (depending on the theorist) constructed by the text. The theory originally emerged from an interest in how “ideological state apparatuses” maintain mass control through what became known as “subject positioning” (Althusser 1971). For Althusser, an individual is “interpellated,” assigned an identity and role by society’s institutions. Thus the film is said to position the spectator much like ideological state apparatuses position “the subject.” The problem is that this position, role, or space is often described as though it were a person. In one prominent book about terminology in film theory, the authors write that in screen theory the spectator or “the viewer” is “an artificial construct produced by the cinematic apparatus” and a “space” that the cinema “constructs.” Yet they go on to assign human characteristics to this constructed “space” or “position,” claiming 250
TF17557.indb 250
16/9/08 15:47:31
SPE C TATOR SH IP
that in viewing a film a “state of regression is produced” and “a situation of belief is constructed” in the position or subject (Stam et al. 1992: 147). One wants to ask how a role or a position can regress or have beliefs. To take another example, theorists may claim that the spectator is a concept or a structural term constructed by the text (not a person), but then write that the classical stylistic system leads to various text– spectator relationships (relationships between texts and persons). In his introduction to film theory, Robert Stam implicitly recognizes this problem (2000: 231); his somewhat telling solution is to distinguish between the “spectator” and the “actual spectator.” One suspects that the better fix would be to restrict the word “spectator” to designations of persons rather than roles, terms, or positions. Mayne, for example, distinguishes between “subjects,” which are roles or positions, and “viewers,” or actual people. Yet the arch critic of screen theory might also point out that the theory depends precisely on this ambiguity in the meaning of “spectator.” It gains its force by seeming to prescribe determinate effects of texts on viewers, claiming that the film “constitutes” or “produces” the “viewer as subject” (Pribram 1999: 149). Yet the screen theorist, when confronted by counterarguments to such determinism, can also disavow the implausibilities of the theory by claiming that it is really about positions, roles, or subjects, and not about actual people at all (Prince 1996). This is similar to Thomas Pavel’s claims for what he calls Foucault’s “l’esquive empiricotranscendentale,” or “transcendental-empirical dodge,” whereby the theorist, when confronted with empirical counterevidence to a historical argument, will counter that the argument is not about facts, but transcendental conditions of possibility (Pavel 1988: 16, 1992: 7). Also interesting is the attempt by some scholars to bypass hypothetical spectators altogether to directly reach the real viewer. This has been a trend of historicalreception studies, which I consider here to be an offshoot of cultural studies. In the introduction to Hollywood Spectatorship, Melvyn Stokes writes that one of the book’s central purposes is to “question the dominance of theoretical views of spectatorship” (Stokes and Maltby 2001: 1). Hollywood Spectatorship includes essays that investigate the actual and historical rather than hypothetical or possible reception of films. Janet Staiger writes that since historical-reception studies is history and not philosophy, it “does not attempt to construct a generalized, systematic explanation of how individuals might have comprehended texts, and possibly someday will, but rather [describes] how they actually have understood them” (Staiger 1992: 8). Staiger also claims that the historical-reception theorist would criticize the notion of the “ideal reader,” by which I take her to mean any model of the hypothetical spectator, as “ahistorical” (Staiger 1992: 8). For Staiger, even theories that hypothesize spectators by identity (race, class, gender, sexual orientation, etc.) or historical period (pre- and post-sound, classical and postclassical, etc.) are deeply suspect and prone to “universalize” viewer experiences (Staiger 2001). Yet any claim that spectatorship can be studied without the benefit of general psychological assumptions about spectators constitutes a naïve empiricism that most reception theorists and cultural studies proponents, Staiger included, would be loathe to embrace. Staiger’s brand of historical-reception studies does make general 251
TF17557.indb 251
16/9/08 15:47:31
CA R L PL A N T I NGA
theoretical claims about how spectators interact with texts. Staiger argues, for example, that spectators are “perverse” in the diversity of ways they interact with films. She also claims that the film text itself has no “immanent” meaning for the spectator, implying that the spectator constructs the film’s meanings. Both of these are elements of a de facto model of the hypothetical spectator. The path to the “real” spectator runs through the fields of history and the formation of general hypotheses about spectator psychology and behavior. The historical-reception theorist may well ask that hypothetical models of the spectator meet some standard of plausibility or testability, but historical-reception studies cannot do away with hypothetical models of the spectator altogether. In arguing this, however, I am not claiming that scholars need a comprehensive a priori model of the spectator. A model of the spectator may describe only selected elements of the spectator’s activities, psychology, or cultural influences.
Models of the spectator How have screen theory, cultural studies, and cognitive theory modeled the spectator? Such a model could be very complex, but I will focus on three issues: (1) relations between text, context, and spectator; (2) degree of spectator activity or passivity; and (3) the spectator’s motivation and psychology. I will also mention and in some cases discuss the major criticisms of each theory. Viewing context, the characteristics of the film itself, and the spectator’s particular subjectivity all must be taken into account in determining actual spectator responses and interpretations. Yet various theories seek to enlarge or diminish the role of one or the other. Screen theory is often thought to foreground the determinative powers of the film text at the expense of both context and spectator, but the issue is actually more complicated than this. If the spectator is constructed or positioned by the film text, then one analyzes the text to determine how such positioning occurs; this explains the focus on textual analysis prevalent in much screen theory. Now of course, the degree to which screen theory can be said to ignore the actual spectator depends on whether one takes this entity “spectator” to be a person or a position, as I have discussed above. But the theory is an empty shell if it does not assume that actual persons occupy the roles constructed by the text. That is why in screen theory the spectator is often described as having drives, an unconscious, desires, and fantasies; thus the “spectator” is not just a role or position, but a model for actual persons. In fact, textual analysis is a viable method of determining spectator response only if (a) actual viewers are thought to share important psychological characteristics; and (b) varied viewing contexts are relatively unimportant in relation to some (but not necessarily all) elements of viewer-text interaction. Screen theory conceives of the spectator as passive, an unwitting victim of a system built to exercise hegemonic control of its capitalist subjects. The institutions and technologies of the cinema are often referred to as an “apparatus” that functions to acculturate subjects to structures of fantasy, desire, dream, and pleasure that fully conform to dominant ideology (Mayne 1993: 18). Since the supposed ideological effects of mainstream films are built into the apparatus and apparently irresistible, a 252
TF17557.indb 252
16/9/08 15:47:31
SPE C TATOR SH IP
viable strategy of resistance is the production of a “countercinema” and “progressive texts” that subvert the regressive pleasures of mainstream film (Comolli and Narboni 1976). It should be noted that the films of the countercinema, such as Jean-Luc Godard’s Weekend (1967) and Sally Potter’s Thriller (1979), were often influenced by Brechtian theory more than by psychoanalysis. Thus the idea of the progressive text does not necessarily depend on a psychoanalytic model of the mind (see Stam 1992; and Lovell 1982; for critiques of Brechtian theory as manifested in film studies, see Plantinga [1997] and Smith [1996]). The psychological model of the spectator employed in screen theory is complex and varied, so this brief account risks overgeneralization (for a fuller description see “Psychoanalysis” in this volume). In screen theory, the viewer is taken to be hedonically motivated to watch films, the sources and workings of the spectator’s desires being identified in the context of some version of the Freudian or Lacanian unconscious. The cinema is thought to be uniquely capable, among the arts, of eliciting unconscious processes similar to daydreaming, night-dreaming, voyeurism, fetishism, early psychosexual fantasies, and primitive oral narcissism, all of which engage “desire.” The theory goes on to harness such processes to subject positioning, such that the subject’s drives, or desires, channel him or her toward repression or socially acceptable patterns of subjectivity – toward the good capitalist subject. Screen theory offers a controversial, yet in some ways seductive, theory of the motivations of the spectator – desire and the drive for pleasure – and of the unconscious processes put into play by film viewing. Few of its competitor theories have yet dealt with the vital issue of spectator motivations and pleasures in such depth (see Plantinga 2009, for an alternative account). Yet screen theory has received significant criticism. From the perspective of cultural studies, it has been critiqued for assuming a passive spectator, for ignoring differences between spectators, for minimizing history and more generally the contexts of film viewing, and for the determinism inherent in its use of psychoanalysis (Mayne 1993: 52). Cognitive theorists and analytic philosophers have critiqued the theory for its bad logic and reasoning. Noël Carroll, for example, has found fault with the protocols of reasoning instantiated in the theory, arguing that its “central concepts are often systematically ambiguous,” often leading to equivocation and argument by pun (Carroll 1988: 226–34). David Bordwell has similarly criticized the reasoning methods of screen theory, such as its purported top-down inquiry impervious to empirical data, argument by bricolage, and associative reasoning (Bordwell 1996a: 18–26). In general, Bordwell and Carroll believe that psychoanalysis should be reserved for special cases, such as irrational spectator behavior, that psychoanalysis might be employed when “there is an apparent break-down in the normal functioning of our cognitive-perceptual processing” (Carroll 1996a: 64) and that “conscious and preconscious” spectator processes are better explained by cognitive science and associated methods (Bordwell 1985: 30). But it is not clear that reserving the application of psychoanalysis to irrational spectator behavior makes much sense. For either psychoanalysis is false as a general theory of the human psyche, in which case it doesn’t apply to irrational or rational behavior, OR it is true, in which case it explains both! 253
TF17557.indb 253
16/9/08 15:47:31
CA R L PL A N T I NGA
Although screen theory does not hold the dominant position in film studies that it once did, its legacy is still substantial. For one, as David Bordwell convincingly argues, many of the dominant assumptions of what he calls “subject position” theory clearly remain in cultural studies (Bordwell 1996a). In addition, the claim for textual power over spectators has not been given up. In her book, which combines elements of screen theory and cultural studies, Michele Aaron argues that “the spectator’s submersion in a submission to the text . . . must be understood as an inevitable part of the act of engagement” (Aaron 2007: 3). In contrast to screen theory, cultural studies seemingly offers a much more open and diverse account of possible film/spectator relationships. Cultural studies scholars, and especially historical-reception theorists, agree that the film text itself has been granted a larger role in determining spectator interpretation and response than it deserves. The focus must now be expanded outward to the film viewing context, and for some scholars, further yet. Rick Altman suggests that we think in terms of the “cinematic event,” in which reception opens “onto an infinite cultural space . . . neither beginning or ending at any specific point” (Altman 1992: 4). Cultural studies thus examines fan magazines, star images, film reviews, advertising and publicity, newspaper accounts of riots and disturbances at screenings, and in general any element of discourse and historical context that bears on viewing experiences. Historical-reception scholars offer case studies that show how contextual factors influence response, and that responses vary over time. They also argue that the film itself has no “immanent” meaning. Thus reception scholars will sometimes deny the distinction between the comprehension of what the film explicitly signifies, and the interpretation of implicit or ambiguous meaning (Staiger 1992: 20), which has the effect of draining the cinematic work of meaning and attributing meaning-making to the spectator. Taken together, the historical case studies, the discourses surrounding a particular film, and the denial of the distinction between comprehension and interpretation might be thought to minimize the importance of the film itself in determining or even influencing spectator interaction and response. While space prohibits a full discussion of these issues, I will make two observations. The first is that the historical-case studies provided to demonstrate diversity of interpretation do not disprove the distinction between comprehension and interpretation, but rather affirm it. All of the diverse uses of and responses to films are grounded on basic agreements about first-level denotative meaning, the very elements that are comprehended rather than interpreted. Second, one might question the implicit assumption that spectators are wholly free to interact with texts in any way they wish and that they actually exercise whatever freedoms they do have. Using the framework of historical-reception theory, one might point out that for many viewing contexts, the most powerful available frameworks for the reception of movies will be limited and conventional, such that the spectator chooses a film, from a narrow range of possibilities, based on the kind of experience the film seems to offer. This choice of which film to view strengthens the impression that the viewing “contract” is freely entered into and increases the likelihood of a congruent or sympathetic viewing. In many cases, then, the spectator does not resist, but rather immerses herself in and actively enjoys, 254
TF17557.indb 254
16/9/08 15:47:31
SPE C TATOR SH IP
that experience, in effect opening herself to the film. This isn’t the only “reading strategy” available, but it is arguably the most widespread and common, whether the film is viewed in a theater or at home. If this is right, it restores a more central place to the film text itself in making meaning and generating response because the conventional receptive context is assumed by filmmakers, exhibitors, and many audiences and is built into the design of films, marketing and advertising campaigns, and the institutions of exhibition and distribution. The spectator’s response is never uniquely determined by the text, but contextual factors strongly support textual influence. With regard to whether the spectator is active or passive, there is an unresolved tension in various strands of cultural studies between, on the one hand, the impetus to attribute a wholesale freedom to the spectator to interact with and respond to films in an infinite variety of ways (the perverse spectator or the “free” consumer) and, on the other, the claim that spectators are “constituted” in discourse, along with the recognition that any historical context prescribes a limited set of “possible” interpretive strategies for the spectator. The possibility of resistance to oppressive discourse is predicated on, for example, the spectator’s ability to actively counter dominant “readings,” yet such resistance implies agency and perhaps even free will. Culture studies proponents who insist that spectators are the products of sociocultural and ideological forces (including discourse) sometimes contrast such models of the spectator with the “humanist subject,” who “creates ‘himself’ ” and “controls the surrounding world” (Pribram 1999: 158). Yet the assumption of spectator agency and the possibility of resistance to social forces suggests a smaller difference between the cultural studies and humanist conceptions of the viewer than is sometimes assumed. Few humanists, after all, would deny that culture and ideology influence viewers. And any spectator with even a modicum of personal agency, one would think, could resist and affect “the surrounding world” and maintain at least some control of her subjectivity. Even if someone cannot exercise direct control over thoughts and feelings, she can exercise indirect control, for example, by leaving the theater or shutting down the DVD player. Last, we come to the spectator as modeled in cognitive film theory. Cognitive film theorists foreground the film text and the spectator and usually minimize the viewing context. This stems in part from the assumption that spectators share certain psychological characteristics and viewing activities. Bordwell argues that for some purposes, film scholars can bypass the nature/culture debate by focusing on “contingent universals,” that is, patterns of behavior and response that spectators share. Such patterns are contingent because “they did not, for any metaphysical reasons, have to be the way they are” and are universal because they are “widely present in human societies.” For the cognitive theorist, diverse viewers approach films, to some extent, with similar psychic hardware. Many cognitive theorists embrace an evolutionary psychology that posits similar adaptive challenges in human history, resulting in common physical and psychological characteristics. As Bordwell writes, the “core assumption here is that given certain uniformities in the environment across cultures, humans have in their social activities faced comparable tasks in surviving and creating their ways of life” (Bordwell 1996b: 91). This has led to the development 255
TF17557.indb 255
16/9/08 15:47:32
CA R L PL A N T I NGA
of common psychological characteristics that are neither wholly cultural nor wholly biological. Such contingent universals – for example, the startle response to loud sounds or inference-making when confronted with a narrative enigma – are assumed by filmmakers in the design of films. One project common to many cognitive theorists is the “naturalization” of film aesthetics, that is, the demonstration that various manifestations of film form and style have roots not only in cultural convention, but in human nature as developed in the broad span of human evolutionary adaptation (Anderson 1996: 23–5; Carroll 2003: 10–58; for a dissenting view, see Choi 2006). The assumption that spectators share characteristics universally is a controversial one, and will be addressed in the next section. While screen theory modeled the spectator as a passive victim of ideology, Bordwell posits an active spectator. “A film,” he writes, “does not ‘position’ anybody” but rather “cues the spectator to execute a definable body of operations” (Bordwell 1985: 29), such as making inferences and testing hypotheses. Bordwell’s constructivism, moreover, posits a viewer who maintains a Gombrichian beholder’s share, not only constructing the film’s meanings but also its fabula, or story (Bordwell 1989: 1–8). Yet not all cognitive film theorists embrace constructivism. Gregory Currie, for one, argues that film viewers typically, or at least often, find rather than construct meaning. “From the point of view of stressing viewer activity, there is nothing to choose between saying that the viewer creates meaning and that she finds it” (Currie 2004: 160; see also Gaut 1995); both constructing and finding are active. Currie argues that Bordwell’s constructivism is not central to cognitive film theory. With regard to spectator activity, it should also be noted that, although both cognitive theory and cultural studies posit an active spectator, the activities of interest differ. The former emphasizes the spectator’s psychological activities in relation to the film text, while the latter focuses on identity formation in relation to films, as well as their political uses. Screen theory holds that the spectator is motivated, and often at an unconscious level, by pleasure and desire. Cultural studies posits a spectator motivated primarily by identity politics, a spectator who finds and identifies with some representations and actively resists others. Cognitive theory in its early years tended to assume a spectator motivated by curiosity and anticipation to understand or comprehend a film. This resulted in various explorations of the means by which spectators comprehend narrative in film viewing (Bordwell 1985; Branigan 1992). Cognitive theorists have also explored the perception of images and the means by which viewers make sense of various film techniques (Currie 1995; Anderson 1996; Carroll 1996b, 2003; Persson 2003). More recently, cognitive theorists and philosophers have begun to explore character engagement or identification (Smith 1995; Gaut 2006; Coplan 2004; Neill 2006) and more broadly, the general emotional and affective responses elicited by films (Carroll 2008; Grodal 1997; Plantinga forthcoming). I have argued that pleasures and desires should be accounted for in any account of film spectatorship, and can be useful if the concepts are used in their folk-psychological senses (Plantinga forthcoming). Cognitive film theory has been criticized from various perspectives. For one, it is said to “universalize” the spectator in much the same way as screen theory does, 256
TF17557.indb 256
16/9/08 15:47:32
SPE C TATOR SH IP
and has shown little interest in spectator difference. Some have found its model of spectatorship to be naïvely putting its faith in science and rational discourse (Stam 2000: 240–1) or to neglect the nonrational elements of spectatorship (Campbell 2005). Its concern with universal spectator psychology is sometimes thought to be trivial (see Robert Stam, as quoted in Quart 2000: 41) or narrow, while cultural studies, in contrast, is thought to be politically engaged. Bordwell and Carroll defend mid-level research on spectatorship that asks and attempts to answer researchable questions rather than engage in broad, top-down theory (Bordwell 1996a; Carroll 1996a). And the cognitive theorist might also answer that there is nothing in principle that would prevent the cognitive theorist from dealing with spectator difference or identity politics, despite the few attempts to do so thus far; in fact, various theories of social cognition would seemingly provide an excellent platform for such explorations.
Spectator difference and similarity The study of spectator difference is a strong current in contemporary film and media studies. The issue deserves an essay unto itself, but here we must be content with a few remarks. Various accounts of spectatorship in film studies often begin with the homogeneous spectator of screen studies and count as a marked improvement the gradual recognition of heterogeneous spectators and the “multiple points of identification” possible in the viewing of any film. Laura Mulvey initiated the discussion of difference with her oft-cited essay “Visual Pleasure and the Narrative Cinema” (Mulvey 1992 [1975]), in which she argued that the spectator implied in mainstream films was male. This both instantiated gender as the major concern of studies of spectator difference for many years, and led to a good deal of scholarship attempting to theorize the place of female subjectivity in the film viewing experience (for an overview see Aaron 2007: 24–50; see also “Gender” in this volume). Subsequent scholarship has branched out to consider spectatorship from the perspective of race, class, ethnicity, and sexual orientation, sometimes in particular historical periods and contexts (see “Race” in this volume). Such studies are undeniably of great interest. If we wish to discover the place of films in the lives of actual persons, spectator difference will have to be taken into account. Yet when the interest in spectator differences leads to a denial of similarities in viewers, there is a problem. As I argued at the essay’s beginning, the study of spectatorship cannot proceed without a model of the hypothetical spectator. This issue has not been sufficiently discussed in film studies (or in philosophy). It is often presumed that the promotion of spectator difference is politically enabling and properly celebrates diverse human groups. Yet the study of difference in itself, like the search for human commonalities, is politically neither progressive nor regressive. The belief in presumed differences between races was the foundation of hundreds of years of racist oppression, and the idea that all people are in some sense equal or equally deserving of basic rights lay at the heart of the Civil Rights movement in the United States. Neither does the summary dismissal of “universals” in human nature derive 257
TF17557.indb 257
16/9/08 15:47:32
CA R L PL A N T I NGA
from careful consideration of the issues. While “universalizing” the spectator is often assumed to be a BAD THING, one sees few discussions of just what this “universalizing” would entail. Are we to deny any similarities between spectators whatsoever? Moreover, in film and media studies, there have been very few discussions of the nature of universals themselves, nothing on the order, for example, of Patrick Colm Hogan’s discussion of the concept (Hogan 2003: 133–9). All of this is to suggest that the heavy promotion of spectator difference and the opprobrium attached to all discussion of universals or human similarities is something of an “idol of the tribe,” a faulty habit of thought, a misplaced orthodoxy. In scholarship about spectatorship, both viewer differences and viewer similarities must eventually be accounted for. See also Interpretation (Chapter 15), Emotion and affect (Chapter 8), Empathy and character engagement (Chapter 9), Consciousness (Chapter 4), Gender (Chapter 13), Race (Chapter 21), Psychoanalysis (Chapter 41), Cognitive theory (Chapter 33), and Phenomenology (Chapter 40).
References Aaron, M. (2007) Spectatorship: The Power of Looking On, London and New York: Wallflower. Althusser, L. (1971) “Ideology and Ideological State Apparatuses: Notes towards an Investigation,” in B. Brewster (trans.) Lenin and Philosophy, New York: Monthly Review Press. Altman, R. (1992) Sound Theory Sound Practice, New York: Routledge. Anderson, J. D. (1996) The Reality of Illusion: An Ecological Approach to Cognitive Film Theory, Carbondale: Southern Illinois University Press. Bordwell, D. (1985) Narration in the Fiction Film, Madison: University of Wisconsin Press. —— (1989) Making Meaning: Inference and Rhetoric in the Interpretation of the Cinema, Cambridge, MA: Harvard University Press. —— (1996a) “Contemporary Film Studies and the Vicissitudes of Grand Theory,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press, 3–36. —— (1996b) “Convention, Construction, and Cinematic Vision,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press, 87–107. Branigan, E. (1992) Narrative Comprehension and Film, London and New York: Routledge. Campbell, J. (2005) Film and Cinema Spectatorship, Cambridge, UK; Malden, MA: Polity Press. Carroll, N. (1988) Mystifying Movies: Fads and Fallacies in Contemporary Film Theory, New York: Columbia University Press. —— (1996a) “Prospects for Film Theory: A Personal Assessment,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press, 37–68. —— (1996b) Theorizing the Moving Image, Cambridge: Cambridge University Press. —— (2003) Engaging the Moving Image, New Haven, CT: Yale University Press. —— (2008) The Philosophy of Motion Pictures, Malden, MA: Blackwell. Choi, J. (2006) “Naturalizing Hollywood? Against the Naturalistic Account of Filmic Communication,” Film Studies: An International Review 8: 149–53. Comolli, J., and Narboni, J. (1976) “Cinema/Ideology/Criticism,” in B. Nichols (ed.) Movies and Methods, vol. 1, Berkeley: University of California Press, 22–30. Coplan, A. (2004) “Empathic Engagement with Narrative Fictions,” Journal of Aesthetics and Art Criticism 62: 141–52. Currie, G. (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge and New York: Cambridge University Press. —— (2004) Arts and Minds, Oxford: Clarendon Press; New York: Oxford University Press.
258
TF17557.indb 258
16/9/08 15:47:32
SPE C TATOR SH IP
Gaut, B. (1995) “Making Sense of Films: Neoformalism and its Limits,” Forum for Modern Language Studies 31: 8–23. —— (2006) “Identification and Emotion in Narrative Film,” in N. Carroll and J. Choi (eds.) Philosophy of Film and Motion Pictures, Malden, MA: Blackwell. Grodal, T. (1997) Moving Pictures: A New Theory of Film Genres, Feelings, and Cognition, Oxford: Clarendon Press; New York: Oxford University Press. Hogan, P. C. (2003) Cognitive Science, Literature, and the Arts: A Guide for Humanists, London and New York: Routledge. Lovell, A. (1982) “Epic Theater and the Principles of Counter-Cinema,” Jump Cut: A Review of Contemporary Media 27: 64–8. Mayne, J. (1993) Cinema and Spectatorship, London and New York: Routledge. Mulvey, L. (1992 [1975]) “Visual Pleasure and Narrative Cinema,” The Sexual Subject: A Screen Reader in Sexuality, London and New York: Routledge. Neill, A. (2006) “Empathy and (Film) Fiction,” in N. Carroll and J. Choi (eds.) Philosophy of Film and Motion Pictures, Malden, MA: Blackwell. Pavel, T. G. (1988) Le mirage linguistic: essai sur la modernisation intellectuelle, Paris: Minuit. —— (1992) The Feud of Language, Oxford, UK; Cambridge, MA: Blackwell. Persson, P. (2003) Understanding Cinema: A Psychological Theory of Moving Imagery, Cambridge and New York: Cambridge University Press. Plantinga, C. (1997) “Notes on Spectator Emotion and Ideological Film Criticism,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, Oxford: Clarendon Press; New York: Oxford University Press, 372–93. —— (forthcoming) Moving Viewers: American Film and the Spectator’s Experience, Berkeley: University of California Press. Pribram, E. D. (1999) “Spectatorship and Subjectivity,” in T. Miller and R. Stam (eds.) A Companion to Film Theory, Malden, MA: Blackwell. Prince, S. (1996) “Psychoanalytic Film Theory and the Problem of the Missing Spectator,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press, 71–86. Pudovkin, V. I. (1949) Film Technique, trans. I. Montagu, New York: Lear. Quart, A. (2000) “The Insider: David Bordwell Blows the Whistle on Film Studies,” Lingua Franca 10(2): 34–3. Smith, M. (1995) Engaging Characters: Fiction, Emotion, and the Cinema, Oxford: Clarendon Press; New York: Oxford University Press. —— (1996) “The Logic and Legacy of Brechtianism,” in D. Bordwell and N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press, 130–48. Staiger, J. (1992) Interpreting Films: Studies in the Historical Reception of American Cinema, Princeton, NJ: Princeton University Press. —— (2001) “Writing the History of American Film Reception,” in M. Stokes and R. Maltby (eds.) Hollywood Spectatorship: Changing Perceptions of Cinema Audiences, London: British Film Institute, 19–32. Stam, R. (1992) Reflexivity in Film and Literature: From Don Quixote to Jean-Luc Godard, New York: Columbia University Press. —— (2000) Film Theory: An Introduction, Malden, MA: Blackwell. Stam, R., Burgoyne, R., and Flitterman-Lewis, S. (1992) New Vocabularies in Film Semiotics: Structuralism, Post-Structuralism and Beyond, London and New York: Routledge. Stokes, M., and Maltby, R. (2001) “Introduction: Historical Hollywood Spectatorship,” in M. Stokes and R. Maltby (eds.) Hollywood Spectatorship: Changing Perceptions of Cinema Audiences, London: British Film Institute, 1–16.
259
TF17557.indb 259
16/9/08 15:47:32
24
S OUND Giorgio Biancorosso The past three decades have seen the emergence of a wave of specialized scholarship that has greatly enhanced our understanding of sound in cinema and raised awareness of its significance to an unprecedented degree. This new development in film studies was very much needed. It was not until the work by Rick Altman and Tom Levin in the early eighties, for instance, that fundamental questions about the impact of sound technology on the perception of auditory events in cinema were first raised (Altman 1985–6; Levin 1984; see also Lastra 2000). Michel Chion’s seminal work, too, emerged in the eighties, and its influence in the Anglo-American world crystallized in the nineteen-nineties (some of its most important implications, in fact, still await further development). Framing much of the work produced in the wake of these authors’ early, influential writings was a discussion of the relative importance of sound versus “the image.” While crucial in instigating a debate on the marginalization of sound in film studies and investigating its causes, comparing sound with the image is in itself of little interest, however. The terms are too vague to be of any use, as both “image” and “sound” are at best approximations for complex, multifaceted phenomena. Moreover, the vastly different roles played by aural and visual cues in a film, despite their convergence in the mind of the spectator into something like a complex gestalt, make any attempt to establish a hierarchy between them simply uninformative. The significance of sound in cinema must be gauged in terms that are germane to hearing. This need not be at the expense of vision, however. In fact, a fertile – indeed, urgent – question is whether it is useful to talk about film sound as a discrete area of inquiry in the first place. The history of filmmaking amply illustrates that sound is a specific area of praxis with its own specialized terms and practitioners. Separating out film sound as a category of the spectator’s experience or an element in the production of meaning, however, is more problematic. For the more one addresses “sound” as a subject, the more one reaffirms the idea of its separation from the other elements of film. This not only legitimizes approaches to film analysis and interpretation that leave out sound as a matter of course but also endorses, however unwittingly, a view of “film sound” as a much more cohesive and unified field of experiences than it actually is. As is well known, the term “sound” refers to dialogue, music, sound effects, and silence, while its cognate “soundtrack” refers to their strategic combination. These
TF17557.indb 260
16/9/08 15:47:32
S OU ND
four broad categories further subdivide into many more types, reflecting a myriad of generic, stylistic, sociocultural, and personal factors at work in the shaping of a film. There is no one sound or sound dimension, but only concrete instances of recorded sounds mixed and reproduced in a certain way, and these can open up a complex, highly diversified world of its own (as the existence of radio dramas proves beyond question). In a film, an individual sonic occurrence may be best understood in relation not to other sounds but to a simultaneous visual shape or suggestion, dramatic lead, or symbolic meaning. Sounds, moreover, reach us in a physically overdetermined fashion, as they feature reverberation, time of decay, intensity of tone, relative loudness, and timbre (to name only a few of their characteristics). Not infrequently, one of these aspects stands out, reaching out, as it were, to an element of the mise en scène. Think, for instance, of how reverberation may link up to the shot of a particular place to conjure the impression of spatial depth, or how the rhythm and pace of a certain sonic pattern, synchronized to images of objects crossing the frame, helps convey the impression of motion. Criticism and analysis must bring out the finely textured connections between the visual and auditory elements of film as they pertain not only to different kinds of sounds but also to various aspects of those sounds. Michel Chion is therefore right when he refers to cinema as a form of “audiovision.” One important corollary of his theory is the provocative claim that “there is no soundtrack,” as “[e]ach audio element enters into simultaneous vertical relationships with narrative elements contained in the image (characters, actions) and visual elements of texture and setting” (Chion 1994: 40). The claim has been challenged on the grounds that sound-to-sound connections are just as important and are symptomatic of a holistic understanding of the soundtrack on the part of filmmakers (Altman et al. 2000: 341). But this seems to nuance and enrich, rather than fundamentally alter, the gist of his argument, especially as it pertains to the experience, as distinct from the production, of a film. Chion’s analyses and neologisms, and above all his ideas of synchresis, causal listening, and the acousmêtre, are eloquent testimony to his attempt to read auditory and visual aspects of film in terms of their relationships, rather than spurious separateness. It is these relationships one must describe, if one wants to recover a full sense of the perceptual richness of the spectatorial experience. It is symptomatic of the formidable grip of old habits and prejudices that Chion’s call for a truly integrated approach to the aural and visual dimensions of cinema has primarily informed scholarly work on sound, but not film analysis tout court. The aptness of the term “audio-vision” notwithstanding, sitting through a fiction film amounts to more than just registering visual and auditory stimuli; rather, it is the active seeking of cues for the construction of a coherent story and parsing of a constantly evolving narrative. As such, it involves the interplay of seeing and hearing, not only with one another, but with previously acquired and newly accumulated knowledge. It is to this as yet unexplored area of the phenomenology of film that I would now like to turn. In the interest of space, I will do so by limiting the discussion to an important area of intense speculation in film theory: the relationship between sounds and their sources. 261
TF17557.indb 261
16/9/08 15:47:32
GIORGIO BI A NC ORO S S O
Edward Branigan observes that “[t]he perceptual difficulty of decoupling sound from a lighted object [understood as its source] is suggested by the fact that it is extraordinarily rare that a film spectator is led to interpret noise nondiegetically” (1997: 110 on unanchorable sounds, see Jost 1989). This difficulty, as we will see, lies at the heart of the art of dubbing as well. But let us stay with the example of noise and ask, what about offscreen noise? Examples are legion. How does one determine that noise is diegetic even when a source is not visible? The spectator may know that there is a source in the scene, either because he or she has previously seen it, or because a character has referred to it, or simply because it seems plausible or he expects it. He or she may infer that the noise is diegetic by its kind or because that makes sense, given the dramatic context at hand. Indeed, the determination may be made, to paraphrase psychologist Jerome Bruner, “before the information given” (1973). If the film has established a pattern of using certain types of sounds at recurring junctures then the spectator will instantly hear a noise as diegetic, simply by its complying with what in narratological jargon is called an infra-textual rule, that is, a pattern or figure that is distinctive of the film and the spectator’s having learned to recognize and anticipate (for example, in his films, Jacques Tati flaunts the recurrence of certain sounds to tease the audience and help them recognize their strangeness). The perception of a sound in film is imbued with beliefs, reminiscences, and expectations about its role in an imaginary construct, even when the source is clearly visible. This can be seen through a number of limit cases. In Woody Allen’s notorious gag from Bananas (1971), the music underscoring Alvin’s daydreaming about his meeting with Mr Vargas turns out to be played by a flesh-and-blood harpist hidden in the closet of his hotel room. One may be tempted to take the pun as paradigmatic of the ease with which music crosses the boundary between the diegetic and nondiegetic. In fact, the laughter with which the audience normally responds to the startling sight of the harpist suggests just the opposite. At issue is not whether music in film is used both diegetically and nondiegetically – it is, as we all know. The scene addresses a more specific question: Can the same music be both nondiegetic and diegetic within the same context? In the case of this excerpt, the answer is that it cannot. The harp glissandos may be plausible as nondiegetic background scoring – notwithstanding the parody implied therein – but it is entirely implausible as an occurrence in the film’s story world, and that is why it cannot ultimately be anchored to a source without generating amusement, surprise, or outright laughter. No matter how much political truth there may be in the joke – if one was ever intended – harpists do not practice in closets. It’s just not done, not even under the most fantastic and ridiculous Latin American dictatorship (the film is set in an imaginary Latin American country meant as a parody of both a fascist and a communist regime). To be sure, Allen’s conjuring trick demonstrates the power of seeing in guiding our understanding of music in film. Presented with the image of the on-screen harpist, one finds oneself forced to attach the music to a source known to be impossible or at best highly unlikely. But the success of Allen’s conjuring trick also depends on the power of the knowledge of that very impossibility – for after all what he is after is a comical effect. Seeing the harpist in the closet, replaying the music in retrospect as if it were 262
TF17557.indb 262
16/9/08 15:47:32
S OU ND
diegetic, and yet knowing that this is at best unlikely, creates a short circuit – the awareness of a contradiction – that plays itself out as laughter. As publicly observable behavior, laughter is the evidence of the complex interplay of hearing, seeing, and knowing that characterizes the perception of sound in film. It may be argued that this interplay occurs only under very unusual circumstances, as one picks up information about the story world, much as a Gibsonian subject detecting invariants and building a picture of the on-screen world, irrespective of concepts and beliefs and the expectations that go with them (Gibson 1966). But whether dormant or concurrent, beliefs about the likelihood, plausibility, or artificiality of the coupling between a sound and an image are a central element of the phenomenology of the perception of sound in film, for they lie at the core of our relationship to film as representation. Whether understood as entertainment, narrative comprehension, or contemplation, the film experience needs a theory of imaginative response as much as a theory of perception to enable the richness of its phenomenology to emerge in full. This is why it is important to distinguish between the perception of film sound and that of sound in general. Consider again the case of noise. While on-screen noise is nearly impossible to hear nondiegetically, due to the difficulty of “decoupling it from a lighted object,” to use again Branigan’s words, the instant recognition of noise as offscreen, as distinct from nondiegetic, is less easily explained (though it is equally common). It may well reflect the exercise of a certain kind of knowledge and thus result from a prediction of how films work (as already mentioned above). It may also reflect patterns of behavior common in everyday life. After all, sounds whose sources remain unseen not only reach us at all times, but are also crucial in guiding our sense of inhabiting a certain kind of space, specifying its properties and suggesting the kinds of activities taking place therein. It is fair to assume that we bring this ability to perceive the space around us through sound to bear on the construction of a diegetic space. Digital Surround Sound depends upon it. The experience of offscreen space, however, is layered in a special way, as it is informed by the awareness that it has been consciously left out of the frame and that this exclusion plays a specific role in an intentionally conceived entity (that is, the film as the product of a creative team working within a tradition and for a certain audience). Whether seen or unseen, sources are of paramount significance for reasons that go beyond their role in determining the status of sounds within the narrative. Upon addressing the question of the perception of sound, Christian Metz argues that “the recognition of a sound leads directly to the question: “a sound of what? . . . In language as the metacode of sounds, the most complete identification is obviously that which simultaneously designates the sound and its source (‘rumble of thunder’)” (1980: 25, 26). Metz attributes this linguistic habit to what he calls “primitive substantialism,” namely the tendency to distinguish “fairly rigidly the primary qualities that determine the list of objects (substances) and the secondary qualities which correspond to attributes applicable to these objects” (27). Descartes and Spinoza are mentioned as the beginners of this tradition, but in the Anglo-American philosophical tradition the locus classicus on the distinction between primary, secondary, and tertiary qualities is Locke’s Essay Concerning Human Understanding (II.viii). This is, in any case, a 263
TF17557.indb 263
16/9/08 15:47:32
GIORGIO BI A NC ORO S S O
metaphysics codified by Aristotle. Don Ihde, in what remains the most comprehensive and persuasive attempt to address the imbalance between vision and hearing in the field of philosophy, finds an anticipation of the doctrine of primary and secondary qualities in Democritus (1976: 9). Indeed, he traces the “visualism” of Western thought back to the Greeks and argues that its condition, one of philosophy’s “original sins,” is the separating of the senses (7). Within the hierarchy of primary and secondary qualities, sound, like color, is ostensibly secondary, and so the auditory features of a sound are felt to be dispensable with the least loss of recognizability. This is more than a convention of language, according to Metz: it is through language, a condition of the perceptual experience itself. Therefore, Metz continues, “if I perceive a ‘rumble’ without further specification, some mystery or suspense remains (horror and mystery films depend on this effect): the identification is only partial. However, if I perceive ‘thunder’ without giving any attention to its acoustic characteristics, the identification is sufficient” (1980: 26). Much has been made of the “adjectival” nature of sound as a symptom of a preference for the visual realm over the auditory one. One needs to be careful, however, in attributing the scarce attention to sound in film studies to a general inattention to sound in the wider culture. Rick Altman, for instance, has suggested that the reasons for the marginal role of sound in film theory and criticism can be found within the history of film, and film criticism, itself: the desire to distinguish cinema from the theater, the conditions of exhibition in the so-called silent era, and the belief that the medium must be defined in terms of an “essence” (which is inevitably identified with the image or an aspect of image manipulation, like editing) (1980: 12–14). Practical factors have also come into play. Sounds unfold in time and they decay. In written commentaries, they can only be evoked or described rather than quoted. Though aspects of the moving image are also impossible to transpose – think of camera movement – others can be visually quoted or referred to without too obvious a misrepresentation of their original manifestation in a film (image composition, framing, the use of a certain camera lens, the costumes, the profile of an actor, his or her makeup, etc.). Citability and transposability, in other words, have played a major role in what writers and theorists have chosen to write about. More prosaically, but just as importantly, the tedious and enormously time-consuming tasks of arranging a set for shooting, positioning a camera, and achieving the most desirable conditions of light have slanted filmmakers’ own jargon toward the visual. Anyone who has spent a day on a film set can testify to this. That so much effort is spent on capturing images does not mean that they are inherently more significant at the point of reception; still, it almost inevitably entails that film directors, when asked to talk about their craft, dwell primarily on the process of image production. To top it all off, the process of production and post-production appears to recapitulate the change of film from “silent” to “sound,” reinforcing what Altman has dubbed the “ontological fallacy,” namely the belief that cinema is essentially visual and that sound is an afterthought (1980: 14). The bias for the visual in film studies has to some extent been exaggerated, 264
TF17557.indb 264
16/9/08 15:47:32
S OU ND
however. In part, such a belief stems from a more liberal way of parsing linguistically the field of the visual vis-à-vis the auditory realm. No one refers to the notes of a Chopin ballade merely as the “sounds of the piano” or to an actor’s soliloquy as “the sounds of a voice.” It would be foolish to do so, though there is much of value in the process of sound production itself (the quality of the instrument, the acoustics of the venue, the physical efforts of the performer, the fullness of an actor’s voice, and so forth). When it comes to music and language, we tend to equate “sound” with the acoustical. We don’t seem to use the term “image” in the same way. Whether we talk about the optical features or representational content of a film shot we say that we are seeing an “image.” Whether we remember props, facts, characters, or their specifically optical mode of presentation, we say of a film that we remember its “images.” Should one restrict the field of the visual to the optical, just as that of sound is restricted to the acoustical, “images” would be understood as carrying considerably less weight in the process of film appreciation. It is to images as paths of access to people and objects that we pay attention, less so to images as such. After all, in everyday life, what sound is an adjective of – if at all – is not an image, but an object or a space. Heidegger puts it well when he says: Much closer to us than all sensations are the things themselves. We hear the door shut in the house and never hear acoustical sensations or mere sounds. In order to hear a bare sound we have to listen away from things, divert our ear from them, i.e., listen abstractly. (Quoted from Ihde 1976: 26; emphasis mine) With this in mind, let us return to Metz’s discussion of the relationship of a sound to its source. He argues that privileging the source object at the expense of the acoustical characteristics of the sound has shaped the way we talk about offscreen sound in film. Indeed, according to Metz, the very term “offscreen sound” reflects this bias, for strictly speaking a sound cannot be “off”: “[E]ither it is audible or it doesn’t exist . . . The situation is clear: the language used by technicians and studios, without realizing it, conceptualizes sound in a way that makes sense only for the image. We claim that we are talking about sound, but we are actually thinking of the visual image of the sound’s source” (1980: 29).While suggestive, as every phenomenologist should know, the expression “offscreen sound” need not be taken as a transparent window onto the experience, let alone an accurate description of it. It is shorthand for a sonic occurrence that may well turn out to be appreciated and discussed in terms of its acoustical characteristics (despite the expression’s emphasis on the fact that the source is not visualized). Understood descriptively, moreover, it does not suggest we “are actually thinking of the visual image of the sound’s source,” as Metz says; rather, it indicates that it is wrong to divorce sounds from their sources, in pursuit of the “purely acoustical,” to begin with. Whether inherited biologically or motivated culturally, our constant preoccupation with what a sound is a sound of is an established and fundamental component of the listening experience. To sever hearing a sound from knowing where it comes from 265
TF17557.indb 265
16/9/08 15:47:32
GIORGIO BI A NC ORO S S O
and/or seeing its source is to offer a truncated picture of the listening experience; it is to talk about “bare sound” or “abstract listening,” to quote Heidegger. Seeing the source helps project some of its qualities as an object back on to the sound through a process of mutual implication and enrichment; attributes of the lighted object are transferred on to the sound, and vice versa. Even the mere awareness of a source, in the case of offscreen sound, initiates a process of mutual implication (not between image and sound but between a concept, or idea, and the sound). Even in monophonic cinema, the offscreen sound of a crowd is the same sound as an on-screen one, only as far as their physical characteristics are concerned. Because of the inseparability of sound from image, they are, in fact, perceived differently. Faced with both the on-screen and the offscreen sound of a mumbling crowd, in other words, we are not dealing with the same sound, with or without the matching image of the source, but rather with two qualitatively different sounds. The expression “offscreen sound,” then, does justice to the interdependence of seeing and hearing in the auditory experience. So far I have spoken of sound as issuing from a source. Much of what we call sound, however, involves two or more physical objects or elements coming into contact in one way or another. The human voice, too, though seemingly generating from one point deep inside the body of the speaker, is the result of a complexly calibrated mechanism involving tissues, cavities, air, and energy. The fact that sounds are typically described with reference to a single or even simple source betrays the anthropomorphization they undergo in common speech, as if the rattling noise of an engine, like the human voice, came from some metaphorical “inside” rather than being produced by the friction of its mechanical components. It is in this respect that language does fail, for it makes us overlook the extent to which a sound, as Ulric Neisser suggests, informs the listener about an event or chain of events (1976: 155). Taking Neisser’s intuition a step further, I want to suggest that as the product, however incidental, of an event, a sound – itself an event – is heard in terms of its sources because of a deep-seated need to read events causally. Think of how we perceive visible motion as the result of a physical cause, be it collusion, initial propulsion, or pull. Dancers exploit the compelling tendency to interpret events causally, inviting us to see their gestures and motion not only as a bodily equivalent or paraphrase of musical rhythm and melodies but also as their effect. By the same token, the anchoring of a sound to a source shown on a movie screen, or known to be offscreen, betrays a tendency to link a perceivable phenomenon to a cause (as distinct from a mere source). Aristotle famously called the causing agent of physical phenomena their “efficient cause.” The efficient cause is something that produces or brings about something else (its effect). In Aristotelian terms, then, the source is the efficient cause of the sound, and hearing a sound in terms of its source may be understood, after Chion, as a form of “causal listening” (1994: 25–8). One can describe the perception of film sound and music within the context of a more skeptical view of causality as the mere relation between two events. Within this conceptual framework, hearing diegetic sound can be described as nothing more than the forging of an imaginary relation between two (real) events: the playback of a sound and a moving image. Independent of whether one privileges the idea of “efficient causation” or that of “relation,” reference to 266
TF17557.indb 266
16/9/08 15:47:32
S OU ND
causality forces one to recognize that the source of a sound is not only its site of provenance but also that which produces it. Upon listening to a character speak, a spectator is simultaneously constructing two kinds of causal relations, since he or she hears the sound of the actor’s voice as springing from the source in the film’s world, while knowing its physical genesis in the film production and playback processes. Attention is typically focused on the former, rather than on the latter, and for good reasons. Aware of the effortlessness with which audience members create imaginary relations between the objects and people they see on the screen and the synchronized sounds they hear, filmmakers have learned how not to worry about fidelity too much. The result of this state of affairs is a certain degree of stylization in the acoustical construction of the story world even in the most self-describedly “realistic” drama (a revealing foil to this is the use of harshly realistic sounds in cartoons). One reason for the considerable scientific and philosophical interest of blatantly intrusive dubbing, such as one finds in Italian, German, or Hong Kong cinema, is that it offers the illusion of a sumptuous, perceptually vivid impression of a causal relation which is known to be purely imaginary – one that is forced down our throats, in fact. As such, dubbing stands in spectacular contrast to our everyday experience of a great many causal relations that, though known to be scientifically true, cannot be grasped through our senses. See also Definition of “cinema” (Chapter 5), Medium (Chapter 16), Music (Chapter 17), and Phenomenology (Chapter 40).
References Altman, R. (1980) “Introduction,” Yale French Studies: Cinema/Sound 60: 1–15. —— (1985–6) “The Technology of the Voice,” Pts. 1 and 2, Iris 3: 3–20; and 4: 107–18. Altman, R., Jones, M., and Tatroe, S. (2000) “Inventing the Cinema Soundtrack: Hollywood’s Multiplane Sound System,” in J. Buhler, C. Flynn, and D. Neumeyer (eds.) Music and Cinema, Hanover and London: Wesleyan University Press. Branigan, E. (1997) “Sound, Epistemology, Film,” in R. Allen and M. Smith (eds.) Film Theory and Philosophy, New York: Oxford University Press. Bruner, J. (1973) Beyond the Information Given: Studies in the Psychology of Knowing, New York: Norton. Chion, M. (1994) Audio-Vision, trans. C. Gorbman, New York: Columbia University Press. Gibson, J. J. (1966) The Senses Considered as Perceptual Systems, Boston: Houghton Mifflin. Ihde, D. (1976) Listening and Voice: A Phenomenology of Sound, Athens, OH: Ohio University Press. Jost, F. (1989) L’œil-caméra, 2nd ed. rev., Lyons: Presses Universitaires de Lyon. Lastra, J. (2000) Sound Technology and the American Cinema: Perception, Representation, Modernity, New York: Columbia University Press. Levin, T. (1984) “The Acoustic Dimension: Notes on Cinema Sound,” Screen 25: 55–68. Metz, C. (1980) “Aural Objects,” trans. G. Gurrieri, Yale French Studies: Cinema/Sound 60: 24–32. Neisser, U. (1976) Cognition and Reality, San Francisco: W. H. Freeman and Company.
267
TF17557.indb 267
16/9/08 15:47:32
25
S TYLE Noël Carroll The notion of style in cinema plays a number of different roles in different contexts and serves a number of different purposes. In certain contexts, the purpose of the notion of style is to differentiate sets of motion pictures. In this regard, we might speak of the style of a period, a school, a movement, a genre, or even a nation. Likewise we might use the notion of style to differentiate one director in contrast to another – for example, to mark what distinguishes Jean Renoir from other directors, such as Sergei Eisenstein. But the goal of stylistic analysis may not be descriptive and/or classificatory in the preceding sense. It may be explanatory. That is, we wish to explain why an individual motion picture is the way it is: why it has the elements it does and why they stand in the relations that they do. In this light, we are asking for an explanation of how the motion picture hangs together. Sometimes this sort of analysis might also be called close stylistic analysis or formal analysis. It is a matter of the analysis of the form of the motion picture – an explanation of the way in which the movie embodies its point or purpose.
Differentiating concepts of style One view of style regards it as essentially a matter of differentiating works of art (Goodman 1978). Several of the prominent uses of the concept of style among cinema scholars aim at contrasting one group of motion pictures from other groups. Some of the leading kinds of cinematic groups that commentators have sought to distinguish include schools and movements (like Italian Neo-Realism, the French New Wave, Structural Film, Dogme 95 Film), period motion picture making (the International Style of Silent Filmmaking), genre style (slasher films, spy thrillers, musicals), national styles (Hong Kong Cinema, Bollywood), and personal or individual oeuvres (usually the body of work of a director, but one might also focus upon the work of a screenwriter, a cinematographer, a set designer, costumer, special effects artist, and so on). One thing that the aforesaid style groupings have in common is that their object typically involves groups of motion pictures greater than one (I say “typically,” since there are examples of single film directors like Charles Laughton or Udi Shankar).
TF17557.indb 268
16/9/08 15:47:32
S T Y LE
In addition, however, there are also stylistic descriptions of a general sort that operate upon the totality of motion pictures, dividing the entire corpus of cinema into different categories. For example, we may refer to certain films as surrealistic (with a small s as opposed to Surrealistic with a capital S, which is the label of a periodspecific movement) in virtue of the oneiric shot transitions they exploit – as in the case of Busby Berkeley’s dance numbers in The Gang’s All Here (1943). We can call this kind of stylistic description universal because we can divide the totality of motion pictures into “surrealist” and “not surrealist” in this sense. That is, for any motion picture, speaking globally and transhistorically, it will fall into this category or not. Similarly, we may call a shot “classical,” if it is symmetrical or balanced, as are many of the compositions in Lang’s Siegfried’s Death (1924). Such stylistic descriptions are compatible with other stylistic descriptions because these other descriptions are being made relative to a more restricted category. So the compositions in Lang’s Siegfried are classical in terms of one dimension of universal categorization, while the film is also an example of German Expressionist Cinema from the perspective of its membership in a specific school, movement, or tendency. Moreover, some of the aforementioned categories may overlap. A fifties’ musical belongs both to a genre grouping and a period-specific category. One thing to iterate here is that, although a motion picture may belong to a universal stylistic class, like that of the classical, it is also likely to belong to other categories of lesser generality, such as German Expressionism. We can call these latter categories group-stylistic categories, since they usually embrace a lower order of generality. Furthermore, many of the group categories mentioned above may intersect. For example, a thirties gangster film belongs to the genre category of the gangster film, while also inhabiting a class of period- and place-specific films. Although I admit that it sounds odd to classify the oeuvre of a single moviemaker, like Ida Lupino, under the rubric of a group category, nevertheless the body of work of such a director will be standardly made up of a group of more than one film. In determining the style of a specific period, we concentrate on the similarities between the motion pictures under consideration. For example, we might describe the style of Soviet expressive realism of the twenties in terms of the recurring use of montage, naturalistic photography (with frequent use of low-key lighting and soft focus), epic compositions (often exploiting low-angle setups) and a tendency to favor the use of mass heroes – the collectivity as the protagonist (Huaco 1965). In addition, it is useful to clarify the profile of a period-specific style by contrasting it with other period- and place-specific styles, as Bazin does when he contrasts the montage style of Soviet Expressive Realism with the deep-focus realism of the sound cinema (Bazin 1967). Since stylistic classifications are relative to different group categories, the same body of work may call for different directions of emphasis. For example, in discussing the work of Buster Keaton as a silent comic filmmaker, we notice features of his manner of filmmaking that correspond with the work of other directors of the period, like Chaplin. However, when analyzing Keaton’s personal style, we look at what differentiates Keaton from his peers. Thus, the characterization of Keaton’s personal style is 269
TF17557.indb 269
16/9/08 15:47:33
NOËL CA R ROLL
likely to exclude a number of the features that comprise the characterization of Keaton as an example of a silent, comic film director. Generally, a motion picture maker possesses a generic period style tacitly. It is presupposed. For example, Vincent Sherman did not self-consciously decide to adopt a thirties’ studio style when he came to make his first film, The Return of Dr X (1939). He found the style readymade, so to speak. Although typically period- and place-specific style is tacit, school or movement style is more a matter of conscious choice. Buñuel and Dali were quite deliberate in their adoption of a Surrealist (with a capital S) style of filmmaking in Un chien andalou (1929), extrapolating, as they did, from the disjunctive imagery of Surrealist poetry and then developing a comparable mode of abrupt editing in which sequences appear to be conjoined for no detectable reason. Among the varieties of group style is genre style. Many motion pictures fall into certain categories with relatively fixed though still variable purposes, whose implementation most often involves recourse to certain reliable, recurring strategies for getting the job done. For example, comedies tend to be lit evenly and brightly in contrast to horror movies which tend to be lit darkly, often exploiting chiaroscuro effects. That is, genres come with certain enduring stylistic tendencies, and the analyst can use those tendencies to differentiate one genre from another. Personal style differs from the other branches of group style canvassed so far in that it targets the work of specific motion picture artists. Although directorial style is the predominant focus, to date, of most cinema scholarship concerned with personal style, as indicated earlier, there is no reason why it cannot take up the oeuvres of screenwriters, cinematographers, etc. The preoccupation with the personal style of directors is often referred to as auteurism (Sarris 1985). Its influence since at least the 1960s has been enormous. As a result, it is perhaps the most common form of stylistic analysis currently practiced in cinema studies. The goal of analyses of this sort is to pith what makes unique the characteristic approach of (usually) the motion picture director to his/her materials. This requires that the cinema analyst home-in on what differentiates how one director constructs his/her imagery from how others do it. Frequently, this involves pairwise comparisons, contrasting, for example, the monumentalism and architectonic nature of Stanley Kubrick’s handling of the deep-focus shot with the more fluid, relaxed, and quotidian approach adopted by Robert Altman. In this, the motion picture analyst is searching for something like the “signature” of the auteur – the ways in which the motion picture maker’s way of doing things sets him/her off from the rest. In this regard, Ozu’s low-angle compositions are an important part of his signature, as are Martin Scorsese’s rapid dolly zooms into close-up. Since there are different orders of style – from the universal to the personal – the possibility for confusion is ripe. Indeed, analysts risk talking past each other, if they speak carelessly about motion picture style, simpliciter. Instead it is advisable that when speaking of style in cinema, analysts qualify their domain of inquiry – that they signal whether they are concerned with universal style or group style, and if group style, whether they are speaking of a generic period style, school or movement style, genre 270
TF17557.indb 270
16/9/08 15:47:33
S T Y LE
style, or individual style. Although this list of styles is not necessarily exhaustive, nor are the categories necessarily mutually exclusive, these classifications often serve a congruent taxonomic function, namely to describe bodies of motion pictures in such a way that we can differentiate certain groups of movies from others. Such categories are indispensable for connoisseurship. However, to make matters even more complex, in addition there are forms of stylistic analysis that serve a different purpose than differentiation. Rather than aspiring to differentiate descriptively various groups of motion pictures from each other, a very central form of stylistic analysis is devoted to explaining why individual moving pictures have been put together as they have. This is a matter of attempting to identify and account for the form or style of the individual motion picture.
The concept of style in the individual work Although a great deal of stylistic analysis is dedicated to differentiating groups of motion pictures from each other, stylistic analysis can also be directed at individual motion pictures and/or parts thereof. This type of analysis constitutes a large amount of what students are taught to do in introductory cinema appreciation classes. Style is generally thought of as the way in which something is done. There is the action and the manner of its execution (Lang 1998). On this construal, the stylistic analysis of a motion picture addresses how a motion picture or a part thereof was constructed or formed, not in terms of its technological provenance (or material cause, as Aristotle would put it), but in terms of its design. In this respect, stylistic analysis is an interrogation of the form of the motion picture. Unfortunately, this observation hardly clarifies the nature of the stylistic analysis of the individual motion picture, unless we can say something about the way in which the notion of form should be understood. That is, such stylistic analysis takes the form of the moving picture as its object of inquiry. But what is form? A common approach to thinking about form, including cinematic form, is to contrast it with content. A motion picture has a content, something it is about, and the form of the motion picture is the way in which that content is articulated or embodied. This view of content, however, is inadequate because it is insufficiently comprehensive. For not all motion pictures have content in the requisite sense. Some films may not be “about” anything. For instance, a flicker film may only be aimed at provoking a certain perceptual experience and not have any subject matter. It is not about a certain perceptual experience: the film is what makes the perceptual experience happen. Nevertheless, I suppose, we would refrain from asserting that a flicker film like this lacked form. Another popular way of thinking of form, in contrast to content, is to think of the form of a motion picture as that which contains the content of the movie after the fashion in which a bottle contains and gives shape to the liquid within it. However, this analogy raises significant problems when we attempt to extend it to motion pictures. It becomes difficult, for example, to determine whether or not we should 271
TF17557.indb 271
16/9/08 15:47:33
NOËL CA R ROLL
call certain elements of a movie instances of the form or the content of the work in question. Consider the point of view of a motion picture. Surely, in ordinary speech, it is natural to regard the point of view of a movie as part of its content. For generally it expresses an attitude toward whatever the movie or sequence thereof is about. And yet, it seems equally compelling to say that the content of a movie is both contained within and organized by its point of view. So the point of view of a motion picture would appear to count as an aspect both of the form and of the content of the work. This may be a view that certain theorists would be willing to endorse enthusiastically. But, be that as it may, such a view by dissolving the distinction from content gets us nowhere in terms of understanding the pertinent concept of form. We need another concept of motion picture form, one not hostage to the concept of content. Returning to ordinary language for inspiration, one notes that two of the most frequent ways of discussing artistic form is to describe it in terms of unity and complexity (aka variety). Both these concepts imply that form is something that has parts. Obviously, something cannot be complex, unless it is made up of at least two diverse elements. To be complex is the contrary of simple – to be comprised of some single stuff. Likewise, to be unified, as opposed to being simple, requires parts that are related in a way that is organized or coordinated, as the recurring motif of clocks serves to reinforce visually the tight-knit structure of the flow of events in The Set-Up (1949). Indeed, most narrative motion pictures organize their diverse episodes in terms of an overarching unity of action through which the protagonists, across a varied series of events, discover a solution to some abiding problem, such as defeating Godzilla or winning the hands of seven brides. Clearly the concepts of unity and complexity are complementary. If something is complex, as opposed to chaotic, the parts in question must be interrelated in some way, while, reciprocally, unity requires parts to unify. Therefore, since perhaps the most fundamental ways of characterizing form involve the concepts of unity and complexity, it seems reasonable to hypothesize that cinematic form has as among its basic ingredients parts and their relations. That is, when in the course of a stylistic analysis of a moving picture we make statements about its form, our statements are generally about the relations between parts of the movie in question. When we say of the restaurant scene in Tati’s Playtime (1967) that the composition of the comic incidents (the parts) is decentered, we are talking about the way in which they have been organized (intentionally interrelated by means of juxtaposition) such that none then has perceptual salience, thereby impelling the eye of the spectator to explore the frame in search of them. In general, it seems reasonable to conjecture that when we offer observations about the form or style of an individual motion picture, we are making statements about the relation or relations between its parts. For the style of the moving picture is a matter of how the parts are put together, or formed. Form statements, then, appear to be ultimately translatable into statements such as “x bears such and such a relation to y” where x and y are parts, of some level of generality, of the motion picture. Even when the relation between the relevant parts is not made explicit, it can usually be fleshed out in terms of talk of parts and their relations. 272
TF17557.indb 272
16/9/08 15:47:33
S T Y LE
If a motion picture’s style is described as “jarring,” then that implies that its parts are related discordantly, where, of course, this may be a commendable or unfortunate feature, depending on the goals of the moving picture. Here it is important to keep in mind that motion pictures have many different kinds of elements and that these elements can stand in many different relationships. Sounds may repeat, functioning as leitmotifs, like the whistle of the child-killer in Lang’s M (1931). Characters can stand in adversarial relationships to each other; this is a matter of dramatic conflict, which is a typical formal feature in narrative movies. Objects and their apparent scale can be juxtaposed to each other – like the gigantic kulak and the miniscule Marfa in Eisenstein’s The Old and the New (1929). This too is a formal relation, as is the equilibrium or disequilibrium of the objects in a shot. Similarly, sequences can contrast with each other in terms of fast-cutting and slowcutting, and this can supply the basis for a formal structure. The form of an individual motion picture or its style, ex hypothesi, consists in how its various parts relate to each other. Motion pictures may have different parts related in different ways. Some of these ways may be coordinated, such as the way that most of, especially, the main characters are related to the plot in most narrative films. Or, the elements may be relatively uncoordinated. The color elements of a studio decor, though related to each other, may not have any significant relation to the dramatic conflict in the narrative. But regardless of whether the sets of relations in a motion picture are hierarchically organized, all of the relations are formal relations, and the style of the individual moving picture, it may be claimed, is a matter of how the parts of the film relate to each other. This idea, then, suggests a conception of stylistic analysis almost naturally. The style or form of a motion picture is the sum total of its formal relations – the sum total of the ways in which all of the parts of the movie relate to each other and gather into further webs of relationships. Such is the object of stylistic analysis. Consequently, stylistic or formal analysis of an individual motion picture involves tracking all of the ways in which the parts of the movie are related to each other, and following the ways in which those relations become parts in larger networks of relationships. This is a very liberal view of cinematic form or style. The way in which any part of a motion picture is related to another part is an instance of cinematic form on this view. Thus, any motion picture that has discriminable parts related in some way or other will have form of the sort that is subject to this kind of stylistic analysis. This approach might be called the descriptive account of form or style in the individual motion picture. According to the descriptive account, any instance of a relation among elements of an individual movie is an instance of cinematic form. On this conception, in order to provide a full account of the form of a given motion picture, one would list or summarize all of the ways in which the parts or elements of the work are related. This approach can be called descriptive because it classifies any relation among elements of a work as an instance of cinematic form, irrespective of any principle of selection. On this view, the ideal stylistic analysis of a movie would be a long list of descriptions of all of the ways in which the elements of a given motion picture were related. Indeed, 273
TF17557.indb 273
16/9/08 15:47:33
NOËL CA R ROLL
certain once popular strategies of cinema analysis, such as Raymond Bellour’s notion of segmentation, have actually converged on the descriptive analysis of motion picture form (Bellour 1976). An admirable feature of the descriptive approach to stylistic analysis is its comprehensiveness; it doesn’t leave anything out. Nevertheless, I do not believe that the descriptive account delivers what we expect from an adequate approach to the stylistic analysis of the individual motion picture. For, the descriptive account is impracticable, precisely because its conception of form (the object of stylistic analysis) is, in effect, too broad. Obviously, we rarely, if ever, encounter anything as exhaustive by way of stylistic analyses as what we would expect if the descriptive account sketched above crystallized our ruling conception of form or style in the individual work. Nor does the lack of such accounts seem purely accidental, or a function of the fact that virtually no one would have the energy to draw up such an account, let alone read one. For, more importantly, the aforesaid descriptive account does not seem as though it offers what we want from a stylistic analysis. For example, we want our stylistic analyses to be selective, whereas it is a mark of the descriptive account that it is nonselective. But, furthermore, we do not want stylistic analyses to be selective because we are lazy. We want them to be selective in order to help us to understand and appreciate the motion picture at hand. A descriptive account of motion picture form is rather like the fantastic map that Borges imagines which is of the exact scale of its referent, while containing every one of its referent’s details. Such a map is useless for guiding us through the territory since it is a mirror of the territory. For, if one cannot find one’s way on the ground, then a perfect replica will be of no help. Likewise, the descriptive account of cinematic form is more like a duplicate of a motion picture’s form than a guide to what is significant for understanding and appreciating the movie. And the reason for this is that it is not sufficiently selective. Surely it seems promising to construe cinematic form in terms of how the motion picture’s parts are related to each other. Yet not all of the relations between the elements of the motion picture are equal. Some are more important than others for the purpose of delivering an understanding of cinematic form. The trick is in determining which ones those are.
Form and function (Carroll 1999, 2003) The descriptive account of cinematic form is inclusive and incorporative. It regards every way in which the elements of a motion picture relate to other elements as part of the form of the work. In this matter, it privileges no relations over others; it is – or would be – purely descriptive. However, this does not appear to be an apt characterization of our actual practices of stylistic analysis, since, in the ordinary run of things, we expect more of such analyses. Indeed, we expect them to be analyses – that is, to have an explanatory dimension. To this end, we are not interested in all the elements of the motion picture and their corresponding relata, but only those that contribute 274
TF17557.indb 274
16/9/08 15:47:33
S T Y LE
to the realization of the point(s) and purpose(s) of the individual work. For, it is just these features that will advance our understanding of the motion picture as a whole. From this perspective, the formal elements and relations in the motion picture that are pertinent for stylistic analysis are those conducive to the realization of the point(s) and purpose(s) of the motion picture. The form of the moving picture, that is, is whatever functions to bring about whatever the movie is intended to achieve. The form of the moving picture is functional; the form of the movie is ideally determined by what the motion picture is supposed to do. For example, in The Lives of Others (2006) there is an attempt to portray the bleakness of living conditions under the former communist regime in East Germany. To underscore viscerally the drabness of it all, the filmmakers shot scenes in which bright-colored objects were systemically removed. As a result, the screen is often continuously dominated by sequences of what feels like monotonous, institutional, unrelenting grayness – the expressive analog on the moviemaker’s palette for existential hollowness. This way of articulating the color scheme of the The Lives of Others is a formal or stylistic choice, i.e., a choice designed to reinforce the point or purpose of the motion picture, which in this case amounts to the theme of the repressive tedium and life-crushing program of the German Democratic Republic. This account of cinematic form can be labeled the functional account. According to the functional account of cinematic form, the form of an individual motion picture is the ensemble of choices intended to realize the point(s) or purpose(s) of the motion picture. Here the notion is that form or style follows function. This approach, of course, assumes that motion pictures have points or purposes. Yet this assumption appears scarcely controversial, once one realizes how very diverse those purposes can be. In some cases, the purpose of a movie may be to propose a theme or point of view, or the purpose may be to foreground an expressive property, or it may be to arouse feelings, including feelings of visual pleasure, in audiences. A motion picture may be about communicating ideas – including ideas about the world or about the nature of film – or it may have no ideas or meanings, but might be merely devoted to promoting a certain sort of experience, such as repose, excitement, suspense, or perceptual delight. Motion pictures can make points, or they may simply have points – for instance, to encourage viewers to use their discriminatory powers keenly and perceptively. It should not be hard to concede that all or nearly all motion pictures have points or purposes – probably, in most cases, more than one – once we think of points and purposes in this broad way. The form or style of the movie comprises the ways in which the work realizes its purposes. And, in this regard, cinematic structure follows function. In The General (1927), for example, Buster Keaton is committed to enabling the audience to seeing exactly how many of the large-scale physical events in the film – such as the decoupling, derailing, and swerving of railroad cars – transpire causally. That is one of the overarching purposes of the film. To this end, Keaton not only deploys highly determinate, deep-focus long shots, with internally pronounced perceptual pathways (such as uniformly articulated railroad tracks), but then rotates the array by means of long-shot field reversals to ensure that the viewer grasps all of the pertinent variables in the scene and their spatial and causal interrelations. 275
TF17557.indb 275
16/9/08 15:47:33
NOËL CA R ROLL
In the famous mortar sequence, he visually tracks the artillery piece from every side, shooting from in front of it and behind it, so that we see not only its probable trajectory, but, then, suddenly the curve in the roadbed that pulls Buster’s locomotive out of the line of fire and saves him from being blown to smithereens. The choice here of long-shot field reversals not only communicates the information that Buster has been saved, but shows us how causally this came about (Carroll 2007). Some commentators might be hesitant to call Keaton’s directorial decision in the previous example a choice (Wollheim 1979). They reason that a choice involves the possibility that things might have been otherwise, but given Keaton’s aims and his already established stylistic proclivities it would have been unlikely that he would have approached the mortar scene differently. Nevertheless, it still makes sense to call his decision a choice insofar as a director working in 1926 on a sequence like the mortar scene would have had a number of different ways of shooting it. One (though perhaps not the actual Keaton) could have done it in close-ups, as a Soviet filmmaker might have, albeit with different results. Call these alternative approaches a repertoire (Ross 2003). It is relative to such a repertoire that Keaton’s decision counts as a choice – the decision to opt for one approach, out of a repertoire of contextually available alternatives, rather than others. The form or style of the individual motion picture consists of the collection of formal choices enabling the realization of the points and purposes of the moving picture. Movies, needless to say, may have more than one point or purpose, and these may or may not be coordinated. Sometimes formal choices reinforce each other, often hierarchically, although sometimes they may not be related at all. Furthermore, formal choices themselves may have more than one function in a given motion picture. Cinematographer Dariusz Wolski’s chalky/charcoal, black-and-white palette in Tim Burton’s Sweeney Todd (2007) functions to underscore the griminess of London, and the moribund aspects of Sweeney Todd (Todd/Todt) and Mrs Lovat, while providing an eminently effective backdrop, colorwise, for all the gushing, red arterial explosions issuing from Sweeney’s razors. Generally, compared with the descriptive account, the functional account will regard fewer of the relations between elements of the motion picture as belonging to the form or style of the movie, since usually not all of the possible relations between elements contribute to the realization of its point or purpose. Of course, if all of the relations contributed to the point or purpose, then the functional account will take note of them all. But this will be the exception rather than the rule, since such cases are genuinely extraordinary. Typically, only some of the relations in the movie will function to advance the purpose of the work, and it is upon these that functional analysis concentrates. Moreover, insofar as these relations are understood in terms of the functions being discharged, this sort of analysis does not merely describe the motion picture but explains why it is the way it is, thereby augmenting our comprehension of the work at hand. Similarly, the functional conception of cinematic form is superior to the notion of form as a container of content. For in speaking of form or style in terms of the intended realization of points or purposes, we are not committed to attributing 276
TF17557.indb 276
16/9/08 15:47:33
S T Y LE
meaningful content to a motion picture. A motion picture may not mean anything, but nevertheless do something, like stimulating a certain perceptual experience. Thus, the functional account is more comprehensive than the view that sees form as correlative to content, just because the functional account provides a way to discuss the forms of contentless motion pictures. Style in an individual moving picture is a matter of form – a question of how the points or purposes of the work are implemented, manifested, or embodied. Something belongs to the form or style of the work if it belongs to the ensemble of choices intended to realize the point or purpose of the motion picture. A formal or stylistic choice has the intended function to realize the point or purpose of the movie, if the point in question is the intended result of the formal choice and if the formal choice occurs in the work in order to secure the point or purpose of the motion picture. In this respect, the form or style of the work is regarded as generative – as bringing to fruition the point or purpose of the work. Stylistic analysis, in this way, contributes to the explanation of how the motion picture achieves its points and purposes. This, of course, involves having a sense of the point or purpose of the work, a project that often involves interpretation, where the point of the work is to make meaning.
The explanatory concept of style with regard to group style Although when applied to groups of motion pictures, the concept of style generally performs the function of differentiating descriptively one kind of movie from others, groups of motion pictures can be characterized functionally wherever and to the extent that it is feasible to assign points or purposes to the groups in question. If, for example, one can correlate an enduring project with a motion picture maker throughout his career, then it may be possible to identify the generative strategies he employs to those ends (along with their evolution over time) over his oeuvre, thereby isolating his style. Stan Brakhage was, for instance, consistently engaged in championing alternative modes of perception and, to that end, he employed a panoply of techniques that subverted perspectival imagery, which he associated with conventional (and conventionally imposed) vision. Thus, his emphasis upon two-dimensional (such as perforations and scratches on an opaque background) and out-of-focus imagery were crucial to his generative style – to the realization of his point, the valorization of unacknowledged or underacknowledged realms of perception. Attending to these stylistic choices, in turn, helps viewers to understand and to appreciate Brakhage’s production. Or, for a more mainstream example, consider the way in which Otto Preminger’s roving camera movements, fluidly tracking scenes from every angle and thereby refraining from privileging the viewpoint of any character over others, serve to express his objective stance to the world of his fictions. Similarly, if one can establish a point or purpose of a group of motion pictures – such as those that comprise a movement or school or genre, one can begin to suggest a functional account of the style of that group. This account of style will be explanatory, since the analysis of the relevant style explains how the works in question succeed in 277
TF17557.indb 277
16/9/08 15:47:33
NOËL CA R ROLL
realizing the points or purposes of the group. Generative accounts of style like this may not coincide with differentiating accounts of style for the same group of motion pictures, since the elements of style that explain how a group of motion pictures achieve their ends may not differentiate that group from other groups. For example, strategies of suspense are integral, generatively speaking, to most science fiction movies, but such structures of suspense do not differentiate science-fiction movies from other popular genres. Of course, one may upon occasion succeed in marrying the differentiating approach to style in groups of motion pictures to the explanatory conception by focusing on the intersection of the generative or explanatory stylistic choices that, at the same time, differentiate the group of movies in question from the relevant comparison classes. Nevertheless, with sufficiently large and varied groups of motion pictures – like national cinemas – there may not be enough sharing of points or purposes among the group to allow for the application of the explanatory concept of style to the group. See also David Bordwell (Chapter 29).
References Bellour, R. (1976) “Segmenting/Analyzing,” Quarterly Review of Film Studies 1: 331–53. Reprinted in P. Rosen (ed.) (1986) Narrative, Apparatus, Ideology, New York: Columbia University Press. Carroll, N. (1999) Philosophy of Art: A Contemporary Introduction, London: Routledge. —— (2003) “Film Form: An Argument for a Functional Theory of Style in the Individual Film,” Engaging the Moving Image, New Haven, CT: Yale University Press. —— (2007) Comedy Incarnate: Buster Keaton, Physical Humor, and Bodily Coping, Oxford: Blackwell. Goodman, N. (1978) “The Status of Style,” in Ways of Worldmaking, Indianapolis: Hackett. Huaco, G. A. (1965) The Sociology of Film Art, New York: Basic Books. Lang, B (ed.) (1979) The Concept of Style, Ithaca, NY: Cornell University Press. —— (1998) “Style,” in Michael Kelly (ed.) Encyclopedia of Aesthetics, vol. 4, Oxford: Oxford University Press. Ross, S. (2003) “Style,” in J. Levinson (ed.) The Oxford Handbook of Aesthetics, Oxford: Oxford University Press. Sarris, A. (1985) “Notes on the Auteur Theory in 1962,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism, 5th ed., Oxford: Oxford University Press. Wollheim, R. (1979) “Pictorial Style: Two Views,” in B. Lang, The Concept of Style, Ithaca, NY: Cornell University Press.
Further reading The following are further reading on this topic: James Ackerman, “Style,” in M. Philipson (ed.) Aesthetics Today (Cleveland, OH: World Publishing, 1961); André Bazin, What Is Cinema?, trans. Hugh Gray, vol. 1 (Berkeley: University of California Press, 1967); David Bordwell, On the History of Film Style (Cambridge, MA: Harvard University Press, 1997), together with his edited special issue of Style, vol. 32, on cinema, and his Poetics of Cinema (London: Routledge, 2007). See also David Bordwell and Kristin Thompson, Film Art: An Introduction, 8th ed. (New York: McGraw-Hill, 2006); E. H. Gombrich, “Style,” in D. L. Sills (ed.) International Encyclopedia of the Social Sciences (New York: Macmillan, 1968); A. Harrison, “Style,” in D. Cooper (ed.) A Companion to Aesthetics (Oxford: Blackwell, 1992); and Jenefer Robinson, “Style and Significance in Art History,” Journal of Aesthetics and Art Criticism 40 (1981): 8–14, “General and Individual Style in Literature,” Journal of Aesthetics and Art Criticism 43 (1984): 5–14, and “Style and Personality in the Literary Work,” Philosophical Review 94 (1985): 227–47.
278
TF17557.indb 278
16/9/08 15:47:33
26
VIOLENC E Stephen Prince Depictions of violence have been an enduring feature of motion pictures since the inception of the medium. Violence has drawn filmmakers to explore its potential for cinematic stylization, and its presence on screen has fueled efforts to censor and regulate the medium. Movie violence has attracted audiences so deeply and incessantly that one might argue with much justification that depictions of violence constitute one of cinema’s essential pleasures and appeals for viewers. Indeed, violent genres, such as westerns, war and gangster movies, emerged very early in cinema and have been with us ever since. Director Brian De Palma (Dressed to Kill [1980], The Untouchables [1987]) observed that cinema, as a medium emphasizing movement and action, is inescapably drawn toward violent action: Motion pictures are a kinetic art form; you’re dealing with motion and sometimes that can be violent motion. There are very few art forms that let you deal with things in motion and that’s why Westerns and chases and shoot-outs crop up in film. (Pally 1984: 14) Because of its enduring presence throughout film history, movie violence has attracted the interest of scholars from diverse disciplines who have written substantial works about its psychological attractions for viewers (Bok 1998; Cerulo 1998; Goldstein 1998), its historical manifestation in the films of particular eras (Prince 2003), its economic functions in a modern media economy (Hamilton 1998), its prevalence in particular genres (Clover 1992; Lichtenfield 2007), and its symbolic functions in contemporary visual culture (Prince 2000; Sharrett 1999; Slocum 2001). Much scholarly attention has focused on questions about the effects of viewing violent films, and anxieties about potential harmful effects have helped to fuel the controversies attaching to movie violence throughout the decades. Fears that viewing violence on screen may lead some viewers to behave violently in real life have motivated the long history of censoring screen violence. In this essay, I will trace some of that history and the assumptions about movie violence that lay behind it, focusing on the United States during Hollywood’s Production Code era, which lasted until the 1960s, and the post-Code era that followed.
TF17557.indb 279
16/9/08 15:47:33
S T EPH E N PR I NC E
The effects of watching violence in film and on television have been extensively studied in the social scientific literature. More than two hundred studies, involving over 43,000 individuals as subjects, point to a link between media violence and aggressive attitudes and behavior in susceptible individuals, generally those who already harbor violent fantasies or have experienced abusive upbringings or have difficulty identifying the boundaries between real life and fictional media worlds (Paik and Comstock 1994: 516–46). Individuals predisposed to act violently are more likely to immerse themselves in violent films, most of which portray violence as an effective means of problem-solving and as being a behavior that does not cause lasting negative consequences in the form of bleeding, bruising, broken bones, emotional suffering, and so forth (McArthur et al. 2000: 164–8). The bulk of the violence shown in films and television is sanitized in this way; injury or death is quick and painless. For psychologists who work from a social learning orientation, such media content carries the lesson for viewers that violence works as an effective means of removing obstacles to one’s goals (Bandura 1973). Paradoxically, violence sanitized of its overly negative components is the type often fingered by empirical research as offering the greatest potential harm: The contextual web of realistic, serious, painful action that surrounds graphic portrayals of violence serves both to outrage viewers to complain about these portrayals and at the same time protects them from negative effects, especially of disinhibition and desensitization. In contrast, it is the nongraphic violence that is surrounded by the much more antisocial web of context. While viewers are much less outraged by this “other” violence – which is much more prevalent on television – they are much more at risk of learning that violence is fun, successful, and non-harmful. (Potter and Smith 2000: 319) The weight of the social scientific evidence tends to undercut the predominant critical framework that has infused popular thinking about violence in the movies, namely, that watching such violence produces a cathartic experience. There is little empirical evidence in support of this idea. The concept of catharsis derives from Aristotle’s Poetics where, in a discussion of tragedy, he observed that by means of language and performance, a tragic play induces pity and fear in the spectator, emotions which the drama then effectively purges, a process denoted by the term “catharsis.” Aristotle, however, was not talking about violence but about the emotions induced by tragedy as a theatrical genre. Nevertheless, although classical Greek drama contained little explicit violence on stage, the idea of catharsis has been absorbed into popular thinking about cinema, where it has furnished a template for thinking about the effects of movie violence. Director Sam Peckinpah (The Wild Bunch [1969]), for example, believed that the graphic bloodshed he put on screen would act as a safety valve for society, enabling viewers to work off their aggressions in the comfort of the imaginary screen world. Peckinpah, in turn, was influenced by Aristotle and by Antonin Artaud’s “theatre of cruelty.” In his influential essays on the place of cruel spectacle in theater, Artaud wrote that such dramatizations would be healthy; that 280
TF17557.indb 280
16/9/08 15:47:33
V IOLE NC E
they would drain social abscesses. “I defy any spectator,” Artaud wrote, “to whom such violent scenes will have transferred their blood . . . to give himself up, once outside the theater, to ideas of war, riot, and blatant murder” (Artaud 1958: 82). However, speaking as a social scientist, George Comstock observed that with respect to media violence: the catharsis hypothesis is often wrongly attributed to Aristotle, but in fact he proposed only that by arousing pity and fear, the dramatic genre of tragedy would lead to their catharsis. He said nothing about aggressive behavior, and he was prescient not to do so. (Comstock 1980: 130) If there is scant empirical support for the idea that movie violence works in terms of catharsis, there is also no simple, one-to-one correspondence between screen violence and the behavior of viewers. Connections or linkages between the screen-world and real-world violence are complex, are not easy to disentangle from other variables, and are never without many forms of complex mediation. Despite this complexity, worries about the formative influences of screen violence are as old as cinema. Modern viewers looking at old movies are unlikely to feel that there is much violence in them, and yet audiences and critics in each historical period have found cinema to be excessive in its presentation of violence. Upholding Illinois state censorship of the western, The Deadwood Coach (1924), for example, an appellate judge condemned what he clearly felt was the film’s excessive violence: The picture portrays, first, a killing, then a fight with the Indians and a stagecoach holdup, and an attempt to kill, then the shooting up of some kind of eating house, and a diving from a window, then a holdup of the Deadwood coach and its destruction, then a killing of the guard, the driver being beaten and tied to a tree, then an arrest, then a breaking of the jail by the roughest element of the town, then a release of a prisoner, and, finally, a so-called desperate fight to hold up the stage, then an attempt to escape, and, finally, a man plunges 1,000 feet to his death on the rocks. (Prince 2003: 18) The classic Universal monster movies – Dracula (1931) and Frankenstein (1931) – may seem quite tame today, and yet, in their period, they were widely condemned for being unacceptably gruesome and horrific. State and municipal censor boards around the country made numerous cuts in Frankenstein, and these boards continued to delete shots and scenes from subsequent horror films, as well as from gangster movies. The latter were decried for having the excessive levels of killing that Tommy guns made possible and for showing the deaths of law-enforcement officers. When Dracula and Frankenstein were re-released as a double-feature in 1938, teachers complained that the films left those schoolchildren who had seen them agitated with fear and anxiety. One teacher wrote a letter of protest to the Hollywood industry’s leading studios and distributors complaining that such films offered viewers an opportunity to feast on horror and murder (Prince 2003: 63–4). These examples suggest that violence on 281
TF17557.indb 281
16/9/08 15:47:33
S T EPH E N PR I NC E
screen is governed by the representational conventions that operate within a given period and in relation to the sensibilities of viewers about the nature of what is real and where the thresholds lie that determine when a representation becomes excessive. Conventions for depicting violence change over time, as do assessments by audiences about what counts as real and convincing. Dracula and Frankenstein were excessive films in their day, just as Saving Private Ryan (1998) and Hostel (2005) have been in more recent periods. Violence on the screen has two components, the depicted behavior and the stylistic means through which it is presented. The behaviors themselves have not changed awfully much over the course of film history. Characters have been beating on, shooting, stabbing, strangling, and mangling one another since film began. But the stylistic designs that filmmakers have employed to present these behaviors have changed considerably in ways that make violent behavior on the screen today more insistent and emphatic. A brief survey of changing conventions in screen violence can indicate how boundaries between acceptable and unacceptable content have shifted. First, an irony needs noting. Screen violence as a concept did not exist in earlier periods. “Violence” as a category of screen content emerges as a perceived phenomenon in the late 1960s, with the fall of screen censorship and the popularity of films that would have been deemed transgressive in an earlier period. Popular violent, sixties films that could not have been filmed in prior decades include Bonnie and Clyde (1967), The Good, the Bad and the Ugly (1967), The Dirty Dozen (1967), and The Wild Bunch. Before the 1960s and the greater creative freedoms for filmmakers which that period ushered in, the film industry had been very concerned about depictions of violence on screen, but this term never appears in memorandums about industry regulation or in the Production Code, which governed Hollywood filmmaking from 1932 to 1968. Instead of “violence” as a discrete category, Hollywood’s in-house regulatory board, the Production Code Administration (PCA), used the terms “brutality” and “gruesome” to describe content that was considered excessive and in need of trimming for the screen. The PCA derived its understanding about what counted as unacceptably brutal content from its study of the kinds of scenes that regional censors had been deleting from films during the 1920s. These included knife violence and scenes of sharp weapons such as arrows or spears sticking in people; shootings in which law officers were killed or in which guns were fired at an excessive rate; beatings that were prolonged or that resulted in visible damage to the victim; scenes of torture, dismemberment or mutilation; and scenes in which the victim was shown to be suffering. The PCA worked with filmmakers to reduce this type of content in Hollywood films. As a result, an elaborately coded visual system of representation emerged in which the camera averted its gaze from brutal violence or in which a symbolic substitute was offered for imagery that could not be shown explicitly. When the gangster Gaffney (Boris Karloff) is machine-gunned in Scarface (1932), viewers see, instead, a bowling pin that totters for a moment and then falls. A teddy bear on fire amid the wreckage of an aircraft tells viewers of The High and the Mighty (1954) that the young son of the film’s hero (John Wayne) has perished in the crash. In Sweet Smell of Success (1957), 282
TF17557.indb 282
16/9/08 15:47:33
V IOLE NC E
a sudden cut to the drummer in a jazz band takes viewers out of a scene in which two brutal cops attack a musician, with the drummer’s activity standing in for the beating that is off-camera. The ideological assumption that guided this policy was that depictions of egregious violence to the human body were distasteful and loathsome and were to be off-limits to filmmakers. Thus, in the 1930s and 1940s, filmmakers might show gunfire blasting into masonry walls, shattering windows, and blowing holes in furniture, but they could not squib actors so as to simulate bullet strikes on the body. One of the anomalies of gangster movies in this period is that drive-by shootings wreak tremendous visible damage on buildings while shooting victims perish unmarked. The foundational convention of violence against people in this period was a representational statement that, far from producing anguish and pain, violent death merely signals the onset of sleep. In The Strange Love of Martha Ivers (1946), Walter O’Neil (Kirk Douglas) shoots Martha (Barbara Stanwyck) in the belly at close range, and she sinks peacefully out of the frame, as if she is falling asleep. This type of tranquil death is common in films of the period, and it strikes modern viewers as being quite unrealistic because they are accustomed to seeing signs of bullet trauma and spasms in the bodies of gunshot victims. Certainly Martha’s narcotized response is a highly stylized manner of representing a violent assault. But as I will discuss in the last section of this essay, movies today are not much better at capturing essential truths about violence. What has changed is that viewers today expect to see visible evidence of physical trauma and pain. These signs, however, are often elaborated and embellished in contemporary film to a degree that becomes nonsensical, as when the villain in an action movie is punctured by scores of bullets, falls down seemingly dead, and then leaps back up to launch another attack on the hero. While the PCA sought to impose limits on screen violence, filmmakers pushed in the opposite direction. They filmed scenes in which characters are skinned alive (The Black Cat [1934]), have a flaming fondue thrown into their face (Raw Deal [1948]), or are thrown down a flight of stairs in a wheelchair (Kiss of Death [1947]). The Second World War opened the door to greater violence in the era’s combat films, such as Bataan (1943), which shows a partial decapitation. The PCA tolerated greater levels of violence in war movies than in gangster and horror movies, but across all action genres the push by filmmakers to do harder violence was unrelenting. Moviemakers were drawn to elaborate ever more stylish and intensified depictions of violence, and once knowledge of how to do gun violence for the camera or how to stage a convincing physical beating became part of the filmmaking community’s stockpile of craft knowledge, such cinema accomplishments could not be unlearned. Thus, while montage editing and slow motion are not established as a normative template for gun violence until the late 1960s, the death of Tony Camonte in Scarface (1932) remained as an example to filmmakers wanting to take cinema further. As Camonte is machinegunned by the cops, his death throes extend across several compositions taken with different camera setups. These prolong his death agony, and actor Paul Muni, playing Camonte, pantomimes the spasms produced by bullet hits, even though no squibs are employed. No slow motion is employed either, but the little montage shows very 283
TF17557.indb 283
16/9/08 15:47:33
S T EPH E N PR I NC E
clearly how editing can effectively slow down, by extending, the temporal duration of a moment of violence. Once the sequence had been accomplished, it remained there for filmmakers to study in subsequent years. Thus, while it is doubtful that a history of screen violence can be written in terms of a movement toward ever more truthful depictions, it is the case that the evolution of screen violence has occurred in terms of a movement toward ever more intense and elaborated depictions. These got their biggest boost in the 1960s, as the power of the PCA waned, along with the legal foundations on which regional censorship rested. Filmmakers rushed forward to pursue their long-nurtured visions of a bloody screen. Acts of violence and types of imagery that would have been censured in earlier decades now flourished. These included the extended beatings of Marlon Brando in The Chase (1966) and Eli Wallach in The Good, the Bad and the Ugly, graphic imagery of arrows penetrating victims in Duel at Diablo (1966), and the incineration of German officers and women in The Dirty Dozen. But it was the use by Arthur Penn and Sam Peckinpah of slow motion and multiple camera montages in Bonnie and Clyde and The Wild Bunch that established a stylistic template for gun violence that predominated for more than a generation. They also employed elaborate squibs to simulate bullet strikes on the characters, overturning a PCA prohibition in flamboyantly bloody fashion. When the Bunch and Bonnie and Clyde are gunned down, their flesh and clothing erupt with force that is both visual and visceral. Peckinpah even went so far as to rig squibs for entrance and for exit wounds and to put pieces of meat inside the explosive so that it would look like bits of the character were blown apart by the bullet. Whereas the ideology governing violent representations during the PCA period in Hollywood stipulated that such representations be clean and sanitized and, oddly enough, that violence did no serious, visible harm to human beings even while killing them, the new approach emphasized the destructive power of violence, especially upon human flesh. Indeed, Peckinpah stated time and again that he wanted to put the sting, the pain, back into movie violence because he considered the existing Hollywood style to be false to reality. Ironically, however, critics and viewers regarded his slow motion and montage editing to be inherently more realistic as a style than the existing norms that he sought to overturn. Peckinpah, by contrast, always insisted that his style was artificial, but that through this artifice he could catch a truth about violence that Hollywood had avoided, namely, that it dehumanizes people, destroys their humanity. He sought to portray this by emphasizing the grotesqueness of a character’s extended death throes. With its lengthy, detailed massacre scenes, The Wild Bunch was the most elaborately violent film that had yet been made, and it had a terrific impact on audiences because it made them conscious of a new way of looking at, and filming, screen violence. There was nothing comparable to that impact until Steven Spielberg released Saving Private Ryan, which hit audiences with the same intensity and revolutionary force as Peckinpah’s film had thirty years previously. Spielberg, too, sought to overcome the existing stylistic templates governing depictions of Second World War combat. Like Peckinpah, his solution was to create a novel way of seeing using a freshly calibrated formal design, one that audiences were not overly familiar with. In this case, it 284
TF17557.indb 284
16/9/08 15:47:34
V IOLE NC E
involved radical experimentation with the speed and angle of the camera’s shutter, manipulation of the camera’s lens, and methods of processing the film conjoined with extremely graphic depictions of bullet wounds and bodies blown apart by explosives. He also replaced the reassuring narrative conventions of Hollywood’s combat films, in which the death scenes of secondary and major characters occurred at clear and preordained points in the story, with a more disturbing, existentialist acknowledgment that amid the chaos of combat death can come at any time for anyone. This existentialist understanding that violent death is inescapable and threatens from all sides is a social perception that separates the contemporary period from earlier eras of screen violence, in which a violent event was understood to be exceptional and atypical. In the old horror movies, the monsters were always killed off unambiguously at the end. Horror, today, ends with the monsters still at large. Moreover, the graphic violence that Spielberg placed for the first time inside a Second World War combat movie had taken root in the horror film a generation earlier. In this respect, the horror genre is a key link between Peckinpah’s experiment with the western in The Wild Bunch and Spielberg’s experiment with the combat film in Saving Private Ryan. The widespread, ferocious, and indiscriminate killing that Peckinpah showed in The Wild Bunch flourished subsequently in horror movies, particularly the graphically violent ones that began to proliferate in the 1980s, where such violence constituted a statement about the essentially pitiless nature of the universe. As these developments suggest, films today are more unsparing of their audiences, are crueler, and are more assaultive, in part because the screen technologies of violence (prosthetic limbs, digital effects) make outré displays easier to achieve today, but also because viewers seem to regard such cruelty as offering a truer portrait of the world they inhabit. Indeed, this is why the brutality that shocked audiences in the 1930s seems pale today. In those earlier films it occurs less frequently and is staged with greater tact and indirection. The cruelty on display in contemporary film is indeed breathtaking. In the horror genre, for example, a cycle of torture films has proven to be very successful at the box office. Saw (2004), Hostel, and their sequels construct narratives around the abduction and slow torture of characters whose physical torment is visualized in lurid and fascinating detail. Mel Gibson’s The Passion of the Christ (2004) revivified the tale of Jesus’ torture, by dwelling at length upon the tearing and rending of his flesh during scourging and crucifixion. Fans of the film reported this to be the most realistic depiction of these events, even though none ever had the misfortune to actually witness a victim scourged with a flagrum or nailed to a cross. Actually, much of what Gibson shows in the film is more in keeping with the conventions of movie violence than with the known facts of the physiological response to scourging and crucifixion. But despite this, the film’s relentless cruelty seemed to provide its fans with what they regarded to be a truer account of Jesus’ last days. That contemporary films are more harsh and detailed in their presentation of violence seems clear enough. But what about the corollary assumption that such an approach offers a truer account of the phenomenon of human violence? While one may grant that viewers perhaps feel more unsettled about their world today than in generations past and that the higher level of cruelty in today’s films is related to this 285
TF17557.indb 285
16/9/08 15:47:34
S T EPH E N PR I NC E
anxiety, two qualifications need to be considered. In these key ways, the world of movie violence in mainstream films remains quite unreal. First, while viewers today are surrounded by media images of great violence – in the movies, on television, and in videogames – all of this is essentially fictitious violence. It is elaborated within a narrative framework that structures it according to moral formulas of punishment and reward, and viewers know at all times that it is fictitious. By contrast, imagery of actual, real-world violence is far more regulated, censored, and tightly controlled than are the fictitious images carried by the media. The killing and maiming of Iraqi civilians, often targeted by US forces, has been a widespread and enduring feature of the Iraq War, but it has been rarely photographed, despite extensive reporting in print media. The US government does not permit any filming of the returning caskets of American soldiers fallen in Iraq. Imagery of the planes hitting the World Trade Center on 9/11 and of the towers falling is readily accessible, but not images of the jumpers. Photos and footage of people falling from the towers have been widely suppressed, especially images of the mangled remains where these victims hit the street. Following Hurricane Katrina, hundreds of bodies littered New Orleans, yet the news media refrained from showing the scale of these deaths. Fictional media violence is rampant in contemporary culture, yet imagery of real death is not, despite its abundance in the world that exists off screen. How then is a viewer to judge the veracity of contemporary cinema’s visual language of violence? It becomes exceedingly difficult because of this paradox – violence as a fictitious construction is amplified and elaborated, while the visual record of actual violence in the world becomes a victim to political agendas that seek its suppression. As a result, viewers have a visual frame of reference that is skewed in terms of the representational conventions that predominate in contemporary media. A second problem must qualify any judgment that the movies today are better at depicting violence than in generations past. This problem is related to the attraction of cinema to action and to movement, which Brian De Palma noted. Cinema’s privileging of action has meant that the representation of violence overwhelmingly occurs in terms of a stylistics of spectacle, in which the emotional buildup and explosive release of violent energy takes precedence. The western outlaws or the gangsters march dramatically toward their climactic confrontation with an antagonist; the viewer keys up with anticipation; and an exciting, extensive action sequence shows a glorious gun battle in which the characters find numerous opportunities for dramatic and heroic deaths. Actual gun violence is different, and what the movies, fixated on the stylistics of spectacle, don’t catch is the aftermath. Here I must speak more personally in order to describe this condition. I have been a professor at Virginia Tech for nearly twenty years, during which time the community has been a peaceful and quiet place to live and work. That changed on 16 April 2007, when a deranged student went on a rampage and shot to death thirty-two people. He killed two students at a dormitory early that morning and then, a few hours later, chained shut the doors to a building housing numerous classrooms. He prowled from room to room, slaughtering as many people as he could. He killed thirty in the building and wounded twenty-five. The shooting was over very quickly. His rampage in the building lasted about fifteen minutes, but 286
TF17557.indb 286
16/9/08 15:47:34
V IOLE NC E
the profound consequences, and indeed the meaning, of his actions were to be found in the aftermath. Stunned family members, friends, and members of the university community struggled to make sense of what had happened. Lives were changed forever, and the web of human relationships showed up as the densely constructed network that it is. In the movies, a killer’s victims are quite discrete and are limited largely to those who are actually killed. Maybe a wife or a friend or a colleague is alive at the end of the story to reminisce about the character that has died, but movies don’t capture the network of social interconnection upon which actual violence has such a terrible impact. In life, those affected are far more numerous because the fallen have families, friends, acquaintances, and those families and friends have other friends and families who are affected because they know someone who was involved. The movies don’t get this web of effects very well, and they certainly don’t get the aftermath of violence very well. The lingering pall that the loss of a loved one creates has never been as interesting for filmmakers as it is to stage and to stylize scenes of violent action. The aftermath is more inward, less visual. Filmmakers, especially in the action genres of popular cinema, are drawn to the mechanics of violence, less often to what violence leaves behind in the world after it is over. But in my community, that’s where the meaning and effects of the killer’s rampage were to be found. Violence, then, is a fundamental category for expression in cinema, one that has always exerted great attraction for filmmakers and audiences, and yet the medium’s grasp of this subject is partial, biased, and certainly misleading. Cinema attaches great stylistic hyperbole to its renditions of violent death, and, as style, violence is something the movies do extremely well. As a human phenomenon, however, violence retains a power and a complexity that film – popular cinema especially – has often failed to grasp. See also Censorship (Chapter 3).
References Artaud, A. (1958) The Theater and Its Double, trans. M. C. Richards, New York: Grove Press. Bandura, A. (1973) Aggression: A Social Learning Analysis, Englewood Cliffs, NJ: Prentice-Hall. Bok, S. (1998) Mayhem: Violence as Public Entertainment, Reading, MA: Addison-Wesley. Cerulo, K. (1998) Deciphering Violence: The Cognitive Structure of Right and Wrong, New York: Routledge. Clover, C. (1992) Men, Women and Chainsaws: Gender in the Modern Horror Film, Princeton, NJ: Princeton University Press. Comstock, G. (1980) “New Emphases in Research on the Effects of Television and Violence,” in E. L. Palmer and A. Dorr (eds.) Children and the Faces of Television, New York: Academic Press. Goldstein, J. (ed.) (1998) Why We Watch: The Attractions of Violent Entertainment, New York: Oxford University Press. Hamilton, J. (1998) Channeling Violence: The Economic Market for Violent Television Programming, Princeton, NJ: Princeton University Press. Lichtenfield, E. (2007) Action Speaks Louder: Violence, Spectacle, and the American Action Movie, Middletown, CT: Wesleyan University Press. McArthur, D., Peek-Asa, C., Webb, T., Fisher, K., Cook, B., Browne, N., Kraus, J., and Guyer, B. (2000) “Violence and Its Injury Consequences in American Movies: A Public Health Perspective,” Western Journal of Medicine 173 (September): 164–8.
287
TF17557.indb 287
16/9/08 15:47:34
S T EPH E N PR I NC E
Paik, H., and Comstock, G. (1994) “The Effects of Television Violence on Anti-social Behavior: A Meta Analysis,” Communication Research 21(4): 516–46. Pally, M. (1984) “ ‘Double’ Trouble,” Film Comment 20 (September–October): 12–17. Potter, W., and Smith, S. (2000) “The Context of Graphic Portrayals of Television Violence,” Journal of Broadcasting and Electronic Media 44(2): 301–23. Prince, S. (ed.) (2000) Screening Violence, New Brunswick, NJ: Rutgers University Press. —— (2003) Classical Film Violence: Designing and Regulating Brutality in Hollywood Cinema, 1930–1968, New Brunswick: Rutgers University Press. Sharrett, C. (ed.) (1999) Mythologies of Violence in Postmodern Media, Detroit: Wayne State University Press. Slocum, J. (ed.) (2001) Violence and American Cinema, New York: Routledge.
288
TF17557.indb 288
16/9/08 15:47:34
Part II
Authors and T rends
TF17557.indb 289
16/9/08 15:47:34
TF17557.indb 290
16/9/08 15:47:34
27
RUDOL P H AR NHEIM Jinhee Choi Since the turn of the twentieth century, the continuing attempt to elevate the status of film, artistic or otherwise, has taken various routes, through either an affiliation with or differentiation from previously established art forms. The film d’art movement in France and Autorenfilm in Germany, for instance, tried to lure the middle-class audiences through literary adaptations, the subject matter of which such middle-class audiences would find sophisticated. Classical film theorists, such as Rudolf Arnheim, André Bazin, and Sergei Eisenstein, took an opposite approach, claiming that there exists a set of characteristics distinctive of the film medium, and that film as art should explore these medium-specific qualities. This essay examines Arnheim’s film theory by tracing out his reasons for thinking of film as an independent art form. Since media convergence has become the norm within the global film industry, one may find the classical film theorist’s urge to explore the peculiarities of the medium rather outdated. Media conglomeration might be an historically contingent phenomenon, one in which film plays a significant part, but aesthetic concerns still remain with regard to the relationship between cinema and other media. Arnheim has reservations about certain technological developments within the history of cinema because he believes that the artistic potential of film lies in its technological limitations. This reservation further invites us to consider and reconsider the relationship between film technology and aesthetics to counterbalance a tendency in the contemporary discourse on film technology that foregrounds the medium’s limitlessness.
The real, the filmic, and the natural Arnheim is concerned with film as art and characterizes his approach as Materialtheorie, according to which the aesthetic principles of a medium should be based upon its material properties (Arnheim 1957c: 2). Like other artistic media – such as pictures, music, dance, and literature – film can serve various functions, one of which is artistic. For Arnheim, not all films are art and they don’t need to be. But to grant film an artistic possibility he had first to disprove the then common claim that the film medium lacks artistic potential. The major criticism leveled against film derived from a certain characterization of its representational capacity: as a photographic
TF17557.indb 291
16/9/08 15:47:34
JI N H EE C HOI
medium film is merely a reproduction of reality. One implication of such a criticism is that the mechanical processes involved in photography and film do not permit the operators’ creative control or intervention. Painters, for instance, can intervene at any moment during the process of painting, from the decision of the subject matter, composition, drawing, and color of pigment, to the thickness of brush stroke. In contrast, filmmakers and camera operators have a limited range of control – such as lighting and the placement of objects during the filmmaking process – while the rest is rendered through a more or less automatic mechanical process. As Noël Carroll notes, such a charge comes from the growing artistic tendency in the late nineteenth and early twentieth centuries toward antimimesis (Carroll 1988: 21). From Baudelaire to Croce, the principal function of art, it is claimed, should not be found in the imitation of nature, and thus film, which excels in “re-presenting” reality, is considered too aesthetically inadequate to belong to the realm of fine art. In his attempt to challenge the aesthetic criticism of photography and film, Arnheim first asks whether the filming process is indeed automatic. A photographic image of a simple object, such as a cube, notes Arnheim, is not automatically rendered. It can either succeed or fail in making an object recognizable to the viewer and thus requires the photographer’s skill to find an adequate angle and proper lighting (Arnheim 1957a [1933]: 9–10). The representational capacity of photography and film, then, is not given, but is something that can be achieved by virtue of the photographer’s skills. Arnheim does not construe the relationship between a photographic image and its referent as a matter of “truth,” or correspondence. Rather, it involves the aesthetic sensibility of a photographer, which can provide an insight into the object. An artistic filmmaker should capture the essence of an object or an event, and there is no set of rules to follow. Arnheim argues, “There is no formula to help one choose the most characteristic aspect: it is a question of feeling” (Arnheim 1957a [1933]: 10). The then common assumption that photography and film cannot be art was the result of a misconception that the filmmaking process is automatic. As Arnheim asserts, it not only requires a level of skill from a filmmaker in order for the audience to discern the object filmed, but also involves an aesthetic sensibility in order to choose and foreground the most salient aspects of the object. Arnheim further asserts that the film medium’s tie to reality does not yield a filming process that is altogether automatic, nor does film replicate the perceptual experience of the real. Arnheim lists a set of attributes of film that differentiate filmic perception from natural perception. From (i) the reduction of depth, to (ii) lighting, to (iii) the delimitation of the screen, to (iv) the absence of the space-time continuum, and to (v) the absence of color and of nonvisual sensory coordination, filmic transformation of reality falls short of rendering an exact replica of natural perception. In filmic images, for instance, sizes and shapes of objects do not remain constant in the way we normally see two distant objects. Instead, the object in the background of a shot looks disproportionately small, due to the camera’s monocular vision (Arnheim 1957a [1933]: 13–14). Moreover, a black-and-white film reduces the natural color scheme to a limited color palette, which results in the reconfiguration among colors and objects. The shade of tree leaves in a black-and-white film may be linked to that of a woman’s 292
TF17557.indb 292
16/9/08 15:47:34
RU D OLPH A R N H EI M
lips through similar shades of gray, inviting the viewer to make a connection that he or she would not normally draw. The characteristics that Arnheim attributes to cinema are not all specific to the medium. Many of them are shared with photography or derived from the fact that film is a photographic medium. Arnheim dwells on one aspect, though, which he believes distinguishes film from both photography and theater. Film produces in the spectator a distinctive spectatorial effect. Arnheim claims that the film image is neither completely two-dimensional nor completely three-dimensional, rendering a “partial” illusion of real space. By “partial,” Arnheim refers to two different aspects: (i) the unreality of filmic space is often unnoticeable; and (ii) information gathered from filmic image is incomplete (Arnheim 1957a [1933]: 28–9). It is the first aspect that demarcates the film medium from the other competing media. I will come back to this point shortly. The second meaning of “partial” is in accordance with the major principles of gestalt psychology. Even the most elementary process of vision does not passively receive data from the real world but creatively organizes sensory raw materials in accordance with a set of principles. In natural perception, we do not need every detail to infer the whole. Likewise, with a few salient aspects of objects and events represented in a film, one can still have a strong sense of the real. But the partial illusion in the second sense is not unique to the film medium but can be shared with any type of art. To what extent, then, does the perceptual experience of film involve an illusion? Arnheim postulates a continuum from abstraction to the real in representing space, with photography as the most abstract, film in the middle, and theater at the most real end. The film spectator does not have a false belief about what he or she sees. He or she does not mistake the film image for the real. But film has the spatial and temporal mobility that both photography and theater lack, without drawing much of the viewer’s attention to it. A film image is bounded by its margins and can easily allow a montage, both within a scene and between scenes. Arnheim claims, “a result of the ‘pictureness’ of film is, then, that a sequence of scenes that are diverse in time and space is not felt as arbitrary. One looks at them as calmly as one would at a collection of picture postcards” (Arnheim 1957a [1933]: 27–8). He continues, “if at one moment we see a long shot of a woman at the back of a room, and the next we see a close-up of her face, we simply feel that we have ‘turned over a page’ and are looking at a fresh picture” (Arnheim 1957a [1933]: 28). Arnheim must have a specific editing system in mind, here, such as the continuity editing of Hollywood, in which the editing pattern is less noticeable. In contrast, theater employs “real” space to a certain extent, which disallows montage within a scene. But Arnheim does not further postulate the relationship between the spectator and the fictional space and/or its ideological implication, which becomes the focal point of the debates of postclassical film theorists, such as Jean-Louis Baudry and Christian Metz, of a psychoanalytical-semiotic bent. Arnheim’s ontological and epistemic observations about film lead him to refute the prevalent criticism of the medium. Film is far from a perfect copy of reality, rendering in the viewer a perceptual experience as an alternative to both natural perception and the perception given by other artistic media. For Arnheim, however, the imperfections of the “mechanical” reproduction process of the film medium should not be 293
TF17557.indb 293
16/9/08 15:47:34
JI N H EE C HOI
dismissed as technical flaws; instead, they should serve as the basis for the artistic use of the medium. In the chapter “The Making of a Film,” Arnheim demonstrates how each “defect” or “drawback” of the film medium can yield artistic effects. His famous quote aptly captures this idea: “Art begins where mechanical reproduction leaves off, where the conditions of representation serve in some way to mold the object” (Arnheim 1957a [1933]: 57). According to Arnheim, the artistic possibility of the medium should be found in its very technical constraints. As Jon Elster notes, creativity comprises both “choice of constraints” and “choice within constraints” (Elster 2000: 176). The filmmaker’s initial choice of the medium would impose certain intrinsic, technical constraints, within which he or she can pursue the medium’s artistic potentials. Consider, for example, the reduced depth of field in the cinema, which can underscore a psychological state of a character. Arnheim takes an example from King Vidor’s The Crowd (1928), which showcases an aesthetic inclination of the medium. A boy hurries home, as he sees a crowd gathering around his house drawn by the sound of ambulance. The camera is located on the second floor, looking down at the door, as the boy enters. As the boy nears the end of the stairs, he appears disproportionately large – an effect of the camera lens – which nevertheless artfully expresses his fear of hearing the dreadful news – his father is dead. The delimitation of the visual field – another technical constraint, as well as aesthetic potential of the medium – carries one of the most significant functions of film: framing. Framing comprises a wider range of film techniques and devices – camera distance (especially close-up), angle, movement, and editing – in contrast to other limitation, such as absence of color or of the space-time continuum, which results from more or less a single device (film stock/development) or technique (editing). In theater, audiences have considerable visual freedom in terms of what to look at; film, by contrast, has a set of devices to direct and control the viewer’s attention. The great silent comedies that Arnheim revisits throughout Film as Art, provide ample instances of gags created through framing. For example, in The Cameraman (1928), a female receptionist, who is the object of affection of the male protagonist, enters the office. The waiting area is not shown in its entirety, and as the camera reframes we see Buster Keaton sitting on a chair at the corner: he has been waiting for her all night. Such a narrative and spatial revelation can come as a pleasant surprise to the viewer, violating the viewer’s expectation of space. Given the artistic potentials of the medium, the goal of filmmakers is not merely to “re-present” reality, unfolding in front of the camera, but to transform its material constraints into cinematic expression. Arnheim does not clearly discuss the scope of “expression” in Film as Art and one must tease out his notion of “cinematic expression” in the light of his later work, Art and Visual Perception. In the latter, Arnheim emphasizes the expressive nature of visual perception in general. Our perceptual mechanism does not merely register sense data but recognizes them as expressive: When I sit in front of a fireplace, I do not normally register certain shades of red, various degrees of brightness, geometrically defined shapes moving at such and such a speed. I see the graceful play of aggressive tongues, flexible striving, 294
TF17557.indb 294
16/9/08 15:47:34
RU D OLPH A R N H EI M
lively color. The face of a person is more readily perceived and remembered as being alert, tense, and concentrated than it is as being triangularly shaped, having slanted eyebrows, straight lips, and so on. (Arnheim 1974: 454–5) The perceptual processes Arnheim describes above differ from scientific observations. If the former translate the raw material into the rubric of expression broadly construed to include both expressive and nonexpressive qualities, the latter deal with the material conditions of objects – weight, shape, and size. Arnheim claims that expression consists of the recognition of the isomorphic relationship between the patterns of the stimuli and those of the expressive qualities (Arnheim 1974: 450). A drooping willow tree does not look sad because it looks like a sad person. Rather, the willow tree expresses sadness by virtue of the shapes of its branches that are structurally similar to that of the intended mood. For Arnheim expression is “an inherent characteristic of perceptual patterns,” not a projection of and/or association with the expressive qualities of human and animate beings (Arnheim 1974: 452). A comparison or analogy with the physical manifestations of human expression comes as secondary (Arnheim 1974: 452). How can film express? In addition to the filming process transforming the real, a film reflects and registers the artistic vision and sensibility of the filmmaker. Arnheim continues, “If expression is the primary content of vision in daily life, the same should be all the more true of the way the artist looks at the world” (Arnheim 1974: 454–5). Cinematic representation is not a vehicle of transferring the real, not merely an instrument of observation, but a means to translate and communicate via the real. The aesthetic appreciation of film should, then, encompass the understanding of content and its expressive appearance. As Arnheim notes, one should be aware that “there stands a policeman” and of “how he is standing” (my italics, Arnheim 1957a [1933]: 43). Film is a medium that displays both the generality and concreteness of the object filmed. A shot of a policeman should bring out the most characteristic aspect of policemen in general, as well as the peculiar aspect of that policeman. A carefully chosen angle will enable the viewer to see the unfamiliar within the familiar. Arnheim’s expressionist aesthetics, however, is heavily circumscribed by “naturalism.” Film art is not an imitation but a transformation of nature: it “strengthens,” “concentrates,” and “interprets,” but not to the extent that it completely “restructures” nature or imposes a new reality (Arnheim 1957a [1933]: 35, 57). With his emphasis on “nature,” Arnheim does not align himself with any specific artistic movement, but rather uses it as an umbrella term, encompassing both his ontological assumptions about the medium and a general aesthetic principle. Arnheim views art as being located somewhere between nature and a complete replica of nature (the complete cinema). In order to be art, film should explore and foreground the formal as well as phenomenal gap between nature and film. But film still should remain faithful to nature. In his analyses of Entr’acte (1924), Arnheim claims that a shot of a dancer taken through the glass floor merely foregrounds the formal similarity between the dancer’s tutu and a flower petal without increasing an artistic value of the film. Arnheim observes that Carl Theodor Dreyer’s treatment of the court scene in The 295
TF17557.indb 295
16/9/08 15:47:34
JI N H EE C HOI
Passion of Joan of Arc (1928) may be visually stimulating in an otherwise monotonous scene, but the camera angles are utterly unmotivated. Arnheim raises a similar yet different criticism against the Soviet montage filmmakers, including V. I. Pudovkin and Sergei Eisenstein, for their use of intellectual montage. Some cases of intellectual montage “violate” reality by breaking both the narrative and spatial unity. In Pudovkin’s Mother (1926), after the mother visits her imprisoned son, his hope of being rescued is expressed by a juxtaposition of shots of him in jail with images of children playing on the field and of ice thawing on a river. According to Arnheim: Putting actual pictures in juxtaposition, especially in an otherwise realistic film, often appears forced. The unity of the scene, the story of the prisoner who is rejoicing is suddenly interrupted by something totally different. Comparisons and associations like the brook and the sunbeams are not lightly touched upon in the abstract but are introduced as concrete pieces of nature – and hence are distracting. (Arnheim 1957a [1933]: 90) Comparing intellectual montage with poetry, Arnheim claims that it is more challenging for the visual medium to convey abstract ideas than the verbal medium. The verbal medium readily allows a conceptual link between formally discrete ideas, due to their weak mental imageries. The concreteness of filmic images, however, would resist such an imposed unity. Interestingly enough, Arnheim evaluates somewhat differently Eisenstein’s use of intellectual montage in The Battleship Potemkin (1925). Although the stone lions seem to rise to roar through the juxtapositions of three successive shots of the three different statues, the apparent unity of the action created by montage is strong enough to make the montage convincing (Arnheim 1957a [1933]: 100). Arnheim does welcome the evocation of an abstract idea through cinematic means, insofar as it does not break the illusion of unity or produces the illusion of a reality. Arnheim praises the ingenuity of Chaplin’s performance in The Gold Rush (1925), in the scene where Chaplin eats his boiled boot and shoelace as if it is a proper meal. Chaplin’s performance effects the kind of expressive engagement with perceptual reality via a formal analogy manifest in the scene; it successfully translates the idea that poverty is a low grade of wealth through the formal similarity between the meal that Chaplin seemingly enjoys and a meal that the rich would consume (Arnheim 1957a [1933]: 144–5). For Arnheim, film is first and foremost a visual medium. Although the artistic effects of the medium should be found in its departure from reality, film cannot completely do away with it. It is no surprise, then, that animation, which does not have any link to reality, is completely deleted from Arnheim’s discussion of film art. However, one must note that Arnheim’s criticisms of certain types of films or style, including abstract film and intellectual montage, are derived not only from his ontological assumption about the medium but also from the aesthetic code in art. Although Arnheim attempts to draw out a “naturalistic” aesthetic from the cinema’s ontological commitment to reality, he equivocates on two different meanings of “natural”: (i) natural in the 296
TF17557.indb 296
16/9/08 15:47:34
RU D OLPH A R N H EI M
sense that it is faithful to reality; and (ii) natural in the sense that it is “subtle” and “indirect.” The latter sense of natural becomes apparent in his discussion of film acting. Arnheim prefers “natural” acting to stylized silent acting. Cinema requires a type of acting different from what is desirable in theater, since the size of screen – and the close-up – increases the legibility of character action. According to Arnheim, everyday actions and gestures are often indeterminate, with their meanings assessed by recourse to the context. In contrast, film acting and gestures contain precision and clarity to the extent that they may be viewed as unnatural. “Naturalness” is a relative concept for Arnheim, judged in the light of convention. An appropriate acting style in slapstick comedy might be unsuitable for melodrama. Nevertheless, argues Arnheim, facial expressions and physical gestures should not be prioritized and serve only as one of the many ways in which character psychology is conveyed. In Les Nouveaux Messieurs (1929), for instance, Suzanne spills tea on the saucer when she learns that her lover will be transported abroad. As the simple phrase “That day we read no more” in Dante’s The Inferno, condenses the love affair of Francesca and Paolo, Suzanne’s behavior indirectly indicates her surprise and disappointment (Arnheim 1957a [1933]: 107, 142). The indirect visual method that Arnheim endorses is, in part, possible by virtue of the cinematic devices – the close-up and framing. Inanimate objects can bear as much significance as actors do, as a result of the film’s capacity to direct the viewer’s attention to any element of mise en scène. But Arnheim’s predilection for such an aesthetics does not necessarily result from the material base of the medium. Rather, Arnheim’s approach to film aesthetics is deductive in the sense that he postulates a certain function and the principles of art in general, and then shows how film can meet such standards. Such an approach, however, has the risk of begging the question. One may legitimately challenge Arnheim’s philosophical premises: Why is expression the principal function of all art? To what extent does the indirect visual method enhance such a function? The burden of proof is on Arnheim (Carroll 1988), and it seems fair to add that he did not ultimately shoulder that burden. So far, we have examined how Arnheim builds his aesthetic theory of film on the medium-specific elements. As each artistic medium employs a distinct material, film should explore its material for artistic purposes. In the next section, I focus on Arnheim’s conception of the “complete cinema,” which is a neglected alternative to Bazin’s (1967 [1945]) idea of the “total cinema.” To what extent does the development of film technology circumscribe the artistic potential of the film medium? To what extent does Arnheim’s rejection of the complete cinema further illuminate his film theory?
The notion of the complete cinema The pursuit of the complete cinema, according to Arnheim, is a natural outcome of “the fulfillment of the age-old striving for the complete illusion” (Arnheim 1957a [1933]: 158). Technological developments would bring the film medium closer and 297
TF17557.indb 297
16/9/08 15:47:34
JI N H EE C HOI
closer to an exact reproduction of reality. Arnheim acknowledges that the history of visual art has been driven by the desire to create a replica of reality – what Bazin later calls the “mummy” complex, an impulse to immortalize an appearance against the passage of time (Bazin 1967 [1945]: 9). At the same time, he claims, there has been a countertendency “to originate, to interpret, to mold” (Arnheim 1957a [1933]: 157). But Arnheim is not concerned with whether the mimetic desire or the expressive tendency provides a better psychological explanation of the evolution of the visual art. His point is rather that in the case of film, the two are often in conflict, with the mimetic desire deterring filmmakers from exploring the expressive qualities of the medium. Technological developments will enhance the medium’s capacity to imitate nature. And the narrower is the gap between the filmic reproduction and reality, the slimmer are the chances for film to become an art. Arnheim’s concern, however, is not that the introduction of technologies like sound, color, and widescreen will remove some of the constraints embedded in the medium. One should recall that, for Arnheim, technological imperfections manifest in the film medium are not, in and of themselves, aesthetic merits: instead they can be employed to produce artistic effects. Any technological novelty has its material constraints, and filmmakers can develop a new set of relationships between a newly developed device and the existing ones. Arnheim even grants an artistic possibility to sound film, although of an accidental kind. “By sheer good luck, sound film is not only destructive but also offers artistic potentialities of its own” (Arnheim 1957a [1933]: 154). Arnheim’s rejection of sound films is based on both industry and aesthetic grounds. I will discuss the industry reason first, and the aesthetic one later. Arnheim argues that when the sound film becomes the industry norm, there will be no room for the silent cinema to exist alongside the sound film (Arnheim 1957a [1933]: 159). This claim is a reflection on the practices of the film industry in general, but also has an aesthetic consequence. Arnheim argues that so long as silent film, sound film, and complete film can coexist, there is nothing inherently deplorable about the invention of the complete cinema. It can serve a function different from that of the silent cinema – such as filming a stage play. But the coming of sound has an aesthetic consequence as well, since it will alter the viewer’s criteria for assessing the “naturalistic” film. As I have mentioned earlier, Arnheim construes “naturalness” as a relative concept. The viewer’s conception of the natural would change in accordance with the introduction of new technologies. Sound film would make silent film feel artificial, as color film would make black-and-white film look unnatural (Arnheim 1957a [1933]: 160). Arnheim’s resistance to sound film is deep-seated and grounded in his conviction that film is a visual medium and that the introduction of sound will hamper the medium from expanding its expressive qualities. In his article entitled “A New Laocoön,” Arnheim considers the artistic potentials of the hybrid arts, through which he attempts to demonstrate that sound film curtails, rather than strengthens, the aesthetic merits of the medium. Arnheim asserts that unlike human expression, which is anchored on the biological unity of a person, hybrid arts lack such a center. 298
TF17557.indb 298
16/9/08 15:47:34
RU D OLPH A R N H EI M
The combination of two sensory modes – for instance, dance and music, or visual action and dialogue – does not automatically guarantee the artistic unity of a work. Artistic unity, according to Arnheim, is achieved at the secondary level, which presupposes both the separation and pa