The Palgrave Handbook Of The Philosophy Of Film And Motion Pictures 3030196003, 9783030196004, 9783030196011

This handbook brings together essays in the philosophy of film and motion pictures from authorities across the spectrum.

1,440 142 9MB

English Pages 1047 Year 2019

Report DMCA / Copyright


Polecaj historie

The Palgrave Handbook Of The Philosophy Of Film And Motion Pictures
 3030196003,  9783030196004,  9783030196011

Table of contents :
Foreword......Page 5
Contents......Page 8
Notes on Contributors......Page 13
List of Tables......Page 21
Introduction......Page 22
Part I: The Medium in Film and Motion Pictures......Page 24
Chapter 1: Film Ontology: Extension, Criteria and Candidates......Page 25
Determining the Extension of “Film”......Page 26
Criteria for an Ontology of Film......Page 30
Realism......Page 32
Illusion......Page 35
Imagination......Page 37
Film as Dream......Page 38
Film as Language or Symbol System......Page 40
Film as Thought......Page 42
Moving Pictures, Moving Images......Page 44
Further Conditions?......Page 47
Bibliography......Page 48
Chapter 2: Medium Specificity......Page 50
What Was Medium Specificity?......Page 51
The Return to Medium Specificity......Page 56
The Issue of Evaluation......Page 65
Bibliography......Page 68
Chapter 3: The Moving Image......Page 69
Why Images? Why Motion?......Page 70
Transparency......Page 74
Depiction......Page 76
Movement and Moving Images......Page 79
Conclusion......Page 87
Bibliography......Page 88
Chapter 4: The Art of Cinematography......Page 90
Preliminary Considerations: Picture-Making......Page 91
The Resources of Cinematography......Page 93
Film Stock and Lab Work......Page 94
Lighting......Page 98
Lenses and Camera Movement......Page 101
The Question of Authorship......Page 107
Bibliography......Page 110
Part II: The Structure of Film and Motion Pictures......Page 113
Chapter Overview......Page 114
Part I: Narration and Narrators......Page 115
Chatman’s Argument......Page 117
Questions About Chatman’s Argument......Page 118
Can the Actual Filmmaker Tell Her Fictional Story?......Page 119
Defending the Ontological Gap Argument......Page 121
Imagined Seeing at the Movies......Page 123
The Imagined Seeing Thesis: How to Formulate It?......Page 125
Part V: Reconsidering the Objection from Absurd Imaginings......Page 127
Conclusion: Further Issues for Cinematic Narration......Page 131
Bibliography......Page 132
Chapter 6: Narrative and the Moving Image......Page 135
Narrative Dynamics......Page 136
The Modality of Narrative......Page 140
Serial Narratives......Page 144
Narrators, Authors, Tone, and Point of View......Page 148
Bibliography......Page 155
Introduction......Page 158
Editing and Rhythm in Film Philosophy......Page 159
Rhythm and the Editor’s Cognition......Page 161
On Asking: What Is Rhythm in Film Editing?......Page 162
Materials of Rhythm: Time, Energy, and Movement......Page 164
Timing......Page 166
Pacing......Page 168
Trajectory Phrasing......Page 170
Purpose of Rhythm: Cycles of Tension and Release......Page 172
Conclusion......Page 175
Bibliography......Page 176
Chapter 8: Animation......Page 179
What Is Animated in an Animated Image?......Page 180
Kinds of Animation: A Taxonomy of Techniques......Page 181
Why Has So Little Philosophical Attention Been Paid to Animation?......Page 187
Appreciating Animation......Page 191
Digital Animation......Page 196
Bibliography......Page 200
Introduction......Page 202
The Place of Sound in the Concept(ion) of a Film......Page 203
The Challenge Against Mimetic Sound......Page 204
Medium Specificity Challenge: The Case of Sound......Page 205
Sound and the Threat to Expression......Page 206
The Conception of Film in Contemporary Philosophical Writings: The Absence of Sound......Page 207
Realism in Film Sound......Page 213
Cognitive Versus Perceptual Illusion......Page 214
Perceptual Realism Versus Convention......Page 218
Transparency......Page 221
Conclusion......Page 224
Bibliography......Page 225
Chapter 10: What Is a Screenplay?......Page 228
Authorship......Page 230
Screenplays and the Ontology of Cinema......Page 234
Screenplays and the Artistic and Ethical Evaluation of Cinema......Page 239
Conclusion......Page 243
Bibliography......Page 246
Part III: Approaches and Schools......Page 248
Chapter 11: Analytic Philosophy of Film: (Contrasted with Continental Film Theory)......Page 249
Introduction: Never Mind the Gap......Page 271
A Long and Winding History......Page 276
Language, Analysis, and Description: On Various Methods......Page 278
Of Science and Scientism......Page 282
A Short History of Historicism......Page 285
Arguing the Points: Clarity Clarified......Page 287
Conclusion: Beyond the Sinne of the Fathers......Page 290
Bibliography......Page 292
Introduction......Page 296
Phenomenology for the Philosophy of Film?......Page 298
Husserlian Phenomenology......Page 300
Husserl on Image Consciousness......Page 305
Perceptual Phantasy......Page 309
Heideggerian Phenomenology......Page 311
Dasein......Page 312
The Existential Structures of Dasein’s Being-There......Page 315
Merleau-Ponty, Sobchack, and the Embodied Film Viewer......Page 319
Bibliography......Page 323
Chapter 14: Ideology and Experience: The Legacy of Critical Theory......Page 325
Historical Background and Theoretical Assumptions......Page 326
The Benjamin-Adorno Debate......Page 329
Cinema and Experience......Page 333
Cinema as a Public Sphere......Page 336
Critical Theory in/and New German Cinema......Page 339
Ideology Critique and Experience......Page 340
Bibliography......Page 343
Introduction......Page 344
Skepticism and Perfectionism......Page 347
The World Viewed: Automatic World Projections and the Hollywood Star......Page 354
The Comedy of Remarriage......Page 358
The Melodrama of the Unknown Woman......Page 361
Reception......Page 363
Secondary Sources......Page 364
Chapter 16: Film Art from the Analytic Perspective......Page 366
Film as Art......Page 367
Gaut’s Three Levels of Discussion and Beyond......Page 369
The Emergence of Analytic Philosophy of Film: Sesonske and Sparshott......Page 375
Highlights in the Analytic Philosophy of Film After Sesonske and Sparshott......Page 380
Persistent Questions......Page 385
Bibliography......Page 386
Chapter 17: Cognitive Theory of the Moving Image......Page 389
Origins and History......Page 391
Grounding Tenets: Cognition and Naturalism......Page 394
Narrative Comprehension......Page 399
“Hot” Cognition and Embodiment......Page 402
Character Engagement, Sympathy, and Empathy......Page 406
Emotion, Affect, and Mood......Page 409
Questions and New Directions......Page 410
Bibliography......Page 412
Chapter 18: Aesthetic Criticism......Page 417
Bibliography......Page 444
Chapter 19: Poststructuralism and Film......Page 448
What Is Poststructuralism?......Page 451
Saussure and Structuralism......Page 452
The Critique of Structuralism and Poststructuralist Turn......Page 456
Poststructuralism in Film Studies......Page 458
Poststructuralism and Carroll’s Critique of ‘Grand Theory’......Page 460
Deleuze’s (Post-Poststructuralist) Film-Philosophy......Page 464
Bibliography......Page 471
Part IV: Philosophy Through Film......Page 473
Chapter 20: Thoughtful Films, Thoughtful Fictions: The Philosophical Terrain Between Illustrations and Thought Experiments......Page 474
I......Page 476
II......Page 480
III......Page 485
IV......Page 491
Bibliography......Page 494
Chapter 21: Contemporary Philosophical Filmmaking......Page 496
Some Preliminaries......Page 498
Raising a Philosophical Question: Waking Life......Page 499
Presenting a Counterexample to a Philosophical Thesis: Crimes and Misdemeanors......Page 503
Illustrating a Philosophical Thesis: Anomalisa......Page 504
Providing Empirical Support for a Philosophical Thesis: The Act of Killing......Page 509
Presenting and Supporting a Philosophical Thesis: Amour......Page 513
Conclusion......Page 515
Bibliography......Page 516
Chapter 22: Filmosophy/Film as Philosophy......Page 517
Film-Philosophy/Film as Philosophy......Page 518
Early Film-Philosophy......Page 519
Cavell......Page 523
Deleuze......Page 526
Thought and Cinema......Page 527
Frampton’s Filmosophy......Page 531
Film-Philosophy/Film as Philosophy......Page 534
‘Bold’ Film-Philosophy (Mulhall)......Page 535
‘Moderate’ Film-Philosophy (Wartenberg)......Page 538
Conclusion......Page 539
Bibliography......Page 541
Part V: Auteur Theory, the Avant-Garde and New Filmmakers......Page 544
Chapter 23: The Auteur Theory in the Age of the Mini-Series......Page 545
Chapter 24: The Question of Poetic Cinema......Page 552
What Is Poetry?......Page 553
Filmed Poems: Adaptation, Illustration?......Page 555
Maya Deren......Page 557
Piotrovskij......Page 558
Pasolini......Page 560
Epstein......Page 562
Poets and the Cinema: Surrealism......Page 565
American Avant-Garde Cinema......Page 567
Bibliography......Page 570
Chapter 25: Avant-Garde Films as Philosophy......Page 573
Avant-Garde Films Illustrating Philosophy......Page 577
Avant-Garde Films Originating Philosophy......Page 585
Avant-Garde Films Enacting Philosophy......Page 591
Avant-Garde Films Occasioning Philosophical Reflection......Page 594
Bibliography......Page 598
Part VI: Documentary......Page 601
Chapter 26: Show and Tell: The Identification of Documentary Film......Page 602
“No, it doesn’t matter”......Page 603
“Yes, it is important to know whether you’re watching fiction or nonfiction film”......Page 607
Showing: The Realist Approach......Page 609
Telling: The Relational View......Page 616
Show and Tell......Page 622
Bibliography......Page 624
Chapter 27: The Autobiographical Documentary......Page 626
Narrative: On the Differences Between Filmic and Literary Autobiographies......Page 629
Fiction and Nonfiction: Perspectives Within Documentary Filmmaking......Page 634
The Autobiographical Act......Page 637
A Brief History of the Autobiographical Documentary in North America......Page 638
The Question of Evidence......Page 640
The Autobiographical Act: An Alternative Perspective......Page 643
Bibliography......Page 647
Part VII: Movies and Society......Page 650
Chapter 28: Feminist Philosophy of Film......Page 651
Images of Women......Page 652
Spectatorship and the Male Gaze......Page 653
Audience-Film Negotiation......Page 657
Cognitivism......Page 660
Ideology Critique......Page 662
Constructive Feminist Philosophy of Film......Page 664
Subversion of Patriarchal Ideas......Page 665
Development of a Resistant Imagination......Page 667
Expansion of the Feminist Imagination......Page 670
Conclusion......Page 671
Bibliography......Page 672
Chapter 29: Race in Film......Page 674
A Brief History of Race in Cinema......Page 675
On Interpretation......Page 681
Aesthetics of Race in Film......Page 684
Double Consciousness of Race in Cinema......Page 685
Political Aesthetics of Race in Film......Page 688
And Yet to Come......Page 694
I......Page 695
II......Page 697
III......Page 701
IV......Page 704
V......Page 707
VI......Page 712
VII......Page 716
Chapter 31: Film, Art, and Pornography......Page 717
Definitions, or What Is Pornography?......Page 718
Can Porn Be Art?......Page 729
The Value of Pornography......Page 742
Conclusion......Page 750
Introduction......Page 752
Propaganda Simpliciter......Page 754
Neutral Propaganda......Page 755
Negative Propaganda......Page 758
Omnipresent Propaganda......Page 760
Top-down Propaganda......Page 765
Heterogeneous Propaganda......Page 769
Bibliography......Page 774
Part VIII: Movies and the Arts......Page 776
Chapter 33: Film and Fine Art: Automatism, Automata and “The Myth of Total Cinema” in The Red Shoes and The Tales of Hoffmann......Page 777
Part One: The Defensive Legacy of Orthodox Film Theory......Page 778
Part Two: Automatism, Automata and the “Myth of Total Cinema”......Page 782
Part Three: The “Possibilities” of Film and Philosophy......Page 789
Bibliography......Page 793
Chapter 34: The Sonic Art of Film and the Sonic Arts in Film......Page 795
Two Overviews: Film Music and Philosophy of Film Music......Page 797
Turn It Up: Music Matters to Movies......Page 802
The Sonic Arts of Film......Page 807
Film Music and Theories of Music......Page 810
Conclusion......Page 812
Bibliography......Page 813
Chapter 35: Adaptation, Translation, and Philosophical Investigation in Adaptation......Page 816
I......Page 817
II......Page 821
III......Page 830
Part IX: Emotions and Psychology......Page 835
Introduction......Page 836
Cognitive Theory of the Imagination......Page 837
Fictional Truth in Film......Page 838
The Nature of the Imaginative Experience Elicited by Film......Page 842
Affective Response......Page 845
Simulation and Identification......Page 849
Bibliography......Page 852
Chapter 37: Empathy and Sympathy: Two Contemporary Models of Character Engagement......Page 855
A Note on Identification......Page 856
Empathy......Page 859
Basic or Lower-Level Empathy: Emotional Contagion and Kinesthetic Mimesis......Page 860
Higher-Level Empathy......Page 864
Sympathy......Page 868
Conclusions......Page 876
Bibliography......Page 877
Chapter 38: Affect and Motion Pictures......Page 882
Theories of Emotion......Page 883
An Integrative Approach......Page 886
Other Affective States......Page 887
Biology Versus Culture......Page 888
Empathy and Alternatives......Page 891
Do Movies Arouse Real Emotions?......Page 893
How Do Movies Arouse?......Page 896
The Paradox of Negative Emotions......Page 898
Emotion and Identity......Page 902
Conclusions......Page 903
Bibliography......Page 904
Chapter 39: Psychoanalysis and the Philosophy of Film......Page 911
Questions About Applying Psychoanalysis......Page 912
Plato on Poetry......Page 913
Why Apply Psychoanalysis?......Page 914
Poetry Without Pleasure......Page 915
Ambiguity of Application......Page 916
Pathography......Page 917
Character, Work......Page 919
The Special Case of Analysts in Films......Page 922
Film Theory......Page 924
Audience as Analysand......Page 927
Feminist Theory......Page 930
Stanley Cavell......Page 931
Bibliography......Page 932
Part X: Alternative Media......Page 934
Chapter 40: The Television Medium......Page 935
An Art of Television?......Page 940
Television Aesthetics......Page 945
Bibliography......Page 954
Introduction......Page 957
Defining Videogames......Page 959
Videogames as Films......Page 961
The Medium of the Videogame and the Art Form of the Videogame......Page 965
The Art Status of Films and Videogames......Page 966
Videogames as Art......Page 967
Videogames and Interactivity......Page 969
Smuts on Interactivity......Page 970
Lopes and Gaut on Weak and Strong Interactivity......Page 971
Self-Involvement and Interactivity......Page 973
The Ethics of Films and Videogames......Page 974
Videogames, Art and Ethics......Page 975
The Ethics of Interactivity......Page 976
Bibliography......Page 978
Technological Variety of Virtual Reality......Page 981
A Brief History of Virtual Reality as an Art Medium......Page 983
VR and Cinema: A Stylistic Continuity or Rupture?......Page 988
Body Ownership Illusions and the Unique Potential of VR......Page 992
VR Uniqueness in Neuroscientific Terms......Page 995
Index......Page 1001

Citation preview

The Palgrave Handbook of the Philosophy of Film and Motion Pictures Edited by  Noël Carroll Laura T. Di Summa Shawn Loht Foreword by  Errol Morris

The Palgrave Handbook of the Philosophy of Film and Motion Pictures

Noël Carroll Laura T. Di Summa  •  Shawn Loht Editors

The Palgrave Handbook of the Philosophy of Film and Motion Pictures

Editors Noël Carroll The Graduate Center City University of New York New York, NY, USA

Laura T. Di Summa William Paterson University Wayne, NJ, USA

Shawn Loht Baton Rouge Community College Baton Rouge, LA, USA

ISBN 978-3-030-19600-4    ISBN 978-3-030-19601-1 (eBook) © The Editor(s) (if applicable) and The Author(s) 2019 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover illustration: TEK IMAGE / Getty Images This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland


There really is no disentangling philosophy and film. Film is part of philosophy; philosophy is part of film. Most major philosophical issues are expressed in film in one way or another because film, properly conceived, is a way of thinking about the world. It’s about the relationship between our perception of the world and the world itself. It’s riddled with fundamental issues of epistemology, philosophy of language, philosophy of perception, philosophy of mind, moral philosophy, and on, and on. Let me give an example, which I use in my recent book The Ashtray: one of my favorite John Ford movies, The Man Who Shot Liberty Valance (1962). Looked at from one perspective, The Man Who Shot Liberty Valance is an extended essay on reference, meaning and the coherence versus the correspondence theories of truth. Following Horace Greeley’s injunction to “Go West, young man,” Ransom Stoddard (James Stewart) takes his law school diploma and hangs out his shingle in Shinbone, a lawless frontier town in a life-or-death struggle with Liberty Valance (Lee Marvin), gunslinger and goon of the cattle barons. In the final shoot-out, Stoddard seems to kill Valance. But when his conscience revolts against his newfound notoriety, Tom Doniphon (John Wayne) recalls for him—and we, the audience, see—that it was really Doniphon who shot Liberty Valance from the shadows across the street. We, the viewers, and a few people in the world of the film—admittedly a fictional world but a world that we can easily imagine—believe that Tom Doniphon shot Liberty Valance. But everyone else believes that Ransom Stoddard shot Liberty Valance. We know that one view is true and one view is false. That is, one view is true and one view is false in the world of the film. And, make no mistake, films endlessly conjure with fictive worlds which are meant to be seen as real worlds in the sense that we are asked to think about what people believe in them and how they act on their beliefs. It’s at the very end of the film where these issues come directly into play. Ransom Stoddard and his wife Hallie (Vera Miles) are returning to Washington v



from Shinbone on the train; the conductor explains to both of them that he will do all in his power to ensure that they make all their connections and arrive in Washington on time. He adds as a rejoinder, “Nothing is too good for the man who shot Liberty Valance.” He says this while looking at James Stewart (Ransom Stoddard). Who is the conductor referring to? Is he referring to Ransom Stoddard or is he referring to Tom Doniphon? The conductor in the world of the film clearly thinks that he’s referring to Ransom Stoddard. But is he? On the one hand, his description uniquely identifies Doniphon; on the other, his intention is to pick out Stoddard. This has become a sticky wicket in the world of language philosophy—endless debates between people who feel that proper names are sticky labels and people who think that proper names are disguised definite descriptions (and, to be honest, a whole lot of complicated examples in between). But the point is that it’s just this kind of ambiguity in language that gives John Ford’s epic an additional oom-pah-pah. Because we’re thrown back on the ambiguities, the ironies, the uncertainties of the film. And in considering them, they multiply before our eyes. Is Tom Doniphon good or bad? He’s good because he allowed Ransom Stoddard to go on and serve his country and attain statehood for his unnamed western state. He’s bad because he committed a cold-blooded murder. A murder no matter how you look at it—possibly justifiable homicide, not according to the law, but by some higher moral principle if there is one. The Man Who Shot Liberty Valance also has much to teach us about ideas of truth and heroism. After all, what can be more heroic than giving up everything for the woman you love and losing that woman in the process, as does Doniphon? What does such a movie have to do with philosophy? There’s a simple answer to that: EVERYTHING. It’s like one of those pictures you’re given and asked, how many animals can you identify in this image? How many philosophical issues can you identify in The Man Who Shot Liberty Valance? My guess is hundreds. Instead of trying to expound them, I would like to offer another example: Carol Reed’s The Fallen Idol (1948). Consider how our picture of the world is changed by adding and subtracting information. In The Fallen Idol, the young protagonist Phillipe (Bobby Henrey) has one version of reality, the police have another. And a man’s life hangs in the balance. Movies are particularly good at this—at providing versions that call attention to what is real and what is imagined. In that sense, they are ultimately epistemic. (I could be, perhaps, charged with indulging in that same kind of device in The Thin Blue Line.) The Fallen Idol is an especially nuanced film in this respect. The different perspectives it presents challenge an audience to confront both the mystery of what happened and its role as investigator and interpreter of that mystery. Near the middle of the film, Phillipe, a diplomat’s son in the care of his embassy’s staff, finds himself in the middle of a domestic crisis. He is roused by a quarrel between his father’s butler, Baines (Ralph Richardson), and Mrs. Baines (Sonia Dresdel). Philippe spies the couple arguing at the top of the



embassy’s grand staircase. He runs out to the fire escape and down a level to get a better view as the fight intensifies and gets physical. Just when it looks like Baines is about to throw his wife down the stairs, Phillipe goes down another level, anticipating her fall. He can’t see that the fight de-escalates as he does. Baines calmly walks off while his wife, who is trying to confront Baines’s mistress, runs to a nearby window and begins to pound on it. The bottom of the window tilts inward from the force of her blows at the top, knocking Mrs. Baines off her feet and down the stairs. It is at this point that Philippe arrives on the next level of the fire escape and has his expectations confirmed when he sees Mrs. Baines’s body tumbling down the last few stairs. As the camera follows Phillipe down the fire escape, we can peer into the building through windows spaced widely apart. We see that Phillipe does not have an unobstructed view of what is going on inside the building. We know, then, that we cannot trust his version of events. But can we trust our own? Phillipe’s perspective is an almost perfect metaphor for film itself. For the now old-fashioned idea of the persistence of vision. A strip of 35-mm film has 16 images every foot and is projected at a film speed of 24 images a second—24 images separated by strips of black. How neatly this matches Phillipe’s warped perception as he runs down the fire escape: the dark, blank wall of the embassy exterior and the images that he sees through the windows. Ultimately, there is no separating philosophy from film or film from philosophy. They’re interwoven with each other. I could name a few more themes, but it’s clearly a partial, incomplete list. Images and reality. The motivation of characters that are partially real and partially make-believe. Documentary (of all stripes and descriptions) and drama. Epistemology, ontology, and everything else you could imagine. The question isn’t whether there is philosophy in film. The question should be: how could it be otherwise? FilmmakerErrol Morris New York, NY, USA


Part I The Medium in Film and Motion Pictures   1 1 Film Ontology: Extension, Criteria and Candidates  3 Frank Boardman 2 Medium Specificity 29 Noël Carroll 3 The Moving Image 49 Nick Wiltsher and Aaron Meskin 4 The Art of Cinematography 71 Patrick Keating Part II The Structure of Film and Motion Pictures  95 5 Silly Questions and Arguments for the Implicit, Cinematic Narrator 97 Angela Curran 6 Narrative and the Moving Image119 Patrick Keating 7 On Rhythm in Film Editing143 Karen Pearlman 8 Animation165 David Davies ix



9 Sound in Film189 Paloma Atencia-Linares 10 What Is a Screenplay?215 Ted Nannicelli Part III Approaches and Schools 235 11 Analytic Philosophy of Film: (Contrasted with Continental Film Theory)237 Richard Eldridge 12 When the Twain Shall Meet: On the Divide Between Analytic and Continental Film Philosophy259 John Ó Maoilearca 13 The Phenomenological Movement in Context of the Philosophy of Film and Motion Pictures285 Shawn Loht 14 Ideology and Experience: The Legacy of Critical Theory315 Espen Hammer 15 Stanley Cavell: What Becomes of People on Film?335 Paul Guyer 16 Film Art from the Analytic Perspective357 Deborah Knight 17 Cognitive Theory of the Moving Image381 Carl Plantinga 18 Aesthetic Criticism409 Andrew Klevan 19 Poststructuralism and Film441 Robert Sinnerbrink



Part IV Philosophy Through Film 467 20 Thoughtful Films, Thoughtful Fictions: The Philosophical Terrain Between Illustrations and Thought Experiments469 E. M. Dadlez 21 Contemporary Philosophical Filmmaking491 Thomas E. Wartenberg 22 Filmosophy/Film as Philosophy513 Robert Sinnerbrink Part V Auteur Theory, the Avant-Garde and New Filmmakers 541 23 The Auteur Theory in the Age of the Mini-Series543 Douglas Lackey 24 The Question of Poetic Cinema551 Tom Gunning 25 Avant-Garde Films as Philosophy573 Malcolm Turvey Part VI Documentary 601 26 Show and Tell: The Identification of Documentary Film603 Vitor Moura 27 The Autobiographical Documentary627 Laura T. Di Summa Part VII Movies and Society 651 28 Feminist Philosophy of Film653 Zoë Cunliffe



29 Race in Film677 Lewis R. Gordon 30 How Do We Look So Far? Notes Toward a Queer-Film Philosophy699 David A. Gerstner 31 Film, Art, and Pornography721 Jacob M. Held 32 Propaganda and the Moving Image757 Sheryl Tuttle Ross Part VIII Movies and the Arts 781 33 Film and Fine Art: Automatism, Automata and “The Myth of Total Cinema” in The Red Shoes and The Tales of Hoffmann783 Kristin Boyce 34 The Sonic Art of Film and the Sonic Arts in Film801 John Dyck 35 Adaptation, Translation, and Philosophical Investigation in Adaptation823 Garry L. Hagberg Part IX Emotions and Psychology 843 36 Imagination and Film845 Jonathan Gilmore 37 Empathy and Sympathy: Two Contemporary Models of Character Engagement865 Daniel Jerónimo Tobón 38 Affect and Motion Pictures893 Jesse Prinz 39 Psychoanalysis and the Philosophy of Film923 Nickolas Pappas



Part X Alternative Media 947 40 The Television Medium949 Ted Nannicelli 41 Videogames and Film971 Jon Robson and Aaron Meskin 42 Virtual Reality as an Emerging Art Medium and Its Immersive Affordances995 Gal Raz Index1015

Notes on Contributors

Paloma Atencia-Linares  is a research associate at the National Autonomous University of Mexico (UNAM). She did her PhD in Philosophy at University College London (UCL), was a lecturer at the University of Kent, UK, for two years and worked for six years as a sound designer for TV and films. Frank  Boardman is Assistant Professor of Philosophy at Worcester State University. He has worked predominantly in the philosophy of art, with excursions into logic, ethics, metaphysics, and epistemology. In the philosophy of film, he has so far been primarily concerned with film ontology, the rhetorical and cognitive values of film, and the nature and criticism of concert films. Kristin Boyce  is Assistant Professor of Philosophy and a faculty fellow at the Shackouls Honors College, Mississippi State University. She received a doctorate in Philosophy from the University of Chicago in 2010. She is the recipient of numerous awards, including an American Council of Learned Societies (ACLS) New Faculty Fellowship, a post-­doctoral fellowship from Stanford University, and a Josephine De Karmán Dissertation Fellowship. Her primary research interests are in philosophy of art, history of early analytic philosophy, and Wittgenstein. Noël Carroll  is Distinguished Professor of Philosophy at the Graduate Center of the City University of New York. He has written over 15 books, including, most recently, Humour: A Very Short Introduction. He has also been a journalist and has written five documentaries. Zoë Cunliffe  is a PhD student at the City University of New York’s Graduate Center. Her philosophical interests include aesthetics, social and political philosophy, and social epistemology. Most recently she delivered a paper, “Narrative Fiction and Epistemic Injustice,” at a meeting of the American Society for Aesthetics. When not doing philosophy, she is usually eating Asian food, binge-watching films and television, or exploring the city.




Angela Curran  is a visiting assistant professor at Kansas State University. She works on topics in ancient Greek philosophy, aesthetics, and philosophy of film, and is author of The Routledge Philosophy Guidebook to Aristotle and the Poetics (2015). E.  M.  Dadlez has a PhD in Philosophy from Syracuse University and is Professor of Philosophy at the University of Central Oklahoma. Her work is mainly on the philosophy of art and literature, and on topics at the intersection (sometimes, more accurately, the collision) of aesthetics, ethics, and epistemology. She is author of various articles on aesthetics and feminist ethics, as well as of What’s Hecuba to Him? Fictional Events and Actual Emotions and Mirrors to One Another: Emotion and Value in Jane Austen and David Hume. She is editor of Jane Austen’s Emma: Philosophical Perspectives. David Davies  is Professor of Philosophy at McGill University. He is author of Art as Performance (2004), Aesthetics and Literature (2007), and Philosophy of the Performing Arts (2011). He is also editor of The Thin Red Line (2008) and co-editor of Blade Runner (2015), both in the Routledge series Philosophers on Film. He has published widely on philosophical issues in the metaphysics and epistemology of art; issues relating more specifically to film, photography, performance, music, literature, visual art, and dance; and issues in general metaphysics, philosophy of mind, and philosophy of language. Laura T. Di Summa  is an assistant professor at William Paterson University. She received her PhD from the City University of New York’s Graduate Center in 2014, where she worked with Prof. Noël Carroll. She has written and presented papers on autobiography and its developments, philosophy of motion pictures, everyday aesthetics, and issues related to the cognitive analysis of visual arts. She has been the Managing Editor of The Philosophical Forum since 2010. John Dyck  is a PhD candidate at the City University of New York’s Graduate Center, finishing his dissertation on aesthetic value. He also writes on philosophy of music. Richard  Eldridge is Charles and Harriett Cox McDowell Professor of Philosophy at Swarthmore College. He has held visiting appointments in Bremen, Erfurt, and Freiburg (Germany); Essex (UK); Sydney (Australia); and Stanford and Brooklyn (US). He has written widely on aesthetics, the philosophies of literature and film, German Idealism and Romanticism, and the philosophy of language, including, most recently, Images of History: Kant, Benjamin, Freedom and the Human Subject (2016) and Werner Herzog: Filmmaker and Philosopher (2019). David  A.  Gerstner  is Professor of Cinema Studies in the Department of Media Culture at the City University of New York’s (CUNY) College of Staten Island. He also serves as a member of the doctoral faculty at the CUNY Graduate Center. His books include: Christophe Honoré: A Critical Introduction



(with Julien Nahmias, 2015), Queer Pollen: White Seduction, Black Male Homosexuality, and the Cinematic (2011, Choice Outstanding Academic Title, 2012), Manly Arts: Masculinity and Nation in Early American Cinema (2006), and The Routledge International Encyclopedia of Queer Culture (editor, 2006— New York Public Library “Best of Reference,” 2007). His co-edited works include Media Authorship (with Cynthia Chris) and Authorship and Film (with Janet Staiger). He is editor of the book series Queer Screens. Jonathan Gilmore  is Associate Professor of Philosophy at the City University of New York, the Graduate Center and Baruch College. He is author of Apt Imaginings: Feelings for Fictions and Other Creatures of the Mind. Lewis R. Gordon  is Professor of Philosophy at the University of Connecticut (UCONN)-Storrs, Honorary President of the Global Center for Advanced Studies, and Honorary Professor in the Unit for the Humanities at Rhodes University in South Africa. His books are Fear of Black Consciousness and his collection of essays 论哲学、去殖民化与种族 (“On Philosophy, Decolonization, and Race”), trans. Li Beilei. Tom  Gunning is Distinguished Service Professor in the Department of Cinema and Media at the University of Chicago, and author of D.W. Griffith and the Origins of American Narrative Film, The Films of Fritz Lang; Allegories of Vision and Modernity, and over 150 articles. Paul  Guyer  is Jonathan Nelson Professor of Humanities and Philosophy at Brown University and Florence R.C.  Murray Professor in the Humanities emeritus at the University of Pennsylvania. He received his AB and PhD from Harvard under the supervision of Stanley Cavell. He is author of ten books on Kant, including Kant’s aesthetics, and of A History of Modern Aesthetics in three volumes (2014). He was General Co-Editor of the Cambridge Edition of the Works of Immanuel Kant, and in that series was co-­editor and co-translator of the Critique of Pure Reason and the Critique of the Power of Judgment and editor and co-translator of Kant’s Notes and Fragments. With Ted Cohen and Hilary Putnam, he edited Pursuits of Reason, a volume of essays on the work of Stanley Cavell (1993). He is a fellow of the American Academy of Arts and Sciences. Garry  L.  Hagberg is James H.  Ottaway Professor of Philosophy and Aesthetics at Bard College, and has also held a Chair in the School of Philosophy at the University of East Anglia. Author of numerous papers at the intersection of aesthetics and the philosophy of language, his books include Meaning and Interpretation: Wittgenstein, Henry James, and Literary Knowledge and Art as Language: Wittgenstein, Meaning, and Aesthetic Theory; his most recent book is Describing Ourselves: Wittgenstein and Autobiographical Consciousness. He is editor of Art and Ethical Criticism and Fictional Characters, Real Problems: The Search for Ethical Content in Literature as well as of his most recent edited volume, Wittgenstein on Aesthetic Understanding. Co-editor of



A Companion to the Philosophy of Literature and editor of the journal Philosophy and Literature, Hagberg is presently writing a new book on the contribution literary experience makes to the formation of self and sensibility, Living in Words: Literature, Autobiographical Language, and the Composition of Selfhood, and editing a volume Stanley Cavell on Aesthetic Understanding. Espen Hammer  is Professor of Philosophy at Temple University. He is author of, among other books, Stanley Cavell: Skepticism, Subjectivity, and the Ordinary (2002), Adorno and the Political (2005), Philosophy and Temporality from Kant to Critical Theory (2013), and Adorno’s Modernism: Art, Experience, and Catastrophe (2015). He is editor of German Idealism: Contemporary Perspectives (2007), Theodor W.  Adorno II: Critical Assessments of Leading Philosophers (2015), and Kafka’s The Trial: Philosophical Perspectives (2018). He is also a co-editor with Peter E.  Gordon and Max Pensky of The Blackwell Companion to Adorno (forthcoming) and a co-editor with Axel Honneth and Peter E. Gordon of The Routledge Companion to the Frankfurt School (2018). Jacob  M.  Held  is Associate Professor of Philosophy and Director of the University of Central Arkansas (UCA) Core (General Education) at the UCA in Conway, Arkansas. His philosophical work focuses on political and legal theory, pornography, and popular culture and philosophy. His most recent works include The Philosophy of Sex (7th edition), edited with Raja Halwani, Alan Soble, and Sarah Hoffman (2017), Stephen King and Philosophy (2016) and Wonder Woman and Philosophy (2017). Dr. Held focuses primarily on scholarly work that is accessible to and engages a broad audience. Outside of the academy his work has been featured in venues such as The Huffington Post, The Guardian, and PhilosophyTalk. Patrick Keating  is an associate professor in the Department of Communication at Trinity University, where he teaches courses in film studies and video production. He is author of Hollywood Lighting from the Silent Era to Film Noir (2010) and editor of Cinematography (2014). His most recent book is The Dynamic Frame: Camera Movement in Classical Hollywood (forthcoming in 2019), for which he was named an Academy Film Scholar by the Academy of Motion Pictures Arts and Sciences. Andrew Klevan  is Associate Professor of Film Studies at the Oxford University. He is author of Aesthetic Evaluation and Film (2018), Barbara Stanwyck (2013), Film Performance: From Achievement to Appreciation (2005), and Disclosure of the Everyday: Undramatic Achievement in Narrative Film (2000). He is also co-editor of The Language and Style of Film Criticism (2011). Deborah  Knight teaches philosophy at Queen’s University, Kingston, Canada. Her early academic career in film studies was cut short because she supported analytic film theory in the days of dogmatic poststructuralism. She



has written on filmmakers including Hitchcock, Eastwood, and Nolan and on film genres including the Western, science fiction, detective fiction, and horror. She has written about the structure of film narratives, emotional engagement with fictional characters, sentimentality, and ethical criticism. Forthcoming work will allow her to revisit Blade Runner as well as her skepticism about the claim that certain feature-length fiction films are philosophical thought experiments. Douglas  Lackey is Professor of Philosophy at Baruch College and the Graduate Center, City University of New York, where he has taught courses on film theory and the history of film. He is interested in connections between perception theory and film, and action theory and dance aesthetics. Shawn  Loht  has written on phenomenology, the philosophy of film, and ancient Greek philosophy. His recent publications include the monograph Phenomenology of Film: A Heideggerian Account of the Film Experience (2017). He has taught in the philosophy departments at Tulane University, Mercer University, and Pennsylvania State University. Aaron  Meskin  is Professor of Philosophical Aesthetics at the University of Georgia. He works on a variety of issues in aesthetics, the philosophy of food, and philosophical psychology. He has authored numerous journal articles and chapters and co-edited five books, including The Routledge Companion to Comics (2016), Aesthetics and the Science of Mind (2014), and The Art of Comics: A Philosophical Approach (2012). Vitor Moura  is an assistant professor at the Universidade do Minho, Portugal, where he presently teaches a number of courses, ranging from logic to aesthetics of architecture. He received his PhD from the University of Wisconsin-­Madison in 2006. His research projects address the issue of intentionalism in aesthetic interpretation, and perception and aesthetic experience. He is also a member of the Executive Committee of the European Society for Aesthetics. Ted  Nannicelli teaches at the School of Communication and Arts, The University of Queensland. He is author of A Philosophy of the Screenplay (2013), Appreciating the Art of Television (2017), and Artistic Creation and Ethical Criticism (forthcoming). John Ó Maoilearca  is Professor of Film at Kingston School of Art, Kingston University, London. He has also taught philosophy and film theory at the University of Sunderland, England, and the University of Dundee, Scotland. He has published ten books, including (as author) Bergson and Philosophy (2000), Post-Continental Philosophy: An Outline (2006), Philosophy and the Moving Image: Refractions of Reality (2010), and (as editor) Bergson and the Art of Immanence (2013) and The Bloomsbury Companion to Continental Philosophy (2013). His last book was All Thoughts Are Equal: Laruelle and Nonhuman Philosophy (2015).



Nickolas Pappas  is Executive Officer and Professor of Philosophy at the City University of New  York’s Graduate Center. He is author of several books, including the Guidebook to Plato’s Republic (now in its third edition) and most recently The Philosopher’s New Clothes: The Theaetetus, the Academy, and Philosophy’s Turn against Fashion. His numerous shorter pieces cover topics in ancient philosophy and aesthetics. Karen  Pearlman  is a senior lecturer at Macquarie University (Sydney) and author of Cutting Rhythms, Intuitive Film Editing (2016). Her creative practice research film Woman with an Editing Bench (2016) won the Australian Teachers of Media (ATOM) Award for Best Short Fiction and the Australian Screen Editors Guild (ASE) Award for Best Editing in a Short Film along with six other film festival prizes. Other publications from Karen’s ongoing research into editing, cognition and feminist film histories include “Editing and Cognition Beyond Continuity” in Projections, The Journal of Movies and Mind (2017), “Documentary Editing and Distributed Cognition” in A Cognitive Approach to Documentary (Palgrave Macmillan 2018), co-editing a special issue of the journal Apparatus titled “Women at the Editing Table: Revising Soviet Film History of the 1920s and 30s” (2018), and the essay film After the Facts (2018), which has screened at many international film festivals, won “Best Editing” in Open Content at the ASE awards 2018, and has been accepted for publication in [in]Transition in 2019. Carl Plantinga  is Professor of Film and Media at Calvin College and former president of the Society for Cognitive Studies of the Moving Image. He is author of Screen Stories: Emotion and the Ethics of Engagement (2018), Moving Viewers: American Film and the Spectator’s Experience (2009), and Rhetoric and Representation in Nonfiction Film (1997). Plantinga also co-edited The Routledge Companion to Philosophy and Film (with Paisley Livingston) and Passionate Views: Film, Cognition and Emotion (with Greg M. Smith). Jesse Prinz  is Distinguished Professor of Philosophy at the Graduate Center of the City University of New  York and was Einstein Visiting Fellow at the Berlin School of Mind and Brain. Prinz is interested in how the mind works and how philosophical accounts of the mental can be informed by findings from psychology, the neurosciences, anthropology, and related fields. His research interests include emotion, consciousness, cultural cognition, concepts, perception, moral psychology, and aesthetics. Much of his work is a continuation of the classical empiricist tradition, which emphasizes the role of perceptual experience and socialization in grounding our cognitive capacities. Prinz is author of over 100 articles, and several books, including: Furnishing the Mind, Gut Reactions, The Emotional Construction of Morals, The Conscious Brain, and Beyond Human Nature. Two other books, Works of Wonder and The Moral Self, are forthcoming, and a book on social construction is in progress.



Gal Raz  is a visiting lecturer at the Steve Tisch School of Film and Television and the Sagol School of Neuroscience, and a researcher at the Sagol Institute for Brain Functions, Tel-Aviv Sourasky Medical Center. His fields of research include brain network dynamics, neuroscience of empathy, cinematic empathy, neuoroaesthetics of motion pictures, audiovisual brain computer interfaces, and virtual and augmented reality. Jon Robson  is Assistant Professor of Philosophy at the University of Nottingham. He has written extensively in a range of areas within aesthetics (especially on the epistemology of aesthetic judgments). Outside of aesthetics, he works in a range of other philosophical sub-disciplines, including epistemology, ethics, metaphysics, and the philosophy of religion. He is co-editor of Aesthetics and the Sciences of Mind (2014) and The Aesthetics of Videogames (2018) as well as co-author of A Critical Introduction to the Metaphysics of Time (2016). Sheryl  Tuttle  Ross is Full Professor of Philosophy at the University of Wisconsin-La Crosse. She researches at the intersection of art, politics and morality. She has written extensively about propaganda, developing the Epistemic Merit Model and its application to a variety of artworks, media and social circumstances. She is currently expanding her research focus to include political humor and aesthetic akrasia. Robert  Sinnerbrink is Associate Professor of Philosophy at Macquarie University, Australia. He is author of Understanding Hegelianism (2007), New Philosophies of Film: Thinking Images (2011), Cinematic Ethics: Exploring Ethical Experience through Film (2016), and Terrence Malick: Filmmaker and Philosopher (2019). He is also on the editorial boards of the journals Film and Philosophy, Film-Philosophy, and Projections: The Journal of Movies and Mind. Daniel Jerónimo Tobón  teaches at the Universidad de Antioquia, Colombia, and is a doctoral candidate at the Universidad Nacional de Colombia. He has written on aesthetics, contemporary Colombian art and film, art and memory, and emotions. Malcolm Turvey  is Sol Gittleman Professor in the Department of Art and Art History and Director of the Film and Media Studies Program at Tufts University. He is also an editor of the journal October. He is author of Doubting Vision: Film and the Revelationist Tradition (2008) and The Filming of Modern Life: European Avant-Garde Film of the 1920s (2011), and co-editor of Wittgenstein, Theory, and the Arts (2001). His Play Time: Jacques Tati and Comedic Modernism will be published by Columbia University Press in 2019. Thomas E. Wartenberg  is a senior research fellow in Philosophy at Mount Holyoke College. His main areas of focus are aesthetics, the philosophy of film, and philosophy for children. Among his publications are Thinking on Screen: Film as Philosophy, Unlikely Couples: Movie Romance as Social Criticism, Existentialism: A Beginner’s Guide, and Mel Bochner: Illustrating Philosophy.



He has edited or co-edited four volumes on philosophy and film, most recently Fight Club (Philosophers on Film). He is Film Editor for Philosophy Now. Nick Wiltsher  is Marie Skłodowska-Curie Individual Fellow at the University of Antwerp’s Centre for Philosophical Psychology. He has previously been employed at Auburn University (AL, US), the Federal University of Rio Grande do Sul (Porto Alegre, Brazil), and the University of Leeds (UK). He works on imagination in the philosophy of mind and on a number of topics in aesthetics.

List of Tables

Table 7.1 Table form distillation of the three questions being asked and the ideas being proposed in response to each Table 7.2 Table form distillation of the three operations, or tools, editors deploy and each of their three suboperations

148 157



At present, we are in the midst of a heyday in the philosophy of film, or, more accurately, the philosophy of the moving image or motion picture, whose practice comprises not only film, but television, video, video games, digital imaging, virtual reality, and technologies yet to come and which are delivered by a comparable array of platforms. Never before have so many philosophers, from such a wide variety of theoretical traditions, taken this measure of professional interest in the movies. These interests, moreover, range across the philosophical division of labor, raising metaphysical, epistemological, ethical, political, and, of course, aesthetical questions with respect to cinema. In this anthology, we have attempted to provide a wide sampling of the many issues from a diverse range of schools that are philosophically interrogating the moving image today. Philosophers such as Hugo Munsterberg and György Lukacs took an early interest in film, commenting upon it even before 1920. Moreover, writers who were often classified as film theorists—such as Béla Balász, Rudolf Arnheim, André Bazin, and V.F. Perkins—were, even when not card-carrying philosophers, nevertheless philosophically informed, and they pursued the philosophical investigation of film in the process of enfranchising film as an art form and establishing what they identified as its constitutive norms. In the English-speaking world, from the standpoint of professional philosophy, the publication in 1971 of Stanley Cavell’s The World Viewed: Reflections on the Ontology of Film was a seminal event. Here was a book-length treatise on film by a member of Harvard University’s philosophy department, arguably one of the most distinguished philosophy departments in the Anglo-American world, whose faculty at the time included many giants of twentieth-century philosophy (such as W.V.O. Quine and John Rawls). Cavell’s work provided both inspiration and legitimatization to younger philosophers who grew up movie-mad and aspired to unite their love of film with their love of philosophy. All the authors in this volume belong to the generations who, if they have not followed Cavell’s lead, have benefitted from the opportunity he made possible, if only institutionally. They have gone where philosophers have never gone xxv



before and recorded in depth the sophistication of the contemporary philosophical conversation focused on the moving image. Perhaps needless to say, this anthology has not covered every imaginable subject in the expanding area of the philosophy of cinema, a moving target, if there ever was one. But we, the editors, have attempted to canvas topics at greater length than previous volumes in this genre, and we have encouraged our authors to attempt original philosophizing in their contributions. In this regard, we hope that this volume will not only track past and recent achievements in the philosophy of the moving image but also pave the way for the future. Noël Carroll


The Medium in Film and Motion Pictures


Film Ontology: Extension, Criteria and Candidates Frank Boardman

Our concern in this chapter is with a set of issues central to the ontology of film. “Ontology” traditionally refers to the branch of philosophy that deals with questions about the being and nature of things. We can safely assume that films exist or even less controversially that there are some films. So, our first question in the ontology of film need not be “are there films?” but rather “what is film?” An even clearer statement of our primary question may be “what sort of thing is film?” Armed with an answer to that question, we could wade more confidently into secondary issues regarding films’ constitutive parts or the conditions under which they persist through time. But alas, while we have no shortage of available theories about the nature of film, there is nothing like consensus around any one of them. This has not been an entirely unfavorable condition, however. A good number of critical insights and useful observations about film have emerged out of philosophers’ debates, arguments and disagreements over film ontology. Another sort of disagreement, though, has been less beneficial. These are disagreements over the question itself, which largely emerge from two sources. First, we don’t enjoy antecedent agreement on the class of things that are films (the extension of “film”). And we can’t expect to agree on what makes things films if we don’t first agree on which things are films. Second, we don’t yet agree on what we’re asking. That is to say, we don’t agree on criteria for an adequate answer. So, instead of offering or defending a particular theory of film ontology in this chapter, I’d like to discuss the proper extension of “film” and propose a set

F. Boardman (*) Worcester State University, Worcester, MA, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




of criteria for selecting an ontological theory of it. I’ll then consider some ­current and likely candidate theories using those criteria. By way of conclusion, I’ll offer some suggestions for ways the discussion might move forward.

Determining the Extension of “Film” We should notice right away that there is no natural or privileged use of “film” or set of films independent of some particular theoretical or conversational interest. A frustrated aficionado of celluloid and acetate film might be interested in the nature of the set of objects that ranges from Louis Le Prince’s experiments to those few hold-out 35-mm theaters scattered around the world. But most of us do not think that the history of film ended with the swift triumph of digital projection in this decade. The films that dominate the cultural landscape are still being made, distributed, seen, discussed and written about, even if there is no longer any film involved. The “film” in this sense relies on Le Prince’s and Edison’s technologies, but doesn’t emerge until others use that technology to say and tell. And it is not entirely clear when that activity starts. Are the Lumiere brothers’ early projected films the start of this history or does it wait on the first narrative film (which may also have been theirs)? We can’t just be interested in narrative films per se, lest we exclude whole swaths of genuine films more likely to be seen at the gallery than the multiplex. But then again, it is not merely the technology of moving images in which we’re interested. Whatever films are, they don’t include security camera footage or Skype conversations. We should think of ourselves as primarily investigating an art form. This is not because we should assume from the outset that there is an ontological difference between last night’s security footage and Rear Window, but because only the art film (understanding that term as widely as is reasonably possible) provides us with a special investigative interest.1 The cultural, aesthetic and historical interests we take in Hitchcock’s masterpiece2 motivate questions about its nature. And because we take similar interests in other films, we should not assume that Rear Window is ontologically sui generis. Just as it is the art form status of films that arouses our interest in its ontology, the ontology of film has figured prominently in the establishment of film as an art. As moving pictures matured from curiosity to art form, the live-­ action narrative film naturally invited comparisons to two close cousins. Focusing on the technology and means of display, the film is kin to photography. It is in fact photography itself, multiplied, put in motion and coupled (eventually) with sound. Focusing instead on their content, films are in the family of theater and continue the ancient history of drama and comedy.

1  For a slightly different (and stronger) take on the discontinuity between the technical and artistic natures of film, see Jacques Rancière, Film Fables, Emiliano Battista, trans. (Oxford: Berg 2001) 11–12. 2  Yes it is.



This situation presented something of a dilemma for the early advocate of films’ art status. On the one hand, the art status of photography was itself in question. On the other, if films are artworks in the theatrical tradition, then the artwork in question may just be the actions filmed rather than the film itself. Thus, the price for resting the possibility of films’ art status on similarity with another art form is that it makes room for a skeptic to claim that film is no distinct form, but rather a mode of presentation for the form to which it’s compared. This central problem is still with us, even if the art status of (some) photography is more of a given today. If film can be an art form, it must be a kind of thing that is (a) capable of being art and (b) different in some significant and relevant way from existing art forms. Given their proximity, the pressing challenge for (b) is to distinguish film from photography and theater, though we want to be able to distinguish it from painting, poetry and music as well.3 We have good reason, then, to want an ontology of film as an art form. And, in fact, it is difficult to see what other compelling reason we could have for wanting an ontology of film. I don’t think I’m going out on a limb by thinking that the ontology of art is just more interesting than the ontology of security or communication technology. Even so, there does not necessarily need to be an ontological difference between the film and the security footage in order for there to be a difference in their art status, unless there is an ontological difference between art and nonart (but that would be another matter). There is, after all, at least a prima facie distinction to be made between ontological and art-status questions. In the present context, the former ask “what makes this thing a film and not something else?” While the latter ask, “what makes this film a work of art and not something that is not a work of art?” There are two reasons, though, to confine the extension of “film” in question to filmic works of art and exclude security camera footage, home movies, Skype conversations and the like. The first follows from the kind of interests we just discussed. It is captured in what David Davies calls the ‘pragmatic constraint’ on the ontology of art: Artworks must be entities that can bear the sorts of properties rightly ascribed to what are termed “works” in our reflective critical and appreciative practice; that are individuated in the way such “works” are or would be individuated, and that have the modal properties that are reasonably ascribed to “works,” in that practice.4

3  I should note that there is no universal agreement on the need to distinguish film from other arts. Alain Badiou, for instance, thinks that “cinema is nothing but editing” and that otherwise it is the “plus-one of the arts” taking from music, stage, literature and even painting without contributing anything essentially artistic of its own. Alain Badiou, Cinema, Susan Spitzer, Trans. (Cambridge: Polity, 2017) 97; 89. 4  David Davies, Art as Performance (Malden, MA: Blackwell, 2004), 18.



This “methodological principle guiding philosophical inquiry,” Davies notices, is at least implicitly at work in many and otherwise various ontologies of art.5 Sherri Irvin later makes explicit use of the same criterion, (rightfully) preferring to call it the “critical practice constraint.”6 Applying the critical practice constraint to film suggests an ontological difference between (for instance) security camera footage and live-action narrative films. It is inappropriate to criticize security footage for its aesthetic or otherwise formal faults. Regardless of how films are properly individuated, the question of individuation does not arise for security footage. Footage has duration, not individuality. And the same film could have different endings. If some duration of security footage has a different ending (and “ending” is not even the right word here), it is a different bit of footage. The second reason becomes apparent when we consider the difference between actual security camera footage and security-camera-footage-like shots in a film. Certain shots in The Wire (2002–2008) or The Bourne Identity (2002), for instance, let you know that a character is being filmed by providing the perspective of the security camera filming. The entire movie Look (2007) is shot from such perspectives. But in these instances, the shots themselves establish the existence of a fictional and unseen security camera. Actual security cameras require the existence of an actual unseen security camera. These are not merely ways of seeing the film and the footage. Something is fundamentally different about the two. One camera is fictional and the other actual. Surely, the difference between the fictional and the actual is ontological. This is not to say, however, that the ontological difference between security camera footage (and the like) and film as we mean to discover it must track the fiction/non-fiction divide. There are, after all, plenty of non-fiction films, both documentaries (in the traditional sense of “documentary,” we’ll consider another in a little bit) and nonnarrative (what are often and sometimes pejoratively called “experimental”) films. We could imagine, in fact, actual security footage being repurposed as such a film. But then it would have been transformed (or “transfigured” in Arthur Danto’s parlance).7 And that transformation is a transformation in the kind of thing that it is. It will go from a mere recording to something with a point, a meaning. It will go from having only duration to an individual identity. Its modal properties will change. It will become subject to aesthetic evaluation. It will, in short, become a new kind of thing. So, only those moving images8 that might be reasonably thought of as artworks are films in the sense under consideration. But there is still some work to be done to say which artworks are members of that class. The first couple of  Davies, Art as Performance, 18–19.  Sherri Irvin, “The Ontological Diversity of Visual Artworks” in New Waves in Aesthetics, ed. Kathleen Stock and Katherine Thomson-Jones (London: Macmillan, 2008), 2. 7  Arthur Danto, “The Transfiguration of the Commonplace,” The Journal of Aesthetics and Art Criticism 33 (1974). 8  Or, as we’ll see a little later, potentially moving images. 5 6



questions involve the technology of displaying films—even (as we discussed above) when the technology that is standard is changing over time. First, is such technology necessary? For instance, if films involve moving images, then so do flip books, and we might be reluctant to admit flip books into the relevant class.9 But perhaps we shouldn’t be too hasty. Bill Brand’s “Masstransiscope” is a series of paintings that appear as an animation to riders of the Q train after they leave Dekalb Avenue on their way to Manhattan from Brooklyn. The moving-image effect is delivered by stationary pictures viewed from a moving train rather than moving pictures viewed from a stationary position. The same visual effect could have been carried out in the normal way, with those particular pictures in motion. In fact, it is theoretically possible to set up giant slides of each frame of The Godfather (1972) separated by black pillars at just the right distance and have a train pass them all at just the right speed while piping in synchronized audio across the loudspeakers. The overall effect would be a viewing of The Godfather by other (albeit rather extraordinary) means. And The Godfather is a film. Then is the technology of film display enough to guarantee a film? Mike Hoolboom’s White Museum consists visually of half an hour of film leader (the empty film used by technicians to mark beginnings and endings) projected on a screen. Most of us would want to count this as a film, even if it is the kind that could only be made as a minimalist comment on the nature of film.10 So, once art status is established or assumed, it seems that uses of certain display technologies are sufficient for a work to be a film. But what about the images that are displayed? Most importantly, is there a useful ontological category that captures both animation and live-action films? Three considerations speak in favor of there being one, or at least our defaulting to the position that there is one in the absence of evidence to the contrary. First, there are the obvious commonalities. Both involve displayed moving images typically (these days anyway) accompanied by sound. Both typically tell stories, involve acting and have a common lineage as narratives in literature and theater. Second, animations and live actions occupy the same place in the cultural landscape. They are shown in the same theaters, on the same televisions and via the same websites. They’re reviewed and discussed in the same sections of newspapers. Third, if they were distinct categories, there would be a great and ever-greater number of bothersome borderline and otherwise unclear cases. There are the mixtures like Who Framed Roger Rabbit? (1988) and The Phantom Menace (1999), entirely Rotoscoped films like A Scanner Darkly (2006), and the action blockbusters that leave virtually no shot untouched by visual effects. Admittedly, none of these considerations are quite definitive. It may be that the right 9  Noël Carroll, for instance, thinks these are outside the relevant class. See “Defining the Moving Image,” in Philosophy of Film and Motion Pictures, ed. Noël Carroll and Jinhee Choi (Oxford: Blackwell, 2006) Especially 131. 10  For more on such cases and an account that puts the sufficiency of cinematic display technology front and center, see Trevor Ponech, “The Substance of Cinema,” The Journal of Aesthetics and Art Criticism 64 (2006).



­ ntological account separates animation from live action. But such an account o has the dual burdens of a proof that we tend to be in error and a diagnosis for that error. Finally, while we’re on the topic of borderline cases, there may be a question about the status of mixed-media works that involve film. Carl Hancock Rux’s Mycenaean involves live acting on a stage accompanied by projected moving images. And the band Neurosis used to accompany their performances with original video projected around them. But these sorts of works shouldn’t over-­ worry us. These hybrid works are plausibly combinations of various art forms.11 It seems unlikely that they would need any special ontological account. They are, as we should expect, works whose parts belong to various categories, whatever those categories turn out to be.

Criteria for an Ontology of Film Having sketched an outline for the extension of the operative sense of “film,” we’re in a better position to provide some criteria for a good theory of film ontology. If the goal of such a theory is to answer the question “What is film?” or “What makes this thing I’m watching a film and not something else?”, then the primary condition on a successful theory is that it provides necessary conditions for membership in that extension that together come as close as possible to being sufficient. That is to say, we’re after a set of conditions, each one of which must be present in something if that thing is to be a film, and (if possible) all of those conditions being present in something guarantees that that thing is a film. It may be that no set of conditions will be sufficient because films may share an ontology with non-films. But we should not assume this at the outset. And even if sufficiency is impossible or undesirable, we should still aim at coming as close as we can to it in order to say something interesting about the nature of film. So, we’re after properties common to everything in the class described above that help us distinguish those things from nonmembers. As should be clear from our earlier conversation, we at least want to be able to distinguish film from other art forms, especially photographs, paintings and theater. Much of the usual criteria for theory selection apply to ontologies of film in a number of interrelated ways. Such theories ought to be consistent, at least in the absence of evidence of some dastardly paradox waiting at the heart of film. They ought to be plausible as well. That is to say, we ought—ceteris paribus— to adopt the ontology that forces us to amend fewer of our present beliefs. Similarly, the best ontological theory will, again ceteris paribus, be simpler. We ought to prefer the theory that forces us to take on fewer new beliefs. Especially important in this context is that we should take on as few new ontological entity types as we reasonably can. The best theory will also produce more 11  My use of “hybrid” here is akin to Jerrold Levinson’s in “Hybrid Art Forms,” The Journal of Aesthetic Education 18 (1984).



­ tility. There is more that we can do with the most preferable theory, be it u opening new lines of research, informing our appreciation of films or otherwise guiding our actions. Finally, explanatory power is as important a criterion for adjudicating between ontological theories of film as it is in other contexts. So what ought to count as data for such a theory? For starters, an ontological theory ought to shed some light on our common experience of film and our everyday discourse about films. The latter will include a wide range of practices, from the writings of professional critics to watercooler talk about recent summer blockbusters. It’s this kind of consideration that motivates the “critical practice constraint” we discussed earlier. There may, however, be something wrong with our common experience of or discourse about film, and we should be open to the possibility that the best ontology of film can show us this. But again, this places a heavy burden of proof and diagnosis on the theory that would attempt to do so. In the absence of both of these, our default demand ought to be that a theory help us explain rather than explain away our typical experiences and practices. Of special interest regarding explanatory power are two puzzles that Danto introduces in “Moving Pictures,” both of which point to crucial and fundamental issues.12 The first, in very Danto-nian style, involves indiscernible counterparts: the photographic slide and the unmoving film image, projected side by side onto a screen or wall. The two are exactly the same to all appearances, and yet there is an important difference between them. Because that difference is a difference in kind or, to be a bit more dramatic, mode of being, it is incumbent on an ontology of film to explain the difference. One way of looking at this problem is that it raises the stakes on a prior condition on an ontology of film: that it should appropriately distinguish film and photography. The second puzzle that Danto considers has—I think unfortunately— received less attention, at least as an issue in film ontology. This is the distinction between what Danto calls a “screenplay” and a “documentary” film. Here, “screenplay” does not mean the textual plan for or documentation of a film, nor is “documentary” the genre designation with which we’re most familiar. Rather, they refer to two ways of viewing a film, which correspond to two ways of viewing given filmed objects. When we watch that famous scene in The Godfather, what we see is the result of the work of a camera trained on Al Pacino walking into a staged restaurant-like setting with a prop gun. When we see Al Pacino acting with a prop in his hand, we view them as motifs and The Godfather as documentary. We are literally seeing what the camera documents. Alternatively, we can see the same thing before us as Michael Corleone walking into a restaurant dining room with a gun in his hand. That is to view Pacino and the prop as models and The Godfather as screenplay. We do not yet have an ontological distinction (much less a problem) but rather two ways of seeing. The ontological issue arises when we notice that some films can only—at least  Arthur Danto, “Moving Pictures,” Quarterly Review of Film Studies 4 (1979).




without error—be viewed as documentaries, others as screenplays or documentaries. No one in the security camera footage is a model, and it is not at all a screenplay. But there’s still no real trouble, since typically no one is acting in security footage. The really difficult questions start when we consider films of plays. When my son acts in a school play, he is a model for whatever fictional character he’s playing. And yet my recording from the third row cannot be viewed as a screenplay even though it is recording someone being a model. I can slip into viewing him as model, but what I’m viewing is him-as-model within the play that is then filmed, not in the film itself. Given the facts (a) that there is an ontological difference between the film that can be viewed as screenplay and the film that cannot and (b) that the screenplay is, among films that may count as art, more common than those that cannot, it is incumbent on an ontology of film as an art form that it account for or at least help explain this difference. Until we have done so, we have not really distinguished film from theater. Incidentally, we also risk turning the in-theater handheld camera bootleg of a movie into its own work. Two one-time ambitions for the ontology of art are notably absent from these criteria—namely that a theory of this sort should provide a critical standard or teleology for film. This is not because these things are undesirable. If further utility is a reason to prefer a theory, then, surely, we should—ceteris paribus—choose the theory that provides benefits as useful as these. This result, however, is just too unlikely to demand of a theory. The diversity of styles and purposes of good films strongly suggests that there is more than one good-­ making quality of film, whatever their ontology. It seems only a vestigial Aristotelianism would have us expect an inference to the goodness of a thing directly from its kind or type.

Candidates With these criteria in hand, let’s evaluate some candidate ontologies of film. Some of these have been introduced for other purposes, though they’re all claims that either have been offered as ontologies of film or might reasonably be so offered. Realism Realism has a special place in the history of film and film theory. In its simplest form, it is the claim that what distinguishes film as an art form is its capacity to transmit, present or re-present reality as it is. André Bazin used this feature of film to distinguish it from other plastic arts.13 On his account, photography is the culmination of a centuries-long fascination with reproducing reality in the plastic arts. Film then expands photography’s capacity to capture reality by providing it with movement. Realism thus provided Bazin and his immediate 13  André Bazin, “The Ontology of the Photographic Image,” in What Is Cinema? Vol. 1, trans. Hugh Gray. (Berkeley: University of California Press, 2005).



followers with three key things: (a) an ontology of film, (b) a justification for the art status of film and (c) a critical standard for film. Regarding (b), the art status of film is achieved by establishing a place for photography and film at the end of a narrative history of art forms. For (c), Bazin thinks that because film is essentially photography, and realism is photography’s great defining achievement, films are at their best when they focus on this realist project to which they’re uniquely suited. Formalism, realism’s primary rival, is (allegedly anyway) more at home in painting and other traditional plastic arts. Our focus is, of course, on (a), especially given that the art status of film is not so controversial, and providing a critical standard is no requirement of an ontological theory. It is worth noting, though, that the photographic (and especially mechanical) nature of film has in other hands been used to cast doubt on its art status.14 Read strictly as an ontological theory, Bazin’s realism is somewhat unclear. First, he makes quite a bit of the “objective” nature of photography and film as opposed to other “subjective” plastic arts like painting and sculpture.15 While it’s true that we can, as Bazin argues, under normal circumstances infer the existence of x from a photograph of x and not a painting of x, this hardly makes photography and film objective, as though consciousness is not involved in choosing, curating and editing shots. In the kinds of photographs and films we’re interested in when we consider them as art forms, such conscious involvement is necessary. Second, he speaks about film as a reproduction of reality rather than—as might seem more natural—a recording of it. This appears to not be a metaphor. As he says at one point: “The reality that cinema reproduces at will and organizes is the same worldly reality of which we are a part, the sensible continuum out of which the celluloid makes a mold both special and temporal. I cannot repeat a single moment of my life, but cinema can repeat any one of these moments indefinitely before my eyes.”16 Bazin seems here and elsewhere to come awfully close to suggesting that there is no important difference between seeing in life and seeing on film, except that the latter has the advantage of potential repetition. But taken literally, this claim entails the absurdity that there is no important difference between the object and its image on film. It is perfectly reasonable that we might be watching Blazing Saddles on television and I say, “I see Cleavon Little there on the screen.” But if you were to then say, “where is Cleavon Little now?” I could not reasonably respond “I just told you, he’s right there on the screen.” But perhaps Bazin has in mind something more plausible, something akin to Kendall Walton’s “transparency thesis” regarding photography and film.17 On  Notably: Roger Scruton, “Photography and Representation,” in Philosophy of Film and Motion Pictures, ed. Noël Carroll and Jinhee Choi (Oxford: Blackwell, 2006). 15  Bazin, “The Ontology of the Photographic Image,” 13–14. 16  André Bazin, “Death Every Afternoon,” in Rites of Realism: Essays on Corporeal Cinema, ed. Ivone Margulies, trans. Mark A. Cohen (Durham: Duke University Press, 2003), 30. 17  Kendall Walton, “Transparent Pictures: On the Nature of Photographic Realism,” Critical Inquiry 11 (1994). 14



this version of realism, photography and film are perceptual aids to seeing what we wouldn’t otherwise be able to see, much like a telescope or a convex mirror in a parking lot. Instead of allowing us to see at great distances or around corners, film allows us to see what has occurred in the past. I’m thus no more committed to saying that Cleavon Little is right there on the screen than I am to pointing at the mirror when you ask where is the car that I see in it, or to the telescope when you ask where is the star that I see in it. In this respect, “transparency” may not be the best term. The first case of transparency we think of is the window. But when I remark on a tree I see through a window I point in the same direction when you ask, “where do you see this tree?” as I do when you ask, “where is this tree?” But when we’re in the garage and I remark on a car around the corner, I point toward the mirror when you ask, “where do you see this car?” and another direction entirely when you ask, “where is this car?” And I point to the telescope when you ask where I see a given star but not when you ask where that star is. It is true that we say things like “I see the star through the telescope” and “I see the tree through the window.” But we risk equivocation when we assume these sentences have exactly the same structure. In one sense, “seeing through” is synonymous with “seeing by means of” (as in “we see unicellular organisms only through microscopes”) and in another “seeing through” is synonymous with “seeing on the other side of without seeing around” (as in “A person can only see through so much water”). We “see through” some things in both senses, eyeglasses, for example. Film, however, is not one of these. This complaint may not amount to much more than terminological quibbling. What is important may instead just be that viewing a photograph or a film allows us to see real things the same way that telescopes and mirrors do, even if not quite transparently. But if it is not actual transparency, then the transparency theorist owes us an account of this particular kind of seeing reality. A number of detractors have objected that no adequately sufficient condition has been offered for this sort of transparency.18 As a first attempt, causal production seems like a decent candidate. The actual car around the corner, plus the mechanics of reflection causes me to see what I do in the mirror. The actual star, plus the mechanics of telescopy causes me to see what I do in the telescope. And Cleavon Little himself, plus the mechanics of cameras and projections, causes me to see what I do on television. But then again, an optical sensor and a description-generating program can mechanically cause a written description of an object, but we wouldn’t call that description “transparent” in any sense. So perhaps visual similarity should be added. But imagine a chess-­ obsessed society in which sensors on the boards of grandmasters transmit their moves to boards set up in living rooms across the country so that each move is 18  Gregory Currie, Image and Mind: Film, Philosophy and Cognitive Science (Cambridge: Cambridge University Press, 1995) especially 61–76; Berys Gaut, “Analytic Philosophy of Film: History, Issues, Prospects,” Analytic Philosophy 38 (1997) 147–8; Carroll, “Defining the Moving Image,” 121.



replicated by automatically moving pieces on boards everywhere. We’d all sit and watch the same games played out on our own boards, which look exactly like the one on which the grandmasters are playing. But those boards would not be transparent as a result. However, even if the transparency theorist has nothing to offer in the way of rebuttal or additional conditions, it is not clear that transparency has thus been refuted. She could, for instance, posit a virtuous circle between transparency and seeing through (in that first sense of “through”). Or she could insist that transparency is a simple, irreducible concept that is (a) given in our perceptual experience of photographs and live-action film and (b) evidenced by everyday language? Then again, transparency of any sort may not provide the best account of those experiences or our discussion of them, even among realist theories. And if not, then it is not clear that anything speaks in favor of transparency. Perhaps we say that we “see” Cleavon Little in Blazing Saddles but not in the most detailed verbal description of him because recognizing Little on film is so very much like recognizing him in real life. But recognizing his description is not at all like recognizing him. The latter first involves recognizing certain shapes or sounds that look and sound nothing like Little himself. Little in life (when he was alive) and Little on film trigger the same recognition capacities. If we can recognize Little in the former circumstance, we can do so in the latter, at least under normal circumstances. But this does not mean that Blazing Saddles presents Little to us, as transparency would have us believe. Perhaps, instead, it represents Little to us. And the specific way in which this pictorial representation via perceptual likeness happens via moving pictures provides film with its unique ontology. This is more or less Gregory Currie’s position, and is at present the most plausible version of realism available.19 But will any version of realism provide an adequate ontology for film? Probably not, if we take seriously the requirement that such a theory explains the difference between the film and the recorded play. If what we see is the object in front of the camera, either because we see it directly or because photography is transparent, then we see only Cleavon Little and never Sheriff Bart, since there is, in fact, no such sheriff. Similarly, if cinematic representation is only carried out through the recognition of perceptual likeness, it is not at all clear how we could distinguish Blazing Saddles, in which it is perfectly reasonable to see Sheriff Bart or Cleavon Little and some rehearsal footage of Blazing Saddles where it would only be appropriate to see Little. Bart and Little, after all, look an awful lot alike. Illusion If film is not essentially the presentation or representation of reality, maybe its nature is to be found rather in its unique capacity to present us with (or cause in us) the illusion of something real. There is something to be said for this view  Currie, Image and Mind, especially 79–112.




as well. We do praise films for “sucking us in” and fault them for ruining our suspension of disbelief. And yet films do not tend to produce genuine cognitive illusions, at least in adults. That is to say, we do not believe what we see in the film is really happening before us, that the perspective of the camera is our own. If we did, we’d have to ask how we fly over cities when we see establishing shots or jump back and forth in a room without continuous movement in multi-­ camera scenes. Nor do we think the content of the film is actually happening, or we would either be inclined to do something about it or have to ask ourselves troubling questions about our omniscient impotence.20 So films do not engender propositional illusions, or false beliefs that x. The aerial establishing shot does not, for instance, produce the illusion that we are flying. But it may still produce a different, non-propositional sort of illusion. The Icaruses among us may feel liberated, while the acrophobes feel some nauseated panic. These feelings are part of the illusion of flight without the illusion that we are flying. Similarly, “Dutch angle” shots may give us the sensation that something is not right with the world. We might call this sort of thing a “sensory” illusion, and it does seem perfectly natural to say that such shots provide the viewer with the sensation of flying or of unease. The problem is that while this sort of sensory illusion may be a capacity of film, it is hardly a necessary condition. The aerial shot may produce an illusory sensation of flying, but do scenes from the film adaptation of Neil Simon’s Lost In Yonkers similarly produce a sensation of being present for a conversation? Is there such a thing? Even if there is and it does, it seems highly implausible that the film does so in some way that is significantly different than a stage production of the play. A somewhat different sort of perceptual illusion (which may be thought of as a subset of sensory illusions) might be thought to be far more common, if not universal, in films.21 Specifically, we might wonder if the unique essence of film is that it produces the illusion of movement. Currie argues against this “weak illusionism,” insisting that the images on the screen really do move. Images, after all, are mind-and-perception-dependent entities. The effect may be pulled off by a quick succession of still pictures, but the image itself is nonetheless moving. Others—notably and I think most successfully Andrew Kania— have insisted that this movement is illusory after all.22 A film of Misty Copeland doing pirouettes appears to be an image of her moving, but some frame images include Copeland’s whole face, some just the right side of it, some just the left, and others none of it at all. How can a faceless image and a full-faced image be the same image? We say it is, so claim the illusionists, only because these differ The reader may notice a certain similarity or affinity here with suggestions in Colin Radford’s treatment of the paradox of fiction in “How Can We Be Moved by the Fate of Anna Karenina?” Proceedings of the Aristotelian Society, Supplementary Volumes 49 (1975) 71. 21  I’m relying on Currie here for the distinction between perceptual and cognitive filmic illusions. Currie, Image and Mind, especially 28–30. 22  Andrew Kania, “The Illusion of Realism in Film,” British Journal of Aesthetics 42 (2002). 20



ent images are united by the illusion of movement that their quick succession produces.23 There is no reason to posit continuity much less identity absent that illusion. I haven’t the slightest idea how to adjudicate between the illusionist and anti-illusionist on this point. I suspect, in fact, that they may be having a purely verbal dispute. The good news is that we don’t need to settle the issue here, as we have some other reasons to think even perceptual illusionism is inadequate as an ontology of film. For starters, plenty of other art forms are capable of producing the illusion of movement, and if we make our characterization of “art form” fine grained enough, there are some entire forms that do so. Ink-­ and-­paper optical illusions are a well-established sort of example, but there are others. Stage settings in plays, operas and ballets sometimes give the illusion of movement to stationary figures, and a note or melody being taken up by different parts of an orchestra at different times creates the illusion that the sound is traveling in space. The illusion of movement is not necessary for film either, as Danto’s example of the motionless film indicates. And finally, the illusion of movement, even if it is a genuine feature of film, does nothing to distinguish the screenplay from the recorded play. One produces the illusion of movement if and only if the other does as well. Imagination There are ways for us to entertain an idea, claim or attitude without being aware of or under the illusion of its truth, actuality or veridicality. One of those is to imagine it. Film may be a means by which we come to imagine the events depicted. This too has at least two versions. On the first, we imagine participating in the experience. Unless we imagine that we are invisible, inaudible and capable of taking on another’s perspective (which would be a lot), we imagine that we are right there in the depicted scene having the perspective of the camera. The same sorts of issues we identified with regard to cognitive illusions apply as well to participatory imaginations. We might be a little more lenient and forgiving, though, as we can imagine all sorts of things we can’t actually believe. We may be able to imagine ourselves flying during an establishing aerial shot, for instance, or having the capacity to move from one place to another without continuous movement during multi-camera scenes. But given a split-screen shot, are we to imagine that we’re in two places at once? Is this even possible? Also, what on earth are we imagining is happening to us when we hear a disembodied voice-over, especially when the speaker is a narrator and not a character in the film? Who is talking to us? When we hear the soundtrack, where is that music supposed to be coming from? Even if we could imagine these things without distracting questions attached, the participatory imagination theory has another problem in just the kinds of  Kania, “The Illusion of Realism in Film,” 253–4.




shots that might seem to give it credence. If films produce imaginations, then the point-of-view (POV) shot invites us to imagine having the perceptual experience of a character fictionally present to the events of the film.24 The POV shot provides a kind of first personal storytelling. But if this is the nature of film, then all shots do this. How, then, do we recognize the POV shot and distinguish it from others? We could avoid all of these issues by instead thinking that films provide for an impersonal imaginative experience. On this view, a film puts us in contact with the content of an imaginative experience without placing us in it. So, when I watch the scene in Blazing Saddles where Sheriff Bart meets the Cisco Kid, I imagine as actual the meeting I’m seeing. I’m aided in this imagination by Cleavon Little and Gene Wilder’s acting, the set, costumes and so on. But I don’t imagine being a third person inexplicably unacknowledged there in the jail cell. This is undoubtedly a more plausible account of our experience of films. Unfortunately, impersonal imagination won’t do much better as an ontology of film. Our first concern should be this: what exactly is the film object here? If it’s the imagination, we don’t experience the same films. This certainly does not conform to our everyday language and critical practices regarding films. If instead the film is an aid to the imagination, we have nothing to distinguish films from other narrative art forms such as plays or novels. Either way, we’re no closer to a unique ontology. And we’re certainly no closer to explaining either of Danto’s puzzles. The still photograph and the motionless film both engage the imagination in the same way (if either of them do). And whatever the difference between the screenplay and the recording of a play is, it can’t be in their capacity to aid imagination. The recording of an excellent play might spark an audience’s imagination much better than a poor screenplay. And an excellent screenplay might do better than a poor recording of a play or a recording of a poor play. Film as Dream The “dream” or “oneiric” theory of film may refer to a number of different theses. Weakly, some films are subject to the kind of Freudian psychoanalysis that is applicable to dreams, and some of the methods of such analyses are valuable mechanisms of film interpretation in general. Ambiguously, Hollywood was once known as a “dream factory.” And confusingly, Christian Metz and Alfred Guzzetti claim that, despite some important differences between watching a narrative film and dreaming: the filmic flux resembles the oneiric flux more than other productions of waking resemble it. It is received, as we have said, in a state of lessened wakefulness. Its signifier (images accompanied by sound and movement) inherently confers on it 24  Notice this presence is fictional even if the film is non-fictional. Real people don’t record their perceptual experiences.



a certain affinity with the dream, for it coincides directly with one of the major features of the oneiric signifier, “imaged” expression, the consideration of representability, to use Freud’s term.25

If dreams do involve the id’s expression via the image content of dreams, this does indeed indicate a certain isomorphism between dreaming and film. But this similarity is not enough to establish the stronger claim that film experience is of a type with dream experience. After all, there are stark differences between the representational natures of dreams and films. For instance, films represent in part via the kind of complex communicative implicatures that mark everyday utterances. Cinematic conventions are so commonly and successfully used to express meaning to an audience that confounding convention can be a successful tool for changing the meaning of an image. A filmmaker’s anticipation of her audience’s expectations regarding these conventions guide her filmmaking decisions and the audience’s expectations of that anticipation guide their interpretive judgments. Film meaning and interpretation is, therefore, cooperative. Even if it is reasonable to think that the id expresses the content of dreams in order to communicate something to the ego (and “communicate” may be a stretch or at best a metaphor here), the communication is hardly cooperative. To bolster the stronger claim, Metz and Guzzetti posit the “lessened wakefulness” of film experience, so that viewing a film takes us into something of a daydream state. By “daydream,” they must mean something more removed from waking experience than mere musing or imagining. But what then? Are there daydreams that genuinely get us closer to the dream state wherein, crucially, we take the image to be real? Maybe we do. I seem to have the experience of daydreaming so deeply that I have to consciously re-recognize the unreality of the daydream’s content. But that does not happen with films. In the darkest theater showing the most realistic film, I only “remind” myself of its unreality when I want to shake off its emotional effects, not to settle what is real and what isn’t. What we need to consider, then, is a strong dream theory that relies on the psychological condition of film engagement rather than any dubious claims about the degree of actual “sleep” involved in our reception of films. And luckily we have such theories. Colin McGinn, for instance, claims that “movies arouse in the viewer the same kinds of psychological mechanisms and processes that characterize the dreaming state.”26 Like the participatory imagination theory we just discussed, the dream theory has a problem with the kinds of cases that might at first glance look like good homes for it. Consider “The Test Dream,” from season five of The Sopranos. The episode’s title comes from an extended dream sequence in which the viewer is taken into one of Tony Soprano’s dreams. This sequence is 25  Christian Metz and Alfred Guzzetti, “The Fiction Film and Its Spectator: A Metapsychological Study” New Literary History 8 (1976): 90. 26  Colin McGinn, The Power of Movies (New York: Vintage, 2007), 102.



remarkably successful in capturing the phenomenal quality of dreams. Tony dreams a symbolically rich semi-linear narrative in which events occur uncaused, the predictability of time and space is fleeting and people from his life assume others’ roles in a way that suggests he is dreaming amalgamations of people. Tony himself “moves” through each inappropriately connected stage of his dream unaware of the abrasiveness of their juxtaposition and unaware that he is dreaming. The sequence is surreal without being disorienting, absurd without being unfamiliar. The aesthetic effect is in fact exceedingly familiar. What is displayed is very much the way we remember dreams. It is, in a word, dreamlike. How, then, does this show the dream theory false? If we have an instance of film that is recognizably dreamlike, doesn’t that at least provide some anecdotal evidence for it? In fact, it does just the opposite. What we should focus on here is not the fact that this sequence is dreamlike but rather that we are able to recognize that it is dreamlike. The sequence is recognizably dreamlike because it engages—through the use of both dramatic and cinematic techniques—the psychological operations characteristic of dreaming. It is remarkable only because not all—in fact very few—films do so. If all film worked on us this way as the strong dream theory imagines, it would be trivial to say that this one particular film is dreamlike. All film would be dreamlike. Compare the situation with film viewing and dreaming to two activities that more clearly involve the same psychological operations: lying and storytelling. These are not the same things and neither is one a version of the other (though they may both be versions of creative imagining). Yet it would clearly be trivial to say that a given instance of lying was storytelling-like. Of course, it was storytelling-like, just like all instances of lying. Lying and storytelling have just the kind of relation that the dream theorist imagines for film viewing and dreaming. But it is not trivial to say that this sequence from The Sopranos is dreamlike. So, it must be false that, in general, viewing films and dreaming involve the same psychological mechanisms and processes. Film as Language or Symbol System If one problem with the oneiric theory of film is its inability to adequately account for the communicative capacity of film, perhaps we should look directly to that capacity for the nature of film. Language, being our primary means of communicating, is a natural place to start. And indeed, so much has been said about the “language of film” that it has become an assumption in some quarters that there is such a thing. But even if there is a language of film, this does not entail that film is essentially a language. English is the language of Moby Dick, Old English the language of Beowulf and First-Order Logic the language of the proofs I should be grading right now. But no one would therefore say that Moby Dick, Beowulf or proofs on an exam are these languages. But if there is a language of film and film is not itself that language, what language is it? More to the point, it is not uncommon to read or hear that film is a specific and unique language, and that is the sort of theory in which we’re interested.



In order to evaluate this theory, we’ll need to first say what a language is. A language is first and foremost a symbol system, though not all symbol systems are languages. There is an unfortunate tendency to conflate the two concepts, but it is worth keeping them distinct. A symbol system may be any set of determinants for the range of “stands for,” “represents” or some other relevantly similar relation. The “check engine” light in my car is part of a symbol system but not a language. The “check engine” example may suggest that a language is a communicative symbol system, and what’s lacking in the car notification system is communicative intention and Grice’s “nonnatural meaning.”27 The light means that there’s a problem with the engine, but only “naturally,” as my hurting elbow means that there’s some problem in my arm. If this is right, then film may be a language after all, as we’ve discussed its communicative nature already. But languages are not only communicative and not all systems of communication are languages. Performative utterances are noncommunicative uses of language. And I may point to my smiling face to communicate my happiness to you.28 A still rough but better approximation: a particular language is a set of symbols and rules that (a) determine when sequences of those symbols are well formed and (b) relate those sequences to meanings. The two key features here are that a language must contain a syntax and semantics, and the latter must be determined by rules of the language rather than anything like natural fit or resemblance. And on this characterization, it is pretty clear that film is not a language.29 True, there may be cinematic conventions that rely upon communicative symbols. Montages mean that the ratio of screen time to fictional time decreases during them—a shot of a building followed directly by an interior shot means that the depicted room is in that building, and if the camera pans away slowly from two characters who might have sex, then they had sex. But these are remarkable cases. Most meanings in film are not delivered by semantic rules or conventions. We need not and should not appeal to any convention to explain why Sheriff Bart is a person and not a horse. If Mel Brooks wanted the Sheriff to be a horse, then he would have had a horse play the part and not Cleavon Little. Maybe we could imagine a convention that would override this natural assumption, one where horses are customarily used to play people. But the fact that a convention could counterfactually override more natural ­depictive assumptions does not mean that those assumptions are themselves merely conventional. Film may fail to be a language and still be a symbol system, even if the “represents” relation is in most cases dependent on a perceptual similarity between the representation and the represented. Whether or not this is a feature of  H.P. Grice, “Meaning” Philosophical Review 66 (1956).  Just to be clear: the smile itself may indicate that I’m happy. It’s the pointing that does the communicating. 29  Currie helpfully provides us with a fairly devastating argument against this sort of theory. Much of what I say in this section echoes that argument. Gregory Currie, “The Long Goodbye: The Imaginary Language of Film,” in in Philosophy of Film and Motion Pictures, ed. Noël Carroll and Jinhee Choi (Oxford: Blackwell, 2006). 27 28



many films will ultimately depend on your preferred theory of depiction. The good news (again) is that we don’t need to decide, since depiction is not necessary for film in the first place. Even if it is appropriate to say that the interview of Shelby Foote in Ken Burns’ Civil War is a depiction of Shelby Foote (a rather large “if”), still there is no depiction in purely abstract films. In its favor, a linguistic or semiotic ontology of film may be capable of explaining one of Danto’s two puzzles. It may just be that the screenplay is in the language of film (whatever that turns out to be) and the recorded play is not. This may be even if the recorded play looks in many ways like a screenplay, just as the sequence “Blund parg remple dwo draguile ringe” may look like a sentence but is not. Of course, a whole lot of work would have to be done to explain how this works in the case of film, and we’ve already considered some reasons to be pessimistic about success. The theory is entirely lost, though, in providing help on the other puzzle. If the unmoving film image is in a language, then the photograph is in the same language. Film as Thought There are two different diagnoses available for the shortcomings of the last few theories. First, we might think that their proponents err in seeking a psychological characterization of something that is not essentially psychological in nature. Alternatively, we could lay the fault in the theorists’ attempts to align film too closely with other, unrelated psychological faculties—an audience’s imaginative or illusory reception, on the one hand, or a filmmaker’s oneiric or linguistic capacities, on the other. If we want to avoid the latter mistake, perhaps what we need is a theory of film that assigns to it a relatively unique psychological function. We have the foundations for such a theory in Gilles Deleuze’s expansive work on cinema. Deleuze rejects purely psychoanalytic and linguistic theories of film, and yet film is for him still essentially bound to human psychology. As he says, “The great directors of the cinema may be compared, in our view, not merely with painters, architects and musicians, but also with thinkers. They think with movement-images and time-images instead of concepts.”30 Film on this view is a set of image-kinds (or, perhaps, kinds of image-making) distinguished by their role in a fairly sui generis psychological process. We cannot, therefore, characterize films and filmmaking using established categories; we can only offer analogies such as Deleuze’s suggestion that images are to filmmaking what concepts are to cognition. What is left for us to do in fleshing out a theory of this kind, then, is not to say what sort of thing the film image is, but rather to provide a taxonomy of film images. We see the beginnings of just such an attempt in Deleuze’s distinction between the movement-image and the time-image. The movement-image, 30  Gilles Deleuze, Cinema 1: The Movement-Image, Hugh Tomlinson and Barbara Habberjam, Trans. (Minneapolis: University of Minnesota Press, 2013) xiv.



born of a number of distinctly cinematographic techniques that Deleuze elides into “framing,” is fundamentally dynamic. It is neither reducible to series of still photographs nor is it a mere illusion produced by their rapid succession. It is characteristic of films that their images have no privileged moments. Deleuze then provides kinds of movement-image categorized by what they are images of—notably perceptions, affects and actions.31 Time-images, which we gain from the montage, are likewise more than a depiction or signifier of time. As the montage relates discontinuous movement-images, it provides—or just is—a distinct form of thought about (and in) time.32 For Deleuze, these image-kinds also have historical significance. The advent of the time-image marks the break between classical and modern film. And the latter is only possible as a postwar phenomenon, just like other forms of thought—notably but not exclusively philosophy. The taxonomy of images that Deleuze provides is both insightful and useful, and it’s possible that he is not espousing a film ontology in quite the same vein as the others we’ve considered here (his intentions are not always so clear). Even so, his work is illustrative of an approach to film ontology about which we ought to have at least two concerns. First, any account that treats film as a mode of thought is going to have to contend with the facts that thoughts typically belong to individuals and films are typically made by many people.33 Given a filmic image, we might wonder whose thought we’re seeing: the director’s, editor’s, cinematographer’s, camera operator’s or set designer’s? There are a couple of strategies we could employ to resolve this sort of issue, though neither are available to Deleuze. First, we might think of a film as presenting a collective thought in much the same way that a legislature can collectively enact a law or a committee can collectively issue a statement. But a necessary feature of these other cases is that the members of the bodies in question communicate and coordinate with one another to reach an agreement on the thoughts they express. This is only possible through the use of articulable concepts, and the thoughts they collectively express employ concepts as well. But the thoughts Deleuze ascribes to film are imagistic rather than conceptual, and it is not at all clear how a group of people could agree on images before they’re produced. We might avoid this problem by positing a kind of collective unconscious that manifests itself in the filmic image rather than in cognitive content. But whatever plausibility the phrase “collective unconscious” borrows from Jungian psychology (and that might not be very much) dissolves when we realize how comparatively local and mysterious this unconscious commonality would have to be. Instead of residing in all people because of a long-shared history, this collective unconscious would reside in a few people making a film together because of extraordinary coincidence.  Ibid., 64–5.  Deleuze, Cinema 2: The Time-Image, Hugh Tomlinson and Robert Goleta, Trans. (Minneapolis: University of Minnesota Press, 1989) 34–8. 33  Deleuze himself tends to focus just on directors. 31 32



Our second problem: while there is clearly something distinct about filmic images, they are not unique as a matter of thought. Setting aside for a moment the aforementioned problems that attend multiple authorship, imagine a film director and a stage director both in total control of their productions. They both must imagine and produce from imagination something for an audience to see. The resulting visual arrays are both bounded or framed, either by the scope of the lens or by the edges of the stage. Stage settings can move to alter perspectives in many of the ways that camera movement does in films. And finally, costume changes, actor switches and discontinuous settings and actions can produce on stage the same temporal imaging that we see in film montages. Moving Pictures, Moving Images Emphasis on movement is a recurring theme in a number of the theories we’ve already considered. For the representational realist like Currie, it identifies the film’s species among the genus of depiction. For the perceptual illusionist, it is the crucial illusion that film generates. And it forms the foundation for a full half of Deleuze’s taxonomy of filmic images. We’ll consider here Arthur Danto’s and Noel Carroll’s ontological theories, both of which put film’s movement front and center. Unsurprisingly, Danto has the advantage of at least attempting to provide in his ontology of film an account of the two puzzles we’ve been discussing. Indeed, his focus is on distinguishing film from photography, on the one hand, and from theater, on the other.34 The former is achieved, he thinks, by recognizing that films are essentially moving pictures. The movement of films is threefold: pictures move in the projector, images move on the screen and cameras move while they record. So, the unmoving-image film is distinct from the perceptually indiscernible photographic slide by virtue of the fact that we watch the one but (at most) see the other. That is to say, it is reasonable to wonder if the film image will move, but not the photograph image. And the unmoving nature of the film image is a stylistic choice of the filmmaker, while for the photographer it is just a function of the medium. As long as we’re heading down this rabbit hole, though, we might wonder about another sort of image projected next to the first two. This one consists of the photograph replicated and pasted over and over onto a 35-mm reel and then projected on the wall. So, we have: (a) a photograph projected onto a wall next to; (b) a perceptually indiscernible still film projected on the wall next to; and (c) a reel of film containing a series of reproductions of (a) projected on the wall.

 Danto, “Moving Pictures.”




Is (c) a film? On the one hand, there would appear to be some tension between counting Hoolboom’s projected leader as a film but not (c). On the other hand, there seems to be a relevant difference between (b), which was shot with a moving picture camera, and (c) which is the same image shot with a still photo camera. For instance, movement in the first camera would have produced a moving picture. Movement in the second camera would have produced a different picture. So (b) at least potentially involves all three kinds of filmic movement that Danto identifies—the camera, the picture and the image, while (c) contains only one. It is not clear if all three are necessary or if they’re individually sufficient for the requisite movement, and I fear we may be approaching the limits of intuition and good philosophical taste. Let’s turn then to the other explanandum, the distinction between the documentary recorded play and the screenplay. According to Danto, the difference is that a recorded play is about what is photographed, while the screenplay is not. The recording of a theater production of Romeo and Juliet is about the actors and what they’re doing on a stage. The screenplay Romeo and Juliet is also a film of actors, but it is about two doomed teenage lovers. Those lovers don’t appear in the recorded stage production at all, only people acting in those roles. As much credit as Danto deserves for taking this problem seriously, we should have at least two concerns about his strategy for solving it. First, I’m not sure the recorded play is about what is photographed either. It is not clear that the recording of the play is about anything. It seems more natural to say it is a recording of a play, not about the play. Second, some non-fiction screenplays are about what is photographed. There is an important distinction to be made between the “documentary” recording, as Danto uses the term in this context, and the “documentary” genre of non-fiction films. The genre-documentary is not a mere recording of some event. Though it is about real events, and, therefore, it is about what is photographed, it can still be a work whose artfulness depends on its moving images. That is to say, there are still two ways of viewing the genre-documentary film. We can view it as recording-documentary when we consider only the events captured on the film or we can view the same footage as a means to the filmmaker communicating something about what is being shown. To view the genre-documentary in this way is effectively to view it as screenplay rather than recording-documentary. And because this is an appropriate way of viewing the genre-documentary, some screenplays are about what they photograph. Carroll prefers “moving image” to “moving picture,” in part to capture the continuity between celluloid and digital films. That Carroll and Deleuze both emphasize “movement” and “image” should not tempt us into exaggerating the similarity of their approaches. Carroll’s “image” does not have the psychological content of Deleuze’s. And for Carroll, movement is an essential feature of film, while for Deleuze the “movement-image” is a historically contingent mode of cinematic thought. Terminology aside, Carroll’s approach is rather more in line with Danto’s. In fact, Danto’s emphasis on the possibility of



movement (at least on first viewings) forms a key part of Carroll’s own ontology of film. It is for Carroll one of five necessary conditions. In sum films (or moving images) are: 1. capable of movement—it is reasonable for first-time audiences to wonder if the image will move or not; 2. detached displays—audiences cannot orient themselves in or to the filmed spaces; 3. (as performances) tokens generated from tokens—a given film performance (i.e., a projection) is produced from a token (present on the reel/ hard drive) of a type (the film itself); 4. (as performances) not distinct artworks—the projection of a film is not subject to artistic evaluation and 5. two dimensional—the film’s display is two dimensional, even if perhaps its appearance isn’t.35 Each of these conditions serves a purpose and helps get us closer to our goal: (1) as we’ve said, distinguishes film from photography and (2) rules out filmic realism. There is an important difference between everyday perception of an object, even mechanically aided perception of objects and seeing them on film. When I see a car in the garage mirror or a star in the telescope, I can always say where I am in relation to what I see. This is very rarely the case when I see something on film. (3) and (4) distinguish film from theater. The performance of a play is a token of a type (i.e., a performance type that may be instantiated only once), and we do evaluate such performances. They have their own directing, acting, staging and so on. Film performances used to require some special skill when they were shown using complicated 35-mm projectors. But the projectionist had no artistic choices to make. And today it is mostly connecting an external hard drive to a projector and pushing a button. (5) ensures that the conditions don’t include things like moving sculptures. Let’s grant that these conditions work as Carroll intends. At best, we have necessary conditions that are not jointly sufficient. But, as Carroll claims, sufficiency might be fool’s errand. Even so, we want conditions that will distinguish film as an art form (as moving pictures, moving images, etc.) from as many of its nearest neighbors as possible. If it is worth distinguishing film from kinetic sculpture, then it’s worth distinguishing a screenplay from a recorded theater production. And Carroll’s conditions don’t do that. Let’s consider, then, a few conditions that might be added to his list that might get us a little closer still.

35  Carroll, “Defining the Moving Image,” especially 124–130. This terminology is characteristic of Carroll’s desire to move us away from theories of film that emphasize its medium.



Further Conditions? Intention might be a natural place to start. Adding a requirement that something is a film only if it is made to be more than a mere recording would emphasize the artifactual and communicative natures of film. More importantly, it would help fill the gap left by Carroll’s conditions mentioned above. The person who sets up a camera at the back of the theater to record a play likely does not intend to make a film so much as record an event. And the in-­ theater bootlegger does not mean to create a new film but to have a recording of the one on the screen. We do not want to rule out the possibility that someone could make a film by editing existing footage that was not shot with the requisite filmic intentions, and so we would want to be agnostic about whose intentions are determinative. But in so doing, we may be inviting other troubles. For instance, intention does not buy us sufficiency. We can try to make a film, successfully record video and audio and end up with a mere recording. There would be a kind of perversion in addition to dishonesty in the theater bootlegger claiming to have made his own work simply by virtue of recording someone else’s. A filmmaker’s intention does not seem to be necessary either. If I install security camera footage in a gallery as a kind of (albeit superficial) comment on the surveillance state, what I have is a work of art. And if the art form isn’t film, I don’t know what it is. But it would be wrong to say that there’s a filmmaker involved. What, then, about the intention of the people acting or otherwise performing? When I set up a camera to record a play, they are acting for the audience, which happens to include a camera. And when I make a film, they are (so to speak) performing for the camera. But again, performing for a camera can’t be necessary without ruling out abstract films, nature films and all the others that involve no conscious performance at all. Perhaps the above example suggests instead that we should borrow a bit from Danto’s and George Dickie’s theories of art status.36 We don’t want to conflate the ontological and art-status questions, but perhaps they’re not entirely separable. After all, we are looking for an account of the kinds of objects that are members of a particular art form. The larger problem is that institutional theories aren’t going to help us solve the central problem. Films can be art, but so too can be plays. I can display a film in a gallery, but so too could I display a play by showing a recording of it. Nor will the particular artworld context of a display fix the art form in question. Films are usually shown in theaters. But when the Metropolitan Opera broadcasts productions in theaters, they do not claim to be showing opera films but rather “transmitting” theatrical opera productions to theaters.

36  Arthur Danto, “The Artworld” Journal of Philosophy 61 (1964); George Dickie, The Art Circle: A Theory of Art (Evanston, IL: Chicago Spectrum Press, 1997).



So, if neither the intention of a would-be filmmaker or performer nor the context of display are going to work, it seems only natural to look to certain features of the object itself to determine the distinction between the film and the recorded theater production. Indeed, the recording of the play and the screenplay will likely look very different. The former, for instance, is less likely to hide stage settings, non-fictional audience members and so on. But it is not impossible for films to include such things. The end of Blazing Saddles might be an example, if not everyone was in on the joke as the fight scene stumbled out onto the studio lot. And a significant portion of the New York City population has accidently appeared in episodes of Law and Order. Also, I could add quite a bit of cinematography and editing to the recording of my son’s play without transforming it into a film in its own right. The Metropolitan Opera, in fact, has added more and more depth to the filming and editing of its recordings. But it is not clear that they at some point started making films. And if they did, we’d still be confronted with a kind of sorites problem: at what point did their recordings transform? What one little change caused the extraordinary change from mere recording to film?

Conclusions? This is where I think we’ve found ourselves: with a significant explanandum for an ontology of film and without an otherwise plausible ontology of film to adequately explain it. Those accounts that might help us distinguish the film from the mere recording are helplessly flawed in other ways. The difficulty of this problem is related, I suspect, to the attraction of such theories. Oneiric and imagination theories connect the experience of film to psychological processes that are themselves somewhat mysterious, and there is something mysterious about film. There is a difference between the film and the mere recording, evidenced by the appropriateness and inappropriateness of certain ways of seeing but not reducible to them. We should not, however, mistake something mysterious for something spooky or inexplicable. In fact, the ontology of film is in a terrific position, as better theories emerge along with more refined questions. We should feel bolstered by evidence of genuine progress and eager to see it continue.

Bibliography Badiou, Alain. 2017. Cinema. Trans. Susan Spitzer. Cambridge: Polity. Bazin, André. 2003. Death Every Afternoon. In Rites of Realism: Essays on Corporeal Cinema, ed. Ivone Margulies. Trans. Mark A.  Cohen, 27–31. Durham: Duke University Press. ———. 2005. The Ontology of the Photographic Image. In What Is Cinema? Vol. 1. Trans. Hugh Gray, 9–16. Berkeley: University of California Press. Carroll, Noël. 2006. Defining the Moving Image. In Philosophy of Film and Motion Pictures, ed. Noël Carroll and Jinhee Choi, 113–133. Oxford: Blackwell.



Currie, Gregory. 1995. Image and Mind: Film, Philosophy and Cognitive Science. Cambridge: Cambridge University Press. ———. 2006. The Long Goodbye: The Imaginary Language of Film. In Philosophy of Film and Motion Pictures, ed. Noël Carroll and Jinhee Choi, 91–99. Oxford: Blackwell. Danto, Arthur. 1964. The Artworld. Journal of Philosophy 61: 571–584. ———. 1974. The Transfiguration of the Commonplace. The Journal of Aesthetics and Art Criticism 33: 139–148. ———. 1979. Moving Pictures. Quarterly Review of Film Studies 4: 1–24. Davies, David. 2004. Art as Performance. Malden: Blackwell. Deleuze, Gilles. 1989. Cinema 2: The Time-Image. Trans. Hugh Tomlinson and Robert Goleta. Minneapolis: University of Minnesota Press. ———. 2013. Cinema 1: The Movement-Image. Trans. Hugh Tomlinson and Barbara Habberjam. Minneapolis: University of Minnesota Press. Dickie, George. 1997. The Art Circle: A Theory of Art. Evanston: Chicago Spectrum Press. Gaut, Berys. 1997. Analytic Philosophy of Film: History, Issues, Prospects. Analytic Philosophy 38: 145–156. Grice, H.P. 1956. Meaning. Philosophical Review 66: 377–388. Irvin, Sherri. 2008. The Ontological Diversity of Visual Artworks. In New Waves in Aesthetics, ed. Kathleen Stock and Katherine Thomson-Jones, 1–19. London: Macmillan. Kania, Andrew. 2002. The Illusion of Realism in Film. British Journal of Aesthetics 42: 243–258. Levinson, Jerrold. 1984. Hybrid Art Forms. The Journal of Aesthetic Education 18: 5–13. McGinn, Colin. 2007. The Power of Movies. New York: Vintage. Metz, Christian, and Alfred Guzzetti. 1976. The Fiction Film and Its Spectator: A Metapsychological Study. New Literary History 8: 75–105. Ponech, Trevor. 2006. The Substance of Cinema. The Journal of Aesthetics and Art Criticism 64: 187–198. Radford, Colin. 1975. How Can We Be Moved by the Fate of Anna Karenina? Proceedings of the Aristotelian Society, Supplementary Volumes 49: 67–80. Rancière, Jacques. 2001. Film Fables. Trans. Emiliano Battista. Oxford: Berg. Scruton, Roger. 2006. Photography and Representation. In Philosophy of Film and Motion Pictures, ed. Noël Carroll and Jinhee Choi, 19–34. Oxford: Blackwell. Walton, Kendall. 1994. Transparent Pictures: On the Nature of Photographic Realism. Critical Inquiry 11: 246–277.


Medium Specificity Noël Carroll

The notion of medium specificity has a long history in the philosophy of art but probably finds its most classic statement in Lessing’s treatise Laocoon. The doctrine of medium specificity, as traditionally construed, holds that art forms are individuated by their physical media which also provide said art forms with norms of excellence that, in turn, are determined by the possibilities and/or limitations of their material constitution. The idea of medium specificity has been especially influential in the history and philosophy of film. Many theorists of the silent film such as Béla Balázs (Balázs 2010; Carroll 2014), Rudolf Arnheim (Arnheim 1933, 1956; Carroll 1988), and Soviet montagists like V.I. Pudovkin (Pudovkin 1958) relied upon the doctrine of medium specificity to establish the artistic credentials of cinema. Subsequently, sound film theorists, such as André Bazin (Bazin 1967) and Siegfried Kracauer (Kracauer 1960) mobilized the notion of medium specificity in order to defend realism as the essence of cinema. Throughout film history, some critics have used the notion of the medium specificity, which they sometimes referred to as the cinematic, as a measure of filmic achievement. A film was good if it exploited the unique properties of the medium; a film was defective to the extent that it was uncinematic—that it failed to take advantage of the potentials of film as film. The concept of medium specificity was also employed in order to advance the case for developing academic film departments. It was argued that since film art was grounded in unique features of the medium, it required its own discipline of inquiry distinct from literary, theatrical, and art historical study as well as from the departments of history, psychology, and sociology. For various reasons, including the philosophical problems to be canvassed in the next section, medium specificity theses fell out of favor in the last quarter N. Carroll (*) The Graduate Center, City University of New York, New York, NY, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




of the twentieth century (see Carroll 1988, 1996a, b, 2003a, b, 2008). However, in the last decade or so, we have seen the return of modified versions of the claims of medium specificity in the work of Berys Gaut (Gaut 2010, 2016), Dominic Lopes (Lopes 2014), and Ted Nannicelli (Nannicelli 2017). These philosophers have expanded the conception of the medium to comprise not only the material characteristics of film but also the film practices. To this extent, they advocate what might be called a praxeological concept of the medium over a material one. Because medium specificity claims have been proposed in different forms at different stages, this chapter will be divided into two main parts, addressing what medium specificity was in the earlier stages of the philosophy of film and what theorists are claiming the thesis amounts to today. Because medium specificity thinking is so intimately connected to film evaluation, I will conclude by briefly sketching an alternative approach to motion picture evaluation, which I believe is superior to the invocation of medium specificity, but that can also do everything that theory in its various forms can legitimately claim to be able to do.

What Was Medium Specificity? As previously mentioned, the notion of the specificity of the medium of film arose early on in film history as a way in which to philosophically enfranchise the emerging technology of film as an art form in its own right (Carroll 1988). Early film theorists—such as Hugo Münsterberg (Münsterberg 1970; Carroll 1996a, b, c, d), Belász, Arnheim, and most of the Soviet montagists— faced two fundamental challenges. The first challenge involved showing that film could be an art in the face of the argument that, due to its photographic provenance, cinema was naught but the slavish, mechanical recording or reproduction of reality, thereby providing no opportunity for creativity, expression, and/or formal invention—that is, the very hallmarks of artistry. Film theorists, like the philosopher Hugo Münsterberg, countered these arguments by maintaining, for example, that cinema, through its physical structuring devices—like the close-up and parallel editing—was mind-like and, therefore, capable of expressive effects and imaginative manipulation. However, even if cinema could deliver art to its audiences, was it an art form in its own right or merely theater in a can—that is, a film can? In other words, was film only a recording device, relaying whatever transpired theatrically before the camera to audiences, or did it have an original artistic contribution of its own to make? Just as one might charge that a CD is not an artwork but simply a means for storing and exhibiting an otherwise autonomous musical artwork, so the suspicion arose that cinema might not be anything more than a mechanism for distribution theater to the masses. Medium specificity theories emerged and flourished in order to block the allegation that cinema was just theater in a can—that there was no autonomous art of cinema, but only filmed plays—recording of famous plays with famous



players. To suppose that there was an autonomous art of film separate and apart from the artistic plays they contained would be like mistaking the potato chip package for the potato chips. Moreover, it was supposed that if film was to be acknowledged as an art form in its own right, it should be in virtue of its capacity, based in its material nature, to do something other than what could be done by another art form, notably theater. Here, there was something like the underlying presupposition that the arts should be subject to the principle of the division of labor. Medium specificity theorists, like Arnheim, another proponent of the silent film (Arnheim 1956), argued that there were specific features of the medium which differentiated film from the other art forms, in particular, theater, AND that in virtue of these unique features, cinema possessed a range of effects that it could discharge better than any other art form. The relevant features of the medium here were thought of either in terms of the stuff that comprised the medium—as in paint is the stuff of painting so celluloid was the stuff of film— that is, the physical medium of film, or the medium was thought of as the instrumentalities used to manipulate the stuff—the camera or editing—the splicing together of pieces of film (Carroll 1996a, b, c, d). Thus, the medium in both the sense of a physical medium and an instrumental medium was material. That was why Arnheim called his theory “Materialgerechtigkeit”—or “doing right by the material” where this meant exploiting the material for the best effects for which it is suited. The pertinent effects here—such as the representations of fast-moving action over great distances—moreover, were not only putative effects that no other artistic medium could deliver as expeditiously, they were also the best effects available by means of cinema as well. Cinema was enfranchised as an autonomous art form via medium specificity thinking because there was something it could do that either other art forms could not do or that cinema could do much better. Thus, cinema had something to add to the system of the arts (a.k.a. the artistic division of labor) and, hence, deserved to be acknowledged as an autonomous art form.1 But also since these effects represented what cinema did best of all the things it did, these effects were ones that cinema artists qua cinema artists were advised to aspire to, since these effects were the very best cinematic effects at which one could aim. So, in addition to enfranchising cinema as an art form, the notion of medium specificity supplied filmmakers with guidance. Namely, aim at those effects as to what to make that in virtue of the material cinematic medium—whether in virtue of its distinctive capacities for representation or its distinctive limitations—cinema delivers whether uniquely or better than any other art medium, such as theater. Moreover, this approach also provides critics and viewers with a means for evaluating what filmmakers produce. Effects produced by cinema-specific 1  Although this strategy for defending the credentials of silent film was pursued by many of its defenders, one notable exception was the Soviet theorist, Sergei Eisenstein (see Carroll 2002).



means—effects often called cinematic—are regarded as positive achievements of film art, while those that do not recruit cinematic ends, perhaps by trafficking in theatrical effects, are to be regarded with disapproval qua cinematic art. On this view, the theatrical set design and acting of Lars von Trier’s Dogville comes off badly even though these choices were perfectly suited to realizing his purpose: to portray American society as governed by implacably oppressive roles for which theater functioned as a symbol. As V.F. Perkins and David Sorfa have observed in their talk in the Film-Philosophy Conference in Oxford in 2015, medium specificity criticism has traditionally imposed an agenda or set of standards dictating what artists should do. Moreover, these standards are general. They provide a way of evaluating any putative specimen of film art. For any work of film art, ask whether it realizes the best effects available in the material medium—that is, best relative to effects of other media, including close rivals like theater, and the best of all the distinctive effects of film. Cinematic excellence will occur, in other words, at the intersection of what cinema does better than the other arts and best of all the things it does as a function of its material medium. Films that aim lower—such as the ones that verge on theatrical effects—are uncinematic, that is, defective as film art.2 Again, it pays to emphasize here that this is a general view. Cinematic excellence is always a function of pursuing effects distinctive in the ways determined by the material means of cinema. This is a much stronger view than the view that sometimes a film may be excellent because the filmmaker has ingeniously exploited some feature of the medium to rewarding effect—as Orson Welles and his cinematographer Gregg Toland did in order to achieve depth of field in Citizen Kane. Nor is this the view that on certain occasions, a film may be defective precisely because the filmmaker has failed to use the medium in such a way as to achieve desired effects—as might occur, for example, if he/she opts for a slow film stock that is inappropriate for shooting at night. For, although the preceding two assessments make reference to the material medium, they are not examples of medium specificity thinking, properly so-­ called, as it has been handed down in the tradition of film theory because these appraisals are local, applying to the effectiveness or defectiveness of the use of the medium relative to the specific filmmakers’ aims. Indeed, the preceding example of using the wrong film stock for nighttime filming, though it refers to the medium, seems grounded in common sense rather than an application of the medium specificity theory or, for that matter, any theory. Furthermore, note that the preceding examples cannot be generalized to all motion pictures because not all motion pictures have the same aims or purposes.

2  Indeed, although medium specificity arguments traditionally promised to distinguish cinema from every other medium, typically, the contrast that was emphasized most was with theater, perhaps for the reason that it was the closest neighbor to film (and not to mention, its nearest competitor for audiences).



Motivating the notion of medium specificity is probably something like the commonplace, practical intuition that one should use the right tool for the job. All things being equal, you should have a church key to open a beer can, not a Phillips head screwdriver. Use a Phillips head screwdriver to turn nuts with an x-shaped notch on top. A Phillips head screwdriver can open a beer can, though somewhat inefficiently and inelegantly. But it is not the right—the most suitable—tool for the job. So, don’t use it for a task for which it is not intended. Similarly, battles and chariot races are more vividly portrayed on film than they are onstage. So, use film rather than theater to portray them. Theater, on the other hand, is better suited to deliver soliloquies. So, present Oedipus at Colonus live; don’t try to film it. Language in cinema should be laconic, in the service of action. Speech in theater is primary; the movement serves to accentuate it. So, give unto theater what is theater’s and to film what is film’s. Moreover, These Prescriptions Are General. Nevertheless, understanding traditional medium specificity theories to be general in scope, they are susceptible to a number of objections. Here are some of the most decisive. Some of the most challenging problems that the medium specificity theories face involve their approaches to critical appraisal. The first problem is that such theories do not seem to accord neatly with how we actually assess achievement with respect to the history of the motion picture. Consider the Oceana Roll sequence in Chaplin’s The Gold Rush. The Tramp, in order to entertain some lady friends, takes two dinner rolls and attaches them to two forks, proceeding then to execute an exquisite mime of a dance—the rolls serving as feet and the forks as legs. It is pantomime pure and simple. Arguably, Chaplin could have done it to equal effect on a stage. For all intents and purposes, it could have been a mere recording of a stage performance. No special features of the motion picture medium like editing or depth of field are in evidence. Thus, it should deserve a fairly low grade from the medium specificity proponent, if not being counted as outright defective—a failure of film as film. On the other hand, the Mack Sennett, Keystone Kop comedy, Lizzies of the Field, traverses an expansive space hyperactively with the eponymous automobiles buckling as they throttle forward at high speed. The film exploits the advantage of editing for representing this type of event, and it could never have been portrayed on stage with comparable effect. Thus, by the light of the medium specificity thesis, Lizzies of the Field would get a passing grade, if not a high one, as a work of film art. In any event, it would get a higher grade than “The Oceana Roll.” But how plausible is that? Given a Sophie’s choice between “The Oceana Roll” and the Keystone Cops’ comedy Lizzies of the Field, what film historians would plump for Lizzies of the Field? Barring allegiance to the notion of medium specificity, as traditionally advanced, would any critic rank Lizzies of the Field more highly as a work of motion picture comedy than Chaplin’s Oceana Roll? “The Oceana Roll” sequence is infinitely more imaginative, in its own right, than the admittedly serviceable slapstick violence of Lizzies of the Field, while,



at the same time, the Oceana Roll sequence exquisitely performs the narrative function of reinforcing the idea of the Tramp’s poetic sensibility and delicacy. What could the friend of medium specificity say on behalf of his higher ranking of Lizzies of the Field? That it was true to the medium—that it is pure film, unencumbered by theater? But what does that amount to? What virtue is there in pure cinema so-called? One problem with this proposal is it makes it sound as though being true to the medium or being pure is good for its own sake or intrinsically good. But that would appear to contradict or, at least, misconstrue the very idea of what it is to be a medium—that is, a means. A medium, in other words, is instrumentally valuable, if it is valuable at all. By fetishizing being true to the medium for its own sake, the advocate of medium specificity theorist appears to have lost sight of the very thing a medium is. Of course, the defender of medium specificity at this point may claim that his view is not that medium purity is good for its own sake, but that it has been proven empirically that impure cinema is defective, whereas pure cinema is always better, and, even, that all the very best cinema is pure. But that claim is hardly sustainable in the face of counter examples like the Oceana Roll. Moreover, there are many more examples than merely this one from Chaplin. For instance, myriad hack TV directors have produced work that is thought to conform more strictly to the alleged dictates of the medium than did Marcel Pagnol, but few, if any, have produced a masterpiece as moving as his film Césare. Furthermore, it should be obvious that things a medium does better than other media are not destined to be excellent. Computer-generated imaging (CGI), considered as a material medium, stages battles between superheroes and behemoths bent on conquering Earth better than other media, notably theater and maybe even photographically based cinema. But the effect is uniformly tiresome rather than excellent as proven by the last 20 minutes of most Marvel blockbusters when our heroes hurl large objects, like trucks, trains, and buildings at the monsters, and the monsters return gargantuan fire with monotonous regularity and ineffectuality. In response, medium specificity theorists sometimes claim that their theory has been empirically validated by the failure of the early silent film d’arts. But perhaps that failure could be ascribed to their lack of sound. Of course, Shakespeare merely mouthed but not spoken is boring. Similarly, medium specificity theorists may assert that their theory best explains the problem with early sound films; they were too theatrical. But perhaps the better explanation was that they were tedious theater. Fans of medium specificity will undoubtedly attempt to declare that my apparent counterexamples are without weight on the grounds that they are not truly excellent. But this attempt at damage control typically risks begging the question, since it usually rests upon presuming the very medium-specific standards it is supposed to be defending. This liability is in evidence in one of the most frequently recurring embarrassments of medium specificity theorizing—namely, that medium specificity theorists do not agree among themselves upon which features of the medium



are pertinent for their prognostications. Indeed, they notoriously line up in opposing camps such as realists versus montagists (e.g. sound film theorists like Bazin versus silent film theorists like Pudovkin). And when one side presents a counterexample to the other, the defending camp cries foul—for example, formalist charge petitio principii when realists, like Kracauer, deny that Hitchcock’s thrillers are exemplary cinema basically on the grounds that it offends the canons of realism that supposedly follow from the photographic nature of the medium. In defense of my own counterexamples to the medium specificity approach, I argue that from the perspective of critical evaluation, artistic excellence is of utmost concern. Moreover, artistic excellence need not be the result of a distinctive use of the medium in the sense that a proponent of medium specificity understands that concept. The “Oceana Roll,” for example, is artistically excellent without essentially deploying any specifically mediumistic resources unavailable to theater. My slogan might be summarized as “Excellence above purity.” But what of the underlying intuition that one should use the right tool to do the job—that you should use a hammer to drive nails, not a shoe, and you should use cinema to do what it is for—to stage battles, not monologues? In fact, the intuition is based on a questionable analogy. Film is not like a hammer. Considered as an art form, cinema does not have a single purpose or even a determinate range of purposes. We think we should choose the right too to do the job because in practical affairs, inefficiency should be avoided. But inefficiency is not an issue. If the result is worthwhile, we will suffer to put the question of efficiency to one side. No matter how many takes Chaplin forced his crew to endure, where the result is excellent, we are unlikely to complain. Yet what about the intuition that there should be a division of labor betwixt the arts? Again, it is inspired by a doubtful analogy. If two art forms, based in different media, can do the same thing, we do not demand that one cease and desist. Rather, we count our blessings. Many early comedy shticks originated on stage, often the vaudeville stage, only to be transferred essentially unaltered to film, but we do not begrudge such routines by Harry Landon, W.C. Fields, or the Marx Brothers for that reason. We are richer for having more excellent things. And that is more important to us than the purity of the media.

The Return to Medium Specificity Although the case against medium specificity may seem formidable, three leading philosophers of art—Berys Gaut, Dominic McIver Lopes, and Ted Nannicelli—have recently come to its defense. One strategy that these philosophers share is an expanded notion of what constitutes the nature of the medium under discussion. Specifically, they think of the medium as comprised not only of material elements—such as the stuff from which moving images are made and the instrumentalities through which that stuff is articulated—but also of practices. One might characterize their conception of the medium as



­ raxeological as distinct from the traditional view of the nature of the medium, p which we might call material. This raises the question, as we will see, of whether these contemporary philosophers are really defending anything worth really calling a medium specificity theory or whether they have changed the subject in a way that is ultimately uninteresting.3 Since Gaut’s position was developed earliest, I will deal with his version first. Gaut sets out to defend three theses which he maintains amount to allegiance to a medium specificity doctrine (Gaut 2010, pp. 286–287). They are: (i) Some correct artistic evaluations of artworks refer to distinctive properties of the medium in which these artworks occur. (ii) Correct explanations of some of the artistic properties of artworks refer to distinctive properties of the medium in which these artworks occur. (iii) For a medium to constitute an art form, it must instantiate artistic properties that are distinct from those that are instantiated by other media. The key modification in Gaut first two theses is their restriction of the scope of the claims of medium specificity. Only some evaluations and only some explanations are said to rely upon the distinctive properties of the medium. But this is far less ambitious than the preceding medium specificity theories which proposed systematic generalizations about the relation of, for example, deviations from medium specificity and artistic defectiveness. Arnheim’s argument against the sound film was a priori, whereas one supposes that Gaut thinks that those evaluations that refer to distinctive features of the medium will be discovered a posteriori. Likewise, Arnheim thought that any deviation from cinematic medium specificity was a defect, whereas Gaut only proposes that some are. Arnheim, in addition, contends that there will be a systematic relation between expressive deviations from the mechanical reproduction of perceptual appearances and artistic achievement, whereas Gaut does not say when reference to the medium will result in cinematic achievement or not. For Gaut, it seems to be a purely contingent matter. His version of medium specificity is far less powerful than Arnheim’s. Consequently, inasmuch as Arnheim is probably the most articulate proponent of medium specificity, I think we may fairly charge Gaut with changing the topic rather than offering a defensible version of something authentically considered a medium specificity thesis, especially if that view is a theory in the sense of plotting systematic regularities. Moreover, since Arnheim’s theory was a general one, I doubt that Arnheim would have denied that some negative evaluations of motion pictures might refer to distinctive features of the medium—like my previous example of some3  The traditional notion of medium specificity was impressively robust in its claims of generality. The qualifications of certain versions of the praxeological approach, notably Gaut’s, reduce the view to scarcely more than banal, common sense.



one trying to shoot a night scene with a film stock that is too slow. Rather, by invoking medium specificity, Arnheim wanted to claim much more, namely that any motion picture trespassing onto the domain of theater would be defective. Furthermore, anyone, including Arnheim, could ground the criticism of using slow film stock for night shooting without appealing to the doctrine of medium specificity. One could simply point out that it is absurd or self-­defeating to shoot at night with film stock that will not register anything perceptible, if you want viewers to see the action. That is, one can reject this particular use of the medium as defective on the common-sense grounds that it patently does not secure the filmmaker’s purpose. Thus, we can see the connection between some evaluations of cinema and features of the medium—the first premise of Gaut’s revisionist view—hardly supports the idea of medium specificity as it has evolved historically. Likewise, Gaut is probably right that some artistic properties of film are connected explanatorily to the medium, especially when it is expanded praxeologically. Disjunctive editing explains the surrealistic, artistic properties of the films by Buñuel and Dali. But Gaut’s qualification of the thesis is not a very daring claim when one considers that film also shares many artistic properties with theater and painting. Such a view is too underwhelming to claim lineage from theorists like Arnheim or Bazin. Gaut defends his claim about the relevance of reference to the medium using two arguments. The first notes that artistic achievement often involves overcoming difficulties posed by the medium. Hence, in order to commend an artist for overcoming the relevant difficulties, one will have to refer to the medium. This argument falls short of dealing with even all the cases that Gaut’s view hopes to lay claim to, let alone to all the pertinent cases for the obvious reason that, for example, not all of the cases of motion picture evaluations that refer to the medium will involve overcoming difficulties. Hitchcock’s mastery of montage is justly lauded, but there is no difficulty that his editing always overcomes. Murnau’s camera movement in Sunrise is eminently fluid, but there is no difficulty that it appears to be transcending. Gaut’s second contention in favor of his view of evaluation is that “There are terms of critical appraisal that refer to differential features of the medium in which it occurs.” But doesn’t this beg the question? Yes, there have been critics who accept versions of the notion of medium specificity—often in terms of the cinematic. So, in fact, they have referred to features of the medium in their critical appraisals. Does the fact that they talk this way always make their appraisals correct? They might be victims of a false theory. Perhaps Gaut is claiming that this is the way that critics should talk. But isn’t that what Gaut should be demonstrating? He shouldn’t merely presume it. Or perhaps he is merely repeating his first premise—namely, that it is sometimes apposite to include reference to the medium in evaluations. That is true, but it is question-



able whether that constitutes a commitment to medium specificity rather than being a matter of simply changing the subject. Gaut not only differs from traditional theorists, like Arnheim, in the scope of his claims. He also disagrees about the nature of what we call “a medium.” For Arnheim, the medium of film is material—a matter of film, cutting, and cameras. Again, that is why he called his view “Materialgerechtigkeit.” Arnheim’s view is also closer to the dictionary meaning of “medium” with reference to art forms. But Gaut, following Richard Wollheim, wants to include practices—uses of the technical resources of an art, of how they are manipulated—under the rubric of “the medium.” Ostensibly, Gaut, again, following Wollheim, motivates this in order to distinguish art forms from non-art activities that employ the same technical resources that related artworks do. That is, art photography and forensic photography may share the same technical resources—cameras and so on. So, what distinguishes them? Practices—the way they are used— Gaut proposes. Traditionally, medium specificity theorists had ways to tell, if only implicitly, art from non-art. For example, Arnheim and Balász held expression theories of art. Their task was, given that, to distinguish the unique art forms that met that task, by explaining how, for instance, film could be expressive. Thus, they did not need to find a way of distinguishing art from non-art within what they thought of as “a medium.” They had what Lopes calls a buck-stopping theory of art at their disposal (Lopes 2014). Consequently, the pressure Gaut feels at this juncture in his view is not something he shares with most of the others we call “medium specificity” advocates. That does not, of course, necessarily count against his theory. But I think the addition of practices to an account of what constitutes a medium makes it difficult to be a robust medium specificity proponent. The reason is because what we think of as discrete art forms, although they may deploy different materials, often share practices. Theater and cinema both share strategies for multiplanar composition in depth, both share method acting, and so on. Also, film, video, and broadcast TV share a vast number of practices, if only for the obvious reason that the later technical inventions imitated film in the process of enfranchising themselves as art. Furthermore, there is the process that I call “the interanimation of the arts”: this occurs when certain art movements that have their home in one traditionally individuated art form begin to be imitated across other art forms—as surrealist painters adapted the practices of surrealist poets and as minimalist filmmakers imitated the practices of minimalist painters and sculptors. Adding practices to what constitutes a medium not only diverges from previously known medium specificity approaches; it also makes establishing the uniqueness of individual media, as commonly conceived, more and more unlikely. Gaut’s way of attempting to individuate media is to regard each medium as a collection of media. That is, nested in each medium are an amalgam of various technical resources and practices. Some of the features in different media



such as multiplanar composition in depth appear at least in theater and film. Nevertheless, for Gaut, media can be still individuated in as much as the exact collection of technical resources and practices that make for one medium are not replicated exactly in any other medium, even though many of them are. This, however, raises a question about whether this is a genuine medium-­ specific approach to film since it will not necessarily forestall theatrical practices in insofar as both film and theater will have many of the same practices, as well as their effects nested in them. Indeed, cinema and theater may have so many practices in common that the only way to individuate them may be to go back to such old-fashioned medium-­specific distinctions as between recording versus live performance, indicating that in certain cases, Gaut’s expanded praxeological concept of the medium is not always relevant to individuating media. And what will Gaut say of the convergent practices of mainstream film, video, and television? Since for Gaut it seems the material medium is not supposed individuate art forms, will he be compelled to say, perhaps counterintuitively, that they are one and the same art form? Gaut’s various concrete attempts to individuate cinema seem to me to fail. At one point, he claims that the moving image is unique to cinema. But if “cinema” is being used here in its conventional sense, this sounds false, since shadow puppetry, video games, flip books, various optical toys like the zoetrope not to mention broadcast TV, and video all involve moving images. Moving imagery is not a sufficient condition for cinema. Nor does it seem that moving imagery is a medium in the normal sense, since moving images are realizable by so many different means of implementation. Is it a medium in Gaut’s sense, one composed of shadow puppetry and so on? That seems improbable since the moving image would not seem to be composed of a coherent set of practices. That, needless to say, does not show that it cannot be a medium in Gaut’s sense. But it does shift the burden of proof to Gaut. Let him tell us the unique package of technical resources and practices that make for the moving image medium as opposed to the more familiar idea that the moving image arts are composed of many discrete media. In a number of cases, Gaut’s suggestions about what differentiates cinema from theater are off the mark. Considerations of space do not allow a review of all of them. But here is a sample of unsuccessful contrasts. He cites editing as unique to cinema, but if the editing in question involves the juxtaposition of scenes and/or images, theater can do it by means of lighting. Gaut cites rack focus as another contrast, yet the effect can be secured via scrims and lighting. Pace Gaut, the appearance of camera movement can be aped and functionally approximated by a revolving stage as in Les Miz. Nor is the close-up effect impossible to approximate in theater. One way involves the use of spotlights and, as Susan Sontag pointed out, magnifying mirrors have been deployed to simulate close-ups in theater (Sontag 1969).



A final problem worth noting with Gaut’s proposal for individuating media may be that it will result in too many media. Avant-garde film and mainstream narrative cinema are generally thought to belong to the same medium. However, given the way in which Gaut counts practices as pertinent to media individuation, they should count as distinctive because of the radically different ways in which they manipulate their technical resources. Indeed, once practices are added into the mix, might not westerns, horror films, romantic comedies, musicals, and so on constitute different media. But surely that is to radically change what is meant by a medium.4 Of course, one may call for such revisionism on the basis of some promise of added explanatory power. But what the advantage would be in this case remains obscure? Recently, the philosopher of television Ted Nannicelli has explicitly endorsed the practice-oriented or praxeological conception of the medium advanced by Gaut especially.5 Since Nannicelli’s approach is perhaps the most developed application of this revisionist viewpoint so far, it is useful to examine it in some detail. Nannicelli asserts that “A medium is something like a cluster of relatively, stable coherent practices of making something in a particular vehicular medium [where a ‘vehicular medium’ is a set of particular materials – a.k.a. a traditional medium]” (Nannicelli 2017, p. 63). Given this conception of the relevant sense of the medium, Nannicelli proposes to individuate movies from television in virtue of their allegedly divergent practices of that which he calls temporality. What he has in mind here is that films putatively are organized to be stand-alone or one-off affairs intended to be watched in one sitting—narratives, in other words, with temporal closure meant to be taken-in in single encounter. Television, on Nannicelli’s account, in contrast to film, is partly characterized by “1) the understanding of ‘the work’ as being temporally subdivided in various ways: series, season, episode, format and so forth and thus potentially having different sets of temporal boundaries, some of which are malleable in a way that the temporal boundaries of films are not, and 2) a diverse and complex set of practices for making these temporally diverse and malleable kinds of works.” (Nannicelli 2017, p. 67). Apparently, Nannicelli wants to demonstrate that the praxeological conception of “the medium” is superior to the traditional material conception inasmuch as the former can differentiate film and television more effectively than the latter.

4  I am not an advocate of the traditional version of the idea of medium specificity, nor am I in favor of making a great deal of the notion of the medium for purposes of evaluation or even individuation. Nevertheless, I do think that if one wants to advance a defense of medium specificity, one needs to employ the notions of medium specificity and the medium as they are traditionally used in order to avoid changing the subject. 5  That is, although Nannicelli acknowledges Lopes’ influence, it is Gaut’s example that primarily guides his approach.



However, if Nannicelli’s alleged distinction is meant to support the praxeological approach as a replacement for the material conception of the medium, it is arguably a failure. Since the earliest days of film, it has featured continuous series (or serials), such as Fantomas, Vampires, The Perils of Pauline, Flash Gordon, Rocket Man, and so forth. Some of these series, like Hopalong Cassidy and Flash Gordon were among the most popular programs on early television, where they had been transposed from the silver screen. Since the self-same serials were staples of film and television, such seriality cannot individuate the two media. Comedy series, like sitcoms, are perennials on television, but they were likewise common in the cinema as evidenced by series like Our Gang Comedies, The Three Stooges, and many others which were also recycled as popular TV programs. Cartoon series like Popeye the Sailor Man and Betty Boop would also appear to fall into this category. Furthermore, the serial form has also been adopted as a feature-length film format. There are temporally continuous movie series such as The Hunger Games, The Maze, Divergent, Twilight, It, Justice League of America, and, of course, Star Wars, among others. And, there are many movie series with discontinuous episodes that are temporally comparable to TV series like Perry Mason with continuing protagonists like Bulldog Drummond in the 1920s; Tarzan, Charlie Chan, Mr. Moto, Nick, and Nora Charles in the 1930s; and James Bond and Indiana Jones in our own day. Moreover, television also employs stand-alone or one-off temporal forms. The most obvious case is the made-for-TV movie, such as Steven Spielberg’s Duel, to mention one of literally thousands of examples worldwide. Early television had a number of programs that specialized in anthology presentations—produced to advertise companies like Alcoa and General Electric—that featured stand-alone dramas like Marty and Requiem for a Heavyweight that were later adapted for the big screen. Less elevated were the weekly anthology series like The Twilight Zone, One Step Beyond, Lights Out, Alfred Hitchcock Presents, Outer Limits, and others which presented a series of discontinuous, stand-alone episodes with closure. Here, Nannicelli refers to the viewer’s experience as piecemeal and processional. But might that not also describe the experience of film series like the Sherlock Holmes series featuring Basil Rathbone and Nigel Bruce. In addition, there are anthology films with multiple stand-alone episodes such as O’Henry’s Full House, I Vinti, Seven Deadly Sins, Paris, J’Aime, and, of course, The Twilight Zone: The Movie.? What Nannicelli seems to have done in setting up his distinction between film and television has been to contrast certain typical kinds of contemporary movies with the kind of television series, like Better Call Saul, that exemplify the so-called present Golden Age of television. At best, this would be to mistake certain selected period-specific trends for an ontological distinction. However, even its historical accuracy is questionable, since the kinds of coun-



terexamples I have raised have occurred throughout the histories of film and television. Apart from these empirical oversights, Nannicelli’s approach leaves us with the possibility of too few media. Since his account of “the medium” seems to be intended to be exclusively praxeological, where all the practices between traditional, materially distinguished media are shared, we would appear to have to say that we have one medium. Arguably, experimental film and experimental video possess all the same practices. Similarly, mainstream commercial cinema and commercial narrative television may also share the same practices. Should we regard these, respectively, as cases of the same media? Surely, this is at least at odds with our current classificatory schemas. And what reasons do we have to abandon that classificatory schema? Lopes’ version of the medium specificity approach shares many of the same problems that beset Gaut’s and Nannicelli’s. Like them, he does not regard media as necessarily just material. They can be physical and/or symbolic, as in the case of literature, and they are also constituted in terms of practices. Moreover, again like Gaut, a medium is nested. Lopes calls the collection of media that make up a medium its media profile. But Lopes’ ultimate aim is not to defend just the medium specificity approach to cinema. Rather, he thinks that he is committed to medium specificity on the way to stalking bigger game—namely, that of addressing the enduring question of defining what it is to be an artwork by maintaining that something is an artwork just in case it is a member of an art kind. This, of course, raises the question of what it is to be an art kind. And this is where the commitment to medium specificity enters the picture for Lopes, since being an art kind requires having a distinctive medium profile which includes practices, notably appreciative practices, that is, practices that are appropriate to understanding work in the art kind. Here is a formal statement of Lopes’ view: X is a work of K [e.g. cinematic art], where K is an art [an art kind or art form] only if X is a work in a medium profile M, where M is an appreciative kind and X is a product of a M-centered appreciative practice P. (Lopes 2014)

Crucial for our purposes, here is that something is a work of a certain art kind—say cinematic art—only if it is the product of an M-centered practice— that is to say a medium-centered-appreciative practice—a practice that focuses appreciation upon the distinctive properties derived from manipulating the physical and/or symbolic resources of the medium. An example of what Lopes has in mind here, derived from his writing on photography, is the use of photography to bracket the object in front of the camera from its natural context and reframe it, in a way that enables us to see the object afresh (Lopes 2003). To appreciate a work of photographic art on Lopes’ view is, among other things, to attend to its unique medium profile, its capacity, for instance, to defamiliarize things by, so to speak, lifting them out of space and time for



decontextualized scrutiny. Thus, we are able to see the strenuous exertion that the runner expends as she strains toward the finish line in a photograph, revealing something to us that we would have missed in the haze of activity on film. Lopes offers the aforesaid formula as framework for research into the nature of individual art kinds. Since this framework has yet to be filled out in a single case, its viability remains to be seen. At best, it is a promissory note. Lopes cites my own theory of the moving image (Carroll 1996a, b, c, d) as the kind of thing he has in mind, although he does not commit himself to it. Nor should he, because, since it is not an account of moving image art, it provides no suggestion about what sort of things are to be appreciated qua moving images, and, if I am right, the moving image is not itself a medium but an effect multiply-­realizable by a range of different media, properly so-called. Furthermore, I do not believe that Lopes’ formula is successful on its own terms. It implies that every work of art is a member of at least one art kind. There are no, in other words, what Lopes calls free agents. Among other things, he says, there are no artworks without media. I think that this is false. For example, I do not think ready-mades can be said to have a distinctive media, since anything can be a ready-made. Also, what is the medium of conceptual artist Robert Barry’s piece “All the things I know but of which I am not at the moment thinking – 1:36 p.m. June 15, 1969.”? Surely the medium was not the card with this statement penciled on it. You might be tempted to say that the medium is thought, as Lopes is, but is thought an artistic medium and if it is, is it a distinctive one? Also, another candidate that would appear to lack a medium profile is performance art which would seem to be open to any sort of human activity including inactivity. Installation art poses a similar challenge in that anything can comprise it and it can serve as a vehicle for a wide range of discourses including but, not confined to, art theory, politics, economics, racism, and so forth. Lopes is aware of challenges like this. He does not reject these works as art. He proposes two strategies to deal with them. In some cases, like Duchamp’s Fountain, he suggests that they be regarded as the early progenitors of what will become an art kind, in this case, the ready-­ made. This strikes me as ill-advised, since it involves Fountain becoming an artwork retrospectively—that is, only after the art kind was consolidated successfully. Rather, I think that it is more plausible to think that Fountain, like all other artworks, was art at the moment of its birth. Another strategy that Lopes proposes in order to deal with the putative free agents I have enumerated is to claim that they are all examples of a specific art kind, namely Conceptual Art. The first thing that needs to be said is that Lopes’ category of Conceptual Art is much broader than usually believed and, frankly, rather ad hoc. It would regard various Fluxus events as conceptual art—such as Alison Knowles and her friends gathering at the Museum of Modern Art regularly to lunch on tuna fish sandwiches. However, I question whether that is correctly classified as Conceptual Art, at least as it is typically construed.



Lopes maintains that so-called Conceptual Art has a media profile including language, and ideas, mostly ideas about art. Of course, whether language and ideas are media is controversial and, in any event, Conceptual Art, as usually understood, is not restricted to trafficking only ideas about art. And it is not clear that every work of performance art must involve language or ideas about art or anything else. It may simply be predicated on provoking a reaction.6 Thus, it is not clear that Lopes’ framework for investigation of art kinds is as general as he maintains. Consequently, there is no reason to think that it fits cinema. In his work on developing this framework for exploring the individual arts, he has not said that much about evaluation. However, he does tend to be emphasizing that the value that individual arts, like photography and paintings, have as or qua photography and paintings. Does he think that this is the only or primary way that art kinds have value—in virtue of the kinds of appreciation that attend to the medium profiles of the art form? If so, that would appear to return us to the problem of fetishizing purity over excellence.

The Issue of Evaluation Although historically the notion of medium specificity played a central role in the artistic enfranchisement of film and video, today, if the idea has any function with respect to those media, it would appear to be primarily in terms of evaluation.7 As we have seen, for Gaut, the idea of medium specificity with respect to evaluation amounts to the claim that some cases of filmic evaluation advert to features of the cinematic medium. For example, Jean Renoir’s irregular panning in some of his films of the 1930s is said to contribute to the feeling of spontaneity in his images which, in turn, contribute to his overall project of realism. That is, Renoir’s camera follows the movement of the characters wherever they go rather than framing the characters in such a way that they must roam within a predetermined, proscenium-­like box. Instead, the camera must track their seemingly spontaneous movements as they plunge off screen, plotting a jagged floor plan, in contrast to the elegant, preplanned, geometric type of crane shot that Hitchcock would employ in order to reveal the key in Notorious. In contrast, Renoir’s shooting style leaves the impression that the characters are free to engage the space as the situation requires rather than being apparently blocked ahead of time in a pre-rehearsed environment, thereby via cam-

6  Moreover, if we accept Lopes’ revision of the notion of Conceptual Art, that will result in having too few media, in terms of how we currently individuate media, since ready-mades and performance art are not typically regarded as belonging to the same medium. Likewise, avant-garde video and avant-garde film will arguable be regarded as the same medium, although that is not how we currently categorize them. 7  One area in which medium specificity argumentation may still be marshaled for the purpose of artistic enfranchisement is in the defense of the artistic status of computer art (see Lopes 2009).



era movement imparting the feeling that the image is more authentic, that is, a more realistic portrayal of human action. But here, a crucial question arises: is it the irregular panning as such—arguably a film specific feature of the medium—that is commendable? Or, is it that this feature has been used in a way that serves Renoir’s purposes, namely, the production of a certain sense of realism? Surely, this use of the medium is commendable because it serves Renoir’s purposes so neatly. Indeed, it embodies or articulates his purposes. The lesson here is that when it comes to evaluating an artistic choice in film (as in the other arts), it is the question of whether or not the artist’s purpose is being served that is paramount, not whether the device is specific to the medium.8 To endorse the latter view is to revert to fetishizing the medium, as if it were valuable for its own sake. A similar point can be made with respect to negative evaluations. One feature of the medium, often cited as essential, is its capacity to record whatever stands before the camera, irrespective of the intentions of the film director or the camera operator. If a bird lands on the lover’s head as he pleads his case to his lady-fair, that untoward event is indelibly ingrained in the film footage. Obviously, this essential feature can give rise to straightforward artistic defects in certain films. For example, in various films of the distant past, including The Ten Commandments, Viking Queen, and Spartacus, characters have been reported to have been seen wearing wristwatches (Givens 1999). This is certainly an artistic defect, even if it is not perhaps one of the highest order. But why is it an artistic defect? Surely not because these films enlisted an essential feature of the medium simpliciter, albeit one supposes accidentally, but because engaging that specific feature of the medium in those particular contexts involved an effect that was at odds with the purposes of these films— which, broadly speaking, was to at least deliver a convincing representation of the past. It was not the use of the medium as such that was artistically dubious. It was the use of the medium relative to the constitutive purposes of the films at hand.9 Had these movies been time-travel films, there would have been no problem. So, once again, it is the purpose behind the use of the device that is relevant to evaluation. And this, of course, is the reason that choices of devices not putatively unique to the medium of film, such as theatrical devices, can also warrant positive evaluations when they serve the designs of the filmmakers. An avant-garde film of the 1970s might, for example, employ highly stylized, artificial acting in order to articulate the theme that the film was constructed without incurring aesthetic offense. 8  This is not to claim that this is the only concern that the critic needs to address when evaluating a movie. One may also need to ask questions about the value of the purpose, including the question of whether the purpose was worth the effort (see Carroll 2016). 9  By “constitutive purposes,” I mean the purposes that make the motion picture artwork, the artwork it is.



The film The Battle of Algiers has been commended again and again for its use of grainy film stock. Obviously, this is not because grainy film stock has something inherently good about it, but because it served the purposes of that particular film by imbuing it with a documentary look, thereby enhancing its realism. Though it was a brilliant choice for the purposes of The Battle of Algiers, it would have been an artistic defect in Pillow Talk. Indeed, it is exactly because of the predominance of purposes that Gaut must limit his claims for medium specificity to at best some instances. But even more significantly, the reasons that some cases of evaluation are connected to features of the medium have less to do with the value of the use of the medium as such, but to the purposes to which the medium is put. And the same points apply even when the medium is conceived praxeologically. As already noted more than once, the difference between Gaut’s revisionist conception of medium specificity and traditional views is that the traditional accounts provided general criteria of cinematic excellence (and defectiveness). This, all things being equal, is an enviable, theoretical virtue. I have been touting the line that purpose is more fundamental to motion picture evaluation than is medium specificity. But how does this approach compare with the admirable scope of the medium specificity approach? Clearly, my competing approach to evaluation is at least as general as the medium specificity view, if not even more encompassing. Why? Because every film has a purpose (or set of purposes, often interconnected), and, for that reason, the question of whether or not it has secured its purposes is at least a part of the process of evaluating it. One counterexample to this conjecture might be a film so confused in its intent that we might be tempted to say that it has none. Fair enough. So, let us amend the view to claim that every movie has an intended, constitutive purpose (or set of such constitutive purposes) and that assessing how the movie manages to discharge those purposes or, more likely, fails to do so is a major part of critical evaluation. Another worry might be films that supposedly eschew any preordained purposes—films whose (avant-garde) produces allege are open to viewers responding to them as they may. But the very ambition to afford such an open structure for viewers is unavoidably itself a purpose and, therefore, not a problem for the purpose-driven view. Among other things, a further advantage of my alternative approach to evaluation over the idea of medium specificity is that my view has the means to avoid the kind of untenable evaluations that entangled the medium specificity view when it was confronted by the comparison of the “Oceana Roll” sequence of The Gold Rush and Sennett’s Lizzies of the Field. For Chaplin’s invention is a more creative way of articulating or embodying his comic purposes than is Sennett’s—notwithstanding the fact that Chaplin’s is theatrical and Sennett’s is putatively cinematic.



Bibliography Arnheim, Rudolf. 1933. Film. Trans. L.M. Sievking and I.F.D. Morrow. London: Faber and Faber. ———. 1956. Film. Berkeley: University of California Press. Balázs, Béla. 2010. Early Film Theory. Trans. Rodney Livingstone. New  York: Berghahn Books. Bazin, André. 1967. What Is Cinema? Trans. Hugh Gray. Berkeley: University of California Press. Carroll, Noël. 1988. Philosophical Problems of Classical Film Theory. Princeton: Princeton University Press. ———. 1996a. Medium Specificity Arguments and the Self-Consciously Invented Arts: Film, Video and Photography. In Theorizing the Moving Image, 3–24. Cambridge: Cambridge University Press. ———. 1996b. The Specificity of Media in the Arts. In Theorizing the Moving Image, 25–36. Cambridge: Cambridge University Press. ———. 1996c. Defining the Moving Image. In Theorizing the Moving Image, 49–74. Cambridge: Cambridge University Press. ———. 1996d. Film/Mind Analogies: The Case of Hugo Münsterberg. In Theorizing the Moving Image, 293–304. Cambridge: Cambridge University Press. ———. 2002. Eisenstein’s Philosophy of Film. In Camera Lucida/Camera Obscura: Essays in Honor of Annette Michelson, ed. Richard Allen and Malcolm Turvey, 127–146. Amsterdam: Amsterdam University Press. ———. 2003a. Forget the Medium! In Engaging the Moving Image, 1–9. New Haven: Yale University Press. ———. 2003b. Kracauer’s Theory of Film. In Engaging the Moving Image, 181–202. New Haven: Yale University Press. ———. 2008. The Philosophy of Motion Pictures. Oxford: Blackwell Publishing. ———. 2014. Béla Balázs: The Face of Cinema. October 148 (May): 53–62. ———. 2016. Art Appreciation. Journal of Aesthetic Education 50 (4, Winter): 1–14. Gaut, Berys. 2010. A Philosophy of Cinematic Art. Cambridge: Cambridge University Press. ———. 2016. Cinematic Art and Technology. In Current Controversies in Philosophy of Film, ed. Katherine Thomson-Jones, 17–35. London: Routledge. Givens, Bill. 1999. Film Flubs: Memorable Movie Mistakes. New York: Citadel Press. Kracauer, Siegfried. 1960. Theory of Film: The Redemption of Physical Reality. Oxford: Oxford University Press. Lopes, Dominic. 2003. The Aesthetics of Photographic Transparency. Mind 112 (July): 1–16. ———. 2009. A Philosophy of Computer Art. London/New York: Routledge. ———. 2014. Beyond Art. Oxford: Oxford University Press. Münsterberg, Hugo. 1970. The Film: A Psychological Study. New York: Dover Press. Nannicelli, Ted. 2017. Appreciating the Art of Television: A Philosophical Perspective. London: Routledge. Pudovkin, V.I. 1958. Film Acting and Film Technique. London: Vision Press. Sontag, Susan. 1969. Film and Theater. In Styles of Radical Will, 99–122. New York: Farrar, Strauss, and Giroux.


The Moving Image Nick Wiltsher and Aaron Meskin

When you watch a film, you typically have an experience rather like that of seeing things move. The questions of how to characterize and theorize that typical experience preoccupy us in this chapter. In addressing this question, we take ourselves to be asking, more or less, what a moving image is, or what sense can be made of the phrase. Moving images occur in a motley of forms and formats: Hollywood movies, film art, CCTV footage, home videos, Skype calls, and so on. There is an interesting, narrower category of artifact in the vicinity, namely cinematic images. These are moving images apt for appreciation as cinematic—constitutive parts of movies, TV shows, and similar. What makes them so apt is what makes them a more specific category of moving image, and this aesthetic question is beyond our purview here.1 Our discussion is the broader category of the moving image, but the chief reason to be interested in moving images is because they are central to the experience of cinema. So we will often refer to film. However, not all images that move are moving images in the sense with which we’re concerned. Roughly, by moving images, we mean the sort of images that are seen when a film is projected at a frame rate that produces the 1  For an approach to cinematic images that we can’t hope to summarize here, see Gilles Deleuze’s heroic attempt to taxonomize the distinctively cinematic forms of movement in both space and time—that is, those forms of movement that cinema is uniquely or distinctively able to show: Gilles Deleuze, Cinema 1: The Movement-Image, trans. Hugh Tomlinson and Barbara Habberjam (London: Athlone Press, 1986), and Gilles Deleuze, Cinema 2: The Time-Image, trans. Hugh Tomlinson and Robert Galeta, (London: Athlone Press, 1989).

N. Wiltsher (*) Centre for Philosophical Psychology, University of Antwerp, Antwerp, Belgium A. Meskin Department of Philosophy, University of Georgia, Athens, GA, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




impression of motion (the standard rate for contemporary film is 24 frames per second) or digital equivalents of this. Whether other sorts of images that move should be assimilated to these moving images is a topic of some contention, as we’ll see. Given that the term “moving image” has two parts, you might expect that a satisfactory characterization of it would say something about images, and something about movement. Broadly, this is what we are going to do, but it is hard to separate the two entirely. Although, as we explain in Section 1, philosophers have suggested that the images involved in films have various distinctive features, it’s their movement that’s their most distinctive and most puzzling feature. So, even though we treat image-questions and movement-questions somewhat separately, it will be apparent that each impinges on the other; one might say we are treating two aspects of the same issue separately rather than discussing two issues pertaining to the same subject. In more detail, here is how things go. In Section 1, we explore preliminary issues concerning the identity of moving images, making clear why movement is thought to be a highly distinctive feature of images typically found in film, and why we are talking about images rather than pictures. In Sections 2 and 3, we address issues about images. These are essentially general issues concerning pictures and images which are specifically relevant to film, namely whether certain kinds of image are transparent, enabling a sort of indirect perception of their object (Section 2) and how to theorize depiction, the distinctive kind of representation of which images are capable (Section 3). The film-specific versions of both these questions are mostly concerned with movement. The big question about the movement of moving images is whether it is real movement, or apparent or illusory movement. We discuss this debate, and alternative accounts of the experience of image movement, in Section 4. We also address the issue of the ontology of moving images—the question of what sort of thing they are—which is, again, entangled with the question of movement. We conclude with some comments concerning the centrality of movement to the literature as it stands and possible ways in which progress might be made.

Why Images? Why Motion? We are concentrating on moving images because it is a commonplace to think that such things are necessary and important constituents of film. So philosophizing about film requires attention to them. They’re almost certainly not sufficient constituents; a movie doesn’t count as such solely in virtue of involving a moving image. But the thought that films involve moving images is intuitively compelling, and indeed embedded in the basic descriptive vocabulary we use regarding cinema. Equally, though, we talk naturally about moving pictures rather than images, and the former is arguably the more common term (it’s the motion picture industry—not the moving image industry). But “picture” and “image” are not synonymous, and so a choice needs to be made.



Noël Carroll prefers the term “images” to “pictures” in this context because “pictures imply recognizable representations, whereas…much of the art that concerns us has been and will be nonrepresentational and abstract”.2 Carroll is correct to say that there are many abstract and arguably nonrepresentational films that need to be considered in giving a general account of the moving image. Consider, for example, Hans Richter’s Rythmus 21 (1921), Stan Brakhage’s hand-painted films such as Water for Maya (2000), and Derek Jarman’s Blue (1993). However, his reasoning is unpersuasive. It is not at all clear that the term “picture” implies recognizable representations. After all, the public has no trouble making sense of the idea that highly abstracted and nonrepresentational paintings are pictures of some kind or other (even if they dispute their artistic status and value). Barnett Newman’s zip paintings and Jackson Pollock’s drip works are pictures in at least some ordinary sense. There are, however, other challenging cases that suggest “image” is the better term. Consider Stan Brakhage’s nonphotographic Mothlight (1962), which was made by inserting various natural objects between two strips of celluloid. There are certainly images on the screen. Are they pictures? Perhaps, but perhaps not. Or, consider films that only involve text, such as Paul Sharits’ Word Movie (Flux Film #29) (1966) or Michael Snow’s So Is This (1982). These films do not involve pictures of any kind. As Gregory Currie puts it, “the word ‘Vienna’, though I may recognize it as referring to Vienna, does not represent Vienna pictorially”.3 Nor should we be tempted by the thought that these movies depict words—such a view would tend to collapse the distinction between presenting and representing linguistic items. Yet it is reasonable to characterize these textual films as composed of images. (Such films might not involve movement. Snow’s film does not. But, as we shall see below, it still falls into the category of the moving image.) These examples suggest that we should talk about moving images rather than moving pictures. But perhaps even this is a mistake: perhaps films don’t involve images at all, let alone moving ones. Trevor Ponech argues that the essence of cinema is “stroboscopic visual display”: a field of projected points of light in rapid, flashing motion.4 The fact that the points of light are themselves in motion doesn’t entail that a viewer will have an impression of motion; a display in stroboscopic motion can show something static. An image may supervene on the display, but it doesn’t have to; the visual things appearing in the display might be “sketchy abstract shapes, an undifferentiated m ­ onochromatic field, words, num-

 Noël Carroll, Theorizing the Moving Image (Cambridge: Cambridge University Press, 1996), xiii.  Gregory Currie, Image and Mind: Film, Philosophy and Cognitive Science (Cambridge: Cambridge University Press, 1995), 7–8. 4  Trevor Ponech, “The Substance of Cinema,” Journal of Aesthetics and Art Criticism 64, no. 1 (2006): 187. 2




bers, or snowy static interference”.5 Ponech thus denies that images are any part of the essence of cinema. Jonathan Walley argues that Ponech mistakenly relies on the historical contingency that most films have been produced in such a way to make an essentialist argument, and that he misidentifies that historical contingency in a number of ways—that is, he gets the technological history and present of cinema wrong.6 One might also wonder whether at least some of the things that Ponech contrasts with images could in fact be subsumed under a more capacious conception of them. Interested readers might also consult Ponech’s reply, but we note the issue principally to set it aside.7 We take it that films do indeed involve images—if not necessarily, at least typically. We will, however, be more cautious about the claims that films have a special or essential relation to movement, and that the images they involve must move. This latter claim has customarily been finessed by careful theorists of film, who note that one may use a film projector to show images of static, unchanging objects. (Chris Marker’s La Jetée approximates this condition.) Yet surely this projection would count as a film, and therefore it can’t be that actual movement of an image is necessary for something to count as such. This point was first made by Arthur Danto, and has been reiterated by Carroll.8 Danto’s suggestion for a finesse is that in the case of still pictures such as slide projections, the lack of movement is “logically determined”, whereas in the case of moving pictures, stillness would have to be the product of “perverse artistic intention” (5). This, however, cannot be right, since the stillness of slide projections is surely not a matter of logic nor even obviously one of necessity; slide projectors are not necessarily static or stable. An alternative approach, which Danto also suggests, is that a properly informed viewer has a “legitimate expectation” that there will be movement in a film, even if it turns out that there is not.9 But it is illegitimate—in fact, according to Carroll, it is “conceptually absurd”—for someone to “expect to see movement in what she knows is a still picture”.10 So, according to Danto and to Carroll, it is a necessary condition of something being a film (and, more generally, an instance of the moving image) that it makes sense to expect it to move, at least on first viewing.  Ponech, “The Substance of Cinema”, 195.  Jonathan Walley, “On Ponech on the Essence of Cinema”, Journal of Aesthetics and Art Criticism 65, no. 4 (2007): 408–12. 7  Trevor Ponech, “Cinema Again: A Reply to Walley”, Journal of Aesthetics and Art Criticism 65, no. 4 (2007): 412–16. 8  Arthur C. Danto, “Moving Pictures”, Quarterly Review of Film Studies 4, no. 1 (1979): 1–21; Carroll, Theorizing the Moving Image, 63–6; Carroll, “Towards an Ontology of the Moving Image”, in Philosophy of Film, ed. Cynthia Freeland and Thomas Wartenberg (New York: Routledge, 1995), 66–85; Carroll, The Philosophy of Motion Pictures (Oxford: Blackwell, 2008), 58–9. 9  Danto, “Moving Pictures”, 5. 10  Carroll, “Towards an Ontology”, 73. 5 6



Again, it is hard to see what role conceptual possibility is playing here, and making films’ connection to movement relative to the reasonable expectations of the typical viewer will not satisfy those seeking a connection intrinsic to the medium.11 One might also wonder whether it’s really so absurd to expect to see movement in a still picture. As James Shelley argues, stills can show things moving; they can show things in motion.12 So it’s not unreasonable to go to the art gallery expecting to see (depictions of) things that are moving. It might be unreasonable to go there expecting to see things move, but a lot of still pictures allow us to see movement happening. From this and similar observations, Shelley draws the conclusion that films have no special or essential relation to the depiction of movement. Moving images, whatever they are, don’t have a capacity to show movement that sets them aside from still images; if we want to find the essence of cinema, we have to look elsewhere. One might argue that it is not the depiction of movement that marks the distinction between moving pictures and still pictures but, rather, the different ways they depict movement. So, for example, Robert Hopkins argues that “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”.13 This is too strong; there could be still (but changing) pictures in which development in color or shade properties determine development in color or shade of properties represented. Perhaps, then, the difference between still and moving pictures is that only in the latter does development in properties of the representation determine the representation of change of position. That is to say, in moving pictures, the ways in which parts of the picture change, or don’t, dictate the ways in which the things pictured move or do not move. However, even this cannot be quite right, since a picture in which mere changes in color determined the representation of change of position need not count as moving. (Imagine a stationary picture of a moving ball in which color properties were used to represent the ball’s position.) Perhaps the thing to say about this case is that this picture doesn’t represent the motion pictorially (although it represents other features pictorially), and that this needs to be added to the modified version of Hopkins’ account; that is, the focus must be on depictive aspects of pictures.14 On the other hand, it is not clear why we should require that moving pictures depict movement rather than depicting something else.

11  Such people might find more appealing Carroll’s idea that films are moving images because they “belong to the class of things for which creating the impression of movement is a technical possibility”. Carroll, the Philosophy of Motion Pictures, 59. For criticism, see Robert Yanal, “Defining the Moving Image: A Response to Noël Carroll”, Film Philosophy 12 (2008): 135–140. 12  James Shelley, “Motion Sickness” (unpublished manuscript, May 2017), Microsoft Word file. 13  Robert Hopkins, “Depiction”, in The Routledge Companion to Philosophy and Film, ed. Paisley Livingstone and Carl Plantinga (London: Routledge, 2009), 65. 14  Hopkins (personal correspondence) suggested this.



It is not clear whether a modified version of Hopkins’ account can be developed that will rebut Shelley’s heterodox conclusion. We shall put aside the issue. But note, again, that an account such as Hopkins’ sounds as if it will be silent regarding the movement of non-depictive images. Our task is to survey current debates about moving images, and those debates are conducted almost exclusively in just the terms whose relevance Shelley is denying: moving images are fundamental to cinema, and moving images bear some special relationship to movement that stills somehow don’t—a relationship bound up with a special capacity to show, exhibit, manifest, or represent movement. But the reader might well find, by the end of our survey, that they feel that much of the debate needs rethinking at the sort of fundamental level that Shelley suggests.

Transparency We began this chapter by saying that films give us experiences rather like those of watching things move. The simplest explanation is that we are, in fact, watching things move. This is to claim that films are transparent. Of course, the claim on its own does not underwrite the natural idea that, in some way or other, the images involved in films themselves are moving, but one could consistently conjoin the idea that films transparently show objects in motion with an account of the motion of the images depicting those objects. The transparency thesis has primarily been explored with regard to photographs, and the debate is primarily conducted in the terms laid down by Kendall Walton.15 His claim is that photographs allow us to literally see their objects; we see through the medium to the thing. This is not to say that photographs aren’t pictures, nor that we do not see the photographs themselves (they are transparent, not invisible), but to claim that they are pictures of a special sort: ones that allow us to indirectly see their objects. (Walton also holds that such pictures prompt us to imagine directly seeing their objects, but this is an implication of his general theory of depiction and is not directly implied by the transparency thesis.) In brief, Walton argues that photographs are transparent because they involve belief-independent, counterfactual dependence on their depicted scenes which preserves real similarity relations. As Diarmuid Costello and Dawn Phillips put it, this thesis has proved “remarkably resilient”, despite the fact that numerous philosophers have found it counterintuitive.16 The somewhat inconclusive debate that has ensued consists mainly in people trying, with greater or lesser degrees of success, to identify relevant discontinuities between the experiences of seeing an object represented in a photograph and seeing that same object. So, for example, Berys Gaut has argued that seeing involves “being in unmediated or direct 15  Kendall L. Walton, “Transparent Pictures: On the Nature of Photographic Realism”, Critical Inquiry 18, no. 1 (1984): 67–72. 16  Diarmuid Costello and Dawn M. Phillips, “Automatism, Causality and Realism: Foundational Problems in the Philosophy of Photography”, Philosophy Compass 4, no. 1 (2009): 6.



contact with an object” and that this requires that rays of light “pass uninterruptedly from it to our eyes”.17 Several other critics have argued that seeing necessarily involves the provision of certain sorts of spatial information concerning the location of objects relative to the viewer, and that photographs do not provide this egocentric information.18 One possible reason why this debate is inconclusive is that Walton’s original claim incorporates, or relies upon, what he later admits is a conceptually revisionary account of ordinary seeing, according to which (for example) one counts as seeing a foot by virtue of seeing a footprint.19 So attempts to grapple with the transparency thesis have to be oriented either to Walton’s revisionary account or to some more ordinary understanding of sight; this raises the danger that the parties to the debate will end up talking past each other. In any case, the transparency thesis is typically held to apply to film as well as to still photographs, both being largely photographic media. But a number of considerations concerning transparency are more germane to films than they are to photographs; here are four. First, it seems that the transparency thesis, even if correct, cannot apply to all films, at least not without some nuance. Some films are animations; some animations not produced by photographic means; so, the transparency thesis cannot apply straightforwardly to them.20 Second, and somewhat related, one can wonder what objects one sees through a putatively transparent film. Most photographs are documentary: they aspire or purport to show things just as they were. Some films are also documentary, in this sense, but a good number are fictions: they tell a story, and they show how some things are in the story by showing us how some things really are (how we get from the latter to the former is not our concern here). The transparency claim is most usually taken as the claim that we really see actors, props, and sets; if, in any way, we “see” characters, items, or venues, it is a special sort of “imagined seeing”.21 However, Hopkins has argued that there is a respectable sense in which one does really see fictional things in fictional films.22 So there is room for debate concerning the objects of transparent seeing through films. Third, it might seem that certain forms of broadcast video introduce factors which make the transparency claim more plausible for moving images than it is for photography. For example, the viewer of the feed of a fixed security camera will know the location of the depicted objects. However, commentators have  Berys Gaut, A Philosophy of Cinematic Art (Cambridge: Cambridge University Press, 2010), 91.  Carroll, Theorizing the Moving Image, 62–3; Carroll, The Philosophy of Motion Pictures, 56–8; Jonathan Cohen and Aaron Meskin, “On the Epistemic Value of Photographs”, Journal of Aesthetics and Art Criticism 62, no. 2 (2004): 197–210; Currie, Image and Mind, 65–7. 19  Kendall L.  Walton, “Looking Again through Photographs: A Response to Edwin Martin,” Critical Inquiry 12, no. 4 (1986): 805. 20  Carroll, The Philosophy of Motion Pictures, 97. 21  George Wilson, Seeing Fictions in Films (Oxford: Oxford University Press, 2011). 22  Robert Hopkins, “What Do We See In Film?”, Journal of Aesthetics and Art Criticism 66, no. 2 (2008): 149–59. 17




generally held that such cases are ultimately not different from the photographic and cinematographic cases because the image itself does not carry the requisite egocentric information.23 But, fourth, one could argue that, even if still photographs are not transparent, and even if moving images are not wholly so, the latter might be transparent with respect to motion. A lot of the arguments against the transparency thesis in the literature rely on finding differences between the ways in which actual seeing reliably informs us about the spatial location of objects relative to us and the ways in which seeing photographs and films fails to do so. As Jonathan Cohen and Aaron Meskin put it: “photography is not transparent, insofar as the visual process of looking at photographs fails to carry egocentric spatial information about their depicta”.24 Now, to see motion, one needs to see an object change its position. This change needs to be relative to a perspective. Cohen and Meskin suggest that properly seeing an object requires reliably carrying information about the spatial relations between the object and the viewer; the question is whether properly seeing movement requires this. Perhaps it is enough for motion-seeing that information is reliably carried to the effect that the object is changing position, whether or not information about its various positions relative to you is also reliably conveyed. If that is so, moving images might transparently show motion—they might allow you to see an object’s motion—even if they don’t transparently show the actual object. Moving images might be transparent in many, some, or no respects. But even those who accept the thesis often say that transparent images can also represent objects in a more mediated manner. So, it’s worth exploring the distinctive sort of representation of which images are capable, particularly their putative capacity to represent movement.

Depiction If films don’t show motion in some literal, transparent sense, perhaps, instead, they represent motion. Films doubtless represent in multiple ways. They often contain linguistic representations of one sort or another, they may well involve sonic representations, and so on. But we’re concerned with the moving image, and so the form of representation we’ll consider is depiction. Depiction is a form of representation by which pictures or images represent their objects. While it’s a type of representation unique to images, it’s not the only way in which images can represent; not everything an image shows or represents is depicted. Comics represent various things without depicting them: for ­example, they often represent speech without depicting it. Allegorical paintings can represent an idea without depicting it; they depict objects that ­somehow represent the idea. So it is an open question whether motion is 23  Carroll, Theorizing The Moving Image, 63; Cohen and Meskin, “On the Epistemic Value of Photographs”, 203. 24  Cohen and Meskin, “On the Epistemic Value of Photographs”, 201.



among the things that films generally depict or among the things they usually represent in some other way. Theorizing about depiction usually starts from static images, typically paintings. Theories of depiction can be grouped into four kinds. According to the first, depiction depends on resemblance or intended resemblance between aspects of a picture and aspects of the depicted object.25 The second kind of theory holds that pictures symbolize what they depict, much as words and sentences symbolize their objects.26 According to this view, depiction has a syntactic structure and semantic scheme that we need to learn in order to grasp the representation. The third kind of theory says that depiction relies on recognitional capacities.27 According to this view, pictures exploit our ability to recognize things by triggering just the same perceptual or neuropsychological capacities; a picture of a toucan depicts a toucan by triggering whatever it is in the eyes, brain, and mind we use to recognize real toucans. The fourth kind of theory says that depiction involves a special sort of experience. This experience is often called seeing-in. The idea is that experience of an image involves seeing, in the marks on a surface, that which the image depicts; the experience has two aspects or folds, one of the surface, the other of the represented object.28 Some have said more about this special experience: for example, that it’s an experience of resemblance, or that it’s an experience with both imaginative and perceptual aspects.29 Without settling the question of which of these theories is right, we can ask whether any of them make it seem plausible that motion is something that can be depicted. Paintings and photographs can certainly show movement, and perhaps, indeed, they depict it. Quite plausibly, different theories will say different things about how, or under what circumstances, images depict motion. A symbolic theorist will be attracted, perhaps, to the example of a cartoon image of a sprinter, with waving lines flowing behind them; these seem to symbolize motion. Partisans of the other approaches may be less likely to say that such an image depicts motion, but ready to say that a picture of, say, a ship under full sail does do so, by virtue of triggering recognitional capacities, acting as an imaginative prop, or through resemblance. 25  For example, Catharine Abell, “Canny Resemblance”, Philosophical Review 118, no. 2 (2009): 183–223; John Hyman, “Depiction”, Royal Institute of Philosophy Supplement 71 (2012): 129–50. 26  For example, Nelson Goodman, Languages of Art: An Approach to a Theory of Symbols (Indianapolis: Bobbs-Merrill, 1968); John Kulvicki, On Images: Their Structure and Content (Oxford: Oxford University Press, 2006). 27  Dominic Lopes, Understanding Pictures (Oxford: Oxford University Press, 1996); Flint Schier, Deeper Into Pictures (Cambridge: Cambridge University Press, 1986). 28  Richard Wollheim, “On Pictorial Representation”, Journal of Aesthetics and Art Criticism 56, no. 3 (1998): 217–26. 29   On experienced resemblance, see Robert Hopkins, Picture, Image and Experience: A Philosophical Inquiry (Cambridge: Cambridge University Press, 1998); On imagination and perception admixed, see Kendall L.  Walton, Mimesis as Make-Believe: On the Foundations of the Representational Arts (Cambridge, MA: Harvard University Press, 1990), 293–352.



But none of these thoughts seem germane to the question of how motion is shown in moving images. The trouble is that there is something peculiarly direct and literal about the relation of moving images to motion. The images don’t represent motion—they display motion, or they’re in motion, in some sense. Just think how unintuitive it is to say that motion is symbolized in an ordinary film; one wants to say that, no, there just is motion there, not a symbol of it. Resemblance is likewise intuitively unsuitable. Similarly, it seems strange to say that one recognizes motion in film, perhaps because motion is not something one generally recognizes in actuality. Recognition is to do with successfully reidentifying objects by their appearance, but motion does not have an appearance in this sense. One does not reidentify motion in different cases by noticing a certain way it looks. Plausibly, one detects motion using some more basic capacity to understand spatial relations.30 The fourth view is less clearly implausible, insofar as it seems that one does experience motion when watching a film. But the question here is specifically whether films represent motion via a special experience, and once more, it seems that, for example, the idea that motion is represented via a special sort of imaginative experience makes a complex mountain out of a simple molehill; the experience is, really, very much like that of seeing motion. However, the idea of seeing-in on which the experiential view is founded does at least allow us to distinguish clearly two aspects of film experience that might be confused or conflated. Suppose that depiction indeed involves an experience with two aspects: one pertinent to that which is depicted and one pertinent to the medium of depiction. It’s common, regarding the latter, to talk of the variegated ways in which surfaces might be marked: they might have colors, textures, and so forth. But the colors that a painting depicts (if it does so) need not be among the colors that the surface actually is.31 Similarly, a painting may depict a shape without any of the marks on the surface actually instantiating that shape. With this in mind, we can distinguish two aspects of films’ movement: the motion that’s depicted and the motion that is (perhaps) a property of the medium in (partial) virtue of which it depicts. One might think that movement is fundamental to the moving image: a property by which it depicts movement. But the movement depicted need not be the movement of the images. The experience of movement in seeing film is intimately connected with the construction of the space of the film by camera angles, tracking, and suchlike. This is also how a spatial perspective is constructed. Constructed spatial perspectives in paintings are plausibly construed in terms of seeing-in: we are aware both of our actual perspective on the painting and the perspective shown on the objects in the painting. And so, similarly, we might think that the motion of moving images involves seeing-in: we are in some sense aware of both the motion of  The term “motion recognition” is typically applied to machine detection of motion.  Dominic McIver Lopes, “Pictorial Color: Aesthetics and Cognitive Science”, Philosophical Psychology 12, no. 4 (1999), 419–423. 30 31



images and the motion of the objects that they show. This, again, is an issue meriting consideration that we can’t give it here.32 We turn now from the image aspect to the movement aspect of moving images.

Movement and Moving Images Suppose moving images don’t transparently, really, show us objects in motion. And suppose that motion is not a depicted property of the objects in the image. Perhaps, instead, we see motion in films because we see images apparently move. Something like this idea is more or less orthodoxy; moving images are images that move, which is why they can show movement. What’s far from settled is how exactly to understand the claim that images move: do they really move or do they just seem to move? Those who think they really move are generally called realists; those who think they just seem to move generally claim that the movement is an illusion, and thus may as well be called illusionists. The debate between illusionists and realists is one of those curious arguments where both sides often claim to be advancing, or vindicating, the common-­sense view of the matter, and so tend to claim that the burden of proof rests with their opponents. For this reason, positive arguments for either side are somewhat thin on the ground, while reasons to doubt the other side’s position proliferate. The illusionist’s version of common sense runs something like this: we all know that films are made up of a series of static images, which seem to move when shown consecutively at a certain speed. So at no point do we really see a moving image; that is impossible, since there is no actual, single, moving thing to be seen. Rather, we experience an illusion of movement. Because illusionism does have a certain intuitive plausibility, one might think that not much more is needed than this.33 Andrew Kania, however, offers something more substantial.34 He argues that a central feature of illusions is that we can be disabused of them, shown that reality is not how our experience suggests it to be. And we can be disabused of the illusion that the cinematic image really moves. If we slow down the film, the movement disappears; we start to see images discontinuously.35 The point is not that the motion disappears, and so must have been an illusion; the point, rather, is that once you understand what’s going on at the lower level, you’ll change your mind about what to say about your experience at the higher level. You will now say something like: I was not really seeing a moving image, just as you will want to say that you were not really seeing a bent stick when you realize that it was a straight stick standing in water. 32  For more on the issue, see Currie’s discussion of whether films represent time and motion by way of presentation of time and motion, or in some less direct manner, in Image and Mind, 96–102. 33  Currie cites several culprits who insouciantly assume the truth of illusionism (Image and Mind, 34). 34  Andrew Kania, “The Illusion of Realism in Film”, British Journal of Aesthetics 42, no. 3 (2002): 243–58. 35  Kania, “The Illusion of Realism”, 248–9.



In defense of realism, Carroll argues that the motion of images does not count as a perceptual illusion.36 Most things are not, at some fundamental level, as they appear to be. For instance, objects that are apparently solid are, at a microphysical level, full of gaps, holes, and air. Nonetheless, it is an objective fact that ravens and writing desks are solid objects; we would be badly mistaken to say that their solidity is illusory. But this means that we have to accept that appearance properties such as solidity are, in some sense, objective response-­ dependent properties. An appearance property counts as real, or objective, if creatures like us, with perceptual capacities like ours, tend to reach intersubjective agreement on the question of whether objects do or don’t have the appearance property in question. By this definition, the motion of moving images is a real, response-dependent property of those images. We will return shortly to this idea of response-dependent motion, but will first finish with Carroll’s argument. Its key move comes as he considers a putative problem with it: doesn’t it obliterate the category of perceptual illusion entirely? After all, a half-submerged straight stick will appear bent to any creature with something like our perceptual capacities; so doesn’t the stick’s bentness count as a real, response-dependent appearance property in Carroll’s view?37 And so are there no perceptual illusions at all? To dissipate this difficulty, Carroll argues that we can be disabused of perceptual illusions, but not of response-dependent appearances, using “our normal perceptual capacities and ordinary or everyday or conventional procedures”.38 We can find out that an apparently bent stick is not actually bent by taking it out of water. We can’t find out that a table is not solid without employing expensive and complicated apparatus. He then suggests that the means by which Kania suggests we disabuse ourselves of the “illusion” of motion are not ordinary, everyday, or conventional procedures; film projection technology is such that slowing a film down counts as an “extraordinary” process. Therefore, filmic motion is as real as the solidity of tables. The illusionist could well respond that there is something unsatisfactory, to say the least, about sorting perceptual illusions from objective appearances on the grounds of how easy it is to complicate or undermine the apparent deliverances of perception. For one thing, Carroll’s distinction is hostage to technological fortune. If, at some point in the future, advances allow an electron microscope in every garage, it will be everyday and conventional to disabuse ourselves of the appearance of solidity; solidity would thereby become a perceptual illusion. Nonetheless, the illusionist doesn’t have all the common sense on their side. The realist’s version of common sense goes along these lines: when we look at a film, we seem to see the image move; this is manifest in the experience we undergo. We should take this experience at face value, unless we have good  Carroll, The Philosophy of Motion Pictures, 89–92.  Carroll, The Philosophy of Motion Pictures, 90. 38  Carroll, The Philosophy of Motion Pictures, 91. 36 37



reason not to; we don’t have good reason not to. So, we should take the experience at face value, and agree that the images really move. This sketched argument for realism is essentially that presented by Currie.39 As you might expect, Currie defends the argument by concentrating on the crucial claim that we have no good reason to doubt that moving images really move. He claims, among other things, that cinematic images must actually move; that is, that one cannot coherently hold that cinematic images exist but do not move, else we’d only see static images in films, and this is wildly implausible.40 In response, Kania argues that many of the considerations Currie advances falsely, or perhaps question-beggingly, assume the presence of a continuous image on the screen.41 However, like Carroll, Currie mainly premises his defense of realism on the claim that the motion of film images is a response-dependent property. He does this by analogy with the idea that colors are response-dependent. Plausibly, we can’t specify what it is to be a certain color without reference to (typical, human) experience of that color; nonetheless, it’s true that things have colors, that colors aren’t illusions. We can say something similar about motion in films: we can’t say what such motion is without reference to our experience of it, but it’s not illusory, all the same. Currie’s argument has an advantage over Carroll’s, in that it doesn’t depend on questionable claims about the response-dependent nature of all appearance properties. Rather, it trades on the common (though eminently debatable) idea that some such properties, paradigmatically colors, are response-­dependent in the relevant sense. Nonetheless, Kania argues that the analogy is poorly drawn. Paradigmatic cases of motion, he says, necessarily involve a thing being in one place at one time, and then in a contiguous place at a contiguous time.42 If we understand motion so, film images don’t involve ordinary movement; they don’t change their spatial location in this manner. There is no image that occupies contiguous spaces. What Currie has done, effectively, is identify the special filmic illusion by calling it response-dependent motion, much as we might call the apparent property of a half-submerged stick response-dependent bentness. This is just the accusation to which we saw Carroll responding with somewhat desperate maneuvers, and it’s not clear what Currie could say against it. However, trading on the notion of response-dependent properties is not the only route open to the realist. Kania points out that, in the argument just ­outlined, Currie seems to distinguish two sorts of motion, and attribute one to images and the other to objects.43 The other option available, with which Currie seems to flirt, is to argue that the film image is a special sort of entity  Currie, Image and Mind, 34–42.  Currie, Image and Mind, 34–5. 41  Kania, “The Illusion of Realism”, 250. 42  Kania, “The Illusion of Realism”, 254. 43  Kania, “The Illusion of Realism”, 258, referring to Currie, Image and Mind, 47. 39 40



capable of real movement: an entity not quite like the medium-sized dry goods which we’re perfectly happy to say move, but enough like peculiar things such as shadows to which we more or less happily ascribe movement. So, we could retain a univocal conception of what movement is, and gently expand the range of things to which we’re willing to ascribe it. Against this, Kania argues that the slippery slope from shadows is far from frictionless. We’re happy to ascribe movement to shadows because the paradigmatic condition of contiguous positioning is fulfilled, and, indeed, we become less happy to do so when it isn’t. But that condition fails with regard to film because we can see (once we inspect the mechanisms of projection) that the condition is not actually fulfilled.44 Nor can the realist argue that our ability to apparently see the image proves that a continuous, moving image really exists; this would, again, be to effectively deny that there are any such things as illusions, since we would then apparently be committed to saying that seeing a bent stick in water proves that there is a bent-stick-entity (rather than, as before, a stick with the property of response-dependent bentness).45 Kania seems entirely right that the realist is faced with two fairly unpalatable options: make distinctions among types of motion, where we might previously have thought there were none, or carve out a class of novel entities capable of real movement. Nonetheless, we might explore further these options and, in so doing, cast some doubt on the illusionist’s position. We can start by asking what, exactly, the illusion that the illusionist attributes to film is meant to be. While the debate has often been conducted with respect or by analogy to such familiar perceptual phenomena as apparently bent sticks and Müller-Lyer lines, the really salient illusions are those of apparent motion: that is, cases where discrete, stationary projections or appearances in different locations at small time intervals produce the impression of motion of a single object. It seems that the illusionist wants to say that cinematic motion is an illusion of just this kind. However, Christoph Hoerl argues that the movement of cinematic images is not an illusion of apparent motion. One of his arguments for this conclusion relies on the observation that in many cases of apparent motion, there is an identifiable gap between the appearances of the two discrete stimuli; the impression of motion is created despite the subject being aware of that gap. The subject sees a dot flash to the left of an array, and then a dot flash to the right of the array, and sees them as distinct dots, but, nonetheless, still has some impression of movement. But, he argues, cinematic motion is not (usually) this sort of illusion, if it’s an illusion at all. The viewer does not see the images ­composing a film as distinct images and, nonetheless, have an impression of movement; the viewer really seems to see one continuous image.46  Kania, “The Illusion of Realism”, 254–5.  Kania, “The Illusion of Realism”, 255–6. 46  Christoph Hoerl, “Seeing Motion and Apparent Motion”, European Journal of Philosophy 23, no. 3 (2015): 692–94. 44 45



Hoerl’s argument depends on the claim that there is a phenomenological difference between experiences of apparent motion and experiences of real motion, and that experiences of moving images are phenomenologically closer to the latter. Apparent motion, he argues, “does not involve the visual presentation of something that is moving”. Rather, “apparent motion displays suggest that there is something moving, because they resemble paradigm cases of visible movement in certain respects”.47 The point is that apparent motion does not involve, or seem to involve, a single spatio-temporal particular reappearing at different points, whereas real motion does involve this appearance. So does cinematic motion, and thus cinematic motion is closer to real motion than it is to apparent motion. However, it’s not clear why the illusionist can’t accept all this and still maintain that the motion of images is illusory. Showing that such motion is not an illusion of apparent motion in Hoerl’s sense doesn’t show that it’s not an illusion at all, and so it doesn’t show that it should be classified as a case of real motion. Hoerl does have other arguments, though, which address both sides of Kania’s dilemma concerning multiplying motion types or pluralizing our ontology. Regarding motion, Hoerl argues that many cases of actual movement are not paradigmatic cases of visible movement. Some things move too slowly, some move too fast, some too far, and some not far enough, for their movement to be visible to us. Tectonic plates move, but we can’t see them doing so. So do insects’ wings, but their movement is too fast for us to track. The paradigmatic cases of visible movement involve spatial and temporal displacements of magnitudes that we humans are well-adapted to detect. Conversely, in cases such as illusions of apparent motion, the spatial and temporal displacements can be of magnitudes similar to those of paradigmatic cases of visible movement. All in all, then, there is an objective property of appearing to be in motion that is not enjoyed by all moving things and can be enjoyed by non-­ moving things.48 Hoerl’s aim is to explain illusions of apparent motion within a direct realist framework—that is, without invoking perceptual contents or sense-data to explain the impression of movement. The point of mentioning the argument here is to contrast it with Currie and Carroll’s notion of response-dependent motion; we leave it to the reader to make comparisons and relative evaluations. Hoerl also claims that Kania’s argument about slowing down the projector relies on a questionable assumption about what’s required for cinematic motion to count as real motion. This assumption is that cinematic motion is dissective. Once again, an analogy with color is employed. A pointillist painting may look green, while actually being made up of yellow and blue dots. Yet the painting really is green; it’s just that its real green-ness is only detectable at a certain distance. This shows that color is not dissective; that is, not every part of a  Hoerl, “Seeing Motion”, 694.  Hoerl, “Seeing Motion”, 686–7.

47 48



colored object needs to be that color. Indeed, conceivably, a painting such as this could be really green even though no part of it is actually green. Similarly, Hoerl argues, cinematic motion is not dissective. Under the right conditions, it looks like the images are moving; why not say they really are moving? If the motion is not dissective, the fact that it disappears under other conditions, or closer examination, does not mean it is not there in the first place.49 Kania, of course, is likely to respond that real motion is dissective, and so if cinematic motion is not, that just shows that we are once again diversifying types of movement and effectively naming the special cinematic illusion. Hoerl anticipates this response by, once more, arguing that we “should allow into our ontology re-identifiable spatial particulars other than physical objects. Once we do, these can clearly undergo movement”.50 So, again, we return to the possibility of diversifying our ontology. Perhaps it is really ontology that is the key to this debate. Both sides invite us to assimilate moving images to a certain kind of artifact, agree that that kind of artifact can or can’t undergo motion, and then agree that therefore moving images can or can’t really move. So the realist asks us to think of them as akin to shadows and to accept that shadows can move; the illusionist says we should accept that moving images are not anything more than the things they’re composed of, namely static images, which clearly cannot move. As this suggests, the illusionist takes it that there is no such thing as the moving image—there just isn’t a single image that could be moving. The realist takes it that there is, in fact, such an image. It’s an interesting question how we might settle this—that is, how we might establish the identity conditions of a moving image—without begging the question against one side or the other. The kind of thing moving images are will affect the kinds of movement of which they’re capable. And so, what one says about one issue will affect what one says about the other. But one might wonder which is the dog and which the tail. Because the question of ontological categorization usually arises in the literature in conjunction with the issue of movement, it’s not clear which way the motivation travels: whether thoughts about ontology motivate thoughts about movement, or vice versa. The reason that the ontological question is difficult, and interesting independently of its influence on the movement question, is the peculiarity of the moving image. On the one hand, moving images do not seem to have the same status as, say, celluloid images, which are fairly clear examples of concreta. As Currie remarks, “there does seem to be a difference between the substantial pictures we make contact with when we look at a painting or a photographic print, and the insubstantial pictures of film—the images on a screen”.51 On the other hand, cinematic images appear to be public in nature and, hence, do not  Hoerl, “Seeing Motion”, 689–90.  Hoerl, “Seeing Motion”, 690. For a similar argument, see Trevor Ponech, “External Realism about Cinematic Motion”, British Journal of Aesthetics 46, no. 4 (2006): 349–68. 51  Currie, Image and Mind, 30. 49 50



seem to be purely mental phenomena like mental imagery. So how shall we understand the image? The first thing to point out is that there are at least two things that might be meant by talk of “the moving image”: on the one hand, there is the visual display that takes up the entire screen or projection surface, and, on the other hand, there are the parts of that image which often appear to move—an image of a droid or a princess, for example.52 It is not always clear which sort of image various authors are referring to, but our primary interest is in the latter sorts of images, since these are the only plausible candidates for movement. The simplest account of moving images is that they are identical to patterns of light on a screen. This seems to be Gaut’s view of what moving images amount to, and on such an account they are real but do not move.53 Currie’s view is more complicated. He appears to identify moving images with patterns of color that are both mind dependent and extrinsically sustained.54 That is, these patterns are dependent both on the responses of viewers and on some sort of technological apparatus, and perhaps also on various screening conditions. So it seems that, for Currie, there is a moving image if typical observers typically experience a moving image under the right conditions, just as red is identifiable as that which is experienced by typical observers under the right conditions. Here, we have response-dependent ontology to go with response-­ dependent movement. But more needs to be said if this ontological conjecture is to support realism about movement, since the illusionist can just respond that to call this a single image is to beg the question—we haven’t established why we should call this a single image rather than the illusion of an image. There is one more argument against illusionism that might be made on the basis of Hoerl’s work, though he doesn’t quite make it himself. His main aim is to argue that apparent motion illusions can be accounted for within a direct realist framework, without appeal to representational content or sense-data. He does so by diversifying types of movement and types of objects capable of movement. Insofar as those are moves typical of realists, they are moves illusionists are likely to reject. This suggests that illusionists ought to be representationalists: they ought to think that illusions of motion are explained by something to do with representational content. In particular, they ought to appeal to content to explain the fact that there is a persistent and continuous object of experience, the moving image, despite their claim that there is no single image that is moving. The obvious representationalist way to do so, as Hoerl recognizes, is to argue that stimuli are “subjectively supplemented”: they are augmented with mental items that fill in the gaps in the series of stimuli such that something  Cf. Currie, Image and Mind, 35–6.  Gaut, A Philosophy of Cinematic Art, 65. 54  Currie, Image and Mind, 30–4. Gaut claims that Currie holds that cinematic images supervene on the light pattern on the screen, but this does not seem right. Currie’s supervenience claim is about the movement of the images. See Gaut, A Philosophy of Cinematic Art, 65, and Currie, Image and Mind, 40. 52 53



constant is presented.55 But one can see that, if one continues along this path, the moving image itself—the continuous thing that is the object of experience—is in danger of becoming a mental item: either an artifact of representational content or a sense-datum. And it may well be that the illusionist, and anyone else, should balk at the idea that moving images are essentially mental items. We haven’t the resources or space to go more deeply into this question, but one thing it does highlight is that the debate between illusionists and realists may in part be down to confusion over the relation between the (intentional) object of experience and the actual perceived object in the case of film. Realists sometimes appear to argue that, since there is a persistent object of experience, there must be a persistent object in the world to be experienced; more precisely, that since the object of experience appears to have a certain property, some actual object must have that property. That’s not a good argument. There might indeed be an actual image with the property of motion, but we can’t infer that from the fact that we experience motion. On the other hand, the illusionist needs to explain how there can be a continuous object of experience here without resorting to or falling into claiming that cinematic images are mental. What this might suggest, in turn, is that the whole focus of this debate is misdirected. One can get a feeling this is the case when one reads Currie claiming that our experience of film represents to us that there are images moving.56 If this is meant as a straightforward report of the experience of watching a film, it is plainly false. As Shelley argues, it is very rarely, if ever, the case that we seem to see images moving when we watch films.57 Rather, we seem to see things moving. It could be that there are images involved, but those are not things we seem to see, not things that figure in the contents of our experiences. And if it doesn’t seem to us in experience that we see an image move, then its seeming movement can neither be real nor illusory; there is no seeming movement of an image to explain. If Shelley is right, the illusionist and the realist are the proverbial two bald men fighting over a comb. Of course, if he’s right, he seems committed to some version of the transparency thesis or a more sophisticated explanation of how cinematic images can depict motion without themselves moving or seeming to move. Before we conclude, we wish to mention briefly one last point about the debate we have just surveyed. Therein, it seems that everyone implicitly thinks they are comparing the experience of film to experience of real motion, and that it is straightforward what that is. It is not, and there are reasons to doubt Currie’s claim that there is no position in the metaphysics of motion that makes a difference to the debate.58  Hoerl, “Seeing Motion”, 682–5. Hoerl does not endorse this thesis.  Currie, Image and Mind, 36. 57  Shelley, “Motion Sickness”. 58  Currie, Image and Mind, 34. 55 56



The usual, intuitive view of what motion is, whose codification is commonly attributed to Bertrand Russell, is that, necessarily, something moves if and only if it is in one place at one time, and at a different place at a different time.59 As a corollary, it is said that for something to be in motion at an instant is for it to have a velocity over intervals. This seems to be more or less the account being assumed by the participants in the debate over filmic motion. Now, elements of the received view have been questioned: for example, John W. Carroll discusses problems with the notion of motion at an instant, and Shieva Kleinschmidt argues that the metaphysical possibility of multiply-located entities presents problems for the stated definition of movement.60 Owing in part to such problems, Graham Priest endorses a “Hegelian” account of motion, according to which motion is inconsistent: “to occupy more than one place (in fact a continuum of places) at the same time, and hence both to be and not to be in some place”.61 It’s an interesting question, deserving further study, whether what we say about real motion might affect what we say about supposed contrast cases such as the movement of images. Such study might also consider Chris Mortensen’s contention, based partially on Priest’s work, that the experience of motion has inconsistent contents: motion perception is, therefore, inconsistent.62 Teasing out the implications of this position for the debate on moving images is beyond our scope here, but it seems clear at least that the “real motion” and “experiences of real motion” with which cinematic motion is compared are less well understood than we might assume.

Conclusion In conclusion, here are two points which emerge from this survey of various ways to account for the apparent motion of the film image, which is so often taken to be a necessary aspect of such images. The first is that, if we are ever going to make sense of the idea of the motion of moving images, a lot is going to depend on how we make sense of adjacent and antecedent notions. Many of the issues revolve around comparisons between “real” things and not-real things: in the case of transparency, real seeing is compared with the ways in which film allows us to see; in the case of depiction, real perspective is compared with constructed perspective; in the case of illusionism, real motion is compared with illusory motion, and real appearance properties with somehow 59  Bertrand Russell, Principles of Mathematics (Cambridge: Cambridge University Press, 1903), sec. 447. 60  John W.  Carroll, “Instantaneous Motion”, Philosophical Studies 110, no. 1 (2002): 49–67; Shieva Kleinschmidt, “At It Again: Time-Travel and the At-At Account of Motion”, Erkenntnis 82, no.2 (2017): 185–98. 61  Graham Priest, In Contradiction: A Study of the Transconsistent, expanded ed. (Oxford: Oxford University Press, 2006), 180. 62  Chris Mortensen, “Motion Perception as Inconsistent”, Philosophical Psychology 26, no. 6 (2013): 913–24.



less real ones. In each case, it could well be that it will be much more clear how we should think about the moving image once it has been established how exactly we should think of the real things with which aspects of it are being compared. As a strategic move, however, sitting on our hands as we wait for the philosophy of perception to be settled has little to recommend it, and we might therefore pay more heed to a second point suggested by the foregoing. The thought that films show movement is obvious and intuitive, as is the thought that this is a distinctive feature of moving images, and so it is natural to focus closely on them and their movement. But the intricacies involved in elaborating and explaining moving images’ motion might lead to the thought that it is hard to say much decisive when considering motion in isolation from other properties and aspects of moving images. And this is true even when considering moving images in isolation from other aspects of film and filmic art. We might do better to treat the motion of the moving image as an important member of a cluster of qualities that make such images important, interesting, and captivating.

Bibliography Abell, Catharine. 2009. Canny Resemblance. Philosophical Review 118 (2): 183–223. Carroll, Noël. 1995. Towards an Ontology of the Moving Image. In Philosophy of Film, ed. Cynthia Freeland and Thomas Wartenberg, 66–85. New York: Routledge. ———. 1996. Theorizing the Moving Image. Cambridge: Cambridge University Press. Carroll, John W. 2002. Instantaneous Motion. Philosophical Studies 110 (1): 49–67. Carroll, Noël. 2008. The Philosophy of Motion Pictures. Oxford: Blackwell. Cohen, Jonathan, and Aaron Meskin. 2004. On the Epistemic Value of Photographs. Journal of Aesthetics and Art Criticism 62 (2): 197–210. Costello, Diarmuid, and Dawn M. Phillips. 2009. Automatism, Causality and Realism: Foundational Problems in the Philosophy of Photography. Philosophy Compass 4 (1): 1–21. Currie, Gregory. 1995. Image and Mind: Film, Philosophy and Cognitive Science. Cambridge: Cambridge University Press. Danto, Arthur C. 1979. Moving Pictures. Quarterly Review of Film Studies 4 (1): 1–21. Deleuze, Gilles. 1986. Cinema 1: The Movement-Image. Trans. Hugh Tomlinson and Barbara Habberjam. London: Athlone Press. ———. 1989. Cinema 2: The Time-Image. Trans. Hugh Tomlinson and Robert Galeta. London: Athlone Press. Gaut, Berys. 2010. A Philosophy of Cinematic Art. Cambridge: Cambridge University Press. Goodman, Nelson. 1968. Languages of Art: An Approach to a Theory of Symbols. Indianapolis: Bobbs-Merrill. Hoerl, Christoph. 2015. Seeing Motion and Apparent Motion. European Journal of Philosophy 23 (3): 676–702. Hopkins, Robert. 1998. Picture, Image and Experience: A Philosophical Inquiry. Cambridge: Cambridge University Press.



———. 2008. What Do We See In Film? Journal of Aesthetics and Art Criticism 66 (2): 149–159. ———. 2009. Depiction. In The Routledge Companion to Philosophy and Film, ed. Paisley Livingstone and Carl Plantinga, 64–74. London: Routledge. Hyman, John. 2012. Depiction. Royal Institute of Philosophy Supplement 71: 129–150. Kania, Andrew. 2002. The Illusion of Realism in Film. British Journal of Aesthetics 42 (3): 243–258. Kleinschmidt, Shieva. 2017. At It Again: Time-Travel and the At-At Account of Motion. Erkenntnis 82 (2): 185–198. Kulvicki, John. 2006. On Images: Their Structure and Content. Oxford: Oxford University Press. Lopes, Dominic McIver. 1996. Understanding Pictures. Oxford: Oxford University Press. ———. 1999. Pictorial Color: Aesthetics and Cognitive Science. Philosophical Psychology 12 (4): 415–428. Mortensen, Chris. 2013. Motion Perception as Inconsistent. Philosophical Psychology 26 (6): 913–924. Ponech, Trevor. 2006a. The Substance of Cinema. Journal of Aesthetics and Art Criticism 64 (1): 187–198. ———. 2006b. External Realism about Cinematic Motion. British Journal of Aesthetics 46 (4): 349–368. ———. 2007. Cinema Again: A Reply to Walley. Journal of Aesthetics and Art Criticism 65 (4): 412–416. Priest, Graham. 2006. Contradiction: A Study of the Transconsistent. expanded ed. Oxford: Oxford University Press. Russell, Bertrand. 1903. Principles of Mathematics. Cambridge: Cambridge University Press. Schier, Flint. 1986. Deeper into Pictures. Cambridge: Cambridge University Press. Shelley, James. Motion Sickness. Unpublished manuscript, May 2017. Microsoft Word file. Walley, Jonathan. 2007. On Ponech on the Essence of Cinema. Journal of Aesthetics and Art Criticism 65 (4): 408–412. Walton, Kendall L. 1984. Transparent Pictures: On the Nature of Photographic Realism. Critical Inquiry 18 (1): 67–72. ———. 1986. Looking Again through Photographs: A Response to Edwin Martin. Critical Inquiry 12 (4): 801–808. ———. 1990. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, MA: Harvard University Press. Wilson, George. 2011. Seeing Fictions in Films. Oxford: Oxford University Press. Wollheim, Richard. 1998. On Pictorial Representation. Journal of Aesthetics and Art Criticism 56 (3): 217–226. Yanal, Robert. 2008. Defining the Moving Image: A Response to Noël Carroll. Film Philosophy 12: 135–140.


The Art of Cinematography Patrick Keating

Over 80 years ago, the Hollywood cinematographer Victor Milner wrote that his job was to “attune the visual mood of the picture to the dramatic mood of the story.”1 British cinematographer Douglas Slocombe defined the function of cinematography in similar terms: “Not merely to reproduce something, but to comment on it and translate it on to the screen in the ‘mood’” that best fit the story.2 For French cinematographer Henri Alekan, the task of the cinematographer was not confined to “humbly reproducing” the subject; instead, the cinematographer was supposed to “create” a shifting series of atmospheres.3 All three practitioners emphasized the cinematographer’s power to give artistic shape to the films they shot. Admittedly, Milner, Slocombe, and Alekan were not unbiased observers; they had an interest in defining their craft under the most favorable terms. But I think they were onto something. The cinematographer is an artist—not just a technician but a picture-maker, a storyteller, and, in some meaningful sense of the word, an author. In this chapter, focusing on the narrative fiction film, I will define and defend the cinematographer’s contribution to film art. Part One clears some ground by clarifying the sort of “picture-making” I have in mind, arguing that a cinematic image may reasonably be described as a picture of a fictional character. Part Two, the heart of the chapter, considers how a cinematographer might use a range of techniques, such as camera 1  Victor Milner, “‘Miscasting’ the Cinematographer,” American Cinematographer 13 (February 1933): 13. 2  Douglas Slocombe, “The Work of Gregg Toland,” Sequence 8 (Summer 1949): 70. 3  Alekan, quoted in René Predal, “Les grands operateurs (IV),” Cinéma 73 173 (February 1973): 93. My translation.

P. Keating (*) Trinity University, San Antonio, TX, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




­ ovement and lighting, to contribute to a film’s larger storytelling goals. Part m Three defends the idea that films are authored collaboratively and explains how cinematographers might be understood as members of a collaborative authorial group. Throughout, I will argue that we can find considerable continuity between the film-driven techniques of traditional cinematography and the computer-driven techniques of today’s digital cinematography. Digital tools increase the filmmaker’s ability to shape the image in meaningful ways, but those abilities have been quite powerful for some time.

Preliminary Considerations: Picture-Making Cinematography is the craft of producing moving images—typically, by photographing a series of still images that produce the appearance of motion when they are projected at a suitable speed. Suppose we are watching the 1944 film Double Indemnity, directed by Billy Wilder and photographed by John F. Seitz. A strip of film, containing a series of stills, is running through the projector. When we look to the screen, we see projected pictures of Phyllis Dietrichson, the femme fatale played by Barbara Stanwyck. Is it fair to describe these pictures as pictures of Phyllis, even though Phyllis is a dramatic role? I think it is. Here, I take inspiration from an important article by Catharine Abell, who has defended the view that a fiction film may “primarily depict” its fictional characters the way a documentary primarily depicts its non-fictional subject. Abell writes, “Cinematic representation, I propose, involves primary depiction. […] The computer-generated film Shrek primarily depicts Shrek, Princess Fiona, and Donkey; films produced by painting onto film stock primarily depict the things that are painted onto it. Likewise, the documentary Crumb primarily depicts Robert Crumb; All About Eve primarily depicts Margot Channing; and Les Triplettes de Belleville primarily depicts the Belleville sisters.”4 This proposal provocatively groups the photographic film All About Eve with various kinds of animated works, suggesting that a fiction movie may depict fictional characters in various ways. To be sure, a photographic film may simultaneously depict actors giving performances, as All About Eve depicts Bette Davis playing the role of Margot Channing. But the fact that the film depicts its actors does not preclude the possibility that it also depicts its characters. Abell’s proposal has the significant advantage of retaining a way of speaking that seems quite natural in ordinary conversation. Looking at a newspaper’s comics pages, one might readily describe certain images as pictures of fictional characters, as when one says that a cartoon is a picture of Charlie Brown, even though Charlie Brown does not exist. Indeed, in an ordinary-language context, I simply cannot imagine someone looking at a relevant Peanuts cartoon and insisting “That is not a picture of Charlie Brown.” It does not seem 4  Catharine Abell, “Cinema as a Representational Art,” British Journal of Aesthetics 50, no. 3 (July 2010): 277.



­ nreasonable to extend this way of speaking about pictures to the cinema. It u would be perfectly understandable to watch Snow White and the Seven Dwarves (David Hand, 1937) and say that we are seeing moving pictures of Snow White. Pushing the case a little farther, the viewer of The Avengers (Joss Whedon, 2012) might look at the computer-generated image of a large green man and describe it, quite reasonably, as a picture of The Hulk. Our usage of the phrase “picture of…” is quite broad—broad enough to include Charlie Brown, Snow White, The Hulk, and other fictional characters. The next step is more challenging. If we are willing to say that all of the images listed above are pictures of fictional characters, then we should be willing to say that the images in Double Indemnity are pictures of Phyllis Dietrichson. Admittedly, there are two immediate problems with this view. First, the images in Double Indemnity are also pictures of Barbara Stanwyck, and it seems odd to say that one image can depict two subjects. However, the oddness recedes when we consider precedents in other arts. As Abell explains, “On the assumption that Rembrandt intended both Bathsheba and Hendrickje Stoffels, whom he used as a model, to be identifiable in his painting Bathsheba, it will turn out to depict both women, since both are identifiable in the picture.”5 If a painting may depict two subjects at once, then so may a film. To make sense of such cases, we can take advantage of the distinction between primary and secondary depiction (a distinction that Abell credits to Dominic Lopes). The images in Double Indemnity primarily depict Phyllis Dietrichson and secondarily depict Barbara Stanwyck.6 A second challenge to my view is the fact that the images in Double Indemnity are photographic, and therefore different from paintings, and different from the images in cartoons, animated films, and movies packed with CGI (computergenerated imagery). A photographic image has a causal relationship with its subject. Because Phyllis Dietrichson does not exist, it is simply not possible to take a photograph of Phyllis Dietrichson. The best one can do is take a photograph of an actor playing Phyllis Dietrichson. This reasoning is sound, but I do not claim that the images in Double Indemnity are photographs of Phyllis Dietrichson. I merely claim that they are pictures of Phyllis Dietrichson, just as the images in Snow White and the Seven Dwarves are pictures of Snow White and the images in The Avengers are pictures of The Hulk. The filmmakers of Double Indemnity intended to tell a story about Phyllis Dietrichson (among other characters). To do so, they made use of several tools—writing, acting, music, costume design, editing, and, yes, photography. Animation is one of cinema’s ways of producing pictures of fictional characters. Photography is another. A hypothetical scenario might make this proposal sound more plausible. Suppose we were to learn that Barbara Stanwyck failed to show up for photography one day on the set of Double Indemnity. And suppose that the ­filmmakers  Abell, “Cinema as a Representational Art,” 275.  Or, if you prefer, the other way around. For my purposes, it is only necessary to establish that it makes some sense to speak of pictures of fictional characters. 5 6



had access to some cutting-edge proto-CGI technology, allowing them to produce pictures of Stanwyck playing Phyllis artificially. In such a case, we might be willing to count the (hypothetical) computer-generated shots of a woman with an anklet as pictures of Phyllis, the same way we might count computergenerated images of a large green man as pictures of The Hulk. I think it would be odd to say that some of these images (the computer-­generated ones) are pictures of Phyllis and that the rest (the photographic ones) are not. Watching the film, they all serve as—and, in my view, are—pictures of Phyllis. For that matter, they are all pictures of Stanwyck. The difference is that only some of them are photographs of Stanwyck. (None of them are photographs of Phyllis.) In 1944, this example would have seemed far-fetched. Now it seems pretty routine. Big-budget Hollywood movies shift between photographic images and CGI all the time, often imperceptibly. Even in the 1940s, a film might have depicted its characters and situations in various ways—photographing actors against rear-screen projections, replacing actors with stunt doubles, or combining two or more settings into one apparent space via an optical printer. Filmmakers had a range of techniques for producing pictures, one of which was photography. There are other important ways of thinking about the problem of cinematic depiction, such as Robert Hopkins’s proposal that we experience a “collapsed” form of seeing-in when watching “two-tier” films and Gregory Currie’s argument that a sequence of cinematic images may “give us information about” the fiction, without performing the impossible task of “representing the characters of the fiction.”7 Both are compelling theories that explain certain problems that I cannot address here. Here, I will continue to rely on the everyday usage of the phrase “picture of…” because I think it will help us understand the sorts of contributions that cinematographers might make to the art of cinema—contributions that most often respond to the problem of picturing characters in expressive ways. Earlier, I wrote that the craft of cinematography is the craft of producing moving images. Now I can make the claim a little more precise. In the realm of the mainstream narrative fiction film, the craft of cinematography is the craft of producing pictures of fictional characters, objects, and places. How might a cinematographer use the resources of the craft to produce those pictures expressively?

The Resources of Cinematography Many works in the philosophy of photography pay special attention to photography’s causal relationship with its subjects. Roger Scruton goes so far as to exclude idealized forms of photography and cinema from the realm of the representational arts because they are incapable (at least, qua photography and 7  Robert Hopkins, “What Do We See in Films?,” The Journal of Aesthetics and Art Criticism 66, no. 2 (Spring 2008): 149–159; Gregory Currie, Image and Mind: Film, Philosophy, and Cognitive Science (New York: Cambridge University Press, 1995), 11.



cinema) of expressing an artist’s intentions.8 By contrast, defenders of the cinema as a representational art, such as Abell, argue that “cinematic works can have aesthetic value in virtue of expressing their makers’ thoughts about their primary subjects.”9 This section adopts the latter strategy. Focusing on the narrative fiction film, I will list some of the techniques a cinematographer might use to craft pictures with an expressive purpose. Film Stock and Lab Work In a Hollywood film, the person in charge of producing the pictures is the cinematographer, who leads several teams of craftspeople, including the camera crew, the grip crew, and the electrical or lighting crew.10 Some industries treat the lighting designer as a separate job. On set, the cinematographer usually answers to the director, who may dictate instructions to the cinematographer with more or less precision. For most of cinema’s history, cinematographers have photographed movies on film stock, such as 16 mm or 35 mm. In the last decade, digital cinematography has become the norm, though some prominent directors and cinematographers still prefer to work with traditional film.11 In both cases, the cinematographer may express ideas by making judicious choices regarding the handling of the medium. Let us consider the case of traditional film first. The choice of film stock can make a significant difference to the “look” of the finished film. The point is obvious when we compare black-and-white stock to color stock, but it remains true when we consider variations within those two broad categories. For instance, two black-and-white film stocks might differ in terms of latitude—that is, their responsiveness to a range of light values. Suppose that two cinematographers—call them Judy and Woody—are photographing the same scene. On set, the brightest part of the scene is 50 times brighter than the darkest part of the scene. Judy has selected a stock with a latitude of six stops. Because one stop is equivalent to a doubling of the amount of light, such a stock is capable of preserving detail as long as the brightest part of the scene is no more than 64 times brighter than the darkest part of the scene. In other words, Judy is capable of registering the entire scene, from the shadows to the highlights. By contrast, Woody has selected a stock with a latitude of five stops. Such a stock is capable of preserving detail as long as the brightest part of the scene is no more than 32 times brighter than the darkest 8  Roger Scruton, “Photography and Representation,” The Aesthetic Understanding (London: Methuen, 1983), 103. 9  Abell, “Cinema as a Representational Art,” 285. 10  In brief, the electrical crew sets up the lamps, but the grip crew modifies the light by setting up stands to create shadows, diffusion, and other effects. The grip crew also works with the dolly and crane, while the camera crew works with the camera and the tripod. 11   Christopher Lucas, “The Modern Entertainment Marketplace, 2000-Present,” in Cinematography, ed. Patrick Keating (New Brunswick, NJ: Rutgers University Press, 2014), 133–140. Christopher Nolan is an example of a director who prefers film.



part of the scene. If Woody opens up the aperture, he will be able to register detail in the shadows, but the highlights in the finished film will register as white, with no detail.12 If he closes down the aperture, Woody will be able to register detail in the highlights, but the shadows will register as black, with no detail. If Woody decides that he does not like these options, he may alter the lighting itself, perhaps adding more illumination to the shadows, thereby ensuring that the scene will “fit” within his latitude of five stops. Both film stocks preserve aspects of the scene’s tonal scale. Within the latitude, darker objects will appear darker in the finished film, and lighter objects will appear lighter. In that sense, the resulting picture resembles the original scene in terms of its patterns of light and dark. Indeed, the causal nature of photography may allow the filmmaker to capture such gradations with exceptional nuance, arguably exceeding the abilities of most painters. Still, the choice to use one film stock rather than another remains a creative choice. Woody might prefer the look of a narrow-latitude film stock, and Judy might prefer the look of a wide-latitude alternative. In addition to the variable of latitude, stocks vary in several potentially meaningful ways. Whether black-and-white or color, some film stocks are more sensitive to (blue) daylight than they are to (orange) tungsten-balanced light. Some film stocks have more visible grain than others. Some film stocks are fast, responsive to low levels of illumination, and some film stocks are slow, requiring more illumination to produce an exposure. When we add the variables of developing and printing, we see that the filmmaker’s ability to shape the image via cinematography is considerable. A few historical examples will give a better idea of the range of options. During the 1930s, the standard practice among black-and-white cinematographers at MetroGoldwyn-Mayer was to add extra illumination to each scene, compensating for the added brightness by reducing the negative’s time in development. This procedure produced the studio’s characteristic low-contrast style, featuring gentle gradations of gray instead of intense highlights and black shadows. Meanwhile, the standard practice at Warner Bros. was the reverse: Cinematographers would use less illumination and the lab would compensate with increased development time, resulting in high-contrast images with darker shadows.13 In each case, the style suited the studio. The wealthy M-G-M could afford to spend more money on electricians to operate the extra lamps, and its glamorous stars benefitted from the softer look. The more cost-conscious Warner Bros. favored a cheaper style, and the studio’s tough-guy stars had less need of glamor.

12  I have simplified the example in various ways, most notably by ignoring the fact that most film stocks are negatives. On the negative, a highlight will appear black, and a shadow will appear clear. These values are reversed on the positive print. If anything, this complication strengthens the case that a cinematographer may express a thought through the handling of film stock, because the process of using a negative to make a positive print introduces another stage where choices must be made, regarding the printing of the positive image. 13  Barry Salt, Film Style and Technology: History and Analysis, 3rd ed. (London: Starword, 2009), 253.



After the end of the studio system, cinematographers had more freedom to develop signature looks. During the 1970s, cinematographer Vilmos Zsigmond gained considerable esteem for his work with “flashing,” exposing the negative to a small amount of light before or after shooting. In films like the revisionist Western McCabe and Mrs. Miller (Robert Altman, 1971), the negative would pick up extra details in the shadows, lowering the contrast while adding a slight sepia tone that mimicked the appearance of nineteenth-century photography.14 The 1990s brought considerable experimentation with film stocks and processing, at a time when digitization loomed on the horizon. By using tungsten-­ balanced stocks outside, a cinematographer could produce a blue look that could be enhanced further with blue filters, as on Jane Campion’s The Piano (1993, photographed by Stuart Dryburgh).15 By mixing together 8 mm, 16 mm, and 35 mm stocks, a cinematographer could alter the film’s grain structure from scene to scene, with equally visible shifts in color and latitude, as on Oliver Stone’s JFK (1991, photographed by Robert Richardson), filmed on a total of 14 different stocks.16 By working with the laboratory to restore silver to the negative or positive during processing, a cinematographer could darken and desaturate the shadow areas, producing deep pools of black, as on David Fincher’s Se7en (1995, photographed by Darius Khondji).17 Even before the considerable expansion of digital tools in the 2000s, cinematographers had a wide range of options at their command. The best cinematographers patterned their decisions, creating a progression that underlined the structure of the story. Consider Ernest Dickerson’s work on Spike Lee’s Malcolm X (1992). The story may be divided roughly into three large sections. In the first section, Malcolm (Denzel Washington) is a minor criminal; in the second, he goes to jail; in the third, he becomes an activist. Each section features a different style. At first, the cinematographer favors highly saturated images, taking full advantage of Wynn Thomas’s colorful production design. Next, Dickerson employs strong contrasts with desaturated colors, as if all the color had been drained out of Malcolm’s life. After Malcolm emerges from prison with a new outlook, Dickerson shifts to a more balanced strategy, reintroducing color while avoiding the garish over-saturation of the first section.18 The visual progression echoes the progression of the narrative, which tells the story of a man who finds his identity after experiencing various alternatives. 14  See Herb Lightman, “The New Panaflex Camera Makes its Production Debut,” American Cinematographer 54, no. 5 (May 1973): 619; and Vilmos Zsigmond, “Behind the Cameras on Heaven’s Gate,” American Cinematographer 61, no. 11 (November 1980): 1110. 15  Ric Gentry, “Painterly Touches,” American Cinematographer 78, no. 1 (January 1997): 56. This article also explains how Campion and Dryburgh produced four distinct color palettes for four different stages of the story in The Portrait of a Lady (1996). 16  Bob Fisher, “The Whys and Hows of JFK,” American Cinematographer 73, no. 2 (February 1992): 45. 17  David E.  Williams, “The Sins of a Serial Killer,” American Cinematographer 76, no. 10 (October 1995): 37. 18  Al Harrell, “Malcolm X: One Man’s Legacy, to the Letter,” American Cinematographer 73, no. 11 (November 1992): 34.



Arguments insisting on the causal nature of cinema and photography often rely on an idealized notion of the practice, asking us to imagine a photograph taken using standard exposure, standard developing time, and standard color timing. As Paloma Atencia-Linares explains, “Philosophical accounts of photography have generally provided an oversimplified view of the photographic practice,” ignoring many crucial steps in the process.19 Suppose you want to take a photograph of a street on a bright sunny day. The range of illumination may exceed the latitude of your film stock, no matter which stock you choose. Do you expose for the highlights, sacrificing detail in the shadows? Do you expose for the shadows, sacrificing detail in the highlights? Or do you set your exposure somewhere in the middle, hoping to capture most of the scene while losing detail in the very brightest and very darkest spots? Each choice produces a different picture. The same logic applies to later stages, such as development and printing. On a major motion picture film, the cinematographer works with a color timer to set the printing lights for each shot. The color timer may start with a default set of printing lights (say, 25-25-25, indicating an equal balance of red, green, and blue light), but those lights can be dialed up or down, altering the color balance and the density of the positive print. To be sure, the camera’s causal process captures an astonishing range of details, including details that may escape the cinematographer’s notice or control. But the existence of a causal connection should not prevent us from noticing how much of the process can and must be guided by choices, from early choices about film stocks to later choices about printing lights. Most of these observations apply, with even more force, to cinematography in the digital age. The role of color timer is now occupied by a color grader, equipped with considerably more powerful digital tools. Whereas a color timer had the ability to alter the overall color balance of a particular shot (for instance, by adding blue to the entire composition), a color grader can now alter the color balance of every detail (for instance, by adding blue to one character’s jacket). Even a film that was photographed on traditional 35 mm film stock may take advantage of the color grader’s services. Cinematographer Roger Deakins photographed O Brother Where Art Thou? (2000, directed by the Coen Brothers) in 35 mm, which was then transferred to a digital intermediate. A color grader adjusted the colors digitally, and the result was printed back onto film stock.20 Since then, the digital intermediate has become the norm in Hollywood production for movies shot on film. Meanwhile, the majority of Hollywood movies have been shot (or “captured”) on digital cameras. Like a film camera, a digital camera may produce pictures with more or fewer pixels  Paloma Atencia-Linares, “Fiction, Nonfiction, and Deceptive Photographic Representation,” The Journal of Aesthetics and Art Criticism 70, no. 1 (Winter 2012): 21. 20  Lucas, “The Modern Entertainment Marketplace,” 134–135. Deakins defends his use of the digital intermediate by situating the technology within a broader history of cinematographic craft. See Roger Deakins, “The DI, Luddites, and other Musings,” American Cinematographer 89, no. 7 (October 2008): 78–83. 19



(roughly similar to grain), with more or less dynamic range (roughly similar to latitude), and with more or less responsiveness to low light levels (roughly similar to speed). A digitally captured image may receive gentle or extensive manipulation in the hands of a grader. Indeed, some cinematographers worry that the digital grader has become too powerful.21 Traditionally, cinematographers made many of their most important decisions on set, while working with the camera, electrical, and grip crews. Now the grader has astonishing new powers to alter the cinematographer’s expressive interpretations. Lighting In The Godfather (Francis Ford Coppola, 1972), the great cinematographer Gordon Willis famously lit actor Marlon Brando from above, casting deep shadows over his eyes. The result is certainly artistic, but what sort of art has Willis produced? From one perspective, Willis’s achievement would appear to be dramatic, rather than cinematic. Like a theatrical lighting designer, Willis and his electrical crew have lit the actor in a particular way, expressing the mysterious danger of his character by making his eyes hard to see. Film lighting does indeed have a great deal in common with theatrical lighting; however, the tasks are not identical for several reasons. First, Willis’s decisions regarding exposure impacted the ultimate appearance of the lighting effect. Willis could have exposed for the shadows, making the eyes visible while leaving the rest of the face overexposed, or he could have exposed for the light, making the illuminated parts of Brando’s face visible while leaving the eyes underexposed. (Willis opted for the latter solution.) Second, the cinematographer needed to take the latitude of the stock into account when making decisions about lighting. Willis could have underexposed the eyes by a small amount, allowing the eyes to remain within the latitude of the stock, or he could have underexposed the eyes by several stops, allowing the eyes to go completely black. In practice, Willis judged the lighting very carefully, allowing us to see just a hint of detail in the eyes—not totally black, but not readily visible, either. Third, Willis needed to take the direction of the light into account, in ways that were not identical to the concerns of a theatrical lighting designer. Consider the problem of “eyelight.” Suppose a theatrical designer set up a lamp that produced a glimmer of light in an actor’s eyes. In the theater, the spectators would be spread out, witnessing the action from multiple angles. Some spectators would see the reflection in one part of the actor’s eye, others would see the reflection in another part of the actor’s eye; still, others might fail to see the reflection at all. In the cinema, the camera typically photographs the actor from one angle at a time. A cinematographer who wants to add an eyelight must place each lamp with precision, taking note of the reflections as they appear from the cam21  See the discussion between cinematographer John Bailey and colorist Stefan Sonnenfeld in Jon Silberg and Steven Pizzello, “Cinematographers, Colorists, and the DI,” American Cinematographer 90, no. 6 (June 2009): 78–82.



era’s “monocular” perspective. A cinematographer who wants to eliminate an eyelight must be equally careful, placing each lamp to ensure that none produces the unwanted reflection. Willis had to work very hard to keep Vito’s eyes as dark as they appear on film. As we have seen, cinematographers took special pride in their ability to adjust lighting over time to suit the changing demands of a story. Recall Victor Milner’s desire to “attune” a film’s lighting moment by moment, Douglas Slocombe’s call to “translate” the mood of the story, and Henri Alekan’s urge to “create” the right atmosphere. In the absence of a universally accepted term, I propose the word “modulation” to describe this technique of altering the lighting as a story develops. We can find modulation on a large scale, as when a film alters its lighting from scene to scene or act to act, and we can find modulation on a small scale, as when a film alters its lighting within a scene. An example of large-scale modulation is Vittorio Storaro’s work on Bernardo Bertolucci’s The Last Emperor (1987). In interviews, Storaro explained that he assigned a dominant hue to each section of the story, moving through all of the colors of the rainbow (red, orange, yellow, green, blue, indigo, and violet) and culminating in white, representing the synthesis of all the colors.22 In the finished film, the distinctions are not as obvious as Storaro’s comments might suggest, but the general strategy of shifting the palette as the story progresses remains quite visible, as in early scenes where Storaro and his crew have flooded the sets with orange light or in later scenes where the nighttime windows are rendered in an almost unnatural purplish blue. As an example of small-scale modulation, consider a scene in Ernst Lubitsch’s Ninotchka (1939). The protagonist, played by Greta Garbo, is living in the Soviet Union, thousands of miles away from her lover in Paris. In one scene, Ninotchka is sitting at a table with three friends. A small lamp hangs overhead. The lighting on Ninotchka’s face mimics the appearance of the overhead lamp by producing shadows under her nose and chin. Within this relatively stable framework, cinematographer William Daniels modulates the lighting’s contrast from shot to shot. Early in the scene, the lighting is relatively high-contrast: The shadows under Ninotchka’s nose and chin are fairly dark. Later, Ninotchka receives a letter from her lover. When she reads the letter, the film cuts to a medium close-up of Ninotchka’s face. She is still lit from above, but now the shadows under her nose and chin receive more fill light, producing a lower-­ contrast image. Then, Ninotchka learns that the letter has been censored. The final shot of the scene shows Ninotchka sitting under the lamp, and the l­ ighting has changed once again. Now the “top-light” effect is a little more extreme, casting longer shadows under her nose and chin, while making her eyes diffi Storaro, quoted in Ray Zone, “The Literature of Light: An Interview with Vittorio Storaro, ASC, AIC,” in Writer of Light: The Cinematography of Vittorio Storaro, ASC, AIC, ed. Ray Zone (Hollywood: ASC Press, 2000), 85–86. I discuss several examples of Storaro’s work, including this one, in Patrick Keating, “What Does It Mean to Say that Cinematography Is Like Painting with Light?,” in Transnational Cinematography Studies, ed. Lindsay Coleman, Daisuke Miyao, and Roberto Schaefer (Lanham, MD: Lexington Books, 2017), 109. 22



cult to see. The shadows are also darker, resulting in more contrast. Notice how the character’s mood changes over the course of the scene: Ninotchka is glum, then hopeful, and then in despair. The scene’s lighting shifts in comparable ways, starting out with a plausible representation of the overhead lamp effect, shifting toward softness when Ninotchka experiences a moment of hope, and then shifting again toward a more brutal top light as Ninotchka’s hopes are dashed. These modulations might seem troubling to some. Should we understand that the lighting in the room is actually changing, as if the lamp had a magical power to respond to Ninotchka’s moods? Or are we supposed to understand that the illumination from the hanging lamp remains constant from beginning to end, regardless of the pictorial evidence that it is changing whenever Ninotchka turns happy or sad? The example illustrates a more general problem. Once we take the time to look, we may notice that films are filled with lighting effects that make no sense. In one scene, a star’s hair may be aglow, illuminated by a powerful backlight, even when there is no light source behind the character. In another scene, the star may be dimly visible, even when the film has established that the character is in a pitch-black room. Notice that the problem is not exclusive to photographic cinema. We might see the same inconsistent lighting in an animated film, or a computer-­generated film, or indeed in a printed comic book or a painting. Perhaps the problem of inconsistency is nothing more than a “silly question.” Many philosophers have argued that it is pointless to ask questions about certain seemingly nonsensical features of represented worlds. Mocking such silly questions, Kendall Walton has asked, “Why do all thirteen of the diners in Leonardo’s Last Supper line up in a row on the same side of the table? […] Must we suspect that they are fearful of facing one another—of kicks under the table or bad breath?”23 Similarly, we might mock questions about inconsistent lighting as obtuse, failing to understand that the shifting lighting serves a pictorial purpose. My own view is that it is useful to draw a distinction between what we see and what we understand. We see pictures of Ninotchka—pictures that are designed in particular ways. On the basis of these pictures, we come to understand the character and her situation. The pictures are means to storytelling ends. Depending on the nature of those ends, the pictures may contain various design features, including some visible inconsistencies. On this view, we see pictures of Ninotchka in and out of the shadows, but the film as a whole does not give us reason to understand Ninotchka’s world as a place where light magically adjusts to her moods. Seeing the pictures is one thing, understanding the story-world is another. Watching a film, I see the pictures moment by

23  Kendall Walton, Mimesis as Make-Believe: On the Foundations of the Representational Arts (Cambridge: Harvard University Press, 1990), 175. I have benefitted from Murray Smith’s discussion of this passage. See Smith, “On the Twofoldness of Character,” New Literary History 42, no. 2 (Spring 2011): 290.



moment, but I develop and revise my understanding of the story-world by taking the norms and logic of the entire film into account. Whether this proposal is persuasive or not, the problem of inconsistent lighting has attracted the notice of cinematographers as well as philosophers. Some cinematographers have rejected illogical lighting shifts in favor of a more consistent approach. Taking note of this approach, the historian Fabrice Revault d’Allonnes has proposed a distinction between “classical” and “modern” lighting. Whereas a classical film like Ninotchka modulates its light to dramatize the changing moods of the story, a modern film represents light as dramatically “indifferent.”24 Pointing to examples from L’Avventura (Antonioni, 1960) to L’argent (Bresson, 1983), Revault d’Allonnes argues that many post-neorealist filmmakers strive to capture the nuances of light without seeking to correlate those nuances with the story’s moods. A character might commit suicide in broad daylight or find love in ominous shadows. One form of Revault d’Allonnes’s modern lighting is naturalism, abjuring the artificiality of Hollywood lighting in favor of an approach that is more faithful to the lighting of the real world. But modern lighting need not be naturalistic; it could just be flat, as if the cinematographer were making no effort to model the subject. The lighting is still expressive—but the idea being expressed is the idea that the light of the world is meaningless or obtuse.25 The “modern” cinematographer is rejecting the role of storyteller in favor of the more elemental role of picture-maker, using the tools of cinematography to draw our attention to the subtleties of light and shade, independent of story concerns. Whether working in the studio period or in the Hollywood of today, the classical cinematographer remains both, a picture-maker and a storyteller. Lenses and Camera Movement Lenses offer several variables to the cinematographer, such as aperture, focus, and focal length. By adjusting the aperture, the cinematographer allows more or less light into the camera. All other things being equal, the resulting images will be lighter or darker. Of course, all other things are rarely equal. In practice, the cinematographer must take several factors into account when selecting the f-stop (a measurement of aperture), including the illumination on the scene itself, the speed of the film stock, the shutter speed of the camera, and the amount of light lost in the lens itself. Unlike the amateur photographer who relies on the camera’s automatic exposure controls, the professional cinematographer judges the f-stop very carefully. An incident light meter may help the cinematographer measure how much illumination is falling on the scene, and a reflective light meter may help the cinematographer measure how much illumination is reflecting back into the camera, but the cinematographer must

 Fabrice Revault d’Allonnes, La Lumière au cinéma (Paris: Cahiers du cinéma, 1991), 9.  Revault d’Allonnes, La Lumière au cinéma, 11.

24 25



i­nterpret each one of these measurements, treating them as pieces of information to be weighed rather than as requirements to be followed. The choice of f-stop inevitably alters the resulting picture’s depth of field. By photographing a scene with a wide-open aperture (say, f/2), the cinematographer produces a shallow-focus image. If the middle-ground is in focus, then the foreground and background will look soft. By photographing a scene with a closed-down aperture (say, f/16), the cinematographer produces a deep-­ focus image, with sharp focus on the foreground, middle-ground, and background. Famously, the cinematographer Gregg Toland photographed several classic films in deep-focus, including Citizen Kane (Orson Welles, 1941) and The Best Years of Our Lives (William Wyler, 1946). Toland used powerful arc lamps to produce intense illumination on each scene, while employing coated lenses to make sure that very little light was lost in the lens itself. These technologies allowed Toland to photograph several scenes using the necessary narrow aperture settings.26 Still, these techniques were not always sufficient. There are several “deep-focus” shots in the film that were, in fact, produced with the aid of special effects, such as rear-screen projection or compositing on an optical printer.27 For instance, when Charles Foster Kane discovers that his wife Susan has poisoned herself, the “deep-focus” image is in fact a composite; the foreground, with its glass of poison looming ominously in the frame, was shot separately with the aid of a matte. The French film critic André Bazin wrote that “depth of field creates a relationship between the viewer and the image which is closer to the viewer’s relationship to reality.”28 Several subsequent philosophers of film have endorsed Bazin’s view, such as Gregory Currie, who argues that the long-take, deep-­ focus style “enhances our ability to detect spatial and temporal properties of the fiction by using the capability we have to detect those properties of things in the real world.”29 In this way, Currie situates Bazinian realism within the context of the “recognition” theory of depiction that he favors elsewhere.30 By contrast, George M. Wilson thinks of the Bazinian aesthetic as a kind of epistemic realism. He writes, “the ways in which movie viewers (fictionally) come to know about the world of the story are, so to speak, realistically presented and articulated.”31 Wilson distinguishes epistemic realism from the perceptual realism associated with photographic transparency. 26  Patrick Ogle, “Technological and Aesthetic Influences on the Development of Deep-Focus Cinematography in the United States,” in Movies and Methods, vol. 2, ed. Bill Nichols (Berkeley: University of California Press, 1985), 58–83. 27  On Toland’s collaboration with other contributors, including special-effects expert Linwood Dunn, see the chapter on cinematography in Robert Carringer, The Making of Citizen Kane, rev. ed. (Berkeley: University of California Press, 1996). 28  André Bazin, “The Evolution of Film Language,” in What Is Cinema?, trans. Timothy Barnard (Montreal: Caboose, 2009): 101. 29  Currie, Image and Mind, 107. 30  Currie, Image and Mind, 80. 31  George M. Wilson, “Imagined Seeing and Some Varieties of Cinematic Realism,” in Current Controversies in Philosophy of Film, ed. Katherine Thomson-Jones (New York: Routledge, 2016), 61.



For my purposes, it is important to note that deep-focus cinematography is a technical achievement—the result of creative choices made by the cinematographer. Indeed, the fact that many of Kane’s deep-focus techniques were produced with the aid of special effects provides further support for the idea that there is considerable continuity between the cinematographic techniques of the twentieth century and the hybrid techniques (mixing together photography and CGI) of the present day. Like Roger Deakins, Gregg Toland was above all a picture-maker, happy to use any available tool, from the arc lamp to the optical printer, to produce meaningful pictures. The creative input of the cinematographer is even clearer in the case of “shallow-focus” images. Here, the cinematographer must work with the director to choose which part of the composition will be in focus. For instance, in Baz Luhrmann’s Moulin Rouge! (2001, photographed by Donald McAlpine), one shot shows Christian (Ewan McGregor) walking away from the stage, while Satine (Nicole Kidman) sings a love song to win him back. Christian is in sharp focus, just left of center. The distant stage, featuring a purple and magenta set that looks like a heart, is centered but soft. The resulting composition pulls our attention in two directions at once—toward the heart-shaped set, with its bright saturated colors and its central position, and toward Christian himself, the only focused figure in the frame. Close the aperture just a touch, and you will throw the entire composition out of balance: The heart-shaped set will look a bit sharper, and we will lose sight of Christian in the shadows. Lenses also introduce the variable of focal length. A cinematographer may choose to photograph a scene with a wide-angle (short focal-length) lens or with a telephoto (long focal-length) lens, producing strikingly different results. Many philosophers have used the example of focal length to make larger points about the philosophy of film. For instance, Catharine Abell cites the use of a wide-angle lens to support her argument that a film can express a thought about a fictional character, while Berys Gaut argues that the distortions associated with different kinds of lenses present a problem for the idea that we imagine seeing the fictional world. (“Are we to imagine,” he asks, “that we have eyes that can change their focal length so that they can mimic the effects of such lenses?”)32 Given its significance to these larger debates, it is important to be precise about how focal length produces its visual effects. One way to characterize the difference would be to say that the wide-angle lens deepens or stretches the space, while the telephoto lens makes the space flatter or more compressed. Another way to characterize the difference would be to say that the telephoto lens magnifies the center of the image, while the wide-angle de-­ magnifies it. I prefer the second way of speaking because I think that the idea

 Abell, “Cinema as a Representational Art,” 283–284; Berys Gaut, A Philosophy of Cinematic Art (New York: Cambridge University Press, 2010), 205. 32



of “stretching” or “compressing” space is misleading.33 Suppose you have a wide-angle lens on the camera. You ask two actors—call them “Humphrey” and “Ingrid”—to stand in front of the camera. Humphrey is standing two feet away from the camera, in a tight framing from the shoulders up. Ingrid is standing four feet away, in a slightly looser framing from the waist up. Because Humphrey is two times closer (two feet away as opposed to four feet away), the image of his face will occupy more screen space. Indeed, to borrow a term from the philosophy of depiction, his “occlusion size” will be four times greater— twice the height and twice the width.34 In general, the viewer of the picture will not see Humphrey as abnormally large. As long as other depth cues are present, the viewer should see Humphrey as normal-sized but closer. Now replace the wide-angle lens with a telephoto lens. (If you are using a zoom lens, you may simply zoom in. A zoom lens is a variable focal-length lens, capable of shifting from short to long and back again.) In contrast to the wide-­ angle lens, the telephoto lens secures a narrower angle of view. It picks out a smaller slice of space and magnifies it. The angle of view is so narrow that now you cannot really see the faces of Humphrey and Ingrid at all; perhaps you see a big close-up of Ingrid’s eyeball, four feet away. Assuming that you would still like to have a good view of both actors’ faces, you must ask the actors to move farther away. Let us say that you place Humphrey 100 feet away, and you place Ingrid 102 feet away. Humphrey is now framed from the shoulders up, and so is Ingrid. Their occlusion sizes are almost identical. Why? The absolute difference in distance remains the same. Humphrey is still two feet closer. But the relative difference in distance is now much smaller. Previously, Humphrey was two times closer—and so his image was four times bigger. Now, Humphrey is just two percent closer—and so the difference in occlusion size is very small. The similarity in relative size may account for the sense that the telephoto image looks “flatter” than the wide-angle image. But doesn’t a zoom-in (or zoom-out) distort the space? There is at least one plausible way to say that the answer is “No.” Put a wide-angle lens back on the camera. Keep the actors far away: Humphrey at 100 feet away, Ingrid at 102. The actors’ relative distances will be roughly the same (approximately a two percent difference), and so again their occlusion sizes will be roughly the same. Indeed, go ahead and take a photograph from this location, with the wide-­angle lens on the camera. And go ahead and take another photograph from this location, with the telephoto lens on the camera. Now “blow up” the wide-­angle shot, magnifying it in a darkroom or on a computer. The resulting image might look grainy, but the blown-up wide-angle shot will look identical—spatially—to the unmodified telephoto shot. In both shots, you will see a ­ shoulders-­ up framing of 33  The following discussion of lenses draws on Bruce Block, The Visual Story: Creating the Visual Structure of Film, TV, and Digital Media, 2nd ed. (Burlington, MA: The Focal Press, 2007). The appendix offers a particularly clear account of Block’s view. If you find Block’s line drawings unconvincing, try photographing some of his arrangements for yourself. 34  I borrow the term from John Hyman, The Objective Eye: Color, Form, and Reality in the Theory of Art (Chicago: University of Chicago Press, 2006), 98.



Humphrey and Ingrid, with similar relative occlusion sizes. Both shots will look equally “flat.” All that the telephoto lens has done is to magnify the center of the image without moving any closer. What about the celebrated shot in Alfred Hitchcock’s Vertigo, where the filmmakers have combined a dolly-in with a zoom-out to produce the image of a stairway “stretching” out in space, as if space were getting deeper? Doesn’t the example prove that the wide-angle lens stretches space—and that the telephoto lens compresses space? Not necessarily. Notice that there are two variables here: focal length and relative distance. Zooming out allows the camera to move closer to the foreground objects. Moving closer to the foreground objects changes the relative distances. Changing the relative distances changes the relative occlusion sizes. Changing the relative occlusion sizes (as well as some other altered depth cues, such as textural diffusion) makes it appear that space is getting deeper. Another oft-noted effect of the wide-angle lens is its tendency to distort a person’s features in a close-up. However, this example supports my point that we always need to take relative distance into account when we seek to make sense of focal length’s effects. Suppose you want to take a close-up of Humphrey. If there is a wide-angle lens on the camera, you will need to ask Humphrey to stand very close to the camera—perhaps just a few inches away. Now Humphrey’s nose will be twice as close as Humphrey’s ears. So Humphrey’s nose will appear unusually large (in terms of its occlusion size), and his ears will appear unusually small. Significantly, Ingrid, standing in the background of the shot, will not look distorted, even though she, too, is photographed with a wide-angle lens. If you were to place a telephoto lens on the camera, Humphrey’s proportions would remain the same. The only difference is that you wouldn’t be able to notice the proportions because you would be looking at a blurry, unrecognizable patch of skin on Humphrey’s face. The defining feature of the wide-angle lens is, quite literally, its wide angle of view. The angle of view is wide enough that an actor can approach the camera and remain comfortably within the limits of the frame. When the actor is that close to the lens, small differences in absolute distance may produce large differences in relative distance, producing large differences in relative occlusion size. The telephoto lens, with its narrow angle of view, requires you to position the actors farther away from the camera, with consequences for relative distance and corresponding consequences for relative occlusion size. None of this means that the cinematographer cannot use focal length in a meaningful way. Quite the contrary: Filmmakers have long used wide-angle lenses, mid-focal-length lenses, and telephoto lenses for expressive purposes, such as characterization, mood, and spatial clarity (or spatial obscurity). For instance, in Julie Taymor’s 2002 film Frida (photographed by Rodrigo Prieto), one shot represents the protagonist’s point of view as she lies in a hospital bed and receives an injection. By photographing the shot with a wide-angle lens, the filmmakers have created a composition with exaggerated occlusion sizes. The syringe looks close to the camera and huge onscreen, but the ceiling panels



look distant from the camera and small onscreen. The resulting shot represents what Frida (Salma Hayak) sees, while expressing how she sees the world around her—as a vast, empty space, dominated by the syringe and the pain it represents. The last cinematographic tool that I will discuss is camera movement. By panning or tilting the camera, the filmmaker changes the camera’s angle of view horizontally or vertically. By dollying or craning, the filmmaker physically moves the camera from one position in space to another. It is quite common for filmmakers to combine camera movements, as when the operator pans to reframe in the middle of a dolly shot. All of these movements can be mimicked in digital cinema, as in the “virtual” shots swooping through various neighborhoods as Judy (the rabbit voiced by Ginnifer Goodwin) takes the train to the city in Zootopia (2016, directed by Byron Howard, Rich Moore, and Jared Bush). Recently, Katherine Thomson-Jones has situated the moving camera within ongoing debates about the relationship between cinematic seeing and cinematic narration. Specifically, she argues that it is sometimes appropriate to imagine moving in response to a moving-camera shot in a fiction film, but she insists, plausibly, that this example of imagined seeing does not require a cinematic narrator.35 Noël Carroll rejects the imagined seeing model altogether, instead situating the moving camera within his account of cinematic narration, whereby filmmakers seek to maximize audience engagement by controlling the spectator’s attention. One of the most powerful tools of attention management is variable framing, which is “fundamentally a matter of changing the position of the camera on the action,” whether through editing or camera movement.36 For instance, in the climactic scene of Jurassic Park (Steven Spielberg, 1993, photographed by Dean Cundey), the youngsters Lex and Tim (Ariana Richards and Joseph Mazello) are hiding in a kitchen as they attempt to elude two velociraptors. In one shot, Tim is cowering in the left foreground, while Lex crawls by in the background. After Lex exits the frame, the camera dollies in to a close-up of Tim. By dollying in, the film mandates that we look at the terrified boy. “Mandate” is a strong word, but it seems appropriate here. To borrow Carroll’s terms, the framing controls our attention through a combination of bracketing, indexing, and scaling.37 Here, the moving frame has bracketed off everything else we might see, such as Lex or the dinosaurs, thereby giving Tim unchallenged priority. Meanwhile, the camera has moved toward Tim, pointing at him as if the camera were an index finger. The movement of the camera also serves to change the scale of the image, making Tim even larger onscreen. Variable framing often serves the function of clarity, directing our attention to the most informative story point. Spielberg and his collaborators remind us 35  Katherine Thomson-Jones, “Narration in Motion,” British Journal of Aesthetics 52, no. 1 (January 2012): 34. The article is a response to George M. Wilson’s defense of the cinematic narrator idea. See “Le Grand Imagier Steps Out: The Primitive Basis of Film Narration,” in Seeing Fictions in Film: The Epistemology of Movies (New York: Oxford University Press, 2011), 29–51. 36  Carroll, The Philosophy of Motion Pictures, 125. 37  Carroll, The Philosophy of Motion Pictures, 125.



that Tim is paralyzed with fear. Movement also denies us access to information we would very much like to have. We would like to know if the dinosaurs are twenty feet away or three feet away. We would like to know if Lex, who just escaped an attack, is now moving into a safer position. By dollying in to such a close view of Tim, the film increases suspense by keeping crucial information off-screen, thereby asking us to fear the worst. The dinosaurs do not appear in this particular shot, but Carroll’s bracketing-­ indexing-­scaling model could apply to any type of shot, whether photographic or computer-generated. At the end of the same kitchen scene, the camera appears to dolly forward into a tight close-up of a velociraptor’s (computer-­ generated) face. As in a purely photographic shot, the movement brackets off extraneous details and indexes the velociraptor’s face, while making the face large enough that we can read the dinosaur’s crafty, malevolent expression. From the standpoint of storytelling, the function of the movement remains the same, whether the shot in question is photographic or digital.

The Question of Authorship Traditional auteur theory excluded most cinematographers from the realm of film authorship. At Movie, a British journal of film criticism that did much to develop the auteur approach in the 1960s, Ian Cameron explained, “The assumption which underlies all the writing in Movie is that the director is the author of a film, the person who gives it any distinctive quality it may have.”38 Of course Cameron admitted that there were exceptions, but the auteurists’ urge to defend cinema’s power as a medium of artistic expression encouraged them to prioritize examples where one individual was clearly in control. Most cinematographers simply were not candidates for single-auteur status. Some even took pride in their willingness to serve the director. According to cinematographer William Clothier, “It’s stupid to fight with directors. If John Ford told me to put the camera upside-down, I’d put the camera upside-down. It’s the director’s prerogative.”39 Since then, a few cinematographers have made more ambitious claims for the craft. Perhaps the most admired cinematographer of the past fifty years has been Vittorio Storaro, who won three Best Cinematography Academy Awards while working for three different directors: Apocalypse Now (1979, directed by Francis Ford Coppola), Reds (1981, directed by Warren Beatty), and The Last Emperor (1987, directed by Bernardo Bertolucci). Pointing out that “cinematography” means “writing with light,” Storaro told one interviewer, “Cinematographers are authors of photography, not directors of photography. We are not merely using technology to tell someone else’s thought, because we

 Ian Cameron, “Films, Directors, and Critics,” Movie 2 (September 1962): 13.  Clothier, quoted in Scott Eyman, Five American Cinematographers (Metuchen, NJ: Scarecrow Press, 1987), 126. 38 39



are also using our own emotion, our culture, and our inner being.”40 In practice, Storaro used a wide range of techniques, such as carefully controlled color schemes, heavily filtered lenses, and the specialized handling of film stocks, to create meaningful patterns, unique to each film. Recently, several philosophers have defended the view that films may be collaboratively authored, a view that is expansive enough to include the authorial contributions of cinematographers. For instance, Berys Gaut has offered an eloquent defense of the multiple-authorship view. Within this broad framework, Gaut proposes two useful distinctions. The first, he writes, “concerns the degree to which creative power in determining the artistic properties of a film is centralized or dispersed.”41 In some films, a single powerful figure, such as a writer-director, might make almost every important decision. But there are many films where tasks are distributed according to a clear division of labor: The writer crafts the script, the director works with the actors, and the cinematographer designs the lighting and compositions. “The second dimension of variation,” Gaut continues, “concerns the degree to which the different collaborators are in agreement over the aims of the film and their role within its production.”42 Even in a film with dispersed responsibilities, the filmmakers may all work together to produce a coherent film—or they might work against each other to produce a work filled with contradictory meanings. As a historian of Hollywood cinema, I think that Gaut is correct to say that most mainstream films are collaboratively authored. The fact that Gaut allows for a wide range of collaborative possibilities—from the hierarchical to the egalitarian, from the harmonious to the conflicted—adds to his model’s appeal. Elsewhere, I have studied how cinematographers of the Hollywood studio system faced several competing demands. Their professional organization—the American Society of Cinematographers—encouraged cinematographers to think of themselves as artists, modulating the lighting to suit the changing moods of each story. But the studios pushed cinematographers to think of themselves as portraitists, responsible for making the stars look good at all times, regardless of story concerns.43 As we have already seen, cinematographer William Daniels altered the lighting from shot to shot when photographing Ninotchka. We might interpret these modulations as storytelling choices, employing gentle contrasts for a moment of optimism and harsh contrasts for a moment of despair. Alternatively, we might interpret these modulations as an attempt to balance the demands of the studio with the demands of the story and its setting, prioritizing glamour in the close shots while prioritizing the plausible representation of a hanging lamp in the wide shots. The two ­interpretations are not mutually exclusive. In fact, it is a mark of Daniels’s achievement that he has accomplished multiple goals with a disarmingly simple shift in contrast. 40   Storaro, quoted in Benjamin Bergery, “Reflections 10: Storaro, ASC,” American Cinematographer 70, no. 8 (August 1989): 70. 41  Gaut, A Philosophy of Cinematic Art, 128. 42  Gaut, A Philosophy of Cinematic Art, 128. 43  Patrick Keating, Hollywood Lighting from the Silent Era to Film Noir (New York: Columbia University Press, 2010), 161.



To be sure, even Gaut’s expansive view of film authorship does not automatically lead to the conclusion that the cinematographer is an author. As several scholars have pointed out, most film crews include caterers, but we usually do not think of caterers as members of a film’s authorial team.44 Summarizing Gaut’s view, Katherine Thomson-Jones offers a good way to make the relevant distinction: “Film interpretation is guided by signs of creative intelligence in the work itself. […] These signs do not need to be signs of a single author; they can be signs of multiple authors who we may imagine as coordinating their creative activities more or less successfully in a more or less unified film.”45 The specific decisions of the caterer usually do not leave signs of creative intelligence in the film; the specific decisions of the cinematographer usually do. Working with the camera crew, the electrical crew, and the grip crew, the cinematographer shapes the look of the film by manipulating on-screen details of composition, lighting, and camerawork. For instance, consider cinematographer Mandy Walker’s contribution to Hidden Figures (2016). In consultation with director Theodore Melfi, production designer Wynn Thomas, and others, Walker eschewed digital capture and photographed the movie on film stock, to better evoke a “Kodachrome” look that evoked the film’s 1960s setting. Most of the film was photographed on 35 mm stock, but early scenes of a protagonist’s childhood were photographed on 16 mm, evoking an even earlier time period by introducing visible grain. Each primary location featured a distinctive color palette, established through costume design, set design, and lighting, all subject to post-production manipulations via a digital intermediate. Houses were depicted in warm colors, in contrast to the desaturated look characterizing the computing rooms.46 The filmmakers also worked out a visual progression through their use of camera angle. Walker explains, “When we were shooting the African American women talking to the white guys, we had the camera just a little bit under their eye-­ line, so they were always looking above them. As the film develops, the camera starts to be on the same level for both characters.”47 In this way, Walker and her collaborators expressed the film’s primary theme concerning the struggle for equality. All of these choices serve as signs of a creative intelligence shaping the film’s visual style in meaningful ways.

44  C. Paul Sellors, “Collective Authorship in Film,” The Journal of Aesthetics and Art Criticism 65, no. 3 (Summer 2007): 269. See also Sondra Bacharach and Deborah Tollefsen, “We Did It: From Mere Contributors to Coauthors,” The Journal of Aesthetics and Art Criticism 68, no. 1 (Winter 2010): 27–28. 45  Katherine Thomson-Jones, Aesthetics & Film (New York: Continuum, 2008), 52. 46  Carolyn Giardina, “How Hidden Figures Got its 1960s ‘Kodachrome’ Look,” The Hollywood Reporter, January 13, 2017, 47  Walker, quoted in Agatha French, “Hidden Figures: An Interview with Cinematographer Mandy Walker,” Cinema Thread,



The increasing digitization of cinema has changed the debate over cinematographic authorship in surprising ways. On the one hand, certain digital tools threaten the cinematographer’s authority. The digital intermediate gave color graders considerable power to change the images that the cinematographers had worked so hard to produce, while computer-generated sequences had the power to exclude cinematographers from the task of image creation altogether. As historian Christopher Lucas explains, the cinematographers of the last two decades “have felt compelled to assert their relevance to the production process as the malleability of digital cinematic images opened the field to more ‘collaborators.’”48 On the other hand, some cinematographers have welcomed the new digital tools as extensions of their traditional powers. Roger Deakins, who championed the use of the digital intermediate, later served as a “visual consultant” for digitally animated films, such as Gore Verbinski’s Rango (2011).49 Even films that employ no cinematographer at all may rely on a conception of the craft that has been around for a century or so. From this point of view, digital cinema can be seen as the fulfillment of the craft’s long-term goal of producing meaningful pictures that express the shifting moods of each story, down to the smallest details.

Conclusion It is possible that digitization will continue to alter or even eliminate the position of cinematographer, who might come to serve as a consultant on the design of looks executed by others. But even those films that are made entirely on computers will still owe a considerable debt to the tradition of cinematography that I have sketched here—a tradition of picture-making, storytelling, and authorship.

Bibliography Abell, Catharine. 2010. Cinema as a Representational Art. British Journal of Aesthetics 50 (3, July): 273–286. Atencia-Linares, Paloma. 2012. Fiction, Nonfiction, and Deceptive Photographic Representation. The Journal of Aesthetics and Art Criticism 70 (1, Winter): 19–30. Bacharach, Sondra, and Deborah Tollefsen. 2010. We Did It: From Mere Contributors to Coauthors. The Journal of Aesthetics and Art Criticism 68 (1, Winter): 23–32. Bazin, André. 2009. The Evolution of Film Language. In What Is Cinema? Trans. Timothy Barnard, 87–106. Montreal: Caboose. Bergery, Benjamin. 1989. Reflections 10: Storaro, ASC. American Cinematographer 70 (8, August): 70–74. Block, Bruce. 2007. The Visual Story: Creating the Visual Structure of Film, TV, and Digital Media. 2nd ed. Burlington: The Focal Press. Cameron, Ian. 1962. Films, Directors, and Critics. Movie 2 (September): 12–16.  Lucas, “The Modern Entertainment Marketplace,” 153.  Lucas, “The Modern Entertainment Marketplace,” 155.

48 49



Carringer, Robert. 1996. The Making of Citizen Kane. rev. ed. Berkeley: University of California Press. Carroll, Noël. 2008. The Philosophy of Motion Pictures. Malden: Blackwell Publishing. Currie, Gregory. 1995. Image and Mind: Film, Philosophy, and Cognitive Science. New York: Cambridge University Press. Deakins, Roger. 2008. The DI, Luddites, and Other Musings. American Cinematographer 89 (7, October): 78–83. Eyman, Scott. 1987. Five American Cinematographers. Metuchen: Scarecrow Press. Fisher, Bob. 1992. The Whys and Hows of JFK. American Cinematographer 73 (2, February): 42–52. French, Agatha. Hidden Figures: An Interview with Cinematographer Mandy Walker. Cinema Thread. Gaut, Berys. 2010. A Philosophy of Cinematic Art. New York: Cambridge University Press. Gentry, Ric. 1997. Painterly Touches. American Cinematographer 78 (1, January): 50–57. Giardina, Carolyn. 2017. How Hidden Figures Got Its 1960s ‘Kodachrome’ Look. The Hollywood Reporter, January 13. Harrell, Al. 1992. Malcolm X: One Man’s Legacy, to the Letter. American Cinematographer 73 (11, November): 28–34. Hopkins, Robert. 2008. What Do We See in Films? The Journal of Aesthetics and Art Criticism 66 (2, Spring): 149–159. Hyman, John. 2006. The Objective Eye: Color, Form, and Reality in the Theory of Art. Chicago: University of Chicago Press. Keating, Patrick. 2010. Hollywood Lighting from the Silent Era to Film Noir. New York: Columbia University Press. ———. 2017. What Does It Mean to Say that Cinematography Is Like Painting with Light? In Transnational Cinematography Studies, ed. Lindsay Coleman, Daisuke Miyao, and Roberto Schaefer, 97–115. Lanham: Lexington Books. Lightman, Herb. 1973. The New Panaflex Camera Makes Its Production Debut. American Cinematographer 54 (5, May): 564–567, 598–599, 611–620. Lucas, Christopher. 2014. The Modern Entertainment Marketplace, 2000-Present. In Cinematography, ed. Patrick Keating, 132–157. New Brunswick: Rutgers University Press. Milner, Victor. 1933. ‘Miscasting’ the Cinematographer. American Cinematographer 13 (13, February): 36–37. Ogle, Patrick. 1985. Technological and Aesthetic Influences on the Development of Deep-Focus Cinematography in the United States. In Movies and Methods, ed. Bill Nichols, vol. 2, 58–83. Berkeley: University of California Press. Predal, René. 1973. Les Grands Operateurs (IV). Cinéma 73 173 (February): 87–95. Revault d’Allonnes, Fabrice. 1991. La lumière au cinéma. Paris: Cahiers du cinéma. Salt, Barry. 1992. Film Style and Technology: History and Analysis. 2nd ed. London: Starword, p. 253. Scruton, Roger. 1983. Photography and Representation. In The Aesthetic Understanding, 102–126. London: Methuen. Sellors, C. Paul. 2007. Collective Authorship in Film. The Journal of Aesthetics and Art Criticism 65 (3, Summer): 263–171.



Silberg, Jon, and Steven Pizzello. 2009. Cinematographers, Colorists, and the DI. American Cinematographer 90 (6, June): 78–82. Slocombe, Douglas. 1949. The Work of Gregg Toland. Sequence 8 (Summer): 69–76. Smith, Murray. 2011. On the Twofoldness of Character. New Literary History 42 (2, Spring): 277–294. Thomson-Jones, Katherine. 2008. Aesthetics & Film. New York: Continuum. ———. 2012. Narration in Motion. British Journal of Aesthetics 52 (1, January): 33–43. Walton, Kendall. 1990. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge: Harvard University Press. Williams, David E. 1995. The Sins of a Serial Killer. American Cinematographer 76 (10, October): 34–42. Wilson, George M. 2011. Le Grand Imagier Steps Out: The Primitive Basis of Film Narration. In Seeing Fictions in Film: The Epistemology of Movies, 29–51. New York: Oxford University Press. ———. 2016. Imagined Seeing and Some Varieties of Cinematic Realism. In Current Controversies in Philosophy of Film, ed. Katherine Thomson-Jones, 57–75. New York: Routledge. Zone, Ray. 2000. The Literature of Light: An Interview with Vittorio Storaro, ASC, AIC. In Writer of Light: The Cinematography of Vittorio Storaro, ASC, AIC, ed. Ray Zone, 79–92. Hollywood: ASC Press. Zsigmond, Vilmos. 1980. Behind the Cameras on Heaven’s Gate. American Cinematographer 61 (11, November): 1110–1113, 1164–1165, 1172–1181.


The Structure of Film and Motion Pictures


Silly Questions and Arguments for the Implicit, Cinematic Narrator Angela Curran

Chapter Overview Fiction films tell a story, with images and sound, and in doing so invite the audience to imagine that specific events have happened, as they are reported and shown in the image and soundtrack of the movie (Carroll 1990, 1996, 2006; Currie 1990; Walton 1990: 39; Lamarque and Olsen 1994).1 But how more precisely does movie narration work on the viewer so that she comprehends the story? Here there is a tremendous and fascinating debate regarding the nature of cinematic narration, or how a movie conveys the story events to the audience. This chapter concerns these disagreements. The focus is on the view that there are ubiquitous, implicit narrators in fiction films. Such a narrator is the agent that is tacitly understood to be carrying out a showing of the story events to the audience from the world of the film fiction. My chapter aims to advance the debate on a problem often raised by philosophers who are skeptical of implied narrators in movies. This is the concern that positing such elusive narrators gives rise to absurd imaginings (Gaut 2004: 242; Carroll 2006: 179–180, 2016). The worry arises because critics maintain that the “Realistic Heuristic” governs our imaginings about fiction. The Realistic Heuristic involves the claim that when we engage with a work of fiction, we “fill in” and draw 1  For an influential discussion of make-believe and the mimetic arts, see Kendall Walton (1990). For recent discussions of fiction and imagination, see Matravers (2014) and Stock (2017). For an accessible overview of some key debates about fiction and imagination, see Stock (2013).

A. Curran (*) Kansas State University, Manhattan, KS, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




i­mplications from what is explicitly true in the fiction based on how things work in real life, unless it is explicitly stipulated to be otherwise. The problem, critics allege, is that when we “fill in” the implications of the implied cinematic narrator’s presence in the story world, absurd imaginings follow. For instance, how is it possible for an implied narrator to convey the story of the Battle of Dunkirk from the scene of the action without getting shot or having to duck bullets? George Wilson, a supporter of some kind of implicit, “narrating agency” in fiction films, maintains that the indeterminate nature of what is true in the story world means that questions about how the implied narrating agency is able to carry out its mission are “silly” ones to ask (Wilson 2011, 2013). The debate between friends and foes of the cinematic narrator has been at a stalemate most centrally because there seems to be no resolution as to whether the questions critics raise about the implied narrator in movies are legitimate ones to ask. In this chapter, I examine how the “absurd imaginings” problem arises for all the central arguments for the elusive cinematic narrator and discuss why the questions critics pose about this narrator are legitimate ones to ask. In Part I, I introduce some terminology relevant to understanding the debate about cinematic narrators. In Parts II, III, and IV, three central arguments—The Narration Implies a Narrator Argument, The Ontological Gap Argument, and the Imagined Seeing Thesis—are considered and assessed. In Part V, we focus on the arguments for and against the claim that positing the implied narrator in movies gives rise to absurd imaginings. In my concluding comments, I briefly discuss directions of research that further inquires into cinematic narration might take. Part I: Narration and Narrators We should clarify some terminology. Fiction films convey a story, which is about something, what we call its fictional content. The story is concerned with giving an account of imagined characters and situations. A film has a plot, an underlying sequence of events as they occur in the story, and narration, the telling or relating of these fictional events to the audience.2 Cinematic narration is the way in which the film tells a story.3 Some of the ways in which we talk about narration in cinema has its origins in literary theory.4 Someone creates or makes a work of literary fiction: this is the actual flesh-and-blood author. The author is something external to the film, its cause or creator. So, for example, Conan Doyle is the author of the Sherlock Holmes stories. An author also sometimes creates a narrator, an internal component of the work that is the fictional voice that recounts the 2  See Wilson 2013 and Livingston 2005 for a survey of some of the points in contention about narrative, in general. For skepticism about the usefulness of talking about cinematic narration, see Pye (2013: 136). 3  For an introduction to the basic principles of cinematic narration, see Bordwell (1985: 48–61) and Carroll (2008: 116–146). 4  See Gaut (2004) and Thomson-Jones (2007).



happenings and situations that take place in the story. For example, in the Sherlock Holmes stories, it is Sherlock’s trusty sidekick, Doctor Watson, who is a character in and the narrator of the Holmes stories. Similarly, someone creates a film: this is the flesh-and-blood filmmaker, the actual person who is the cause or creator of a film. Where it is reasonable to think that one individual exercises the most significant control over the movie, we can speak of the filmmaker as the counterpart of the author (Livingston 1997). Alternatively, when it makes sense to think of the movie as the creative product of a group of individuals, such as the director, the screenwriter, the cinematographer, and so on, we can say the movie is the collaborative project of multiple filmmakers (Gaut 2010: 128–132). When literary theorists talk about how works of fiction convey points of view on the events in the story, they often use the concept of the “implied author.” This is a hypothetical construct whose viewpoint on what happens in the story world makes itself clear in the text (Booth 1961: 70–71; Nehamas 1981). Likewise, some use the term “implied filmmaker” for the hypothetical agent who is responsible for the sensibility and attitudes manifest in the film’s narration.5 We said that a narrator is a fictional character that recounts the goings on in the story. Some novels have explicit character-narrators, such as Doctor Watson in the Sherlock Holmes novels or the character of Esch, who tells her story in the first-person in Jesmyn Ward’s Salvage the Bones. In these novels, it is fictional that the characters are telling the story. However, some hold that in every literary fiction, there are implicit narrators—fictional beings who recount the story events as real to the reader. These narrators are implicit, not explicitly introduced, and they have no interaction with the other fictional characters in the story. The actual author cannot tell the story because she does not believe the events in her story happened. Instead, there must be a narrator who is part of the story world and who believes the characters and events exist and reports them as fact to the reader. The debate over cinematic narration concerns whether we should make the same move and say that there are implicit narrators in movies by whose actions we come to know about the depicted events in the world of the film. One central point of contention is whether the same reasons that some say there are implicit narrators in literary fictions carry over to support the claim that movies standardly have implicit narrators as well. A second is how best to describe the specific imaginative experience of the audience who watches a movie. We can understand this point of contention as a question about what the audience at the movies is “mandated” to imagine. 5  See Wilson’s discussion of the implied filmmaker of Letter from an Unknown Woman (1948) in Wilson (1986: 134–9). Some, such as David Bordwell, reject the notion of an “implied filmmaker” and, instead, prefers to talk about how the “narration itself ” cues the viewer to be surprised, sympathetic, and so on (Bordwell 1985: 62). Greg Currie (1995a, b) uses the concept of an “implied filmmaker” to address unreliable narration in fiction films.



The idea is that a fiction film “mandates” or requires that the viewer imagine various things as part of their correct comprehension and appreciation of the movie’s narrative. The second point of contention is then: are viewers at the movies “mandated” to imagine just the fictional contents of the story? Alternatively, are they also required to imagine how it is that they come to learn of the story events? If so, is it standard for viewers to learn about the world of the story through an implicit, fictional narrator? If the answer to this last question is “yes,” then we say that the implicit narrator “mediates” our access to the story events and the narrator presents those events to us “indirectly” (Walton 1990: 357). Now, to illustrate, some fiction films mediate our access to the story by using characters from the story that the film explicitly introduces as the tellers of the tale, as happens in Shawshank Redemption (Drabont, 1994) or Murder My Sweet (Dmytryk, 1944). Films also tell the story by using omniscient narrators such as the voice-over narrator in The Royal Tenenbaums (Anderson, 2001). In this case, the narrator belongs to the fictional world and reports the events as if they happened, but is not involved in any of the story events. But what does the audience imagine when they watch a fiction film where the story is not told either by a character-narrator or by an omniscient, third-­ person narrator, as is the case with The Wizard of Oz? Some claim that every fiction film has an implicit, fictional narrator who is responsible for conveying the story, as a whole, to the audience. Call this the “Ubiquity Thesis” (see Kania 2005: 47). We now turn to examine arguments for this thesis. Part II: Narration Implies a Narrator Chatman’s Argument Why should we think that there are ubiquitous fictional narrators in movies? Seymour Chatman has argued for the implied cinematic narrator, simply by considering what is implicit in the concept of narration (Chatman 1990: 128; see also Levinson 1996: 252). Thus, Chatman’s argument is known as the Analytic or A Priori Argument.6 His argument is that the meaning of the concept of “narration” logically implies there must be a narrator. The Narration Implies a Narrator Argument: Stage One (Chatman 1990: 113–15) 6

1. Every narrative is an activity, the act of telling or showing a story. 2. Activities must have agents. 3. The agent of a narration is its narrator. 4. Therefore, necessarily, for every act of storytelling, there is a narrator.

 Gaut (2004: 235–236) calls it the former; Kania (2005: 47–48), calls it the latter.



Stage Two (Chatman 1990: 133–4) 5. Fictional films contain narratives. 6. Therefore, necessarily, for every fiction film there is a fictional narrator. Chatman responds to David Bordwell, who proposes that narration is a process or activity of selecting, arranging, and rendering story material (Bordwell 1985: xi). Chatman counters by saying that activities require agents; there is no doing without a doer (Stage One above). In the case of cinematic narration, the agent carries out the showing of the story to the audience. The narrator is not an actual human being (see Stage Two above), so this means the narrator cannot be the actual filmmaker who creates the film. Also, Chatman argues the narrator cannot be the implied filmmaker. Narration, the act of telling a story, involves communication between a sender and a receiver, whereas the implied filmmaker (hypothetically) invents the narrative, but does not communicate it to the audience, says Chatman (1990: 130). Narration implies someone or something that narrates: therefore, in literary works and fiction films, there is a narrator, distinct from the actual filmmaker, who uses the soundtrack and the series of edited photographic images to convey the story.  uestions About Chatman’s Argument Q Some question the claim that narration logically implies a narrator. For example, David Bordwell’s view is that every property attributed to a narrating agent can instead be ascribed to the film’s “narration itself” (Bordwell 1985: xi). Critics say in reply that this involves an inappropriate personification of the filming process, or it is a shorthand device for saying there is a narrating agency doing the narration, which does not get rid of an intentional agency doing the storytelling (Gaut 2010: 200; see also Currie 1995a, b: 247–9).7 So many are inclined to accept that narration implies a narrator. The central problem with Chatman’s argument is that it fails to establish that narration requires a fictional narrator. For even if we restrict the argument to fictional narratives, and we suppose that the claim that (a) there is a telling or narration of a fictional story, entails the claim that (b) there is someone who tells the story, it does not follow without some further argument that (c) there is fictional narrator or teller of the story. For it could be the author who is the one who tells the story.8

7  For the same reason, some who accept the idea that storytelling or narration is an intentional activity reject Kendall Walton’s suggestion that there could be a “naturally occurring” and nonintentionally produced story, for instance, cracks in wood that seem to tell a story, provided the audience standardly decided to use such things as “props” in their game of make-believe (Walton 1990: 52). 8  See Kania (2005: 48), Köppe and Stürhring (2011), Gaut (2004: 235–237), and Wilson (1997: 299–300).



Indeed, some maintain that given that narratives are artifacts that are made to communicate a story, the clear choice for the person who tells the story is the actual author or filmmaker. For the narrator is the person whose intentions have to be understood if the story is to be correctly appreciated and that would be the actual person who created the story (Currie 2010: 66). If this line of criticism is successful, the intentional nature of a film, as a work designed to communicate a story, leads back to the actual author as the storyteller, not toward an implied cinematic narrator. And the attempt to say that logic requires there are implicit fictional narrators in every film fails. Finally, when we think of explicit narrators, we normally suppose it is appropriate to ask things such as, how does the narrator know about these things? And, what is its point of view? (Currie 2010: 66). But there are no answers to these questions about an implicit narrator, who is not explicitly introduced as the agent giving us access to the story. Chatman tries to maintain that questions about how the narrator works are “non-questions” not in need of an answer (Chatman 1990: 130). But as we will see, such queries about the implicit narrator are not disposed of so easily. Part III: The Ontological Gap Argument  an the Actual Filmmaker Tell Her Fictional Story? C The argument we will discuss in this section aims to make up for the deficits in Chatman’s argument. It does not try to reason, a priori, from the concept of narration to the existence of a fictional narrator. Rather, the argument is that implicit cinematic narrators are needed to explain the nature of our engagement with fiction films. The argument, presented by Jerrold Levinson in a rich discussion of film music and narrative agency (and recently defended in Wilson 2011 and Matravers 2014), has come to be known as the “Ontological Gap Argument.”9 Ontological Gap Argument: 1. Reason demands an answer to the question of what makes possible our knowledge of the story events. 2. Only fictional beings can have access to events in the world of the fiction. 3. Therefore, only a fictional narrator can convey to us the knowledge of the events in the world of the fiction. 4. We do have knowledge of the story events in film fiction. 5. Therefore, there is a fictional being, an implicit, cinematic narrator who is responsible for conveying the knowledge of the events in the story.10  See Levinson (1996: 252–256); so-named by Andrew Kania in (2005).  Levinson describes the cinematic narrator as an agent who provides access to the story world. But in a note, he also endorses another role for the cinematic narrator: the cinematic narrator is also responsible for crafting the plot, the underlying sequences of events in the story (see Levinson 1996: 280, footnote 21). 9




Note that the conclusion of the argument, (5) above, applies to all fictional movies. Thus, if the Ontological Gap Argument is correct, the Ubiquity Thesis is established. Implicit fictional narrators are ubiquitous; they are standard in all fiction films. Such implied narrators are “the best default assumption available for how we make sense of narrative fiction film” (Levinson 1996: 252). But does the argument stand up? If this is the correct way to formulate Levinson’s argument, then several problems arise. One worry is that the argument is undermined by its assumptions (Kania 2005: 48–49). Premise (2) says that only fictional beings can have access to the world of fiction. But the cinematic narrator, Levinson supposes, is part of the story world. It follows that the audience cannot have access to the fictional narrator and its fictional narration since this occurs in the world of the story. So implied fictional narrators do not help with the worry about how we “access” the story world. Also, Noël Carroll raises a Platonic ThirdMan style objection. The Ontological Gap Argument maintains that we need a fictional being to access items in the fiction. Since the cinematic narrator is part of the story world (even if it does not interact with the other denizens in the world), then we are off on an infinite regress, and we will need another cinematic narrator to access the first cinematic narrator, and so on! (Carroll 2006: 179). Another point of contention is the claim that only a fictional being can show the story events to the audience (premise 3). It follows that the actual or implied filmmaker cannot show or narrate the goings on in the story. The filmmaker, either actual or implied, can show us images, for instance, shots of the sets on the lot of Universal Studio, which are filmed to represent Frankenstein’s castle. However, the filmmaker cannot show the audience the fictional goings on in Frankenstein’s castle, for they stand apart from it, in the outside (actual) world (Carroll 2016: 117). Carroll challenges the Ontological Gap Argument by testing our intuitions about scenes in which movie directors appear as themselves, a not uncommon practice in cinema (Carroll 2016: 121). Carroll discusses the case of Bergman’s Persona (1966), where toward the end, we see documentarystyle scenes of Bergman and the camera crew (Carroll 2016: 121). What are we supposed is going on in this scene? It is natural to suppose that the filmmaker, Bergman, is appearing in the story, not some fictional stand-in. For, if contrary to fact, Bergman was to have a heart attack in the scene, who would we say died? Carroll maintains that our intuition tells us that it would be the actual director, Bergman, not some fictional doppelganger, Carroll maintains. These and similar examples (for instance, Hitchcock making a cameo appearance in his films) give us reason to question the sharp dichotomy between the fictional world and the actual filmmaker on which the Ontological Gap Argument rests.



 efending the Ontological Gap Argument D In reply, Levinson is likely to say the above formulation of his argument misses his central point. Viewers at the movies imagine that they are receiving visual information from the story world. The actual filmmaker cannot convey visual information from the fictional world. Only a narrator operating from within the world of the fiction, for whom the events are “real and reportable” can give the viewer the sort of perceptual access that she imagines herself to have (Levinson 1996: 255). Indeed, Derek Matravers recently argues that Levinson’s point involves the “standard view” that to “imagine a story” involves something like, “to imagine it is a report of actual events” (Matravers 2014: 123). However, it is far from clear that this is the consensus view on what is involved when viewers imagine a story, by either reading a literary fiction or watching a fiction film. Critics insist that an important point is being begged in describing what the viewer imagines that she is seeing is a “report” of events, if this is taken to imply the viewer imagines she is seeing a visual recounting of actual events (Carroll 2016: 126). Instead, other explanations of how a movie works on the viewer so she understands what is going on or true in the story are available. Noël Carroll proposes the view, for example, that what is so in fiction is whatever the maker or makers of the fiction intended the audience to imagine (Carroll 2016: 122). Call this the Imagination Account of Fiction. If the actual filmmakers of An American Werewolf in London (John Landis, 1981), for example, mandate that the audience imagines that American college student, David Kessler, is bitten by a werewolf and turns into one, then it is true in the fiction that this is so. If Carroll is right, there is no need to posit a fictional narrator, reporting the story events as if they are real, to explain how a viewer understands what goes on in the story. Narrative comprehension comes about through the “fictive intent” of the work’s creator (Carroll 2006: 176). Derek Matravers, however, insists that the Imagination Account of Fiction is mistaken (Matravers 1997: Chapters 3 and 7). He firmly rejects the idea that fiction can be defined by the creator’s mandate to imagine the story contents. For fictions mandate that we imagine, as well as believe, various things are so in the story world. For example, a viewer of Nowhere in Africa (2002, Caroline Link) is mandated to imagine various things, such as that Walter and Jettel Redlick are a Jewish couple that is forced to relocate from Nazi Germany in 1938 to a farm in Kenya. However, the director intends that we also believe various things, for instance, that Kenya is in Africa. Is it possible for the Imagination Account of Fiction to account for the fact that there are truths in fiction we are mandated to believe as well as imagine? In a very recent book-length treatment of imagination and fiction, Kathleen Stock suggests that it can. She defends what she calls “extreme intentionalism,” the view that the fictional content of a work is what the author intended the reader to imagine (Stock 2017). In response to Matravers, she argues that the total content of a fiction is stored in the mind of the reader and then marked as “imagining.” When the author intends the reader to believe various things are



so in the story, the reader indexes these truths as beliefs. Thus, Stock responds that the Imagination Account can acknowledge that an author intends that some of a work’s contents are to be believed and not just imagined (Stock 2017: 168). Noël Carroll explains how this type of mental compartmentalization might work. The default assumption is that what is true in the fiction is what the filmmaker mandates that the viewer imagine. However, as we work to comprehend the story’s narrative, depending on the film, we might then “suspend” the mandate to imagine various things as so in favor of a mandate to believe these things instead (Carroll 2016: 124). For example, as we comprehend the story in Gone With the Wind (1939, Fleming), we suppose the author mandates us to imagine that certain things are so in the life of Scarlett O’Hara, but then come to understand that we need to believe various things (for instance, that Atlanta is in Georgia) (Carroll 2016: 124). We might say that in Carroll’s view, as we engage with a work of fiction, we go through a process of “reflective equilibrium” in which we measure hypotheses about what goes in the story against the evidence that is presented in the story figure out what we are mandated to imagine versus what we are mandated to believe. To be sure, more could be said about how this happens, as Carroll acknowledges. But in principle, we see how a response to Matraver’s objection to the Imagination Account of Fiction works. Carroll’s reply to Matravers also has implications for another point that often comes up in the discussion of the Ontological Gap Argument. For in defending the argument, Levinson seems to make use of what is known as the Assertion Argument.11 According to this view, movie narration works the way in which some think that narration in literary fiction works. Literary works employ declarative sentences to report the goings on of characters and events in the story, as does the first sentence of Harry Potter and the Chamber of Secrets “Not for the first time, an argument had broken out over breakfast at number four Privet Drive.” Declarative sentences make utterances or assertions, and it is natural, the argument goes, for the reader to imagine that where there is an assertion, there is an asserter, the implicit narrator of the story.12 George Wilson builds on the Assertion Argument to argue that we need implicit narrators in literature for the reader to understand what is true in the fiction versus what is just supposition (Wilson 2007: 82–83). If, to use a version of Wilson’s example, “Katie loves Hubble. Many people thought this was true. But was is so?” there is a question if we are supposed to take “Katie loves Hubble” as something we imagine is true in the story versus something we are just supposed to consider as a possibility. Wilson’s idea is that to figure out that “Katie loves Hubble” is true in the story, we need to determine whether we should imagine a fictional, implicit narrator is “asserting” that this is so.

 Carroll (2006: 197), Thomson-Jones (2009: 299); see also Matravers (2014: 123).  Walton (1990: 265), Matravers (2014: 122).

11 12



However, we can see how Carroll and others might reply that no such imagining of a fictional narrator is necessary. The reader faces a choice of taking “Katie loves Hubble” as true in the fiction versus something she is simply supposed to wonder about or entertain as a possibility. Her task then is to ­determine whether the author mandates her to imagine that “Katie loves Hubble” is true in the story or whether “Katie loves Hubble” is something she should just consider as a possibility in the story. To determine the author’s intentions, she asks which hypothesis makes the most sense of the narrative as a whole. No fictional narrator is needed. Wilson replies that while it is possible to figure out what is true in the story without imagining a narrator asserting it to be so, it is “extremely strained and artificial” to not have an implied narrator be the way one ascertains the truth in a fiction (Wilson 2011: 120).  onclusion: Where Do Things Then Stand with the Ontological Gap C Argument? To accept it, we must make several assumptions about viewer’s experience at the movies. First, viewers at the movies imagine they are receiving a report from inside the story world; second, that to comprehend what is true in the fiction, we need to imagine a fictional presenter asserting or reporting that things are so in the story world. Implicit narrator skeptics call both these assumptions into question. Instead, they propose the Imagination Account of Fiction: that we can comprehend what is true in the story by what the author mandates us to imagine. The Imagination Account faces some challenges, specifically the fact that sometimes we are mandated to not imagine things are so in the story world, but also believe them. This is a challenge that proponents of the Imagination Account acknowledge they have to meet. But they maintain that there is ample motivation to do so. There is simply no evidence that the “plain viewer” at the movies imagines herself to access the story events through the mediation of a fictional narrator (Carroll 2016: 126). As noted, in Part I, some suggest that the reasons for thinking there are implicit narrators in literature also support narrators in movies. However, novels convey a story with words, while the use of images is central to storytelling in films. In the next section, we look at George Wilson’s formidable argument, which takes into account the nature of cinematic narration as visual storytelling. Part IV: The Imagined Seeing Thesis I magined Seeing at the Movies Like Levinson, Wilson is interested in how audiences at the cinema say that, in some sense, they “see” or make perceptual contact with the fictional events and characters in the drama.13 Wilson acknowledges that the audience does not

13  Wilson (2011: 7). Wilson is inspired by a fascinating discussion of visualization at the theater and at the movies in Williams (1973).



literally see the characters in the narrative story, yet he wants to take seriously their talk of “seeing” the characters in the story. To do so, Wilson argues for a distinctive thesis regarding movie narration: the Imagined Seeing Thesis, the view that movie narratives work on viewers by prompting them to imagine that they see the characters and events in the story, or “imagine seeing” for short. There are many questions that philosophers raise about the Imagined Seeing Thesis. One basic question is: just what is imagined seeing? Is it just a manner of speaking? Should the phenomenon Wilson calls “imagined seeing” be analyzed or reduced to other sorts of imagining? For instance, perceptual imagining, which is counterfactually dependent on perception (Currie 1990: 181–185) or “seeing-in,” where one imagines one thing (a photographic image of Cary Grant) as another (an image of Roger Thornhill).14 Is the Imagined Seeing Thesis an empirical claim about how viewers, in fact, engage, with movies? (Stecker 2013: 153). If so, what is the empirical evidence in support of it? Imagined seeing is often differentiated from impersonal imagining, or imagining that certain things are so in the fiction, for example, that Harry Potter is a student at Hogwarts, the school for wizards.15 In contrast, imagined seeing is a form of personal imagining, for I place myself into the content of what I imagine, for instance, I imagine that I see Dorothy and Toto arriving in the land of Oz.16 Imagined seeing is thought to be a kind of experiential imagining because when a visual representation induces imagined seeing in the viewer, it is said to induce an experience one thinks of as “as if” one were actually seeing the events and characters in the fiction (Wilson 2011: 73, 2013: 167). Talk of imagined seeing gives Wilson a distinctive way to argue for his version of the implicit cinematic narrator, a “minimal narrating agency” (Wilson 2011: 112). This minimal narrating agency has no personal characteristics; its only function is to show the story events to the audience. Thus, Noël Carroll has dubbed this argument the “Seeing/Showing” Argument for the implicit cinematic narrator (Carroll 2016). The Seeing/Showing Argument: 1. Movie narration works on the audience by standardly prompting them to imagine that they see the story events from the fictional world. 2. If (1), then standardly, in all fiction films, there must be a fictional presenter, an implicit, minimal narrating agency that shows the audience the events from the world of the fiction. 14  For discussions of seeing in, see Wollheim (1998: 217–238), Hopkins (2008, 2016), and Stecker (2013). For a response to Wollheim that imagined seeing should not be understood in terms of seeing in, see Walton (2002). 15  Some philosophers, such as Noël Carroll, Colin McGinn, and Greg Currie maintain that imagining at the movies is standardly impersonal imagining. See Carroll (1995: 98–99, 2006, 2016), Currie (1991, 1995a, b), and McGinn (2005) and Gaut (1998: 333–334, 2010: 217). 16  For personal versus impersonal imaginings, see Currie (1990: 181–185).



3. Therefore, standardly, in all fiction films, there is a minimal narrating agency that shows the audience the events from the world of the fiction. If the Seeing/Showing Argument is correct, then, standardly, in every fiction film there are implicit “narrating agencies” that mediate our access to the story worlds and the claim that every fiction film has an implicit narrator is confirmed. But how sound is the argument?  he Imagined Seeing Thesis: How to Formulate It? T The Imagined Seeing Thesis needs to be refined, as Wilson recognizes, because there are some puzzling questions that arise when we try to take seriously viewers talk that they imagine seeing the events and characters from the fiction. The problem with imagined seeing arises from the following claims: if the audience imagines that they are seeing the story events, then they imagine seeing them from a series of definite visual perspectives. If they imagine seeing from a visual perspective, then they also imagine that they see from a vantage point that is within the story world. This would be the account of the IST that George Wilson calls “Face-to-Face-Imagined Seeing”: Face-to-Face Imagined Seeing Thesis: When the audience watches a fiction film, they are prompted to imagine that they are seeing the story events by standing face-to-face with them (Wilson 2011: 36).

But Face-to-Face Imagined Seeing gives rise to a host of perplexing questions. Is it plausible to think that as we watch Christopher Nolan’s Dunkirk (2017), for example, we are mandated to imagine that we are there on the beach at Dunkirk as bullets fly and the Allied Forces are rained down with bullets?17 What do we then imagine about how we are able to dodge bullets? And when we imagine seeing a murder in the story that is stipulated to be unseen, doesn’t that involve us in engaging in contradictory imaginings, that (a) we imagine that it is true that the murder is unseen and (b) we imagine seeing the murder (Currie 1995a, b; Carroll 1995, 2005, 2016; Gaut 2010)? Wilson rejects Face-to-Face Imagined Seeing because he does not think it is true to the viewer’s experience. It is not part of our engagement with a movie that we imagine ourselves located within the story space, at the viewpoint implied by the vantage point of the motion picture shot. So Wilson instead favors this version of imagined seeing: Mediated Imagined Seeing: The audience imagines that they see a recording of the events in the story world that has been photographically derived in some undetermined way (Wilson 2011: 89).

17  But see Thomson-Jones (2012), who argues that some films prompt the audience to imagine that they are moving within the world of the film.



According to Mediated Imagined Seeing, the audience imagines that they see the story events, indirectly, through seeing a motion-picture-like recording of them. Just how this recording is obtained is not part of what the audience needs to imagine (Wilson 2011: 89–91). Wilson has a battery of arguments in favor of Mediated Imagined Seeing. His most central point is that Mediated Imagined Seeing is the best way to explain the way that aspects of cinematic construction such as color, grain, focus, camera angle, and editing mediate the audience’s imagined seeing of the characters and action. For example, when the viewer sees the action jump quickly from one time and place to the next, as happens in the final shots of North By Northwest, what does the viewer imagine is going on? According to Mediated Imagined Seeing, she does not explain this as the actual filmmaker’s decision to move the action by having a rapid cut. Instead, the viewer imagines that there is some editing going on at the level of the fiction, through the action of the implicit, minimal narrating agency. One worry is that, like its cousin, Mediated Imagined Seeing, is open to the objection that it gives rise to absurd imaginings (Gaut 2004: 242; Carroll 2006: 179–180). For if the viewer imagines she is watching motion picture shots of actual events, then she will need to imagine the implications of this, and embarrassing questions follow such as how is the fictional narrator able to record the incidents and go unnoticed? How can there be a recording of a story, such as The Ten Commandments (Demille, 1956) that takes place in B.C.E., before the invention of the camera? Wilson is aware of these objections and says that we may imagine that we are watching segments of the story world via “naturally iconic shots,” shots that are causally dependent on the scene but need not be produced by a camera. So, imagining the presence of a camera or other recording device need not be part of what it is that the audience imagines (Wilson 1997: 113, 2011: 48). Thus, the burden of Wilson’s reply to the concern about absurd imagining is that viewers can imagine things without having to imagine the implications of what they are imagining (Wilson 2013: 161). Wilson supports this claim he derives from some work by Kendall Walton (Walton 1990: 174–182). Walton’s idea is that there are questions about fictions whose answers are not specified and so these questions are “silly,” pointless, and inappropriate to ask. Wilson takes up Walton’s idea and illustrates it with the example from Flash Gordon in the old black and white science fiction serials (1986). In the story, we suppose that Flash Gordon has a viewing machine that enables him to see anywhere in the universe, but such a device violates the laws of physics, as we know them. Wilson maintains that it is a silly question for the viewer to imagine how such a device works, for this is indeterminate or not specified in the Flash Gordon stories (Wilson 1997: 314–315). The same is true if viewers had to imagine the implications of imagining they are watching a recording of actual events. With this move, Wilson tries to fend off the absurd imaginings objection by saying that questions about how the recording of events comes about are silly ones to ask.



In reply, critics such as Berys Gaut and Noël Carroll insist that what is known as the “Realistic Heuristic” governs our imaginings about fiction. Their idea is that when we engage with a work of fiction, we “fill in” and draw implications from what is explicitly true in the fiction based on how things are in the real world, unless it is explicitly stipulated to be otherwise.18 In the Flash Gordon serials, the story makes explicit that the screening devices work as shown. So, we go along with this feature of Flash’s screen, just as we go along with the idea, in other stories, that there are wizards that can perform magic, there are zombies that are dead and alive, and so on (Carroll 2006: 181). In short, critics charge that Wilson’s defense rests on a misleading analogy (Carroll 2006: 181). Because Flash’s screening device is explicitly introduced to work as represented, we do not take issue with it. We suspend “default realism” and do not expend energy worrying about it. However, no one clues us into the implicit narrator: it is, after all, an implicit feature of the narration, not explicit. So, we cannot fend off worries about how the cinematic operator with the thought “just accept the filmmaker says this is how things work” (Carroll 2006: 125). Because there is not, in other words, an exceptions clause for the cinematic narrator, the Realistic Heuristic licenses us to imaginatively fill in the implications of its presence in the story world (Carroll 2006: 181). Thus, the debate between Wilson and his critics concerns whether the questions that critics ask about the operation of Wilson’s version of the implicit cinematic narrator are “silly” ones to ask. This is the question we examine in the next section. Part V: Reconsidering the Objection from Absurd Imaginings To review, by far, the most serious objection that friends of the implicit cinematic narrator face is the concern about absurd imaginings. This question dogs all versions of the cinematic narrator we have discussed. In response to this problem, Chatman maintained that questions about how the narrator comes to have its knowledge are “non-questions” (Chatman 1990: 130). Levinson had to fend off Kania’s concern that it is not possible to say a fictional narrator is our guide to the story’s sights and sounds without embarrassing questions about the narrator following. Also, Wilson faces the objection that his Mediating Imagined Seeing thesis cannot avoid the sort of absurdities that have plagued other formulations of imagined seeing. The question we must now, then, try to sort out is whether the critics’ questions about the cinematic narrator are legitimate ones to ask. Concerns about improbabilities in works of fiction go back to Aristotle’s Poetics, where he said that ideally there should be no improbabilities in the plot (Poetics 1460b27). Drama is an imitation of human action and life. Dramas that have improbable incidents, especially in the plot, undermine the sense that goings on in the drama work as they do in real life. For things in real life obey  Gaut (2004: 245), Carroll (2006: 181, 2016: 120).




cause and effect, and the audience’s emotional response to the story depends on their making a connection between the fiction and everyday life. Thus, plays that build to a narrative resolution by having a deus ex machina solution, such as Medea improbably spirited away in a chariot at the last minute, are to be avoided, unless there is some overriding reason to include them (Poetics 15.1454a37). Also, when the poet must include them, to achieve a certain effect, the artist should find ways to de-emphasize them, for instance, by including them in the “backstory” of the larger story world from which the play draws, and not as part of the events that are dramatized in front of the audience on the stage (Poetics 15.1454b5-7). Kendall Walton voices similar ideas when he advances the “Reality Principle” (Walton 1990:144–151). To comprehend and appreciate the story, we must “fill in: a great deal of information that is not explicitly represented.19 The Reality Principle directs an appreciator of fiction to “fill in” or infer from what is explicitly presented in the fiction, based on the idea that the fictional world operates as the real world does. So, for example, to appreciate the Harry Potter stories and movies, we have to infer things that the novel does not explicitly introduce: even though Harry is a wizard and can perform magic, in every other relevant respect, Harry is like ordinary “muggles”—he is not immortal, has parents, and so on. Thus, Walton’s “Reality Principle” is very much like Carroll and Gaut’s “Realistic Heuristic.” One might say that, in general, the Realistic Heuristic, the idea that we fill what is true in the fiction according to how things work in real life, is sound. For an important, if not universally accepted, way of thinking about fiction is that it is capable of affecting a change in the audience’s view of themselves and the world.20 To do so, works of fiction must present representations of human action that are broadly realistic and true to life, unless things are specified otherwise in the story (for instance, Flash Gordon’s viewing screen or magic in the world of Harry Potter). Further, there is an important practical reason for assuming that the Reality Principle holds. It is just not possible for an author or filmmaker to fill out everything that is true in the story world she creates for the reader or viewer. To do so would run the risk of distracting from the appreciation of what is essential and relevant to know and what is not. Instead, with the Reality Principle, the creator of the fiction can leave certain fictional truths implicit, and we use how things work in the real world to fill in information about the story world. When the story world departs from the real world, this exception can be explicitly introduced. So, the Reality Principle provides a practical way for the appreciator of a fictional work to “fill out” the story world, without leaving the contents of the story world mostly unspecified (Gaut 2004: 245). 19  The problem of just what an appreciator of fiction “fills in” as she comprehends a story is a subject of great debate. See, for example, Lewis (1978), Beardsley (1981: 242–247), Walton (1990: 144–161), Lamarque (1990), and Lorand (2001). 20  Catherine Wilson (2004), Elisabeth Schellekens (2007).



In response, Wilson is likely to insist that there are paradoxes and inconsistencies at the base of many fictional narratives. A prime example that Wilson gives comes from the Adventures of Huckleberry Finn. It is fictional that Huck, a barely literate young man who goes stir crazy when he sits for too long, wrote the 300-page memoir about his adventures (Wilson 1997: 309). If we ask how this could be, we would get tangled up in imagining absurd scenarios that distract from our proper appreciation of the story. Wilson suggests we are just not meant to imagine the implications of how such a narrative feat is possible, and so the Realistic Heuristic is blocked. We need not imagine all the implications of how Huck could have crafted the tale. In other words, we should not expend energy worrying about the embarrassing questions about the cinematic narrator/fictional shower because they have no answer within the world of the story, and so they are silly questions to ask (Wilson 1997: 308–9; Wilson 2011: 48). It might appear that we have arrived at a stalemate or impasse between philosophers on both sides of the debate. However, we might make headway by looking more closely at some of the examples that are often given of silly questions to ask. If these paradigm examples are like, in relevant respects, questions one might ask about the cinematic narrator, then Wilson’s argument might be plausible. On the other hand, if the agreed upon examples are not sufficiently like asking questions about the cinematic narrator, then we might have some reason for thinking that Wilson’s response to his critics is not plausible. Centrally, the silly questions that Walton considers to be inappropriate are so because it is wrong to seek an answer in the world of the fiction. Instead, we find the answer outside the story world, by considering the generic conventions that govern the art form in question. For example, the audience at the opera does not ask why it is that the characters in opera often spend their last moments singing, while they are passing away in excruciating pain! (Walton 1990: 177). Only someone who was ignorant of the conventions of opera, as a particular art form, would ask this question. Further, it is fictional in the play, The Belle of Amherst, that the character, Emily Dickinson, is shy and retiring, a person of few words. However, the actor who impersonates Dickinson has many lines and commands the attention of the audience (Walton 1990: 176). It is silly or inappropriate to ask why a quiet person is talking so much and expect the answer to be found within the terms of the story. The conventions and nature of a play of this sort require that the character talk a lot to convey the poet’s thoughts and feelings. In these types of cases, as Greg Currie notes, there is no one-to-one correspondence between the features of the representation, for instance, of Othello and the features that Othello has in the story world (Currie 2010: 59). That is, a well-informed spectator of theater understands that the poetic speeches that Othello offers, for example, do not reflect a quality in the character of Othello, but are there instead to stir the emotions with the beauty and eloquence of the quality of the language. As Currie puts it, “While the words uttered by the actor constitute great poetry, they are not represented as representing great poetry in the mouth of the character” (Currie 2010: 60).



Generalizing from these examples, we can explain why the questions one could raise in these cases are inappropriate ones to ask. A question is a silly one, in these cases, to ask, provided the answer is not found within the terms of the fiction but instead is explained by the nature of the genre of the artwork and its associated conventions. We can also see how accepting that these cases involve silly questions is compatible with holding our imaginings about fiction are, in general, governed by the Realistic Heuristic. It is not given as true in the fiction that Othello is a fine poet or speechmaker: this is a feature of how Othello is represented due to the artistic requirements of Shakespeare’s plays. As such, we are not mandated to imagine the implications for what is true in the story world when Othello makes a beautiful speech. Similarly, we are not required to infer what follows from imagining that barely literate Huck Finn authors a 300-page memoir. We understand it is Mark Twain’s words that are the source of the fictional memoir, not Huck Finn’s, because fictional narratives have to be crafted by actual authors. So, we understand that we do not need to infer what follows for the story world if we imagine that we are reading a lengthy fictional memoir that Huck pens. Now what follows for the debate between Wilson and his critics over the absurd imaginings? Recall that on Wilson’s Mediated Imagined Seeing thesis, when we watch a movie, we are to imagine that we are watching a recording of the story events taken from within the fictional world. Critics ask, how was such a recording made? How could there be a recording of events if the story is set in a time before the camera was invented? And, further, if we suppose that some naturally occurring camera is the source of the shots we are seeing, what are we to imagine about point-of-view shots? How can naturally occurring cameras get inside people’s heads? Are the questions that his critics pose silly ones to ask? Recall that according to Wilson’s Mediated Imagined Seeing, we are to imagine that it is true in the story world that such a recording was made. For recalling the Ontological Gap Argument, the narrator or narrating device has to be imagined to be part of the fictional world in order for us to imagine that what we are seeing is a recording taking place from within the story world. The objection then is that once we imagine a recording takes place in the world of the fiction, we must imagine what follows from this, in accordance with the Realistic Heuristic. But when we imagine what follows from the presence of a recording of the story events, absurd imaginings follow. Wilson insists that to seek answers to these questions is inappropriate. But our examination of some central cases of silly question pertaining to fiction suggests that the questions about how the recording comes about are legitimate ones to ask. For we said that there is no reason to think that it is true in the fiction that Huck is a literate or capable of sitting still long enough to write a 300-page memoir. And there is no reason to think it is true in the fiction that Othello, a brash man of action, makes beautiful speeches, and so on with the other examples we looked at. But Wilson would have us imagine that it is true



in the fiction that there is a recording of the story events or fictional facts. This means, as Wilson’s critics charge, that once we suppose that the implicit narrator (or minimal narrating agency) is part of the fictional world, it is reasonable to fill in the implications of its presence there. And when we do, we get tangled up in the embarrassing questions about the cinematic narrator that we have rehearsed in this section.21 Conclusion: Further Issues for Cinematic Narration If we are skeptics about imagined seeing as the way movies work on us as viewers, is there another way to explain the phenomena to which Wilson’s work draws our attention? For Wilson is insistent that any account of how movie narratives engage our imagination will need to address the way in which viewers at the movies describe their experiences as that it is “as if ” they are seeing segments of the story world. A concern with explaining the impression that we are making perceptual contact with the story world also clearly motivates Levinson’s Ontological Gap Argument. But explaining our engagement with movies in terms of imagined perceptual relations faces problems, as we have seen. How might we undertake to explain how film as a distinctively visual form of storytelling works on us, the viewers? Greg Currie proposes that we distinguish visual fictions from nonvisual fiction by how film narration determines or conveys the story content. Currie uses the term “perceptual imagining” to mark out the distinctive kind of imagining movies prompt in viewers (Currie 1991: 140, 1995a, b: Chapter 6). When a viewer watches a movie, it is the viewer’s actual perception of a visual image that prompts her imagining the story’s contents. In visual fictions, the viewer’s imagining of story events is then counterfactually dependent on looking at images.22 Noël Carroll discusses the distinct perceptual and cognitive faculties that movie narration engages, as a form of pictorial comprehension (Carroll 2008: 108–115). Movies present familiar scenes and characters even if they are ones the filmmaker makes up. Thus, movies mobilize the same capacity for object recognition that we employ in everyday life. Therefore, one might say that the “Recognition Prompt” view can explain why viewers report that their experience is “as if ” they see the characters in real life, without positing they stand in an imagined perceptual relation to them. What they are reporting is a sense of recognition of something previously encountered in perception, not an imagined seeing of them. A further possibility is to hold that the notion that viewers at the movies look at the moving pictures on the screen and “see in” to them the characters  See also Curran (2016: 103–106).  Wilson remains open to the possibility that what he means by “imagined seeing” at the movies is what Currie means by perceptual imagining. See Wilson (2011: 75–76). Currie revisits his views about imagined seeing at the movies in Currie (2018). 21 22



and situations that the image depicts (Wollheim 1998: 217–238; Hopkins 2008, 2016). Seeing in is not an imagined seeing of characters but a seeing in which one thing (a movie shot of Ingrid Bergman) is taken as a representation of another (Ilsa Lund). The virtue of “seeing-in” is that it can account for the role that perception plays in imagining story content, while also being able to explain how the viewer can appreciate the properties of the moving shot as an image or representation (Stecker 2013: 153–4).23

Conclusion Cinematic narration is the way in which movies tell their stories to an audience. The overall question we have looked at here is how do movies work on us so that we come to imagine the story events? There are two broad areas of contention. The first concerns whether to comprehend what is true in the story— we need to imagine a fictional presenter who reports or shows that things are so in the story. Alternatively, is the Imagination Account of Fiction right that we comprehend the story in virtue of the actual filmmaker’s mandate to imagine things are thus and so in the story? The second concerns whether audiences at the movies standardly imagine seeing the characters and story events. Or do we instead imagine that certain things are so in the story world, without imagining we are seeing these incidents? We have seen that the issue of how we imaginatively “fill in” the implications of what is explicitly the case in the fiction is central to resolving both issues. An exciting line of further inquiry is whether the Imagination Account of Fiction can offer a complete answer to how viewers comprehend a story. If the argument in this chapter is correct, there is sufficient reason to hope it can do so.

Bibliography Beardsley, M. 1981. Aesthetics: Problems in the Philosophy of Criticism. Indianapolis: Hackett. Booth, W. 1961. The Rhetoric of Fiction. University of Chicago Press. Bordwell, D. 1985. Narration and the Fiction Film. Madison: University of Wisconsin Press. Carroll, N. 1990. Philosophy of Horror, or Paradoxes of the Heart. London: Routledge. ———. 1995. Review: Mimesis as Make-Believe. The Philosophical Quarterly 45 (178): 93–99. ———. 1996. Fiction, Nonfiction and the Film of Presumptive Assertion. In Film Theory and Philosophy, ed. Richard Allen and Murray Smith. Oxford: Oxford University Press. ———. 2005. Introduction: Part IV: Film Narrative/Narration. In Philosophy of Film and Motion Pictures, an Anthology, ed. Noël Carroll and Jinhee Choi, 175–184. Oxford: Blackwell Publishing. ———. 2006. Introduction to Part IV: Film Narrative/Narration. In N. Carroll and J. Choi (2004), 175–184.  For Wilson’s view on imagined seeing as “seeing in” look at Wilson (2013: 167–168).




———. 2008. The Philosophy of Motion Pictures. Malden/Oxford: Blackwell Publishing. ———. 2016. Motion Picture Narration. In K. Thomson-Jones, ed. (2016). Carroll, N., and J. Choi. 2004. Philosophy of Film and Motion Pictures: An Anthology. Malden/Oxford: Blackwell Publishing. Chatman, S. 1990. Coming to Terms: The Rhetoric of Narrative in Fiction and Film. Ithaca: Cornell University Press. Curran, A. 2016. Fictional Indeterminacy, Imagined Seeing, and Cinematic Narration. In K. Thomson-Jones, ed. (2016). Currie, G. 1990. The Nature of Fiction. Cambridge: Cambridge University Press. ———. 1991. Visual Fictions. The Philosophical Quarterly 41 (163 (April)): 129–143. ———. 1995a. Image and Mind: Film, Philosophy, and Cognitive Science. Cambridge: Cambridge University Press. ———. 1995b. Unreliability Refigured: Narrative in Literature and Film. The Journal of Aesthetics and Art Criticism 53 (1): 19–29. ———. 2010. Narratives and Narrators: A Philosophy of Stories. Oxford: Oxford University Press. ———. 2018. Visually Attending to Fictional Things. In Perceptual Imagination and Perceptual Memory, ed. Fiona Macpherson and Fabian Dorsch, 186–208. Oxford: Oxford University Press. Davies, David. 2010. Eluding Wilson’s “Elusive Narrators”. Philosophical Studies 147: 387–394. Gaut, B. (1997). Film Authorship and Collaboration. In R. Allen and M. Smith, eds. Film Theory and Philosophy. Oxford University Press: 149–72. ———. 1998. Imagination, Interpretation and Film. Philosophical Studies 89: 331–341. ———. 2004. The Philosophy of the Movies: Cinematic Narration. In P.  Kivy, ed. (2004), 230–253. ———. 2010. A Philosophy of Cinematic Art. Cambridge: Cambridge University Press. Hopkins, R. 2008. What Do We See in Film? The Journal of Aesthetics and Art Criticism. 66 (2): 149–159. ———. 2016. Realism in Film and Other Representations. In K.  Thomson-Jones, ed. (2016). Kania, A. 2005. Against the Ubiquity of Fictional Narrators. The Journal of Aesthetics and Art Criticism 63: 47–54. Kivy, P., ed. 2004. The Blackwell Guide to Aesthetics. 1st ed. Malden: Wiley-Blackwell. Köppe, T., and J. Stürhring. 2011. Against Pan-Narrator Theories. Journal of Literary Semantics 40: 59–80. Lamarque, P. 1990. Reasoning What Is True in Fiction. Argumentation 4: 333–346. Lamarque, P., and S.H.  Olsen. 1994. Truth, Fiction, and Literature. A Philosophical Perspective. Oxford: Oxford University Press. Levinson, J. 1993. Seeing, Imaginarily, at the Movies. The Philosophical Quarterly 43: 70–78. ———. 1996. Film Music and Narrative Agency. In D. Bordwell and N. Carroll, eds. Post-Theory. Reconstructing Film Studies. The University of Wisconsin Press, 248–82. Lewis, D. 1978. Truth in Fiction. American Philosophical Quarterly 15 (1): 37–46. Livingston, P. 1997. Cinematic Authorship. In R. Allen and M. Smith, eds. Film Theory and Philosophy. Oxford University Press, 132–48. ———. 2005. Narrative. In The Routledge Companion to Aesthetics, ed. Berys Gaut and Dominic McIver Lopes, 2nd ed., 359–370. Oxon: Routledge Press. ———. 2013. The Imagined Seeing Thesis. Projections 7: 139–146.



Lorand, R. 2001. Telling a Story or Telling a World. British Journal of Aesthetics 41 (4): 425–443. Matravers, D. 1997. The Paradox of Fiction: The Report Versus the Perceptual Model. In Emotion and the Arts, ed. Mette Hjort and Sue Laver, 78–93. Oxford: Oxford University Press. ———. 2014. Fiction and Narrative. Oxford: Oxford University Press. McGinn, C. 2005. The Power of Movies: How Screen and Mind Interact. New York: Pantheon. Nehamas, A. 1981. The Postulated Author. Critical Inquiry 8: 133–149. Pye, D. 2013. Seeing Fictions in Film. Projections 7: 131–138. Schellekens, E. 2007. Aesthetics and Morality (Bloomsbury Aesthetics). Continuum. 63–92. Stecker, R. 2013. Film Narration, Imaginative Seeing and Seeing-In. Projections 7: 147–154. Stock, K. 2013. Imagining and Fiction: Some Issues. Philosophy Compass 8 (10): 887–896. ———. 2017. Only Imagine. Oxford: Oxford University Press. Thomson-Jones, K. 2007. The Literary Origins of the Cinematic Narrator. British Journal of Aesthetics 47 (1): 76–94. ———. 2009. Cinematic Narrators. Philosophy Compass 4 (2): 296–311. ———. 2012. Narration in Motion. British Journal of Aesthetics 52: 33–43. ———., ed. 2016. Current Controversies in the Philosophy of Film. New York/London: Routledge. Walton, K. 1990. Mimesis and Make-Believe. On the Foundations of the Representational Arts. Cambridge, MA: Harvard University Press. Walton, K. 2002. Depiction, Perception, and Imagination: Responses to Richard Wollheim. The Journal of Aesthetics and Art Criticism 60 (1): 27–35. ———. 2008. Marvelous Images: On Values and the Arts. Oxford: Oxford University Press. Williams, B. 1973. “Imagination and the Self”, in his Problems of the Self: Philosophical Papers 1956–1972, 26–45. Cambridge: Cambridge University Press. Wilson, G. 1986. Narration in Light: Studies in Cinematic Point of View. Baltimore: Johns Hopkins University Press. ———. 1997. Le Grand Imagier Steps Out: The Primitive Basis of Film Narration. Philosophical Topics 25: 295–318. Wilson, C. 2004. Literature and Knowledge. In Philosophy of Literature: Contemporary and Classic Readings, ed. E. John and D. Lopes. London: Blackwell. Wilson, G. 2007. Elusive Narrators in Literature and Film. Philosophical Studies 135: 73–88. ———. 2011. Seeing Fictions in Film: The Epistemology of Movies. Oxford: Oxford University Press. ———. 2013. Seeing through the Imagination in the Cinema. PRO 7: 155–171. Wollheim, R. 1998. On Pictorial Representation. The Journal of Aesthetics and Art Criticism 56: 217–233.


Narrative and the Moving Image Patrick Keating

How does a film tell a story? The simplicity of this question hides a number of puzzles. Some scholars try to unpack these puzzles by focusing on the word “story.” What sorts of situations might constitute a story? How are those situations linked? Others have focused on the word “tell.” Does a film tell a story or show it? And who does the telling (or showing)? A third alternative is to focus on the word “film.” Does film offer distinctive resources for storytelling? What resources does it share with verbal narrative? In this chapter, I approach all of these puzzles by starting with the “how.” How does a film tell a story? Over time. Whether film or novel, narrative is temporal. Whether narrative or not, so is film. Part One endorses a “rhetorical-functionalist” theory of narrativity, placing special emphasis on the temporal effects of prospection, retrospection, and re-­ cognition. Part Two develops this time-based approach further by considering the modality of narrative—that is, the way narratives appeal to our sense of what might happen. Part Three expands the focus beyond film to include television, where many narratives are told in serial form, stretching across multiple episodes or seasons. Part Four turns to a difficult problem that has received a great deal of attention in the philosophy of film—the problem of the cinematic narrator. After reviewing some of the key positions in this debate, I argue that we can understand cinematic narration without the overly personalized notion of the narrator. I also propose that the purpose-oriented logic of the rhetorical-­ functionalist approach to narrative gives us good reasons to hold on to the “implied author” as a useful concept for narrative analysis.

P. Keating (*) Trinity University, San Antonio, TX, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




Narrative Dynamics In “What Is a Temporal Art?,” Jerrold Levinson and Philip Alperson have proposed over a dozen ways that an art form might be considered temporal. The cinema fits into almost every category. For instance, like music and dance, the cinema requires time in its presentation: “The parts of the artwork are not all available at any one moment, but only consecutively.”1 Narrative movies are temporal in another sense: As the sequence onscreen unfolds, typically over the course of a couple hours, the story-oriented spectator must make sense of another sequence, the sequence of changes in the fictional world, a sequence that may cover hours or years or centuries. Narrative theorists have proposed various terms for these two sequences, including the structuralist pair story-­ discourse and the formalist pair fabula/syuzhet.2 In a recent essay on narrativity, Meir Sternberg has offered several possible pairs: “actional vs. presentational or rhetorical, mimetic vs. communicative, narrated vs. narrative, told vs. telling/ reading dynamics.”3 Whichever terms we use, the process “entails an interplay between the one sequence’s flow of development and the other’s flow of disclosures—between the two great sources of narrative change, in the world itself and in our knowledge about it, respectively.”4 Though primarily concerned with literature, Sternberg’s evocative language, contrasting a flow of development with a flow of disclosures, seems particularly apt for cinema. Watching a narrative film, the pictures and sounds offer a flow of disclosures; those pictures and sounds reveal and conceal the details of the developing story-world. Some observers might question the wisdom of splitting these two sequences at all. Don’t we just see the story-world unfolding onscreen? As debates in the philosophy of depiction suggest, such a question might be answered in many ways—for instance, by appealing to ideas of recognition or imagined seeing. For now, I want to argue in favor of the dual-sequence proposal by giving an idea of its explanatory power. The “development and disclosures” model sharpens our awareness of a crucial feature of cinematic storytelling: its selectivity. We do not see the story-world in its entirety; we see a selection of the story-­ world, represented in framed pictures. Consider two scenes from Clarence Brown’s 1926 film Flesh and the Devil, where the camera’s movements shape our understanding of the story-world by controlling what is inside the frame and what is outside it. Set in the nineteenth century, the film tells the story of a woman, Felicitas, who tries to destroy the friendship between two men, Leo and Ulrich. In one scene, Leo (John Gilbert) and Felicitas (Greta Garbo) are 1 Jerrold Levinson and Philip Alperson, “What Is a Temporal Art?,” Midwest Studies in Philosophy 16 (1991): 441. 2  On the differences between the story-discourse pair and the fabula/syuzhet pair, see Seymour Chatman, “Towards a Theory of Narrative,” New Literary History 6, no. 2 (Winter 1975): 295–296. 3  Meir Sternberg, “Narrativity: From Objectivist to Functional Paradigm,” Poetics Today 31, no. 3 (Fall 2010): 636. 4  Sternberg, “Narrativity,” 637.



embracing. The door opens, revealing Felicitas’s heretofore unintroduced husband, an angry older man named Count von Rhaden (Marc McDermott). Seeing his wife with Leo, Rhaden raises his left hand in astonishment. The camera dollies closer as Rhaden clenches his fist in anger; onscreen, it appears as if the husband is crushing the lovers with his hand. The next shot cuts to Felicitas. She has seen her husband, but Leo has not. In one sense, the dolly-in toward the clenched hand supplies privileged information; we see what the characters do not. Leo and Felicitas are oblivious to the clenched hand; even Rhaden himself is probably unaware of it, so immersed is he in his own rage. The film tells its story by focusing our attention on a crucial detail, demanding that we notice its significance. Still, our privileged view remains partial. The hand itself occludes our view of Leo and Felicitas, and the tighter framing leaves Rhaden’s face off-screen. These exclusions affect our experience of time by making us want to know more: in the short term, we want to know when the lovers will notice Rhaden’s presence; in the longer term, we want to know if Rhaden’s anger will pose a threat to the lovers. Rhaden challenges Leo to a duel, which is represented in the next scene in a celebrated moving-camera shot. First, we see seven men silhouetted against the sky. The camera (mounted on a vehicle) dollies back rapidly, as the two duelists take their paces and the four “seconds” run to safety below the horizon. Now only one man remains in the shot: the “impartial,” who raises and lowers his hands to signal for the men to fire. From off-screen, we see two puffs of smoke. The seconds re-enter the shot and run for the sides of the frame while the screen fades to black. This shot is remarkably opaque. We see seven men, but we do not see their faces. Then we see only three men, and then only one—and the one we care about, Leo, is off-screen. The firing of the guns leaves us as uncertain as ever, since we see no bodies fall. The quick fade-out forces us to wait for the next scene to find out if Leo has survived or not. (He has.) In the previous scene, the camerawork was informative, dollying in to draw our attention to a key detail. Here, the camerawork explicitly conceals information. Moving the camera backwards makes the edges of the frame salient, emphasizing the process of inclusion and exclusion. Again, the framing shapes our experience of time. Suspense about an upcoming duel is converted into intense curiosity about a duel that just happened. These shifts are profoundly temporal, not just because the story-world is changing, but because our understanding of that world must be revised with every new disclosure. As Eyal Segal explains, this temporal way of thinking about a film or novel places special emphasis on functions: “Unlike most narratological approaches, Sternberg’s defines this essence of narrative not in the mimetic terms of represented or narrated action, but rather in the rhetorical-communicative terms of narrative interest.”5 One way of defining narrative is to prioritize the shared 5  Eyal Segal, “The ‘Tel-Aviv School’: A Rhetorical-Functional Approach to Narrative,” Current Trends in Narratology, ed. Greta Olson (Berlin: Walter de Gruyter, 2011), 302.



properties of stories, as opposed to plots, thereby favoring one sequence over the other (for instance, by stating that a story must consist of two or more linked events). By contrast, the rhetorical-functionalist scholar tries to keep both sequences in mind at all times, considering how the interplay between the flow of disclosures and the flow of development produces effects, most notably the temporal effects of prospection (looking ahead toward the future), retrospection (looking back to a known gap in the past), and re-cognition (rethinking events we thought we understood). For the viewer, interest is sparked by a gap—that is, by a salient unknown. There are countless things we do not know about the story-world, but the film makes certain gaps in knowledge salient, as when a character says that she is going to an audition tomorrow (prompting us to wonder if she will get the part) or when another character finds a dead body in the living room (prompting us to ask who killed the victim and why). Characterizing the process of prospection as a kind of suspense, Sternberg explains: “Suspense arises from rival scenarios envisaged about the future: from the perceptible discrepancy between what the telling lets us readers know about the happening (e.g. a conflict) at any moment and what still lies ahead, ambiguous because as yet unresolved in the told world, at least not to our knowledge.”6 In Flesh and the Devil, when the duel scene begins, Leo’s fate has not yet been resolved. It is a salient unknown in the future. Because Leo is a sympathetic protagonist, we may hope that he will survive, and we may fear that he will be injured or killed. Sternberg contrasts the future orientation of suspense with the past orientation of curiosity. A mystery may skip over a murder, generating the reader’s curiosity about who committed the crime: “Knowing that we do not know, we then go forward with our mind lingering on the gapped antecedents, trying to infer (bridge, compose) them in retrospect.”7 In the duel scene, as soon as we see the two puffs of smoke off-screen, we know that the duel has already happened; it becomes a gap in the past. The moving camera has ensured that we notice this gap by making the frame and therefore off-screen space unusually salient. Re-cognition or surprise is also oriented toward the past, but in a different way. According to Sternberg, “We must be lured into a false certainty for a time about time past. […] The narrative first unobtrusively gaps or twists its chronology, then unexpectedly discloses to us our misreading in ignorance and enforces a corrective rereading in late re-cognition.”8 In these terms, the initial appearance of Rhaden qualifies as a surprise. Until this point, the film has not disclosed the crucial fact that Felicitas is married. Perhaps we assume that she is unmarried; perhaps we just do not think about her marital status at all. Either way, we must revise our understanding of the story-world when Rhaden appears. What seemed like a romantic relationship was a dangerous 6  Sternberg, “Narrativity,” 640–641. Sternberg’s definition of suspense is admittedly broad, and some readers might prefer to use “prospection” as a less emotionally laden term. For a more narrowly targeted theory, see Noël Carroll, “Toward a Theory of Film Suspense,” Theorizing the Moving Image (New York: Cambridge University Press, 1996), 94–117. 7  Sternberg, “Narrativity,” 641. 8  Sternberg, “Narrativity,” 641.



affair all along. The dolly-in toward Rhaden’s hand forces us to confront Leo’s ignorance—and our own. Whether appealing to prospection, retrospection, or re-cognition, the film shapes the spectator’s experience by disclosing developments over time. Within film studies, the leading exponent of Sternberg’s theory of narrative has been David Bordwell. In the co-authored volume The Classical Hollywood Cinema, Bordwell argued that we may think of a Hollywood narrative as a chain of tightly linked causes and effects.9 For Bordwell’s admirers, this observation helps explain why mainstream films are so easy to follow. For Bordwell’s detractors, the causal chain model seems overly reductive, eliminating everything that does not follow in a straight line. My own view is that the causal chain is the action-oriented portion of what is ultimately Bordwell’s much richer “functionalist” argument, in which the film cues the spectator to make various inferences about the story (fabula), thereby activating the emotionally charged play of prospection, retrospection, and re-cognition that Sternberg discusses. Although it can be useful to think of a classically constructed film as a straight line of causes and effects, I prefer another metaphor that Bordwell proposes: the “winding corridor” that shifts and curves, suggesting a clear path to follow but keeping us guessing all the while.10 Early causes point us toward later effects, but not in a way that everything seems inevitable. Will the lovers meet? If so, how? Will the villain be defeated? If so, how? If anything, Hollywood’s corridors have grown more winding or even crooked in recent years, to the point that some of Hollywood’s most complex narrative films mislead their spectators for long stretches of time before revealing their “third-­ act twists,” as in The Sixth Sense (M.  Night Shyamalan, 1999), The Prestige (Christopher Nolan, 2006), and Arrival (Denis Villeneuve, 2016).11 But the winding corridor was always a guiding principle of Hollywood storytelling, with its long-standing appeals to suspense, curiosity, and surprise. Another cinema scholar who has examined how a film might disclose its developments over time is V.F.  Perkins. Perkins’s approach differs from Bordwell’s in various ways; for one thing, Perkins subscribes to the view that the causal chain is overly reductive. But the two share an interest in the ways a film might shape our experience moment by moment. In an essay on the concept of “worldhood,” Perkins writes, “Since the film’s characters are in a world, their knowledge of it must be partial, and their perception of it may be, in almost any respect, distorted or deluded.”12 A world must be understood from 9  David Bordwell, “The Classical Hollywood Style, 1917–1960,” The Classical Hollywood Cinema: Film Style and Mode of Production to 1960, by David Bordwell, Janet Staiger, and Kristin Thompson (New York: Columbia University Press, 1985), 13. 10  Bordwell, “The Classical Hollywood Style,” 37. 11  On recent puzzle films and twist films, see David Bordwell, The Way Hollywood Tells It: Story and Style in Modern Movies (Berkeley: University of California Press, 2006), 73–82. 12  V.F. Perkins, “Where Is the World? The Horizon of Events in Movie Fiction,” in Style and Meaning: Studies in the Detailed Analysis of Film, ed. John Gibbs and Douglas Pye (Manchester: Manchester University Press, 2005), 26.



a point of view, and the individual point of view is always limited. Crucially, Perkins extends this insight outward: “That applies to us, too, as observers of their world and their understandings.”13 No matter how informative the storytelling, there are aspects of the film’s story-world that will remain forever unknown to us. This obscurity is not a flaw but a simple fact of world-making that the artful filmmaker may turn to advantage by choosing with care which details to disclose and when to disclose them. Bounded by a frame, cinematic representation is always selective and fragmentary: “We are offered an assembly of bits and pieces from which to compose a world.”14 The spectator uses this sequence of fragments to make sense of the film’s world—a process that is always partial and in flux. All along, the spectator is guided by an awareness that the film is a purpose-built construction. Because a film represents a world, we may always ask, “Why is the movie, now, showing us this and not that?” Watching a narrative film is a deeply temporal experience, not just because the film takes time to pass through the projector, but also because the images onscreen provides a sequence of disclosures—disclosures that help us make sense of another sequence, that of the story-world.

The Modality of Narrative To understand a simple causal chain, you need to understand what happens and why. But most films offer a more complicated pattern, engaging our interest by asking us to consider what might happen—whether it ends up happening or not. We might think of this aspect of narrative as an appeal to “modal” thinking. How might a narrative engage our understanding of possibilities? One strategy is to rely on characters to open up this more expansive view. Marie-Laure Ryan defines embedded narratives as “story-like constructs contained in the private worlds of characters.”15 When a character hopes that something will happen, the reader understands the target of their hopes as a “virtual event.” Hopes, fears, beliefs, doubts—any of these private feelings may generate a virtual event. Some of those events may indeed become actualized in the world of the story (as when a character’s worst fears are realized), but others may not (as when a fear proves unfounded). Either way, the virtual event serves to shape our experience of the narrative as it unfolds in time, orienting us toward possibilities in the past, present, and future.16 Somewhat differently, Gerald Prince has proposed the category of the “disnarrated,” referring to “events that do not happen but, nonetheless, are referred to (in a negative or hypothetical mode) by the narrative text.”17 In a novel, a narrator might explain  Perkins, “Where Is the World?”, 26.  Perkins, “Where Is the World?”, 26. 15  Marie-Laure Ryan, Possible Worlds, Artificial Intelligence, Narrative Theory (Bloomington: Indiana University Press, 1991), 156. 16  I discuss Prince and Ryan elsewhere in Patrick Keating, “Narrative Dynamics and the Competitive Reality Show,” Storyworlds 5 (2013): 55–75. 17  Gerald Prince, “The Disnarrated,” Style 22, no. 1 (1988): 2. 13 14



a series of hypothetical events while pointing out that they did not happen in the fictional world. Such disnarration may sharpen our understanding of an actual event’s significance. Note that Prince’s category overlaps with but differs from Ryan’s. For Ryan, the virtual event requires some sort of mental state, as when a character believes or fears that something might happen. For Prince, an event may be disnarrated—presented as a hypothetical but unrealized possibility—whether a character is aware of the possibility or not. The functionalist approach I have been advocating places special emphasis on the logic of possibilities. Discussing the dynamics of prospection, retrospection, and re-cognition, Sternberg writes, “Let me just point out their inherent modality. Arising from a gapped future or past, all three dynamics entail multiple (ambiguous, uncertain, hypothetical, reversible) gap-filling, necessarily a matter of ‘possibility or probability’ rather than ‘fact.’”18 A narrative is modal because it appeals to our understanding of what might happen (or might have happened), not just what does happen. When we wonder what will happen next, we are in the grips of an emotionally engaging story. To be sure, some gaps are temporary rather than permanent. When a detective solves the crime at the end of a classically constructed mystery, uncertainty is replaced by certainty. But the finality of the ending may feel all the more satisfying because of all the uncertainty that has come before.19 Certain cinematic genres and techniques appeal to modal thinking explicitly. For instance, David Bordwell has analyzed “forking paths” films like Sliding Doors (Peter Howett, 1998) and Too Many Ways to Be No. 1 (Ka-Fai Wai, 1997), which represent alternative timelines without necessarily establishing which one (if any) is the “true” timeline.20 (As Bordwell points out, many of the films are less radical than they appear, offering various clues to suggest that the last timeline should be given priority.) Other films depict characters’ imaginings as if they were actual events in the story-world. In High Fidelity (Stephen Frears, 2000), one scene shows the protagonist Rob (John Cusack) meeting his insufferably pretentious rival Ian (Tim Robbins). Unexpectedly, Rob starts to insult Ian—but then the film returns to an earlier moment in the conversation, and we realize that Rob simply imagined insulting him. The film then repeats the joke, first showing Rob attempting to punch Ian and then showing Rob killing Ian with the help of his friends. Each time, the film returns to the initial conversation, marking Rob’s increasingly extreme behavior as an increasingly fanciful bit of wish fulfillment. This technique has become a comedy-film cliché, but more dramatic films have used such virtual events to profound effect. At the end of Spike Lee’s 25th Hour (2002), there is an extraordinary  Meir Sternberg, “If-Plots: Narrativity and the Law-Code,” in Theorizing Narrativity, ed. John Pier and José Ángel García Landa (Berlin: Walter de Gruyter, 2008): 34. 19  As Eyal Segal explains, “A successful conclusion of the investigation thus resolves the curiosity gaps about the crime mystery—and simultaneously the suspense gaps regarding the course of the investigation.” See Segal, “Closure in Detective Fiction,” Poetics Today 31, no. 2 (Summer 2010): 167. 20  David Bordwell, “Film Futures,” Poetics of Cinema (New York: Routledge, 2008), 171–187. 18



sequence in which a father (Brian Cox), hoping that his son Monty (Edward Norton) will not report to jail, delivers a long monologue explaining how the son might go on to live a productive life if he runs away. The film illustrates the father’s words quite vividly, revealing the sights and sounds of a future that will not happen. Indeed, the sequence goes on for so long that a spectator might wonder, “Wait—this is a fantasy, right?” Similarly, at the end of La La Land (Damien Chazelle, 2016), the two protagonists, now living separate lives, share a fantasy of how their love might have developed differently. The stylized set design, reminiscent of ballet sequences from 1950s musicals, clearly marks the sequence as a fantasy, and yet the musical number goes on for so long that a spectator might doubt its status, at least for a moment. In Prince’s terms, Lee’s film disnarrates the future: Chazelle’s film, the past. Although representations of imagined events are not hard to find, they remain exceptions to the norm whereby mainstream films simply depict what happens. However, we should not dismiss modality as a curiosity, relevant only for films that feature explicit “what if” sequences. Applying the functionalist model to cinema, Inbar Shaham has shown how Hollywood genres deploy patterns of forecast, enactment, and report in distinctive ways, as when the heist film offers a detailed forecast of a future theft, generating a set of predictions that will shape our understanding of the heist itself, which may or may not go according to plan.21 More broadly, any narrative film may mandate its spectators to consider a wide range of possible outcomes, whether they are actualized or not. Consider Carol Reed’s 1949 film The Third Man, based on a screenplay by Graham Greene. The film offers an array of characters with conflicting goals and shifting beliefs. To borrow Ryan’s terminology, each articulation of a goal or belief produces a “virtual event.” Holly (Joseph Cotten) arrives in Vienna expecting to meet Harry (Orson Welles). At first, his expectation is thwarted; later, it is realized. The military policeman Calloway (Trevor Howard) aims to convince Holly that Harry was guilty of involvement in a murderous racket. Calloway fails to accomplish his goal at first, but then he succeeds. Harry’s former lover Anna (Alida Valli) believes that Harry is dead. We think that she is correct, but then we learn that she is wrong. Then Harry is killed, and she forms a new belief that Harry is truly dead—a belief we now know to be correct. To understand any of these events as they unfold in time, we must understand a great deal that goes beyond the “objective” events, because we must understand what the characters think will happen, what the characters believe about what has happened, and what could end up happening whether the characters expect it to happen or not. The crucial scene when Holly learns that Harry is alive evokes surprise in the narratological sense of the term. In the story-world, with its flow of developments, Harry was alive the whole time. In the filmic sequence, with its flow of disclosures, the truth of Harry’s survival remains concealed for half the film. 21  Inbar Shaham, “The Structure of Repetition in the Cinema: Three Hollywood Genres,” Poetics Today 34, no. 4 (Winter 2013): 442.



Harry’s sudden appearance mandates an act of re-cognition—a quick reshuffling of previous assumptions. Significantly, Harry’s appearance does not come as a total shock. Prior to the moment of revelation, the film has offered several hints, encouraging the spectator to think of Harry’s survival as a possibility. These hinting scenes show how a filmmaker might take advantage of cinematic resources like composition and lighting to suggest a possibility without actually showing it. For instance, when the still unrevealed Harry sneaks into a doorway, the camera is positioned up high and far away. Centering ensures that we notice the mysterious man, but angle and scale keep his identity hidden. A few moments later, a cat approaches and plays with Harry’s shoelaces, thereby providing another clue to the mysterious man’s identity, since we know from a previous scene that the cat liked only Harry. Though probable, the possibility that the man is Harry remains unconfirmed; because the cat is filmed in close­up, the man’s face remains off-screen. A few moments later, a drunken Holly tries to taunt the man in the doorway. Cinematographer Robert Krasker’s lighting is carefully arranged, allowing us to see the man’s shoes, while keeping his face in total darkness. Finally, a neighbor turns on a light, conveniently illuminating Harry’s face. The camera dollies in, unmistakably directing our attention to a long-withheld fact: Harry is alive! Camerawork, dialogue, framing, lighting—all have worked together in a play of concealment and revelation, hinting at a possibility before finally offering confirming proof. Noël Carroll’s theory of erotetic narration provides a useful way of analyzing films that unfold in this teasing way. Writing about popular movies, he writes, “At one level, the plot is a network of events and states of affairs held together by the cement of causation. Yet, at another level—namely, the level of rhetorical address—a typical movie narrative is a network of questions and answers, where the questions are self-generated but then finally resolved.”22 Early scenes often generate macro-questions sustained over a large part of the film. When Calloway tells Holly that Harry was a ruthless criminal, the information generates questions about the past (Was Harry guilty?) and the future (Will Holly be able to vindicate his friend?). About halfway through the film, the question is resolved decisively. Holly and Anna both come to agree that Harry was guilty. The question is closed, for them and for us. But the surprise revelation that Harry is alive introduces a new macro-question: Will Holly help Calloway apprehend Harry? This question sustains our interest right up to the climactic scene, when Holly kills Harry. At a narrower level, individual scenes might generate micro-questions, answered soon after they are posed.23 When Holly runs away from a group of henchmen, we ask, “Will Holly escape?” This question generates suspense for about a minute of screen time, as long as it takes for Holly to elude his would-be captors. Together, the macro-questions and the micro-questions keep us watching from beginning to end.

 Noël Carroll, The Philosophy of Motion Pictures (Malden, MA: Blackwell, 2008), 136.  Carroll, The Philosophy of Motion Pictures, 137–138.

22 23



All of this might make it seem like a mainstream movie is little more than an easily solved puzzle, but a skillful film may use its question-and-answer structure to generate thematic richness. The Third Man introduces the question “Will Holly help Calloway apprehend Harry?” to develop a contrast between Holly and Anna. Both characters care deeply about Harry, but they respond to the news about Harry’s guilt differently. Anna insists that her love for Harry has not changed because of what she has learned about him, and she refuses to betray her beloved by helping Calloway. Anna’s steadfastness qualifies any admiration we might feel regarding Holly’s decision. The film gives us good reasons to hope that Calloway will convince Holly to help, reminding us that Harry’s crime was lethal and deserving of punishment. And yet the film also gives us good reasons to criticize Holly, whose mercy killing of Harry stands in such sharp contrast to Anna’s abiding love. In this way, the film has deployed its question-and-answer structure to raise difficult, possibly unanswerable questions about what we owe to someone we love. Not all films deploy twists and turns in the manner of The Third Man, but the modal model is broadly applicable, even in much simpler films. A minimal story about a character with a straightforward goal appeals to a viewer’s understanding that the character may or may not achieve it. A predictable film relies on us to make predictions, selecting outcomes from multiple possibilities. Because a film unfolds in time, future revelations (including revelations about the past) are always uncertain.

Serial Narratives All of my examples so far have involved feature films with defined endings, but there are many works of moving-image art that extend their stories across multiple episodes. Indeed, it could be argued that serial storytelling has become the dominant norm in American media industries. Just as many of the most prestigious television works are long-running series, such as The Sopranos (2000–2007) or The Good Wife (2009–2016), many of the most profitable big-­screen films are sequels, as in the Harry Potter series or the Marvel Cinematic Universe. In an elegant analysis of the prime-time serial (PTS) format, Michael Z. Newman argues that television producers have responded skillfully to a distinctive set of constraints: “Given the incentive to produce narratives that engage audiences week after week, television has developed a powerful mode of storytelling.”24 The solutions involve a distinctive approach to time: “Looking at the PTS’s narrative form, we may consider it to have three storytelling levels for analysis: a micro level of the scene or ‘beat,’ a middle level of the episode, and a macro level of greater than one episode, such as a ­multi-­episode arc.”25 At a small-scale level, a scene must deploy narrative infor24  Michael Z. Newman, “From Beats to Arcs: Toward a Poetics of Television Narrative,” The Velvet Light Trap 58 (Fall 2006): 17. 25  Newman, “From Beats to Arcs,” 17.



mation in an artfully redundant way—repeating just enough information to ensure that novice viewers of the program will not get confused, without becoming so repetitive that devoted fans will get bored. At the next level, an episode must provide some sense of closure, while leaving enough storylines dangling to keep the viewer tuning in for more. At the highest level, a multiepisode arc must reward long-time viewers for their investment in the show’s characters, without compromising the interest of particular scenes or episodes. For instance, The Good Wife develops multi-season arcs concerning Alicia Florrick’s (Juliana Margulies) increasingly successful career as a lawyer and her increasingly troubled relationships with her boss Will (Josh Charles) and her long-­unfaithful husband Peter (Chris Noth). Individual episodes center on the specific legal cases that Alicia’s firm wins or loses, while individual scenes show Alicia confronting and overcoming local obstacles along the way. Throughout, the show introduces gaps to play on our feelings of narrative interest, producing both long-term suspense (Which man will Alicia choose?) and short-term suspense (Will Alicia win this week’s case?), as well as various forms of curiosity (Did Alicia’s client commit the murder or not?) and surprise (I thought Alicia was losing the case; I didn’t realize she had devised the perfect plan to win!). As we have seen, a typical popular movie achieves closure by answering all of its questions (or, at least, the most salient ones) by the end of the film.26 Certain highly episodic television shows adopt a similar structure. The characters may remain the same from season to season, but each episode stands more or less on its own, as in mysteries that introduce new suspects and new solutions every week. By contrast, the highly serialized form of the soap opera refuses closure by delaying answers systematically.27 As Newman points out, the prime-time serial is a mixed form, “a hybrid of episodic dramas and serials such as soaps and miniseries.”28 A beat might introduce a micro-question that is answered after the commercial break or develop a macro-question that is answered at the end of an episode or sustain an even larger macro-question that is stretched over multiple episodes or seasons. Again, it is useful to think of questions and answers in relation to other narrative functions, such as characterization, lest we turn TV shows into mere guessing games. Unlike a two-hour movie, a multi-season television show can examine a character’s psychological growth (or decline) in extraordinary detail and nuance, sometimes approaching the complexity of a novel. In any given scene, our understanding of a character’s behavior may be enriched by our understanding of the character’s past. While a series offers writers an opportunity to examine a character in depth, it also poses a significant storytelling ­challenge, requiring the writer to balance the interests of long-time viewers 26  Noël Carroll, “The Power of Movies,” Theorizing the Moving Image (New York: Cambridge University Press, 1996), 89. 27  Noël Carroll, “As the Dial Turns: Notes on Soap Operas,” Theorizing the Moving Image (New York: Cambridge University Press, 1996), 121. 28  Newman, “From Beats to Arcs,” 16.



with the needs of novices seeking to comprehend basic information. With this in mind, Jason Gendler has examined how television writers create psychologically rich situations “through the information established within an individual episode itself (what I call episodic enrichment), versus the degree to which rich situations are created by relying on information accrued over the course of previous episodes (serial enrichment).”29 He concludes that highly serialized shows like Mad Men actually rely on episodic enrichment more than we might suspect, along with a third category (blended enrichment), whereby our understanding of a character’s psychology is enriched by earlier scenes within the episode as well as earlier scenes within the show as a whole. Many of the problems of serial storytelling are not specific to film or television but appear throughout the more general category of “serial fictions.” Andrew McGonigal has used this term to refer to a class of fictions “whose generation and reception is (i) connected in an interesting way to distinct, relatively discontinuous episodes of installments that are (ii) appropriately construable as taking place in a single fictional world.”30 Examples include long-running comic books, such as The Amazing Spiderman; novels originally released in a serialized format, such as The Pickwick Papers; and the stories of Sherlock Holmes. Within this broad category, there are a number of distinctions: for instance, between fictions that move toward completion and fictions that remain open-ended or between fictions that indicate the temporal ordering of specific episodes and fictions that leave important facts about temporal order indeterminate.31 McGonigal is particularly interested in the problem of fictional truth—a problem that certain serial fictions raise by introducing troubling contradictions over the course of the series. In the original episode of Star Wars (1977), also known as Star Wars IV: A New Hope, there is considerable evidence that Luke is not the son of Darth Vader, most notably the fact that the trustworthy Obi-Wan Kenobi explicitly tells Luke that Vader killed Luke’s father. In the sequel, The Empire Strikes Back (1980), Darth Vader tells Luke that he is Luke’s father, and the remaining films in the series (both sequels and prequels) ask us to assume that Vader’s statement is true. Imagine a spectator in 1977 who watches Star Wars and states that Luke is Vader’s son. Now imagine a spectator in 2007 who watches Star Wars and states that Luke is Vader’s son. McGonigal’s intuition—which I share—is that the first statement is somehow worse than the second. His own solution appeals to relativism, making no appeal to an “absolute” fictional world. “Whether a given proposition is true-according-to-the-fiction,” he writes, “is something that always is implicitly relative to a context of assessment.”32 Other scholars have proposed different solutions, such as Ben Caplan’s “work contextualism,” whereby “the 29  Jason Gendler, “The Rich Inferential World of Mad Men: Serialized Television and Character Interiority,” Projections 10, no. 1 (Summer 2016): 40. 30  Andrew McGonigal, “Truth, Relativism, and Serial Fiction,” British Journal of Aesthetics 53, no. 2 (April 2013): 165. 31  McGonigal, “Truth, Relativism, and Serial Fiction,” 165. 32  McGonigal, “Truth, Relativism, and Serial Fiction,” 178.



content of the movie changes across contexts,” and Lee Walters’s “invariantism,” whereby “current installments of a serial fiction represent defeasible evidence for what is true according to the maximal fiction of which it is a part.”33 My own view is that fictional worlds, whether serialized or not, often contain inconsistencies, precisely because fictional worlds are means toward rhetorical ends. As a standalone work, Star Wars seeks to establish Darth Vader’s status as a formidable villain. As a sequel pointing ahead to at least one more successor, The Empire Strikes Back seeks to enrich Luke’s characterization, to establish a macro-question that will stretch on into the next episode, and to produce a moment of surprise (or, in the case of my 10-year-old self, utter astonishment) by forcing us to re-cognize aspects of Star Wars that we thought we understood. Even today, if we view the original film in light of its own rhetorical goals, we may reasonably conclude that Luke is not Vader’s son. If we view Star Wars in light of the larger series’ rhetorical goals, then we may reasonably conclude that he is. Questions about what happens in the story-world are complicated further in works of “transmedia” storytelling. As Henry Jenkins explains, The Matrix trilogy (1999–2003) confronted the challenge of telling its story across three distinct films, two of which were made several years after the first was completed. Even more remarkably, the franchise told portions of its story through various supporting materials, such as video games and comic books. These supporting materials depicted crucial events in the story-world’s causal chain. It simply was not possible to understand the films’ story-world fully by seeing the films alone; a committed spectator needed to seek out the supporting materials, as well.34 In this case, the directors of the films (the Wachowskis) were closely involved with the production of the supporting materials. But what should we make of unauthorized expansions upon existing fictional worlds, such as fan fiction? It is increasingly common for fans of film and television shows to generate their own narrative content, in the form of short stories or videos, re-imagining characters in creative and sometimes radical ways. As Marie-Laure Ryan points out, such works bear an interesting relationship to authorized or “canonical” works of serial fiction. A work of fan fiction may project a fictional world that is distinct but related to the fictional world of the canonical text. By contrast, a later episode of a television series unambiguously projects the same story-world as earlier episodes do.35 Consider a hypothetical example. In the fifth season of The Good Wife, the beloved character Will dies unexpectedly. If a fan were to write and film an episode of The Good Wife in which Will survives, the fan’s 33  Ben Caplan, “Serial Fiction, Continued,” British Journal of Aesthetics 54, no. 1 (January 2014): 73; Lee Walters, “Serial Fiction, the End?,” British Journal of Aesthetics 55, no. 3 (July 2015): 337. Thanks to Andrew Kania for pointing me toward this interesting debate. 34  Henry Jenkins, “Searching for the Origami Unicorn: The Matrix and Transmedia Storytelling,” Convergence Culture: Where Old and New Media Collide (New York: New York University Press, 2008), 95–134. 35  Marie-Laure Ryan, “Transfictionality across Media,” in Theorizing Narrativity, ed. John Pier and José Ángel García Landa (Berlin: Walter de Gruyter, 2008), 388–392.



intervention would not change the story-world of the original show. My intuition is to say that Will would still be dead. All this talk of worlds might sound overly metaphysical, losing sight of the functionalist perspective I have been championing so far. To put the case in more functionalist terms, let us start with the assumption that the story-world of The Good Wife is a construction—a construction that is always subject to revision during viewing in light of the show’s flow of disclosures. When I, a devoted fan of the show, watch Season 5, Episode 15 (the episode in which Will dies), I use information from previous episodes to make sense of the twists and turns in this particular episode. At the same time, I use information from the episode I am watching to revise my understanding of previous episodes (pondering whether the circumstances of Will’s death cast his previous behavior in a new light) and make predictions about future episodes (wondering how Alicia will react to Will’s death). But suppose I were to watch the hypothetical fan video in which Will survives. As with the real episode, I would be warranted in using information from previous authorized episodes to make sense of the fan video’s twists and turns. However, I would be reluctant to use the fan video to warrant significant re-cognition of previous episodes; if the fan video were to claim that Will and Alicia had been married all along, that would not make it so for the canonical episodes. Nor does the fan video warrant prospection toward future authorized episodes, though it might warrant prospection toward future fan fictions. As this example suggests, the concept of the “canonical” text may raise another set of problems—problems regarding the authorship of fictional works. I consider the problem of authorship in the next section.

Narrators, Authors, Tone, and Point of View In the philosophy of film narrative, one of the liveliest debates concerns the status of the cinematic narrator, a concept that may be distinguished from the figure of the character narrator. Many films, such as Julie & Julia (Nora Ephron, 2009) and The Shawshank Redemption (Frank Darabont, 1994), employ character narrators—fictional characters who recite their stories. The character narrator poses a distinct set of philosophical problems. In Julie & Julia, there are several passages of voice-over narration based on the letters, books, and blogs written by the main characters. Do these voice-overs express the internal thoughts of the characters writing the words or the internal imaginings of the characters reading them? In The Shawshank Redemption, Red (Morgan Freeman) narrates much of the story in a voice-over. To whom is Red speaking? Is he speaking to us? If so, how can a fictional character speak to real spectators? These are interesting questions, but the problem of the cinematic narrator raises a different set of puzzles. Here the question is whether or not the film itself has a narrator—some storytelling agency that organizes the ­pictures and sounds, including but going well beyond the voice-overs of Julie, Julia, and Red. Some scholars, such as Seymour Chatman, have argued that all



films have narrators, whether they have character narrators or not.36 Other scholars, such as David Bordwell, have argued against the idea of the cinematic narrator.37 As in literary theory, the question of the cinematic narrator raises related questions about authors, both real and implied. My own view is that the cinematic narrator points us toward a cluster of important issues regarding a film’s tone and point of view, but that we can usually explain those issues more efficiently by appealing to the (implied) author. For George M. Wilson, the real interest of these questions does not lie solely in the problem of the narrator; rather, it lies in the even more fundamental problem of what it is to see a fiction in a film.38 Specifically, he argues in favor of two closely related ideas, which he calls the Imagined Seeing Thesis and the Fictional Showing Hypothesis. He writes, “If, in watching a movie, viewers imagine seeing the narrative action on screen, then presumably they thereby imagine that the projected motion picture images they are watching are, in some way, ‘showing’ the narrative action to them.”39 The idea that spectators imagine seeing the action requires some additional imagination regarding the showing. Significantly, Wilson does not believe that viewers imagine seeing the fictional world directly, as if they were invisible observers looking over the shoulders of the characters. Instead, he argues that viewers “imagine themselves seeing those fictional constituents through the mediation of the onscreen moving images, images that fictionally have been transparently derived from the dramatized situations of the story.”40 Wilson’s emphasis on mediation is salutary because it encourages us to consider how photography, editing, and sound shape the viewer’s experience. The case for Imagined Seeing supports the case for Fictional Showing, which in turn meets a necessary condition for the claim that films have audiovisual narrators who fictionally recount the events. I do not have the space (or, frankly, the expertise) to make a case for or against the Imagined Seeing Thesis. Instead, I merely remark that Imagined Seeing is only one of many possible ways of theorizing what we see when we look at the cinematic image. Other theories, such as “recognition” accounts or “resemblance” accounts, might provide less immediate support for the Fictional Showing hypothesis, thereby making the case for cinematic narrators less pressing.41 For instance, in a response to Wilson’s book, Robert Stecker has proposed Richard Wollheim’s 36  Seymour Chatman, Coming to Terms: The Rhetoric of Narrative in Fiction and Film (Ithaca: Cornell University Press, 1990), 133. 37  David Bordwell, “Three Dimensions of Film Narrative,” Poetics of Cinema (New York: Routledge, 2008), 122. 38  George M.  Wilson, Seeing Fictions in Film: The Epistemology of Movies (New York: Oxford University Press, 2011), 126. 39  Wilson, Seeing Fictions in Film, 54. 40  Wilson, Seeing Fictions in Film, 89. 41  For a survey of approaches, see Catharine Abell and Katarina Bantinaki, “Introduction,” in Philosophical Perspectives on Depiction, ed. Abell and Bantinaki (New York: Oxford University Press, 2010), 1–23.



“seeing-in” theory as an alternative to Imagining Seeing, partly to cast doubt on related arguments concerning fictional showing and effaced narrators.42 However we theorize seeing movies, I think there is another reason why the narrator idea merits attention, even from its critics. As narrator skeptic Katherine Thomson-Jones explains, “The primary motivation for arguing that films and other kinds of narrative art always have narrators is the observation that a story is always told in a certain way. This accounts for the tone of the work, or the set of attitudes manifest in the way that characters and events are described or depicted.” After listing some of the ways a film might manifest such an attitude (for instance, through cinematography or editing), Thomson-Jones explains, “When we pick up on the attitudes manifest in a film’s style, we naturally want to assign these attitudes to someone. And if we want to assign them to someone inside the story, we assign them to a narrator.”43 This argument sounds like an endorsement of the cinematic narrator, but Thomson-Jones goes on to argue that the idea’s appeal is illusory, the result of a bias toward analogizing film with literature. If we think of film as analogous to theater, then the urge to look for a cinematic narrator becomes less acute.44 We might want to assign the film’s attitude to someone—but that someone need not be a figure “inside” the story-world. I am sympathetic to Thomson-Jones’s approach, which criticizes the overly literary notion of the cinematic narrator while acknowledging the value of an approach that helps us characterize the distinctive way a film’s story is told. The challenge is to develop an account of a film’s attitude without (or without necessarily) appealing to the figure of a narrator. A number of film scholars have addressed this challenge in recent years. In another thoughtful response to Wilson’s book, Douglas Pye argued that we can preserve Wilson’s nuanced approach to close analysis while attributing the relevant choices to a film’s author.45 Influenced by Perkins’s work on worldhood, Pye himself has written major works on the difficult concepts of “tone” and “point of view.” Regarding the former, he proposes four possible “axes” of tone: “Attitudes implied to the film’s subject matter; attitudes implied to the film’s audience; attitudes implied to the conventions the film employs or invokes; attitudes implied to the film as a film.”46 While appealing to our understanding of how a film’s world is represented, Pye wisely warns against making too sharp a distinction between “how” and “what.” The details of the world may express tone just as well as the ­techniques of its framing. In contrast to tone, “point of view” is more closely related to the kinds of access a film’s sequence of images and sounds may offer 42  Robert Stecker, “Film Narration, Imaginative Seeing, and Seeing-In,” Projections 7, no. 1 (Summer 2013): 147–154. 43  Katherine Thomson-Jones, “The Literary Origins of the Cinematic Narrator,” British Journal of Aesthetics 47, no. 1 (January 2007): 78. 44  Thomson-Jones, “The Literary Origins of the Cinematic Narrator,” 90. 45  Douglas Pye, “Seeing Fictions in Film,” Projections 7, no. 1 (Summer 2013): 137. 46  Douglas Pye, “Movies and Tone,” in Close-Up 02, edited by John Gibbs and Douglas Pye (New York: Wallflower Press, 2007), 30.



us. As Pye explains, “The idea that limits imposed on the spectator’s or reader’s access to the story are significant and highly variable is perhaps the central importance of the concept of point of view.”47 Considering a film’s point of view involves a consideration of the film’s patterns of access as they develop over time. Some films pattern their disclosures by “following” a single character from scene to scene. However, Deborah Thomas cautions against the temptation to associate a film’s point of view with that of a specific character. She writes, “A film’s point of view is clearly not reducible to that of the characters—or even a privileged character—within it, but includes an attitude or orientation toward the various characters (whether one of ironic detachment, sympathetic involvement, moral condemnation, or whatever) as well as some sort of epistemological relationship which is never precisely one of identity (where we see and know precisely what they do, nothing more nor less), and a spatial positioning which is not identical with theirs.”48 The term “orientation” captures the idea well. One orients oneself with respect to something else. The patterns in the film guide us toward adopting a particular posture toward the characters. Consider an example from Harry Potter and the Prisoner of Azkaban (2004), directed by Alfonso Cuarón. The film’s organization keeps us tightly attached to Harry (Daniel Radcliffe), who appears in virtually every scene, and the sound design and camerawork occasionally represent his subjective experiences. For much of the film, Harry (Daniel Radcliffe) worries that the escaped convict Sirius Black (Gary Oldman) plans to kill him. In a crucial scene, Harry and Hermione (Emma Watson) find an injured Ron (Rupert Grint), who points to the other side of the room. The film cuts to a new shot, dollying along a set of dog’s footprints before revealing Sirius (who can turn into a dog), standing menacingly behind a door. It is as if we are seeing Sirius through Harry’s eyes, focusing his attention on the surprising threat. However, it would be an oversimplification to say that the shot is simply taken from Harry’s point of view, as there are several factors that complicate the issue considerably. First, the shot does not represent Harry’s exact position in space. The camera moves toward Sirius, but Harry stands still. Second, Harry is not the only character looking toward Sirius. When Ron points at Sirius himself, Harry and Hermione turn together to look at the footprints on the ground; the subsequent shot represents their shared point of view. This is a significant choice, given the film’s larger strategy of comparing Harry and Hermione, each of whom has strengths and weaknesses that the other lacks. Third, there is the obvious but important fact that we cannot see Sirius from Harry’s point of view because Harry is in the Shrieking Shack and we are sitting in a theater looking at pic47  Douglas Pye, “Movies and Point of View,” Movie 36 (2000): 2. Pye makes this comment in the context of his discussion of George Wilson, Narration in Light: Studies in Cinematic Point of View (Baltimore: Johns Hopkins University Press, 1986). 48  Deborah Thomas, Beyond Genre: Melodrama, Comedy, and Romance in Hollywood Film (Moffat, UK: Cameron and Hollis, 2000), 20.



tures of Harry and his friends—who, in any case, are not real. These pictures provide us with carefully managed access to the story-world and its characters. Harry believes that Sirius is a mortal threat, but we have an entirely different relationship to this mysterious figure. We may already suspect that Sirius is not what he seems, and we almost certainly assume that Harry will survive the confrontation, even if we do not yet know how. Rather than fear that Harry will die, our real concern is that Harry will give in to his darker urges, reacting to the threat of Sirius with violence. The emphasis on Hermione’s shared point of view balances our Harry-centered fears with the Hermione-centered hope that she will help Harry survive the ordeal with his good nature intact. Rather than put us directly in Harry’s position, the film pushes us to take a more complex orientation toward the protagonist by adopting an understanding but critical attitude toward Harry’s confrontational response. So far, I have been describing this scene using a depersonalized vocabulary, suggesting that “the film” directs our attention and shapes our orientation. I think that there are merits to this depersonalized approach. The film does not sympathize with Harry and Hermione; guided by the film, we do. We experience a complex blend of sympathies in response to the film, with its sequencing of pictures and sounds. Similarly, David Bordwell prefers to speak of a film’s “narration” (or, sometimes, “narrative dynamics”), setting aside talk of narrators and implied authors as unnecessary.49 Even more broadly, Pye prefers to speak of the “movie,” a product informed by complex intentions—not just the telling of stories but also the building of worlds and the elaboration of ideas.50 I share the skepticism of Thomson-Jones, Bordwell, and Pye about the value of the term “cinematic narrator” as a tool for film analysis. However, two caveats are in order. First, though the term “narrator” strikes me as misleadingly anthropomorphic, the theory of film narrative must start with the assumption that the movie mediates our (oft-changing) understanding of the story-world. Analyzing the movie’s patterns of mediation, whether we attribute those patterns to a narrator or not is a step toward understanding how the movie produces its effects. Second, for all my hesitations about the personalized connotations of the word “narrator,” it seems fair to say that some useful phrases will necessarily appeal to some kind of storytelling agency. For instance, I have argued that the film has “a strategy of comparing Harry and Hermione.” The appeal to strategy is an appeal to purposes. The question is: Whose purposes? Some would answer this question by citing a single author, such as the director Alfonso Cuarón or the author of the original book, J.K. Rowling. Others would appeal to multiple authors, on the grounds that most films are made by a collaborative group of artists and technicians. Still others might appeal to the notion of the “implied author,” understood as a construct of the text, or perhaps as an authorial persona that is manifest in the text. Berys Gaut has surveyed these possibili Bordwell, “Three Dimensions of Film Narrative,” 122.  Pye, “Seeing Fictions in Film,” 136.

49 50



ties, and more, with care.51 Like Gaut, I believe that most mainstream films have multiple authors, a category that may include directors, writers, actors, cinematographers, and more.52 As a historian, I spend most of my time studying real authors in collaborative and sometimes competitive relational dynamics, and I worry that the term “implied author” places too much emphasis on unity to account for real films that manifest contradictions and compromises. Nevertheless, I think the term “implied author” identifies something important about our engagement with narrative films because it points us to purposes that are, literally, implied by those films. Let us consider the logic of purposes more carefully. Recently, some philosophers of film have proposed a two-part model, distinguishing the logic of the film’s world from the purpose-driven logic of the film’s design. For instance, Gregory Currie has argued that we may approach a narrative from an internal or external perspective. Internally, “we speak and think directly of the characters and events in the story”; externally, “we see a vehicle, something that represents a sequence of events in virtue of the activity of an agent we call the author.”53 When a character behaves strangely, we might make sense of the behavior by asking about the character’s goals. Alternatively (or, in addition), we might make sense of the behavior by considering how the passage furthers the author’s goals. These ideas echo some important discussions within the rhetorical-­ functionalist framework. According to Meir Sternberg and Tamar Yacobi, when seeking to make sense of an unexpected feature of a text, a reader may appeal to several mechanisms of integration. One option is to treat the unexpected feature as a mistake. Perhaps we assume that the text was misprinted— or that the film was projected out of focus. In such a case, we are choosing not to look for a purpose behind the feature.54 But sometimes we do look for purposes. Within the broader framework of integration, the theory of motivation seeks to explain how we make sense of a text while “regulated by a sense of the text’s purpose.”55 As Sternberg explains, a work of fiction (be it a novel, painting, or film) may be structured according to “a pair of motivational logics, mimetic (world-like, referential, fictional) as against aesthetic (rhetorical, ­communicative, functional).”56 Among this set of pairs, I find the binary “fic51  See the chapter on cinematic authorship in Berys Gaut, A Philosophy of Cinematic Art (New York: Cambridge University Press, 2010), 98–151. 52  Gaut, A Philosophy of Cinematic Art, 125. 53  Gregory Currie, Narratives & Narrators: A Philosophy of Stories (New York: Oxford University Press, 2010), 49. See also Murray Smith’s discussion of the “referential” and “formal” aspects of a film, in Smith, “On the Twofoldness of Character,” New Literary History 42, no. 2 (Spring 2011), 289. 54  For a summary of the theory of integration, see Meir Sternberg and Tamar Yacobi, “(Un)reliability in Narrative Discourse: A Comprehensive Overview,” Poetics Today 36, no. 4 (December 2015), 402–412. 55  Meir Sternberg, “Mimesis and Motivation: The Two Faces of Fictional Coherence,” Poetics Today 33, no. 3–4 (Fall–Winter 2012): 413. 56  Sternberg, “Mimesis and Motivation,” 368.



tional vs. functional” to be particularly apt, suggesting the contrast between explanations that appeal to features of the fictional world and explanations that appeal to the work’s purposes. Crucially, the two logics are not equal: “The internal tensions between [the] two modes—as alternative, ‘mimetic’ vs. ‘aesthetic,’ fictional vs. functional, logics of patterning—are always resolvable and always in favor of the second mode, by way of a higher teleological explanation.”57 Here, the word “teleological” indicates that we make sense of a curious feature by treating it as part of a larger design, seeing the text as a purposefully made construction. The traits of the story-world (its objects, its characters, its events) are understood as means to rhetorical ends. To return to my earlier example from The Third Man, consider the moment when the light turns on, suddenly revealing Harry Lime’s presence in the shadowy doorway. Within the story-world, the moment is carefully motivated. Because Holly was sad, he got drunk. Because Holly was drunk, he started yelling at the mysterious man across the street. Because Holly was yelling, a neighbor turned on a light. Because a neighbor turned on a light, Harry’s identity was revealed. Behind these fictional motivations there lie functional motivations—most notably, the purpose of delivering a major surprise, requiring spectators to reorganize their understanding of previous events. Although these particular features are doubly motivated, certain choices regarding the film’s camerawork are motivated functionally but not fictionally. Within the fictional world, there is no particular reason why the camera should have framed the cat so tightly; after all, there is no camera in the fictional world at all. The framing choice makes sense, not in light of the film’s world (where the camera does not exist) but in light of the film’s purpose, hinting at Harry’s presence before making the decisive revelation. These hints shape our attitude toward Holly, allowing us to remain one step ahead of the film’s likable but foolish protagonist, who does not see the revelation coming at all. This two-tiered model of motivation also explains cinematic techniques like camera movement. In Flesh and the Devil, the dolly-in toward the husband’s clenched fist does not represent anyone’s motion within the fictional world; no one is moving toward the hand. A Hollywood filmmaker might call the movement “unmotivated,” but if anything the (functional) motivation for the camera movement is overly obvious, as if the filmmakers were calling out to the audience, “Look here! See how angry this man is!” By contrast, the movement toward Sirius in Harry Potter and the Prisoner of Azkaban is doubly motivated. Why the sudden movement toward Sirius? Because Harry and Hermione have focused their attention on the threat. But also for a purpose: to direct our attention to a plot point. Or, more fully, for a multileveled purpose: to direct our attention to a plot point through the mediation of Harry and Hermione, thereby deepening our attachment to these two characters who are experiencing this threat together.

 Sternberg, “Mimesis and Motivation,” 411.




Sternberg’s theory of motivation presupposes an implied author. So, too, does Currie’s theory of “internal” and “external” explanations.58 My own view is that the “implied author” concept does important work, even if the term strikes me as a little too anthropomorphic, conjuring up the image of a magical storyteller who doesn’t really exist. In defense of a less anthropomorphic version of the concept, it can be useful to remember that no one, to my knowledge, is literally claiming that a mysterious being called the “implied author” actually made the film. If anything, the causal relationship runs the other way. The implications are in the film; the movie implies a (shifting) set of purposes as it unfolds in time. Of course, a team of real people made the film, guided by purposes that may have been in unison or in conflict; the word “implied” does the work of reminding us that we are not in contact with the real authors. We are in contact with the film, and we make sense of the film by considering its implications. H. Porter Abbott suggests the term “inferred author,” a term that nicely captures the fact that “we often differ with each other (and no doubt the author as well) in the views and feelings we attribute to the implied author.”59 Whichever term we use, the implied author is understood here as a construction, always subject to debate.

Conclusion There are many issues in the philosophy of film narrative that I have been unable to address here, such as unreliable narration, interactive narration, and the problem of imagination, to name a few.60 Instead, I hope to have offered a perspective from which to address such problems—a perspective that places special emphasis on the temporality of narrative. Whether we are watching a film or a TV show, we experience the moving-image artwork over time. When the moving image tells a story, it represents a world that develops over time, as well. To understand how narrative works in moving-image media, we must consider the ever-shifting relationship between these two sequences.

Bibliography Abbott, H. Porter. 2008. The Cambridge Introduction to Narrative. 2nd ed. New York: Cambridge University Press. Abell, Catharine, and Katarina Bantinaki. 2010. Introduction. In Philosophical Perspectives on Depiction, ed. Catharine Abell and Katerina Bantinaki, 1–23. New York: Oxford University Press.  Sternberg and Currie differ on other points, for instance, on the need for a narrator as mediator. 59  H. Porter Abbott, The Cambridge Introduction to Narrative, 2nd ed. (New York: Cambridge University Press, 2008), 85. 60  For an introduction to these debates, see the following: on unreliable narration, Gregory Currie, Image and Mind: Film, Philosophy, and Cognitive Science (New York: Cambridge University Press, 1995), 260–280; on interactive narration, Gaut, A Philosophy of Cinematic Art, 224–243; on (and against) imagination, Derek Matravers, Fiction and Understanding (New York: Oxford University Press, 2014), 146–157. 58



Bordwell, David. 1985. The Classical Hollywood Style, 1917–1960. In The Classical Hollywood Cinema: Film Style and Mode of Production to 1960, ed. Janet Staiger Bordwell and Kristin Thompson, 1–84. New York: Columbia University Press. ———. 2006. The Way Hollywood Tells It: Story and Style in Modern Movies. Berkeley: University of California Press. ———. 2008a. Film Futures. In Poetics of Cinema, 171–187. New York: Routledge. ———. 2008b. Three Dimensions of Film Narrative. In Poetics of Cinema, 85–134. New York: Routledge. Caplan, Ben. 2014. Serial Fiction, Continued. British Journal of Aesthetics 54 (1, January): 65–76. Carroll, Noël. 1996a. As the Dial Turns: Notes on Soap Operas. In Theorizing the Moving Image, 118–124. New York: Cambridge University Press. ———. 1996b. The Power of Movies. In Theorizing the Moving Image, 78–93. New York: Cambridge University Press. ———. 1996c. Toward a Theory of Film Suspense. In Theorizing the Moving Image, 94–117. New York: Cambridge University Press. ———. 2008. The Philosophy of Motion Pictures. Malden: Blackwell. Chatman, Seymour. 1975. Towards a Theory of Narrative. New Literary History 6 (2, Winter): 295–318. ———. 1990. Coming to Terms: The Rhetoric of Narrative in Fiction and Film. Ithaca: Cornell University Press. Currie, Gregory. 1995. Image and Mind: Film, Philosophy, and Cognitive Science. New York: Cambridge University Press. ———. 2010. Narratives & Narrators: A Philosophy of Stories. New  York: Oxford University Press. Gaut, Berys. 2010. A Philosophy of Cinematic Art. New York: Cambridge University Press. Gendler, Jason. 2016. The Rich Inferential World of Mad Men: Serialized Television and Character Interiority. Projections 10 (1, Summer): 39–62. Jenkins, Henry. 2008. Searching for the Origami Unicorn: The Matrix and Transmedia Storytelling. In Convergence Culture: Where Old and New Media Collide, 95–134. New York: New York University Press. Keating, Patrick. 2013. Narrative Dynamics and the Competitive Reality Show. Storyworlds 5: 55–75. Levinson, Jerrold, and Philip Alperson. 1991. What Is a Temporal Art? Midwest Studies in Philosophy 16: 439–450. Matravers, Derek. 2014. Fiction and Understanding. New York: Oxford University Press. McGonigal, Andrew. 2013. Truth, Relativism, and Serial Fiction. British Journal of Aesthetics 53 (2, April): 165–179. Newman, Michael Z. 2006. From Beats to Arcs: Toward a Poetics of Television Narrative. The Velvet Light Trap 58 (Fall): 16–28. Perkins, V.F. 2005. Where Is the World? The Horizon of Events in Movie Fiction. In Style and Meaning: Studies in the Detailed Analysis of Film, ed. John Gibbs and Douglas Pye, 16–41. Manchester: Manchester University Press. Prince, Gerald. 1988. The Disnarrated. Style 22 (1): 1–8. Pye, Douglas. 2000. Movies and Point of View. Movie 36: 2–34. ———. 2007. Movies and Tone. In Close-Up 02, ed. John Gibbs and Pye, 1–80. New York: Wallflower Press. ———. 2013. Seeing Fictions in Film. Projections 7 (1, Summer): 131–138.



Ryan, Marie-Laure. 1991. Possible Worlds, Artificial Intelligence, Narrative Theory. Bloomington: Indiana University Press. ———. 2008. Transfictionality Across Media. In Theorizing Narrativity, ed. John Pier and José Ángel García Landa, 385–417. Berlin: Walter de Gruyter. Segal, Eyal. 2010. Closure in Detective Fiction. Poetics Today 31 (2, Summer): 153–215. ———. 2011. The ‘Tel-Aviv School’: A Rhetorical-Functional Approach to Narrative. In Current Trends in Narratology, ed. Greta Olson, 297–311. Berlin: Walter de Gruyter. Shaham, Inbar. 2013. The Structure of Repetition in the Cinema: Three Hollywood Genres. Poetics Today 34 (4, Winter): 437–518. Smith, Murray. 2011. On the Twofoldness of Character. New Literary History 42 (2, Spring): 277–294. Stecker, Robert. 2013. Film Narration, Imaginative Seeing, and Seeing-in. Projections 7 (1, Summer): 147–154. Sternberg, Meir. 2008. If-Plots: Narrativity and the Law-Code. In Theorizing Narrativity, ed. John Pier and José Ángel García Landa, 29–107. Berlin: Walter de Gruyter. ———. 2010. Narrativity: From Objectivist to Functional Paradigm. Poetics Today 31 (3, Fall): 507–659. ———. 2012. Mimesis and Motivation: The Two Faces of Fictional Coherence. Poetics Today 33 (3–4, Fall–Winter): 329–483. Sternberg, Meir, and Tamar Yacobi. 2015. (Un)reliability in Narrative Discourse: A Comprehensive Overview. Poetics Today 36 (4, December): 327–498. Thomas, Deborah. 2000. Beyond Genre: Melodrama, Comedy, and Romance in Hollywood Film. Moffat: Cameron and Hollis. Thomson-Jones, Katherine. 2007. The Literary Origins of the Cinematic Narrator. British Journal of Aesthetics 47 (1, January): 76–94. Walters, Lee. 2015. Serial Fiction, the End? British Journal of Aesthetics 55 (3, July): 323–341. Wilson, George M. 1986. Narration in Light: Studies in Cinematic Point of View. Baltimore: Johns Hopkins University Press. ———. 2011. Seeing Fictions in Film: The Epistemology of Movies. New York: Oxford University Press.


On Rhythm in Film Editing Karen Pearlman

Introduction The issue of rhythm in cinema is enormously complex and still not well understood. Bordwell and Thompson, Film Art: An Introduction (1997: 196) I think all good editors have to be good dancers Robert Dalva, editor (cited in Gross 2009: 37)

What is rhythm in film editing? Philosophical discussions of rhythm in film are divided in their treatment of the subject. Cognitivists and analytic philosophers treat it warily, if at all, as an object of study that is complex and hard to understand. On the other hand, continental philosophers embrace rhythm as a subject, but can  make their discussions complex and hard to understand. This chapter aims to provide a bridge between the two. It attempts to demystify what rhythm is, how it is shaped, and what it is for, while still respecting that it is, in both a film editor’s and an audience’s experience, a felt phenomenon. The intention is to provide an analytic discussion that offers an explanatory framework for the embodied and affective experience of edited rhythms. The chapter begins with a brief discussion of the issue of different philosophical perspectives on rhythm in film. It then offers a cognitive framework for understanding the expertise that film editors activate when shaping rhythms. The question of what rhythm actually is, is considered by synthesizing my own firsthand knowledge of editing practice, and that of other practitioners, with theoretical and philosophical discussions of film. The filmmakers’ perspective, while less common My thanks to Ilona Hongisto and Catalin Brylla for their comments and suggestions in the development of this chapter. K. Pearlman (*) Macquarie University, Sydney, NSW, Australia © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




in contemporary philosophizing, has substantial precedent in the writings of Esfir Shub, Dziga Vertov, and Sergei Eisenstein, for example. It is used here to support the argument that understanding the cognitive complexity of the editors’ actions in shaping rhythm sheds light on its effect on audiences. The other unusual perspective in the mix is that of dance theorists whose insights about rhythm in movement help to support the argument that movement is primarily what is being shaped into affective rhythms through the editing of moving images.

Editing and Rhythm in Film Philosophy The question of what rhythm is in film arises with the first film philosophers. Very early in the history of theoretical writings on the medium of film, Harvard experimental psychologist Hugo Munsterberg wrote about the vitalizing power of rhythm in the movies, underscoring its capacity “to excite and to intensify the personal feeling of life and to stir the depths of the human mind.” (1916: 220) Munsterberg saw cinema as a “psychotechnology” endowed with a distinctive power over the viewer’s existential reality—and as such also a continuation of the various types of mechanical instruments developed to measure and quantify the human mind. (Henriques et al. 2014: 9)

This précis of Munsterberg’s ideas reveals a dichotomy present in discussion of rhythm in film from as early as 1916. The schism is between theorizing film as a deeply felt, almost spiritual experience and as a technological phenomenon that could measure and quantify, and thus presumably itself be measured and quantified. This dichotomy is particularly salient to my discussion of rhythm. The analytic approach offers specific and contained explanations of film techniques, which are useful to filmmakers and theorists. However, the study of rhythm is a study of something that is not, or is not primarily, created or encountered analytically.1 Dictionary definitions of the word “rhythm” frequently emphasize that “rhythm is a felt phenomenon” (Brogan 1993: 1068). Or, as Jean Mitry puts it, rhythms consist in “relationships that are felt” (1997 [1963]: 115, ital. in original). This quality of being felt and created through feeling is significant. It is this quality that causes rhythmic creativity to be characterized as subjective and ineffable in writings on the craft of editing and to be handled warily in cognitive theory. This chapter aims to provide ideas about rhythm in film editing that do not disrupt the felt experience, but do make the case that, although rhythm is a felt phenomenon, it is not just felt. Its properties can be identified. The processes of creating it can be articulated. The purposes it serves in engaging spectators in films can be understood. 1  There are, of course, important exceptions to this in the philosophizing of rhythm generally. See, for example, Rhythmanalysis: Space, Time and Everyday Life, Lefebvre (2004).



The dangers of this analytic approach must first be acknowledged, though necessarily briefly. Merleau-Ponty’s description of cinema applies to the experiences of a film’s rhythm, as shaped by editing: “not a sum total of images, but a temporal Gestalt…” (Merleau-Ponty 1964: 54). Films experienced in the wild (Hutchins 1995) (as opposed to classrooms or laboratories) are not g ­ enerally experienced as “discrete cinematographic elements (visual, aural, and editorial) but in these elements’ meaningful totality as a temporal and sensual configuration that is grasped” (Sobchack 2008: 439). The danger in breaking down a “meaningful totality” into component parts is that it actually does not function as component parts. Analogous to this concern would be breaking a bicycle down into its component parts. Doing so might reveal the parts needed to make a bike and the mechanics of their relations, but the bike cannot be ridden when it is in pieces. Films can’t be screened when they are in pieces either. The editor has to shape those pieces through selection and limitation of movement into a rhythm. It may therefore be useful to know what the component pieces of rhythm are. The question arises: can philosophy contribute further refinements to the question of what are the components of cinematic rhythm, how are they shaped, and what are they for, without slicing cinema into a dysfunctional bicycle? Perhaps due to the resistance of “flow” (Greek: rhythmos) to analysis, the design of rhythm as the kinetic, energic, and temporal patterning shaped by editing a film has had little attention from analytic or cognitive studies. Cognitive studies in editing such as Tim Smith’s study of continuity and attention (2012), James Cutting’s work dissecting films into the categories that make up its material/surface form (see, e.g., Cutting 2016) and Heimann et al.’s 2016 study of neurological responses to edits offer empirical evidence of biological mechanisms that respond to edits and editing techniques, but not rhythm in film editing. Noël Carroll’s discussion of point-of-view editing reveals evolutionary mechanisms that ground the expectation of a shot of someone looking to be followed by a shot of what they see (see Carroll 1996), but the rhythmic dimension of this is not discussed. One analytic text that does make mention of rhythm in film editing is Film Art. Bordwell and Thompson describe a number of attributes of rhythm in their discussion of the “Rhythmic Relations between shot A and shot B” (Bordwell and Thompson 1997: 278–280). This pragmatic discussion, which is found in a book that covers the whole gamut of film techniques and is directed to the attention of undergraduate students, is necessarily cursory. Most of the operations it describes will be subsumed under the operations I will call “pacing,” as in frequency of cuts rather than rhythmic relations as a whole. However, Bordwell and Thompson’s neo-formalist approach, which is generally to observe and describe techniques being used in the composition of films, will be applied to the materials with which editors are working and the tools they have for working with them. Doing so will help develop an understanding of what an editor is doing when shaping rhythm, which in turn can help us to understand what rhythm is when experienced in a finished film.



Rhythm and the Editor’s Cognition Once editing of film evolves, as it did very early in cinema history, to become part of expressive narration of ideas, it involves creative decision-making (Orpen 2003; Pearlman and Heftberger 2018; Pearlman et al. 2018). Editors are responsible for the final decision-making about structure and rhythm of the film that ultimately reaches an audience. (Pearlman 2009, 2016). Editors’ creative skills have some basis in the built-in meaning-making and expressive capacities of humans, but are not, as skills, “natural” or even “instinctive.” They are culturally learned and developed forms of expertise. Thus, I propose, understanding editors’ expertise helps to reveal what it is that they are expert at shaping when they shape rhythm. The inclusion of the editor’s cognition in the analysis of rhythm in film editing is further supportable given that the editor is positioned, on a production crew, as the film’s first viewer. Robert Wise is just one of many esteemed editors and directors who would agree that “as the editor you are the audience” (Wise, cited in Orpen 2003: 7). As the film’s “first audience,” the editor embodies several key attributes which have been associated with affective responses to film in cognitive film theory. One that is particularly salient is embodied simulation (ES), as described by Gallese and Guerra: ES has been proposed to constitute a basic functional mechanism of humans’ brain, by means of which actions, emotions and sensations of others are mapped onto the observer’s own sensory-motor and viscero-motor neural representations. Such theory was triggered by the discovery of mirror neurons in the macaque monkey brain. Mirror neurons are motor neurons that typically discharge both when a motor act is executed and when it is observed being performed by someone else. (Gallese and Guerra 2012: 184)

The theory that mirror neurons are the source of embodied simulation continues to be controversial (see Hickok 2014). However, whether it is specifically through mirror neurons or these neurons are simply metaphorical for a much more complex imitative system in human brains, it is plausible to suggest that editors respond to movement that they see in the unedited filmed material. Some part of what they see or hear in the uncut material’s movement will provoke an empathetic response, and that part will be selected and juxtaposed with another part that also has a qualitative affect. Putting two shots together, each of which inherently has rhythm, makes a third rhythm, which is not the same, or even just the sum of the first two (see Eisenstein 1942, 1949; Kuleshov in Levaco 1974). So, the edit begins to have a rhythm of its own. At this point, editors cannot simply recognize a “right” rhythm, as an audience might. Their own internal rhythms must come into play to shape rhythm through an editing process. As editors begin to do more than respond to  existing rhythms, we draw on rhythms known to us through what Arnold Modell might call ­“corporeal imagination” (2003), as well as latent rhythms in the film material, to create the finished film’s rhythm.



Modell’s phrase “corporeal imagination” suggests that the editor’s embodied and extended cognition responds directly to movement they see in the material. Applying cognitive dance theorist Dee Reynold’s phrase “kinaesthetic imagination” (2007) takes this a step further. “Kinaesthetic imagination” describes choreographers’ capacities to draw on their own feeling for movement and imagine creative patterns and flows of movement. I propose that this is also a skill that editors develop. They use embodied cognition, corporeal, and kinesthetic imagination in concert with various forms of procedural and cultural expertise, to construct coherent movement phrases from disparate fragments of moving images. They apply this enactive cognition to several kinds of movement (e.g., narrative movement and visual movement,) to shape persuasive (Orpen 2003) rhythmic experiences of films. Modell and Gallese and Guerra tie their cognitive neuroscience directly to phenomenology. They take a cognitive approach to the understanding of a felt phenomenon, and in this I follow them. Scholars working on the linguistics-­ based Cinematic Poetics research project in Germany ( also could be said to tie the two traditions together in their analysis of rhythm. Films move their spectators. The film’s communication with spectators can be understood as a vital form of aesthetic composition and as bodily and sensory responsiveness. In this sense, when we say, the film moves the spectator, it is not meant to be understood metaphorically but quite literally: Film images develop as movement patterns, combining different staging tools like sound composition, montage rhythm, camera movements, and acting to one temporal gestalt. They do literally move spectators, because they organize their perception processes in the temporal course of the film reception. (Scherer et al. 2014: 2083)

The notion that films “literally” move spectators’ bodies is ratified in scientific analyses of human response to movement (e.g., Heimann  et  al. 2016; Gallese and Geurra 2012; Grodal 1997). The notion of the film as a “temporal gestalt” implicitly points to the work of the editor in piecing together filmed fragments into a unified experience. That film images “develop as movement patterns” (Scherer et al. 2014: 2083) is not a naturally occurring phenomenon, but one which is shaped by the expert embodied, embedded, and enactive cognitions of the editor. Thus, by analyzing the editor’s work of shaping rhythm, I aim to shed light on what part rhythm plays in the experience of an audience.

On Asking: What Is Rhythm in Film Editing? Lewis (2008) writes of Mitry that: [h]is intellectual project was ambitious, Aristotelian in nature. What are the material, efficient, formal, and final causes of the cinema? What is it made of? How does it work? To what effect? To what end? (397)



The approach I have taken to analyzing rhythm in film editing was originally intended to follow these august thinkers and ask these same questions. However, the inquiry took a “cognitive turn” at a certain point and landed on a variation. Thus, asking “What is rhythm is in film editing?” is actually asking only three questions: The first of these is, like its Aristotelean antecedent, “What is it made of?” I argue that the materials for making rhythm are time, energy, and movement, with movement being the material through which the other two are experienced. The second question is a variation on “How does it work?” which asks about how an editor works on it, or how the materials from which rhythm is made are shaped into a significant form. The argument here is that the editor has three cognitive/conceptual tools or operations that they draw on: timing, pacing, and trajectory phrasing. Even though all of them are at work simultaneously in the fluent editor’s expert work with filmed material, each of these can, and will, be broken down into more refined operations. Analyzing how an editor works on rhythm reveals something about how edited rhythms work because the words for the editor’s tools—timing, pacing, and trajectory phrasing—are also words for what audiences ultimately experience. The questions “To what effect? To what end?” are here amalgamated to ask “What is rhythm?” in the sense of “What is its purpose?” This question returns us to the definitions of rhythm that position it as a felt experience. The proposal is that its purpose is to modulate feeling and attention into “cycles of tension and release” (Pearlman 2009, 2016). A further purpose (effect and end) of these cycles is to “synchronize” (Pearlman 2009, 2016) an audience to the movement of images and sounds, emotions, and events in a film (Table 7.1). However, before elaborating on the materials, tools, and purposes of rhythm in film editing, it is important to note some limitations on this inquiry. I am not asking what is good rhythm. The “goodness” of rhythm is contingent, at the very least, on the intentions of the filmmakers and the cultural context in which it is being received. What is a “good” cycle of tension and release, shaped with “good” timing, from “good” movement in the captured material in, for ­example, the United States in 1918, may be considerably less viable as an affective experience in Nigeria in 2018. Nonetheless, it is still a rhythm constructed by an editor from the same materials, with the same tools and, generally, for the same purpose.

Table 7.1  Table form distillation of the three questions being asked and the ideas being proposed in response to each Questions

Ideas Proposed

What are the materials of rhythm? What are the editors’ tools? What is the purpose of rhythm?

Time, energy, movement Timing, pacing, trajectory phrasing Cycles of tension and release



Therefore, I am not asking about the experience of individual constructions of film rhythms—which film flows well and which one does not. I am also not asking about difference between rhythms of cultures, genres, or forms. Instead, I propose a model through which film editing can be understood, and which can be applied to diverse films for testing, expanding, or refuting.

Materials of Rhythm: Time, Energy, and Movement It is a contemporary film industry truism that good editing is “invisible.” This means that, as in many arts, the experiencer’s attention is drawn to the cumulative effect, which in film is usually called “story,” rather than to the craft of its construction. Editing is, of course, not invisible. If it were, movies themselves would be invisible. However, what we see is not the edits; we see that which is edited. The “editing” is experienced not as a series of joins, or even a series of shots joined together. The experience of an edited film is an experience of flow of movement: movement that has been shaped, limited, and designed into cinematically expressive phrases. Movement, then, with its expressive possibilities, is the real concern of editing theory and philosophy. Reducing the materials available from which to craft cinematic rhythms to time, energy, and movement, three elements present in every aspect of life and being perceivable by humans, may seem to be a simplification. That, in a sense, is the point. The materials from which narrative and non-narrative experiences of cinema could be crafted are exceedingly diverse. Their number and diversity has increased exponentially since the invention of the film camera and the establishment of the two major forms of editing (commonly known in English as continuity editing and montage editing). Boundaries of propriety and access, and affordances of film equipment, which may once have limited what could be filmed and edited, have fallen away, and nothing whatsoever comes to mind as potentially excluded. Therefore, any attempt to theorize editing and rhythm in film editing usefully must first reduce these possibilities to something they have in common. Movement, being the root word of cinema (kine), is the clearest choice. Rhythm in the arts, as in the sciences, is understood to be patterned movement over time. This movement may be physical/physiological, molecular/ biological, conceptual, verbal/written, visual, aural, or a blend of modes, as it is in cinema. Waking/sleeping, eating/digesting, working/resting, and inhaling/exhaling are just some of living beings’ ways of participating in the rhythms of the world, of surviving by oscillating or moving with the rhythms of their physical world. Going beyond rhythmic survival, and into rhythmic creativity, as film editors must, it is useful to stream the complexities of types of movement being shaped in cinematic constructions to three broad and overlapping categories.



Movement of events—the rise and fall of perceivable changes to narrative, also sometimes called plot or structure. Duration (time) and emphasis (energy) of the movement of events is crafted by editors in editing processes. Movement of emotions—the dynamics of character emotions as they are expressed in performances, the tonal aspects of spaces and situations, and the affective valence of sequences, all move over time and through degrees of intensity (energy). Editing shapes these movements. Movement of image and sound—the visible and audible experiences of movement in the captured material. Patterning of rate of movement (time) and its force (energy) may be designed for optimal smoothness (see Pudovkin 1949) or explosiveness (see Eisenstein 1949), but in either case, it is patterning of movement. Movement happens in time, and it is impelled by energy, but we cannot see time and energy. So, editors use movement, movement of events, emotions, images, and sounds to shape expressive patterns of time and energy. While editing has been hidden under the distracting cloak of “invisibility,” and overshadowed in film theorizing by a preoccupation with narrative, some useful theory has been developed in dance to explain the significance of shaping experiences of time and energy by shaping movement. The “Choreographic Cognitions” team conducted a study of what dancemakers shape and viewers respond to in danced movement. They write: [T]he artistry of movement is in trajectories, transitions, and in the temporal and spatial configurations in which moves, limbs, bodies, relate to one another… In a dynamical system, time is not simply a dimension in which cognition and behaviour occur but time, or more correctly dynamical changes in time, are the very basis of cognition. (Stevens et al. 2000: 4)

Movies are also dynamical systems. The actions and behaviors that are directed, performed, recorded, and shaped into narratives and narration are movements. They may also be facts, symbols, provocations, and so on. However, the argument being made here, as it has been in dance and in some editing theory (see the bodies of work by Eisenstein, Vertov, and Shub, in particular), is that without energy (dynamic) changes in time, the events, emotions, and images are inert. They lack significance or discernable intention. For a quick experience of how this works, type the words “go home now” and get your computer’s voice to read them out. The time between words and emphasis on words will be even, so you cannot tell what intention impels the words— is it fear or delight? Without dynamical changes in time, it is not really possible to understand the affective significance or intention behind the words. Understanding energy as intention draws on movement/dance analysts Rudolf Laban and Irmgard Bartenieff’s ideas about effort. Effort, as described in their in-depth studies of movement, roughly translates as the attitude and intention behind movement which informs the way it is done (see Bartenieff 1980). The kind of effort with which a person moves is what they mean or



intend with movement. A punch means aggression, violence, forceful intentions, if its effort is aggressive, violent, and forceful. A punch can also be playful. It may originate from a playful state of mind, and the effort that propels it will be entirely different. It may move along the same spatial pathway as an aggressive punch, be a similar speed, and have the same shape, but its effort, or energy, will tell us that it means something different. Narrative context will, in a movie, be important to understanding whether the effort, energy, or intention is aggressive or playful. However, it is worth noting, that, given two performances—one aggressive and one playful—it is actually the editor, working in collaboration with the director, who selects which one to use. The editor composes the series of movements, selected from the recorded material, into the moment-to-moment sense of causality we call “narrative.” In film editing, editors are rarely simply making an experience of time, energy, and movement; they are also shaping story, character relationships, and other kinds of information. Furthermore, film editors rarely work exclusively with human movement. However, in shaping the rhythm of the film, time, energy, and movement are the salient factors. They shape the qualitative experience of the story and information. The movement through time and energy of all of the filmed images is shaped into phrases of related movements and grouped emphasis points. These phrases are then varied, juxtaposed, interpolated, and shaped within themselves and in relation to one another to make the overall experience of time, energy, and movement in a film that is known as rhythm. The next section examines the specific tools an editor has for the shaping of rhythm in film, and considers how they shape an audience experience of edited films.

Editor’s Tools: Timing, Pacing, and Trajectory Phrasing Shaping rhythm is complex, due to the many layers of movement being shaped, purposes for which it is being shaped, parties with investment in its ultimate shape, and equipment available for shaping it. However, from the editor’s point of view, it is possible to analyze the shaping of movement of events, emotions, images, and sounds into three operations: timing, pacing, and trajectory phrasing. These operations each has three suboperations or aspects, and each of these addresses some part of the shaping of the film, and the audience’s experience of it, as a constructed rhythm. For the film philosopher interested in becoming a film editor, more detail about these three operations can be found in both editions of Cutting Rhythms (2009, 2016). Timing There are three aspects of timing to be considered when discussing rhythm in film editing: (1) choosing a frame; (2) choosing duration; and (3) choosing the placement of the shot. Each of these choices is a functional aspect of shaping



rhythm. Choices made, of course, imply choices not made—movements not included, and this too is of considerable significance in shaping rhythm, because rhythm is not just the generalized flow of movement but the limitation and definition of the flow of movement. 1. The choosing a frame sense of timing is the tool at work in firmly limiting the movement of one shot by choosing the precise frame on which to begin and end it. Think of a child holding a crystal goblet on a stone patio. The editor might choose to show us first the stone floor, then the glass held over it by a child’s greasy fingers. Our minds do the work of connecting these shots and fearing the collision of expensive crystal on unforgiving flagstone. (This shot is an example of creating the tension side of the cycles of tension and release that will be discussed below.) But on which frame does the editor cut away from those fingers, and what is implied by the choice of different frames? Cutting before the glass leaves the hand implies its crash is accidental. Cutting on the frame where fingers deliberately begin to lift their grip might imply threat, an intention to execute the destructive action. Cutting a frame or two after the glass has been released might reveal the hand freezing in fear of consequence. Thus, the choice of frame on which to cut shapes the movement to three different narrative implications. We could summarize these as: “the child drops the glass” or “the child destroys the glass” or “the child fears punishment.” These different implications would be surmised from the choice of frame on which the editor cuts. 2. Choosing duration is distinct from choosing the precise frame on which to cut, because, although a shot may change meaning quite dramatically by holding or dropping a frame, the feeling of its duration is not really affected by one frame (which is only small fraction of a second). A ten-­ second shot will feel long if it is juxtaposed with a series of one-second shots. The same 10-second shot, used in the same context, will still feel just as long if it is actually only 9 seconds and 20 frames. And the same 10 (or so) second shot will feel quite short if juxtaposed with a series of 60-second shots. The feeling of a shot’s duration is created by the relative durations of the shots near to it and the concentration of information, movement, and change within it (which is discussed under pacing, later). Perception of duration, as part of the felt experience of rhythm, is also richly entangled with what Mitry calls the “relationships of intensity” (1997 [1968]: 125). Intensity is discussed as an element particularly relevant to the design of trajectory phrasing, later. 3. Choosing the placement of the shot. The decision about where to use shots is also called timing. This sense of timing refers to where in the sequence or ordering of shots a particular reveal of information gets placed. Think of this aspect of timing as analogous to placing a punch line in the context of a joke. A comedian will not start with the punch



line; they will place that piece of information deliberately in relation to a setup that frames it. Otherwise it isn’t funny. This sense of timing has significant implications for the shaping of cycles of tension and release. The shaping of an edit rarely goes according to script, and the editor has a lot of control over the timing of release of information. If we think of narrative tension as the creation of a question in the viewer’s mind (see Carroll 1996), and release as the resolution of that question, then we can see that this sense of timing, of where to place the reveal of information, is a domain in which the editor’s choices are very powerful in shaping the audience experience. Pacing Pacing as an operation in shaping a film edit is often conflated, confusingly, with timing (see, e.g., Reisz 1953). They do overlap, of course, but, in fact, they are different operations. Pacing refers not to the frames, durations, or positions of shots, but to modulation of (1) rate of cutting; (2) rate of movement within a shot; and (3) rate of movement overall. 1. Pacing, as in rate of cutting, concerns how often cuts occur per second or minute or hour. This is not just another way of saying “duration of shots,” although the two ideas do overlap.2 Pacing in this sense can easily be seen when the rate of cutting occurs in patterns, for example, accelerating the number of cuts per minute as a chase gets closer to its climax. (In this case, the durations of the shots get shorter, and the two meanings overlap.) However, pacing, as in rate at which cuts occur, is also a factor in the rhythm of film—even when it is not patterned by design. For example, cutting frequently around a conversation may make the performances seem edgier or sharper. Here, we are not looking at durations of shots directly, but at the content curve of movement within the shots, and either cutting it very sharply, which creates a sense of the pace being quicker, or leaving it loose, with full arcs of movement intact, which makes the pace seem slower. Thus, the pacing, in the sense of the rate at which the cuts occur, manipulates the sensation of the movement of the conversation. Further, the rate at which cuts occur defines the rate at which new visual sensations are introduced. Every cut is in itself a change, so a lot of cuts make a faster rate of change. 2. Pacing also refers to the juxtapositions of rates of movement or change within shots. Imagine these actions occurring on one five-second shot: glass shatters—child cowers—father looms—friends disperse. Now, visualize 2  For more on rates of cutting and duration of shots, see Salt (1974) and the work of scholars in the area of cinemetrics:



that shot in juxtaposition with another five-second shot, in which: hands get cut—dog yelps—child wails—broom descends. The pacing of the sequence may be seen as very fast, even though the cuts are relatively infrequent. Making one cut in ten seconds is not a “fast” rate of cutting in contemporary cinema (see Salt 1983; Bordwell 2002). If the editor chooses to present each of these events in its own shot, thereby making a cut every three seconds, rather than one cut in ten seconds, they would make the rate of change slow down. These events, each shown in a single 3-second shot, would take 24 seconds, not 10. The editor would have sped up the rate of cutting, but the overall effect would be of slowing the pacing. A different kind of tension is created by each choice. Cramming action into ten seconds and seeing it overlap inside shots, albeit necessarily wider shots, will force the movement of the viewer’s eyes rapidly around the frame, trying to absorb the full content curve in each of the overlapping movements in the shot. Tension arises not only from the action but from the necessity of working hard to comprehend it and surmise its implications. Spreading the action over eight closer shots, each containing a single action and lasting three seconds might create tension that is more like dread. The slow unfurling of actions delays resolution of the situation and forces a more intimate experience of the full content curve—the “preparation,” “execution,” and “rest” (see Schmidt 1996) of each action. There is, of course, a third choice: the editor could cut together just the close-ups, using only the peak of each one’s movement quotient, for example. That would increase the rate of cutting to eight shots in eight seconds, making a faster rate of cutting and a faster rate of movement within shots. There is also a 4th choice, a 44th choice, and a 444th choice. At least. There are multiple shots, say, in this case, 10, that have multiple frames, say, 24 per second × 7  seconds each, and any one of these frames could be juxtaposed with any other frame for any duration. Most of these choices can be chunked down by expert editors and easily discarded, without being tried. Some are tried, usually iteratively, meaning we refine until it feels right in the moment and then refine again when that moment becomes part of rate of change overall. . The rate of change overall aspect of pacing refers to rate movement or 3 change of the overall film. A film’s pacing may be the rate at which events move in the film or the rate at which movement of images or emotions occur in the film. In the audience experience of rhythm, this aspect of pacing is necessarily an outcome of the other two aspects of pacing (discussed above). It is felt not as a series of distinct choices, but as a flow of these choices into one another to produce experiences which are generally summed up in one word such as “fast,” “slow,” or “uneven.”



Pacing is very important, especially for the creation of sensations of time and energy, but it is not by itself rhythm. Although rate of movement is significant to rhythm, a rhythmically designed expressive film also requires phrasing of the movement trajectories. Trajectory Phrasing “Trajectory phrasing” is a term designed to cover an area of editing rhythms which is not precisely covered by saying “timing” or “pacing.” Trajectory phrasing describes the manipulation of energy in the creation of rhythm. The word “trajectory,” according to the Random House Dictionary, means “the path described by a body moving under the action of given forces.” So, “trajectory” describes a combination of the direction of a movement and the energy that propels it. “Trajectory phrasing,” then, is joining together movement trajectories in different shots to shape the flow of energy between and through them. This is done by choices of takes and positioning of cuts. The three operations that “trajectory phrasing” describes are (1) linking and colliding trajectories; (2) selecting energy trajectories; and (3) stress. 1. Linking or colliding trajectories: In practice, this means looking at aspects of movement, such as screen direction and asking if they should link or collide. A smooth cut is one in which movement from right to left in one shot is matched with movement from right to left in the next shot. A cut in which movement from right to left is collided with movement from left to right, or simply unmatched in spatial organization and energy, is what Eisenstein might have called a “conflict.” Eisenstein favored the creation of these collisions, famously stating, “montage is conflict” (1949: 38). He writes about several things which could be put in to conflict, including: “close shots and long shots, pieces of graphically varied directions, pieces resolved in volume, pieces resolved in area, pieces of darkness, pieces of lightness” (Eisenstein 1949: 39 italics in original). Significantly for this argument, Eisenstein’s list includes “conflicts between an event and its duration” (1949: 39, italics in original), revealing that the shaping of movement trajectories is a more comprehensive montage action than just the joining of two shots. Trajectories of events, like emotions, are experienced through their visible expression in movement. This movement can be linked, or collided, or edited along a spectrum between the two. The smoothness or conflict in its phrasing will be one of the things that impacts on audience comprehension and felt experience.



2. Selecting energy trajectories: This involves selecting different takes for the variations on use of energy or effort within them. Returning to the scene of the smashed glass on the stone patio, take, for example, the p ­ resence of notoriously difficult to direct animals and children. The script reads: dog yelps—child wails, but this is at best a guideline. The dog might howl and the child cower. The dog might sniff and the child shrug. We might have shots of all of these variations, and cutting them together into the story would change the trajectory phrasing and create a differently nuanced narrative. If the dog obliges with a convincing yelp, the phrase of movement would have a different dynamic—and timing—and significance—if it is followed by the child’s wail, or cower, or shrug. Any of these shots can be cut together to match (link smoothly), but the rhythm of the scene is shaped by the choices of movement energy. The question for the editor is: which gestures, with their particular emotional valence as expressed by effort, will be cut together to create the trajectory that best expresses the intended movement of emotions in the film? Which shots create or release tension as appropriate to the desired significance of events and character relationships? 3. Stress. This refers to creating emphasis points, or stress accents, by manipulation of the trajectory of movement. Stress in movement is gradations of strength or intensity of energy. In our hypothetical sequence, above, the shots of wailing, cowering, or shrugging may be of equal duration. The rates of movement or change within them are comparable, but the stress (effort) they contain and the stress (emphasis) they create are different. Both the shot size and the energy being expended within the shot contribute to the energy accent it makes. By phrasing the trajectories of movement, editors are working in collaboration with everyone from directors and actors to sound designers and composers to put together an appropriately dynamic flow of energy in shot-to-shot juxtapositions, scenes, sequences, and the whole film. What is ultimately considered “appropriate” may vary from the original intention, due to variations in performances, issues with the script, unexpected production problems, and many other things. Thus, an editor’s operations are often understood as “problem-­ solving.” However, it is important to understand this problem-­ solving as a creative act. The phrasing of trajectories is creatively phrasing the flow of movement, time, and energy into rhythms (Table 7.2). An editor uses the tools of timing, pacing, and trajectory phrasing when cutting rhythms. The creation of rhythm in film editing will generally rely on all of these tools and operations being employed simultaneously, or in close alternation, because, as in dance, rhythm in film “is not just a duration of time, accented by stresses. It is also the result of the interaction of Effort combinations with variations” (Bartenieff 1980: 75).



Table 7.2  Table form distillation of the three operations, or tools, editors deploy and each of their three suboperations Editors’ Tools or Operations

Aspects of Each Operation


Choosing a frame Choosing duration Choosing placement of shot Rate of cutting Rate of movement or change within shots Rate of movement or change overall Linking or colliding trajectories Selecting energy trajectories Stress


Trajectory phrasing

Purpose of Rhythm: Cycles of Tension and Release I turn now to the purpose of rhythm in film editing, which I propose is the creation of cycles of tension and release. “[A]lternating tension and rest” (Mitry 1997 [1963]: 104) is core to Mitry’s definition of rhythm as a whole. The slightly more active “tension and release” is a variation borrowed from choreographers’ more colloquial discussion of danced movement. It is used here, in the first instance, as shorthand for three things: 1. The “on-off ” pulsing of bodily rhythms as they move from exertion to relaxation. For example, inhalation is exertion, while exhalation is relaxation. Similarly, the heart is a muscle that contracts (tension) and releases. As has been demonstrated in a range of experiments, watching films has a direct influence on these kinds of bodily tension and release in the viewer. 2. “Tension and release” is also shorthand for a discussion of a range of kinds of experiences of emotions. Characters and social actors undergo heightening and relaxing of emotional states as they encounter problems and solutions in their worlds. For example, they create tension for themselves, for one another, and for audiences, in the heat of dramatic conflict and release the pressure they have built if/when they reconcile. 3. Finally, “tension and release” is also shorthand for the ways that narrative events excite cognitive attention with questions and release it by resolving them. Returning to the crystal goblet hovering over the flagstone in the child’s greasy grip, we can see that “tension” is potentially created or released in every shot-to-shot juxtaposition. Indeed, this might be an effective way of summarizing the editor’s contribution to the sense of causality in narrative films. As has been well rehearsed in writing about editing over the years (see Kuleshov in Levaco 1974), the viewer’s mind will put two shots together to make an infer-



ence about them. The glass hovers: tension, the glass smashes: release and a new tension. The first tension, “Will the glass get broken?”, is resolved by it breaking. The break raises a new question: “Will there be consequences of the breakage?” Now the editor has a choice in how this tension about the consequence of breakage is refined and resolved. If the editor cuts back to the child’s hand frozen in shock or fear, the question might be refined as something like: “Will the child get in trouble?” If the editor cuts to Dad looming, the question might be more like, “Will he hit the child?” If the editor cuts to a smiling gardener with a broom, the tension will likely dissipate, even lose interest as a question altogether, as we surmise, “Ah, he’ll clean it up.” This example of tension and release in a shot-to-shot juxtaposition reveals how tension can be built, dissipated, and rebuilt at the level of image and sound. The editing composes a phrase of movement from glass hovering, to smash, to hand retracting. These are simple movements in image and sound. Our minds do the work of connecting them causally. The sequence also clearly reveals movement of emotion. The verbs are affectively inflected. So, we follow along with the sequence through feelings that might flow in this way: apprehension (hovering), shock (smashing), and fear (retracting). Two points are worth noting here. First, our kinesthetic empathy with the movement of image and sound triggers our affective engagement. Our embodied knowledge of the movement feelings of hovering, smashing and retracting trigger apprehension, shock, and fear. Second, as an audience, the depth and direction of our feeling is being managed by the editor. As noted earlier, the editor’s choice of which shot to cut to after the smash will cognitively (though perhaps not consciously) trigger different questions and hypotheses. Each of these will vary the qualitative shade of fear we feel from a lot (Dad looming) to a little (gardener smiling). The sequence is also a movement of an event, call it: “the glass smashes.” How the editor chooses to make this event move will determine our sense of how significant an event it is in the narrative. It could be the most significant event of the story, in which case the editor might choose to manipulate duration and energy (through timing, pacing, and trajectory phrasing) to center our attention on it commensurately. Or, it may barely be an event. It could be just something that occurs in the narrative of, for example, “the family celebrates.” In this case, the breaking glass incident might be one of many at the party, and thus rhythmically shaped to move appropriately. Like the movement of emotion, our cognitive assessments in the movement of events are grounded in our kinesthetic empathy with the movement visible and audible in the composition of shots. The logic is this: the composition of “hovering, smashing, and retracting” triggers apprehension, shock, and fear. Apprehension, shock, and fear would give rise to the question: “Will the child get in trouble?” Changing the physical (image and sound) sequence of movement to “hovering, smashing, and smiling” would change the emotional “ride” we take with the movement to apprehension, shock, and relief. This sequence of emotions would facilitate a different question, something like: “Will the party continue?”



Tension and release are thus created, by the editor, at all three levels of rhythmic construction: image/sound, emotion, and event. Tension and release are also experienced, by an audience, physically, emotionally, and cognitively. Although it is possible, and in some cases desirable, to describe these different kinds and sources of tension and release, it is important to note that all three are grounded in the embodied experience of perceptible movement. Further, all of these levels of tension and release, and their cinematic triggers, are operating at the same time. They are profoundly entangled, both in the film-­making process and in the film-watching process. Pulling them apart provides, perhaps, an image of different scales of tension and release, or different causes within the flow, but ultimately these differences are resolved into one “sensory (visual and sound), kinetic, intensive, affective, rhythmic” experience (Deleuze 2005: 29). The proposition is that watching a film is a physical experience of patterns of movement. Films synchronize the audience’s physical, emotional, and cognitive experience by getting us to embody, empathize, and conceptually participate in their movement. Just as all of the elements discussed here are functionally entangled (e.g., time and energy cannot be separated from movement), the bodies that rhythm addresses are also functionally entangled: address to the body through patterns of movement is address to its physical, emotional, and cognitive capacities as one living, breathing entity. Interestingly, continental and cognitive philosophers can agree on this. One writes: When it comes to the capture and modulation of the rhythms of heartbeat and emotions taking place in the cinema, for instance, we can speak about a mode of power that operates on the level of sensorimotor bodily adjustments. (Henriques et al. 2014: 15)

The other writes: When a viewer chooses to watch a film, he thereby chooses to be cued into having constant fluctuations of heartbeat, perspiration, adrenalin-secretion and so on. (Grodal 1997: 42)

There are clues in the word choices, but one would be forgiven for not knowing which was written by the continental philosopher and which by the cognitivist, since they agree: the comprehension of a film is physical, first. Thus, we might say that when effectively designed, and optimally drawing on the resources available in the uncut materials, rhythms are the movement of the film composed in such a way as to influence the viewer’s pulse, breath, and attention. A film’s significance is not just “this happened and then that happened.” A film’s impact is in the way that this, then that, happened, including how fast or slow or bumpily or smoothly or forcefully or limply.



Rhythm plays a crucial role… in the way the story is told, in the game of revealing and withholding story information from the viewers to maximize both their active involvement in anticipating the events and their passive abandon to the story’s events. (van Leeuwen 1985: 186)

So, whereas events, characters, and images trigger specific emotions, expectations, and ideas, the rhythms of these modulate the rise and fall of the tension and release—the “resonance of bodily reactions” (van Leeuwen 1985: 186)— with which we follow them.

Conclusion Rhythm in film editing is time, energy, and movement shaped by timing, pacing, and trajectory phrasing for the purpose of creating cycles of tensions and release (Pearlman 2009, 2016). Rhythm is, unsurprisingly, a body thing. Our bodies navigate rhythmically in order to survive and thrive in our rhythmic universe. We participate in the oscillations of natural and human-devised contexts by oscillating with them. It is therefore also unsurprising that we shape the rhythms of movies with the same materials from which we and our rhythmic contexts are made—time, energy, and movement. The tools with which we shape rhythms are dictated by the affordances of these materials: timing, pacing, and trajectory phrasing are the operations afforded by moving images to shape the rise and fall of narrative, affective, aesthetic, and embodied experiences. Comprehension and description of these experiences are not exclusive to the domain of continental or cognitive philosophy. Rhythm is a body thing, self-­ evidently—our hearts, breaths, and gaits are all rhythmic. But it is just as self-­ evidently a “mind” thing. We perceive, inquire, analyze, and understand in cycles that parse information, ideas, and concepts into modulated formation and release. Editors call the complex cognitive activity of shaping rhythms “intuitive” (see Oldham 1992, 2012)—meaning they do it by feel, not by conscious calculation. The editor’s cognitions are “hybrids, unevenly distributed across social, technological, and biological realms” (Sutton 2006: 239). Editors deploy cognitive actions distributed across their brains, bodies, and the filmed material as it passes through the editing technology. Editors’ expertise draws on knowledge of the construction of story, their own felt experiences, their kinesthetic empathy with movement in the filmed material, and their capacity to align how they feel when watching versions of the different rhythms they are creating to how their audience is likely to feel (see Pearlman 2018). Thus, editors’ expertise involves assessing the possibilities in their edit room environment and the possible impacts in the final viewing environment (e.g., the cinema); using that expertise to chunk down and discard potentially thousands of options; and shaping the flow of time, energy, and movement into patterns known as rhythms.



Rhythm in film is complex, but it can be understood. Shaping rhythm is not unlike shaping dance, and thus it does not hurt for editors to be good dancers, or at least have heightened capacities to tune to rhythmic possibilities and shape phrases of movement into significant form. The editor’s understanding and their dancing might be standing in, here, for mind and body, or cognition and the felt experience. This discussion has provided an explanatory framework for the embodied and affective experience of edited rhythms, which demonstrates that edited rhythms act on bodies and minds in a functionally integrated way. Thus, we can say what rhythm is, how it is shaped, and what it is for, while still respecting that it is, in both a film editor’s and an audience’s experience, immediate and embodied.

Bibliography Bartenieff, Irmgard. 1980. In Body Movement: Coping with the Environment, ed. Dori Lewis. New York: Gordon and Breach Science Publishers. Bordwell, David. 2002. Intensified Continuity Visual Style in Contemporary American Film. Film Quarterly 55 (3): 16–28. Bordwell, David, and Noël Carroll, eds. 1996. Post-Theory, Reconstructing Film Studies. Madison: University of Wisconsin Press. Bordwell, David, and Kristen Thompson. 1997. Film Art, an Introduction. 5th ed. New York: McGraw Hill. Brogan, T.V.F. 1993. Rhythm. In The New Princeton Encyclopedia of Poetry and Poetics. Princeton: Princeton University Press. Carroll, Noël. 1996. Theorizing the Moving Image. Cambridge: Cambridge University Press. Cutting, James E. 2016. Narrative Theory and the Dynamics of Popular Movies. Psychonomic Bulletin & Review 23: 1713–1743. s13423-016-1051-4. Deleuze, Giles. 2005. Cinema II, the Time Image. Reprint. London: Bloomsbury Academic. Eisenstein, Sergei. 1942. The Film Sense. Trans. and Ed. Jay Leyda. New  York: Harcourt, Brace. ———. 1949. Film Form: Essays in Film Theory. Trans. and Ed. Jay Leyda. New York: Harcourt, Brace. Gallese, Vittorio. 2011. Embodied Simulation Theory: Imagination and Narrative. Neuropsychoanalysis 13(2): 196–200. 080/15294145.2011.10773675 Gallese, Vittorio, and Michele Guerra. 2012. Embodying Movies. Cinema 3 Journal of Philosophy and the Moving Image 3: 183–210. ———. 2014. The Feeling of Motion: Camera Movements and Motor Cognition. Cinéma & Cie XIV: 22/23. Grodal, Torben Kragh. 1997. Moving Pictures: A New Theory of Film Genres, Feelings, and Cognition. New York/Oxford: Oxford University Press.



Gross, Daniel. 2009. Making A Story Move: The Art of Film Editing. University of Connecticut. &context=srhonors_theses Heimann, Katrin S., Sebo Uithol, Marta Calbi, Maria A. Umiltà, Michele Guerra, and Vittorio Gallese. 2016. Cuts in Action: A High-Density EEG Study Investigating the Neural Correlates of Different Editing Techniques in Film. Cognitive Science 41 (6): 1555–1588. Henriques, Julian, Milla Tanen, and Pasi Valiaho. 2014. Rhythm Returns: Movement and Cultural Theory. Body and Society 20: 3–29. 34X14547393. Hickok, G. 2014. The Myth of Mirror Neurons: The Real Neuroscience of Communication and Cognition. W.  W. Norton. books?id=LK5bAwAAQBAJ. Hutchins, Edwin. 1995. Cognition in the Wild. Boston: MIT Press. Lefebvre, Henri. 2004. In Rhythmanalysis: Space, Time and Everyday Life, ed. Stuart Elden. London/New York: Continuum. 015.1052662. Levaco, Ronald. 1974. Kuleshov on Film. Oakland: University of California Press. Lewis, Brian. 2008. Jean Mitry. In The Routledge Companion to Philosophy and Film, eds. Paisley Livingston, and Carl Plantinga. London: Routledge. https://doi. org/10.4324/9780203879320.ch37. Merleau-Ponty, Maurice. 1964. Sense and Nonsense. Trans. Hubert L.  Dreyfus, and Patricia Allen Dreyfus. Based on 3rd edn. Evanston Illinois: Northwestern University Press. Mitry, Jean. 1997. The Aesthetics and Psychology of the Cinema. Trans. C. King, Reprint, Illustrated. Bloomington: Indiana University Press. Modell, Arnold H. 2003. Imagination and the Meaningful Brain. Cambridge, MA: MIT Press. Oldham, Gabriella. 1992. First Cut, Conversations with Film Editors. Berkeley/Los Angeles/California: University of California Press. ———. 2012. First Cut 2: More Conversations with Film Editors. Berkeley/Los Angeles/ California: University of California Press. Orpen, Valerie. 2003. Film Editing: The Art of the Expressive. London: Wallflower. Pearlman, Karen. 2009. Cutting Rhythms, Shaping the Film Edit. 1st ed. New York/ London: Focal Press. ———. 2016. Cutting Rhythms: Intuitive Film Editing. 2nd ed. New York: Focal Press. ———. 2018. Documentary Editing and Distributed Cognition. In A Cognitive Approach to Documentary Film, ed. Catalin Brylla and Mette Kramer. Basingstoke: Palgrave Macmillan. Pearlman, Karen, and Adelheid Heftberger. 2018. Editorial: Recognising Women’s Work as Creative Work. In Women at the Editing Table: Revising Soviet Film History of the 1920s and 1930s, eds. Adelheid Heftberger, and Karen Pearlman. Special Issue of Apparatus. Film, Media and Digital Cultures in Central and Eastern Europe 6. Pearlman, Karen, John MacKay, and John Sutton. 2018. Creative Editing: Svilova and Vertov’s Distributed Cognition. In Women at the Editing Table: Revising Soviet Film History of the 1920s and 1930s, eds. Adelheid Heftberger, and Karen Pearlman. Special Issue of Apparatus. Film, Media and Digital Cultures in Central and Eastern Europe 6.



Pudovkin, Vsevolod. 1949. Film Technique; and Film Acting: The Cinema Writings of V.I. Pudovkin. Trans. Ivor Montague; Introduction by Lewis Jacobs. New York: Lear. Reisz, Karel. 1953. The Technique of Film Editing. 1st ed. Boston: Focal Press. Reynolds, Dee. 2007. Rhythmic Subjects, Uses of Energy in the Dances of Mary Wigman, Martha Graham, and Merce Cunningham. Hampshire: Dance Books. Salt, B. 1974. Statistical Style Analysis of Motion Pictures. Film Quarterly 28 (1): 13–22. ———. 1983. Film Style and Technology: History and Analysis. London: Starword. Scherer, Thomas, Sarah Greifenstein, and Hermann Kappelhoff. 2014. Expressive Movements in Audio-Visual Media: Modulating Enactive Experience. In Handbücher Zur Sprach- Und Kommunikationswissenschaft/Handbooks of Linguistics and Communication Science (HSK) 38/2, eds. Jana Müller, Cornelia Cienki, Alan Fricke, Ellen Ladewig, Silva McNeill, and David Bressem, 2081–92. De Gruyter. https:// Schmidt, Paul. 1996. Meyerhold at Work. New York: Applause. Smith, Tim J. 2012. The Attentional Theory of Cinematic Continuity. PRO 6 (1): 1–27. Sobchack, Vivian. 2008. Phenomenology. In The Routledge Companion to Philosophy and Film, ed. Paisley Livingston and Carl Plantinga, Routledge. New York. https:// Stam, Robert. 2000. Film Theory: An Introduction. Malden: Blackwell. Stevens, Kate, Shirley McKechnie, Steven Malloch, and Agnes Petocz. 2000. Choreographic Cognition: Composing Time and Space. In Proceedings of the 6th International Conference on Music Perception and Cognition, ed. J.A.  Sloboda C.  Woods, G.B.  Luck, R.  Brochard, F.  Seddon, and S.  OʼNeill. Keele: Keele University. Sutton, John. 2006. Distributed Cognition, Domains and Dimensions. Pragmatics & Cognition 14 (2): 235–247. Van Leeuwen, Theo. 1985. Rhythmic Structure of the Film Text. In Discourse and Communication. New Approaches to the Analysis of Mass Media, ed. Teun A. van Dijk, 216–234. Berlin: De Gruyter.


Animation David Davies

If cinema, broadly construed, is taken to be the medium of the moving image, then animated cinema differs from other kinds of cinema in the kinds of processes generative of this image. In such talk of the ‘moving image’, the ‘movement’ or change in question is a feature of the content perceivable in the image by an attentive observer. Our perceptual engagement with a non-moving image, such as a still photograph, may take place over an extended period of time, and, as spectators, we may ourselves move relative to the photograph (or, indeed, the photograph may move relative to us, as when we observe a photographic image from the window of a slowly moving train). But there is no movement of or change in the elements entering into the perceived content of the image. On the other hand, in perceptually engaging with a cinematic image, such as a shot in a motion picture, we usually see the things in the picture as moving or changing. I shall begin (section “What Is Animated in an Animated Image?”) by asking what it is for a moving image to be animated, and thus, what it is for something to be an instance of animated cinema. This provides the basis for distinguishing different kinds of cinematic animation (section “Kinds of Animation: A Taxonomy of Techniques”), and may also help to explain (section “Why Has So Little Philosophical Attention Been Paid to Animation?”) why, traditionally, so little philosophical attention has been paid to animated cinema per se and why recent developments in cinematic technology have already begun to remedy this deficiency. I shall comment here on the famous exchange between Stanley Cavell and Alexander Sesonske over the status of animated movies. In the remainder of this chapter, I shall examine (section “Appreciating Animation”) certain epistemological and ontological dimen-

D. Davies (*) McGill University, Montreal, QC, Canada © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




sions of animated cinema and look at the conditions under which animated cinema can be art. I conclude (section “Digital Animation”) by reflecting on the significance of the use of digital technology in the production of animated cinema.

What Is Animated in an Animated Image? Etymologically, to animate something is to give it life, soul, movement or expression. A speaker can become animated when she speaks on a topic that engages her, for example. But it is merely necessary and not sufficient for something to be an animated image that (1) the attentive receiver perceives the content of the image as moving or changing. For one thing, even in the case of some non-cinematic images—certain Italian Futurist paintings, for example— the impression of movement is conveyed even in the absence of the perception of actual movement in the elements that make up the perceptual manifold. So we have to insist that it is perceived actual movement in the latter sense that is required for moving images in general, whether animated or not. Second, animated cinema is standardly contrasted with live-action cinema. In the latter, the presented sequences of images in which we perceive movement are elements in shots that are the result of capture by a camera that continuously records movements and changes taking place in pro-filmic space. What is further required for animated cinema is that (2) the perceived movement or change in the content of the presented sequence of images not be the result of the capture of movement or change occurring independently of the image-­ making process, but be a result of (other) features of the image-making process itself. In looking at a sequence of images resulting from the capture of the performance of a dancer by a non-digital movie camera, for example, we observe something that meets the first of these conditions—we perceive the dancer’s movements, as captured by the camera, in looking at the presented images—but not the second—the movements we perceive in the sequence of images owe their reproducibility but not their existence to the image-making process productive of the recording. It is sometimes said that the distinctive feature of animated moving images is that they give the ‘illusion’ of movement and change. For example, the 2017 Wikipedia article on ‘Animation’ defines the term as ‘the process of making the illusion of motion and the illusion of change by means of the rapid succession of sequential images that minimally differ from each other’. But we need to be clear wherein this ‘illusion’ resides. While this has been disputed,1 some have claimed that when we view the aforementioned recording of a dancer, there is the illusion of movement and change because the immediate source of our visual experience is a rapid succession of still images of the dancer projected

1  Gregory Currie, Image and Mind: Film, Philosophy, and Cognitive Science (Cambridge: Cambridge University Press, 1995), 34–42.



onto a screen.2 The purported illusion here resides in the relation between the sequence of still images of the dancer that are the immediate source of our visual experience and the motion and change of the dancer that we see in watching the movie. In the Wikipedia definition of animation, on the other hand, the proposed ‘illusion’ presumably resides in the absence of pro-filmic events that both (1) have the properties of movement and change that we see in attending to an image and (2) are causally responsible for the fact that, in looking at the images, we see the movements and changes that we do. (I shall ignore, in this context, subtleties relating to the possibility that pro-filmic movements and changes can play an intentionally mediated causal role in relation to the movements and changes that we see in an animated movie—if, for example, pixilation is used to reproduce movements or changes previously recorded on film.) A further point to be stressed here is that, while the movement and change that we see in an animated moving image must be a product of the image-­ making process rather than a property of a thing imaged that is causally responsible for the perceived movement and change, a thing imaged that exists independently of the image-making process may be what we see in looking at the image and may be otherwise causally implicated in the perceived movements or changes. As we shall see below, one process used in the production of cinematic animation—pixilation—creates the illusion of movement and change by combining a sequence of individual still images of a three-dimensional (3D) entity—a person or object—in minimally different positions, with each shot in the sequence being separately posed.

Kinds of Animation: A Taxonomy of Techniques The class of animated moving images, as we have seen, is defined in terms of how the motion or change perceived in a moving image is generated. In the case of the animated image, we have said, this is generated through features of the image-making process other than the capture of pro-filmic movements and changes occurring independently of that process. What kinds of features of the image-making process play this role in animation?3 In perhaps the simplest cases, all that is necessary is a manual process of image generation that produces a sequence of images where each differs minimally from its predecessor in a given scene and a way of presenting this sequence of images to the eye of the receiver at a speed sufficient to produce the appearance of change or movement in the visually presented content. ‘Flip-books’ manually activated by the viewer meet these conditions, as does the much more sophisticated but 2  See, for example, Francis Sparshott, ‘Basic Film Aesthetics’, in Film Theory and Criticism 3rd edition, ed. Gerald Mast and Marshall Cohen (New York: Oxford University Press, 1985), 284. 3  For a brief overview of animation processes, see David Bordwell and Kristin Thomson, Film Art: An Introduction 5th edition (London: McGraw Hill. 1997), 46–49. For a more extensive survey, see Paul Wells, Fundamentals of Animation (Lausanne: AVA, 2006).



­ re-­photographic zoetrope, where images painted on the interior of a cylinder p produce the appearance of movement in the depicted scene when the cylinder is rotated and the viewer looks through slots in its wall. Flip-books and zoetropes are examples of animated cinema in the broad sense of cinema identified earlier. Neither, however, would qualify as cinema in the narrower sense of the word with which many of us are more familiar. Cinema in this narrower sense involves a sequence of (traditionally) photographic images presented to viewers on a screen of some kind. To say that the images are photographic is to say that they are the result of a trace-making process of ‘writing with light’, the images originating in the causal agency of what Henry Fox Talbot4 termed the ‘pencil of nature’. In the case of most of the photographic images of interest to us, the image also reflects intentional rendering of the trace-making process or its result, something that may involve pre-capture choices as to camera, lens, focal length, perspective on a subject, and lighting and/or post-capture choices involving such things as selecting particular images from a contact sheet and the use of developmental processes such as cropping, dodging and burning. As we shall see in the following section, the tendency to think of cinema, including animated cinema, in terms of this narrower conception is largely due to the hegemony of the camera and the projector in the production of moving images, whether animated or not, from the time of the invention of such technological devices until relatively recently. This has also arguably contributed to the marginalisation of cinematic animation in the philosophical literature (see section “Why Has So Little Philosophical Attention Been Paid to Animation?”). For, to the extent that animated cinema is taken to differ from non-animated cinema primarily in terms of how the camera is used and what is placed in front of it in order to produce a moving image, it is tempting to think, as many have done,5 that the interesting philosophical questions about the moving image are inherited from its photographic nature. As noted earlier, animated cinema (in the broad sense) is usually contrasted with moving images that capture ‘live action’. David Bordwell and Kristin Thompson speak of ‘live-action’ cinema in terms of ‘continuously filming an ongoing action in real time’.6 It is not enough for ‘live-action’ cinema that we have ‘the photographic recording of real people and events’7 since, as we shall see shortly, at least one kind of animation (pixilation) would seem to satisfy this definition. Until the advent of digital technology that was usable in animation processes, the production of animated moving images of any degree of complexity relied upon the use of the camera. The latter was used to produce a 4  William Henry Fox Talbot, The Pencil of Nature (London: Longman, Brown, Green, and Longmans, 1844). 5   See, for example, Roger Scruton, ‘Photography and representation’, in The Aesthetic Understanding: Essays in the Philosophy of Art and Culture (London and New York: Methuen,1983); William J. Mitchell, The Reconfigured Eye: Visual Truth in the Post-Photographic Era (Cambridge MA: MIT Press,1992). 6  Bordwell and Thompson, Film Art, 46. 7  Berys Gaut, A Philosophy of Cinematic Art (Cambridge: Cambridge University Press, 2010), 7.



series of individual photographic images where, as in the case of the ‘flip-book’, the photographic content of each image differed minimally from that of its predecessor in a given scene. As in the case of live-action film, the impression of movement and change can be produced by projecting the resulting series of photographic images onto a screen at an appropriate speed. Animated cinema of this sort relies on the same facts about human perception that explain how we see continuous movement and change when viewing the sequence of images making up a photographic recording of live action.8 What distinguishes different kinds of photographically based animated cinema is the nature of the things put before the camera in order to produce a sequence of images capable of being used in this way. In the case of what is often termed ‘traditional’ animation, the things photographed are drawings or paintings, or combinations thereof. Just as, in the case of a flip-book, we get the impression of movement and change by ‘flipping’ a sequence of minimally differing individual drawings, the same kind of impression can be produced if we photograph such a sequence of drawings one at a time and then ‘screen’ the resulting series of images either by means of a projector or on a television. This allows us to transcend the physical constraints on the construction and use of flip-books, and as a result, it is possible to produce moving images of much longer duration and narrative complexity in this way. Equally obviously, however, if the movements and changes presented are complex, involve various different elements in the represented scene and are presented against a variegated background, it will require a considerable amount of skilled labour to produce the individual drawings to be photographed. A major breakthrough in traditional animation technology came in 1915 with the invention by Earl Hurd of the ‘cel’ technique. Cels are transparent rectangular sheets of celluloid upon which marks can be made by means of tools capable of applying media such as paint or crayon. Animation artists can inscribe, on individual cels, marks representing particular elements that are partly constitutive of the content to be represented in a given image belonging to a series that is to be photographed to represent a given scene in an animated movie. As in the case of the drawings used in traditional animation prior to the invention of the cel technique, the elements represented in a given image that is part of a scene may include a more or less unchanging background, stable objects that may change their positions in relation to that background and things whose movements and changes are represented in the sequence of images making up that scene. By breaking down the total content of individual images into such component elements and inscribing different elements onto separate cels—an opaque background on one cel, foregrounded objects relative to which the characters are moving on other cels and a moving or changing character—say a dog—frozen at a particular point in its movement or change on yet another cel, the animation artist can reconstruct the required content for each image by stacking the relevant cels on top of one another. The cel 8

 See, for example, Gaut, A Philosophy of Cinematic Art, 8.



representing the background is placed at the bottom of the stack and the other cels can be suitably ordered on top of this. The resulting stack can then be photographed to produce one in the sequence of photographic images required for presentation of the scene in question. The next image in this sequence can be produced by generating another stack of cels, reusing some of those just used—in particular the cel representing the background—and creating the new cels necessary if the screening of the sequence is to give the impression of movement and change. One of the cels required for this following image will represent the next stage in the movements of the dog—in our example—to be represented in the animated film. The cel technique reduces the amount of labour involved because there is no need to exactly redraw the background or the relatively immobile elements each time in generating a sequence of images. It also allows for the labour required for the entire process to be divided up between a team of animators with different individual responsibilities. An alternative technique that was used in the first feature-length animated movie—Lotte Reiniger’s 1926 film The Adventure of Prince Achmed—is to use two-dimensional (2D) jointed cut-­ outs that can be posed against a painted background and photographed to produce the individual images in a sequence that represents movements and changes. The second major class of photographically based animation techniques also involves the generation of the appearance of movement and change through combining a series of individual photographic images whose representational content undergoes minimal change from shot to shot. In this case, however, what are placed before the camera are not themselves images—whether drawings or cels or cut-outs—but three-dimensional objects that are minimally repositioned between successive shots. There are a number of different varieties of what is termed ‘stop-motion’ animation. These differ according to what it is that is manipulated or reposed between the taking of the individual photographic images that make up the material basis of the film. Because of their malleability, clay and plasticine have been popular media, and the manufacture and manipulation of clay figures is best known through the technique of ‘claymation’ used in films such as Gumby and Wallace and Gromit. A second kind of stop-motion animation employs three-dimensional models and structures that are manipulable in virtue of various articulations built into them. Such techniques are used in Charlie Kaufman’s Anomalisa, for example. Third, stop-­ motion animation may employ real people or objects where these can be reposed between individual shots. In the case of ‘pixilation’, as this technique is usually called, real people and real things enter into the content of the moving images, but the images do not result from the capture of real movements and changes on the part of those entities, even though the entities are represented as moving and changing in the ways that result from the stop-motion techniques employed. Pixilation has been used in many music videos. All of the animation techniques listed thus far fit our intuitive idea of a movie as something that involves the use of photography in the production



stage and the use of screens in presenting the movie to an audience. It is worth noting, however, that animated movies that employ film can be produced without using film photographically to ‘capture’ pro-filmic objects or events. Some experimental film-makers (e.g. Stan Brakhage, in Mothlight) produced animated films by directly marking the individual frames on a spool of film to produce the appearance of movement or change when the resulting images are projected. The development of digital technology has made it possible to produce animated movies, or to incorporate animated elements in otherwise ‘live-action’ movies, without relying upon the photographic capture of pro-filmic things and events such as drawings, structures or real people and objects. Digital animation, to which we shall return in the final section of this chapter, can itself take many forms. The simplest examples involve the use of a computer to perform the tasks performed manually and/or photographically in traditional or stop-motion animation, such as the construction and combination of cels. The resulting sequence of images can then be either composed into a computer file—to be screened on a computer screen or by means of a digital projector— or converted into analogue form and encoded on a film strip for non-digital screening. Certain kinds of computer software are able to ‘bridge’ between cels to produce sequences of images representing continuous movement and change without the need to produce individual cels that represent the bridged elements. Digital technology also allows for the manipulation and transformation of images resulting from photographic capture to produce movements and changes that originate in the digital processing rather than in the events captured, thereby satisfying our definition of animation. But digital technology also allows for the production of moving images that are entirely the result of digital processing: software programs or algorithms can be used to directly generate digital files that, when screened, are experienced by the viewer as moving images. Such animated sequences can also be digitally combined with live-action images to represent things and events that do not exist pro-­ filmically—for example, the dinosaurs in Jurassic Park, the eponymous King Kong in the remade edition and Gollum in The Lord of the Rings. A principal exponent of this kind of digital animation has been the Industrial Light and Magic company founded by George Lucas. One technique that has been used in various ways in animated cinema since its early days is rotoscoping, and the latter raises some interesting questions about the definition of an animated image presented in section “What Is Animated in an Animated Image?”. Rotoscoping uses live-action footage as the basis for the production of what are generally taken to be animated films. The rotoscoper takes a live-action shot of certain pro-filmic events and traces out the moving figures or objects in the individual frames constitutive of the shot. These tracings—which, if they are only of the moving or changing elements, will abstract from background features of the traced frames—can then be artistically manipulated in various ways—for example, by painting in details or by imposing the traced figures on a new background. The (modified or u ­ nmodified)



tracings can then be photographed to produce a sequence of images that reproduce the pro-filmic movements and changes captured in the original live-­action footage. In a number of recent ‘animated films’ by directors with prior artistic credentials (e.g. Richard Linklater), use is made of much more sophisticated computerised techniques, such as the ‘interpolated’ rotoscoping developed by Bob Sabiston in the 1990s. I shall say more about some of these recent films at the end of section “Appreciating Animation”, but it is worth noting here that, while most films that employ rotoscoping also use other more standard animation techniques, a film that used only rotoscoping might fail to count as animated cinema given the way in which the latter was defined above. For, in what we might term a ‘strictly’ rotoscoped film, the movements and changes perceived by the viewer are the result of the original live-action capture, even if the process of rotoscoping does not preserve the background of the original capture and presents only traced images of the movements of the central figures or objects that were the subjects of that capture. While the image-making process might substantially change the appearance of the moving and changing objects, it would be the live-action footage that is the source of the movements and changes themselves, in virtue of its having captured movements and changes occurring pro-filmically. An interesting example that might seem to exemplify this kind of strictly rotoscoped film is Loving Vincent (2017), co-directed by Dorota Kobelia,9 This film, about events in the life of Vincent Van Gogh, was initially shot in live action in a studio with actors performing against a green screen. After editing, a crew of experienced artists were assigned separate shots in the film and used oil paint to paint the individual frames constitutive of their assigned shots. Over 100 professional artists were employed in making the film, and their activity resulted in 65,000 oil paintings, many of which were not included in the final print. In painting the frames, the artists used Van Gogh’s own visual vocabulary for both the characters (rotoscoped from the live-action footage) and the backgrounds. It appears, however, that some shots in the film were animated in a more traditional way. In the BBC documentary on the making of the film, one painter is shown producing the 12 images that make up a one-second shot of a woman turning around in front of a church. The woman’s movements were not, it would seem, a result of live-action capture. Do examples like this suggest that we need to rethink our definition of animated cinema, or should we respond that the precise movements and changes that we see in such a film reflect not only what was captured through live-­ action but also the work that the artists have done with the resulting frames? This is a question for which we are unlikely to have, and perhaps should not seek, a clear and principled answer. In this respect, it resembles the question (see section “Digital Animation”) as to when the digital manipulation of a 9  See the 2017 BBC documentary on the making of Loving Vincent, accessed 28 September 2017,



photographic trace results in a loss of the indexicality usually taken to be definitive of photographic images. It is important to note that a given animated movie may, in fact, result from the use of many different animation techniques. The 2002 Festival at the Centre International du Cinéma d’Animation, at Annecy in France, for example, included films containing one or more of the following techniques: 2D and 3D computer animation; claymation; sand; recyclomation; drawing on cels; cut-outs; puppets; live action; ink, pencil, or gypsum on paper; pixilation; animated objects; paint on glass; engraving on film; and photography. One particular short animated film—(The Rise and Fall of the Legendary) Anglobilly Feverson, edited by Rosto—while lasting less than ten minutes, involved a remarkable range of different techniques: Cut-outs, Pixilation, 2D Computer, Animated objects, Paint on paper, Puppets, 3D Computer, Photocopies, Photos, Collage, Mixed techniques and Internet Software.

Why Has So Little Philosophical Attention Been Paid to Animation? Until very recently, little if any serious philosophical attention has been paid to animated cinema. As we shall see, even the attention paid to it recently has been parasitical upon a broader interest in the bearing of digital technology on long-standing philosophical issues concerning photography and film. We shall look at the relationship between animation and digital technology in the final section of this chapter. Here our concern is with why, prior to the interest in digital animation, so little philosophical attention has been paid to animation per se. For example, the voluminous Routledge Companion to Film and Philosophy,10 published less than ten years ago, contains neither an entry on animation nor any references to it in the Index. Part of the reason for the lack of interest in animation on the part of philosophers of film and film theorists is that, for the most part, their interest has been in establishing that the moving image is a serious medium, and indeed a potentially artistic medium, in the face of the criticisms levelled, from the outset of photography, at the products of photographic processes. The latter were characterised as ‘merely mechanical’, and as capable only of documenting pro-filmic things and events.11 To the extent that cinema in the narrow sense is taken to be the result of a photographic process, challenges to the artistic pretensions of photography also infected such pretensions on the part of film. We find a concern with responding to these challenges in the opening pages of Rudolph 10  Paisley Livingston and Carl Plantinga, eds., The Routledge Companion to Philosophy and Film (London & New York: Routledge, 2009). 11  For detailed histories of these criticisms, see Noël Carroll, Philosophical Problems of Classical Film Theory (Princeton: Princeton University Press, 1988), 20–29; Patrick Maynard The Engine of Visualization: Thinking Through Photography (Ithaca: Cornell University Press.1997), 269–77, 290–93.



Arnheim’s Film as Art,12 where they are countered by a detailed account of the ways in which the film-maker can use both composition and editing to represent a subject or scene in a certain way, thereby ‘embodying in pictorial form a thought about’ that subject—the very thing that Roger Scruton later argued could be done in paintings but not in photographically generated images.13 Another central figure in classical film theory, Andre Bazin, stressed the ways in which film could serve our interests in artistic realism through exploiting the possibilities not of editing but of cinematography.14 In a different vein, Jean-­ Luc Godard’s cinematic practice and theorising were concerned with exploring the social and political possibilities of film, and, indeed, with whether the medium could offer an ‘epic’ cinema analogous to Brecht’s ‘epic’ theatre. Stanley Cavell,15 whom some have viewed as the last of the classical film theorists, had a more limited goal although one pursued with great philosophical seriousness: he sought to establish the intellectual and affective power of mainstream Hollywood cinema. For philosophers of film or film theorists having these kinds of agendas, animation, if attended to at all, might seem to be part of the problem rather than part of the solution. The majority of animated films were intended for a mass audience, often a predominantly juvenile one. The makers of such films had little obvious concern with artistic values or social and political critique. Furthermore, animated cinema seems to offer relatively little in the way of the things cited by philosophers and theorists in responding to challenges to the seriousness of cinema in the narrow sense, as illustrated earlier. Composition and editing might seem to play limited expressive roles in non-digital animation, in part because the very labour-intensive nature of the production of animated cinema dictates that the entire film be carefully storyboarded before the actual generation of the images begins, and also in part because the kind of compositional changes possible in live-action cinema require considerable work if they are to be the result of non-digital animation. Cinematography has no place in analogue animated cinema, even if animators can try to generate images that mimic cinematographic achievements, and there seem to be few opportunities, in animated movies, for the kinds of emotional engagement achievable with characters portrayed by skilled actors, even though the medium can be used to produce certain kinds of emotional effects—consider the portrayal of the death of Bambi’s mother! (For responses to some of these claims, see section “Appreciating Animation”). In place of these ‘artistic’ elements central to the making of traditional live-­ action cinema with artistic pretensions, the salient elements in the making of most predigital animation cinema were perceived to be the artisan-like skills  Rudolph Arnheim, Film as Art (Berkeley CA: University of California Press, 1957).  Scruton, ‘Photography and Representation’. 14  André Bazin, What is Cinema? Vol 1, trans. Hugh Gray (Berkeley: University of California Press, 1967). 15  Stanley Cavell, The World Viewed (Cambridge MA: Harvard University Press, 1971); The World Viewed: Enlarged edition (Cambridge MA: Harvard University Press, 1979). 12 13



required to generate the sequences of images used in animation, the division of labour and semi-production-line methods used to generate such sequences given the availability of cel technology, and the simplicity and limited expressive potentiality of the image content. In short, cinematic animation did not obviously engage with the kinds of interests driving most theoretical and philosophical reflection on cinema, whether broadly or narrowly construed: (1) an interest in the use of the cinematic medium to make narrative artworks, where the skills of directors, cinematographers and actors are utilised in the interests of telling a compelling and perhaps socially valuable story, and (2) an interest in the use of different potentialities of the medium for expressive or realistic purposes. As the brief survey in the previous section might suggest, the interesting questions about animated cinema seem to relate to the different techniques used, but these techniques, over and above the visual arts skills used in producing drawings or cels, seem to be of technological rather than artistic interest. Furthermore—something to which we shall return later—there was for some time considerable resistance in philosophical writing on the arts to thinking that the history of making of an art object plays a crucial part in the appreciation of the resulting artwork. Thus, if there were certain animated films that were held to have artistic or social value, this was taken to depend upon what they shared with live-action cinema rather than upon what distinguished them from it. This is not to say that animation has been altogether excluded from the discourse surrounding classical film theory. But even its most prominent intrusion into this discourse turns out, upon closer inspection, to be in the interests of clarifying certain features of live-action cinema. I refer here to the widely discussed exchange between Stanley Cavell and Alexander Sesonske. Sesonske,16 reviewing Cavell’s The World Viewed, challenged Cavell’s definition of cinema in the narrow sense on the grounds that it excluded animated cinema. For Cavell, the material basis for moving pictures is a succession of automatic world projections. ‘Automatic world projections’ are the result of photographic capture of pro-filmic events. Sesonske notes that Cavell is assuming here that moving pictures are products of photographic processes, and that his principal claims about the ontological and epistemological significance of cinema rest on this assumption. Because a moving picture, as an automatic world projection, is a trace of some independently existing reality not otherwise accessible to us, to screen a moving picture is also to screen the viewer from this reality, and thereby to raise through its very nature traditional philosophical concerns about scepticism. The camera, says Cavell, ‘is fated to reveal all and only what is revealed to it’: whatever is screened in a moving picture must have existed before the camera but is no longer accessible to us. Sesonske raises two problems with such a view. First, if the elements of which a photographically generated moving picture is composed are ‘­automatic world16  Alexander Sesonske, review of The World Viewed, by Stanley Cavell, The Georgia Review 28 (4) (1974), 561–70.



projections’, it doesn’t follow that the moving picture itself is such a projection, or that the actions, events and relations that are screened must have existed as part of the pro-filmic reality. To think that it does follow is to ignore the significance of editing, inter alia, and also the distinction between the pro-filmic events—actors on a film set—and the fictional ‘world’ presented to the viewer. Second, to think that all photographically generated motion pictures are ‘automatic world projections’ in the designated sense is to ignore those animated motion pictures produced by photographing either drawings or cels, as in ‘traditional’ animation, or three-dimensional objects, as in various kinds of stop-motion animation. The films produced by the Disney Studios featuring ‘stars’ such as Mickey Mouse and Donald Duck were experienced by audiences no differently from photographically generated ‘live-action’ movies, Sesonske maintains. In watching such films, we experienced ‘a world present to us while we were not present to it, with the same immediacy and conviction, the same sense of moving through its space, the same feeling of intimate acquaintance with its inhabitants’.17 But, as Sesonske stresses, the world present to us in looking at such animated movies is not one that existed prior to the moment of projection. All that existed prior to the latter were the drawings or other entities recorded photographically in the individual shots making up the sequence of images that present to us the world we experience in watching the movie. Sesonske observes that the role of editing in bringing into existence the world we view in watching an animated film reminds us that, even in the case of live-action cinema, the picture itself is not an ‘automatic world projection’. This undermines the attempt to confer on our movie experience the kind of philosophical significance claimed by Cavell: ‘the fact that the camera necessarily did its work at some time before the finished film is screened does not at all necessitate that the world I see is a world past’.18 Indeed, Sesonske’s point, in introducing animated cinema, is less to make a case for the latter than to give a clear example of what he takes to also be the case in the kinds of Hollywood movies dear to Cavell. In the latter as well as in the former, the world projected depends upon various choices made in both production and post-production, and only in virtue of such choices is what is projected sometimes the world as it exists independently of the activities of the film-makers. This point is crucial when we consider Cavell’s response to Sesonske.19 He simply denies that cartoons, and presumably animated moving pictures more generally, are movies. They are not movies, he maintains, because the laws governing the fictional worlds of cartoons are unclear, and as a result of this and the other ways in which the narrative is presented, we cannot be engaged by them emotionally and intellectually in the ways that we can with live-action cinema. Since his concern in The World Viewed is with movies taken to be things that, when well done, can engage us in these ways, he sets aside Sesonske’s criticisms as irrelevant to his intended subject. As noted earlier, it is  Sesonske, review of The World Viewed, 563.  Sesonske, review of The World Viewed, 566. 19  Cavell, The World Viewed: Enlarged Edition, 166ff. 17 18



the achievements of Hollywood cinema that he wishes to explore and account for, and it is films that have the ability to achieve such things that he is defining as movies. As may be clear, however, this response misses Sesonske’s point entirely if, as I have suggested, animated moving pictures are primarily used to illustrate the constructed nature of projected worlds in general. Even in the kinds of Hollywood films celebrated by Cavell, we do not coherently engage with the projected worlds as ‘automatic world projections’.

Appreciating Animation In this section, I want to look at continuities and discontinuities between the appreciation of animated movies and the appreciation of other forms of cinema. Reflecting an ambiguity in the notion of ‘appreciating’ any artefact intended to entertain, stimulate, inspire, challenge or otherwise engage a receiver, there are a number of distinct issues here. In one sense, to appreciate such an artefact is to have the kind of experiential engagement with it that its maker(s) intended or that it affords independently of its maker(s)‘s intentions. Insofar as the artefact in question presents to the receiver some kind of perceptual or conceptual manifold and that it is possible, whether or not in accordance with the maker’s intentions, to grasp certain features of that manifold and relate them to one another and to things external to the manifold in valuable ways, the question about appreciation is a question about the kinds of cognitive capacities involved in achieving such a grasp. ‘Cognitive capacities’ here may involve perceptual or conceptual skills, motor skills of various kinds and affective capacities. But we also talk of appreciating an artefact as the kind of artefact that it is, where this requires not so much our receptivity to what a manifold has to offer us but something closer to an understanding of how, and how well, the artefact embodying such a manifold serves its intended purposes and how it does so in virtue of the ways in which it has been crafted by its makers. And a third question is whether a given artefact lends itself to the kinds of appreciation—in both of these senses—that might be thought to be distinctive of, or appropriate for, the things that we think of as artworks, in some sense of this term. I shall consider how the appreciation of animated cinema compares to that of ‘live-action’ cinema in each of these senses of ‘appreciation’. The movements and changes, and the things moving and changing, that are represented in animated cinema sometimes differ significantly from the movements and changes, and things moving and changing, represented in live-­action cinema, and from those movements, changes and moving and changing things encountered in our non-cinematic experience. Indeed, as we shall see later, one of the attractions of animated cinema is that it enables us to present to the receiver things and events unlike anything she could encounter in the actual world or in any cinematic images derived by simply capturing pro-filmic events. Whether it is a matter of large talking mice or aliens and superheroes with powers no human agent possesses, our engagement with animated movies is often motivated by the pleasures we derive from exploring such supra-­real worlds.



But, in spite of this, it isn’t clear that the cognitive and other capacities upon which we draw in watching animated movies are significantly different from those that we employ in cinematic experience more generally. Indeed, our ability to effortlessly process the visual information furnished by even the more outlandish animated movies testifies to our mental versatility. It is plausible to see this as a further extension of the visual and cognitive capacities employed in making sense of pictures more generally, and live-action movies in particular. In considering our ability to understand what is being portrayed in live-action cinema, Gregory Currie20 argues that, in grasping the content of representational visual images, we utilise the same perceptual skills that enable us to recognise the things represented. In recognising a picture of a horse, we mobilise the same set of feature detectors that enable us to recognise, with considerable reliability, actual horses. He talks here of the ‘natural generativity’ of pictorial representation—depiction—contrasting this with verbal representation— description. While in the case of pictures, I can usually recognise an x-depiction when I can recognise an x, we can grasp the content of a description only if we know the relevant linguistic conventions. Extending this idea to animation, we can say that we recognise the nature, movements and mental states of an animated character by drawing upon the same sorts of discriminatory capacities that are engaged in watching live-action movies, where the latter in turn draw on the same capacities that we exercise in recognising and making sense of the kinds of actual things represented. Makers of animated movies depend upon our ability to extend these kinds of recognitional skills to animated representations of horses, mice and ducks. It is a salient fact about us—upon which the very possibility of such animated movies depends—that we do process the information in such images in this way, and, indeed, seem willing to invest almost anything represented as being capable of self-generated motion with a mental life. This is not to ignore the conventional dimension in much animated cinema. The representation of a horse in Japanese anime stands in a relation to the tradition of stylised representation in Japanese drawing, just as European and American animations typically mirror stylised modes of representations in Western visual art, but this is not usually an obstacle to recognition for those unaware of those traditions. Nor is it clear that there is a striking difference in subjective aspects of our experience of animated and live-action movies—this is the substance of Sesonske’s plausible claim (cited earlier) about our engagement with the worlds of animated and live-action cinema. In the second sense of appreciation distinguished earlier, the salient question is whether knowledge of the particular techniques employed in producing a given animated movie is necessary for appreciating that movie as the artefact that it is. One’s response to this question is likely to mirror one’s views about the appreciation, in this sense, of artefacts in general. For those who subscribe

 Currie, Image and Mind, 80–88.




to broadly ‘empiricist’ views of appreciation,21 appreciating an artefact as the artefact that it is may involve some kind of understanding of how the artefact does what it does, but does not concern itself with how this functionality was conferred upon it by the actions of its makers. Empiricism as a view about the appreciation of artworks has been challenged by a number of theorists over the past 40 years.22 An influential resource in these debates has been the art historian Michael Baxandall.23 Baxandall prefaces his defence of an anti-empiricist account of the appreciation of pictorial artworks by considering what it would be to appreciate an artefact like the Forth Railway Bridge. To appreciate the latter, he argues, is to apprehend it as ‘a solution to a problem in a situation’. Full appreciation, in the sense that concerns us here, of an artefact or artwork requires that we also appreciate the achievements of the makers, what have been termed the ‘achievement properties’ of the work. Those who are moved by these kinds of general arguments for the relevance of achievement properties to appreciation in our second sense are likely to hold that the appreciation of an animated movie requires that we comprehend the various processes of animation that went into its creation. Indeed, it might be thought that animated movies provide a particularly striking example of—and argument for—the more general thesis concerning the appreciative relevance of achievement properties.24 As we saw in section “Kinds of Animation: A Taxonomy of Techniques”, not only is there a plethora of different animation techniques, but there can also be animated movies with perceptually very similar properties that are the result of radically different processes of animation. Where a single movie can draw upon a wide range of different animation techniques, it can be argued that we need to have recourse to information and evidence not given to us in an unmediated perceptual engagement with such an artefact in order to properly appreciate it as the artefact that it is. Furthermore, if we are ignorant of the workings of the cel technique, we would radically overestimate the amount of labour involved in producing even a fairly simple animated movie, such as a Betty Boop cartoon, since we would assume that each photographed frame had to be produced as an individual image in all of its detail. Similar issues arise in the case of films made using rotoscoping, as discussed in section “Kinds of Animation: A Taxonomy of Techniques”. Our third question was whether there are animated movies that are properly accorded the kinds of appreciation—in both senses—that we accord to those things that we take to be artworks (‘properly’ here in the sense that these movies are indeed correctly taken to be artworks rather than ‘properly’ in the sense 21  For example, Clive Bell, Art (London: Chatto and Windus,1913); Alfred Lessing, ‘What is wrong with a forgery?’, Journal of Aesthetics and Art Criticism 23.4 (Fall 1965), 461–71. 22  For example, Gregory Currie, An Ontology of Art (New York: St Martin’s Press,1989), chapter 2; Jerrold Levinson, ‘Evaluating music’, in Musical Worlds, ed. Philip Alperson (College Park PA: Penn State Press,1998), 93–107; David Davies, Art as Performance (Oxford: Blackwell, 2004), chapters 2 & 3. 23  Michael Baxandall, Patterns of Intention (New Haven CN: Yale University Press, 1985). 24  Davies, Art as Performance, 68.



that it is productive to treat them as if they were artworks). Caution is necessary because the term ‘art’ is used in different senses in raising such questions. In a broad sense most familiar in the case of paintings, an artefact counts as art if the processes and materials entering into its making are the same as those that enter into the making of things that are unquestionably artworks. In this sense, what a child produces in nursery school when using crayons and paint to make marks on sheets of paper is art. However, we are less inclined to classify as ‘art’ even in this broad sense every sequence of sounds produced by the playing of anything reasonably classified as a musical instrument, or every photograph taken by a tourist on holiday. Fictional narratives generated by children might be called ‘literary art’, in this broad sense, but it sounds odd to say so save in very particular cases. In a narrower sense, however, we speak of ‘works of art’ with the intention of picking out artefacts that are accorded a distinctive role or value in a culture. Part of the confusion in debates about the artistic status of ‘Outsider Art’ or ‘tribal’ artefacts is a result of failing to distinguish these two senses in which something can rightly be described as art.25 The confusion is compounded when we consider a category like ‘mass art’ that comprises both things that are art in the narrower sense and things that are art in the broader sense. Noël Carroll26 rightly points to the significance of the intentions of the makers in delimiting the scope of the category of mass art, but on many conceptions of ‘high’ art this allows for some works of mass art to also qualify as high art (e.g. Orson Welles’s A Touch of Evil). Most animated cinema falls clearly under the label ‘mass art’—this is particularly the case with animated films made for television viewing such as The Flintstones, Yogi Bear, The Simpsons and South Park. But it is undeniable that, on any plausible account of what qualifies certain motion pictures as cinematic art in the narrower sense, there are works of animated cinematic art in this sense. Although, as we noted in section “Why Has So Little Philosophical Attention Been Paid to Animation?”, classical film theorists interested in defending the artistic pretensions and ‘seriousness’ of moving pictures paid scant attention to animation, there is no reason why the kinds of features to which they pointed in their defences of cinematic art cannot also be realised in animated cinema. Animated films, no less than live-action films, can embody in pictorial form thoughts about their subjects and thereby qualify as ‘expressive’ in Arnheim’s sense, for example. Animated films differ from films that use live actors and locations, in that the movements and changes perceived in the temporally unfolding visual manifold are the result of one or more of the kinds of animation processes described in section “Kinds of Animation: A Taxonomy of Techniques”. But many of those who work as animators certainly think of what they do as making artistic contributions to a film ­comparable to 25  See David Davies, ‘On the Very Idea of “Outsider Art”’, British Journal of Aesthetics 49.1 (Winter 2009), 25–41; Larry Shiner, ‘“Primitive Fakes”, “Tourist Art” and the Ideology of Authenticity’, Journal of Aesthetics and Art Criticism 52.2 (Spring 1994), 225–34. 26  Noël Carroll, A Philosophy of Mass Art (Oxford: Clarendon Press, 1998), 196.



the contributions made by actors. Berys Gaut27 notes here the common practice of referring to animators as ‘actors with pencils’. If there are works of animated movie art in the narrow sense, this raises at least two questions: (1) who are the artists to whom such works should be credited, and (2) for what kinds of reasons might a director aspiring to produce a cinematic work meeting whatever we take to be the conditions for being cinematic art choose to produce an animated work. (1) Berys Gaut28 has argued that cinematic artworks are better thought of as multiply rather than singly authored, given the different kinds of contributions that directors, screenwriters, actors, score-writers and cinematographers, inter alia, make to the artistic qualities of a feature film. In arguing for this conclusion, Gaut rejects the idea that, because the director exercises ‘sufficient control’ over the creative contributions of others, she is the sole author of the films she directs. While he argues convincingly against this claim in the case of live-action cinema, it might be wondered whether there might be a stronger case, on such grounds, for the single authorship of animated movies. It might be said that the contributions of many different individuals to the construction of the cels used in the production of Disney animated movies, for example, are strictly supervised by the film’s directors, and are in any case acts of craft rather than of art. Possibly for these kinds of reasons, it was the practice in early animated cinema to give all authorial credit for animated movies to the person(s) who oversaw the processes of animation, and some have argued for genuine single authorship in more recent animated films.29 But contemporary descriptions within the industry of animators as ‘actors with pencils’ suggest that Gaut’s arguments against the ‘sufficient control’ strategy, if sound for live-action cinema, are also sound for animated cinema. (2) One reason for making an animated cinematic artwork was alluded to earlier. Animation can make visually real for an audience fictional worlds that depart in striking ways from the actual world, and can do so using either more traditional methods of animation or digital animation interpolated invisibly into live-action footage. A more interesting reason has been voiced by Richard Linklater in interviews about two of his films that use Bob Sabiston’s ‘interpolated rotoscoping’, Waking Life and A Scanner Darkly. Speaking of the former film, Linklater30 notes that it was first shot as a live-action film and then animated by the computer-­ enhanced interpolated rotoscoping process. Linklater makes clear that  Gaut, A Philosophy of Cinematic Art, 138–39.  Berys Gaut, “Film Authorship and Collaboration”, in Film Theory and Philosophy, ed. Richard Allen and Murray Smith (Oxford: Clarendon Press, 1997), 149–72. 29  For example, Paul Wells, Animation: Genre and Authorship (London: Wallflower Press, 2002). 30  Richard Linklater, interview (2001) with Spence D. on A Waking Life, accessed 03 September 2017, 27 28



the film was not originally conceived as an animated film, and would in fact not have been completed and released had it not been for his encounter with Sabiston’s rotoscoping process. Given that the film deals in ‘unrealities’, Linklater states, he was unhappy with the ‘look’ of the live-action footage but found what he was seeking in the animated images produced by rotoscoping: ‘Up to that point the film was too blunt, too realistic. I think to make a realistic film about an unreality the film had to be a realistic unreality. You know, the animation is that. It’s realistic because it’s real voices, real gestures, real people and yet it’s unreal because it’s ultimately an animation of that. And yet it’s real world based, so it’s the perfect kind of place to watch this movie from’. As is clear from these remarks, the motivation in this case for producing an animated film was an aesthetic one, given the content of the film. In the case of Kaufman’s Anomalisa, shot in stop-motion animation using multiple 3D-printed manipulable figures produced for each of the principal characters, the motivation relates less to the look of the film and more to the metaphorical implications of the animation process itself. Kaufman,31 asked about his reasons for spending two years in producing a film that could otherwise have been shot in a matter of months, stated that ‘there’s something about this type of animation that communicates fragility and humanity and brokenness … because it’s all handmade, and because it’s an imperfect process’.

Digital Animation As noted earlier, increased philosophical attention to animation in the past decade or so has been prompted by a more general interest in the use of digital technology in the production of both still and moving images. As some recent commentators have noted, the use of digital technology can play a number of significantly different roles in the making of wholly or partly animated moving pictures. Some have also claimed that this helps to illuminate the philosophical significance of animation. One writer who has drawn such a connection between philosophical issues relating to animated cinema and philosophical issues posed by digital cinema is Philipp Schmerheim,32 in an article that discusses inter alia the Cavell-Sesonske dispute. Sesonske, as we have seen, points to animated cartoons—of an analogue nature—that fail to satisfy Cavell’s description of movies as ‘successions of automatic world projections’.33 Such films may be projections of a world that comes into existence at the time of 31  Charlie Kaufmann, interview about Anomalisa, Fresh Air, NPR, December 22, 2015, accessed 30 October 2017, 32  Philip Schmerheim, ‘Scepticism’, in The Routledge Encyclopedia of Film Theory, ed. Edward Branigan and Warren Buckland (London and New York: Routledge, 2014), 413–19. 33  Cavell, The World Viewed, 72.



screening, but they are not projections of aspects of the actual world captured by a camera. Sesonske’s claim is in part epistemological—we engage with the screened world in the same way whether it be a result of live-action capture or a result of traditional or stop-motion animation. But, as I suggested earlier, his larger point seems to be that nearly all screened worlds are constructed worlds: the world of Jules et Jim is no less constructed than the world of a Disney cartoon even if live-action capture of movement and change in pro-filmic events plays an essential part in the former but not in the latter. Schmerheim ignores this more general point and suggests that Sesonske’s arguments prefigure the challenge that is posed to Cavell by digital cinema: they illuminate what would be required of ‘a “digital version” of Cavell’s film ontology under the assumption that animated cartoons share basic features with forms of digital cinema, such as extensive use of animated elements, which in turn have a weaker indexical relation (or none whatsoever) with whatever it is we call reality’. ‘Indexicality’ here is a matter of one thing’s indicating something else in virtue of standing in the right kind of causal relation to it. Certain marks in the sand, for example, indicate the recent passing of a bear, and the rings visible when one fells a tree indicate the age of that tree. Similarly, the distribution of marks on the surface of an analogue photograph indicates certain aspects of pro-filmic reality at the time when it was taken. This is because of the ‘mechanical’ nature of the photographic process. It is in virtue of being mechanical recordings in this sense that photographic images are indexical. It is this fact about the making of a photographic image that leads Lev Manovich,34 for example, to describe traditional cinema, using analogue photography, as grounded in ‘deposits of reality’ and as ‘an attempt to make art out of a footprint’. Indexicality is central to a still image’s being photographic because it is a consequence of the trace-making process whereby photographs are generated. Some theorists have maintained that one loses indexicality, and, consequently, the photographic status of the image, when an image is generated digitally. The most influential proponent of such a view is William J.  Mitchell.35 Mitchell argues that the digital image is non-photographic because it is shorn of those features traditionally held to distinguish photography from painting. Whereas the traditional photographic process allows the photographer relatively little opportunity to change the resulting image, the digital image is ‘inherently mutable’: ‘The essential characteristic of digital information is that it can be manipulated easily and very rapidly by computer. It is simply a matter of substituting new digits for old’.36 Furthermore, because we are aware of this mutability, we cannot with any confidence draw conclusions about the nature of 34  Lev Manovich, ‘What is Digital Cinema?’, published online in 1995, accessed at http:// on 1 November 2010. No pagination. 35  William J. Mitchell, The Reconfigured Eye: Visual Truth in the Post-Photographic Era (Cambridge MA: MIT Press, 1992). 36  Mitchell, The Reconfigured Eye, 7.



pro-filmic reality from features of the digital image. In looking at the latter, we have no way of determining which features of the image indicate aspects of the pro-filmic world and which reflect the intentional activity of the photographer. This calls into question the traditional way of marking the difference between (indexical) photographs and (non-indexical) paintings. Traditional photographs ‘were comfortably regarded as causally generated truthful reports about things in the real world. But the emergence of digital imaging has irrevocably subverted these certainties, forcing us to adopt a far more wary and more vigilant interpretive stance’.37 Thus, Mitchell maintains, ‘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’.38 Manovich draws an even stronger conclusion. Where Mitchell claims that digital photography departs from analogue photography as much as the latter departs from painting, Manovich39 claims that the inherent mutability of the digital image ‘erases the difference between a photograph and a painting’. Should we agree with Mitchell and others about the epistemological and ontological implications of the move from (still and moving) analogue images to digital images? And to what extent does the debate about the latter point to philosophically distinctive features of animated moving images? These questions are best addressed in the context of a more general account of how digital technology can enter into the making of still and moving images. In his A Philosophy of Cinematic Art, Berys Gaut distinguishes three ways in which this can happen.40 First, it may play a part in encoding the results of a trace-making process of ‘writing with light’. Most obviously, this occurs when pro-filmic events are captured with a digital camera. Such a camera may differ from a camera employing photochemical technology only in how it records the light rays stemming from the pro-filmic event. While the trace produced photochemically by a standard analogue camera is a negative, the trace produced by a digital camera is a bitmap in which ‘the integer...stored at each pixel... encodes information about the light emanating from the part of the object that the pixel represents’. Second, digital technology can enter into the generation of an image through ‘painting’, the alteration, by means of a software editing tool, of the information encoded by the pixels in a bitmap. Finally, an image can be directly generated by a computer through the execution of an algorithm, often by means of a computer-generated 3D model to which the algorithm is applied. Digital animation is the result of the use of digital technology to produce a moving image where the appearance of movement and change results from features of the image-making process other than the capture of pro-filmic  Mitchell, The Reconfigured Eye, 225.  Mitchell, The Reconfigured Eye, 4. 39  Manovich, ‘What is Digital Cinema?’. 40  Gaut, A Philosophy of Cinematic Art, 16. 37 38



movements and changes. Digitally animated cinema is then one kind of digital cinema, but digital processing does not by itself result in an animated image. For example, early Dogme films such as Thomas Winterburg’s Festen were shot with digital cameras, but are not examples of cinematic animation. Images resulting from the digital capture of pro-filmic movements and changes are no more animated in virtue of this fact than are images resulting from analogue capture of the same pro-filmic events. It is only when moving images result from digital processing falling into Gaut’s second two categories that we may wish to speak of the images as animated. As noted in the survey of animation techniques in section “Kinds of Animation: A Taxonomy of Techniques”, this may be a matter of generating the appearance of movement and change by using a software tool to modify the result of digital or analogue capture of live-­ action events, or it may be the result of generating movements and changes by purely computational means through running various kinds of programmes that either do computationally what was traditionally done manually or generate image content ‘from whole cloth’. Just as recourse to digital technology in the making of a moving picture does not in itself result in an animated film, so such recourse in the making of still or moving images does not, pace Mitchell, necessarily involve a loss of indexicality. Gaut argues41 that many digitally produced images preserve indexicality, because, in the case of these images, digital technology is used only in the capture of pro-filmic events or in forms of ‘painting’ that reproduce in a digital key the kinds of ‘intentional rendering’ of a photographic trace familiar from the post-capture processing of analogue images. While, as Mitchell stresses, we may not be able to tell by simply looking at a digital image, still or moving, the extent of the digital processing involved in its genesis, indexicality depends not upon our possessing such knowledge but only upon the actual process productive of the image. But, even if we grant this, we might wonder whether Mitchell’s point can be made with respect to digital animation and, in fact, to animation more generally? Is it not distinctive of animated moving images that they lack indexicality and as a result cannot be sources of information about the pro-filmic world? And does not the receiver’s awareness of this fact call into question Sesonske’s claims about the subjective similarities between our experiences of live-action and of animated cinema? I shall conclude by making two points that support Sesonske’s claims, and in so doing, count against attempts to draw significant philosophical consequences from reflections on the presence or lack of indexicality in certain kinds of cinema. First, as we noted earlier, some animation processes, such as pixilation, preserve indexicality with respect to persons and objects, even if this is not the case for the events into which those persons and objects are represented as entering. Second, if it be claimed that most animation processes do not preserve this kind of indexicality, and at best involve an indexical relation between the film and the things entering into the animation  Gaut, A Philosophy of Cinematic Art, 48–9.




process—for example, drawings and cels in the case of ‘traditional’ analogue animation—we can reiterate what I take to be Sesonske’s main point against Cavell: in the case of live-action non-documentary cinema, many, if not all, of the pro-filmic objects or events ‘indicated’ by the film are not what the film is about: Truffaut’s film is about Jules and Jim, and neither Jules nor Jim exists independently of the constructed world of the film. They can no more stand in an indexical relation to the projected images than can Mickey Mouse.

Bibliography Arnheim, Rudolph. 1957. Film as Art. Berkeley: University of California Press. Baxandall, Michael. 1985. Patterns of Intention. New Haven: Yale University Press. Bazin, André. 1967. What Is Cinema? Vol 1. Trans. H. Gray. Berkeley: University of California Press. BBC. 2017. Documentary on the Making of Loving Vincent. news/av/41422698/loving-vincent-the-first-fully-painted-film. Accessed 28 Sep 2017. Bell, Clive. 1913. Art. London: Chatto and Windus. Bordwell, David, and Kristin Thompson. 1997. Film Art: An Introduction. 5th ed. London: McGraw Hill. Carroll, Noël. 1988. Philosophical Problems of Classical Film Theory. Princeton: Princeton University Press. ———. 1998. A Philosophy of Mass Art. Oxford: Clarendon Press. Cavell, Stanley. 1971. The World Viewed. Cambridge, MA: Harvard University Press. ———. 1979. The World Viewed: Enlarged Edition. Cambridge, MA: Harvard University Press. Currie, Gregory. 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, David. 2004. Art as Performance. Oxford: Blackwell. ———. 2009. On the Very Idea of ‘Outsider Art. British Journal of Aesthetics 49 (1): 25–41. Gaut, Berys. 1997. Film Authorship and Collaboration. In Film Theory and Philosophy, ed. Richard Allen and Murray Smith, 149–172. Oxford: Clarendon Press. ———. 2010. A Philosophy of Cinematic Art. Cambridge: Cambridge University Press. Kaufman, Charlie. 2015. Interview about Anomalisa Broadcast on December 22nd 2015 on the NPR Programme Fresh Air. frame-by-frame-filmmakers-make-the-mundane-miraculous-in-anomalisa. Accessed 30 Oct 2017. Lessing, Alfred. 1965. What Is Wrong with a Forgery? The Journal of Aesthetics and Art Criticism 23 (4): 461–471. Levinson. 1998. Evaluating Music. In Musical Worlds, ed. Philip Alperson, 93–107. College Park: Penn State Press. Linklater, Richard. 2001. Interview with Spence D. on A Waking Life. http://ca.ign. com/articles/2001/10/20/interview-with-richard-linklater. Accessed 3 Sep 2017. Livingston, Paisley. 2009. In The Routledge Companion to Philosophy and Film, ed. Carl Plantinga. London/New York: Routledge. Manovich, Lev. 1995. What Is Digital Cinema? Published Online in 1995. Accessed at on 1 Nov 2010. No pagination.



Maynard, Patrick. 1997. The Engine of Visualization: Thinking Through Photography. Ithaca: Cornell University Press. Mitchell, William J. 1992. The Reconfigured Eye: Visual Truth in the Post-Photographic Era. Cambridge, MA: MIT Press. Schmerheim, Philip. 2014. Scepticism. In The Routledge Encyclopedia of Film Theory, ed. Edward Branigan and Warren Buckland, 413–419. London/New York: Routledge. Scruton, Roger. 1983. Photography and Representation. In The Aesthetic Understanding: Essays in the Philosophy of Art and Culture. London/New York: Methuen. Sesonske, Alexander. 1974. Review of Stanley Cavell, The World Viewed. The Georgia Review 28 (4): 561–570. Shiner, Larry. 1994. ‘Primitive Fakes,’ ‘Tourist Art,’ and the Ideology of Authenticity. The Journal of Aesthetics and Art Criticism 52 (2): 225–234. Sparshott, Francis. 1985. Basic Film Aesthetics. In Film Theory and Criticism, ed. Gerald Mast and M. Cohen, 3rd ed. New York: Oxford University Press. Talbot, William Henry Fox. 1844. The Pencil of Nature. London: Longman, Brown, Green, and Longmans. Wells, Paul. 2002. Animation: Genre and Authorship. London: Wallflower Press. ———. 2006. Fundamentals of Animation. Lausanne: AVA. Wikipedia. 2017. Animation. Accessed 15 Aug 2017.


Sound in Film Paloma Atencia-Linares

Introduction When we think about films nowadays, what comes to mind are typically audio-­ visual works; works where both sounds and images are constitutive of the experience we have of films. Yet, for decades, sound was neglected from theoretical writings on film. Many texts and courses on the Philosophy of Film lack entries or sessions on sound and, with remarkable exceptions, for a long-time film theory was mainly focused on the image. This is arguably not very different among practitioners—as Randy Thom, director of sound design at Skywalker Sound, claims: “probably the biggest problem with sound in film is that writers and directors […] tend to think in visual terms”.1 Fortunately, things are changing fast and, at least on the theoretical side, discussion on sound in film has become much more prominent. Professionals and academics from a variety of disciplines such as Musicology, Cognitive Sciences, Cultural Studies and Film Theory are engaging with the multifarious aspects of film sound—a wide range of books, articles, doctoral thesis, journals, blogs and podcasts specialized in film sound have proliferated. Unfortunately, though, Philosophy—at least in the analytic tradition—has not followed this trend, and even when the Philosophy of Film has been a prolific field in the last twenty years, research on the philosophy of sound in film is, comparatively speaking, very much underdeveloped.

 Randy Thom, interview with Glenn Kiser and Michael Coleman, Dolby Institute Podcast Series, podcast audio, August 25, 2015. 1

P. Atencia-Linares (*) University of Mexico, Mexico City, Mexico © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




This entry is focused on the philosophy of film sound in the analytic tradition and its limitations; drawing from current literature in other fields, it aims to complement some philosophical discussions typically centered on the image with their aural counterparts. The first subsection provides an overview of early skeptical views on the contribution of sound to the artistic status of films; the second subsection critically questions the status of sound in contemporary philosophical conceptions of film and suggests a revision that takes seriously the idea of film as an audio-visual medium. The second section provides a proof of concept: it deals with a central topic in the philosophy of film—realism— showing how the discussion of sound may open new directions along three different dimensions: illusion, likeness and transparency.

The Place of Sound in the Concept(ion) of a Film Our contemporary idea of a prototypical film involves sounds as well as images. These elements are not only integrated and part and parcel of what constitutes a given token cinematic work, but they are also significant contributors to its artistic value. The films Whiplash (Chazelle 2014), Blue Velvet (Lynch 1986) or Apocalypse Now (Coppola 1979), for instance, would not be the cinematic works of art they are if they did not have sound. However, as prominent and artistically significant sound seems to be in our current idea of films, this has not been reflected in many theoretical conceptions of film which frequently do not take sound as a constitutive element of the category. In the first part of this section (“Sound and Film as Art: The Early Days”) we will see that some early theorists were reluctant to consider sound as an integral part of the medium for purported expressive and aesthetic reasons.2, I will present their arguments as three conceptual challenges, for which the most forceful response came from History: the development of sound films swept away these theorists’ claims. In the second part of this section (“The Conception of Film in Contemporary Philosophical Writings: The Absence of Sound”) we will see that, although contemporary philosophers do not deny the expressive power of sound in film, the conception of film that transpires from some of their writings is one where the auditory component is simply absent. The moving image is not only much more prominent than sound, but it often figures as the sole defining feature of what counts as a cinematic work. I will analyze contemporary philosophical conceptions of film and will claim that the emphasis on the visual in current definitions of film may have led philosophers to neglect sound in film from their research. I will question these views and sketch a proposal for an alternative methodology regarding the conception of film and the inclusion of sound.

2  Most of these critics of sound in film did not oppose to the use of music as an accompanying element of film. What they rejected was the incorporation of dialogue and, to some extent, naturalistic sound effects.



Sound and Film as Art: The Early Days The first two decades following the invention of cinema were a dialectical battleground: at stake was the recognition of film as a legitimate art form rather than a mere scientific curiosity or popular entertainment. Silent film had undergone remarkable progress toward this end, giving champions of film-as-an-art arguments to believe they were winning the battle. In this context, the incorporation of sound was considered a setback to this cause—especially in the form of dialogue, and to some extent also sound effects. There were at least three challenges that the then new technology faced in order to achieve the status of art. While all these affected the visual as well as the auditory elements of film, it was ultimately sound which was considered more problematic. Let us examine these three challenges separately.  he Challenge Against Mimetic Sound T According to a tradition of thought, something cannot qualify as art if it is a mere (mechanical) reproduction of reality. The status of photography as an art was questioned on the grounds that it was conceived of essentially as an automatic mechanism that recorded the appearance of reality. Since cinema was fundamentally photographic in nature, its status as an art faced the same challenge than photography. The addition of sound made things worse, for sound was just considered as a mere recording of voices and sounds of the world. As Sergei Eisenstein and others put it, “[sound] not only hinder[s] the development and perfection of the cinema as an art but also threaten[s] to destroy all its present formal achievements”.3 Eisenstein, as well as other filmmakers and theorists of the silent era, argued that far from being a mere reproduction of reality, film—and particularly its photographic substratum—transformed reality in a systematic way. In their view, sound, be it in the form of dialogue or realistic sound effects, did not achieve this transformation. Eisenstein appealed to the idea of neutralization with respect to the photographic image.4 When objects are photographed, he claimed, they are turned into neutral blocks of material devoid of their ‘real’ meaning; when reassembled in the cinematic montage, they acquire different significance. Cinema, through montage, does not mimic reality, it transforms it. Hugo Münsterberg held a similar idea: “the work of art shows us the things and events perfectly complete in themselves, freed from all connections which lead beyond their own limits, that is, in perfect isolation”5 and “the photoplay [by which he meant basically the silent film]

3  Sergei M.  Eisenstein, Vladimir I.  Pudovkin and Alexandrov, “A Statement” in Film Sound: Theory and Practice, ed. Elisabeth Weis and John Belton (New York: Columbia University Press, 1985), 83. 4  Sergei M.  Eisenstein, “Beyond the form”, in The Eisenstein Reader, ed. Taylor Richard (London: British Film Institute, 1999), 82–92. 5  Hugo Münsterberg, “The Photoplay: A psychological Study”, in Hugo Münsterberg on Film: The Photoplay and Other Writings, ed. Allan Langdale (New York: Routledge, 2002), 117.



tells us the human story by overcoming the forms of the outer world”.6 Now, (synchronized) sound, these theorists feared, threatens this process of isolation or neutralization. Dialogue and realistic synchronized sound effects, they thought, bring back the original meaning and context to the photographed objects. This, they surmised, makes films become more naturalistic or realistic, which, in turn, hinders their aesthetic or artistic character. If film was going to be conceived of as an art form, then it would be better off without sound.7  edium Specificity Challenge: The Case of Sound M Following the prescriptions of modernism, a medium justifies its becoming a proper art form if there is a specific characteristic, or a range of characteristics— expressive, representational or formal—intrinsic to that medium that distinguishes it from other media or art forms. Moreover, if artists want to excel in their practice of a given art form, the doctrine of medium specificity recommends that they should exploit the specific effects, features or possibilities that stand out in the medium.8 In the case of composite art forms that combine more than one medium of expression—such as opera or theater—there should be a predominant medium that represents the essence of such art form. In opera, for instance, the predominant medium would be dramatic music; in theater, it would be speech. Now, for critics of sound in film such as Rudolph Arnheim, the predominant medium of film is indisputably the moving image. It is this feature that distinguishes it from other art forms. Sound, in turn— especially in the form of dialogue—is not specific to film and thereby hinders its artistic vocation and represents a threat to the authenticity of the image. For Arnheim, “it was precisely the absence of speech that made the silent film develop a style of its own”, and “the predominance of the word would lead to the theatre”.9 Münsterberg held a similar idea. For him “a photoplay cannot gain but only loose if its visual purity is destroyed”.10 As critical as both Münsterberg and Arnheim were with the addition of dialogue, they were more permissive with the incorporation of music, since they saw in it some instrumental value. However, they emphasized that music as well as sound effects were not and should not be considered central or specific to film. As Münsterberg  Münsterberg, “The Photoplay: A psychological Study”, 129.  Early critics of sound in film generally thought that the addition of sound in the form of dialogue and sound effects was a temporary trend eventually doomed to fail. However, if sound was indeed there to stay, they believed, the only way in which it could make an aesthetic contribution would be if it was not fully synchronized with the image (See Pudovkin 1985; Eisenstein et al., 1985). 8  For a criticism of the medium specificity thesis, see (Carroll 2008, 35–52). For a qualified defense of it, see (Gaut 2010, 282–307). None of these mentions specifically the case of sound in film. 9  Rudolph Arnheim, “The New Laocoon”, in Film as Art (California: California University Press, 1957), 224. 10  Münsterberg, “The Photoplay: A psychological Study”, 145. 6 7



claimed, “we must […] disregard the accompanying music or the imitative noises […] they are accessory, while the primary power must lie in the content of the pictures themselves”.11 S ound and the Threat to Expression A third impediment that some early theorists believed sound imposed on film’s path to become an art form was a purported limitation on its expressive resources. Here again, the main criticism was directed at sound in the form of dialogue, and its main champion was, again, Rudolph Arnheim. For him, “not only does speech limit the motion picture to an art of dramatic portraiture, it also interferes with the expression of the image”.12 Silent film, he believed, had managed to perfect the art of expressing a dramatic situation by means of the posture and facial expression of actors and, in so doing, film had developed a style of its own. But the talkies came to replace “the visually fruitful image of man in action with the sterile one of the man who talks”.13 Moreover, one important achievement Arnheim ascribes to silent film in contrast to, say, opera or theater, is that since it is not constrained by dialogue, film does not emphasize the human figure but is more concerned with human beings set off against their natural setting. This, he claims, was “destroyed by the talking film: it endows the actor with speech, and since only he can have it, all other things are pushed into the background”.14 Siegfried Krakauer and Béla Balazs also shared, to a certain extent, the suspicion that the addition of sound might be a threat to film’s expressive powers.15 Balázs, as well as Krakauer, maintained that human beings have learned, by our use of language, to approach the world mainly by abstraction and conceptualization and have thereby lost connection with the expressivity of our bodies and gestures. One of silent cinema’s great achievements, they thought, is that it brought back our ability to read bodies and facial expressions. This, in turn, favors universal communication. Adding dialogue to films would be tantamount to going back to an impoverished language-centered communication.16 Unlike Arnheim, however, Balázs and Krakauer recognized that sound—even dialogue—when used wisely, could eventually contribute to the artistic value of a film. For this, sound has to be used cinematically, which for them meant mainly subordinated to the image. Film, they recommend, should de-­emphasize language in favor of the material qualities of sound and the irrational (non-­ literal or not clearly semantic); film should discover noise and the “acoustic environment. The voice of objects, the intimate language of nature” (my italics).17  Ibid., 83.  Arnheim, “The New Laocoon”, 228. 13  Ibid., 229. 14  Ibid., 227. 15  Béla Balázs, Erica Carter and Rodney Livingstone, Béla Balázs: early film theory: Visible man and The spirit of film (New York: Berghahn Books, 2010), 183. 16  Ibid., and Siegfried Krakauer, “Dialogue and Film,” in Film Sound: Theory and Practice, ed. Weis Elisabeth and John Belton (New York: Columbia University Press, 1985). 17  Balázs, Béla Balázs: early film theory: Visible man and The spirit of film, 185. 11 12



In sum, for many practitioners and theorists writing on the early decades of cinema, the moving image, rather than sound, was the predominant medium, the element that should be favored, the most expressive and the one which ultimately determines the conception of film as an art form. If sound, especially dialogue, could not be avoided, then it should be asynchronous, subordinated to the image and preferably non-literal or emphasizing the formal aspects of sound rather than its semantic content. History has swept away these three challenges. In hindsight, it seems clear that early critics’ fears and conception of film as an art were misguided. They confused a period-specific situation of cinema with the essence of film. Time and the developing and sophistication of the cinematic practices have shown that film is an art form not in spite of the inclusion of sound, but partly because of it. However, it is worth noticing that, even nowadays, many filmmakers who are mostly praised for their creative and artistic use of sound in film exploit it in a way that is partly reminiscent of the recommendation of early theorists. Critics praise filmmakers such as Andrei Tarkovski for emphasizing the materiality of sounds and noise—“the splashing of water, the crunching of glass underfoot, the haunting sirens of locomotives, the squeal of metal wheels”,18 Jean-Luc Godard’s use of asynchronous sound and sounds that mismatch with the image—such as sounds of animals or ambient sound replacing human voices, thereby also de-emphasizing the voice semantic content; David Lynch’s expressive use of noise and seemingly irrational voices, or Jacques Tati’s relegation of dialogue in favor of ambient sounds. The Conception of Film in Contemporary Philosophical Writings: The Absence of Sound In the past twenty years, the Philosophy of Film has become a prolific field of research. Unlike early theorists, contemporary philosophers writing on film are not preoccupied with proving that film, let alone non-silent film, can be art. This is a given. The discussion has focused on many other aspects of cinema. Remarkably, sound remains neglected in most contemporary philosophical texts on film, at least in the analytic tradition. One possible explanation for this is that, still today, philosophers conceive of cinema as an essentially visual medium. They might not deny that sound is an important element of film or doubt its contribution to the artistic value of cinema, but the way they define or think of the medium ultimately determines the priorities on their research agenda. What evidence do we have that philosophers have a visually centered conception of cinema? If so, why is this the case? Are they right? And if they are not, why does it matter? I will try to answer these questions in what follows. Defining concepts is a philosophical task par excellence—philosophers aim at clarifying the nature of a given object of study and the category of being under 18  Peter Green, Andrei Tarkovski. The Winding Quest (London: The Macmillan Press, 1993), 103.



which it falls under. The philosophy of film is no exception and, while not all philosophers devote much space to analyze in detail the nature of film, they do, in one way or another, lay bare the boundaries of what they take their research object to be. And what we find is that they systematically conceive of cinema as a fundamentally visual medium. Here are some representative cases taken from three influential monographs on the philosophy of film written in the last twenty years. From the very introduction of the book A Philosophy of Cinematic Art, Berys Gaut makes it clear what he takes his object of study to be: cinema is the medium of the moving image (…) ‘Movies’ and ‘motion pictures’ are terms that capture the phenomenon: we are discussing pictures that move (my italics).19

Similarly, when describing his conception of the essence of cinema, Gregory Currie claims the following: [T]he cinema, by which name I denote the medium to which particular movies belong, is a visual medium (…). Movies often have other properties, accessible through senses other than sight (…). Auditory properties are the most obvious and widely used (…) but all these, including sound, are incidental accretions so far as cinema itself is concerned, because there can be, and in fact there are, works in the medium which eschew all sensory engagement except the visual (my italics).20

A third case in point is the definition or conceptual analysis that Noël Carroll provides of the notion of film or moving image, as he prefers to call it: x is a moving image if and only if (1) x is a detached display or a series thereof; (2) x belongs to the class of things from which the promotion of the impression of movement is technically possible; (3) performance tokens of x are generated by templates that are themselves tokens; (4) performance tokens of x are not artworks in their own right; and (5) x is two-dimensional. Notice that each of these five conditions is alleged to be necessary and to be conjointly sufficient.21

Unlike Gaut and Currie, Carroll does not explicitly mention that moving images are essentially visual, but sound—or the possibility thereof—is certainly not considered a necessary or essential condition. Clearly, this does not mean that these theorists completely disregard sound in film, nor do they deny that the prototypical instance of film involves a sub-

 Berys Gaut, A Philosophy of Cinematic Art (Cambridge: Cambridge University Press, 2010), 1.  Gregory Currie, Image and Mind. Film Philosophy and Cognitive Science (Cambridge: Cambridge University Press, 1995), 3. 21  Noël Carroll, The Philosophy of Motion Pictures (Oxford: Blackwell Publishing, 2008), 72. 19




stantial use of sound.22 The kind of definitions that are at stake here are not of the prototypical film or moving image. The task of these type of analyses—especially those like Carroll’s—is, on the one hand, to provide necessary (and sometimes sufficient) conditions that, as a matter of logic, identify the elements without which a given object cannot qualify as moving image; on the other, to distinguish instances of works that fall under a certain category, for example, moving images, from others. In Carroll’s definition, for instance, being a detached display distinguishes things shown in films from things seen face to face. The possibility of movement distinguishes film from other detached displays such as paintings or photographs. The fact that films are performances generated by templates and the token generated are not works of art themselves differentiates them from instances of theater plays—where tokens are generated by interpretations and not templates and each performance of a play is indeed a potential work of art. And being two-dimensional tell films apart from, say, moving sculptures. The use of sound, even when pervasive, is not an element of differentiation between films and other art forms. Moreover—and more crucially for contemporary theorists—there are instances of film without sound, so the presence of sound cannot be an essential or necessary condition for a work to be a film—or so people like Currie claim. But is this really so? And, even if it is, how useful or recommendable are these definitions for understanding film as an object of study? Let us begin with the first question. The obvious way to support the claim that there are instances of films without sound is to point to the silent film era.23 Now, although traditional history of film has made a sharp distinction between silent and sound film, stressing the idea of there being a break between the two practices has been a blunt simplification. It is arguable that silent film was ever fully silent.24 Music and sometimes sound effects were a “crucial component not just of silent film exhibition but of the industry as a whole”.25 Before the 1920s and the era of synchronized sound, not only there was a robust practice of accompanying films with live music (which was frequently composed for the film),26 22  Carroll, Currie and Gaut do discuss certain issues related to sound in their works, but the main focus remains on the image. 23  For a parallel argument, see Altman’s “Historical Fallacy”. See Rick Altman, “Four and a Half Fallacies”, in Sound Theory, Sound Practice (New York: Routledge, 1992), 35–37. 24  For more insight on the issue of sound in silent film, see Rick Altman Silent Film Sound (New York: Columbia University Press, 2004), xi; Rick Altman, “The Evolution of Sound Technology”, in Film Sound: Theory and Practice, ed. Elisabeth Weis and John Belton (New York: Columbia University Press, 1985) and James Buhler and David Neumeyer, “Music and the Ontology of the Sound Film: The Classical Hollywood System”, in The Oxford Handbook of Film Music Studies, ed. David Neumeyer (Oxford: Oxford University Press, 2013). 25  Buhler and Neumeyer, “Music and the Ontology of the Sound Film: The Classical Hollywood System”19. 26  In this sense, it would count strictly as film music under definitions such as Jeff Smith’s or Noël Carroll and Margaret Moore’s. See Jeff Smith, “Music”, in The Routledge Companion to Philosophy and Film, ed. Paisley Livingston and Carl Plantinga (New York: Routledge, 2009), 84–195; Noël Carroll and Margaret Moore, “Music and Motion Pictures”, in The Routledge Companion to Philosophy and Music, ed. Theodor Gracyk and Andrew Kania (New York: Routledge, 2011), 456–467.



but there were various technologies that managed to combine moving images with (separately) recorded sound. The fact that the soundtrack was not attached to the same physical object as the image or that sound was an element that had to be either played separately from a recording or performed live, does not necessarily mean that it was not a constitutive element of films. Making this assumption would beg the question. It would exclude sound as a constitutive element of film presupposing an idea of film (a physical substrate or recording that combines sound and image) that in itself excludes an in principle legitimate way in which sound could be an intrinsic part of film.27 It could be argued that it is an exaggeration to claim that accompanying sound (in cases where, say, musical scores were improvised or not made for the film) was an intrinsic part of the cinematic works. Also, even if it were, conceptually, one can think of purely silent films, and that would be a sufficient counterexample for the claim that sound is a necessary element of film. This is a fair claim. However, two things are worth considering.28 Firstly, it is conceivable also to think about a film without images—or at least representational images; for one thing, as Altman reminds us “[d]uring many periods when cinema was heavily marked by its relation to the music industry, music accompanied by a blank screen has regularly been recognized as cinema: the long overtures to the early Vitaphone sound-ondisk features, the intro of a film’s theme song before the images or its continuation after the post-credits, and the use of a totally black screen in recent music videos”.29 For another, there are films, such as Derek Jarman’s Blue where the only thing we see is a static shot of a saturated blue color filling the screen—no representational image—and, in the background we hear sounds, voice-over and music. Disqualifying this work as a film would certainly bead hoc. Now, even when we can conceive of films without (representational) images, authors never consider the possibility of not counting images as a necessary or essential component of films. Yet, they take a different attitude with respect to sound. Secondly, as Carroll mentions, there are also films without movement. Jarman’s Blue is again an example, but there are others such as Chris Marker’s La Jetée or Michael Snow’s One Second in Montreal. This, however, does not mean that we should disqualify movement—or the possibility thereof—as an essential element of film. Why is this? Well, according to Carroll, this is because movement “is a permanent possibility in cinema; (…) static films belong to the class of things where the possibility of movement is always technically available in such a way that stasis is a stylistic variable in films in a way that it cannot be 27  Notice that this is, arguably, what Carroll does when he claims that conditions 3 and 4 of his definition are necessary for the definition of film: “(3) performance tokens of x are generated by templates that are themselves tokens; [and] (4) performance tokens of x are not artworks in their own right”. If sound of “silent” films was counted as a constitutive element of films, as I have argued, then there would be a part of certain films whose performance would not be generated by a template but by an interpretation and whose performance tokens could eventually be considered artworks in their own right. 28  For a similar argument, see Altman’s “Ontological Fallacy”. Altman, “Four and a Half Fallacies”, 37–39. 29  Ibid., 38.



with respect to still pictures”.30 Now, we could make a parallel argument in the case of sound. Synchronized sound recorded in the same physical substrate as the image only became a possibility in the 1920s, but live or recorded, but not synchronized, music and sound effects were a technical possibility from the beginning. Hence, if they were not used, it was perhaps out of a stylistic choice or general preference. It could be claimed that this statement is trivial, for accompanying music or sound conceived in such a way is a possibility for basically all art forms—a painting or a sculpture can always be accompanied by music—and this in no relevant sense means that we should therefore count sound as an essential or constitutive element of painting or sculpture. But this is not the case. The claim is not trivial because, unlike painting or sculpture, film had an aspiration to be an audio-visual media from the beginning. The first attempts to synchronize sound with image are as old as cinema itself31 and, even as early as the 1910s “Cameraphone, Chronophone, Cinephone and dozens of other competing systems were not only invented (…), they were installed in hundreds of theaters across Europe and from coast to coast in the United States”.32 In fact, one of the founding fathers of cinema, Thomas Alva Edison, declared as early as in 1894 the following: “the idea occurred to me that it was possible to devise an instrument which should do for the eye what the phonograph does for the ear, and that by a combination of the two all motion and sound could be recorded and reproduced simultaneously”.33 One could argue, then, that sound has always been, not only in one way or another a possibility for cinema, but also an element that was part of the idea of film since the beginning—the fact that the most appropriate, technically sophisticated and financially convenient technology only developed in the late 1920s is just a historical contingency. Now even if these arguments were not convincing, and theorists were right that, as a matter of logic, sound, unlike (the moving) image, is not a necessary element for some work to qualify as a film, one could ask the following question: to what extent does this conception of film help us to understand and appreciate actual films or current film practices? Certainly, none of the definitions provided are meant to help us appreciate films and, strictly speaking, they do not need to. Conceptual analyses and definitions can be pursued for a variety of purposes and one could argue that the only thing theorists are trying to do is delimit the boundaries of their object or study. But even if their definitions were right, they might have had pernicious consequences. Concepts are central for philosophers and, once the object of study—film, in this case—is defined fundamentally in visual terms, explicitly or implicitly, this sets the theoretical agenda and constrains it to the investigation of what is taken to be the  Noël Carroll, The Philosophy of Motion Pictures (Oxford: Blackwell Publishing, 2008), 60–62.  See, for instance, the early experiment by William Dickson to synchronize sound in a short film for Edison’s Kinetophone project. This experiment dates from 1894–1895. 32  Altman, “Four and a Half Fallacies”, 36. 33  Cited in Rick Altman, Silent Film Sound (New York: Columbia University Press, 2004), 78. 30 31



essential or necessary, which are issues related with the visual image. This, one could argue, has contributed to halt the progress of the research on sound in film within the domain of philosophy. Perhaps a better way to develop a philosophical conception of film could be to rely on the prototype theory—a theory of concepts whereby categorization is not about finding the essential elements or necessary and sufficient conditions for something to fall under a concept. Rather, it relies on a sort of comparison between the constituents that instances of that concept typically have in common. So, for example, a prototype theory of film would count sound as well as the moving image as constituents of the class of things called film because both these elements are typical or recurrent in the prototypical film as we know it, even when there might be films which lack sound or movement. In the spirit of a prototype theory is the notion of Categories of Art as developed by Kendall Walton.34 Very roughly, categories of art are ways in which we classify representational works in terms of their standard, variable and contrastandard properties—standard properties are those that are typically present in works that fall into that category; variable properties are those that may or may not be present, but they are not particularly characteristic of the category and thereby do not typically play a role in categorization. Finally, contrastandard properties are those that are rarely found in instances of that class and tend to disqualify works as part of such category. The idea is that this way of classifying works guide our appreciation, since categories stand as contrast classes that help us compare works in that category relative to their properties. This is not the place to develop an alternative theory or philosophical conception of film along this line, but if such a line would be pursued, the idea would be to count sound as a standard property of films, along with others such as the presence of moving images or a narrative, and so on, that are typically present in films. Following the theory then, finding an instance of a work without sound—being it performed live or recorded and synchronized—would be a rarity and would therefore stand out in terms of appreciation: appreciators would then need to ask themselves what explains the phenomenon, whether it is a historical or technological limitation imposed by the time when the film was produced, a stylistic choice or a genre convention. This is, of course, a very schematic suggestion of how alternative conceptions of film could be developed. But the main point is that, if philosophical conceptions of film take seriously that film is an audio-visual medium, and that far from being an incidental accretion, sound is a standard property of film as a class, then we might expect more philosophical discussions on film sound.

 Kendall L. Walton, “Categories of art”, in Philosophical Review 79, 3, (1970): 334–367.




Realism in Film Sound Let us now consider a topic in the philosophy of film where the discussion of sound, so far neglected, could play a significant role. The debate on realism in film has focused almost exclusively on the moving image—and specifically, on its photographic nature.35 Once again, the auditory dimension is, for the most part, ignored. This is remarkable because sound is frequently cited as an element that substantially contributes to generate the impression of reality that film has been said to confer. Moreover, sound in film raises similar, but, nevertheless, specific, concerns to those that have been noticed with respect to the (moving) image. There are many ways in which it is possible to talk about realism in film,36 but current discussions in analytic philosophy of film concentrate mainly on three specific approaches to the topic, and I will restrict my discussion to these37: (1) realism as the generation of illusions; (2) realism as perceptual likeness as opposed to conventionalism; and (3) realism as (photographic) transparency. All these three topics have been usually discussed with respect to the visual dimension of film, but there are relevant parallels in its auditory counterpart.38 Moreover, the case of sound can provide new insights into both current philosophical views and accounts coming from film theory and other disciplines. Film theorists and other scholars, for instance, frequently mention that sound generates various sorts of illusions and, unlike in many of the examples of the visual realm, where some philosophers have argued against there being any actual illusions, in the case of sound, it is indeed correct to talk about illusory experiences—or so I will suggest. The philosophical idea of perceptual realism, in turn, sheds light, among other things, onto the notion of fidelity—a notion that is frequently discussed in film theory with respect to sound in film—and provides an argument against views such as Chion’s who holds that fidelity is, to some extent, a matter of convention. Finally, although the notion of transparency has been developed mostly in relation to the photographic image, sound recordings can also be said to be transparent. Moreover, the case for transparency in the latter might be more appealing than in the former.

35  This does not mean that the issue of sound is completely absent—see, for example, Andrew Kania, “Realism”, in The Routledge Companion to Philosophy and Film, ed. Paisley Livingston and Carl Plantinga (New York: Routledge, 2009), 237–47. 36  Berys Gaut mentions up to seven distinct dimensions of realism. See Gaut, A Philosophy of Cinematic Art, 97. I restrict to these three because they are the most recurrent discussions. 37  I restrict my discussion to these three approaches not because others are less interesting or worth developing, but for the sake of focus. This chapter concentrates mainly on the analytic philosophy of film, and it aims to complement current visual-centric discussions with the aural counterparts; for this reason, I discuss realism in sound following the structure of current literature in the field. 38  The case of transparency is an exception. Although most of the discussion has focused on photographic images, the transparency of sound recordings has indeed been developed.



Cognitive Versus Perceptual Illusion A recurrent effect that has been attributed to the use of sound in film is the generation of a variety of illusions. David Bordwell, for instance, claims that “offscreen sound can create the illusion of a bigger space than we actually see”.39 With recent multichannel technologies, the audience can hear sounds coming from the left and right sides, behind and even above them.40 Maria Pramaggiore and Tom Wallis reiterate this claim when they affirm that “sound is a powerful tool for helping filmmakers create the illusion that the world of the story extends beyond the boundaries of the frame”.41 We can call this, the illusion of extended space. According to Chion, another effect of the audio-­visual illusion is that sound “raises the possibility of sleight-of- hand effects: sometimes it succeeds in making us see in the image [something] that isn’t even there”.42 Let us call this the illusion of addition. Other illusory experiences attributed to sound in film are the illusion of unity—that is, sounds and images come from different channels and location and yet are, nevertheless, perceived as coming from the same source. This, together with the power for engagement that film music, has also been recurrently claimed to promote, contributes to the audience’s sense of absorption in the world of the story which is frequently conceptualized in terms of suspension of disbelief. As Annabel Cohen puts it, “through music […] film spectators become part of the crowd watching the game, forgetting the theatre seat and the screen”.43 This is also an idea echoed by Claudia Gorbman who claims that “music removes barriers to belief; it bonds spectator to spectacle, it envelops spectator and spectacle in a harmonious space”.44 Now, contemporary philosophers of film have been very critical with the idea that the experience of film is in any sense illusionistic. Gregory Currie, for instance, has claimed that “film has considerable powers to engage and to persuade, but these powers are not accounted for in terms of illusion” (my italics).45 Currie distinguishes between two different kinds of illusion—cognitive and perceptual illusion—and denies that the experience of watching films engenders any of these. 39  David Bordwell and Kristin Thompson, “Fundamental Aesthetics in Sound in Film”, in Film Sound: Theory and Practice, ed., Elisabeth Weis and John Belton (New York: Columbia University Press, 1985), 193. 40  At the time this chapter is written, the most recent technology is Dolby® Pro Logic® IIz, which introduces two front height channels, so as, for example, re-create the experience of rain falling one’s roof. 41  Maria Pramaggiore and Tom Wallis, Film: A Critical Introduction (London: Laurence King Publishing, 2005), 209. 42  Michael Chion, Audio-Vision. Sound on Screen (New York: Columbia University Press, 1994), 5 and 12. 43  Annabel Cohen, “Film Music. Perspectives from Cognitive Psychology”, in Music and Cinema, ed. James Buhler, Caryl Flynn and David Neumeyer (New England: Wesleyan University Press, 2000), 361. 44  Claudia Gorbman, Unheard Melodies: Narrative Film Music (Bloomington: Indiana University Press, 1987), 55. 45  Currie, Image and Mind, 19.



Applied to the case of film, ascribing cognitive illusions to the audience would be tantamount to saying either that (1) the audience come to believe that the events shown in the film are present to them in the theater or that they are somehow present in these events while they happen on the set or (2) that when they see fictional things on the screen they believe they are really present to these things, or these things to them.46 The claims made by Cohen and Gorbman, mentioned above, that give voice to the idea of suspension of disbelief, could—if taken literally—be expressions of this type of illusion.47 But, strictly speaking, it does not seem to be the case that sound or music really makes us forget the theater seat and the screen or removes barriers to belief regarding the differentiation of the space occupied by the audience in the theater and the space of the story (be it fictional or non-fictional). As Currie mentions, this would be at odds with the behavior displayed by normal audience at the theater. If music really had the power to make us forget our current location and believe just for a moment that “we are part of the crowd watching the game”— as Cohen seems to suggest if we take her literally—one could expect, for instance, that when watching Rudo y Cursi (Cuarón, 2008), some member of the audience, believing that she is really in the presence of Gael García Bernal playing soccer, would feel compelled to approach him and ask him for an autograph. But this is certainly an unusual behavior that we rarely—if ever—see happening on a film theater. Moreover, as various philosophers have claimed, one just has to reflect on cases when we are watching a horror film in which someone is about to be killed by a machete-wielding mass murderer. If, by the effect of music or other formal filmic device, we really came to believe, even for a moment, that we are in the same space as the victim, we would either call the police, run or help the victim. But this is, again, something that hardly ever happens.48 The intuition behind the idea of suspension of disbelief and Cohen’s and Gorbman’s quotations are better understood by an appeal to the notion of transportation—49 a non-illusory experience of being carried away by a story due to induced empathy for the characters and the audience’s imaginings regarding the story.50 Music, as well as sound more generally, can certainly play  Kania, “Realism”, 238.  A charitable interpretation could be that Cohen and Gorbman’s phrases are merely figures of speech or exaggerations of the phenomenon, and one should take them with a grain of salt. However, it is, nevertheless, worth clarifying the phenomena. 48  Currie, Image and Mind, 24; Kendall Walton, Mimesis as Make Believe (Harvard University Press, 1990), ch.5; Noël Carroll, The Philosophy of Horror or Paradoxes of the Heart (New York: Routledge, 1990), ch.2. In addition to the argument from behavior, there are other arguments theorists put forward against the idea of suspension of disbelief or cognitive illusion induced by film. For the sake of economy, I restrict the counterexamples to the argument from behavior. 49  Moreover, at least in the case of Cohen, it is plausible to think this idea is closer to what she is proposing. 50  See Richard J. Gerrig, Experiencing narrative worlds: on the psychological activities of reading (New Haven: Yale University Press, 1993); Melanie Green and Timothy Brock, “In the mind’s eye: Transportation-imagery model of narrative persuasion”, in Narrative impact: Social and cognitive foundations, eds. Melanie Green, Jeffrey Strange and Timothy Brock (New York: Psychology 46 47



a role in promoting this experience of deep engagement in the narrative by affecting our emotional reactions.51 Consistent with this view put forward by cognitive psychologists, philosophers have also suggested that a better way to explain this phenomenon is by appealing to the fact that the audience imagine rather than believe the events presented in the films.52 Alternatively—or complementing the idea of transportation—the phenomenon of deep engagement with the events presented in the film can be explained in terms of allocation of attention.53 Attention allows to deal with the brain’s restricted capacities, for example, to process a multiplicity of stimuli. It is frequently mentioned in the literature that one of the powers and functions of sound in film is to attract and guide the attention of the audience to different elements of the film.54 So, it is plausible to claim that sound, together with other formal and narrative devices put at play in the production and projection of a film, capture most of our attentional resources and direct them toward processing the events presented in the story at the expense of the thoughts about the actual situation of the audience in the theater. This does not mean, however, that we forget that we are in the theater or that we come to believe that the events depicted in the film are part of our current reality. Selectivity of attention by no means involves or entails cognitive illusions or suspension of current beliefs. It just means that our attentional resources are focused on the story and stimuli coming from the film and thereby diverted from the current state of the audience.55 Now, few theorists endorse the idea of suspension of disbelief when taken literally as the generation of cognitive illusions. But a more plausible view, and one that is reasonable to ascribe to most of the authors quoted above, is that sound generates perceptual illusions. Following Currie, a perceptual illusion occurs when you cannot help but perceive certain phenomenon as being in a certain way even when you believe that is not the case and that your experience Press, 2013),315–341; Tom Van Laer, Ko De Ruyter, Luca M. Visconti and Martin Wetzels, “The Extended Transportation-Imagery Model: A Meta-Analysis of the Antecedents and Consequences of Consumers’ Narrative Transportation”, Journal of Consumer Research 40, no. 5, (2014): 797–817. 51  See Lars Kuchinke, Hermann Kappelhoff, and Stefan Koelsch, “Emotion and music in narrative films: A neuroscientific perspective”, in The Psychology of Music in Multimedia, ed. Siu-Lan Tan, Annabel J. Cohen, Scott D. Lipscomb, and Roger A. Kendall (Oxford: Oxford University Press, 2013), 118–138; Annabel J. Cohen, “Music as a Source of Emotion in Film”, in Handbook of Music and Emotion: Theory, Research, Applications, ed. Juslin, Patrik and John Sloboda (Oxford: Oxford University Press, 1993), 879–908. 52  See Walton, Mimesis as Make Believe and Currie, Image and Mind. 53  Atencia-Linares, Paloma and Miguel Ángel Sebastián, “Narrative Immersion as an Attentional Phenomenon” (manuscript); See also Liao, Shen-yi, “Immersion is attention” (manuscript). 54  See, for example, Münsterberg, The Photoplay, 79–88; David Bordwell and Kristin Thompson, Film Art. An Introduction (New York: McGraw Hill, 2013), 268–269. 55  This is consistent with an idea that Cohen herself suggests, namely that the increased activation of mental resources involved in processing the musical element of film “heightens our sense of diegetic film world”. In Cohen, “Film Music. Perspectives from Cognitive Psychology”, 366.



is misleading.56 The Müller-Lyre example is a case in point: you are presented with two lines of equal length, but you perceive one as being longer than the other due to the fact that the arrowheads at their endpoints point to different directions. You may know that the lines are indeed equal—so you do not hold a false belief—but your belief does not penetrate or change your perception. Currie denies that watching films involves any kind of perceptual illusion either.57 However, arguably sound in film provides many cases of straightforward perceptual illusions. Take, for instance, the following case mentioned by Michael Chion whereby sound affects the way we see things presented in film: We find an eloquent example in the work of sound designer Ben Burtt on the Star Wars saga. Burtt had devised, as a sound effect for an automatic door opening (think of the hexagonal or diamond-shaped automatic doors of sci-fi films), a dynamic and convincing pneumatic “shhh” sound. So convincing, in fact, that, in making The Empire Strikes Back, when director Irving Kershner needed a door-­ closing effect he sometimes simply took a static shot of the closed door and followed it with a shot of the door open. As a result of sound editing, with Ben Burtt’s “psssht,” spectators who have nothing before their eyes besides a straight cut nevertheless think they see the door slide open.58

This, it seems to me, is a clear case where, even if we are told that the door does not really move, unless we watch the film in slow motion—with the ensuing sound distortion—which is not the normal condition of film viewing, we would still have the perceptual impression that there is movement where there is none. This, in other words, is a perceptual illusion induced by sound effects. But this is not the only case where sound in film engenders perceptual illusions. Sound effects, in most cases, are manufactured by Foley artists or synthesized in post-production rather than recorded in the pro-filmic scene. Moreover, the soundtrack is not only produced separately from the image, but the reproduction of image and sound occurs through different channels. The sounds’ effects and voices we hear on film do not come from the objects and events we see on screen, but from various speakers located around the theater. And yet, given some conditions, for example, that the sounds are properly synchronized and that they are congruent with the image,59 we cannot help but hear the sounds as coming from, or being emitted by, the objects, people and scenes we

 Currie, Image and Mind, 28–9.  In particular, Currie discusses the case of the purported illusion of movement and claims that this is not an illusion—images, for him, really move. (Currie 1995, 34–47). However, he seems to reject perceptual illusions generated by film in general. See (Currie 1995, 28). 58  Chion, Audio-Vision, 12. 59  Annabel J. Cohen, Kelti MacMillan, and Robert Drew, “The role of music, sound effects and speech on absorption in a film: The congruence-associationist model of media cognition” Canadian Acoustics, 34, 3 (2006): 40–41. 56 57



see onscreen. This, we can call, the illusion of unity and is an instance of the very well-studied illusory phenomenon of the ventriloquist effect.60 These are two clear examples of perceptual illusions where sound affects what we see in the image and how we see it. Other cases mentioned by film theorists such as the purported illusion of extended space, where offscreen sound is perceived as an extension of the space of the story, might be more controversial, for it does not create the appearance that we perceive something that is not there or transforms our visual perception. One could claim that if we hear the sound of trees coming from the rear speakers, one is literally hearing this sound; we are not associating it to any object located in the image—so there is no illusion of unity and, one could argue, no illusion at all. However, one reason why this phenomenon can indeed be called illusory is because viewers do not hear offscreen sounds as external to the film—if we hear, say, a door knocking on the right-hand side of the screen, we typically do not think that there is someone knocking a door in the adjacent room. In a way, we cannot help but associate this sound to the space of the image, a space that we do not see but we, nevertheless, assume or imagine is there. In this sense, the offscreen extended space can also be understood as a perceptual illusion. These are only a few examples where sound may indeed show that, contrary to what philosophers like Currie have claimed, film does, in certain circumstances, generate perceptual illusions. Perceptual Realism Versus Convention Another dimension in which realism in film has been discussed is in terms of likeness. In a way, it is very intuitive to see why sound—in any of its forms: dialogue, music or sound effects—can contribute to convey a more life-like cinematic experience: the experience of the real world is one in which sound is omnipresent; things in the world like doors, explosions, falls or trains make noise. However, it also takes no more than a minute to realize that the idea of likeness is not entirely unproblematic: likeness, when applied to sound in film, cannot mean that sounds are ‘true to’ the world since, as we know, many sound effects are not recordings of the individuals we see on the screen but of entirely different things. For example, the flapping of a bird’s wings that we hear in some films is not typically produced by the actual bird flapping its wings, but by, say, some plastic gloves shaking.61 To make sense of the idea of likeness, given this phenomenon, film theorists appeal to the notion of fidelity:

60  Bjoern Bonath et al., “Neural Basis of the Ventriloquist Illusion”, Current Biology Volume 17, Issue 19, (2007), 1697–1703; Gregg. H. Recanzone, “Interactions of Auditory and Visual Stimuli in Space and Time” in Hearing Research, 258, 1–2, (2009): 89–99. 61  Frantzolas, Tasos “Everything you hear on film is a lie”, TED video, 16:33, filmed February 2016, posted October 2016, on_film_is_a_lie?language=en



From the filmmaker’s standpoint, fidelity has nothing to do with what originally made the sound in production […] We do not know what light sabers really sound like, but we accept the whang they make in Return of the Jedi as plausible. If the viewer takes the sound to be coming from its source in the diegetic world of the film, then it is faithful, regardless of its actual source in production. Fidelity is thus purely a matter of expectations. (My italics).62

Realism, understood as fidelity, is not a question of indexicality—whether the sound is really a recording of the actual source or not—but a matter of fitness with the image, that is, whether the sound fits with the expectations of the audience. Expectations, following Bordwell, are sometimes generated by the perceived likeness of the sound heard in the film with what the sound that the audience associate with a given source, given their experience—for instance, if the sound of say a dog depicted on screen resembles in some way the sound of dogs that the audience has experienced—or simply by how convincing the sounds become when paired with an image, regardless of the audience’s previous experience or lack thereof. Now, for theorists such as Chion, fidelity is a “tricky term”,63 and the perceived realism is, to a great extent, a matter of convention: In order to assess the truth of a sound, we refer much more to codes established by cinema itself, by television, and narrative-representational arts in general, than to our hypothetical lived experience […] The codes of theater, television, and cinema have created very strong conventions, determined by a concern for the rendering more than for literal truth. We are all thoroughly familiar with these conventions, and they easily override our own experience and substitute for it, becoming our reference for reality itself […] film as a recording art has developed specific codes of realism that are related to its own technical nature.64

Following this view, fidelity or perceived realism has little to do with likeness and much more to do with codes and conventions with which we have become familiar throughout our experience with film and representational arts more generally. If we perceive these conventions as realistic, it is due to our familiarity with them. Now, some philosophers have resisted the characterization of the idea of this dimension of realism in conventional terms. There might be sounds or images that we may have indeed learned to associate to certain ideas and, to that extent, they can be taken to be conventional or codified—for example, the sound of a bell tolling with the idea of death. However, this is not the predominant case. When speaking about realism as likeness with respect to the image, philosophers have claimed that pictures are representational systems different from, say, language, inasmuch as they depend on our natural perceptual capaci Bordwell and Thompson, Film Art. An Introduction, 283.  Chion, Audio-Vision, 98. 64  Ibid., 107–8. 62 63



ties, so that in order to understand them, we do not have to learn specific rules or codes, we just deploy our natural ability for perceptually recognizing objects.65 Applied to the case of sound in film, this notion can help to understand the notion of fidelity. The idea would be that sound in film is realistic, faithful or life-like not because it is “true to” its source, but because it triggers the same perceptual recognitional capacities that are triggered when we hear sounds in the real world.66 For example, when I hear the sound of a telephone ringing in real life, I use the same perceptual auditory capacities that I use to recognize the sound of a telephone in a film. These capacities allow certain plasticity or levels of tolerance. So, in the same way as we can recognize a woman in a portrait by Picasso, even when the shape does not entirely resemble a real woman, we can also recognize the sound of bones breaking when we hear the sound of a celery being cracked. This is not because we have learned to associate the sound of cracking a celery with the sound of breaking bones; this is not a convention; we need not have experienced this conjunction of sound and image before in any other film in order to experience the sound of the celery cracking as faithful or perceptually realistic. If we experience the sound as perceptually realistic, according to this view, it is because there are some acoustic features that the sound of bones breaking and the cracking celery share that trigger the same perceptual recognitional capacities. In this respect, arguably, our auditory system is even more flexible than our visual system for, as Chion points out, “the figurative value of a sound […] is usually quite nonspecific”,67 many sources produce sounds that, to the human average ear sound very much alike and “we rarely recognize a unique source exclusively on the basis of sound we hear out of context”.68 Also, and importantly, understanding fidelity in terms of perceptual capacities for sound recognition does not mean that we need to have prior auditory experience of the sounds that we hear in film. Just as I can recognize an echidna by seeing first a picture of an echidna,69 I can get an idea of how a shotgun sounds by listening to a given film—even when the sound we hear is not produced by an actual shotgun. Of course, sounds can be misleading and fail to match the real sound, like hearing synthesized “boings” when in a comic film, characters jump on a mattress. But the advocate of fidelity understood in terms of perceptual realism can claim that our auditory system functions in a way that, given certain conditions, sounds are perceived as fitting certain visual 65  See Noël Carroll, Theorizing the Moving Image (Cambridge: Cambridge University Press, 1996), 49–74; Currie, Image and Mind, 79–90; Dominic Lopes, Understanding Pictures (Oxford University Press, 1996). Currie does mention that this idea can be applied to the case of sound in film, but he does not develop it further. Currie, Image and Mind, 88). 66  For a more extensive application of this idea to the case of sound in film, see Birger Langkjær, “Making fictions sound real. On film sound, perceptual realism and genre”, MedieKultur 48, (2010): 5–17. 67  Chion, Audio-Vision, 23. 68  Ibid., 26. 69  For an argument in the case of images, see Currie, Image and Mind, 86–87.



stimuli—and this is the case in film as well as in real life. For instance, when information from different sensory modalities—for example, auditory and visual—is synchronized in time, it tends to be perceived as unified. Moreover, “binding these two stimuli together is done naturally and effortlessly”.70 Synchronicity may not always be enough for perceiving a sound as fitting a given source in the sense required by the notion of fidelity. However, it is also possible to appeal to the idea of congruency. When sounds are perceived as congruent in structure with a given visual stimuli—where congruency can be interpreted in a variety of dimensions such as, say, rhythm, tempo or intensity of movement—they are usually associated in meaningful ways. For example, a pattern of music—or sound effect—fluctuating at a given tempo and pitch and a ball bouncing at a given height and speed can be perceived as congruent in pattern and, in turn, be experienced as faithful to the visual stimuli.71 Perceived congruency between visual and auditory information is a natural phenomenon of our perceptual system, not an ability that we learn by repetition, and it may provide an explanation as for why certain sound effects such as the whangs of the light sabers are perceived as fitting, or faithful to, the image. It is certainly true that filmmakers have adopted certain techniques throughout film’s history that have become conventional or standard due to the effectiveness in being perceived as realistic by the audience. But it is plausible to claim, in line with the perceptual-recognitional model, that they are perceived as realistic not so much because they are conventions with which the audience has become familiar. Rather, they have become conventional or standardized in the film practice because they “exploit and accommodate the cognitive processes people use to perceive the physical world”.72 Transparency A third dimension along which realism in film has been discussed concerns the photographic nature that underlies most films (although clearly not all). Inasmuch as sound recordings are also mechanical registers and bear a similar realistic phenomenology, claims regarding the photographic realism can be applied—with interesting qualifications—to the case of sound in film.73  Recanzone, “Interactions of Auditory and Visual Stimuli in Space and Time”.  See Cohen, “Film Music. Perspectives from Cognitive Psychology”. The purpose of Cohen’s experiment with the bouncing ball is to show that sound—or music in her case—affect the (emotional) meaning or interpretation of images they accompany. However, the CongruenceAssociationist model she provides as an explanation for this phenomenon, can arguably also be applied to the case of fidelity. 72  Todd Berliner and Dale J.  Cohen, “The Illusion of Continuity: Active Perception and the Classical Editing System”, “Journal of Film and Video 63, no. 1 (2011): 44–63. Berliner and Cohen take this approach to explain classic editing devices for continuity in film. 73   These parallelisms—specifically regarding the claim of transparency—have indeed been explored to a certain extent in the literature. However, given that these discussions are scattered and not always in texts concerning film, it is worth compiling in this entry the arguments for transparency of sound recordings that can be applicable to film. 70 71



Photographic realism has sometimes been understood in terms of transparency. The claim that photographs are transparent amounts to saying that, when you see a photograph of your  grandmother, you literally—although indirectly—see her.74 This does not mean that photographs are transparent in the sense that we do not see the photographic surface at all, or in the sense that we are under the illusion of seeing the object face to face. The point is, rather, that seeing things in photographs belong to the same natural kind as seeing objects in mirrors, through telescopes, microscopes or other prosthetic devices. In this sense, one could argue, sound recordings are also transparent, and the most compelling claim for the case of film are, probably, sound recording of voices.75 The idea would then be that, when we hear, say, an actress’ voice in the cinema, we literally hear her voice. Now, the objections raised against photographic transparency may also apply to the case of sound recordings. Here are some of them.76 (A) Hearing recordings, one could claim, unlike hearing someone speaking directly, fail to covary with respect to changes and movements in the egocentric location of the source of the sound. Hence, hearing people’s voices in recordings cannot literally be a case of hearing,77 (B) vision and audition enable us to locate ourselves in space with respect to the things we see or hear, but this is not the case with photographs or sound recordings78 and (C) the fact that a reproduction or representation is mechanical and preserves similarity relations as photographs or sound recordings do not entail that in seeing or hearing the reproduction one is literally seeing or hearing the actual object—when one sees a perfect cast reproduction of a Trojan column done by mechanical means, for instance, one would never say that one is literally seeing the real column.79 It is as difficult to establish the extent to which some of these objections are decisive against transparency of sound recordings as they are against the transparency of photographs, for it is not entirely clear what the necessary conditions for our concept of perceiving—seeing or hearing—ultimately are, and the intuitions around the very notions of literal perception are vague. However, there 74  Kendall Walton, “Transparent pictures: On the nature of photographic realism”, in Noûs 18, 1 (1984):67–72. 75  Andrew Kania has developed this point with respect to musical recordings. See Andrew Kania, “Musical recordings” Philosophy Compass 4, 1 (2009b): 22–38. I base part of this discussion on his. 76  Theorists cited pose these objections for the case of photographs. I “translate” them for the case of sound recordings, but, strictly speaking, the authors are only committed to these objections for the case of vision. 77  Jonathan Cohen and Aaron Meskin, “On the epistemic value of photographs”, Journal of Aesthetics and Art Criticism, 62, 2 (2004):197–210 and Bence Nanay, “Transparency and sensorymotor contingencies”, Pacific Philosophical Quarterly 91, 4 (2010): 248–57. Nanay’s view is weaker than Cohen and Meskin but follow similar lines of argument. They pose the objection specifically to the case of images. 78  Carroll, Theorizing the Moving Image, 61–2; Currie, Image and Mind, 65–9. 79  Gaut, A Philosophy of Cinematic Art, 89.



are plausible responses one can raise to cases (A)–(C). Objection (A) would entail, for instance, that we do not literally hear the voice of a friend when speaking through Skype, since while I move and change my position in my room with my computer, I change my position with respect to the source of the voice—my friend—and yet the sound does not change.80 This seems very unintuitive. (B) in turn, as Andrew Kania has noted, would mean that “you do not literally hear a person whispering in an echo cathedral, unless you can point to where they are”, which again, is unintuitive. Moreover, “it is not clear that one ever locates oneself spatially with respect to an object solely on the basis of one’s (…) auditory experience”.81 Finally, while (C) may be appealing for the case of photographs, it is not so much for the case of sound recordings since it is not clear that the object of perception in the case of the sound recording— the actor’s voice—is different from the object of perception in the case of the sound heard when the actor is in front of you, as it is in the case of photographs. As Gaut himself acknowledges, the transparency of sound recordings might, in principle, be more appealing for the following reasons: (1) sound recordings, unlike photographs do not have surface, so it is more intuitive to characterize the experience as literally “hearing the source of the sound, rather than merely hearing a sound image”82; (2) whereas even with the best quality photograph, it is very easy to distinguish a photograph of Morgan Freeman from Morgan Freeman himself, it can be very difficult to distinguish, just by listening, a very good recording of Freeman’s voice from actually hearing Freeman speaking (Gaut 2010, 95).83 Gaut, however, resists to accept these differences between sounds and images as decisive to support neither transparency in sound recordings nor in photography. He appeals to the case of virtual reality, where there is no perceived surface, and differentiating the image and the scene is not possible. And yet, he claims, the objection raised in (C) still holds—seeing the virtual reality images or listening to the sound recording does not amount to seeing or hearing the object itself. Now, the substance behind the response to Gaut’s objection might be deeper than Gaut assumes: there might be differences between the metaphysics of sounds and images that may make the case for transparency of sound recordings, indeed, more feasible than photographic transparency. Michael G. F. Martin, for instance, suggests that In sound recording, one captures and reproduces the very sounds that were made on some particular occasion, and there is no image or other representational device involved in the recording and reproduction. In photography, one captures and re-presents the visual appearance of an object, but the individual photo-

 Cf. Cohen and Meskin, “On the epistemic value of photographs”, 89.  Kania, Musical Recordings, 10–11. 82  Gaut, A Philosophy of Cinematic Art, 94. 83  Ibid., 95. 80 81



graphed and the event captured are not, themselves reproduced or now available for sight.84

For Martin, the object of perception in the case of the sound recording of a voice is the sound of the voice itself which is an individual,85 and it is the same individual that you could have heard had you been present when it was recorded—of course, it would have been a different instance of hearing but the content would have been the same.86 Photographs, on the contrary, he claims, do not reproduce the individual captured (the depicted person or event)—this is a metaphysical impossibility of images, including virtual reality images or holograms—but only represent its appearance. If Martin is correct and there is a metaphysical distinction between the objects of perception in the case of sound recordings and photographs, the case of transparency may be, indeed, more plausible for the former rather than the latter.

Conclusion Although analytic philosophy of film has been a prolific field of research in the last thirty years, sound is still substantially neglected in this domain of philosophy. This chapter provided a critical analysis of this limitation and suggested some ways in which current philosophical discussions typically centered on the visual image could be complemented and enriched taking into account their aural counterparts. Firstly, the chapter traced the historical roots of this theoretical oblivion—the reluctance of early theorists to consider sound as an integral part of the medium because it threatened to disqualify film as a legitimate art form (section “Sound and Film as Art: The Early Days”). Secondly, it tried to show that, although modern analytical views on film do not share the same concerns as early theorists, the conception of film they provide is equally ­essentialist, visual-centric and characterizes film in terms of necessary and sufficient conditions that aim to distinguish the medium from other art forms. These conceptions of film, in turn—I conjectured—, explicitly or implicitly, set the theoretical agenda and favor an emphasis on visual aspects of film at the expense of aural ones (section “The Conception of Film in Contemporary Philosophical Writings: The Absence of Sound”). The final part of section “The Place of Sound in the Concept(ion) of a Film” also suggests that a more promising line of conceptualizing film—and of giving sound a more prominent role—is to focus on prototypical (or standard) features of film rather than in essential or distinctive ones.  Michael G. F. Martin, “Sounds and Images,” The British Journal of Aesthetics 52, 4 (2012), 332.  The relevant individual here is not the person who emitted the voice, but the sound of the voice itself. 86  An objection to this view cannot be that the content of my perception in the case of the sound recording cannot be the same as the voice perceived directly because the sound recording, could, for example, be noisier (as an old recording). I can see my friend Pau sitting on the chair in front of me with and without glasses. In both experiences I would be seeing Pau—the same individual— even when in one case I see him blurrier. 84




Subsequently, the section on “Realism in Film Sound” provides a proof of concept: there are various discussions on realism in film where sound could have been relevant  that have been neglected do to the overemphasis on the image. This section tried to show how the discussion of sound in film—partly incorporating ideas developed in other fields such as Film Theory or Cognitive Psychology—may shed light onto current discussion of central topics such as that of film realism. In particular, I argued, this can be done along three dimensions frequently discussed by analytic philosophers of film: (1) while philosophers have claimed that film does not generate illusory experiences, the case of sound may stand as an illustrating counterexample; (2) the notion of fidelity of sound as it has been conceived by some film theorists can be better understood along the lines of the notion of perceptual realism as it has been explained by philosophers; and (3) the controversial idea of transparency mostly applied to the photographic image may turn to be much more appealing to understand sound recordings.87

Bibliography Altman, Rick. 1985. The Evolution of Sound Technology. In Film Sound: Theory and Practice, ed. Elisabeth Weis and John Belton, 37–43. New  York: Columbia University Press. ———. 1992. Four and a Half Fallacies. In Sound Theory, Sound Practice, 35–45. New York: Routledge. ———. 2004. Silent Film Sound. New York: Columbia University Press. Arnheim, Rudolph. 1957. The New Laocoon. In Film as Art, 199–230. California: California University Press. Balázs, Béla, Erica Carter, and Rodney Livingstone. 2010. Béla Balázs: Early Film Theory: Visible Man and the Spirit of Film. New York: Berghahn Books. Berliner, Todd, and Dale J. Cohen. 2011. The Illusion of Continuity: Active Perception and the Classical Editing System. Journal of Film and Video 63 (1): 44–63. Bonath, et al. 2007. Neural Basis of the Ventriloquist Illusion. Current Biology 17 (19): 1697–1703. Bordwell, David, and Kristin Thompson. 1985. Fundamental Aesthetics in Sound in Film. In Film Sound: Theory and Practice, ed. Elisabeth Weis and John Belton. New York: Columbia University Press. ———. 2013. Film Art. An Introduction. New York: McGraw Hill. Buhler, James, and David Neumeyer. 2013. Music and the Ontology of the Sound Film: The Classical Hollywood System. In The Oxford Handbook of Film Music Studies, ed. David Neumeyer. Oxford: Oxford University Press. Carroll, Noël. 1990. The Philosophy of Horror or Paradoxes of the Heart. New  York: Routledge. ———. 1996. Defining the Moving Image. In Theorizing the Moving Image, 49–74. Cambridge: Cambridge University Press. ———. 2008. The Philosophy of Motion Pictures. Oxford: Blackwell Publishing. 87  I am very grateful to David Teira, Miguel Ángel Sebastián and Shawn Loht for providing useful comments to improve this chapter.



Carroll, Noël, and Margaret Moore. 2011. Music and Motion Pictures. In The Routledge Companion to Philosophy and Music, ed. Theodor Gracyk and Andrew Kania, 456–467. New York: Routledge. Chion, Michael. 1994. Audio-Vision. Sound on Screen. New  York: Columbia University Press. Cohen, Annabel J. 1993. Music as a Source of Emotion in Film. In Handbook of Music and Emotion: Theory, Research, Applications, ed. Patrik Juslin and John Sloboda, 879–908. Oxford: Oxford University Press. ———. 2000. Film Music. Perspectives from Cognitive Psychology. In Music and Cinema, ed. James Buhler, Caryl Flynn, and David Neumeyer, 360–378. New England: Wesleyan University Press. Cohen, Jonathan, and Aaron Meskin. 2004. On the Epistemic Value of Photographs. Journal of Aesthetics and Art Criticism 62 (2): 197–210. ———. 2008. Photographs as Evidence. In Photography and Philosophy: Essays on the Pencil of Nature, ed. Scott Walden, 70–90. West Sussex: Blackwell. Cohen, Annabel, Kelti MacMillan, and Robert Drew. 2006. The Role of Music, Sound Effects and Speech on Absorption in a Film: The Congruence-Associationist Model of Media Cognition. Canadian Acoustics 34 (3): 40–41. Currie, Gregory. 1995. Image and Mind. Film Philosophy and Cognitive Science. Cambridge: Cambridge University Press. Eisenstein, Sergei M. 1999. Beyond the Form. In The Eisenstein Reader, ed. Taylor Richard, 82–92. London: British Film Institute. Eisenstein, Sergei M., V.I.  Pudovkin, and Alexandrov. 1985. A Statement. In Film Sound: Theory and Practice, ed. Elisabeth Weis and John Belton, 83–85. New York: Columbia University Press. Gaut, Berys. 2004. Film. In The Oxford Handbook of Aesthetics, ed. Jerrold Levinson, 627–643. Oxford/New York: Oxford University Press. ———. 2010. A Philosophy of Cinematic Art. Cambridge: Cambridge University Press. Gerrig. 1993. Experiencing Narrative Worlds: On the Psychological Activities of Reading. New Haven: Yale University Press. Gorbman, Claudia. 1987. Unheard Melodies: Narrative Film Music. Bloomington: Indiana University Press. Green, Peter. 1993. Andrei Tarkovski. The Winding Quest. London: The Macmillan Press. Green, Melanie, and Timothy Brock. 2013. In the mind’s Eye: Transportation-Imagery Model of Narrative Persuasion. In Narrative Impact: Social and Cognitive Foundations, ed. Melanie Green, Jeffrey Strange, and Timothy Brock, 315–341. New York: Psychology Press. Kania, Andrew. 2009a. Realism. In The Routledge Companion to Philosophy and Film, ed. Paisley Livingston and Carl Plantinga, 237–247. New York: Routledge. ———. 2009b. Musical Recordings. Philosophy Compass 4 (1): 22–38. Krakauer, Siegfried. 1985. Dialogue and Film. In Film Sound: Theory and Practice, ed. Weis Elisabeth and John Belton, 126–142. New York: Columbia University Press. Kuchinke, Lars, Hermann Kappelhoff, and Stefan Koelsch. 2013. Emotion and Music in Narrative Films: A Neuroscientific Perspective. In The Psychology of Music in Multimedia, ed. Siu-Lan Tan, Annabel J.  Cohen, Scott D.  Lipscomb, and Roger A. Kendall, 118–138. Oxford: Oxford University Press. Laer, Van, Ko De Ruyter Tom, Luca M.  Visconti, and Martin Wetzels. 2014. The Extended Transportation-Imagery Model: A Meta-Analysis of the Antecedents and



Consequences of Consumers’ Narrative Transportation. Journal of Consumer Research 40 (5): 797–817. Langkjær, Birger. 2010. Making Fictions Sound Real. On Film Sound, Perceptual Realism and Genre. MedieKultur 48: 5–17. Lopes, Dominic. 1996. Understanding Pictures. Oxford: Oxford University Press. Martin, Michael G.F. 2012. Sounds and Images. The British Journal of Aesthetics 52 (4): 331–351. Münsterberg, Hugo. 2002. The Photoplay: A Psychological Study. In Hugo Münsterberg on Film: The Photoplay and Other Writings, ed. Allan Langdale. New York: Routledge. Nanay, Bence. 2010. Transparency and Sensorymotor Contingencies. Pacific Philosophical Quarterly 91 (4): 248–257. Pramaggiore, Maria, and Tom Wallis. 2005. Film: A Critical Introduction. London: Laurence King Publishing. Pudovkin, Vsevolod I. 1985. Asynchronism as a Principle of Sound Film. In Film Sound: Theory and Practice, ed. Elisabeth Weis and John Belton, 86–91. New York: Columbia University Press. Recanzone, Gregg H. 2009. Interactions of Auditory and Visual Stimuli in Space and Time. Hearing Research 258 (1–2): 89–99. Smith, Jeff. 2009. Music. In The Routledge Companion to Philosophy and Film, ed. Paisley Livingston and Carl Plantinga, 184–195. New York: Routledge. Walton, Kendall L. 1970. Categories of Art. Philosophical Review 79 (3): 334–367. ———. 1984. Transparent Pictures: On the Nature of Photographic Realism. Noûs 18 (1): 67–72. ———. 1990. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge: Harvard University Press.


What Is a Screenplay? Ted Nannicelli

Putting aside a number of complexities about atypical forms and functions, most screenplays are, roughly speaking, written plans for the creation of motion pictures. The voluminous “how-to” literature testifies to this characteristic function. The best-known screenwriting “guru,” Syd Field, begins his recent book, The Definitive Guide to Screenwriting, by canvassing a number of plausible analogies or metaphors: “What is a screenplay? A guide, or an outline for a movie? A blueprint, or a diagram? A series of images, scenes, and sequences that are strung together with dialogue and description?”1 Eventually, he settles for the following description: “A screenplay is a story told with pictures, in dialogue and description, and placed with the context of dramatic structure.”2 Other how-to manual writers tend to concur, although often emphasize the screenplay’s typically intermediate nature as a story to be used as a plan for the creation of a motion picture. Consider the definition Alex Epstein offers in Crafty Screenwriting: Writing Movies that Get Made: “A screenplay is writing intended to be turned into a film. It’s a hundred-odd pages held together by brass bands, in which you have written down whatever you want the audience to see and hear in your movie.”3 Still within the realm of “how-to” manuals, but more theoretically sophisticated is Lance Lee’s A Poetics for Screenwriters. According to Lee, “A screenplay is a filmed, enacted, immediate, sequential symbolic imitation of an action with a complete Beginning, Middle, and End  Syd Field, The Definitive Guide to Screenwriting (London: Ebury Press, 2003), 9.  Ibid., 10. 3  Alex Epstein, Crafty Screenwriting: Writing Movies that Get Made (New York: Henry Holt, 2002), 1. 1 2

T. Nannicelli (*) The University of Queensland, Brisbane, QLD, Australia © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




that falls within the larger fundamental story pattern, possessing an inherent significance, generated by the action of characters attempting to and ultimately resolving conflict-generating problems, whose outcome [i.e. a motion picture] embodies the vision of the screenwriter….”4 However, even the more theoretically informed, classroom-oriented “how­to” manuals also emphasize the intermediary nature of the screenplay—its status as a planning document. According to Craig Batty and Zara Waldeback, “A screenplay, whether for film or television, is a blueprint for the next stage of production; a document with words and actions ready to be realized.”5 Consider, too, Anthony Friedman’s description: “A screenplay or script is […] a visual blueprint for production, laying end to end the particular scenes employing the specific terminology of the medium to describe what is to be seen on the screen and heard on the sound track.” Clarifying Field’s description of a “story told with pictures,” with which we started, Friedman continues, “It is visual writing because it is the precursor to production in a visual medium.”6 For the purposes of teaching people what screenplays are and how to write them in the context of Hollywood motion picture production (and other national cinemas that share important features of Hollywood’s industrial organization), there is nothing wrong with the above definitions of the screenplay as a planning document whose story will be realized as a film. For one thing, such how­to manuals cater to people who want to be successful in the Hollywood system. For another, if manual readers do successfully sell a screenplay, they also need to be aware that, as they are often warned, once the screenplay is sold, their creative input is generally no longer welcome as the film moves into production. However, there are other traditions of filmmaking and other purposes for which people write and read screenplays—traditions and purposes that, understandably, are not represented in how-to manuals targeted at aspiring Hollywood movers and shakers. And although these alternative traditions and purposes of screenwriting are outside of the mainstream, they raise interesting philosophical questions about what an adequate definition of the screenplay would be. In previous publications, I have argued that a variety of screenplays that are written solely to be read as works of literature in their own right—rather than to serve as planning documents for motion pictures.7 For example, there is a genre of fan-fiction known as “script-fic,” in which fans of a particular story-­world use the screenplay as a form for sharing their stories with one another.8 Perhaps the most theoretically interesting sort of “script-fic” is that which continues the nar Lance Lee, A Poetics for Screenwriters (Austin, TX: University of Texas Press, 2001), 30.  Craig Batty and Zara Waldeback, Writing for the Screen: Creative and Critical Approaches (Basingstoke: Palgrave Macmillan, 2008), 2. 6  Anthony Friedman, Writing for Visual Media (Burlington, MA: Focal Press, 2014), 199. 7  Nannicelli, A Philosophy of the Screenplay, especially chapters 1, 2, and 9. 8  See Ted Nannicelli, “The Ontology and Literary Status of the Screenplay: The Case of ‘ScriptFic’,” in Journal of Literary Theory 7, no. 1–2 (2013): 135–153. 4 5



rative of a cancelled television show, such as Buffy the Vampire Slayer. Surely, the authors and readers of such screenplays are not under the illusion that their scripts will be “discovered” and used in production. Rather, their creations are intended to offer a reading experience that is worthwhile for its own sake. The use of the screenplay as a literary form is not unique to fan authors, however. For some well-known novelists, the use of the screenplay form appears to be a stylistic choice that is either particularly suited to the sort of story they want to tell or that functions to help them compose a distinct literary work— that is a novel. For example, in earlier research, I discuss the way in which Cormac McCarthy’s work on two screenplays in 1984 eventually led to the publication of his novels Cities of the Plain and No Country for Old Men.9 Here is one final example: I have on my bookshelf a handsome volume of screenplays, published in 2003, by E.L. Doctorow. Interestingly, for the purposes of the volume, Doctorow extensively revised the screenplay used to make the film Ragtime (1981)—evidently for the purpose of affording readers a worthwhile reading experience rather than for the planning of a remake of the film. Such examples constitute further objections to the intuition that screenplays have a kind of “proper function”—as certain other artifacts plausibly might.10 For the present purpose, however, let us focus on screenplays in their familiar, Hollywood-style industrial context, where they are safely characterized as having the primary or proper function of serving as plans for the creation of motion pictures.

Authorship One pre-theoretically plausible intuition that is sometimes voiced in popular critical discourse and that is worth exploring in a more sustained fashion is that screenwriters are, in at least some cases, the authors of screen works. According to a stronger and more tendentious version of this idea, a screenwriter is the sole author of a screen work. In practice, one rarely encounters this sort of claim, but it is advanced from time to time, often to counter a perceived overemphasis on the authority of the director and usually in reference to a putative screenwriting auteur.11 In such cases, the designation of authorship tends to have an honorific function: the screenwriter is dubbed the sole author to emphasize her or his artistic achievement. On this view, a screenwriter whose particular (critically lauded) signature style or sensibility permeates a screen work qualifies as an author. Now, the claim that a screenwriter may be the sole author of a screen work might immediately strike one as too strong: it is hard to think of possible examples even when considering plausible screenwriter  Nannicelli, A Philosophy of the Screenplay, 17.  See Lynne Rudder Baker, “The Ontology of Artifacts,” Philosophical Explorations 7, no. 2 (2004): 99–112. 11  For a recent example, see David Kipen, The Schreiber Theory: A Radical Rewrite of American Film History (New York: Melville House, 2006). 9




auteurs such as Ernest Lehman (whose scripts were directed by Robert Wise, Alexander Mackendrick, and Alfred Hitchcock) or Charlie Kaufman (whose scripts have been directed by Spike Jonze and Michel Gondry). However, the claim may gain some succor by considering screenwriter “showrunners” of serialized television programs. One interesting feature of contemporary serialized television is that the writing and directing of episodes tend to be rotated among a number of individuals. Partly for this reason, there has been a recent tendency to attribute authorship of such programs to the “showrunner”—the individual who is putatively responsible for the program’s overall style and storyline. Of especial importance for the present purpose is that most television showrunners are, first and foremost, writers. Consider, for example, Aaron Sorkin’s role on The West Wing and The Newsroom, Alan Ball’s role on Six Feet Under and True Blood, David Simon’s role on The Wire, Vince Gilligan’s role on Breaking Bad, and Nick Pizzolato’s role on True Detective. None of these showrunners directed more than a handful of episodes and, in some cases, did not direct any. There may be showrunner-directors as well: one thinks here of Lena Dunham’s Girls and Jill Soloway’s Transparent, for example. But for the moment, the point is simply that if there are any screenwriters who are the sole authors of screen works, they are likely to be television showrunners. Thus far, we have been operating on intuition. The above-mentioned showrunners might seem like plausible examples of screenwriters who solely author screen works based on what we know of their creative contributions and a tacit conception of authorship. We need to be careful here, though, because it is also in the industry’s interests to develop author-personas that can be used to brand and market products.12 That is to say, whether Aaron Sorkin really ought to be regarded as the author (or even an author) of The West Wing or The Newsroom demands that we appeal to a more nuanced conception of screen authorship and, possibly, adduce more specific information about his actual creative contributions. More broadly, in order to determine whether screenwriters are ever the authors of screen works, we need a clearer understanding of what screen authorship is. Although we have, thus far, considered screen authorship as pertaining solely to individuals, it is surely plausible that in a collaborative art form like motion pictures, authorship is often collective. Indeed, it is likely that a weaker claim about screenwriter authorship—namely that sometimes screenwriters are among a screen work’s co-authors—is the most defensible. In any case, in order to see whether we screenwriters are ever the sole authors or among the joint authors of a screen work, we need to first clarify our concept of authorship. 12  See, for example, Denise Mann, “It’s Not TV, It’s Brand Management TV: The Collective Author(s) of the Lost Franchise,” in Production Studies: Cultural Studies of Media Industries, ed. Vicki Mayer, Miranda J.  Banks, and John T.  Caldwell, 99–114 (New York: Routledge, 2009). Note that one can accept this moderate claim without a wholesale endorsement of some of the more extravagant assertions of critical theory pertaining to the author as merely a construct of (post)industrial capitalism.



In the philosophy of film, Paisley Livingston has provided the most compelling argument for the possibility of sole or single authorship of mainstream movies. Underlying the argument is Livingston’s subtle and cogent general account of authorship in the arts, according to which “the word ‘authorship’ is best used…to classify accomplishments that we evaluate as instances of expressive or artistic behaviour in various media, where authorship also involves exercising sufficient control over the making of the work as a whole.”13 This sort of “sufficient control” account of screen authorship, in particular, has a precedent in V.F.  Perkins’s classic Film as Film, but is given fuller elaboration in Livingston’s work. Further, on Livingston’s view, sufficient control is necessary but not sufficient for authorship to obtain. Also necessary, according to Livingston, is “the intentional realization of another type of goal in a work— namely, expression,” where “expression” is to be understood in the broad sense of “intentionally making an utterance or work that provides some indication that some psychological state or attitude, broadly defined, obtains in the author.”14 To make things a bit more concrete, a central question is whether a screenwriter ever successfully realizes her or his expressive or artistic aims while exercising sufficient control over the work as a whole. The recognition that many screenwriters meet the first condition seems to be what motivates familiar claims that screenwriters are among a screen work’s co-authors. This claim was perhaps first voiced in a sustained fashion by Richard Corliss, whose book, Talking Pictures, attempts to redress auteur theory’s oversight of screenwriters. Rather than arguing that screenwriters, rather than directors, are the sole authors of films, Corliss demonstrates that they are vital collaborators or (although he doesn’t use this term) among a film’s multiple authors.15 A more formal, sustained version of Corliss’s proposal might resemble something like Berys Gaut’s account of multiple authorship in the cinema. Gaut claims, “in light of the fact of art of artistic collaboration, we should admit that mainstream films have multiple authors.”16 More specifically, according to Gaut, “composers, set designers, directors of photography, and so on, can express their attitudes through their contributions.”17 Yet, as Livingston notes, it is implausible that any and every member of a production team who expresses her attitudes in the film is thereby one of its authors. The problem is not, necessarily, the practical one of an enormous proliferation of authors. Rather, it is that not all of these creative contributions are equal, so the people who make them do not deserve equal shares of praise or blame that would be 13  Paisley Livingston, Cinema, Philosophy, Bergman: On Film as Philosophy (Oxford: Oxford University Press, 2009), 71. 14  Livingston, Cinema, 70. 15  Richard Corliss, Talking Pictures: Screenwriters in the American Cinema (New York: Overlook, 1985). 16  Berys Gaut, A Philosophy of Cinematic Art (Cambridge: Cambridge University Press, 2009), 125. 17  Gaut, A Philosophy of Cinematic Art, 121.



implied by designating them all as authors. This is one reason why something like a sufficient control condition on authorship is also needed. The sufficient control condition advanced by Livingston should not be understood, as it sometimes has been, as motivated by a perceived need to designate a single person as the author of a screen work or to distinguish authored films from un-authored films.18 Livingston explicitly allows for the possibility of what he calls “joint authorship.”19 Whether an individual or multiple people are involved in the creation of a work, sufficient control over the work as a whole is necessary for authorship in part because it would seem that the successfully realization of an intention to express some attitude depends on it and in part because the attitudes actually expressed by the work are not simply the sum total of every collaborator’s contribution. Rather, the attitudes expressed by a work are a matter of how the different components of the work are structured—how they are organized in relationship to one another. And the ability to make decisions about those relationships—about the overall structure of the work—depends on having a certain amount of control over the work as a whole. Hence, the sufficient control condition. Whether a screen work is ever solely authored is, then, a question that needs to be investigated on a case-by-case basis. In principle, there is nothing to preclude a screenwriter from being the sole author of a screen work or one of the work’s joint authors. For example, Sarah Cardwell makes a plausible case that British television screenwriter Andrew Davies ought to be regarded as the sole author of many of the programs he scripted in virtue of collaborating with others in such a way that allowed him to retain control over the expression of his “ideas, intentions, and voice,” thus establishing “a discernible ‘authorial signature’ across his oeuvre.”20 Nevertheless, if we accept Livingston’s account of screen authorship, it seems that such cases are exceptional; more commonly, screenwriters often make essential contributions to screen works but are rarely the authors, individually or jointly, of those works. Why? Prima facie, most screenwriters appear to lack sufficient control over screen works as wholes. But this is another intuition that demands further analysis. In particular, its plausibility depends upon how we are to understand the idea of the work as a whole— that is, how we conceive of the ontology of screen works. In fact, Gaut claims that acknowledging the ontological differences between literature and film lends support to his multiple-authorship view. So, an excursion into the ontology of cinema can help us refine our thinking about screen authorship and the screenwriter’s relationship to it. It will also afford an opportunity to explore a question that is interesting in its own right: what is the ontological relationship between a screenplay and a screen work?

18  See C. Paul Sellors, “Collective Authorship in Film,” Journal of Aesthetics and Art Criticism 65, no. 3 (Summer 2007): 263–271. 19  Paisley Livingston, Cinema, Philosophy, Bergman, 72–76. 20  Sarah Cardwell, Andrew Davies (Manchester: Manchester University Press, 2005), 18.



Screenplays and the Ontology of Cinema One intuitively plausible conception of the screenplay’s relationship to the film is as analogous to a blueprint’s relationship to a finished building. Moreover, there is historical evidence that this was (and still is) how screenplays function in relationship to screen works at least in Hollywood and similar industrial contexts. For example, Janet Staiger, perhaps the first contemporary film historian to note this analogy,21 writes of the continuity scripts for Hollywood feature films of the 1930s and 1940s: “These scripts, of course, were related to the written form of stage plays. However, their relationship to the finished film was much different from that of the drama script to a theatrical performance. The continuity script was a precise blueprint of the film for all the workers.”22 In this section, I will argue that there is something fundamentally right about this understanding of the screenplay’s relationship to a screen work, I will defend it from recent criticism, and I will try to sharpen the claim in such a way that illuminates our prior question about authorship. Recent theorizing about the screenplay’s relationship to screen work has greeted the blueprint view with skepticism. For example, according to Steven Maras, Staiger’s discussion of the screenplay as a kind of blueprint constitutes a “constraining frame” that “inadvertently set[s] up normative conceptions of the roles and functions of the script” insofar as it naturalizes an understanding of screenwriting as sharply divided from production and, thus, a particular, industrial mode of film practice.23 For the present purpose, we can leave Maras’s ideological objections to this putative shortcoming of the blueprint metaphor aside and focus on the supposed inadequacy of it as an account of the ontology of film (although these are intertwined on his view). Maras claims that the blueprint conception of the screenplay (and the conception of the screenplay as a “sovereign” form more broadly) “can…be said to de-legitimate a broader theory of scripting that includes the script, mise-en-scene and other aspects of filmmaking.”24 More specifically, the sort of broader conception of “scripting” (or “screen writing”) Maras has in mind “can refer to writing not for the screen, but with or on the screen. It can refer to a kind of ‘filmic’ or ‘cinematic’ or audiovisual writing….”25 Somewhat ironically, this is a revisionist rather than descriptive account of screenwriting, which Maras at times seems to recognize: “[This approach] means rethinking the demarcation between creation and 21  To the best of my knowledge, the metaphor was first used in 1926 by Russian theorist and critic Viktor Shlovsky. See Denise J.  Youngblood, Soviet Cinema in the Silent Era 1918–1935 (Austin, TX: University of Texas Press, 1991), 68. For a skeptical discussion of the history of thinking of the screenplay as a blueprint, see Steven Price, A History of the Screenplay (Basingstoke: Palgrave Macmillan, 2013). 22  Janet Staiger, “Blueprints for Feature Films: Hollywood’s Continuity Scripts,” in The American Film Industry, ed. Tino Balio (Madison, WI: University of Wisconsin Press, 1985), 173. 23  Maras, 41–42. 24  Maras, 43. 25  Maras, 1–2.



interpretation, writing and performance, and the boundaries between what is a matter of style in cinema and what is a matter of screenwriting.”26 On Maras’s broad understanding of the concept, it seems that “screen writing” determines most or perhaps all of a screen work’s constitutive properties. That is, on Maras’s account, the properties that allow us to identify and individuate a particular screen work are established by “screen writing” in this broad sense. The distinctive color palette and muted tones of a film like A Most Violent Year (2014) are, on this account, the results of “screen writing” inasmuch as the concept includes “writing with light.” Likewise, the original dance routines of a movie such as La La Land (2016) are, on this view, the products of “screenwriting” insofar as the concept includes “writing with dance.” And, although Maras says nothing about music, and there is no English equivalent for “cinematography” or “choreography” in the context of music, perhaps the friend of the broad view of “scripting” would even want to claim that the score of a picture like The Hateful Eight (2015) is the upshot of “screen writing” to the extent that the concept encompasses “writing with sound.” Now, on the one hand, one could accept this broad view of “screen writing” but note that, nevertheless, it is “screen writers” in the broad, revisionary sense rather than “screenwriters” in the standard, descriptive sense who are responsible for many of the stylistic properties that are constitutive of screen works. Moreover, even the “screen writers” who are responsible for such properties (in standard terms, people like the cinematographer, the art director, the choreographer, the composer) seem unlikely to be the authors of a work if one accepts something like a sufficient control condition on authorship. On the other hand, there is good reason to reject the broad conception of “screen writing” on the grounds that it depends on an equivocation between two distinct senses of writing in play here—one literal and one metaphorical. The proposal seems to gain traction by leaning on the metaphorical sense of the term, but turns out to be uninformative because in this context, “screen writing” is just being used interchangeably with “filmmaking.” Moreover, the proposal muddies the conceptual waters by exploiting the ambiguity of our language rather than disambiguating the various concepts to which a single term may refer. However, accepting Maras’s broad conception of “screen writing” is not necessary to register a cogent point he makes—namely that screenwriting practices (in the literal sense) are heterogeneous and vary across filmmaking contexts. This is also one of Steven Price’s central objections to the blueprint metaphor: “[it] only makes sense in an industrial context, and cannot define all of the various kinds of text that circulate as ‘screenplays.’”27 When the blueprint metaphor is taken too literally—which, as Price’s research shows, it often is— this claim is true. Only in industrial contexts, such as that detailed by Staiger, does the screenplay function as a detailed plan for a film that “merely” needs to  Maras, 3.  Steven Price, The Screenplay: Authorship, Theory and Criticism (Basingstoke: Palgrave, 2010), 45.

26 27



be executed by craftspeople. Even then, Price observes, screenplays tend not to specify precise technical detail in the way that blueprints do. Moreover, Price is correct to say that this sort of conception of the screenplay—as a mechanistic, ultimately disposable set of instructions—has blinkered us to the aesthetic qualities of screenplays as literary objects.28 Nevertheless, there is a crucial similarity between screenplays and blueprints, or architectural plans more broadly, at which Price hints but does not pursue. The analogy is most accurate in conceptualizing the ontological relationship between instructions and end product. Despite not probing it further, the point seems to have occurred to Price, who, at the end of his discussion of the blueprint metaphor, cites film theorist Peter Wollen contrasting the relationship between screenplay and film with that between musical score and performance. Price rightly senses that there is an important disanalogy here—an intuition I have tried to sharpen elsewhere with arguments outlined below.29 This is not to say that grouping together screenplays, architectural plans, and scores is in itself a conceptual misstep. On the contrary, there is at least one very good reason to consider these things as of a kind: they are all instructions for the creation (or, in metaphysically more neutral terms, the generation) of distinct artworks. They are, in Stephen Davies’s terms, “notations.”30 But although Davies remains agnostic about whether they are the same sort of notations, there are good reasons to make some further distinctions here. Davies convincingly argues that musical scores are “work-specifying” or “work-­ prescriptive” notations: “[A] score is a musical notation the main purpose of which is to serve as a work prescription. It records a set of instructions, addressed to performers, the faithful execution of which generates an instance of the piece it specifies.”31 Without becoming sidetracked by complex debates about the ontology of music, we can say that a plausible corollary of this view is the idea that musical works are abstract entities that are created in virtue of the composition of a score but which require a performance for their instantiation. Importantly, for our purposes, the work’s constitutive properties are established by the score, lending credence to the idea of the composer as the author of the work (if not performances of it). So far, so good. A trickier matter, as Davies notes, is whether architectural plans are work-specifying in this way. It seems implausible to regard architecture as a performing art, but, Davies, claims, “if we allow that the architectural work is created when the plan is done, whether or not the building is built, just as we allow for finished but unperformed musical works and plays, the plan

 Ibid., 44–46.  Ted Nannicelli, “Instructions and Artworks: Musical Scores, Theatrical Scripts, Architectural Plans, and Screenplays,” British Journal of Aesthetics 51, no. 4 (October 2011): 399–414. 30  Stephen Davies, “Notations,” in A Companion to Aesthetics 2nd ed., edited by Stephen Davies, Kathleen Marie Higgins, and Robert Hopkins (Wiley-Blackwell, 2009), 441–443. 31  Stephen Davies, Musical Works and Performances: A Philosophical Exploration (Oxford: Oxford University Press, 2001), 100. 28 29



might best be regarded as work-specifying.”32 Again, there are complex ontological questions that need to be skirted here—in particular, whether architecture is a single-instance or multiple-instance art. Yet a brief examination of our appreciative practices suggests that architectural works (at least those in the Western tradition) are physical objects which are at least partly identified and individuated by the materials with which they are created and their spatiotemporal locations.33 Admittedly, it seems clear that architectural works can survive various sorts of restoration as the (numerically) same works. For example, the Guggenheim Museum designed by Frank Lloyd Wright has been subject to several such restorations, and today, we regard the museum as the numerically same, continuous entity as that which was originally erected. But if the Guggenheim were levelled, it seems clear that we would regard the work as destroyed—as lost to us—whether or not Wright’s plans were extant. For following those plans to “rebuild” the Guggenheim would not, it seems, “re-­ instance” the same work, but rather create a distinct (albeit closely related) work. The upshot of the thought experiment is this: if works of architecture are at least partly identified and individuated by their physical qualities and spatiotemporal locations, then architectural plans are not work-specifying in the same way musical scores are. In contrast to musical works, which are purely abstract objects created in virtue of the completion of a score, architectural works are not created by the completion of architectural plans because they have physical qualities among their constitutive properties. If they did not, we would not value them as we do; we would not spend so much time and effort preserving and restoring old buildings if it were possible to simply generate equivalent “instances” of them. So, the architect’s role is not strictly equivalent to that of the composer. If an architect is to be regarded as “the author” of a building, her control must extend over the construction of an actual edifice; otherwise, she is the author of a work only insofar as her plans comprise paintings, drawings, sculptures, and so forth.34 In contrast, the composer is the author of a work simply in virtue of completing her notations. Because musical works are abstract entities, they are complete once the composer prescribes the conditions under which they can be instanced. Thus, in the context of music, the composition of a notation affords the composer control over the constitutive properties of the overall work in a way that the composition of a notation in architecture does not. In the former case, notations are work-specifying; in the latter, they are not. Thus, in the former case, one who composes a notation, thereby authors a distinct work; in the latter case, one who composes a notation is only the author of that notation.

 Davies, “Notations,” 441–442.  The following argument, advanced in more detail in Nannicelli, “Instructions and Artworks,” is indebted to Davies’s line of reasoning in “Is Architecture an Art?” in Philosophical Perspectives on Art (Oxford: Oxford University Press, 2007), 129–145. 34  On this point especially, I indebted Davies’s arguments in “Is Architecture an Art?” 32 33



This line of reasoning clearly bears upon the crucial question for our purposes here: Are screenplays work-specifying notations akin to musical scores or are nonwork-specifying notations like architectural plans? Elsewhere I have argued that screenplays are not work-specifying because no cinematic work is created in virtue of the creation of a screenplay.35 The faithful execution of the instructions in a musical score will necessarily instantiate a single, self-same musical work in performance. Cinema, however, is not a performing art, and the faithful execution of the instructions in a screenplay will result in multiple, distinct cinematic works rather than multiple instances of a single cinematic work. This is so in part because cinematic works have, among their constitutive properties, features that are only established in the production and post-­ production processes. Qualities such as color palette, performance, camera movement, editing, and post-production special effects are all, plausibly, among a cinematic work’s identifying and individuating properties. Using the screenplay Carl Mayer wrote for F.W. Murnau’s Sunrise: A Song of Two Humans to make a contemporary film in color and with synchronous sound would result in a new cinematic work rather than a new instance of Murnau’s masterpiece. Indeed, if all extant copies of Murnau’s film were destroyed, we would regard the film as lost to us; we could not generate a new instance of it by following the instructions in Mayer’s screenplay.36 This is the important sense in which screenplays are, indeed, like blueprints or architectural plans. Because screenwriters like Mayer typically have little or no input over many of the constitutive properties of the cinematic works made from their screenplays, let alone the organization of those features in the shaping of the work as a whole, it is implausible that they are correctly regarded among the authors of those cinematic works. Just as an architect would, on the current account of authorship, need to exercise sufficient control over the erection of the building as a whole to count as its author, a screenwriter would need to exercise sufficient control over the creation of a whole screen work in order to be regarded as its author. Because screen works are partly identified and individuated by features such as performance, camera work, editing, mise-en-scene, and so forth, over which screenwriters typically do not have control, screenwriters are rarely, if ever, among the authors of screen works. Nevertheless, screenwriters are authors in a more limited sense: they are the authors of their screenplays. But although screenplays are admittedly among the most central constitutive properties of screen works, the authorship of—no matter how original, distinctive, detailed, or whatever—does not entail the authorship of the distinct, cinematic work made from it.  Nannicelli, “Instructions and Artworks,” and Nannicelli, A Philosophy of the Screenplay.  The argument in this paragraph should be understood as contingent upon our actual cinematic practices being as they are rather than involving a modal claim. That is, we can imagine our creative practices being slightly different such that cinema was a performing art, and the faithful execution of the instructions in screenplays resulted in multiple instances of the same cinematic work. In such a context, screenplays would correctly be regarded as work-specifying notations. See Nannicelli, “Instructions and Artworks.” 35 36



Screenplays and the Artistic and Ethical Evaluation of Cinema The claim that screenwriters are typically not, if ever, the authors (or among the multiple authors) of screen works might strike some readers as counterintuitive. For example, one might insist that, despite the above arguments, it still seems implausible to deny a screenwriter authorship in certain cases where her or his “authorial signature” clearly overwhelms or, at least, matches the creative input by, say, the director. I suspect that something like this thought motivates Berys Gaut’s account of multiple authorship, according to which we ought to regard as authors all members of the production team who express their attitudes through their contributions. Although Gaut focuses on acting, I think certain screenwriting cases might constitute better support for his view. In this section, I will sketch a few examples that an advocate of the multiple-­authorship view could enlist to advance a counterargument to the account outlined thus far. I will try to show, however, that accurate, fine-grained artistic, and ethical evaluations of cinematic works necessarily depend upon a more restricted conception of authorship along the lines of what Livingston proposes—one that enlists sufficient control of the work as a whole as a necessary feature of authorship and, thus, entails that most screenwriters are not the authors of cinematic works. Consider, as a potential counterexample to the view I have advanced thus far, Glengarry Glen Ross (1992). The film’s most obvious association is with David Mamet, who wrote the screenplay, based on his own 1984 Pulitzer Prize–winning stage play. Moreover, at the time the film was made, Mamet was already an accomplished film director as well as a screenwriter, having written and directed the critically acclaimed House of Games (1987) and Homicide (1991) as well as the 1991 book, On Directing Film. Yet the director of the film version of Glengarry Glen Ross is James Foley, whose career is characterized by workmanlike efforts. Suppose, for the sake of argument, that Foley had complete creative control over filming and post-production—that Mamet’s input did not extend beyond the writing of the screenplay. Do we still want to say that Mamet is not among the film’s authors? Perhaps in this sort of case, we may be tempted to think that if the sufficient control condition denies Mamet authorship over Glengarry Glen Ross, so much the worse for the sufficient control condition. Now, the intuition underlying this sort of objection would seem to be this: because Glengarry Glen Ross is a significant artistic achievement, because Mamet has a track record of achievement in writing and directing for the screen (not to mention writing for the stage), and because Foley is an improbable source of the film’s achievements, it seems as if Mamet ought to be regarded as at least one of the film’s authors. In other words, the motivation underlying the intuition that Mamet is the film’s author is that he deserves credit for the film’s artistic success. In response, we may note, first of all, that identifying individuals as the work’s authors is not necessary to properly credit them for their



contributions to the work’s artistic merits. As a matter of fact, key contributors such as writers, actors, cinematographers, and editors are often recognized for their achievements by industry and guild awards, as well as, to a lesser extent, popular criticism. But in any case, it is simply a conceptual error to assume that denying such contributors the status of authors is tantamount to ignoring their artistic achievements. Secondly, although the multiple-authorship view Gaut advances is, in part, motivated by the desire to democratically spread the credit for artistic success, the sword cuts the other way as well: authors are responsible and sometimes blameworthy for artistic shortcomings or failures. Consider, for example, what is plausibly the most significant artistic demerit of Glengarry Glen Ross: the rapid cutting from one medium close-up to another during dialogue scenes. The editing is clearly intended to drive the scenes’ dramatic momentum and complement the clipped, punchy dialogue being delivered by the characters. However, the writing and the acting can both stand on their own. The cutting proves to be gratuitous and distracting, drawing our attention away from the dialogue and performances it is intended to complement. Now, for the sake of argument, accept the plausible assumption that Mamet had no say about the film’s editing. The problem with the multiple-authorship view is that it, nevertheless, burdens Mamet with the responsibility for an artistic demerit over which he had no control. Indeed, the gratuitous editing actually undermines the contribution Mamet did make—the writing—and detracts from its power. It makes little sense to saddle Mamet with responsibility for this artistic shortcoming. Yet this is just what the multiple-authorship view does because it attributes authorship, in a broad and coarse-grained fashion, to all of the members of a production collective of a work regardless of the nature of their contributions. Another sort of example to consider here involves bad movies that are made from good screenplays. Industry lore has it that one cannot make a good film from a bad screenplay, but one can make a bad film from a good screenplay. Such cases, like the case of Glengarry Glen Ross, again illuminate the pitfalls of attributing authorship to screenwriters with the hope of securing credit for their artistic achievements. For example, although Eternal Sunshine of the Spotless Mind is an excellent film directed by Michel Gondry from a Charlie Kaufman screenplay, Gondry’s first production of a Kaufman screenplay, Human Nature, was a mixed success. Although this is speculative, one gets the sense in the performances and overall tone of the film, that Gondry simply failed to successfully realize the balance between irony and whimsy that Kaufman’s scripts characteristically strike. Indeed, it is plausible that what ­artistic merits the film has are due to Kaufman’s script. This ostensible fact, along with Kaufman’s distinctive writing style, might seem to warrant the ascription of authorship of Human Nature to Kaufman as well as to Gondry (assuming, just for the sake of argument, that the alternative is that Gondry is the sole author of the film).



However, identifying Kaufman as one of Human Nature’s authors would unfairly burden him with a number of the film’s artistic flaws over which he had no control whatsoever. As indicated above, one of the film’s chief weaknesses is plausibly the way inconsistencies in tone undermine its more thoughtful (if ironic) handling of serious questions about human nature and human society. This problem is most evident when the film slips into gross-out comedy mode—for example, the scene in which Puff visits a strip club and one of the dancers uses her legs to pull his head into her crotch. At such moments, the film sacrifices sly and intelligent commentary on the pretentions of societal mores for cheap humor that distracts our attention from its substantive themes. Importantly, the strip club scene is much subtler in Kaufman’s screenplay: it simply describes Puff sitting at the club in a drunken stupor. So, none of the artistic flaws that attend this scene in the film are in fact Kaufman’s responsibility. Yet ascribing authorship to Kaufman would place the responsibility and blame for just these sorts of artistic flaws on his shoulders. True, if we regarded Kaufman and Gondry as co-authors, Gondry would share in the responsibility for such artistic flaws. But why should Kaufman bear any responsibility at all when he evidently had no control over the shaping of the parts of the film in which the flaws in here? The multiple-authorship view’s inability to parcel out credit and blame in sufficiently precise ways becomes even more unsettling when a screen work suffers not only from artistic flaws but from moral flaws. For when screen works are morally blameworthy, the consequences faced by those held responsible can be much more serious. Although there is no space to properly review the literature here, there are plausibly a number of ways in which artworks, including screen works, may be subject to ethical criticism. Most notably, ethical criticism of screen works is warranted in cases where those works endorse or seek to elicit from the audience’s morally blameworthy attitudes.37 Now, certainly, in many cases of this nature, screenwriters, as much as—if not more—than any other members of the production team are responsible for such attitudes. Consider the typically sexist dialogue that is part and parcel of many classical Hollywood films. Many otherwise great films of this era suffer from moral blemishes insofar as they seek to elicit the audience’s sympathies for blatantly misogynistic characters or comic amusement at misogynistic actions or lines. Think, for example, the scene in The Public Enemy in which Tom Doyle tells Kitty, “You’re a swell dish. I think I’m going to go for you.” Or, later, just before he shoves a grapefruit in her face: “I didn’t ask you for any lip. I asked if you had a drink.” Indeed, men telling women to shut up and provide them with alcohol is a motif in classical Hollywood. In a similar, unfortunate scene 37  For good discussions of this sort of claim and the interaction between art and ethics more broadly, see Noël Carroll, “Art and Ethical Criticism: An Overview of Recent Directions of Research,” in Art in Three Dimensions (Oxford: Oxford University Press, 2010), 235–271; and Berys Gaut, “Art and Ethics” in The Routledge Companion to Aesthetics, 3rd ed., ed. Berys Gaut and Dominic McIver Lopes (New York: Routledge, 2013), 394–403.



from an otherwise excellent film made about 20 years later, The Asphalt Jungle, Dix Handley asks Doll, “Why don’t you quit cryin’ and get me some bourbon?” For our purposes, the point here is that to the extent such scenes suffer from moral blemishes, it is plausible that the screenwriters are to blame because they are responsible for the dialogue. So, too, are any members of the production collective who had the power to change or remove the dialogue. However, it is not necessary for us to deem these screenwriters the authors of the films in order to criticize for the moral blemishes for which they are responsible. Furthermore, there are good reasons to refrain from identifying those screenwriters as authors. For there are other ways in which screen works may be morally flawed that are beyond the control of screenwriters and, thus, not plausibly their responsibility. Typically, once shooting begins, screenwriters have very little control over anything—if they are on set at all. The most visible work that happens on set is the collaboration between director and actors— oftentimes an intense, intimate, interpersonal activity. As such, while film sets are the sites of cooperative work and collaborative achievement, they are also sometimes the backdrop for disagreements, fights, slights, and ethical failings. Consider this example from the news cycle a few years ago: As The Hollywood Reporter recently noted, “One of the most notorious scenes in cinema history, the Marlon Brando and Maria Schneider butter rape scene in Bernardo Bertolucci’s Last Tango in Paris, is making headlines once again, 44 years after the film’s debut. In a recently resurfaced video interview from 2013, Bertolucci confirms that Schneider, who died in 2011, did not know the details of the rape scene ahead of time, and that the graphic nature of the scene was improvised on set.”38 Now, Bertolucci and Brando did not actually rape Schneider, but it is plausible that they deceived, humiliated, and violated her. If we accept the account of events that both Schneider and Bertolucci offer, then what we see when we watch the film is a fictional rape, but a real violation of another sort and the real humiliation of Schneider. For this reason, it is plausible that the Last Tango in Paris itself is morally blemished and, perhaps, artistically flawed—that is, the film is morally and artistically flawed in virtue of what was done to Schneider during its creation.39 Now, given the plausible idea that authorship essentially involves control over the work as a whole and, as such, responsibility for the work as a whole, we 38  Ariston Anderson, “Hollywood Reacts With Outrage over ‘Last Tango In Paris’ Director’s Resurfaced Rape Scene Confession,” The Hollywood Reporter (December 3, 2016), available at (accessed December 8, 2016). The video is available here: 39  For a supporting argument in defense of this claim, see Ted Nannicelli, “Moderate Comic Immoralism and the Genetic Approach to the Ethical Criticism of Art,” Journal of Aesthetics and Art Criticism 72, no. 2 (Spring 2014): 169–179. I take it that in the Last Tango in Paris case, the genetic approach to ethical criticism ought to be especially plausible, given that the moral flaw inheres in the work’s manifest properties—that is, the actual violation and humiliation of Schneider is visible on the screen.



ought to say that the author(s) of Last Tango in Paris (whoever they are) bear responsibility for the moral and artistic flaws that result from what Schneider experienced in the process of creating the film. (This is not to say the film’s authors are the only ones who bear responsibility; Brando clearly bears some responsibility but is not plausibly an author of the film.) Yet, surely, the responsibility for these ostensible moral and artistic failings belongs in no way to the film’s co-screenwriters, who may have scripted a fictional rape scene but had nothing to do with the real violation and humiliation that occurred during the filming of the scene. So, in this case, and others like it, attributing authorship to a film’s screenwriters threatens to unfairly burden them with responsibility for moral and artistic flaws over which they had no control. Cases like these constitute one sort of important reason that we ought not regard the co-­ screenwriters as among the film’s authors.

Conclusion Taken together, the above analyses of screen authorship generally, the ontology of cinema, and the artistic and ethical evaluation of cinematic works support the intuitions with which this chapter began: screenwriters are important creative contributors to screen works, but they are rarely the authors of those works. The reason, in summary, is that screenwriters rarely meet the plausible necessary condition of having control of and responsibility for the screen work as a whole. As we saw, this is, in part, because screen works have many of their constitutive properties as a result of creative activity in which the screenwriter has no involvement whatsoever, let alone control over. Moreover, we saw that in some such cases, it is right and proper that screenwriters are not regarded as authors because if they were regarded as authors, they would be unfairly burdened with responsibility and blame for artistic and moral flaws over which they had no control. Nevertheless, this conclusion may leave us wanting a bit more of a positive account of screenwriters’ roles in the creation of screen works. There are at least two points we can add. First, as hinted at the start of the chapter, the foregoing account allows for the conceptual possibility that a screenwriter may be the author (or among the authors) of a screen work. On the account of authorship endorsed here, authorship of a given work is determined on a case-­ by-­case basis, and there is nothing, in principle, to preclude screenwriters from authoring screen works if they meet the sufficient control condition and the expression condition. So, it remains possible and, perhaps, plausible, that, say, head writers of long-form television shows such as Louis C.K., Vince Gilligan, and David Milch are the authors of those shows, given the amount of creative control they reputedly have. However, we must also note that, in such cases, those individuals are not the authors of the television programs merely in virtue of being head writers. And the scripts for those television programs still leave many of the programs’ constitutive features unspecified and undetermined. That is, these programs, like all screen works, have constitutive features that are



established after the writing of the screenplay and that are, therefore, beyond the control of the screenwriter qua screenwriter. If individuals like CK, Gilligan, and Milch are the authors of television programs, it is because they are not only head writers but also executive producers who have control over the shaping of the programs as wholes—and responsibility for them. This still seems to leave the screenwriter and the screenplay with less value and significance than what we often intuit they actually have. The second and final point may go some way to ameliorating this result. It is possible that the conclusion at which we have arrived doesn’t feel entirely satisfactory because, in contrast to the descriptive analysis of authorship accepted here, our appreciative practices often operate with a tacit, honorific conception of authorship. So, when we deny that a screenwriter like Mamet is the author of the film Glengarry Glen Ross, it may feel as if we are in some sense denying Mamet’s artistic achievement. Conceptually speaking, however, this need not be the case at all. We simply need to be clearer about what we mean when we use the term “authorship” and develop a finer-grained account of the agency behind the achievement of screen works. True, identifying an individual as the author of a screen work makes her a candidate for praise for the work’s artistic achievements. But it does not follow that the achievements are hers alone. For another possibility is that significant amounts of credit are due to creative contributors who are not among the work’s authors in the descriptive sense above, yet whose agency contributed to the artistic achievement of the work. Thinking in parallel terms about literature, it is plausibly the case that, say, Raymond Carver is the sole author of his short stories and poems, although his editor, Gordon Lish, undoubtedly deserves some of the credit for the artistic achievements realized by Carver’s works. We don’t need to make the overly strong claim that Lish is a co-author of those works in order to properly credit him for his contributions, though. Although only Carver had control over the works as wholes (and, as such, is the sole author of them), Lish deserves and can be allocated credit for his creative contributions by recognizing (1) that authors are not the only people responsible for the constitutive features, including artistically meritorious features of their works and (2) non-author creative contributors are candidates for praise and credit for such features insofar as they exercised agency over them. Elsewhere, I have advanced this sort of argument regarding the appreciation of television, and we can also apply it, mutatis mutandis, to the appreciation of the creative contributions of screenwriters.40 In appreciating the artistic achievement of any given feature of a work, we can ask how the feature happened to get there—how the work happened to be shaped in just that way. Authors deserve some credit for such achievements because they ultimately have control over the work as a whole. Yet in many cases, the artistic achievement that inheres in a particular feature of a screen work is a result of the creative contri40  The following discussion is drawn from Ted Nannicelli, Appreciating the Art of Television: A Philosophical Perspective (New York: Routledge, 2017), 42–45.



bution of some other agent who is not the author—say, an actor, the cinematographer, or the screenwriter. These contributions and the individuals who make them deserve credit, of course, and we do not lack a means to parcel it out if they are not among the works’ authors. We can acknowledge their creative contributions by noting that their actions—their agency—are the sources of the artistically meritorious features. Unlike authorship, which is an all or nothing game since someone either has control over the whole or does not, agency is a scalar concept, admitting of fine-grained degrees of distinction. Another analogy I have used elsewhere is with moral agency in the context of ethical criticism. We think of people like Tony Soprano, who has control over the planning and execution of a crime, as the “authors” of those crimes, and, as such, as bearing the most responsibility for them and deserving the most blame. Yet cases like this often involve other moral agents—accomplices—who have some lesser degree of responsibility and deserve a lesser degree of punishment. In such cases, we are, of course, interested in the “authors” of the crimes, but not exclusively so. We need the concept of moral agency to make the appropriate gradations of moral responsibility and parcel out the appropriate amount of blame and punishment to the accomplices of the crime. One virtue of focusing on agency as well as authorship in the appreciation of screen works (and other art contexts) is the analogous way in which it finely attributes credit and responsibility for a work’s artistic merits and flaws, respectively, among the relevant contributors, including its author(s). For example, attending to agency as well as authorship could potentially help adjudicate and resolve certain disputes such as that pertaining to the relative creative contributions Orson Welles and Herman Mankiewicz made to the screenplay for Citizen Kane. Plausibly, one of the sticking points in this case has been the way in which the dispute has assumed that only one of the men can be the screenplay’s author and, thus, deserving of all of the credit for its artistic achievement. An agential approach would recognize that, authorship aside, both men made important creative contributions and deserve credit.41 In short, a focus on agency might achieve what the model of multiple authorship unsuccessfully attempts: that is, it might allow us to more precisely assign credit for a work’s artistic merits. An emphasis on agency provides a cogent way of affirming the value of the contributions of individuals like screenwriters and assigning appropriate credit for them without simultaneously burdening those individuals with responsibility for features of the work over which they have no control. A focus on agency gives us the tools to acknowledge witty dialogue, intricate plotting, complex character development, and restrained subtext as artistic merits and contributions to a screen work’s overall 41  In this example, another option would be to recognize this as a case of joint authorship along the lines of what Livingston proposes. I am admittedly describing this particular case rather simplistically in the interest of space. For a sustained, detailed discussion, see Robert L. Carringer’s The Making of Citizen Kane, Revised Edition (Berkeley: University of California Press, 1996).



artistic value without becoming ensnared in debates about authorship. It allows us to more clearly and precisely appreciate and credit the artistic contributions that screenwriters make to screen works.

Bibliography Anderson, Ariston. 2016. Hollywood Reacts with Outrage Over ‘Last Tango in Paris’ Director’s Resurfaced Rape Scene Confession. The Hollywood Reporter, December 3. Available at Accessed 8 Dec 2016. Baker, Lynne Rudder. 2004. The Ontology of Artifacts. Philosophical Explorations 7 (2): 99–112. Batty, Craig, and Zara Waldeback. 2008. Writing for the Screen: Creative and Critical Approaches. Basingstoke: Palgrave Macmillan. Cardwell, Sarah. 2005. Andrew Davies. Manchester: Manchester University Press. Carringer, Robert L. 1996. The Making of Citizen Kane, Revised Edition. Berkeley: University of California Press. Carroll, Noël. 2010. Art and Ethical Criticism: An Overview of Recent Directions of Research. In Art in Three Dimensions, 235–271. Oxford: Oxford University Press. Corliss, Richard. 1985. Talking Pictures: Screenwriters in the American Cinema. New York: Overlook Press. Davies, Stephen. 2001. Musical Works and Performances: A Philosophical Exploration. Oxford: Oxford University Press. ———. 2007. Is Architecture an Art? In Philosophical Perspectives on Art, 129–145. Oxford: Oxford University Press. ———. 2009. Notations. In A Companion to Aesthetics, ed. Stephen Davies, Kathleen Marie Higgins, and Robert Hopkins, 2nd ed., 441–443. Malden: Wiley- Blackwell. Epstein, Alex. 2002. Crafty Screenwriting: Writing Movies that Get Made. New York: Henry Holt. Field, Syd. 2003. The Definitive Guide to Screenwriting. London: Ebury Press. Friedman, Anthony. 2014. Writing for Visual Media. Burlington: Focal Press. Gaut, Berys. 2013. Art and Ethics. In The Routledge Companion to Aesthetics, ed. Berys Gaut and Dominic McIver Lopes, 3rd ed., 394–403. New York: Routledge. ———. 2009. A Philosophy of Cinematic Art. Cambridge: Cambridge University Press. Kipen, David. 2006. The Schreiber Theory: A Radical Rewrite of American Film History. New York: Melville House. Lee, Lance. 2001. A Poetics for Screenwriters. Austin: University of Texas Press. Livingston, Paisley. 2009. Cinema, Philosophy, Bergman: On Film as Philosophy. Oxford: Oxford University Press. Mann, Denise. 2009. It’s Not TV, It’s Brand Management TV: The Collective Author(s) of the Lost Franchise. In Production Studies: Cultural Studies of Media Industries, ed. Vicki Mayer, Miranda J. Banks, and John T. Caldwell, 99–114. New York: Routledge. Maras, Steven. 2009. Screenwriting: History, Theory and Practice. London: Wallflower. Nannicelli, Ted. 2011. Instructions and Artworks: Musical Scores, Theatrical Scripts, Architectural Plans, and Screenplays. British Journal of Aesthetics 51 (4): 399–414. ———. 2013a. A Philosophy of the Screenplay. New York: Routledge.



———. 2013b. The Ontology and Literary Status of the Screenplay: The Case of ‘Script-Fic. Journal of Literary Theory 7 (1–2): 135–153. ———. 2014. Moderate Comic Immoralism and the Genetic Approach to the Ethical Criticism of Art. Journal of Aesthetics and Art Criticism 72 (2): 169–179. ———. 2017. Appreciating the Art of Television: A Philosophical Perspective. New York: Routledge. Perkins, V.F. 1972. Film as Film. Baltimore: Penguin. Price, Steven. 2010. The Screenplay: Authorship, Theory and Criticism. Basingstoke: Palgrave. ———. 2013. A History of the Screenplay. Basingstoke: Palgrave Macmillan. Sellors, C.  Paul. 2007. Collective Authorship in Film. Journal of Aesthetics and Art Criticism 65 (3): 263–271. Staiger, Janet. 1985. Blueprints for Feature Films: Hollywood’s Continuity Scripts. In The American Film Industry, Revised Edition, ed. Tino Balio, 173–192. Madison: University of Wisconsin Press. Youngblood, Denise J. 1991. Soviet Cinema in the Silent Era 1918–1935. Austin: University of Texas Press.


Approaches and Schools


Analytic Philosophy of Film: (Contrasted with Continental Film Theory) Richard Eldridge

The terms “analytic” and “Continental” are evidently enough mismatched: “analytic” more or less names a method, while “Continental” names a place. Not all practitioners of an analytic method work outside Europe, and not all of those whose work is more socially hermeneutic work inside Europe. A number of theorists both analyze the mechanisms of film construction and consider the social significance of film interpretively. Arguably, the most interesting and important theorists do both. Yet these terms have, nonetheless, passed into common-enough usage as markers of two distinct, broad overall styles of thinking about film. “Continental” generally indicates an interest in conceptions of an emergent, socially and historically formed human subject understood according to some combination of terms derived from Hegel, Freud, Saussure, and structuralism, in the regressive social functions of typical narrative films as such, and in avant-garde works taken as cutting against the grains of both Hollywood movies and ordinary social life. In contrast, analytic philosophers of film tend to be case-by-case pluralists. They are more likely to focus closely on things that happen in individual films and to pay attention to contingent decisions about shots, sequencing, lighting, dialogue, and so on, that are made in courses of production, embodied in particular cases, and aimed at producing specific effects. Their arguments are more immediately empirical, though their empirical descriptions often have theoretical presuppositions, implicit or explicit. Continental film scholars are more likely to call what they do to film theory, while analytic scholars are more likely to describe themselves as doing philosophy of film. This terminological arrangement mirrors similar distinctions R. Eldridge (*) Swarthmore College, Swarthmore, PA, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




between literary theory and philosophy of literature and between art theory and philosophy of art. The two styles of work on film are represented in two excellent but quite distinct summary anthologies: Critical Visions in Film Theory1 (2010) for Continental film theory and Philosophy of Film and Motion Pictures (2006)2 for analytic philosophy of film. Neither volume includes any essay or excerpt that is included in the other. Exceptions and qualifications to this distinction between Continental film theory and analytic philosophy of film are legion, but as these anthologies illustrate, there are these two broad, overarching, distinct stylistic tendencies in the academic study of film. The major early academic film theorists Hugo Munsterberg, Rudolf Arnheim, and Erwin Panofsky were German emigrés who argued in favor of receiving film as a form of art. They challenged dismissals of films as nothing but cheap entertainments, and they were concerned to establish that film has distinct and striking powers of expression and artistic presentation. Hence, they attended simultaneously to properties of films as a medium of photographically produced moving images and to mechanisms of expression and of the presentation of meaning. As the reception of film as a medium of art became well established, however, partly through their arguments and, more significantly, through the achievements of major directors such as Orson Welles, John Ford, Jean Renoir, and Sergei Eisenstein, their arguments came to be taken for granted, and academic film studies began to split along stylistic lines that distinguished Continental from analytic approaches in studies of art and in the humanities generally. In part, this developing split responded to two more or less distinct traditions of directorial style and its theory. More avant-garde filmmakers were largely centered in Europe and were more influenced by montage editing and by silent film. Their work was more likely to call attention explicitly to the filmed image as artful construction and more likely to offer intellectual provocation than comfortable enjoyment. This directorial tradition includes Sergei Eisenstein, Fritz Lang, and Luis Bunuel, among others.3 Major American  Eds. Timothy Corrigan and Patricia White with Meta Mazaj (Boston: Bedford/St. Martin’s).  Eds. Noël Carroll and Jinhee Choi (Malden, MA: Wiley-Blackwell). 3  In his landmark essay, “The Evolution of the Language of Cinema,” André Bazin contrasts (a) the relatively more montage-oriented style that derives from silent film, that “evoked what the director wanted to say,” and that “insidiously substituted mental and abstract time [in place of] … real time” with (b) the more explicitly realist style, developed through Orson Welles’ use of the deep-focus shot that built on “the realism of sound” (introduced in 1927) to develop “a reborn realism” (Bazin, “The Evolution of the Language of Cinema,” in Bazin, What is Cinema?, Volume 1, trans. Hugh Gray [Berkeley: University of California Press, 1967, pp. 23–40, at pp. 39, 38). Even here, however, enormous qualifications are necessary, many of which Bazin supplies. FrancoSpanish Surrealism (Bunuel, Dali), German Expressionism (Lang, Murnau), and Soviet Montage (Eisenstein, Vertov) are distinct avant-garde traditions. Realism in sound films was established in the 1930s well before Welles by Ford, Hawks, and Capra, among others working in Hollywood and by Jean Renoir and others working in France, all drawing on an earlier realist narrative style in Griffith, Flaherty, Chaplin, and others. There are strong narrative realist elements (along with intensive stylization) in Lang, Murnau, and Dreyer, and so on. Bazin’s broad defense of realism 1 2



directors in contrast worked more centrally within the Hollywood studio system. Their work made special use of the powers of the movie camera to track the world—for example, the motions of Fred Astaire and Ginger Rogers shot in a single continuous take, or the unfolding of a landscape in a panoramic shot of Monument Valley in a Western—and it was more likely to offer narratives for enjoyment than immediate provocation. This directorial tradition includes Howard Hawks, John Ford, and Frank Capra, among others. This distinction between two directorial traditions is, however, likewise crude. There are often narrative elements, sometimes absorbing ones, in montage-­oriented European avant-garde films, and American narrative films experiment in shot angle, editing, and point of view. Jean Renoir is a European realist, and Stan Brakhage is an American hyper-experimentalist. Directors such as Welles, Hitchcock, Ray, Scorsese, Altman, and Kurosawa are major makers of narrative films that are formally innovative and provocative. Since thought can be invited and advanced by narratives, including entertaining ones, and avant-garde images can be narratively compelling, the distinction between narrative entertainment films and intellectual provocation films is far from absolute. Nonetheless, Continental film theory and analytic philosophy of film tend to be responsive in the first instance to one of these two broad directorial traditions, with Continental film theory taking its cues from the Eisenstein, avant-­ garde, montage tradition and with analytic philosophy of film guided by the Ford, narrative, continuous action tradition. But just as traditions of directorial style have increasingly overlapped and converged—where would one put Steven Soderbergh or Andrzej Wajda?—so too have Continental and analytic film theories begun to overlap and converge, with each style of analysis enriching the other.4 And as with directorial styles, there have always been theorists, such as Jean Mitry, who use both analytic and hermeneutic methods and are not fully locatable in either camp. The best practitioners of both analytic philosophy of film and Continental theory are concerned both with broad social and cognitive functions of film as a distinctive medium as such and with particular decisions made and particular effects achieved in particular cases. Continental theorists are more likely to focus on the social and ideological functions of film, and their arguments are likely to proceed from strong premises about subject formation and the political shaping of interests. Noël Carroll, the most important contemporary analytic philosopher of film, has rightly criticized what David Bordwell dubbed SLAB theory (after Saussure, Lacan,

against avant-gardism set the stage, however, for the reaction against realism and in favor of avantgardism that emerged in the 1960s’ structuralist and post-structuralist film theory. 4  As we shall see, the best contemporary Continental film theory that focuses on social meanings and embodied experiences of specific films is no longer “Continental” in the sense of the high structuralist and post-structuralist theory of the 1960s and 1970s practiced by Christian Metz and Stephen Heath, among others, just as the best contemporary analytic film theory is no longer dominated by appeals to cognitive science.



Althusser, and Barthes)5 as practiced by film theorists such as Christian Metz, Stephen Heath, Kaja Silverman, and Terese de Lauretis for being “monolithic” in arguing that all films as such, just by presenting visual narratives that are to some extent intelligible, thereby reinforce already existing social ideologies involving the acceptance of such things as traditional gender roles, the existence of a real world that science is able to represent correctly, and the authority of a control-seeking rational economic agent. Why, Carroll reasonably asks, should we think that all narrative films do or must fulfill such ideological functions, when there is quite evidently an enormous variety of movies that are very different from one another? Why not go pluralistic and look at different films case by case?6 Against Carroll’s relaxed pluralism, SLAB theory argues that we ought to remember that at least the social world and its roles, and perhaps aspects of the natural world as well, are matters of construction and convention that can and perhaps should be changed. Why is an outcropping that reaches up five miles from the ocean floor just barely to break the water’s surface counted as an island, not a mountain? Why are the bodies of women presented more often than the bodies of men as objects of visual pleasure in films, and why are men more frequently the heroes of action films than women? Here, the guiding assumptions and questions of Continental film theory often have some kinds of empirical support, implicit or explicit, even if the readings of individual films that it generates are often (but not always) predictable, heavy handed, and monolithic. The bodies of women are more often presented as objects of visual pleasure than are the bodies of men (though there are exceptions: think of Cary Grant, the young Henry Fonda, or the young Brad Pitt, among others). Men are more often (but not always) the heroes of action films than women are. There are no films in which awkward women are saved by encounters with magic pixie dream boys. Arguably, then, there are two kinds or levels of phenomena that we would like to understand and explain: (a) what is more often or typically (but not always) the case in many (especially commercial successful) films, many of which (though not all) do reinforce widely prevalent ideological assumptions and (b) the distinctive successes—artistic, narrative-cognitive, and emotional-­ aesthetic—of some individual films,7 together with the production decisions that yield these successes in particular cases. 5  Bordwell coined the term Saussure, Lacan, Althusser, and Barthes “(SLAB) Theory” in “Historical Poetics of Cinema,” in The Cinematic Text: Methods and Approaches, ed. R.  Barton Palmer (New York: AMS Press, 1989), pp. 369–98, at p. 385. Bordwell and Carroll together discuss this approach under the title “Grand Theory” in their editors’ introduction to Post-Theory: Reconstructing Film Studies (Madison, WI: University of Wisconsin Press, 1996), pp. xiii-xvii. 6  Noël Carroll, “Prospects for Film Theory,” in Post-Theory: Reconstructing Film Studies (pp. 37–68). 7  Contra what SLAB theory sometimes suggests, these successes are by no means limited to experimentalist or avant-garde works produced outside the (Hollywood) mainstream. Who would deny the artistic as well as commercial successes of Ford, Hawks, Hitchcock, Coppola, or Scorsese?



The first major more or less analytic philosophical book on film is Stanley Cavell’s 1971 The World Viewed. Cavell draws significantly on already existing film theory, especially Panofsky and André Bazin, and he addresses the function of film as art in very broad terms rooted in his general philosophy of art, especially modernist art. Memorably, he writes that “Apart from the wish for selfhood (hence the always simultaneous granting of otherness as well), I do not understand the value of art.”8 Here, art in general is cast as offering us “the route back to our conviction in reality,”9 as though apart from the experience of art our lives were primarily matters of aimless drifting or the expenditure of labor in hyper-conventionalized and unsatisfying routines of work. Beyond this general view of the nature and value of art, however, Cavell develops a rich account (citing Panofsky) of “the unique and specific possibilities of the new medium.”10 This account emphasizes the ontological distinctiveness of the photographic image compared with the painterly image. About a photograph, it always makes sense to ask what was cropped out or otherwise left out of the frame, whereas a painting may include at whim anything a painter chooses to insert. A photograph is to a significant degree mechanically produced, so that it captures the world, while a painting is manually and artisanally produced. Adopting the tropes of Bazinian realism, Cavell remarks that the presentation of the real is achieved in a photograph, and so also in a film, “automatically, …magically, …without my having to do anything, …[thus] satisfying the wish for the world recreated in its own image,”11 so that I might experience its unfolding as meaningful without having myself to bear the burdens of responsibility and responsiveness within it. “The altering frame is the image of perfect attention”12 to the unfolding of significance within the real world, as the camera tracks the progress of a complete action.13 Second, film presentations (as Munsterberg and Arnheim had already noted) are strikingly different from theatrical presentations in having possibilities of close-ups, flashbacks, tracking shots, shifts in points of view, and so forth. No more, however, should we assume a priori that successful film must be primarily realist rather than avant-garde. Who would deny the artistic and commercial success and interest of late Godard, Maya Deren, Dziga Vertov, or Ang Lee? Neither avant-gardism nor realist narrative is either necessary or sufficient for artistic success. 8  Stanley Cavell, The World Viewed [1971], Enlarged Edition, (Cambridge, MA: Harvard University Press, 1979), p. 22. 9  Ibid. 10  Ibid., p. 30. 11  Ibid., p. 39. 12  Ibid., p. 25. 13  Cavell’s Bazinian realism about photography and photographically produced film (as opposed to cartoons or, more recently, computer generated imagery (CGI) movies) is consistent with—in fact, it presupposes and implies—the further thought that the presentation of the real that is achieved is itself also the result of directors’ and other producers’ decisions about angle of shot, lighting, focus, and so forth. See Richard Eldridge, “How Movies Think: Cavell on Film as a Medium of Art,” Estetika: The Central European Journal of Aesthetics, LIVIII, No. 1 (2014), pp. 3–20, for a full development of this point.



Hence, their possibilities of presenting content resemble those of novels as much or more than those of stage plays. In addition, movies are “screened and viewed,” not “performed” like plays.14 As a result, film acting is distinctively different from stage acting. Again, developing a thought of Panofsky’s, Cavell writes that “For the stage, an actor works himself into a role; for the screen, a performer takes the role onto himself ”15 as the camera explores a particular actor’s actually existing physiognomy, bearing, mien, and gait, most prominently by way of close-up. Hence, it is natural for us to refer to Humphrey Bogart or Katherine Hepburn movies, in which they play different characters while we see them in each of them. Cavell’s talk of a broad function of film as art—answering to the wish for selfhood (to be achieved within worldly activity, in relation to others)—may seem to be in line with Continental film theory’s focus on functions of film as such, and, to some extent, it is. After all, Cavell draws significantly upon Bazin and Panofsky, who themselves are working within a recognizably Hegelian tradition of thinking about art and social life that also figures within Continental film theory. In addition, the major inspiration for Cavell’s work on the nature of the human subject is Ludwig Wittgenstein, who would have counted as a core analytic philosopher in 1971, but who also describes the lives of human subjects in terms that have some Hegelian resonances.16 But there are also some important differences that locate Cavell more firmly within a broadly analytic tradition of thinking about the arts. Cavell’s account of the powers and interests of art is distinctly modernist and individualist, in dwelling on successes in making meaning in the face of social ideologies, sometimes via traditional narrative means, rather than on the pervasive reinforcement of those ideologies by non-avant-garde art. While Cavell accepts that human subjects as bearers of discursively structured points of view are socially formed (along more or less Freudian lines), he does not indict the social world as a whole for thus forming and normalizing them, and his modernist picture of the achievement of full and authentic subjectivity is both anti-collectivist and normatively contentful, in contrast with European structuralist and post-structuralist thinking about social life. Second, Cavell proceeds primarily by way of readings of individual films, especially in his two later books on film genres: Pursuits of Happiness: The Hollywood Comedy of Remarriage (1984) and Contesting Tears: The Hollywood Melodrama of the Unknown Woman (1997). Third, Cavell offers a fairly explicit ontological-functional definition of film as art: a sequence of automatic world projections that invites and sustains the conviction of viewers in the meaningful reality of what is presented, and he identifies specific mechanisms of film production that enable films to fulfill this function in their distinctive ways: the  Cavell, The World Viewed, p. 122.  Ibid., p. 27. 16  On the role of Hegel and post-Kantian European thinking more generally in Wittgenstein’s thought, especially as Cavell receives and develops it, see Eldridge, Leading a Human Life: Wittgenstein, Intentionality, and Romanticism (Chicago: University of Chicago Press, 1997). 14 15



close-up, the use of the star, the use of character types (the dandy, the military man, the clown), and deliberated genre elaboration and intertextuality. This detailed attention to definition, genre, and mechanism is significantly more prominent in analytic philosophy of film generally than in Continental film theory. Arthur Danto picks up many elements of Cavell’s definition of film. While Danto does not deny the magic of the experience of movies, he focuses somewhat less than Cavell does on a general artistic function of films as such, and he recasts Cavell’s points in a somewhat different vocabulary. Drawing on ideas that appear in his general philosophy of art in his 1981 The Transfiguration of the Commonplace, many of which were worked out earlier in his 1964 “The Artworld” and 1973 “The Last Work of Art: Artworks and Real Things,” Danto argues that films, like art in general, present a world that is already fully formed by its makers and closed to the viewer’s intervention. It is, Danto, tells us, “essentially true of art …that it logically excludes its spectator from the space and often the time that it occupies.”17 In the case of films, the presentation of a world is visual, not verbal alone (as in literature). Like Cavell, Danto notes that films are mechanically screened or shown, not performed (like plays), and that two prints of the same film are two tokens of the same type (unlike two visually similar paintings that retain distinct-type identities and meanings as singular objects, despite their similarities).18 Unlike paintings, which are viewed, films are watched: they present, or in principle can present, unfoldings of events as things to be watched-for. Even a perfectly static film only of the title page of War and Peace is different from a projected slide image of that title page, since in the film, but not in the slide image, something could move, if the filmmaker chose to have that happen.19 Like Cavell, Danto more or less takes it for granted that the images presented by films are photographically produced. Cavell acknowledges the existence of cartoons, but he holds that they are different enough from (other) movies that a different account of their possibilities of artistic achievement is called for. In cartoons (and in pure CGI productions), there are, arguably, “no real laws [of nature] at all.” Characters may float, fly, or stand still in the air, and “their bodies are [or may be] indestructible, one might almost say immortal,”20 as in Wile E. Coyote’s eternal recoveries from being squashed by an anvil. The natural tendencies, as it were, of cartoons and CGI films are to take fantasy, dreams, science fictions, talking animals, and superheroes as their subject matters. Danto develops this point about the distinct artistic possibilities of (pri17  Arthur C. Danto, “Moving Pictures,” Quarterly Review of Film Studies, 4, 1 (Winter 1979), pp. 1–21; reprinted in Philosophy of Film and Motion Pictures, eds. Carroll and Choi, pp. 100–112, at p. 100. 18  Ibid., p.  101. In the terminology of Nelson Goodman in Languages of Art (Indianapolis: Hackett Publishing Company, 1976), film is an allographic medium of art while painting is autographic. 19  Ibid., p. 102. 20  Cavell, The World Viewed, p. 170.



marily) photographically produced films by noting that photographs have a different kind of meaning or semantic content from non-photographically produced images. X is a photograph of Y if and only if (a) X is caused by Y (with some degree of automatism or mechanism) and (b) X denotes Y pictorially, so that the semantic value (its being true or false of something) varies with its semantic content (what it depicts). Though there can be blurry or misleading photographs of an object Y, and though photographic images are (normally) subject to manipulations by their makers, who make decisions about lens aperture, film stock, shutter speed, focus, and frame, there is no such thing as a false photograph of Y.21 Given its partly automatic-causal relation to what it depicts, the camera is especially well suited to exploring the looks of actual things. Danto distinguishes between the situation in painting, where an actual person (Mrs. Siddons, say) may be used as model for The Muse of Tragedy, and the situation in film, where an actual person (Harrison Ford, say) is used as an occasion for photographic exploration (as in close-up). In the latter case, according to Danto, the particular physical being of the actor can become foregrounded, in such a way that the actor can “swamp the role he or she is playing.” In general, “nothing counts as a different performance” of the same work by a different actor, as particular actors function as particular motifs to be explored by the camera.22 Were a new actor to play a particular role from an already existing film in a new movie, the result would be a remake, not a new performance of the same film. The result, according to Danto, of the fact that photographically produced films display a world narratively and visually by way of the camera’s exploration of really existing things, scenes, and people is a particularly intimate involvement on the part of individual spectators with unfolding meaning. Invoking a line from Proust’s description of the experience of theater, Danto remarks that in the experience of movies, “chaque spectateur regardait …un décor qui n’était que pour lui, quoique semblable aus milliers d’autres que regardait, chacun pour soi, les restes des spectateurs; each spectator viewed … a scene which existed for himself alone, though closely resembling the thousand other scenes viewed by the other spectators, one by one.”23 We can feel, that is, as if the movie is intimately unfolding in a meaningful way for us, one by one. Kendall Walton similarly argues that photographs and films, unlike paintings, possess an inherent realism, even when plots are fantastic. Unlike a painting, a photograph, and so too a film, functions as an aid to or “tool for vision”24 in seeing an actual thing, something like the way in which we see actual things prosthetically, as it were, through a telescope or a mirror. For example, when I  Danto, “Moving Pictures,” p. 104.  Ibid., p. 107. 23  Ibid., p. 101. 24   Kendall L.  Walton, “Transparent Pictures,” Critical Inquiry 11 (December 1984), pp. 250–273; excerpted as “Film, Photography, and Transparency,” in The Philosophy of Film, eds. Thomas E. Wartenberg and Angela Curran (Malden, MA: Blackwell Publishing, 1988), pp. 70–76 at p. 71. 21 22



see Aunt Mabel grimacing in a photograph, “I actually see, through the photograph, the grimace that she effected”25 on the photograph’s occasion. As for Danto, a blurry photograph of X is still a photograph of X. Though intentional manipulations of lens aperture, shutter speed, lighting conditions, and so on are normal, it is also the case that “our photographic equipment and procedures happen to be standardized in certain respects,” and in any case, whatever mediations, mechanical or intentional, are in place, both photographs and films possess a distinctive “transparency”26 so that they put us in visual touch with an actual object. Moving photographic images—films—may then be used, according to Walton, as props in games of make-believe, as we are guided, for example, to imagine Han Solo shifting into hyperspace by actually seeing Harrison Ford move a lever.27 Against the emphases of Cavell, Danto, and Walton on the photographic basis of film and on the transparency of the photographic image in presenting its object, Noël Carroll has developed a more general definition of film that is meant to capture cartoons and completely CGI-produced movies as well as traditional, mostly photographically produced films. We do think of photographic films, cartoons, and CGI productions as movies. Moreover, in general, in the arts, the physical basis of an art form does not by itself determine artistic possibilities and success conditions. For example, visual images can be made from oils, tempera, or water colors, among other things, and they can be rendered on canvas, paper, wet plaster, wood, or cave walls, and yet the products are still paintings. We think of them as paintings, and we experience them as paintings, albeit that they also display various more specific modes of artistic achievement. Sculptures may be made of wood, marble, or cast bronze. Typically, there are both physical bases and “different, nonconverging potentials and possibilities”28 for any medium of art. Why not say the same thing about movies: many different physical bases, many different (related, but not convergent) possibilities of artistic success? Instead of thinking of physical bases  Ibid., p. 73.  Ibid., p. 76. 27  Gregory Currie has argued against Walton and all versions of a transparency thesis by noting that “seeing a photograph of X is a matter of seeing a representation of X rather than of seeing X itself ” (Image and Mind: Film, Philosophy, and the Cognitive Sciences [Cambridge: Cambridge University Press, 1995], p. 51, emphasis added). While it is certainly true that what we see directly (Walton might say) when we see a photograph of X is the photograph, that is, a representation of X, it is also true that (as Currie concedes) “the representations photographs give us are certainly very different in kind from those we get by drawing and painting” (p.  51). The issue then is whether these differences can adequately be understood and explained without saying, as Walton does, that we also see X prosthetically (as it were) by or in our seeing of the photographic representation (of X). For more on visual depiction, with particular reference to Walton on making-believe and Wollheim on the twofoldness of representation (seeing the painting as a painting versus/and seeing what a painting is a painting of ), see Eldridge, An Introduction to the Philosophy of Art, 2d. ed. (Cambridge: Cambridge University Press, 2014), pp. 38–44. 28  Carroll, “Defining the Moving Image,” in Carroll, Theorizing the Moving Image (New York: Cambridge University Press, 1996), pp. 49–74; reprinted in Philosophy of Film and Motion Pictures, eds. Carroll and Choi, pp. 113–133, at p. 116. 25 26



of media as somehow determining possibilities of success, it is apt to take “stylistic ambitions [with respect to storytelling, audience involvement, compelling imagery, etc.] [to] dictate the production or choice of media.”29 Some films are photographically produced and transparent in presenting their objects, but some films (CGI, animated cartoons) aren’t. Given the comparative production costs of paying enormous salaries to stars to be photographed versus paying ordinary wages to teams of graphic designers, “photographic film may represent but a brief interlude in the art form.”30 Articulating the basis of this thought, Carroll then lists five conditions that are necessary for anything to count as a moving image (movie, film, television show, etc.)31: (a) an image is presented in a detached display (we are unable physically to orient ourselves in relation to or to enter the space of the presentation, as we can for a stage play); (b) motion on the part of objects presented within the image or on the part of the frame (camera panning, zooms) is a possibility (as Danto had argued); (c) this display is generated from a token, not via an interpretation (in the way in which plays are interpreted by directors, actors, set designers, etc.); everything is fully fixed within the unfolding images as screened or shown; (d) screenings or transmissions are not artworks in their own right (unlike stage performances, which involve additional art beyond the script alone); and (e) the display is two dimensional (this is true even for 3D movies, which are displayed on a 2D screen).32 This is all very sensible, and there is some point to emphasizing both the similarities between our experiences of photographically produced movies, cartoons, and CGI films and the freedom of filmmakers to use whatever physical bases they wish in order to achieve whatever effects they wish. At the same time, one may wonder (as Cavell does explicitly) about the extent to which our experiences of photographically produced movies and cartoons (and CGI) are similar. Each physical basis may offer its particular possibilities of distinct artistic magic, with the comparative realism of photographic movies contrasted with the freer fantasias of cartoons and CGI. One may also wonder, moreover, about the extent to which filmmakers consciously choose particular physical bases for their distinct artistic possibilities, which possibilities they explore and develop through the making of the work. The look and feel of successful photographically produced films do seem to vary together with choices of subject matter: the more directly transparent the presentation (with more continuity editing), the more realistic the narrative. Contrast, for example, the directly documentary, photographically produced Nanook of the North with the animation-­produced flights of fantasy of Fantasia or The Rabbit of Seville. It is, however, difficult to argue convincingly about the nature of film experience.  Ibid., p. 117.  Ibid., p. 122. 31  The five conditions are listed and argued for in Ibid. pp. 124–130. 32  This is true even for 3D movies, which are displayed on a 2D screen; this condition is necessary in order to rule out ballerina music boxes, for example. Ibid., p. 130. 29 30



Some are likely to feel that the photographic film-cartoon-CGI similarities in their experience are obvious and important; others are more likely to be struck by differences. Moving beyond the question of definition and maintaining a pluralist stance toward the different kinds of successes, artistic and otherwise, achieved in many films, analytic philosophers of film have developed a piecemeal approach to analyzing both individual cases and genres. This approach, generally known as New Cognitivism for its frequent use of data from psychology about emotional response, was initiated by the film scholar David Bordwell’s 1985 Narration in the Fiction Film. Bordwell opens his book by reviewing Aristotle’s account in the Poetics of the aims of tragic drama and of the plot structures, choices of protagonists, and management of points of view that are crucial to achieving those aims. He includes “A Sketch for a Psychology of Filmic Perception and Cognition,” and he discusses distinct narrative strategies in a range of films, including detective films, melodramas, Westerns, and modernist art films. Together with his then University of Wisconsin colleague Noël Carroll, Bordwell edited the 1996 anthology Post-Theory: Reconstructing Film Studies that functioned both as a manifesto for the approach and as a bringing together of work by major contributors to New Cognitivism, many of them analytic philosophers of film—Noël Carroll, Gregory Currie, Cynthia Freeland, Flo Leibowitz, Jerrold Levinson, Paisley Livingston, and Alex Neill—as well as sympathetic major film scholars: David Bordwell, Carl Plantinga, and Murray Smith, among others. Carroll produced a range of monographs and essay collections that developed the approach from 1988 on: Philosophical Problems of Classical Film Theory (1988), Mystifying Movies: Fads and Fallacies in Contemporary Film Theory (1988), The Philosophy of Horror, or Paradoxes of the Heart (1990), Theorizing the Moving Image (1996), Interpreting the Moving Image (1998), Engaging the Moving Image (2003), and The Philosophy of Motion Pictures (2008), among others. Together with their collaborators, Bordwell and Carroll significantly reshaped contemporary film studies, even while Continental film theory remained dominant in many major film studies departments, where work continued to be oriented primarily by post-­ structuralist philosophy and literary theory and to be directed to broader questions about social functions and ideologies.33 33  But see note 4 on contemporary developments. In his important 2007 essay “An Elegy for Theory,” the film scholar D. N. Rodowick traces the emergence of the New Cognitivism not only to the invocation of ideas from cognitive science and experimental psychology but also both to a generalized frustration with failures of specificity in readings of individual films within Continental film theory and to developing interests in film history and in emerging new media, especially digital media. “Since the early 1980s,” as he puts it, film theory in Film Studies departments has been marked by “a decentering of film with respect to media and visual studies and by a retreat from theory.” Rodowick notes that Bordwell’s Narration in the Fiction Film is shaped not only by his reliance on psychology but also by his engagement with “concrete problems of aesthetic practice” (“An Elegy for Theory,” October, 122 (Fall 2007), pp. 91–109; reprinted in Critical Visions in Film Theory, eds. Corrigan, White, and Mazaj, pp. 1110–1126 at pp. 1111, 1113).



The development of New Cognitivism has centered around three related but distinguishable broad topics: narration, genre, and emotion. As already noted, in his 1985 book, Bordwell outlined distinct narrative strategies that work effectively within different genres in order to maintain the emotional engagements of viewers. Subsequent work within the New Cognitivist approach goes into more detail. For example, Gregory Currie usefully distinguishes between character desires that focus on the fates of characters in a film and narrative desires that focus on the development of a film.34 Viewers of Casablanca may both wish for Rick and Ilsa to stay together (while recognizing, too, that it is better that they do not) and wish for Casablanca to end appropriately with Rick turning to Louis with the movie’s memorable closing line. Both character desires and narrative desires can arise with respect to both fiction and documentary films. Similarly, adapting insights from literary narrative theory, Murray Smith notes that it is important to distinguish recognition (of a projected image as an image of a character), alignment or focalization (in which the experiences and attitudes of particular character who may or may not be sympathetic are  presented via point-of-view shots or otherwise highlighted), and allegiance (when viewers’ sympathies for a particular character are motivated by a film).35 Though they often occur together, recognition, alignment, and allegiance are independent of each other, and they can and do come apart. There are often “complex patterns” of a film’s mobilization of recognition, alignment, and allegiance that “may preclude or transcend a single, strong engagement with a single character.”36 In general, in order to give a full and accurate account of viewers’ emotional engagements, it is important to distinguish multiple levels and foci of response, all of which may be in play simultaneously, rather than focusing only on a single dominant response to all films as such, as in some Continental film theory. In general, it is important to be clear about how, as a result of filmmakers’ decisions, movies guide and manage viewers’ responses through their narrative structures. Adapting terminology from Russian Formalist narrative theory, Bordwell distinguishes between the fabula of a narrative, or the events of a presented story as they actually occur (or are made-believe to have occurred) in chronological time, or the order of occurrence, and the syuzhet of narrative, or the order in which the events are actually presented by the film, or the order of telling.37 As Munsterberg and Arnheim noted, these can come apart, given the availability in film construction of flashbacks and flash forwards. These 34  Gregory Currie, “Narrative Desire,” in Passionate Views: Film, Cognition, and Emotion, eds. Carl Plantinga and Greg M. Smith (Baltimore: Johns Hopkins University Press, 1999), pp. 183–99; excerpted in The Philosophy of Film, eds. Wartenberg and Curran, pp. 139–47 at p. 140. 35  Murray Smith, Engaging Characters: Fiction, Emotion, and the Cinema (Oxford: Clarendon Press, 1995), pp.  73–5, 81–95, 108; reprinted in The Philosophy of Film, eds. Wartenberg and Curran, pp. 160–69. 36  Ibid., p. 168. 37  Bordwell, Narration in the Fiction Film (Madison, WI: University of Wisconsin Press, 1985), pp. 48–61.



techniques can be used for either avant-garde experimental effects, as in Alain Resnais’ Last Year at Marienbad, or to motivate the viewer’s reflective engagement with a more directly commercial film, as in Marc Webb’s 500  Days of Summer. George Wilson has carried out a major study of how filmmakers manage points of view and viewers’ access to the events presented in a variety of ways. Wilson is interested in particular techniques of editing or shot sequencing that can give rise to impressions of causality on the parts of viewers. For example, a three-shot progression in Orson Welles’ The Lady from Shanghai suggests to viewers that a woman’s hand pushing a button mysteriously causes an in-fact spatiotemporally unrelated traffic accident involving a car driven by men. Since the causality here cannot be actual, we are forced either to take the sequence to be an editing error or, more likely, to take it “to make a kind of general, metaphorical statement, interjected here by the filmmaker, about, say, the uncanny power and influence of the woman over the men.”38 Wilson goes on to discuss in detail films by Fritz Lang, Alfred Hitchcock, Max Ophuls, Josef von Sternberg, and Nicholas Ray, focusing on how viewers are guided in particular ways in their making sense of what is going on by filmmakers’ decisions about framing, lighting, close-up, and sequencing. Building on work by Aristotle and J. L. Mackie, Noël Carroll has developed a general theory of narrative, according to which a narrative, unlike an annal, chronicle, or mere list of events, must have “a unified subject” (however complex) and display (however indirectly) “perspicuously time-ordered” events that are connected in that earlier events in the fabula or chronological order must function as nonredundant members of a set of unnecessary but sufficient conditions for the occurrence of later events.39 More informally, earlier events must be presented (even if presented out of chronological order in the syuzhet) as relevantly causing or bringing about or producing later events. As Wilson’s work argues, there are a variety of techniques available to filmmakers for establishing the existence of relations of relevant bringing about between events and viewers’ uptakes on them. Causal connections may be shown directly, as in when we see Sam Spade shoot Canino in The Big Sleep, thereby causing his fall and death. Or they may be presented more indirectly, as in when we see close­up shots first of Bruno Antony’s highly polished two-tone wingtips, then of Guy Haines’ more ordinary, less polished brogues in Strangers on a Train—a sequence that suggests characterological differences that will play themselves out causally over the course of the story. In the actual world, and therefore too in the world that has been photographically captured on film, many events cause and are caused by a wide range of prior events (event-causality is overdetermined). The lighting of a match 38  George M. Wilson, Narration in Light: Studies in Cinematic Point of View (Baltimore: The Johns Hopkins University Press, 1986), p. 2. 39  Carroll, “On the Narrative Connection,” in Carroll, Beyond Aesthetics: Philosophical Essays (Cambridge: Cambridge University Press, 2001), pp. 118–133, at pp. 121, 123–24.



may initiate the smoking of a pipe, start a forest fire, or function as a signal to a confederate. A death may be caused all at once by a bullet, by a hand that pulled the trigger, and by offhand insult that motivated the trigger-pulling. No film can focus on all events related as causes and effects. Hence, filmic narratives that do focus on specific cause-effect relations must have some particular point in doing so. Carroll has argued that a narrative (unlike a chronicle) and, in particular, a narrative that “sustains closure” (unlike an indefinitely unfolding soap-opera), “is a network of questions and answers.”40 Directors and other constructors of film narratives exploit the fact that “some scenes or sequences evoke questions; others answer said questions directly.”41 Thus, we are directed by the presentation of Nanook and his companions as setting out on a seal hunt to wonder or ask whether they will succeed in killing a seal and so in acquiring the food, sealskin hides, and tusks that they need. Likewise, when Joan of Arc is arrested and put on trial for heresy, we wonder and ask whether she will be convicted and whether she will confess to heresy and fabrication in order to avoid severe punishment or whether she will rather remain faithful to her experience and accept being burned at the stake. Narrative closure occurs when (and only when) the most important questions that have been raised for the audience by the opening scenes have been fully and appropriately answered: the couple in the romantic comedy end up together (or not, if things haven’t “worked” between them), or we learn who shot Liberty Valance. Both fictional and documentary films often run from one to two hours, and they typically aim at narrative closure and achieve it, if they are minimally well done. More complex, interesting, and better-constructed films will raise more complex questions and arrive at answers in complex but satisfying ways. As Currie has noted, audiences typically narratively desire that a satisfying appropriate end be reached (however bad that end might be for the movies characters, as in the complete eradication of the conquistadors in Werner Herzog’s Aguirre: The Wrath of God), and moviemakers typically seek to satisfy this desire. But there are some exceptions: works that are more open-ended and episodic, without definite ends, such as Michal Apted’s 7 Up series of films (now at eight films and still possibly going) or Andy Warhol’s eight-hour, five-­minute Empire featuring continuous slow motion, fixed camera footage of the Empire State Building. Even in these cases, however, that lack full narrative closure, the audience is guided through a structure of multiple questions and answers: we wonder what has happened over the last seven years to each of Apted’s characters, what will happen to them next, and why; and we both wonder what is going on in the Empire State Building as its lights go on and off and why Warhol chose to present these incidents in this way (we struggle to find a thematic meaning for the movie that explains how and why it is minimally plotted as it is). In general, it is part of the structure of a film that it is experienced as an unfolding narrative whose events are to some extent held together by  Carroll, “Narrative Closure,” Philosophical Studies 135 (2007), pp. 1–15 at p. 5.  Ibid.

40 41



­relations of bringing about through which initiated questions are answered (or revised, dismissed, and displaced by others). As already noted, Wilson42 and Bordwell have completed detailed studies of specific techniques through which films establish relevant relations of bringing about and reach their ends in narratively plausible ways. The same or similar enough narrative strategies of prompting questions or concerns and providing satisfying answers or resolution are frequently employed in many films that thus form a genre. Plausibly, uses of these strategies have both cognitive-psychological and social explanations. Carroll has attended closely to the ways in which filmmakers who make melodramas, horror films, and suspense movies both presuppose and mobilize cognitive and emotive processing capacities on the parts of individual viewers. Within these genres (and others), a filmmaker chooses subject, camera position, composition, lighting, editing, music, and dialogue in order to present an object or person as falling under a certain concept. For example, a person A may be presented as threatened by a dangerous (because dangerous-looking) ax-wielder around a corner. Carroll dubs this initiating presentation that raises concern in the viewer for the well-being and fate of the person thus threatened “criterial prefocusing.”43 This criterial prefocusing establishes an “emotive focus” on the part of the viewer: a pro or negative attitude, a concern, held by the viewer with regard to the possible or likely fate of the threatened character (as in Currie on character desires), and the movie reaches its end or ends in answering the question of what will happen, hence satisfying the viewer’s concern. Within specific genres, criterial prefocusing and emotive focus are established in more or less standardized ways. For example, in a horror film, an attractive protagonist will be presented as menaced or threatened by an antagonist that is disgusting or loathsome as a result of being in some way impure: that is, in being formless (a blob or slime), ontologically mixed (zombies, the living dead; reptilian quasi-­ human aliens), or unnatural (giant spiders or worms, sea monsters, outrageously insane madmen, werewolves, etc.)44 The use of standardized criterial prefocusing, emotive focus, and question-­ and-­answer narrative structure is more dominant in, and more explanatory of, quite standardized genre films than of the achievements of high art in individual cases. In singularly successful cases of film art, the presentation is likely to be more mixed. For example, we feel more sympathy for King Kong than we do for his captors and tormentors, despite the giant ape’s unnatural size. Yet here too the same cognitive processing capacities are called into play by the film: the reason we feel more sympathy for Kong is that he is presented as suffering his torments (and so having an emotional life with which we s­ ympathize)— 42  Wilson, Narration in Light; Bordwell, Narration in the Fiction Film, and Bordwell, The Way Hollywood Tells It: Story and Style and Style in Modern Movies (Berkeley and Los Angeles, CA: University of California Press, 2006). 43  Carroll, “Film, Emotion, and Genre,” in Passionate Views, eds. Plantinga and Smith, pp. 21–47; reprinted in Philosophy of Film and Motion Pictures, eds. Carroll and Choi, pp. 217–33, at p. 222. 44  Ibid., pp. 227–29.



criterial prefocusing—while his pursuers are presented as mostly anonymous, intersubstitutable functionaries. Still more complex, original cases that cross genre lines or that lie largely outside fixed genres will require detailed analyses of multiple devices of presentation in order to see how they guide the complex, modulating understandings and emotional responses of their viewers. Yet even in these complex, original cases, it is plausible to suppose that there are multiple uses of criterial prefocusing, emotive focus, and causal and question-­ answering narrative structures, as more complex, mixed trajectories on the parts of multiple, more complex, mixed characters are developed, so that more complex, mixed understandings and felt responses are set up in viewers. That is, it seems plausible to suppose that there are some useful generalizations available from cognitive psychology and the theory of emotions that characterize and explain how individual viewers understand and respond to at least some details of almost all films, and supplying these useful generalizations (that also guide filmmakers’ decisions) is exactly what Bordwell, Carroll, Wilson, Currie, and others have done. At the same time, it also seems reasonable to consider distinctly social and psychoanalytic causes of both film genres and narrative structures, as those working more within the style of Continental film theory have done. Taking as his “preliminary working hypothesis,” the thought that “the determining, identifying feature of a film genre is its cultural context”—that is, “its [presented] community of interrelated character types whose attitudes, values, and actions flesh out dramatic conflicts within that community,” the film theorist Thomas Schatz argues that the choice and the development of action within a presented film world are determined by “our shared needs and expectations as audience members …for animation and resolution of basic cultural conflicts.”45 Viewers expect and demand the presentation of various social conflicts between lived values (and types who embody them) that run through their actual lives: for example, honor-clan-family versus public order, democracy, and justice (as in The Godfather), or community-normality-stasis versus threat, impurity, corruption (as in horror and alien films), or marriage-settled town life-agriculture versus violence-nomadism-isolation-transgressive individuality (as in Westerns).46 In broad terms, the iconography of a film—its narrative and visual coding—is determined by an ideology (a sense of social and value conflicts and of routes of address to them) under which viewers actually live.47 Films will both sell better and be more readily enjoyed if they speak to value conflicts that are present in the lived social experiences of viewers.48 45  Thomas Schatz, Hollywood Genres: Formulas, Filmmaking, and the Studio System (New York: Random House, 1981), excerpted in Critical Visions in Film Theory, eds. Corrigan, White, and Mazaj, pp. 454–65, at pp. 455, 460. 46  Ibid., p. 457. 47  Ibid., p. 456. 48  Rodowick notes that if the role of individual cognitive psychology is over-emphasized (at the expense of social facts) in determining the structures of films, then what results is a promotion of a “concept of the [decontextualized] rational agent [in production, spectating, and theorizing] …



Nonetheless, despite the importance for successful films of a continuing basic “concern for certain basic cultural issues,”49 change and variation are both possible and normal. Though many social conflicts at an abstract level are “essentially unresolvable [and] irreconcilable,”50 the specific shapes these conflicts take within actual life may change, and so may attitudes toward them. In addition to these external, social causes of variations in film iconography, there are also internal, formal causes, as filmmakers seek to present cultural worlds on film in ever more original and complex ways.51 Frequently, a given genre will develop from an initial transparency and realism in presenting social conflict toward increasing opacity, formal concern, ambiguity, and thematic complexity, as those working within a genre question and revise its conventions (in ways that viewers recognize and even expect).52 Consider, for example, the development of the Western from Stagecoach and High Noon to Little Big Man and Dances with Wolves. Echoing Schatz’s account of increasing opacity, ambiguity, formal concern, and thematic complexity, Bordwell has described the so-called art-cinema that became prominent in the late 1950s and early 1960s (Godard’s Breathless, Truffaut’s The 400 Blows, Antonioni’s L’avventura and L’Eclisse, Resnais’s Last Year at Marienbad) as a form of realism that both responds to and expresses a new sense of social life as alienating. These films have “a drifting episodic quality,” as they display a “broken teleologism” in tracking a “biography of the individual” that is marked by “picaresque subjectivity” rather than an encounter with a clear problem situation and the achievement of a solution to it.53 “Events may lead to nothing”; as these films reflect and present a “realization of the anguish of ordinary living” and a “dissection of feeling” as episodic and dissociated.54 Coherent trajectories on the part of protagonists through continuous, intelligible space-time are often dropped in favor of montage editing that presents flashbacks, fantasies, or directorially introduced analogies and in a perspective that strives to be free of ideological positioning and to assert an epistemology that is value neutral.” When this happens, “the activity of theory is given over to science,” ethical and political concerns are displaced, and “philosophy itself begins to lose its autonomy and identity” as a form of humanistic critique (Rodowick, “An Elegy for Theory,” pp.  1114, 1115). Pointedly, Carroll denies any commitment to “film theory [as] a science,” and he urges a dialectical-critical conception of theory (Noël Carroll, “Prospects for Film Theory,” pp. 59, 56). 49  Ibid., p. 462. 50  Ibid., p. 461. 51  Ibid., p. 462. Recently, Robert Pippin, working from a Hegelian background, has completed two significant studies of film genres as embodiments of responses to socially shared political and moral anxieties and concerns: Hollywood Westerns and American Myths: The Importance of Howard Hawks and John Ford for Political Philosophy (New Haven: Yale University Press, 2010), and Fatalism in American Film Noir: Some Cinematic Philosophy (Charlottesville, VA: University of Virginia Press, 2012). 52  Ibid., pp. 463–64. 53  David Bordwell, “The Art Cinema as a Mode of Film Practice,” in Bordwell, Poetics of Cinema (New York: Routledge, 2008), pp. 151–69; reprinted in in Critical Visions in Film Theory, eds. Corrigan, White, and Mazaj, pp. 559–73, at p. 561. 54  Ibid.



symbols. A sense of the presence of the director as auteur is more prominent in the audience’s experience, as viewers are compelled to ask what the story is, who is telling it for which reasons, and why it is being told in this way (picaresquely, and through directorial “stylistic signatures,” often involving freeze frames, fast cuts, and unusual camera angles, as well as flashbacks).55 Whatever their prominent formal self-markings as works of directorial art, however, these films also present (often difficult) stories that viewers may be presumed to process and understand, however incompletely and improvisatorially, through the general cognitive procedures outlined in New Cognitivist philosophy of film. Moreover, more traditional, continuity-edited movies continued to be made alongside self-consciously art-cinema films. Today, the situation is that of a widened palette of possibilities of both plot (both problem encounters plus solutions and more episodic drifts by individuals) and presentation (both continuity editing and montage jump cuts, flashbacks, and freeze frames). Directors such as Steven Soderbergh, Danny Boyle, and Sam Mendes have learned to combine techniques of both continuity editing and montage editing to produce compelling movies that invite viewers’ attentions to their filmic surfaces while also remaining narratively and depictively coherent. Within an expanded palette of possibilities, the choices made by directors (along with screenwriters and other producers) will typically involve a mixture of cognitive psychological-­ information processing causes, social causes, and formal (originality-seeking causes). Both Continental film theory and analytic philosophy of film have begun to respond to these important, complex cases, so that they have begun somewhat to converge with each other in their attentions simultaneously to all of these causes. Drawing significantly on New Cognitivism, but going beyond it to discuss not only narrative, emotion, and genre but also affect, fantasy, social pleasure, and sensuality, Carl Plantinga, focusing primarily on the Hollywood fiction film, has developed a comprehensive account of the spectator’s experience. According to Plantinga, not all pleasure enjoyed in the experience of film is regressive (contra Freud), though some is; conversely, not all pure clarity of attention without emotion is progressive (contra Brecht), though some is.56 Instead of focusing on a single emotional experience of all films as such, we should pay attention to “the fundamental tenet of a cognitive approach …that the spectator’s affective experience is dependent on cognition, on mental activity cued not only by film form but also by story contents.”57 In addition, however, to involving cognition of both film form and story contents, the experience of viewers is also shaped both by the practices, institutions, and roles of movie-­ going, which involve “a continuous background experience of artificiality,”  Ibid., p. 563.  Carl Plantinga, “Notes on Spectator Emotion and Ideological Film Criticism,” in Film Theory and Philosophy, eds. Richard Allen and Murray Smith (Oxford: Clarendon Press, 1997), pp. 372–93; excerpted and reprinted as “Spectator Emotion and Ideological Film Criticism,” in The Philosophy of Film, eds. Wartenberg and Curran, pp. 148–59, at pp. 148–51. 57  Ibid., p. 152. 55 56



and by “a striking sensual and perceptual experience not found in literature” (as both Murray Smith and Laura U.  Marks have emphasized).58 Plantinga goes on to distinguish at least 15 different aspects—the list has overlapping elements—of the viewer’s experience in responding to a film: cognitive processing (inference making, question asking, etc.), visceral responses [as to a loud noise on a sound track or a sudden burst of light filling the screen], character engagement (including sympathy, primitive empathic responses such as emotional contagion and antipathy), emotional responses to unfolding narrative scenarios and both artifact and meta-emotions, emotions elicited by the viewer’s idiosyncratic associations and memories, moods, pleasures, desires, sexual arousal, various kinds and degrees of kinesthetic turbulence, affective mimicry, …reflex actions, and other sorts of affect in response to narrative, spectacle, setting, music, and the spectator’s own prior subjective experience.59

In general, “affective experience and meaning are neither parallel nor separable, but firmly intertwined,”60 and their developing modulations in the viewer are shaped by a wide and interacting set of film features and devices. In this respect, as Berys Gaut has noted in arguing for the medium specificity of film experience and film achievement (against Carroll’s anti-medium essentialism, but including not only cartoons and CGI films along with photographically produced films but also potentially video games as works within the film medium), film “instantiates artistic devices that are distinct from those that are instantiated by other media.”61 Moreover, part of the viewer’s engagement can also be due, Plantinga notes (as Schatz argued), to the fact that some “popular narratives, in embodying virtual solutions to traumatic problems, …play a role in the development of what might be called distributed or social cognition”62 of the conflicts that run through some stretches of joint social life. Insofar as films can both address or thematize shared problems of individual and social life and can do so through distinct presentational means, it seems natural to wonder whether at least some films arrive at answers to issues about meaning, value, and knowledge in human life. That is, can films as films do philosophy—make progress not only in illustrating philosophical views or claims that might and must be argued for elsewhere and otherwise but also in lending argumentative support to them. In Unlikely Couples: Movie Romance as Social Criticism, Thomas E. Wartenberg argues that “popular film …can be a locus for reflection on the sorts of issues that have traditionally been the 58  Ibid. See also Laura U. Marks, The Skin of the Film: Intercultural Cinema, Embodiment, and the Senses (Durham, NC: Duke University Press, 2000). 59  Plantinga, Moving Viewers: American Film and the Spectator’s Experience (Berkeley: University of California Press, 2009), p. 140. 60  Ibid., p. 2. 61  Berys Gaut, A Philosophy of Cinematic Art (Cambridge: Cambridge University Press, 2010), p. 300. 62  Plantinga, Moving Viewers, pp. 225–26; emphasis added.



domain of philosophy.”63 Drawing on a range of films including It Happened One Night (also treated at length by Cavell in Pursuits of Happiness), Pretty Woman, Guess Who’s Coming to Dinner?, Jungle Fever, Mississippi Masala, and The Crying Game, among others, Wartenberg argues that films can critique social hierarchies; they can challenge ossified expectations about couple formation through strategies of inversion, counterexample, transformation, and destabilization in presenting processes of unlikely couple formation to which viewers’ sympathies and allegiances are mobilized by the qualities of the characters’ specific interactions.64 Hence, contra some strains of over-essentializing Continental film theory that dwell on narrative film as primarily an instrument for the reinforcement of gender and class ideologies, films can do social criticism, and some films do it effectively. In Thinking on Screen: Film as Philosophy, Wartenberg extends this line of argument about the powers of film to “make arguments, provide counterexamples to philosophical claims, and put forward novel philosophical theories”65 to a wider range of philosophical topics, including exploitation (Modern Times), skepticism (The Matrix), autonomy (Eternal Sunshine of the Spotless Mind), and practical wisdom (The Third Man). It seems exactly right to say that these films address these issues, thematize them, and motivate conclusions about them on the parts of viewers. Here, philosophers who demand proof or deductively valid arguments for interesting conclusions from innocuous premises are likely to balk at the thought that these movies amount to doing philosophy in the central meaning of that phrase. Against them, however, one may also wonder whether canonical works of philosophy have ever succeeded in addressing significant issues about meaning, value, knowledge, and reality via demonstrative argument alone, without also relying on imagery, patterns of emplotment, and the mobilization of emotion in their audiences.66 Beyond, moreover, Wartenberg’s efforts strictly to show that some films conform to some of the canonical procedures of properly academic philosophical argumentation such as the supplying of counterexamples and the posing of thought experiments, one might wish more explicitly to broaden the understanding of genuine philosophical argumentation to include the distinctive powers of film to capture things on film, present narratives, mobilize allegiances, and so on. Perhaps film

63  Thomas E. Wartenberg, Unlikely Couples: Movie Romance as Social Criticism (Boulder, CO: Westview Press, 1999), p. xv. 64  Ibid., pp. 236–40. 65  Wartenberg, Thinking on Screen: Film as Philosophy (London: Routledge, 2007), p. 9. 66  On the limitations of demonstrative argument in philosophy in general and on the inevitability of at some point moving “beyond proof,” see Eldridge, Leading a Human Life: Wittgenstein, Intentionality, and Romanticism, especially pp. 1–15, 264–290, and Friedrich Waismann, “How I See Philosophy,” in Contemporary British Philosophy, 3d. series, ed. H. D. Lewis (London: George Allen and Unwin, Ltd., 1956), pp. 445–90. Perhaps “learning to see” is as much or more central to philosophy as demonstrative argument.



can not only do philosophy but also help us to understand how to do philosophy in more authentic and fulfilling ways.67 Some recent work has embraced all of the special presentational powers of photographic film as a medium of art; the wide range of viewers’ cognitive, emotional, and affective engagements with film (together with specific devices for achieving them) that have been explored in the New Cognitivism; and the ability of films productively to thematize and address issues about meaning, value, and knowledge as they are experienced in individual and social life. Robert Sinnerbrink draws on both Cavell and Deleuze in developing what he calls cinematic ethics or an account of film as a medium of ethical experience.68 Sinnerbrink is more interested in “film-philosophy,” that is, “a way of thinking at the intersection between philosophy and film, linking the two in a shared enterprise that seeks to illuminate the one by means of the other.”69 Sinnerbrink moves unhesitatingly from Deleuze on the action-image and the time-image, to Cavell on film as a sequence of automatic world projections that aid in undoing skeptical despair, to Carroll and Bordwell on devices for the mobilization of viewers’ emotions. Similarly, Daniel Martin Feige draws on Hegel on  See the sympathetic reviews of Thinking on Screen that also make this critical point by Cynthia Freeland, “Comments on Thomas Wartenberg’s Thinking on Screen: Film as Philosophy,” Projections: The Journal for Movies and Mind 3, 1 (Summer 2009), pp. 100–109, and Eldridge, “Philosophy In/Of/As/And Film,” Projections: The Journal for Movies and Mind 3, 1 (Summer 2009), pp. 109–116, as well as Wartenberg’s “Response to My Critics,” Projections: The Journal for Movies and Mind 3, 1 (Summer 2009), pp.  117–25. Part I of Paisley Livingston’s Cinema, Philosophy, Bergman: On Film as Philosophy (Oxford: Oxford University Press, 2009) carefully surveys the film-as-philosophy issues. Livingston concludes his discussion and defends film as philosophy by arguing that “what must be rejected …is the idea that we must make a choice between doing philosophy with film and doing philosophy with the linguistic and conceptual tools with which philosophy has been done prior to the advent of cinema” (p. 56). Progress can be made by taking philosophy, filmmaking, and film criticism to be practices that work well when in dialogue with each other. Livingston goes to investigate Ingmar Bergman’s explorations of persistent human irrationality and cruelty in detail. 68  Robert Sinnerbrink, Cinematic Ethics: Exploring Ethical Experience through Film (London: Routledge, 2016). Here, Sinnerbrink develops a suggestion of Rodowick’s that film theory and criticism in the style of both Deleuze and Cavell might amount to “intertwining projects of evaluating our styles of knowing with the examination of our modes of existence and their possibilities of transformation” in “a fluid, metacritical space of epistemological and ethical self-examination” (Rodowick, “An Elegy for Theory,” p. 1118). Against Rodowick, Malcolm Turvey has argued that philosophy should have at best a propadeutic role in clarifying certain basic concepts in film theory such as perception. Against film theory as philosophy, Turvey specifically recommends that “we … use our expertise—gained from watching large numbers of films, observing them and the responses of viewers to them carefully, and learning about the contexts in which they were made and exhibited—to evaluate the theories we take from other disciplines in terms of whether they successfully explain (or not) film” (Turvey, “Theory, Philosophy, and Film Studies: A Reply to D. N. Rodowick’s ‘An Elegy for Theory,’” October 122 (Fall 2007), pp. 110–20 at p. 120). Here, Turvey goes too far in taking philosophy, contra Wittgenstein, to consist in observing and explaining rather than in clarifying (and potentially changing) conceptual and practical commitments—a critical activity that philosophy, film, and film criticism in dialogue might best exemplify in helping us to see our lives and our lives with films more clearly (See notes 65 and 66.). 69  Ibid., p. 14. 67



r­ elations between art and life to develop a picture of film as a dynamic, processual, critical, open-ended interrogation of how we are doing in forming and undertaking to live according to cognitive, ethical, political, and moral commitments.70 Echoing Cavell, Martin Seel has emphasized both the “heightened passivity” of viewers, who typically (as in the experience of music) cannot withdraw their attentions from a filmworld that continuously unfolds before them temporally, and the ways in which movies can distinctively draw on and refigure presentational devices from other forms of art: images and spatial points of view from painting and photography, narrative and focalization from literature, scoring from music, and acting and dialogue from theater.71 In Film Worlds, Daniel Yacavone treats movies-as-art as all at once presenting a world on film (often captured photographically), as ordering and representing events in the film world diegetically, as engaging viewers’ emotions, as cognitively and symbolically thematizing the events they present, as ‘working through’ and disclosing existential truths about human commitments within worldly practices, and as formally achieving a global affective presence that invites and sustains an audience’s absorption.72 As this recent work illustrates (and as the work of Cavell established, along with the work of his predecessors Munsterberg, Arnheim, Panofsky, Bazin, and Mitry), it is both possible and fruitful in the study of film to attend both to specific films and specific presentational strategies (as in much analytic philosophy of film) and to the mesmerizing powers of (the best works within) the film medium as such, which powers are frequently deployed creatively to address matters of essential human concern (as in the best Continental film theory). The study of film as a medium and as a form of art has been immensely enriched by both analytic philosophy of film and the work of many film scholars who have been influenced by ultimately Hegel-­ derived theories of the formation of the subject and of the inextricably cognitive-­emotional-aesthetic-social functions of art. Film study as a whole is now in a more mature, richer, more interesting, and fruitful stage of its development than ever before.73

70  Daniel Martin Feige, Kunst als Selbstverständigung (Münster: Mentis, 2012), especially “Film,” pp. 171–90. 71  Martin Seel, Die Künste des Kinos (Frankfurt a. M.: Fischer, 2013). 72  Daniel Yacavone, Film Worlds: A Philosophical Aesthetics of Cinema (New York: Columbia University Press, 2015). 73  I am grateful to Daniel Yacavone for spectacularly detailed, focused, and useful comments on an earlier version of this essay.


When the Twain Shall Meet: On the Divide Between Analytic and Continental Film Philosophy John Ó Maoilearca

Introduction: Never Mind the Gap Let us begin with a rather dramatic scene, set by Joseph Westfall in the opening to his recent collection, The Continental Philosophy of Film Reader: In general, I do not think it is useful to respond to the claim that Continental philosophy is not really philosophy, both because the term ‘philosophy’ has been contested since the time of the Greeks (and thus what ‘counts’ as philosophy is not easily settled), and because, when the charge is made against Continental philosophers that they are not engaged in the practice of philosophy, it is not typically made as a means of opening a discussion, but of closing one. It is a dismissal of the relevance of what Continental philosophers have to say before what they have said has been seriously considered, and is thus not a position that invites (or even tolerates) a Continental response.1

The purpose of this chapter is neither to refute nor corroborate Westfall’s statement in full, but rather to help us see how we might move beyond its depressing prognosis. The reasons for this are based on recent evidence that, in fact, a conversation is taking place, if not between any particular Continental and Analytic figures, then at least between their ideas, their themes. In other words, 1

 Joseph Westfall, The Continental Philosophy of Film Reader (London: Bloomsbury, 2018), 6.

J. Ó Maoilearca (*) Kingston School of Art, Kingston University, London, UK © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




there is evidence that beyond certain fixations on writing style and a rather unfortunate history of hyperbole, Continental and Analytic philosophy of film have much to discuss in terms of what they actually do with their subject matter (even if not in terms of what they say that they do). In even other words, then, the differences are meta-philosophical—they concern how philosophy of film per se is seen, how it is defined rather than how it is practiced. And, as we shall see, that metaphilosophy has its roots in history, to be precise, the history of Analytic and Continental philosophy. It is that history that we investigate here, both to chart the genealogy of two philosophical “worldviews” so to speak, and also to update and temper the divide with examples of contemporary practice from working film philosophers of every hue.2 On this front, then, we would partly agree with the Analytical film philosopher Murray Smith when he writes that “the image of a mutually exclusive and exhaustive contrast between Analytic and Continental philosophy is in many respects misleading and inadequate as a picture of western philosophy, past and present.”3 In various ways, Smith is correct here, but still not completely—as we hope to show. Consequently, beyond fulfilling the much-needed function of surveying where film philosophy has come from alongside how it has developed more recently, the chapter also forwards the aforementioned thesis, namely that among contemporary practitioners of film philosophy, there is a vast amount to share from all sides in what they practice or actually do rather than in what they say about what they do (especially as the latter is aligned with certain avowed philosophical lineages). According to the Analytic philosopher Ilham Dilman, genuine philosophy is concerned with how we think, are aware of our surroundings, and speak and communicate, as well as the nature of language, mathematics, the empirical sciences, psychology, sociology, and, of course, philosophy itself. It attempts to “clarify” these topics, concerning itself with “sense or meaning” all the while.4 Brenda Almond, another Analytic, adds that it looks for continuities, structures, and patterns. It seeks to get nearer to truth. It also extols the virtues of honesty, openness, courage, impartiality, and respect for persons.5 Alternatively, according to the figure who is, for many, the consummate Continental philosopher, Jacques Derrida: “I must honestly say that now, less than ever, do I know 2  Unsurprisingly, perhaps, even our choice of the term “film philosophy” as opposed to “philosophy of film” (or “philosophy of cinema”) might be contested by some: it is adopted in order to maintain a bivalent sense where philosophy might reflect on film (captured in the more standard philosophy of film) as well as, at least as a nonstandard possibility, where film might reflect on philosophy. For more on this, see John Mullarkey, Refractions of Reality: Philosophy and the Moving Image (London: Palgrave Macmillan, 2009). 3  Murray Smith, “Film Theory meets Analytic Philosophy; or, Film Studies and L’Affaire Sokal,” Institute for Cognitive Studies in Film and Video Electronic Newsletter – Special Edition, vol. 3 no. 1 (November 1997): 4. 4  Ilham Dilman, “Can Philosophy Speak about Life?” in The Impulse to Philosophise, ed. A. Philips Griffifths (Cambridge: Cambridge University Press, 1992), 118. 5  Brenda Almond, “Philosophy and the Cult of Irrationalism.” in The Impulse to Philosophise, ed. A. Philips Griffifths (Cambridge: Cambridge University Press, 1992), 201, 215.



what philosophy is. […] It is as impossible to say what philosophy is not as it is to say what it is. In all the other disciplines…there is philosophy.”6 Differences rarely come starker than that. Yet, one can also find the Analytic philosopher Stuart Brown partially agreeing with Derrida, taking the view that philosophy has no essence but, as a term, covers a range of diverse activities with only a family resemblance and nothing more substantial than that.7 One difference undone, then? How many more might there be to come? In what follows, we attempt to track down a good few more of various types, some residing in the history of philosophy, others in the practice of philosophical analysis, in attitudes to science and scientism, history and historicism, and finally in methodological concerns with argumentative rigor, and stylistic concerns with clarity of expression. When we subsequently look at how these variables manifest themselves within the field of film philosophy, our ecumenism will be given even further motivation: though not all of the oppositions will necessarily collapse, a number will certainly wobble, if not fall, under the strain of closer examination. And this is especially true of the most recent practitioners of film philosophy: the twain do indeed meet, but it is through their actual practice of film and philosophy, that it happens. For some, like Simon Glendinning, what lies behind the term “Continental philosophy” is in fact an Analytic (or Anglo-American) fantasy, a fabrication with no substance at all: efforts to find an internal unity to the Continental collection [of philosophies] will always either underpredict or overpredict because the only perfect predictor is one that acknowledges that the set comprises the distinctive ‘not-part’ part of analytic philosophy: it is a unity of exclusion, not a unity of inclusion.8

“Philosophically” speaking Glendinning is correct: there is no such thing as “Continental philosophy” per se, for what counts as “Continental” has always been flexible, changing, and strategically constructed to oppose its (equally mutating) Anglo-American “Other.” Moreover, the very names given to this dyad, “Anglo-American and Continental,” or “Analytic and Continental,” are either a sham geo-cultural distinction or a category error, as has long been recognized.9 The title “Continental Philosophy” is simply a name for the Anglophone reception of European ideas, which has mostly filtered out the parts 6  Jacques Derrida, “Deconstruction and the Other,” Contemporary Approaches to Philosophy, ed. Paul Moser and Dwayne Mulder (New York: Macmillan, 1994), 373. 7  Stuart Brown, “On Why Philosophers Redefine their Subject,” in The Impulse to Philosophise, ed. A. Philips Griffifths (Cambridge: Cambridge University Press, 1992), 55, 57. 8  Simon Glendinning, The Idea of Continental Philosophy: A Philosophical Chronicle (Edinburgh: Edinburgh University Press, 2006), 116. 9  Three of the pillars of so-called Anglo-American thought, Frege, the Vienna Circle, and the Polish school of Logic, all flourished in Europe, whereas, conversely, there are numerous “Continental departments” in North America, not to mention one or two practitioners in Britain. And things become more confused when one realizes that Finland is strongly Anglo-American, and that Analytic philosophy can be found on various other continents. So much for maps.



it finds indigestible (neo-Thomism, French spiritualism, French epistemology of science, or lebensphilosophie, for instance, being typically neglected outside of France and Germany). Fixing the extra determination of “post-Kantian” as an adequate descriptor of all philosophy on the Continent since 1804 only worsens this illusory metonym, for, though a good deal has happened in French and German philosophy after Kant, that does not mean it was because of Kant: “post Kant ergo propter Kant” is as fallacious as it appears.10 In addition, there are few philosophical themes that are exclusive to the European continent, nor any outside the continent that are confined to “Anglo-­ American” philosophy. The same can be said vis-à-vis method: pace the ill-­ coined name “Analytic philosophy,” we shall see that there are actually few methodological barriers that abide between the two traditions either (and especially as practiced in contemporary Analytic and Continental film philosophy). In fact, it is extremely difficult to make any distinction stand up to historical, methodological, or philosophical scrutiny (though we investigate each of these, nonetheless). The presence or absence of a host of properties, philosophical and non-philosophical, has been cited as a decisive causal factor in the origin and maintenance of the current partition in Western philosophy, some of them rather obvious, some less so. Among the more obvious we have national character, political history, geographical proximity, institutional procedure, the language barrier, methodology, or different philosophical interests; among the more subtle candidates are a difference in style or mood, different philosophical lineages (most often spawned by Frege and Husserl, one of which we return to later), the supposed fact that Analytic thought is uniquely objectivist, individualist, and attentive to science (or “scientistic”), or the supposed fact that Continental thought is uniquely subjectivist, collectivist, and attentive to history (or “historicist”). Yet, as many commentators have come to see, this segregation neither fully succeeds nor fully fails to map clearly onto any geographical, historical, methodological, or philosophical difference.11 If these factors are pertinent at all, then they are at best tendencies, directions more or less followed by both rather than substantive divides. Nonetheless, were one to accept Glendinning’s premise of Continental philosophy’s insubstantial content, such relational or differential meaning does have significance for many in the Anglophone world (both Analytic and “Continental”) all the same, even if only to mark, as previously mentioned, how philosophers view themselves and the consequences of this view: it has less to do 10  For more on this hyperbolic use of Kant, see John Ó Maoilearca and Anthony Morgan, “Transcendental Authority: A Conversation with John Ó Maoilearca” in The Kantian Catastrophe? Conversations on Finitude and the Limits of Philosophy, ed. Anthony Morgan (Newcastle: Bigg Books, 2017), 223–35. 11  See Chase and Reynolds for two fairly recent attempts at a more ecumenical approach to themes in Analytic and Continental thought. James Chase and Jack Reynolds, Analytic versus Continental: Arguments on the Methods and Value of Philosophy (Durham: Acumen, 2011); Postanalytic and Metacontinental: Crossing Philosophical Divides, ed. Jack Reynolds, James Chase, Edwin Mares, and James Williams (London: Continuum, 2010).



with what they think about when they philosophize, than where they ­philosophize, with whom they philosophize, and what they say about their philosophical practice to each other. There may, in fact, be many, perhaps a majority, on both sides who want to believe that their camp is on the right side of history, but such a lack of mutual respect, if it persists, stems from earlier generations of philosophers (the attitudes of Anscombe and Ryle, or Heidegger and Derrida, being exemplary).12 Richard Rorty—himself something of an ecumenist on these matters—put it as follows: …students of analytic philosophy were encouraged to keep their reading in literature well clear of their philosophical work and to avoid reading German philosophy between Immanuel Kant and Gottlob Frege. It was believed that reading Hegel rotted the brain. (Reading Friedrich Nietzsche and Heidegger was thought to have even worse effects – doing so might cause hair to sprout in unwanted places, turning one into a snarling fascist beast.)13

Yet, following Glendinning’s relational stance, it is certainly very noticeable that the perceptions of what counts as non-Analytic thought have been ever changing. By the 1950s, for example, the work of Kant had entered into the Analytic fold and away from its previous position (according to some in the Vienna Circle) at the cusp of the philosophical rift; while Hegel, once the paradigm of nonsense for many in the Frege-Russell-Carnap school, has, since the 1990s, become a figure with whom Analytic thought can do business (in the work of Robert Brandom and John McDowell, for instance).14 And perceptions evolve. In coming to the work of Paisley Livingston, Thomas Wartenberg, and Noël Carroll, say, or (for the “other” side), Robert Sinnerbrink, Catherine Wheatley, and Patricia Pisters, therefore, we are no longer dealing with the attitudes and “optics” of the older generation of Franco-German “masters,” be it Carnap and Frege or Lacan and Deleuze, whose no-go areas were immutable: these are practicing film philosophers arguing about the philosophy of film in extremely compatible ways (whether they know it or not)—as I endeavor to explain. Leaving aside for now what content might distinguish the Analytical from Continental, we can still ask why they became distinct, that is, we can finally turn to the question of historical origins. In doing so, however, we should take care to avoid anachronism: too may zealots of the Analytic and Continental divide have dabbled in spurious historical lineages, with false, retrospective images of continuity created in order to harden the divide, both synchronically and diachronically. As another ecumenist, David Cooper, writes: “Hume [a favourite forbear for some Analytic thinkers] needs to be translated, perhaps  See Eric K. Tranoy, Philosophy Today, Volume VIII (1964): 156–7.  Richard Rorty, “Deconstruction and Circumvention,” Critical Inquiry 11 (1984): 21. 14  For more on this, see Paul Redding, Analytic Philosophy and the Return of Hegelian Thought (Cambridge: Cambridge University Press, 2007). It was Peter Strawson who helped to rehabilitate the image of Kant for Analytic philosophy in the 1950s. 12 13



inaccurately, into a more ‘formal’ idiom to count as an Analytical philosopher on the most favoured characterisations.”15 Indeed, on the question of personal identity, for instance, one can compare Hume as much with Derrida and Deleuze as one can with Russell (in fact, Deleuze’s first book was a study of Hume).16 So let us look a little harder at this history, all the time remembering that it may well be even more significant to intertwine this history with the self-­image of those philosophers who embody it.

A Long and Winding History For certain philosophers, the Continental-Analytic divide has a very simple historical origin: “It took Hitler to drive analytic philosophy out of Europe.”17 The ensuing geographical and institutional divides (which academics ended up in UK and US universities, which remained in Europe) may thereby have some part in constituting the identity of these traditions: departmental solidarity, group cohesiveness, or just mental inertia. Analytic philosophy, for instance, would simply be the philosophy which holds sway among philosophers teaching in Anglophone academies. It would be, as A.J. Mandt puts it, “a community of discourse” rather than an ideology.18 What is deemed non-genuine philosophy would simply be what falls outside a specific community of discourse. That community is historically produced through a range of variables, philosophical (shared convictions, texts, arguments, and analyses), and non-­ philosophical (overt political strategies/intrigues, shared teachers, journals, schools, buildings, seminar rooms, and so on). One might still argue, therefore, that beyond such historically produced divisive effects, there might remain between the traditions underlying arguments and positions that are shared. Alas, such a real continuity of philosophy would be difficult to entertain, according to Mandt, for few individuals (on either side) can have thought through every aspect of his or her historical inheritance: some, if not most of it, must be taken as given.19 One Analytic philosopher who did attempt to unearth such a conceptual archaeology was the renowned Analytic philosopher of language, Michael Dummett. In his account, it is Frege and Husserl who stand at the source of the conceptual divide between Continental and Analytic thought in the twen15   David E.  Cooper, “The Presidential Address: Analytical and Continental Philosophy,” Proceedings of the Aristotelian Society, vol. 94 (1994): 1; and he continues: “Nietzsche and Kierkegaard had to await resurrection through the efforts of Jaspers and others before entering the mainstream of European philosophy.” 16  See Gilles Deleuze, Empiricism and Subjectivity: An Essay on Hume’s Theory of Nature, trans. Constantin V. Boundas (New York: Columbia University Press, 1991). 17  W.D. Hart, “Clarity,” in The Analytic Tradition: Meaning, Thought and Knowledge, ed. David Bell and Neil Cooper (Oxford: Blackwell, 1990), 198. 18   A.J.  Mandt, “The Inevitability of Pluralism: Philosophical Practice and Philosophical Excellence,” in The Institution of Philosophy, ed. A. Cohen and B. Desai (Chicago: Open Court, 1989), 83. 19  Ibid., 87–9, 95, 99.



tieth century. According to Dummett, both sought to renew philosophy through foundationalist means, but each took that quest in a contrasting direction. In search of strong objectivity, Frege looked to linguistic sense, Sinn: he took the dispositional (or “mention”) view of sense to infer a Platonic third realm in which it resides. Conversely, Husserl utilized “meaning” in a more general manner beyond language alone: this meaning was later re-dubbed “noema,” something akin to a mental act that informs not only language but also any conscious, intentional act. This phenomenological intentionality is not psychological in the sense of being voluntary or an independent act separate from physical reality: the “meaning constitutive act” is composite—part mental and part physical. This was not sufficient for Frege, however: wanting the strongest objectivity possible, he placed Sinn or “Thought” (the sense of a sentence) in this Platonic, objective realm. For Frege, Thoughts are not in the mind: while being the contents of psychological acts of thinking, they themselves are not psychological. Thinking is a mental act, but a Thought, unlike a mental image or a sensation, is not a mental object. Hence, starting with the Fregean stance, an objective science of thought is possible. And this stance lies at the origin of Analytical philosophy, with all its concerns for objectivity, scientificity, logical analysis, and, of course, language (both formalized and ordinary). The road travelled from this origin has been long and winding (after the 1950s, few followed either the Platonism or logical atomism of the early Fregeans like Russell, for instance) but its abiding concerns remain. While Frege deposited Thoughts in a third realm beyond mind and matter, after the 1950s Analytic philosophy largely ignored that move and instead radicalized the connection between communication and thought: instead of language merely transmitting thought, it actually generated it.20 Not benefitting from Quine’s much later use/mention distinction, Frege took the dispositional (mention) view of sense to infer a third realm in which it resides, whereas an occurant (use) view would have led him to see sense as the grasp of a communal language.21 And this communal use leads us on, not to further formal, logical analysis, but to a more ordinary, lived approach. Hence, we find Wittgenstein, for example, saying, in a post-Tractarian mode, that “the investigation of grammar is fundamental in the same sense in which we may call language fundamental – say its own foundation.”22 Dummett’s account, though not without its critics, offers us a useful conduit to film philosophy.23 Though other histories are available (circling around e­ xistential  Michael Dummett, Origins of Analytic Philosophy (London: Duckworth, 1993), 26.  Ibid., 13, 109. 22  Ludwig Wittgenstein, “Philosophy,” Contemporary Approaches to Philosophy, ed. Paul Moser and Dwayne Mulder (New York: Macmillan, 1994), 128. 23  See Chase and Reynolds, Analytic versus Continental, 2: “We can add to Dummett’s claims about origins. Early Twentieth-century analytic philosophy was also influenced by Polish logic, mathematics and associated philosophical work…and by ongoing developments in logic and the philosophy of mathematics in the German-speaking world (an enormous group, including among 20 21



consolation, clarity, or even time itself ),24 Dummett’s version builds on an original difference in attitude to perception, one that could be taken as definitive, not only for the subsequent development of Western philosophy but also for the seemingly divergent attitudes taken by different traditions of film philosophy, as we shall soon see. Husserl claims that essences are grasped through intuited meaning; Frege claims that Thoughts are examined through an analysis of language: herein lies the future divergence of Continental philosophy (dominated by phenomenology) and Analytical philosophy (dominated by linguistic analysis and, subsequently, cognitivism), some say.25 Yet this bifurcation may also create a dilemma for the philosopher of film, for is she/he to take the object of cinema as essentially linguistic or as something that goes beyond language, making it more suitable to a Husserlian, phenomenological approach? Fregean Thoughts are objective on account of being communicable, unlike sensations or mental images. But are films linguistic or sensory, are they—at least for any potential philosophical purposes—conceptual and so linguistic, or sensational in a non-linguistic fashion? If the maxim attributed to Stanley Kubrick was correct, namely that “if it can be thought or written, it can be filmed,” then it would seem that the post-Fregean approach might be the wisest one to adopt.

Language, Analysis, and Description: On Various Methods The significance of language for Analytically oriented film philosophers is, of course, obvious. Here is Paisley Livingston as a case in point when tackling the Continental film philosophy of Daniel Frampton: …it is not simply false or incoherent to write, as Daniel Frampton writes, that ‘film possibly contains a whole new system of thought, a new episteme’. It is also coherent to say that possibly film does not contain any such new system or episteme. As such a new, ineffable system of thought cannot, in principle, be articulated verbally; what is ruled out is the possibility of having or providing any linguistically mediated grounds for believing in the existence of a particular case in which wisdom of this sort has been acquired.26 others David Hilbert, Kurt Gödel and Tarski again); by the mid-twentieth century Scandinavian analytic philosophers were also influential (consider Eino Kalla, Georg Henrik von Wright, and Jaakko Hintikka). In the other direction, much early-twentieth-century anglophone [sic] philosophy in the United States and United Kingdom fell outside the analytic movement, was openly hostile to it, and indeed was straightforwardly Kantian or Hegelian in inspiration. The analytic dominance in the United Kingdom and the United States after the Second World War is far too easily read back into the earlier twentieth-century philosophical history of both countries, filtering what we notice in the journals, appointments and monographs of the time.” 24  See Tranoy, “Contemporary Philosophy,” and John McCumber, Time and Philosophy: A History of Continental Thought (Durham: Acumen, 2011). 25  Harold Durfee, Analytic Philosophy and Phenomenology (The Hague: Martinus Nijhoff, 1976), 41. Naturally, Durfee’s point predates the cognitivist paradigm shift that really begins in the 1980s. 26  Paisley Livingston, Cinema, Philosophy, Bergman: On Film as Philosophy (Oxford: Oxford University Press, 2009), 23–4.



Thomas Wartenberg similarly writes that “philosophical arguments rarely involve empirical evidence that can be presented without verbal language” (even as he confesses to the immediacy that the visual element of film offers over its textual dimension).27 Gregory Currie’s Image and Mind also tussles with the word/image duality of film: while putting great emphasis on language for his evidence—discussing tense, meaning, and the languages of art—Currie pushes hard against seeing the combination of images in film being like that in language—contra the semiotic theories of Christian Metz. Yet it is precisely on account of such balanced approaches that we have a dilemma concerning the status, as analytic, of such film philosophers. For Dummett, Analyticism is that tradition of philosophy which holds that only a philosophical account of language leads to a philosophical account of thought. Sentence structure and thought structure hang together. This is Frege’s legacy (with or without the Platonism).28 So, if Analytical film philosophy is not so linguo-centric, must we conclude that Dummett is wrong, or is it that so-called Analytical film philosophers like Wartenberg and Currie have been miscategorized? As for the other, phenomenological side, the line of influence stemming from Husserl down along the “Continental” track of film philosophy is comparatively clearer. Aside from the strong phenomenological inheritance in film theory (such as Andre Bazin’s democracy of vision, which leans heavily on Merleau-Ponty’s philosophy of ambiguity), the use of phenomenology in film philosophy was much in evidence as early as Stanley Cavell’s The World Viewed, with its own Heideggerian inheritance of Weltanschauung (or “world picture”) as well as its unconcealed ontological ambition to discover the essence of the cinematic image.29 A good deal of the wider phenomenological inheritance, however, formed around Merleau-Ponty’s work in particular, given its emphasis upon describing the bodily, lived experience (of perception) and its priority over our language-based intellect. In 1945, Merleau-Ponty himself had said that “if philosophy is in harmony with the cinema…it is because the philosopher and the moviemaker share a certain way of being, a certain view of the world.”30 This emphasis has only been deepened by 25  years of subsequent phenomenological analysis and description, from Vivian Sobchack’s The Address of the Eye (speaking of how “our embodiments differ and our situations change, so the film’s activity of sign production and its meaning change for us”) to Laura U. Marks (on the “skin of film”), Mauro Carbone (the “flesh of images”), and, most recently, Jenny Chamarette (stressing how “phenomenologies of film inform phenomenological accounts of film, which are co-informed

 Thomas E. Wartenburg, Thinking on Screen: Film as Philosophy (Oxford: Routledge, 2007), 80.  Dummett, Origins of Analytic Philosophy, 4, 6–7, 10. 29  See Stanley Cavell, The World Viewed: Reflections on the Ontology of Film, second edition, (Cambridge, Mass.: Harvard University Press, 1979). 30  Maurice Merleau-Ponty, “The Film and the New Psychology,” in Maurice Merleau-Ponty, Sense and Nonsense, tr. Hubert L. Dreyfus and Patricia Allen Dreyfus (Northwestern University Press, 1964), 59. 27




by encounters with film phenomena”).31 The latest phenomenological takes on film philosophy also incorporate descriptions of the “film world” (a notion going back to Cavell’s world views, but now updated), be it understood as “at once cognitive and immersive and ‘sensuous’” (in Dan Yacavone’s Film Worlds) or hermeneutically, as shown by Alberto Baracco’s Hermenenutics of The Film World, which demonstrates, via Paul Ricouer, how the “filmgoer’s perception and interpretation leads to the construction of a film world”.32 This prevalence of phenomenology in Continental approaches to film leads us to a further question of method, that of description versus analysis. The French Analytic philosopher Pascal Engel (one of a fairly rare species, as one might imagine) lists four principles by which the Analytic and Continental can be seen to separate: for the former, philosophy is a communal exercise (albeit avowedly contrarian); hence, it can progress if clarity and argument are pursued; nobody is a genius. For the latter, philosophy is a solitary pursuit; as there is no truth or objectivity, it cannot progress (making it closer to literature); anybody of worth is a genius.33 Most importantly, though, Analytic philosophers are said to follow formal analysis, while the Continentals pursue phenomenological description. But is this true? It can be argued that not all Analytics (or analyses) are equally “Analytical.” Of course, at least in its beginnings, Analytic philosophy claimed to be the logical analysis of language. To quote Joseph Owens: “Just as chemistry understands and explains molecules by dividing them into atoms of which they are composed, so logical atomism analyzes philosophical objects by breaking them up into their ultimate logical components.”34 G.E. Moore, for example, was famous for always asking “what exactly do you mean by that?”35 The call for exactitude brings discreteness with it: to analyze is to break up and thereby to clarify. According to Marjorie Grene, behind this view are four fundamentally empiricist assumptions: of data, of their logical separability, of Hume’s associationism (by which data are connected), and of a Hobbesian materialism (whence they came).36 And here, indeed, we have at least one Analytic film philosopher (Noël Carroll) being true to form in apparently 31  Vivian Sobchack, The Address of the Eye (Princeton: Princeton University, 1992), 305; Jenny Chamarette, Phenomenology and the Future of Film (Basingstoke: Palgrave Macmillan, 2013), 48. I could also here mention the work of Jennifer Barker, Kate Ince, Katharina Lindner, and Davina Quinlivan. 32  Daniel Yacavone, Film Worlds: A Philosophical Aesthetics of Cinema (New York: Columbia University Press, 2015), xiv; Alberto Barraco, Hermeneutics of The Film World: A Ricœurian Method for Film Interpretation (London: Palgrave Macmillan, 2017), 146, emphasis mine. 33   Pascal Engel, “Continental Insularity: Contemporary French Analytical Philosophy,” Philosophy, Volume XXI (1987):3–4. We will return below to the themes of progress through argument and clarity. 34  Joseph Owens, “Analytic and Continental Philosophy in Overall Perspective,” Modern Schoolman, Volume LXX (1993): 137. 35  See Marjorie Grene, Philosophy In and Out of Europe (Berkeley: University of California Press, 1976), 17. 36  Ibid.



adopting this very stance: “Parts and relations, then, are basic ingredients of film form. When we make statements about the form of a film, we are speaking of relations between parts of the film.”37 So far, so analyzed. Yet not all Anglo-American philosophy might seem so “analytic” as its forebears—as Owens also reminds us: Subsequently, as analytical philosophy passed through the stage of logical positivism and the phases of ‘ordinary language’ and elucidation and therapy, it rejected the older logical atomism as in fact a metaphysics. Yet it retained the atomistic tendency towards ‘breaking up’ objects into their component elements.38

To understand analysis through atomism is to misconstrue the purpose of analysis. For instance, while rejecting the old logical atomism, even the likes of the later Wittgenstein can be found arguing that it is no use rummaging through problems unsystematically as one looks through a drawer: one must “examine one thing after another methodically.”39 And so we can have Thomas Wartenburg following in that same spirit of systematic analysis: While I disagree with Peirce’s attempt to model philosophy on the natural sciences, I do think that he made an important point about the conduct of philosophical inquiry. In place of the single chain, he suggested a series of interlocking strands, none of which might be able to bear the full weight of a heavy object, for the interlocking strands would distribute the load, thereby keeping the object supported even when one of the strands was weak or, even, broken.40

Neither a single chain nor an irreducible whole—but not a collection of atoms either: instead, we’re offered a set of interlocking strands. Yet, of course, evidence for systematic method can be discerned just as well within phenomenological description as any other approach.41 What might remain a crucial difference, however (and one that brings us back to the Frege-Husserl ­divergence in fact), is that where phenomenology pursues a rigorous analysis and description of subjective aspects of film experience (appearances or “phenomena”), Analytic film philosophy is after something more objective, grounded not solely in language but also in science.  Noël Carroll, Engaging the Moving Image (New Haven: Yale University Press, 2003), 137.  Durfee, Analytic Philosophy and Phenomenology, 5; Owens, “Analytic and Continental Philosophy in Overall Perspective,” 137. 39  Wittgenstein, “Philosophy,” 138–9. 40  Wartenberg, Thinking on Screen, 133. 41  We need think only of Cavell’s descriptions in The World Viewed of color in various films, say, for an example of this: see Stanley Cavell, The World Viewed, Expanded Edition (Cambridge, Mass.: Harvard University Press, 1979), 80–101. Cavell, of course, can be seen (like Rorty) as something of an overt bridging figure between Continental and Analytic tropes on a number of fronts, and as such, his film philosophy can be seen to prefigure what I am arguing has become an unacknowledged commonality of thematic interest among more recent generations of film philosophers. 37 38



Of Science and Scientism The centrality of science actually raises problems for a certain view (such as Dummett’s) of where Analytic philosophical interests best lie. As Pascal Engel writes: there is the growing propensity of many (mostly American) philosophers to think that philosophy is a sort of subpart of science, either because they conceive philosophy as ‘naturalized epistemology’ in Quine’s sense, or because they see it as a part of cognitive science. There is a tension between the ‘basic tenet’ [of the primacy of language of analytic philosophy] and this second trend because according to the latter the philosophy of language becomes dependent upon the philosophy of mind, and maybe the science of mind.42

Be that as it may, Engel goes on to say that naturalism, seen as the position that all human nature and behavior is a part of nature as science explains it, has been a constant temptation of the Anglophone tradition.43 Russell, for instance, believed in two things: logic and physics—and we are bound to believe in modern physics, he said, “on pain of death”: “Science is at no moment quite right, but it is seldom quite wrong, and has, as a rule, a better chance of being right than the theories of the unscientific. It is, therefore, rational to accept it hypothetically.”44 In Analytic film philosophy, that commitment to the value of science can appear in various forms. Some of them are methodological, as when Carroll portrays the “scientific method as a useful guide to the sort of rational enquiry that film theorists pursue,” or when Wartenberg insists that [t]hought experiments themselves play another important role in natural science. Some scientific theories are themselves elaborate thought experiments in which an idealized model of the real world is created. It is argued that the idealization should be taken as an approximation to the real world that only adds more complexity to the idealized model.45

Others are more substantive in their approach, taking film experience to be best approached as a part of cognitive science, a view that (probably most famously) David Bordwell has argued for since the late 1980s alongside others including Torben Grodal, Paisley Livingston, Joseph Anderson, Carl Plantinga,

 Engel, “Continental Insularity,” 6.  Pascal Engel, “Interpretation Without Hermeneutics: A Plea Against Ecumenism,” Topoi (1991): 138. 44  Bertrand Russell, My Philosophical Development (Nottingham: Spokesman, 2007), 17, quoted in Grene, Philosophy In and Out of Europe, 14–15. 45  Noël Carroll, Theorizing the Moving Image (Cambridge: Cambridge University, 1996), 322; Wartenberg, Thinking on Screen, 64. 42 43



and Murray Smith.46 Currie’s Image and Mind was one of the first substantial philosophical interventions in this mode, arguing that “cognitive science combines rigorous and clear argument with a commitment to the most demanding standards of testability we can devise,” while Smith’s recent work, Film, Art, and the Third Culture (subtitled A Naturalized Aesthetics of Film), indicates that this research “paradigm” remains in good health such that the “mirror” stage now has nothing to do with Lacan’s babies and everything to do with our brain’s neurons.47 By contrast, some traditional Continental philosophy stricto sensu, as seen in Heidegger’s work, for instance, infamously discounted science as a shoddy picture of the human condition: science busies itself essentially with “theorising the regulation of the possible planning and arrangement of human labor.”48 Science is about the manipulation of human reality. So, when Heidegger looks at death, for example, it is with little regard for biology or psychology; it is an existential analysis first and foremost.49 Yet Continental philosophers’ attitudes toward science actually range across a variety of positions, from the belief that it has a genuine but limited role (Bergson, Bachelard, Canguilhem), through the view that its subject matter represents an impoverished or abstract aspect of reality or the “life-world” (Husserl, Merleau-Ponty, and many other naturalistic phenomenologies), to the idea that its underlying ideology is probably (if we were to spend time finding out) replete with inconsistencies and aporia (Derrida), or that it is a mostly historical construct made to discipline and ­control the human subject (Foucault).50 On this view, science would be an instrumentalist, mathematizing, positivist, derivative, or even mythological

46  See David Bordwell, “A Case for Cognitivism,” Iris, vol. 9 (1989); Torben Grodal, Moving Pictures: A New Theory of Film Genres, Feelings and Cognition (Oxford: Clarendon Press, 2000); Livingston, Cinema, Philosophy, Bergman; Joseph D.  Anderson, The Reality of Illusion: An Ecological Approach to Cognitive Film Theory (Carbondale: Southern Illinois University Press, 1998); Carl Plantinga, Moving Viewers: American Film and the Spectator’s Experience (Berkeley: University of California Press, 2009); Smith, “Film Theory meets Analytic Philosophy.” 47  Gregory Currie, Image and Mind: Film, Philosophy, and Cognitive Science (Cambridge: Cambridge University Press, 1995), xv; Murray Smith, Film, Art, and the Third Culture: A Naturalized Aesthetics of Film (Oxford: Oxford University Press, 2017). Smith thinks of his naturalism as a broad stance (a “third culture”) that can accommodate artefacts of the “manifest image” or phenomenological perspective within the “scientific” one (to use the terminology of Wilfred Sellars, which he does) without reduction: this is a “synoptic” approach. But wherever there is irreconcilability, the scientific image wins (as Sellars would also conclude). The first-person perspective is often incorrect, and the illusions and errors it manifests are best corrected by a representational theory of mind (222), as Sellars too advocated, though one adopted by Smith via a hybrid form of cognitivism (“neuroscience and evolutionary theory,” 163). 48  Martin Heidegger, “The End of Philosophy and the Task of Thinking,” in Martin Heidegger, Basic Writings, ed. D.F. Krell, second edition (London: Routledge, 1993), 434. 49  Martin Heidegger, Being and Time, trans. Joan Stambaugh (Albany: State University of New York Press, 1996), 292. 50   See John Mullarkey, “Philosophie au naturel,” Becoming Human, ed. Paul Sheehan (Connecticut/London: Praeger, 2003), 55–66.



form of reasoning, one that has constructed an essentialist idea of “nature” through reduction, repression, and simplification.51 There remains, nonetheless, that particularly French empirical and epistemological tradition which cannot be ignored (the aforementioned Bergson, Canguilhem, Bachelard, but also Simondon and Serres). And this muddies the distinction. What they offered was a scientific philosophy with a double grounding in experiment and argument, interlinked by philosophical speculation. As Bachelard put it in The New Scientific Spirit, “to the extent that hypotheses have been linked to experiment, they must be considered just as real as the experiments themselves. They are ‘realized.’”52 This dialectic of experiment and concept, each subverting the sufficiency of the other with recalcitrant data, each supplementing the other without reducing itself to that other, can also be seen in the work of Bergson and, even, his protégé, Deleuze. This is evidenced in the number of post-Deleuzian and naturalistic contributions to Continental film philosophy. Patricia Pisters in The Neuro-Image, for example, is perfectly at ease when analyzing the neurological events underlying film experience, while Felicity Colman’s Film Theory: Creating A Cinematic Grammar is part of a trend in “new materialist” thought emerging not only within film philosophy but across the humanities.53 In this respect, what David Papineau writes is highly pertinent here: “nearly everybody nowadays wants to be a ‘naturalist’ but the aspirants to the term nevertheless disagree widely on substantial questions of philosophical doctrine.”54 “Nature,” it would seem, has many natures. Whether or not Analytic film philosophers would agree that this is the correct approach to take in our field toward the sciences is doubtless arguable, but that a new generation of Continental film philosophers are using science in a positive fashion, and that this is a significant development, is undeniable.

51  See Cooper, “The Presidential Address,” 10; Jacques Bouveresse, “Why I Am So Very UnFrench,” in Philosophy in France Today, ed. Alan Montefiore (Cambridge: Cambridge University Press, 1983), 24. 52  Gaston Bachelard, The New Scientific Spirit, trans. Arthur Goldhammer (Boston: Beacon Press, 1984), 6. 53  See Patricia Pisters, The Neuro-Image: A Deleuzian Film-Philosophy of Digital Screen Culture (Palo Alto: Stanford University Press, 2012); Felicity Colman, Film Theory: Creating a Cinematic Grammar (New York: Wallflower Press, 2014). Where this materialism deems itself “new” (erroneously or not), is in its understanding of matter, which is not physicalist in the nineteenth-century sense, that is, with physics deemed the supervening science that deals with ultimately inert and atomistic quantities in calculable, determined motion. The new materialisms are “non-reductive,” taking matter as a complex, self-organizing phenomenon. Loosely following Gestalt principles, matter is deemed to form complex, irreducible wholes that are not the sum of their parts. The quantum mechanical language of “entanglement” and “superposition” is cited as contributing to this “intractive” nature (as in the work of Barad), as are slightly older vocabularies coming from the science of complexity which talks of “life” as comprising complex, material, and nonlinear dynamic systems. 54  David Papineau, Philosophical Naturalism (Oxford: Blackwell, 1993), 1.



A Short History of Historicism If a putative Analytic scientism is the would-be “Mr. Hyde” to the scientific approach of “Dr. Jekyll,” then historicism would be the evil alter ego to Continental philosophy’s abiding concern with history (both its own and in general). The charge of offering merely a “history of ideas” is often made against what Continental philosophy regards as an absolutely indispensable method. Using historical context to ground, debase, or pluralize an idea, is a constant characteristic of much Continental thought. There was an exception to this in the synchronic views of the Structuralists in the 1950s and early 1960s, yet, for the most part, and following Hegel, Continental philosophy has thought it possible to historicize everything—perception, knowledge, rationality, the body, gender, and, of course, “nature”—one of the most overdetermined concepts there is (according to Raymond Williams).55 And film experience too could be taken in this direction in various permutations of the “ways of seeing” thesis from Walter Benjamin to John Berger—what Bordwell dubs film theory’s “deeply Hegelian idea.”56 The Analytic charge against this is that the context of discovery is being confused with the context of justification: genealogical argumentation is genetic fallacy. So, while history is important, it is never a philosophical end in itself, being always either ad hominem or ad populem in essence. Rational argument (and logic) must remain immune to history if we are to acquire knowledge at all.57 Hence, as Joseph Margolis put it, “the sense of history within the Analytic tradition is remarkably thin”—and proudly so!58 Admittedly, to the philosopher enthralled by the resources of scientific analysis, the older approaches and ideas of philosophy must look extraordinarily crude.59 Behind this neglect of history within Analytic philosophy (outside of an overt “history of philosophy” setting) lies the assumption, according to Stuart Brown, that philosophy’s history is a site of error and confusion and that “the true identity of the subject has at last been identified in our own century.”60 Yet we may still ask: what are the consequences, if any, of this difference for contemporary film philosophy practice? Though the construction of gender can obviously be seen as historical (and cultural), the theme of gender is tackled with equal concern for its nuances and multiplicity by both sides, be it Tina Chanter, Lucy Bolton, or Davina Quinlivan in Continental film philosophy or  See Raymond Williams, Keywords: A Vocabulary of Culture and Society (Oxford: Oxford University Press, 1976), 165ff. 56  David Bordwell, On the History of Film Style (Cambridge, Mass.: Harvard University Press, 1997), 10, 144. 57  Bouveresse, “Why I Am So Very UnFrench,” 23, 20; Grene, Philosophy In and Out of Europe, 27. 58  Joseph Margolis, “A Sense of ‘Rapprochement’ Between Analytic and Continental Philosophy,” History of Philosophy Quarterly, Volume II (1985): 224. 59  See Reedway Dasenbrock, Redrawing the Lines: Analytic Philosophy, Deconstruction, and Literary Theory (Minneapolis: University of Minnesota Press, 1989), 5. 60  Brown, “On Why Philosophers Redefine their Subject,” 45. 55



Cynthia Freeland, Angela Curran, and Carol Donelan in Analytic film philosophy.61 Where they do nevertheless differ may well be in a certain political tone, however, but this would be connected with broader meta-philosophical differences concerning politics and reason. To begin with, in the Continental style of philosophy, there is frequently a political atmosphere attaching to its self-styled “master thinkers,” from Sartre to Derrida, Deleuze to Badiou. In France, it is said, the development of a political position remains the decisive test for any new philosophy, irrespective of its ostensible theme.62 Analytic thinkers, on the other hand, regard any political reflections they may forward as strictly separate from their philosophy (unless it be under the banner of “political philosophy”): it is politics “off duty,” according to David Cooper.63 The Analytic tendency to think of philosophy as a discrete activity, a tool kit that can be taken up and put down at will, is a marked further difference. By contrast, philosophy for many Continentals is omnipresent and everywhere: as we heard Derrida report: “it is as impossible to say what philosophy is not as it is to say what it is.” As such, philosophical practice comes closer to a sense of “ideology,” an omnipresent political dimension in thought. Hence, the need for constant political and cultural critique: philosophy must be engagé, simply because every power structure and cultural edifice is, in a certain sense, already philosophical (including gender, but also what counts as the human subject or human reason). This contested notion of the self or subject should give us pause. Of course, as a Humean legacy, the question of personal identity has long been central to Anglo-American thought, but only as a metaphysical matter (e.g., with occasional ethical implications, as in the work of Derek Parfit).64 But for Continental thinkers, the “deconstructions” of the subject are far more directly historical and cultural from the outset65: the self that is decentered is not simply a metaphysical continuity, but a social norm, a perceived standard of reason and moral responsibility, a socially and technologically created subject. Any Analytic ­presumption of a single type of subject (that does or does not exist) would be deemed naïve by comparison. Remember how Brenda Almond argued that philosophy extols the virtues of honesty, openness, courage, impartiality, and respect for persons? Indeed, she also argues that the vocation or “enterprise” of philosophy is that of liberalism: the ever-expanding enfranchisement of the 61  See Lucy Bolton, Film and Female Consciousness: Irigaray, Cinema, and Thinking Women (London Palgrave Macmillan, 2011); Tina Chanter, The Picture of Abjection: Film, Fetish, and the Nature of Difference (Bloomington: Indiana University Press, 2008); Cynthia Freeland, The Naked and the Undead (Boulder, Co. Westview, 2001); Davina Quinlivan, The Place of Breath in Cinema (Edinburgh: Edinburgh University Press, 2014); Angela Curran and Carol Donelan, “Gender,” in The Routledge Companion to Film and Philosophy, ed. Paisley Livingston and Carl Plantinga (London: Routledge, 2009). 62  Vincent Descombes, Modern French Philosophy (Cambridge: Cambridge University Press, 1980), 7. 63  Cooper, “The Presidential Address,” 4. See Dummett, Origins of Analytic Philosophy, 194, on his own activities. 64  See Derek Parfit, Reasons and Persons (Oxford: Oxford University Press, 1984). 65  Cooper, “The Presidential Address,” 16–17.



marginalized into liberating (intellectual) activity. Philosophizing is a recognition of human “commonality.”66 The value of philosophy lies in what it does for “human thought and life.”67 Here we see the liberal humanism many Continental thinkers would repudiate as a form of essentialism profoundly open to historical deconstruction.

Arguing the Points: Clarity Clarified The reader might at this stage be wondering why we have pursued our discussion of science into an excursus of history and the construction of the self: what has this to do with the origins and differences between Analytic and Continental film philosophy? The reason is that such methodological concerns (science and/or history) merge into stylistic concerns because, for many, the Continental view is that style (or form) is central to thought, and that, moreover, it cannot be dissociated from “mental” content. Indeed, for Rorty, the differences between the Analytic and other forms of philosophy are “relatively unimportant – a matter of style and tradition rather than a difference of ‘method’ or of first principles.”68 Yet mere style might nonetheless remain an immovable obstacle to rapprochement between the two forms of film philosophy if rigorous argument and clarity of expression are intertwined with that style, as some have suggested.69 And, according to D.W. Hamlyn, argument is indeed (Analytic) philosophy’s “life-blood” and that it would be “nothing without that argument.” For Almond, the “single core element” of Analytic philosophy is its “use of argument.”70 Certainly, the centrality of argument to film philosophers like Thomas Wartenberg reflects not only in his style of philosophizing but also in his estimation of film’s very potential to be philosophical: “films can make arguments, provide counterexamples to philosophical claims, and put forward novel philosophical theories.”71 Murray Smith characterizes Analytic philosophy by its “explicitness, precision, and clarity in argument, in contrast to the ambiguity and obscurity that often besets writing inspired by the Continental tradition,” which is one reason why, he continues, “the questions that Continental philosophy seeks to answer are best answered by embracing the analytic modes of enquiry.”72 So, when Bordwell, for example, writes of Noël Carroll’s argumentative prowess in terms of how he “operates on a level playing field; anyone with an argument can get into the game, but then skill will be required to keep up,” his apparently starry-eyed endorsement is not unmoti-

 Almond, “Philosophy and the Cult of Irrationalism,” 214.  D.W. Hamlyn, Being a Philosopher: History of a Practice (London: Routledge, 1992), 173. 68  Rorty, Philosophy and the Mirror of Nature, 8. 69  See Engel, “Interpretation Without Hermeneutics,” 137. 70  Hamlyn, Being a Philosopher, 168–9; Almond, “Philosophy and the Cult of Irrationalism,” 215. 71  Wartenberg, Thinking on Screen, 9. 72  Smith, “Film Theory meets Analytic Philosophy,” 2, 3. 66 67



vated.73 Contrast these views with that of a Continental master like Heidegger, who famously declaimed that “all refutation in the field of essential thinking is foolish,” and a gulf of understanding about argument might become evident.74 Yet, once again, we would be wrong to make this assumption, especially about the generation of Continental film philosophers who have been practicing their discipline for the past 10–15 years, as opposed to those French “masters” who have made occasional forays into film theory (Badiou, Rancière, Agamben, and even Deleuze). Take Robert Sinnerbrink, for instance. He is both a Hegel scholar and a film philosopher who writes about film and philosophy from the standpoint of metaphilosophy and philosophical ethics. In both cases, moreover, he is obviously adamant about the value of argument: I argue for greater interactive engagement between the rationalistic style of traditional philosophy of film, and the minor, interdisciplinary tradition of what I call film-philosophy, an alternative approach that combines aesthetic receptivity to film with philosophically informed reflection. My aim is to elucidate the productive possibilities for rethinking the film-philosophy relationship that are opened up by the encounter between new philosophies of film and film-philosophy, as contrasting yet complementary ways of exploring the philosophical dimensions of moving images.75

Similarly, Catherine Wheatley, who has written on both Michael Haneke’s cinema and Stanley Cavell’s film philosophy, argues on the theme of ethics with evidence drawn from various sources (at one point citing Simon Blackburn on ethics together with Hegel in the same paragraph).76 She also draws on the moral philosophy of both Kant and Cavell in order to examine the ethics of the film-­ viewing experience in order to establish an “ethical theory of spectatorship”: I shall demonstrate that in Haneke’s works aesthetic reflexivity is conducive to the spectator’s moral reflexivity. By placing reflexive techniques within new frameworks, I argue, Haneke is able to co-opt the spectator into a uniquely moral relationship to the film. […] On an implicit level, the films prompt their ­spectators to ask: How are we complicit with the apparatus? What are the moral consequences of this? Why, upon watching Haneke’s films, do we so often feel irritated, cross, even guilty?77

 Bordwell, “Foreword,” in Noël Carroll, Theorizing the Moving Image, xi.  Heidegger, “The End of Philosophy,” 239. 75  Sinnerbrink, New Philosophies of Film (London: Continuum, 2011), 3. Sinnerbrink has also written usefully on questions of style in what he dubs as a difference between the “rationalist and romanticist styles of thought” of Analytic and Continental modes respectively; see Robert Sinnerbrink, “Questioning Style,” in The Language and Style of Film Criticism, ed. Alex Clayton and Andrew Klevan (London: Routledge, 2011), 39. 76  Catherine Wheatley, Michael Haneke’s Cinema: The Ethic of the Image (New York: Berghahn Books, 2009), 2. 77  Ibid., 5. 73 74



Above all, she argues that Haneke’s work brings into play “the Kantian conception of the ethical agent as caught between two impulses: the impulse towards rationality and responsibility on the one hand, and the impulse towards pleasurable experience and away from unpleasure on the other.”78 So far, so argumentative—or to parody the philosophical behaviorist: if it looks like an argument, swims like an argument, and quacks like an argument, then it is probably an argument. Unless, of course, one decides that it is not a good argument—lacking evidence (of the right kind), being unclear, being too “inductive,” and so on. That, of course, is every critic’s prerogative, but it does smack of the “no true Scotsman” fallacy—permitting ad hoc redefinitions to preserve one’s position in the face of counterexamples, to wit, that Continental film philosophy does not, or cannot, argue. This is the reason why we spent time looking at historicism and the subject, because what counts as a good, clear, and rational argument (or reasoning subject) may not be one universal standard, but a fluctuating, historical set: not all Analytics argue like Rudolf Carnap or Richard Montague, but neither do Continentals argue (or not argue) like Heidegger or Agamben. Now this may appear to be begging the question in favor of the kind of historicism privileged by the Continental philosopher, but a very similar relativity is also found in Analytic thought. Take the old bugbear of clarity, for instance. For Murray Smith, the Analytic mode of enquiry offers “clarity in argument, in contrast to the ambiguity and obscurity that often besets writing inspired by the Continental tradition.” Certainly, clarity is often used as a stick like this to beat the Continental approach—but, we might ask, which clarity, and whose precision, are actually in question here? The early Wittgenstein, for example, was obsessed with Fregean clarity, a clarity that meant rigor, precision, and exactness (each with their own definitions). Subsequently, of course, he came to see clarity more in terms of perspicuity—a clarity that would lead to the disappearance of many philosophical problems. Indeed, perspicuity may be inversely related to rigorous explication, if resolution by perspicuity is not understood as finding an answer, but as evading a problem. It can even be argued that Wittgenstein certainly seems to be leaving the methodology of the sciences and approaching an epistemology of “understanding,” or Verstehen, a category closer to ontological themes (emotions, being-in-the-world, etc.).79 Some might conclude, then, that the notion of clarity itself needs clarifying, though if this were to be a foundational exercise, it would obviously be a circular one.80 What we can only provisionally deduce here is that notions of “clarity” are multiple, the privileging of any one being both highly contestable, and, at best, honorific (“yay” always being better than “boo”).

 Ibid., 7.  Hart, “Clarity,” 208, 210, 214–15, 219. 80  Durfee, Analytic Philosophy and Phenomenology, 4. 78 79



Conclusion: Beyond the Sinne of the Fathers It would be evident to the reader by now that this chapter is not tackling the notorious Saussure, Lacan, Althusser, Barthes “SLAB” (so-called) group of theorists who were supposedly so influential on film’s “Grand Theory” of the 1970s and 1980s. Saussure, Lacan, Althusser, and Barthes were, of course, very important for Structuralist thinking, understood as the dominant theoretical paradigm in France through the 1960s, being eventually exported to the US and the UK a decade later. None of these figures wrote on cinema to any extent, however, and though Louis Althusser was a philosopher teaching at the Ecole Normale Supérieure in Paris for much of his professional career, his original contributions were almost entirely to Marxist theory. Slavoj Žižek has, of course, brought Lacanian metapsychology into dialogue with German Idealism in his own film philosophical writings, but we are not going to look any closer at his work here.81 Indeed, neither are we going to address any of the forays into film by that generation of Continental philosophers prevalent from the 1980s onwards, that is, those of Deleuze, Badiou, Rancière, and Agamben— and for the very same reason. It is not because their ideas are in themselves uninteresting, or that they do not flout some of the negative stereotypes generated around Continental film philosophy (but actually derived more from the SLAB theorists than these next-­generation thinkers), such that some of their work would be worth comparing with that of Analytic thinkers: it is simply because, on the one hand, if we are to compare like with like, then we should look at full-time film philosophers from both sides, Continental and Analytic, rather than at part-time incursions from any quarter.82 And, on the other hand, the figures of Deleuze, Badiou, Rancière, Agamben, or Žižek actually constitute the thematic material for much contemporary film theory with Continental leanings, so to speak. Most of this, however, does not come from full-time film philosophers but from Deleuzian or Žižekian theorists, say, making contributions to film theory alongside many other areas. The question of thematics is a significant one, nonetheless. We have looked at the origins and influences on Analytic and Continental film philosophy in terms of philosophical history, science and anti-science, historicism, methodology, and style. But what of any general thematic distinction between the Analytic and the Continental? Are there different thematic choices made by the respective traditions? Are ethics or aesthetics, for instance, the property of one and not the other? Hardly, for ethical themes are as prevalent in the work of Mary Litch, Cynthia Freeland, or Daniel Shaw as they are in that of Robert

81  See Slavoj Žižek, Enjoy Your Symptom! Jacques Lacan in Hollywood and Out (London: Routledge, 2001) for an example of this kind of work. See also Mullarkey, Refractions, 58–77, and for his “exchange” with Bordwell, Mullarkey, Refractions, 68–70. 82  This is why I have not looked at the work of Robert Pippin (see Pippin, Hollywood Westerns and American Myth: The Importance of Howard Hawks and John Ford for Political Philosophy (New Haven: Yale University Press, 2011)), for example, as an Analytic film philosopher either.



Sinnerbrink or Catherine Wheatley (as we saw).83 Conversely, Livingston, Smith, and Carroll all discuss film aesthetics in great detail, yet it is neglected by Wartenberg in Thinking On Screen (whereas aesthetic discussions loom large in the work of Daniel Frampton, John Ó Maoilearca, and William Brown).84 Besides, part of what appears to be a thematic separation may be merely a presentational issue.85 Where a Continental film philosopher might address issues titled “bodily intentionality,” the “multiplicity of the self,” the “radical contingency of the event,” or the “indeterminacy of the sign,” an Analytic film philosopher might more normally investigate areas going by the name “the mind-body problem,” “cognitive psychology,” “relativity physics,” or the “opacity of demonstratives.” Each might thereby see the other’s research interests as at best irrelevant, at worst arcane, even as, beneath the different terminologies (and style), they actually do not substantially differ. Across a long and supremely varied history, there have been myriad philosophies that have each looked so different to its peers and predecessors alike that it is conceivable that, all else being equal, there may only be two ways one can be sure of obtaining common ascent that X is a philosopher: if she writes about others who we are in agreement to call “philosopher”; or, if she writes about topics (subject matter) which we recognize as topics that our agreed philosophers have written about previously; and that, in both cases, she writes about them in a style which we recognize as a style that our agreed philosophers have used previously.86 This is an acutely reflexive or meta-philosophical criterion, such that when Pascal Engel goes so far as to identify the standards of Analytic philosophy simpliciter with “the standards of good, professional and serious philosophy,” we cannot but envy his certitude.87 We have argued instead that among many contemporary practitioners of film philosophy, there is much to 83  See Mary Litch, Philosophy Through Film (London: Routledge, 2002); Freeland, The Naked and the Undead; Daniel Shaw, Morality and the Movies: Reading Ethics Through Film (London: Continuum, 2012); Sinnerbrink, New Philosophies of Film; Wheatley, Michael Haneke’s Cinema. We should also mention the Continental work of Sarah Cooper, Lisa Downing, and Libby Saxton: see, for example, Sarah Cooper, The Soul of Film Theory (London: Palgrave Macmillan, 2013); Lisa Downing and Libby Saxton, Film and Ethics (London: Routledge, 2009). 84  See Daniel Frampton, Filmosophy (New York: Wallflower Press, 1996); John Ó Maoilearca, All Thoughts Are Equal: Laruelle and Nonhuman Philosophy (Minneapolis: University of Minnesota Press, 2015); and William Brown, Supercinema: Film-Philosophy for the Digital Age (New York: Berghahn, 2015). 85  The question of representationalism, as seen when Gregory Currie states (Image and Mind, 2) that “film is a representational medium,” might be one thematic difference between Analytic and Continental approaches (and certainly is one when considering “master” thinkers like Deleuze and Badiou, for example—see Mullarkey, Refractions, 55–7). But for full-time Continental film philosophers today, this too would overstep the mark if it were taken as a definitive contrast: many of the latter treat cinema as representation too, though not so much via the research paradigm of cognitive psychology as is true of the former. 86  I say “all else being equal,” in that I have left aside other variables such as institutional politics or racial and gender bias (unconscious or not) that would also influence this ascription. Of course, in most situations all else is not equal. 87  Engel, “Interpretation Without Hermeneutics,” 138.



share in what they practice, irrespective of the different philosophical traditions with which they identify (if they do so at all). In other words, the commonality resides in what they do, not in what they say that they do (“I am Analytic, so I do x”—“I am Continental, so I do y”).88 In our still emerging interdiscipline, its later generations need no longer live by the outworn edicts of its putative elders, especially given that these old words, those of an irreconcilable twain, often belie our new deeds.89

Bibliography Almond, Brenda. 1992. Philosophy and the Cult of Irrationalism. In The Impulse to Philosophise, ed. Philips Griffiths, 201–218. Cambridge: Cambridge University Press. Anderson, Joseph D. 1998. The Reality of Illusion: An Ecological Approach to Cognitive Film Theory. Carbondale: Southern Illinois University Press. Babich, Babette. 1994. Philosophies of Science: Mach, Duhem, Bachelard. In Twentieth-­ Century Continental Philosophy. Routledge History of Philosophy, ed. Richard Kearney, vol. 8, 144–183. London: Routledge. Bachelard, Gaston. 1984. The New Scientific Spirit. Trans. Arthur Goldhammer. Boston: Beacon Press. Baracco, Alberto. 2017. Hermenenutics of the Film World: A Ricœurian Method for Film Interpretation. London: Palgrave Macmillan. Barad, Karen. 2007. Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Durham: Duke University Press Books. Barker, Jennifer M. 2009. The Tactile Eye: Touch and the Cinematic Experience. Berkeley: University of California Press. Bell, David, and Neil Copper. 1990. The Analytic Tradition. Oxford: Blackwell. Bolton, Lucy. 2011. Film and Female Consciousness: Irigaray, Cinema, and Thinking Women. London: Palgrave Macmillan. Bordwell, David. 1989. A Case for Cognitivism. Iris 9: 11–40. ———. 1996. “Foreword” to Noël Carroll. In Theorizing the Moving Image, ix–xii. Cambridge: Cambridge University. ———. 1997. On the History of Film Style. Cambridge, MA: Harvard University Press. Bouveresse, Jacques. 1983. Why I Am So Very UnFrench. In Philosophy in France Today, ed. Alan Montefiore, 9–33. Cambridge: Cambridge University Press. Brown, Stuart. 1992. On Why Philosophers Redefine Their Subject. In The Impulse to Philosophise, ed. Philips Griffiths, 41–58. Cambridge: Cambridge University Press. Brown, William. 2015. Supercinema: Film-Philosophy for the Digital Age. New York: Berghahn.

88  We do not have the space here in which to extend this formative power of practice beyond film theory and through film practice itself, but two examples of theory being so formed after practice come in Ó Maoilearca’s All Thoughts Are Equal (which shapes its arguments around the structure of a film, Lars Von Trier’s and Jørgen Leth’s The Five Obstructions, 2003), and Brown’s Supercinema, which generates a concept of the posthuman on the basis of digital cinema. 89  I thank Laura Cull Ó Maoilearca and Catherine Wheatley for their comments on an earlier draft of this chapter: their specific corrections and more general suggestions made it immeasurably better.



Carbone, Mauro. 2016. The Flesh of Images: Merleau-Ponty Between Painting and Cinema. Albany: State University of New York Press. Carroll, Noël. 1996. Theorizing the Moving Image. Cambridge: Cambridge University Press. ———. 2003. Engaging the Moving Image. New Haven: Yale University Press. Cavell, Stanley. 1979. The World Viewed: Reflections on the Ontology of Film. Expanded edn. Cambridge, MA.: Harvard University Press. Chamarette, Jenny. 2012. Phenomenology and the Future of Film: Rethinking Subjectivity Beyond French Cinema. London: Palgrave Macmillan. Chanter, Tina. 2008. The Picture of Abjection: Film, Fetish, and the Nature of Difference. Bloomington: Indiana University Press. Chase, James, and Jack Reynolds. 2011. Analytic Versus Continental: Arguments on the Methods and Value of Philosophy. Durham: Acumen. Colman, Felicity. 2014. Film Theory: Creating a Cinematic Grammar. New  York: Wallflower Press. Cooper, David E. 1994. The Presidential Address: Analytical and Continental Philosophy. Proceedings of the Aristotelian Society 94: 1–18. Cooper, Sarah. 2013. The Soul of Film Theory. London: Palgrave Macmillan. Curran, Angela, and Carol Donelan. 2009. Gender. In The Routledge Companion to Film and Philosophy, ed. Paisley Livingston and Carl Plantinga, 152–161. London: Routledge. Currie, Gregory. 1995. Image and Mind: Film, Philosophy, and Cognitive Science. Cambridge: Cambridge University Press. Dasenbrock, Reedway. 1989. Redrawing the Lines: Analytic Philosophy, Deconstruction, and Literary Theory. Minneapolis: University of Minnesota Press. Deleuze, Gilles. 1991. Empiricism and Subjectivity: An Essay on Hume’s Theory of Nature. Trans. Constantin V. Boundas. New York: Columbia University Press. Derrida, Jacques. 1994. Deconstruction and the Other. In Contemporary Approaches to Philosophy, ed. Paul K.  Moser and Dwayne H.  Mulder, 368–382. New  York: Macmillan. Descombes, Vincent. 1980. Modern French Philosophy. Cambridge: Cambridge University Press. Dilman, Ilham. 1992. Can Philosophy Speak About Life? In The Impulse to Philosophise, ed. Philips Griffiths, 109–124. Cambridge: Cambridge University Press. Downing, Lisa, and Libby Saxton. 2009. Film and Ethics. London: Routledge. Dummett, Michael. 1993. Origins of Analytic Philosophy. London: Duckworth. Durfee, Harold. 1976. Analytic Philosophy and Phenomenology. The Hague: Martinus Nijhoff. Engel, Pascal. 1987. Continental Insularity: Contemporary French Analytical Philosophy. Philosophy XXI: 1–19. ———. 1991. Interpretation Without Hermeneutics: A Plea Against Ecumenism. Topoi 10: 137–146. Frampton, Daniel. 2006. Filmosophy. New York: Wallflower Press. Freeland, Cynthia. 2001. The Naked and the Undead: Evil and the Appeal of Horror. Boulder: Westview. Glendinning, Simon. 2006. The Idea of Continental Philosophy: A Philosophical Chronicle. Edinburgh: Edinburgh University Press. Grene, Marjorie. 1976. Philosophy In and Out of Europe. Berkeley: University of California Press.



Griffiths, Philips A., ed. 1992. The Impulse to Philosophise. Oxford: Blackwell. Grodal, Torben. 2000. Moving Pictures: A New Theory of Film Genres, Feelings and Cognition. Oxford: Clarendon Press. Hamlyn, D.W. 1992. Being a Philosopher: History of a Practice. London: Routledge. Hart, W.D. 1990. Clarity. In The Analytic Tradition: Meaning, Thought and Knowledge, ed. David Bell and Neil Cooper, 197–222. Oxford: Blackwell. Heidegger, Martin. 1993. The End of Philosophy and the Task of Thinking. In Martin Heidegger, Basic Writings, ed. D.F. Krell, 2nd ed., 431–449. London: Routledge. ———. 1996. Being and Time. Trans. Joan Stambaugh. Albany: State University of New York Press. Ince, Kate. 2016. The Body and the Screen: Female Subjectivities in Contemporary Women’s Cinema. London: Bloomsbury. Lindner, Katharina. 2017. Film Bodies: Queer Feminist Encounters with Gender and Sexuality in Cinema. London: I.B. Tauris. Litch, Mary. 2002. Philosophy Through Film. London: Routledge. Livingston, Paisley. 2009. Cinema, Philosophy, Bergman: On Film as Philosophy. Oxford: Oxford University Press. Mandt, A.J. 1989. The Inevitability of Pluralism: Philosophical Practice and Philosophical Excellence. In The Institution of Philosophy, ed. A. Cohen and B. Desai, 77–101. Chicago: Open Court. Margolis, Joseph. 1985. A Sense of ‘Rapprochement’ Between Analytic and Continental Philosophy. History of Philosophy Quarterly II: 217–231. McCumber, John. 2011. Time and Philosophy: A History of Continental Thought. Durham: Acumen. Merleau-Ponty, Maurice. 1964. The Film and the New Psychology. In Sense and Nonsense, ed. Maurice Merleau-Ponty. Trans. Hubert L. Dreyfus and Patricia Allen Dreyfus, 48–59. Evanston: Northwestern University Press. Montefiore, Alan, ed. 1983. Philosophy in France Today. Cambridge: Cambridge University Press. Moore, G.E. 1994. What Is Philosophy? In Contemporary Approaches to Philosophy, ed. Paul K. Moser and Dwayne H. Mulder, 103–124. New York: Macmillan. Moser, Paul, and Dwayne Mulder, eds. 1994. Contemporary Approaches to Philosophy. New York: Macmillan. Mullarkey, John. 2003. Philosophie au naturel. In Becoming Human, ed. Paul Sheehan, 55–66. Connecticut/London: Praeger. ———. 2009. Refractions of Reality: Philosophy and the Moving Image. London: Palgrave Macmillan. Ó Maoilearca, John. 2015. All Thoughts Are Equal: Laruelle and Nonhuman Philosophy. Minneapolis: University of Minnesota Press. Ó Maoilearca, John, and Anthony Morgan. 2017. Transcendental Authority: A Conversation with John Ó Maoilearca. In The Kantian Catastrophe? Conversations on Finitude and the Limits of Philosophy, ed. Anthony Morgan, 223–235. Newcastle: Bigg Books. Owens, Joseph. 1993. Analytic and Continental Philosophy in Overall Perspective. Modern Schoolman LXX: 131–142. Papineau, David. 1993. Philosophical Naturalism. Oxford: Blackwell. Parfit, Derek. 1984. Reasons and Persons. Oxford: Oxford University Press.



Pippin, Robert. 2011. Hollywood Westerns and American Myth: The Importance of Howard Hawks and John Ford for Political Philosophy. New Haven: Yale University Press. Pisters, Patricia. 2012. The Neuro-Image: A Deleuzian Film-Philosophy of Digital Screen Culture. Palo Alto: Stanford University Press. Plantinga, Carl. 2009. Moving Viewers: American Film and the Spectator’s Experience. Berkeley: University of California Press. Quinlivan, Davina. 2014. The Place of Breath in Cinema. Edinburgh: Edinburgh University Press. Redding, Paul. 2007. Analytic Philosophy and the Return of Hegelian Thought. Cambridge: Cambridge University Press. Reynolds, Jack, James Chase, Edwin Mares, and James Williams, eds. 2010. Postanalytic and Metacontinental: Crossing Philosophical Divides. London: Continuum. Rorty, Richard. 1980. Philosophy and the Mirror of Nature. Oxford: Blackwell. ———. 1984. Deconstruction and Circumvention. Critical Inquiry 11: 1–23. Shaw, Daniel. 2012. Morality and the Movies: Reading Ethics Through Film. London: Continuum. Sinnerbrink, Robert. 2011a. New Philosophies of Film: Thinking Images. London: Continuum. ———. 2011b. Questioning Style. In The Language and Style of Film Criticism, ed. Alex Clayton and Andrew Klevan, 38–53. London: Routledge. Smith, Murray. 1997. Film Theory Meets Analytic Philosophy; or, Film Studies and L’Affaire Sokal. Institute for Cognitive Studies in Film and Video Electronic Newsletter – Special Edition 3 (1, November): 111–117. ———. 2017. Film, Art, and the Third Culture: A Naturalized Aesthetics of Film. Oxford: Oxford University Press. Sobchack, Vivian. 1992. The Address of the Eye: A Phenomenology of Film Experience. Princeton: Princeton University Press. Tranoy, Eric K. 1964. Contemporary Philosophy – Analytic and Continental. Philosophy Today VIII: 155–168. Wartenburg, Thomas E. 2007. Thinking on Screen: Film as Philosophy. Oxford: Routledge. Westfall, Joseph. 2018. The Continental Philosophy of Film Reader. London: Bloomsbury. Wheatley, Catherine. 2009. Michael Haneke’s Cinema: The Ethic of the Image. New York: Berghan Books. Williams, Raymond. 1976. Keywords: A Vocabulary of Culture and Society. Oxford: Oxford University Press. Wittgenstein, Ludwig. 1994. Philosophy. In Contemporary Approaches to Philosophy, ed. Paul K. Moser and Dwayne H. Mulder, 125–140. New York: Macmillan. Yacavone, Daniel. 2015. Film Worlds: A Philosophical Aesthetics of Cinema. New York: University of Columbia Press. Žižek, Slavoj. 2001. Enjoy Your Symptom! Jacques Lacan in Hollywood and out. London: Routledge.


The Phenomenological Movement in Context of the Philosophy of Film and Motion Pictures Shawn Loht

Introduction This chapter surveys some of the founding principles of phenomenology insofar as they pertain to the philosophy of film and motion pictures, broadly construed. In recent years there have been no shortage of applications of specific schools of phenomenology to film (e.g. Levinasian, Merleau-Pontyian, Husserlian).1 There have also been numerous comprehensive surveys of phenomenology’s application to film in the form of book chapters, journal articles, and dedicated journal issues.2 This follows multiple earlier waves of phenomenology’s influences on film criticism, theory, and philosophy, which variously  These include Daniel Yacavone, Film Worlds: A Philosophical Aesthetics of Cinema (New York: Columbia University Press, 2014); Hunter Vaughan Where Film Meets Philosophy (New York: Columbia University Press, 2012); Jenny Chamarette, Phenomenology and the Future of Film (Basingstoke: Palgrave Macmillan, 2012); Vivian Sobchack, Carnal Thoughts: Embodiment and Moving Image Culture (Berkeley: University of California Press, 2004); Allen Casebier, Film and Phenomenology: Toward a Realist Theory of Cinematic Representation (Cambridge: Cambridge University Press, 1991). 2  Dedicated journal issues include the 2016 volume of the journal Studia Phaenomenologica and a 1990 issue of Quarterly Review of Film and Video. Other, single-author pieces include Vivian Sobchack, “Phenomenology,” in The Routledge Companion to Film and Philosophy, ed. Carl Plantinga and Paisley Livingston (Abingdon: Routledge, 2009), 435–45; Daniel Yacavone, “Film and the Phenomenology of Art: Reappraising Merleau-Ponty on Cinema as Form, Medium, and Expression,” New Literary History 47 (2016): 159–86; David Sorfa, “Phenomenology and Film,” in The Routledge Encyclopedia of Film Theory, ed. E. Branigan and W. Buckland (Abingdon: Routledge, 2014), 353–58. 1

S. Loht (*) Baton Rouge Community College, Baton Rouge, LA, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




arose in the twentieth century in the work of the Cahiers du Cinema circle, the theorists Christian Metz and Jean Mitry, and Stanley Cavell, among others. In some instances, these diverse contributions have been labeled as naive realism or stylized criticism3; in other cases, it has been recognized that phenomenology is a powerful tool uniquely qualified for this work.4 However, up to the present, most other surveys of phenomenology in the context of the philosophy of film have failed to give careful attention to the wide-lens concern of what phenomenology is, or the manner in which phenomenology’s historical guiding threads bear relevance for the philosophy of film. In the following, what I aim to do is highlight key aspects of phenomenology, considered in its own right, that are relevant for bread-and-butter issues in the philosophy of film. I shall do this by sketching the philosophical frameworks of three of the most important figures of the phenomenological movement—Edmund Husserl, Martin Heidegger, and Maurice Merleau-Ponty—and subsequently highlighting areas of theoretical import for the philosophy of film. In the course of my discussion, I wish to place some emphasis on phenomenology’s potential to address issues of film’s ontology. I begin by considering in general how phenomenology can be understood, and what relevance it holds for philosophizing about film. Phenomenology tends not to admit of a neat and tidy definition. This limitation is due to the fact that phenomenology constitutes more of a philosophical methodology than a circumscribed area of study.5 It concerns not a “what” for the philosopher to take up, but a “how.” As a result, one can usually at best provide a loose list of traits agreeably viewed as characteristic of phenomenology. It is often said that phenomenology is “the study of appearances.” Accounting for a phenomenon (i.e. that which appears) requires describing the first-person, subjective experience of that very phenomenon. Phenomenology entails describing the way things are given insofar as they appear or disclose themselves, while at the same time suspending or bracketing all preexisting theories that might otherwise prejudice judgment.6 Likewise, this approach entails describing the underlying conditions of the phenomenon or appearance to occur at all. For instance, to account for the phenomenology of space requires not just a description of one’s naive experience of space but also the conditions for space itself to be given. Phenomenology also takes up the conditions of the subjectivity underlying experience in general. Phenomenology involves articulating the relationality between the subject and thing, intending agent and intentional object.7 3   Christian Ferencz-Flatz and Julian Hanich, “Editors’ Introduction: What is Film Phenomenology?” Studia Phaenomenologica XVI (2016): 12–13; Sobchack, “Phenomenology,” 435. 4  Dudley Andrew, “The Neglected Tradition of Phenomenology in Film Theory,” in Movies and Methods, Vol. 2, ed. Bill Nichols (Berkeley: University of California Press, 1985), 625–32. 5  Robert Sokolowski, Introduction to Phenomenology (Cambridge: Cambridge University Press, 2000), Ch. 13. Dermot Moran, Introduction to Phenomenology (London: Routledge, 2000), 3. 6  Moran, Introduction to Phenomenology, 11, 21. 7  Dan Zahavi, Husserl’s Phenomenology (Stanford: Stanford University Press, 2004), 13. Thomas Sheehan, “Heidegger and the Right Heideggerians: Phenomenology vs. Crypto-Metaphysics,” Kronos VI (2017): 84–85.



Among its most important commitments, phenomenology highlights the disclosive aspect of experiences, viz. that phenomena or appearances are disclosures to the dative, human agent. Phenomenology supposes an implicit unity of knower and known, such that the human agent is to be understood as fundamentally with things, enveloped in the world, rather than placed into it as an alien being. There is an inclination away from Cartesianism or Kantianism where the starting problem is to understand the conditions of mind’s penetration into world, its “transcendence.” For reasons such as these, phenomenologists tend to regard truth, language, and meaning as derivative upon subjectivity’s existential, world-borne character. A related feature of phenomenology’s emphasis on first-person subjectivity is the view that phenomenology can be equated with ontology. This is to say, the essences of things are coextensive with their description. A long-standing dictum of phenomenology that often gets parroted in this context is its claim that “consciousness is always consciousness of.” On the one hand, this statement refers to the fact that to be conscious of this or that thing means there is always some correlate to which an act of thought corresponds; in other words, there is no consciousness devoid of an object. On the other hand, crucial for phenomenology is the assumption that whatever one is conscious of, or can become conscious of, has a reality. The crux here is that what I can describe in my first-person experience, evanescent or fleeting as it may be, nonetheless has a truthful basis, as I am merely describing what I have become conscious of. Here lay a double-sided aspect of phenomenology that highlights both its appeal and its limitations. In Husserl’s major works, especially Logical Investigations and Ideas, it is proposed that human intellect possesses the power to intuit essences. The essences of things do not require scientific or conceptual analysis. Rather, essences are rock-­bottom, elementary phenomena available to the ordinary reflective mind. However, the intuition of essences does not equally claim the potential for absolute knowledge. Phenomenology operates with the stipulation that things always contain a hidden aspect, a “back” that is inaccessible to perception or intellection but still integral to the essence of the thing (given that we could not behold it any other way). Because phenomenology emphasizes the disclosive nature of appearances, viz. that appearances are given to the human agent, it posits that the essences that present themselves to intuition are limited to precisely this givenness. This is equally expressive of the embodied, existential finitude of the human agent to whom phenomena occur. One cannot step outside of oneself and realize a God’s eye view of the relation between herself and things.

Phenomenology for the Philosophy of Film? The question remains as to what use phenomenology might have for enhancing philosophical examination of film. First, insofar as film-viewing comprises a first-person, subjective experience of meaning, phenomenology is particularly suitable for addressing such experience. Indeed, film would seem to provide an exemplar case of the sort of phenomenon to which phenomenology can be



applied because films operate as dynamic instances of disclosure over and above providing static, pictorial depiction. Films comprise appearances specifically rendered to and for a human viewer. In brief, films have their essence in being viewed. As Merleau-Ponty has commented on this score, films are “gestalts,” presenting worlds in irreducible, ready-made form.8 Consequently, to analyze film-viewing phenomenologically has less to do with metaphysical issues such as the ontology or nature of “moving pictures” because these issues do not feature into first-person experience of viewing a film. The same goes for neurological concerns one might leverage from a cognitive science point of view; my first-person experience of seeing King Kong carry Jane to the top of the Empire State Building simply has no truck with the firing of certain nerve fibers in my brain.9 Phenomenology generally aims to treat its subjects in a holistic fashion. This has the advantage of emphasizing phenomena as they occur in lived experience versus isolating and analyzing singular aspects removed from original experience. Because film-viewing consists of so many experiential potencies that work in tandem (e.g. emotional affect, suspense, visual association, narrative comprehension, sound), phenomenology can help to emphasize the holism in which these occur. Films are experienced; as experienced, their ability to foster meaningful disclosure transcends specific sensory elements that one might distinguish in analysis.10 A perk of this limitation is that a phenomenological approach to the philosophy of film also serves to highlight the unity of viewer and viewed. Insofar as phenomenology asserts the presence of the viewer with things, a phenomenological perspective can highlight the existential, middle-­ voiced relation in which viewer is present to film and film discloses to viewer. The screen-viewer relationship is not one of action at a distance, in which philosophy has the burden to account for how pictures zoom across space and into the eye and brain. It is therefore not inappropriate to conclude here that the phenomenology of film is just as much a phenomenology of the viewer.11 Finally, although much more could be said by way of introduction to this subject, a phenomenological account of film stands to help us better appreciate film’s character vis-à-vis other related art forms such as photography, painting, or theater. For, just like these art forms (and others), we can observe that film-­ viewing has its own brand of intentionality, viz., its own way of being directed toward its objects and constituting meaning.12 An example would be the inter8  Maurice Merleau-Ponty, “The Film and the New Psychology,” in Sense and Non-Sense, trans. Hubert L. Dreyfus and Patricia Allen Dreyfus (Evanston: Northwestern University Press, 1964), 48–62. Also see Robert Sinnerbrink, New Philosophies of Film (London: Continuum, 2011), 38. 9  Zahavi, Husserl’s Phenomenology, 13. 10  Sinnerbrink, New Philosophies of Film, 38. 11  Ferencz-Flatz and Hanich, “What is Film Phenomenology?” 13–14. 12  Robert Sokolowski, “Husserl on First Philosophy,” in Philosophy, Phenomenology, Sciences: Essays in Commemoration of Edmund Husserl, ed. by Carlo Ierna, Hanne Jacobs, and Filip Mattens (Dordrecht: Springer, 2010), 12, 17. Zahavi, Husserl’s Phenomenology, 14.



play of presence and absence fostered by film’s use of montage to foster meaning and convey a narrative. Another example is film’s use of shot/reverse shot editing.13

Husserlian Phenomenology Of the three historical philosophers I will treat in this chapter, Edmund Husserl (1859–1938) has the least to say about film directly. Yet, the foundation his work provides for the phenomenological movement is surely the most crucial and broad to comprehend for our purposes. Moreover, central to Husserl’s program throughout his career is the nature of images, particularly as these relate to word and thought. In what follows, I dissect some general theoretical aspects of Husserl’s phenomenology that have relevance for philosophizing about film, after which I take up his writings on the perception of images, what he calls “image consciousness.” Husserl’s first major work, Logical Investigations, aims to describe the a priori conditions of “ideality,” the notion that meaning is ideal, universalizable, and communicable. In his attempt to refute psychologism, Husserl’s interest is to illustrate why cognition—particularly in its guises of truth and meaning—is not the result of a mental act. In the first volume of the Investigations, entitled Prologemena to Pure Logic, Husserl observes that if cognition were merely a psychological phenomenon, then this would undermine the possibility for scientific knowledge because it would reduce truth to a psychological principle.14 Psychology would be defining what truth is, whereas, for a science to make truth claims ultimately requires truth to be grounded outside of that very science. Husserl wants instead to describe how truth and meaning are acts that a rational agent attains through intuitions, by which the agent achieves insight into the ideal or “categorial” structures of things.15 Much of Husserl’s focus concerns consciousness and, especially, the concept of intentionality, which Husserl regards as the underlying condition of any conscious state. Intentionality for Husserl refers to any state in which one is conscious of, or directed toward, objects or states of affairs. Intentionality is not simply epistemological but, instead, encompasses any way in which one might be minded toward something. It thus includes aesthetically oriented states such as hearing music or viewing a painting, examples which each refer to a specific kind of intentionality with its own locus and content.16 Husserl does not fret over the question of whether objects or reality are mind-independent. Decisive about Husserl’s conception of intentionality is its “existence-independency.”17 Intentionality originates in conscious acts, not objects. 13  Orna Raviv, “The Cinematic Point of View: Thinking Film with Merleau-Ponty,” Studia Phaenomenologica XVI (2016): 163–83. 14  Edmund Husserl, Logical Investigations, Volume 1, trans. J.N. Findlay (Atlantic Highlands: Humanities Books, 2000), 168–74. 15  Zahavi, Husserl’s Phenomenology, 13. 16  Ibid., 14. 17  Ibid., 21



Intentionality is coextensive with meaning. “Meaning” for Husserl refers to the relationship one constitutes with the objects of one’s intentionality.18 Furthermore, meaning always also hinges on intending the matter “as” this or that. Again, this is because of intentionality’s seat in conscious acts, not in any sort of realism with which things disclose themselves.19 Meaning read off of intentionality emphasizes the holism of concrete first-person experiences over and above the sense content out of which experiences originate. The meaning of an intention is the essence of that intention,20 whereas the raw sense content upon which my intention is founded is merely lived through, or as Husserl describes in Ideas, “apperceived.”21 Finally, intentionality in Husserl’s casting has a crucial teleological aspect. Husserl distinguishes two facets of intentionality in which meaning is realized, namely, “meaning-intention” and “meaning-fulfillment.”22 Meaning-intention characterizes intentional states to which some completion, or fulfillment, is outstanding. Any example of a film with an erotetic narrative structure nicely fits the bill here, in that such films convey their story by initiating meaning-­ intentional states where meaning, or in this case, narrative resolution, is gradually filled in for the attentive viewer. In meaning-fulfillment, one’s intentional object becomes “self-evident,” viz., what one was intentionally minded toward, perhaps questioningly, is now present and confirmed. Consider as an example the discovery in Hitchcock’s Psycho (1960) that Norman Bates is the killer. There is a meaning-fulfillment during the climactic final scene in the mansion, where Norman is revealed masquerading as his mother as he attempts to murder the detective. This meaning-fulfillment completes the various meaning-­ intentions leading up to this, where small clues lead one to suspect something fishy about Norman, the goings-on in the Bates house, and the hitherto unsolved murders. A further meaning-fulfillment comes in the explanation of Norman’s insanity conveyed by the psychiatrist during the film’s final scenes; the viewer’s meaning-intention toward Norman’s motivation and psychosis is fulfilled by the psychiatrist’s discourse. For Husserl, the broader, decisive fact about meaning-intention and meaning-fulfillment is that they comprise all intentionalities or conscious states.23 As Husserl describes, “All perceiving and imagining is, on our view, a web of partial intentions, fused together in the unity of a single total intention.”24 In terms of describing the film-viewing experience, these elements of Husserl’s phenomenology of intentionality help to articulate the cognitive structures underlying the projective, additive character of watching a film and forming an understanding, or meaning-fulfillment, of what the film presents. In this light, the viewer’s achievement consists in her realization  Ibid., 23.  Ibid., 24. 20  Ibid., 25. 21  Ibid., 26. 22  Husserl, Logical Investigations, Vol. I, 250. 23  Husserl, Logical Investigations, Vol. II, 707. 24  Ibid., 701 18 19



of meaning via the intentional states afforded by cues a film presents in image, word, and sound. An important distinction is that as film viewers, we not only see the depiction of film images as, for example, those of Norman Bates, a cheap motel, and a creepy mansion, but also, through these, we realize meaning by fulfilling meaning-intentional states. For Husserl, the fact that intentional states typically behold their objects in partial or aspectual views, which can in turn be synthesized into single, comprehensive intentions of things, is of a piece with consciousness’ character of intentionally projecting beyond the merely present. Husserl says, “Only in this way can we understand how consciousness reaches out beyond what it actually experiences. It can so to say mean (emphasis mine) beyond itself, and its meaning can be fulfilled.”25 To my mind, this projective, beyond-reaching character of film-viewing is one of the most decisive aspects for describing the phenomenology of film. To view a film in its entirety provides a comprehensive, singular intentional fulfillment made up of a multiverse of smaller, synthesized meaning-intentions and meaning-fulfillments. A further feature of film illuminated by the Husserlian interpretation is the capacity of film images and sequences to convey meaning. This result is opposed to a classic argument in the philosophy of film that holds that photographs, and by extension, film and moving-image media, are incapable of representing anything, such that, for example, a photograph of a house simply comprises the image of whatever was in front of a camera when someone snapped the shutter. In this view, photographs have no meaning in their own right; it is simply that which is depicted, or the way it is depicted, which may contain some representative, meaningful content.26 From the Husserlian perspective, the deeper import lay in the fact that meaning is achieved through the object as a result of the intentional state occasioned by that object. The object for its part (e.g. the photograph, the film image), while certainly real, is simply a vehicle that carries sense-data, sense-data we see through or “apperceive,” such that we experience ideal meaning. On the whole, a suggestion we can leverage is that film’s meaning resides in the disclosive, meaning-fulfilled state fostered by viewing. It does not make phenomenological sense to characterize the meaning of films or their images without describing the holistic intentional structure they involve. To propose that films or their images possess meaning in the sense of a property, in the end, comprises simply an abstraction of a film removed from the first-­ person experience in which it has its life. Whereas the Logical Investigations provides an account of the experience of meaning and truth, Husserl’s second major work, Ideas, addresses the conditions underlying consciousness of these. The approach Husserl adopts in this text is often called “transcendental” phenomenology. A foundational notion is that of subjectivity, or the “transcendental ego,” regarded as fundamentally  Ibid.  Roger Scruton, “Photography and Representation,” Critical Inquiry 7:3 (1981): 577–603. More recently, see Robert Hopkins, “The Real Challenge to Photography (as Communicative Representational Art),” Journal of the American Philosophical Association 1:2 (2015): 329–48. 25 26



constitutive of experience. Husserl arrives at this view through what are known as the various “reductions.” The “ontological” or “phenomenological” reduction establishes a distinction between consciousness of experience versus consciousness of reality.27 Stated concretely, as a subject, I am able to discern my consciousness of my own experience as fundamentally distinct from my consciousness of the things that occasion that experience. The disclosures I encounter in my first-person experience are thus to be considered the rock-­ bottom phenomena, behind which there is not a further unknown object for philosophy to posit. This discovery provides justification to the phenomenologist to give closer attention to the subjective aspect of experiences and the contribution subjectivity makes to these: “we do not simply focus on the object exactly as it is given, we also focus on the subjective side of consciousness, thereby becoming aware of our subjective accomplishments and the intentionality that is at play in order for the object to appear as it does.”28 A counterpart notion Husserl introduces alongside the phenomenological reduction is called the “epoche”; by this notion is meant the suspension or “bracketing” of all theorization and other modes of comportment that stand to color one’s perspective. In performing the epoche, we withhold any judgment about our subject matter, in order to allow the givenness of things in intuition to inform us. In brief, in using the epoche, we aim to have experience inform our theories.29 The epoche and the reduction afford us the ability to distinguish between our everyday, unreflective experience in the natural attitude and the consciousness of our experience of disclosure in the phenomenological attitude. Whereas the default human mode of everyday experience simply takes things for given as they appear, not questioning how they come to be and so forth, by performing the reduction, we are able to consider how things come to be disclosed to us in that everyday first-person experience. Husserl suggests that one aspect of experience we are better able to appreciate by virtue of epoche and reduction is the fact that the various ways we are conscious of objects each has an essence, an essential manner of constituting its meaning. In other words, just as we can colloquially say consciousness is full of various kind of experiences, for example, wishing, willing, remembering, perceiving, judging, so too do these sorts of experiences each involve a particular mode of relating to one’s object and, indeed, of having that object present itself.30 As Husserl describes it, each type of consciousness has an essence that we can isolate through taking up the phenomenological attitude. For instance, consider the modes of first-person givenness one can readily highlight in the experience of remembering. To experience remembering a past event entails re-presenting to oneself a past perception or series of perceptions. Yet, it is not  Zahavi, Husserl’s Phenomenology, 50–51.  Ibid., 51. 29  Ibid., 45. 30  Edmund Husserl, Ideas: General Introduction to Pure Phenomenology, trans. Boyce Gibson (New York: Collier Books, 1962), 235ff. 27 28



a perception of events in the guise of seeing them right now in front of one. In remembering, we re-present the events as past, no longer present, while nonetheless still factual for us, as “having happened.” This method of describing remembering differs from a naturalistic account, where one might make an inductive claim that memory involves the burning of neural pathways in certain areas of the brain. The crux is that only the first-person account described in the phenomenological attitude can convey this mode of consciousness as it is experienced. What Husserl terms the “eidetic reduction” also speaks to this issue. The eidetic reduction is a method of paring intuitions down to their objects’ essential characteristics, their Platonic eidos. According to Husserl, one can perform the eidetic reduction by abstracting all nonessential features of the object in question, stripping the object to the features without which it would not be conceivable. Husserl describes the eidetic reduction this way: “whatever in purely immanent and reduced form is peculiar to the experience, and cannot be thought away from it, as it is in itself, and in its eidetic setting passes eo ipso into the Eidos, is separated from all Nature and physics.”31 The thrust of the eidetic reduction is that it comprises a method for thinking critically about the object one is engaged with, in first-person experience, and without recourse to extraneous theses or models. I suggest that Husserlian phenomenology offers fertile ground for considering film in these guises. For example, I suggest that film-viewing is a specific type of consciousness, that this consciousness has a unique essence, that this consciousness involves a specific way of constituting its object(s), and that this object has a unique essence. These claims stand to clarify long-standing dialogue about both what film is and whether film and motion pictures possess a set of necessary and sufficient conditions. A widely accepted view in today’s philosophy of film circles is that of the Carroll/Bordwell school, according to which film cannot be prescribed any sort of essence or aesthetic teleology.32 This view is largely a response to “grand theories” of film, in which past theorists and interpretive schools pitched sweeping theoretical apparatuses (e.g. Marxism, semiotics, psychoanalysis) as constitutive for film viewership. In Carroll’s account, film is best understood under the more inclusive category of “moving images,” where these are understood to be two-dimensional displays, detached from their subjects, containing the possibility to give the appearance of motion.33 Carroll highlights the scientifically justified concept of pictorial recognition as particularly apposite for describing the human facility to recognize the subjects of photographs and drawings as depictions of their originals.  Ibid., 240–41.  David Bordwell, “Contemporary Film Studies and the Vicissitudes of Grand Theory,” in PostTheory: Reconstructing Film Studies, ed. David Bordwell and Noël Carroll (Madison: University of Wisconsin Press, 1996), 3–36; Noël Carroll, “Prospects for Film Theory: A Personal Assessment,” in Post-Theory: Reconstructing Film Studies, ed. David Bordwell and Noël Carroll (Madison: University of Wisconsin Press, 1996), 37–68. 33  Noël Carroll, The Philosophy of Motion Pictures (Malden: Blackwell, 2008), Ch. 3. 31 32



In Carroll’s account, the film medium simply exemplifies the fact that human beings naturally see photographic and cinematographic images as images of this or that; there is not any great mystery to the phenomenon, nor any further penetration on the viewer’s part expressive of a deeper phenomenon (e.g. transparency, illusionism, the mirror).34 Vis-à-vis the Bordwellian/Carrollian position, I suggest that the phenomenological approach emphasizes describing the structures underlying the first-­ person experience, the intentionality bound up with film-viewing. But it also avoids construing films as self-standing objects that can be philosophically analyzed outside of the first-person experience they engender.

Husserl on Image Consciousness While Husserl’s extensive writings contain scant references to cinematic images or arts, a number of unpublished writings and notes address what Husserl refers to as “image consciousness.”35 These texts have been collected in Volume XXIII of the collected works (Husserliana), published under the title Phantasy, Image Consciousness, and Memory.36 It is important to acknowledge at the outset that these texts do not comprise a unified view by any means. Husserl’s position on key issues changed over the decades that this subject occupied him. Nonetheless, one can extract several illuminating insights from Husserl’s various explorations on image consciousness and its variants in service of the philosophy of film. By “image consciousness” Husserl means our capacity for seeing images. He largely treats imagination in a mold similar to Aristotle, where this faculty enables one to represent objects via images.37 This representation can take place through either physically present images, for which Husserl reserves the term “image consciousness” proper, or through “phantasy,” the capacity for seeing images conjured through one’s own imagination. Naturally, memory also figures into this account, as memory comprises the reproduction of previously experienced events in the form of images. Decisive is Husserl’s phenomenological account of how images are constituted, that is, the act of meaning that underlies image consciousness.38 For Husserl, the constitution of images consists of three perceptual moments: the physical thing, the image object, and the image subject.39 The physical image is the material foundation, the physical thing in which the image occurs. In a sculpted bust, this would be the clay or bronze out of which the bust is made; in a painting, it would be the canvas and pigment. The image object occurs in  Ibid., 108ff.  Edmund Husserl, Phantasy, Image Consciousness, and Memory (1898–1925), trans. John B. Brough (Dordrecht: Springer, 2005). 36  Ibid. 37  Aristotle, De Anima, Book III. 38  Husserl, Phantasy, Image Consciousness, and Memory, 19–20 [18]. 39  Ibid., 21 [19]. 34 35



the shapes, lines, or other visual cues that bear resemblance to a subject that is known or recognized. So a bust of a person’s head will contain shapes, lines, curves, and contours that resemble to the viewer the shape of a person’s facial features, forehead, hair, and the like. A photograph of a person will have the same effect, containing colors, contrasts of light and dark, and so forth, that convey to the viewer similarity to a person known in real life. And finally, the image subject is the matter of depiction, whatever is meant in the presentation.40 Husserl specifies that one sees the image subject in the image object; this is to say that the image subject is beheld intentionally, through the image object. The image “presents the subject but is not the subject itself.”41 Accordingly, if I view a portrait of America’s first president George Washington hanging on a museum wall, while the visual features of the portrait as image object convey to me a resemblance or likeness of George Washington, it is as the image subject that I actually intend George Washington the man. In the terminology of the Logical Investigations, the viewer’s meaning-intention toward the image subject is predicated on an intentional state of assuming the representation in the image object comprises how the subject Washington would appear if he were present.42 A crucial feature of image consciousness for Husserl is the conflictual nature bound up with viewing images. The framed picture of my wife that sits on my bookshelf coheres with its surroundings by virtue of being an inanimate object alongside other inanimate objects. Yet, this framed picture poses a conflict by virtue of its depicting power. It is not merely a physical thing, but also contains an image, particularly in a fashion that these two, while in mutual tension, do not cancel one another. I constitute the image object intentionally; it is an ideal object, different from the physical thing in which it occurs.43 Hence, there are two different types of perceptual apprehension, one of which involves something physically present and the other a perception of an image object I know not to be present.44 A second conflict lay in the fact that, as Husserl maintains, image objects do not exist; as he writes, they are constituted with “the ­characteristic of unreality.”45 Image objects are (speaking in terms of the real) actually nothing more than the material in which they occur. The black and white I perceive in an old photograph comprise the physical makeup of this photograph; the image object I intend has no actual reality. Husserl describes that whereas image objects certainly appear, fostering the appearance of the image subject in the image, image objects are an appearance of “a not now in the now.”46 As such, image objects are constituted ideally; they are not part of  Ibid., 19 [18].  Ibid., 20 [18]. 42  Peter Shum, “The Evolution of Husserl’s Concept of Imagination,” Husserl Studies 31 (2015): 216. 43  Ibid., 217. 44  Mion, “Husserl and Cinematographic Depictive Images,” 272. 45  Husserl, Phantasy, Image Consciousness, and Memory, 51 [47]. 46  Ibid., 51 [47]. 40 41



the fabric of empirical experience but, instead, are meant intentionally.47 A third conflict distinguishes the image object and the image subject. As Husserl describes it, while image object yields only one appearance, its viewing involves two apprehensions (Auffassungen), viz., taking the image in two distinct ways. Image objects initiate one apprehension of a likeness and one of the actual subject depicted.48 Husserl writes: “We have only one appearance, the appearance belonging to the image object. But we have more than the one apprehension…in which this image becomes constituted for us. If this were not the case, nothing else but the image could be meant.” Husserl elaborates: “If the conscious relation to something depicted is not given with the image, then we certainly do not have an image.”49 What should we take from these explorations? As was observed earlier in a reading of the Logical Investigations, Husserl’s account of image consciousness reveals that the central act of meaning in beholding an image lay in one’s directedness toward the image subject, which occurs “in and with” the image object.50 The image subject makes itself felt to consciousness specifically as an image representation, viz. as one thing seen through another.51 How does this sketch graft onto the image consciousness involved with cinematic representation or film-viewing? On the one hand, there are some simple correspondences. Cinematic images appear through a physical substratum, usually a screen, which may be housed in a television set, smartphone, or other device. Cinematic images involve image objects. That is, they reveal perceptual content that represents through resemblance or likeness. For instance, the figures I see onscreen exhibit the same shape and color as the human beings I know from elsewhere. And cinematic images present actual image subjects, not merely blank image objects that I do not see through. For instance, I see and intend Jodie Foster and Anthony Hopkins in the film The Silence of the Lambs (Jonathan Demme, 1991) by way of the film’s image objects that resemble these real-life people. As noted already, Husserl never provides a robust account of cinematic representation. The passage in which he most directly speaks to issues of cinematic representation reads this way: “It pertains to an image that the depictive image, understood as image object, has a ‘being’ that persists and abides. This persisting, this remaining unchanged, does not mean that the image object is unchanging; indeed, it can be a cinematographic depictive image.”52 The context of this passage concerns the ontological status of image objects insofar as they are able to remain the same throughout a duration or other change, such as a depicted object remaining persistent in the course of cinematic frames. Husserl suggests that the constitutive act is the same in parallel examples, for instance, hearing a piano piece multiple times and each time hearing the same  Shum, “The Evolution of Husserl’s Concept of Imagination,” 217.  Husserl, Phantasy, Image Consciousness, and Memory, 29 [30–31]. 49  Ibid., 32 [31]. 50  Ibid., 29 [28]. 51  Ibid. 52  Ibid., 645 [546]. 47 48



melody. The underlying condition Husserl is attempting to highlight in this discussion is that image objects contain a persistence of being, through which they can maintain their identity through changes. This persistence is constituted by the viewer ideally, which is to say, not in real empirical terms, but, instead, on the level of one’s intentionality, one’s mental directedness toward the object. As has been suggested in the literature, this passage indicates both that Husserl understands cinematic representation as a type of image consciousness following the three-moment structure outlined earlier and that Husserl understands the intentionality as directed toward ideal image objects that outwardly change while maintaining a persistent identity, a position that would not be out of place in Bazin’s observations regarding film as the preservation of objectivity.53 In this light, cinematic representation can be described in terms of the intentional ideality experienced in the image objects of cinematic photoplay.54 I will round out this discussion by highlighting some additional items in Husserl’s account of image consciousness that are especially apposite for film and phenomenology. A question that emerges out of the discussion of the world-bearing character of image consciousness concerns the sort of presence or actuality images bear, particularly if we observe that the images of film and other depictive media present self-contained worlds that seem real unto themselves.55 A crucial qualifier of Husserl’s later accounts of image consciousness56 is the attitude in which the viewer regards the image subject’s existence. In this account, image objects can be viewed “positionally” or “nonpositionally.” Husserl justifies this revision with the claim that every intentional state is positional or nonpositional,57 that is, entails a positing or nonpositing judgment regarding the existence of its subject. But this does not otherwise impact the objectivity of one’s intentional state toward what is depicted in the images. In this way, I can still be intentionally directed toward image subjects, including fictional ones, regardless of whether I believe them real or not. And my judgments about these image subjects still hold good as well, insofar as they remain under the governance of my original positing or nonpositing attitude. This distinction of Husserl’s separates one’s belief or “doxic attitude” toward the depiction from the intentions and judgments one can make about what is in the image. As a result, one can still comport oneself toward the image subjects as if they are actual within the nexus in which they appear.58 A broader result 53  Claudio Rozzoni, “Cinema Consciousness,” Studia Phaenomenologica XVI (2016): 305. Andre Bazin, “The Ontology of the Photographic Image,” in What is Cinema? Vol. I, ed. Hugh Gray (Berkeley: University of California Press, 1967), 14–15. 54  John Brough summarizes Husserl’s position on the cinematic arts. See John Brough, “Showing and Seeing: Film as Phenomenology,” in Art and Phenomenology, ed. Joseph Parry (Abingdon: Routledge, 2010), 192–93. 55  Husserl, Phantasy, Image Consciousness, and Memory, 50 [46–47]. 56  Circa 1912 and onward. 57  Ibid., 430–31 [358–59]. 58  Ibid., 486 [413], 537 [452].



for image consciousness in the case of film-viewing is that one can relate to situations, scenes, and characters just as one does to those in real life. So, for instance, although I view the image object that depicts Jason Vorhees in the Friday the 13th movies (Various, 1980–present) in a nonpositional attitude, not taking his existence to be given, I can still take any number of other attitudes regarding Jason that are logically salient, for instance, that he cannot be killed, that no one can escape him, and so on. Likewise, I can fear for the characters in the film who will fall victim to Jason’s violence, just as I would in a parallel reallife situation of people being hunted by a crazed murderer. And the same goes for my jumping in my seat as a result of a scare in which the villain spontaneously appears without announcement, ready to strike. Husserl maintains that in each case such intentional states, emotional reactions, and the like are genuine; when these instances occur, I am not making-believe or mimicking a belief or emotion but actually undergoing these intentional states.59 A thesis Husserl expresses at various points in these texts is that any perception can be intentionally experienced “as if ” it were real, even those we consider imaginary. One can perceive a situation in a private fantasy, or in the image consciousness occasioned by a film or theatrical performance, and still experience fear, joy, excitement, arousal, and so forth just as if these were genuine experiences.60

Perceptual Phantasy A final, decisive revision of these ideas occurs in Husserl’s writings circa 1917–1918. In this later work, Husserl eschews depiction in his account of image consciousness. The motivation for this revision appears to stem from a deepened appreciation of how image consciousness occurs in works of art. On the one hand, Husserl suggests that image consciousness in an aesthetic attitude can take on the form of “perceptual phantasy,” where one directly perceives entities whose outward, phantastical look is self-constituted. The image object readily presents itself in a phantastic guise, without viewer contribution. Imagination is thus “immediate.”61 An exemplar instance for Husserl’s later thesis is theatrical performance. The viewer of a stage play does not view it through representative image consciousness; the play does not depict something represented from elsewhere, as it were (though peripheral elements such as a wall painted to look like the front of a castle may be depictive). One directly perceives the image subjects, in an “as if ” attitude. For instance, I take in the Shakespeare play King Lear and perceive the lead actor “as if ” he is King Lear. Or as Husserl sometimes puts it in these late texts, I “quasi-perceive” King Lear; I do not look at the actor and hold that he “represents” or “depicts” the true King Lear. Husserl writes that in the case of theatrical performance, “we live in a world of perceptual  phantasy; we have ‘images’ within the cohesive unity of one image,  Ibid., 554 [465–66].  Ibid. 61  Shum, “The Evolution of Husserl’s Concept of Imagination,” 218. 59 60



but we do not for that reason have depictions.”62 Husserl elaborates that we can enact this comprehensive state of perceptual phantasy precisely because we can take the entire image world as null; we can perceive what we see as annulled (or in the earlier terminology, “nonpositionally”) with respect to reality. The presentative, illusory aspect of the play is temporarily concealed, while a selfconstituting productivity emerges.63 Following on this, Husserl observes that certain art media are effective in substituting performance for actual, lived reality. A real perception of a human being giving a speech, for instance, often has the same outward look as a theatrical performance experienced in the mode of perceptual phantasy, by virtue of each instance sharing a common stock of perceptual material, for example, a live human being giving a speech, employing cadence, affect, pauses, humor, and so forth. As Husserl describes, the perceptual appearances of certain things “easily change into other perceptual modes of appearance, and do so in a way that the stock of what is genuinely perceived is common.”64 In sum, for Husserl, perceptual phantasy is a mode of intentionality we activate by viewing images or performances as if they really were happening. Alternately stated, the human subject is able to annul for a time the perceptual attitude of actual experience in instances where the images or performances graft onto actual experience in one or more ways. It seems to me that film-viewing follows a similar mold—perhaps even tailor-made—of immediate imagination enabled by the annulling of reality, where one views images that have a shared, common stock with the images experienced in genuine perception. That is to say, film-viewing is predicated on immediate imagination because the images and performances of film are often comprised of the stuff of life (e.g. people, places, situations, plots), where these images are simply perceived by the viewer in an “as-if,” reality-­annulled guise. This last is especially apposite if we consider film images to be communicative images that present the perceptions of the filmmaker. Seen this way, film images can operate as perceptual phantasy by virtue of the viewer seeing the world as if they were the filmmaker, adopting the filmmaker’s attitudes or points of view. A key distinction of this later thesis is its highlighting of the relative passivity of the viewer, whose perceptual phantasy is by and large controlled by the “quasi-real” material provided in the performances or images. That is, only the later account of perceptual phantasy qua annulled reality seems to allow for my experience of immediately imagining, and thus, quasi-perceiving, for example, Clarice Starling as the main character of The Silence of the Lambs. Whereas the earlier tripartite model seems only to operate at the level of perceiving an image object who resembles the actress Jody Foster and the consequent intention of the image subject as, in fact, the actual Jody Foster. In truth, as we know, when the acting is sufficiently good and the drama sufficiently compelling, the actress disappears, and we do, in fact, see just the character. From the Husserlian  Ibid., 616 [514–15].  Ibid., 618 [516]. 64  Ibid., 619 [517–18]. 62 63



standpoint, however, the key observation to make is that these results are features of the film viewer’s intentional consciousness, that is, constitutive structures through which filmic disclosure occurs. Is it fitting to discard the earlier tripartite model of image consciousness entirely in favor of the later pure phantasy/immediate imagination view? I suggest that phenomenologically speaking, we probably want to hold on to both. I say this under the auspice that a phenomenological approach in philosophy involves suspending judgment and examining subjects from multiple angles, where different, complementary points of view may come into focus, with no single account providing the whole story.

Heideggerian Phenomenology Martin Heidegger (1889–1976) is not typically regarded as a philosopher whose work has relevance for the philosophy of film. This despite the fact that there have been numerous writings in philosophical film criticism that read Heideggerian philosophy into the narratives and themes of specific films and the bodies of work of specific filmmakers. Because Heidegger is regarded as an existentialist philosopher, his writing often serves to illuminate films that exhibit an existential bent or that grapple with life-and-death issues. The aspects of Heidegger’s philosophy that I aim to appropriate presently, however, stem from his analytic of subjectivity, or “Dasein,” in his early masterwork Being and Time. As with my account of Husserl earlier, my main interest is to highlight the theoretical aspects of Heidegger’s philosophy that stand to unravel the phenomenology of film as it pertains to film-viewing and the various structures this involves. Before doing this, however, it may be instructive to address a few of Heidegger’s central criticisms of film and picture media. Interestingly, Heidegger’s main criticism of film falls under the somewhat reductive notion that the advent of film and photography expresses modern subjectivity’s newfound capacity for representing anything and everything to itself, regardless of whether all things of the universe can and should be brought into view.65 For Heidegger, this attitude of the modern subject is fostered by the more pervasive sweep of technology. Technology expresses for Heidegger not simply an area of scientific knowledge—technology is a manifestation of being, in which all things reveal themselves as ready and available for human disposal. Film and photography are thus emblematic of this totalized availability of all things.66 For Heidegger, this state is problematic because it renders the film or photographic subject consumable and disposable. The film image cheapens what is depicted just by virtue of making everything picture-able. 65  See, for instance, Martin Heidegger, “The Thing,” in Poetry, Language, Thought, trans. Albert J. Hofstadter (New York: Harper Perennial, 2001), 163. 66  Martin Heidegger, “The Age of the World-Picture,” in Off the Beaten Track, ed. Julian Young and Kenneth Haynes (Cambridge: Cambridge University Press, 2002), 57–72; Martin Heidegger, “The Question Concerning Technology,” in Basic Writings, Second Ed., Revised and Expanded, ed. David Farrell Krell (New York: Harper, 2003), 307–41.



Unlike in the ancient world, reality no longer refuses any point of view to the human agent. The challenge posed to a reader of Heidegger is thus to consider how film can be described phenomenologically in a fashion that avoids some of Heidegger’s objections.67

Dasein With Being and Time’s concept of Dasein, Heidegger means existence or, literally, “being-there,” as it pertains to a subject. Dasein does not specifically refer to the human being, but instead, it is meant to encapsulate any being with subjectivity. (The “replicants” of the Blade Runner movies  (Ridley Scott, 1982; Denis Villeneuve, 2017) can be understood as Daseins though they are not human.)68 A defining feature of Dasein is that it is concerned about its own being; its being is an issue for it.69 Furthermore, Dasein’s “being-there” expresses the notion that Dasein is ontologically constituted by outward “projection,” or ek-sistence, thrown open potentiality.70 Hence, the phenomenology of Being and Time is often labeled “existential” phenomenology. Heidegger takes up Dasein in Being and Time as a means for engaging what he deems to be the historically forgotten questioning into “the meaning of being.” As Heidegger sees it, philosophers in the present time no longer ask after the meaning of being as such but instead take being to be given in terms of substantial presence, or, following the Cartesian turn, subjective representation. Heidegger’s motivation for focusing on Dasein in Being and Time stems from the observation that the meaning of being is an issue posed by and for Dasein. Dasein is the being who asks after being’s meaning and to whom this meaning is intelligible.71 Heidegger’s analytic of Dasein takes shape in a description of the “existentials” or categories that comprise Dasein. These existentials are the ontological features comprising Dasein’s structure.72 In Husserlian terms, Dasein’s existentials are what is left over when one performs the epoche and reduces all extraneous features of Dasein back to its essence. Among these existentials are what Heidegger calls “being-in-the-world.” This concept describes the fact that Dasein always has a surroundings, a background frame of reference in which things have meaning. Dasein is never “worldless.” The term “world” is not meant in the sense of a container or vessel that holds everything; instead, it 67  For a fuller account of film in the context of Heidegger’s views on technology and art, see Shawn Loht, Phenomenology of Film: A Heideggerian Account of the Film Experience (Lanham: Lexington Books, 2017), Ch. 3. For additional commentary regarding Heidegger’s silence on film, see Brian Price, “Heidegger and Cinema,” in European Film Theory, ed. Temenuga Trifenova (New York: Routledge, 2008), 108–121. 68  Stephen Mulhall, On Film, 3rd ed., 19–30. 69  Martin Heidegger, Being and Time, trans. Joan Stambaugh and Dennis Schmidt, Revised Ed. (Albany: SUNY Press, 2010), 41 [40–41]. 70  Ibid., 43 [43–44]. 71  Ibid., 1–7 [2–7]. 72  Ibid., 44 [44–45]



carries a notion simply of Daseins always having an abode or environment.73 According to Heidegger, the primacy of world in Dasein’s constitution is revealed in Dasein’s engagement with the tools and implements with which one performs everyday tasks.74 The use of tools to perform tasks is underwritten by Dasein’s engagement in its surroundings, or equally, its projected involvement in its world. For instance, I look for my car keys in order to drive to the supermarket, and this reveals my involvement in commerce beyond my household, my nutritional dependence on agricultural, mass-scale food production, and so forth. Heidegger highlights that world becomes most evident when the object I seek is not readily available to suit my need.75 If my car keys are lost, then the larger network of purposes and needs to which my keys afford me access reveals itself. My fundamentally world-oriented existence is revealed in this moment. Heidegger uses the discussion of Dasein’s being-in-the-world to draw out two other important world-oriented existentials, namely, Dasein’s spatial existence and Dasein’s co-existence with other Daseins. First, the phenomenology of lived experience reveals Dasein’s existence as spatial in the sense of the nearness and farness of one’s occupations, or of what is available in one’s lived surroundings; the notion is that lived experience does not present space in the guise of a Cartesian grid where objects have an objective, homogeneous distance from one another. Rather, things are near or far depending on whether one is engaged with them.76 For instance, talking to a friend via a Skype video call can bring this friend phenomenologically closer than, say, the neighbor at a house down the street who I am not thinking of at all. Heidegger labels Dasein’s capacity to bring things near “de-distancing” (ent-fernen); this concept expresses the phenomenon underlying any possibility of having “space.”77 Heidegger’s account of Dasein also describes Dasein’s existence as one of a shared world. Dasein is not to be understood as a solipsistic being. Rather, Dasein is fundamentally born into and oriented toward a world of other Daseins. Heidegger labels this existential “being-with” (Mit-sein), indicating that Dasein’s who-ness or personal identity is essentially public and shared.78 The phenomenological evidence for this view is, again, Dasein’s engagement in tasks. The tasks we perform are typically done in cooperation with other Daseins. Or if not, the tools and equipment we use are furnished by other Daseins. Our surroundings reveal the presence and activity of other Daseins as well. And our concerns overlap with those of others; we understand the other Daseins of our world to have the same or similar concerns as we do. The existence of these other Daseins is bound by self-concern in the same way ours is. Hence, we can exercise care toward others, by helping them to realize their  Ibid., 53 [53]ff.  Ibid., 67 [67]ff. 75  Ibid., 72 [72]ff. 76  Ibid., 99 [102]ff. 77  Ibid., 102 [105]. 78  Ibid., 114–17 [116–21]. 73 74



interests (which, in turn, are often bound up with ours). Perhaps most important for Heidegger here is that Dasein’s self-understanding does not self-­ generate; instead, it derives from the everyday being of these other Daseins, whom Heidegger calls “Das Man,” or the “they.”79 Alternately stated, each of us does not derive our individual personal identity, our understanding of who we are, from ourselves. Rather, we each take the dominant strain of our identity according to the levelled-down norms in which others understand themselves.80 To sum up the Heideggerian elements enumerated thus far, while none of them speaks explicitly to film, the principal point to leverage from this perspective hinges on Dasein’s existentiality. Because the existentials of being-in-the-­ world, de-distancing, and being-with have their origin in Dasein’s own essence, they are part and parcel of Dasein’s being. As underlying structures of Dasein’s ontological constitution, these are not items added on to Dasein after the fact. Nor are they characteristics that one can choose to exercise or not. For me to be a Dasein entails that I am being in the world, de-distancing, and being-with. To consider the relevance of these items for the film-viewing experience, the existentiality of viewership seems nowhere more exemplified than in the de-­ distancing aspect upon which film-viewing is predicated. For films bring distant objects close and present them to us. As classical theorists such as Bazin and Cavell have observed, the film image has a way of rendering its subject present; in this vein, their meaning was largely about the existential presence to which the film image affords one, not the actual presence of the filmed object. What Heidegger’s concept of Daseins helps to illuminate is the projective character of the viewer. In Heideggerian terms, it is the viewer, by virtue of Dasein’s de-­ distancing, projective character who traverses the “distance” to what is depicted on screen. The screened content can become existentially near because Dasein’s existential character is just this, to remove distance. If this were not the case, then no amount of transparency or telescopic power in the film camera would have any effect for Dasein.81 In brief, the close proximity to the screened image in film-viewing is an existential feature of viewership. Following up on this equation, I suggest that film-viewing is existentially rooted in the viewer’s being-in-the-world. That is, I suggest that to view a film is an extension of one’s inherent world-oriented bearing. On the one hand, this is a matter of definition; Dasein simply is being-in-the-world, which is to say, its being is “worlded,” interpreting itself and its engagements with reference to its world, and vice versa. But I suggest that phenomenological observations of  Ibid., 123 [126].  Ibid. 81  I have in mind here Kendall Walton’s seminal argument regarding film images understood as transparent pictures, where the film viewer sees the actual thing just as in viewing through a periscope or telescope. Kendall Walton, “Transparent Pictures: On the Nature of Photographic Realism,” Critical Inquiry 11 (2) (1984): 246–77. This projective character is likewise what distinguishes the Heideggerian reading from Deleuze’s concept of cinema, which, while emphasizing a dimension of phenomenological disclosure, also relies on a decidedly more passive model of viewership. 79 80



the everyday film-viewing experience also bear this out. It is evidenced by the hermeneutic dimension of appropriating or inhabiting a film’s world and, thus, by experiencing this world meaningfully. By and large, films in the traditional guise of Hollywood International present their stories and are composed with reference to singular, relatively cohesive worlds, worlds which oftentimes have a hypnotic, seductive, or otherwise immersive character of their own. In addition, these are typically worlds in which characters have tasks and plots, where they use practical wisdom and know-how to solve problems. And they are ostensibly worlds we interpret from our own situation. We judge characters, situations, and conflicts in terms of how you or I would react, what you or I would do, or what a “regular person” (Heidegger’s Das Man) would do, and so forth. Thus, insofar as Dasein entails being “with” other subjects, film worlds are both inter-subjective and intra-subjective. To recall Dasein’s literal definition of “being-there,” I am present to the world of a film and, mutatis mutandis, the film world is equally present to me. While there is no space here to provide an exhaustive list of examples, a few brief illustrations may suffice to render my account more concrete. The first-person perspective afforded by the camera eye provides one example of Dasein’s existential presence in the film world, for the film viewer does not see simply what someone else sees. The viewer sees it in the first person. By virtue of being-there, the film viewer cum Dasein appropriates the view of the camera eye. I suggest that a similar logic reveals our engagement with the “tools” or “equipment” appearing in the film world, when we consider the manner in which props often comprise meaningful objects for a film’s world. Consider the prominent placement of a large chest in the foreground of the principal shot throughout the 1948 Hitchcock film Rope. This film not only presents a narrative of two men trying to hide the murder of the man stowed in the chest; it also relies on the visual disclosure afforded by the chest itself. The visual obtrusiveness of the chest throughout the narrative becomes a constant throb of agony for the two protagonists as they attempt to weasel out of their situation by telling a suspicious Jimmy Stewart a series of elaborate lies. Important for our purposes, the chest becomes a point of suspense for the viewer, such that its problematic status (“What do I do with a dead body of a friend, whom I had clear motive to murder?”) reveals the larger, shared world of social commitments and norms to which both the characters and the viewer belongs. The wooden chest can emerge as a problematic item in the hermeneutic of the visual narrative just because of the conflict posed by its world-oriented significance.

The Existential Structures of Dasein’s Being-There The last set of concepts I wish to highlight in Heidegger falls under what he calls the “existential constitution” of Dasein’s “there.”82 Dasein has a constitution that fosters its specific ways of being present and, conversely, the ways  Ibid., 130 [134]




things are present to it. Heidegger terms this general phenomenon “disclosedness” (Erschlossenheit), to emphasize Dasein’s character of, on the one hand, not being closed off but instead being open to the world, and on the other hand, bringing illumination to what it presences, bringing a clearing or lighting to what it encounters such that things can reveal themselves.83 In other words, phenomenological disclosure for Heidegger is actually bidirectional, not solely dependent on Dasein’s existential projection or dative disclosure to Dasein. The concepts at work here are three in number: attunement, understanding, and discourse. They are “equiprimordial,” in that they function as three overlapping, irreducible dimensions of Dasein’s constitution. “Attunement” (Befindlichkeit) is the ontological condition for having moods and for things to be present in mood-colored aspects.84 Heidegger observes that moods in the everyday sense comprise specific ways of being for Dasein, such that they can color one’s perception and judgment and can also cause emotional affect. At the same time, moods are not under Dasein’s complete control; they can assail from without,85 as would be familiar to those who suffer from anxiety or depression. Such moods come seemingly of their own power and cannot be instantly willed away. A mood such as depression is unpleasant not just because of the emotional affect it causes but also because it causes the things in one’s world to lose positive meaning. In this account, then, mood represents the “ontic” dimension of Dasein’s underlying ontological character of attunement. Attunement is the existential structure through which such disclosures occur for one. Heidegger uses fear as an example to flesh out this structure. The experience of fear demonstrates that one is attuned to one’s surroundings such that one understands harm to be possible for oneself or others.86 In my first-person experience of fear, what I fear is not simply an object such as a villain but, equally, the harm that can befall me from without. This experience of fear therefore reveals both my nonsolipsistic being-in-the-world and, in Heidegger’s locution, a specific structure of my being “there.” I suggest that this phenomenology is an essential aspect to the way films use moods as a means for presenting their situations and objects in guises that can affect us as viewers. To cite Rope again as an example, one mood that this film conveys is suspense. For a viewer to experience suspense during Rope is predicated on the narrative’s setup of a situation whose outcome is uncertain and, in any case, uncomfortable. The experience of suspense in turn colors what is disclosed during the course of the film. As viewers we constantly await the moment when the protagonists’ plot is discovered or when they accidently give up their secret. Our being-there, our presence in this situation, is one of disclosures that are suspenseful. Things appear to us in a suspense-driven light. Again,

 Ibid., 129 [132–33].  Ibid., 130 [134]. 85  Ibid., 133 [136]. 86  Ibid., 136 [140]. 83 84



Heidegger’s point is that mood-driven disclosures would not occur if Dasein’s being-in-the-world were not constitutionally attuned, a mode of Dasein’s being. This structuration is similarly borne by the existential Heidegger labels “understanding” (Verstand). At its root, understanding comprises Dasein’s capacity to realize through its disclosures the possibilities of itself and its surroundings. Equally, it comprises Dasein’s character of projection (as with the existential of de-distancing, mentioned earlier), or in this case, to be forward-­ looking, able to envision goals and the ways of achieving them.87 In this light, Heidegger observes that understanding is the condition for a concomitant existential, “interpretation” (Auslegung), the capacity to understand things in this or that way and to appropriate them accordingly. Interpretation thus comprises activating possibilities disclosed in understanding and developing them further.88 For an example, one might think of the meaningful disclosures a child’s understanding reveals in a forest. The child encounters a stick as an interesting object and interprets the stick in a play-acting light, say, as a magic wand. More deeply, in order even to encounter the stick as a fallen limb broken from a tree branch, or to walk in the forest in the first place, requires an act of interpretation. At the root, understanding and its activation in interpretation express that every state of visual perception involves “seeing as,” that is, seeing the object and thus taking it as this or that.89 The existential of understanding recasts the earlier concepts of being-in-the-world, and especially de-distancing, by emphasizing that it is Dasein which projects itself into and illuminates its world, by virtue of disclosing the “there” as a possibility and meaningfully adopting it, interpreting it as containing one meaning or other. As noted earlier in the analysis of Husserl, film-viewing reveals this phenomenology on multiple levels. It has long been observed that film-viewing involves seeing a screen as a projection surface and seeing screened images as depictions of actual things. Films involve seeing a series of screened images to comprise a narrative, involving people one interprets as actors, in a setting one interprets as, for example, New York City, and so on. This is possible because Dasein has the existential capacity to interpret things and surroundings for this or that purpose. Over and above one’s interpretive engagement with the components of the screened image of film, however, as understanding, Dasein fosters the very potency by which films can present meaningful images, scenes, and narratives at all. This is because, existentially speaking, Dasein’s understanding is disclosure and realization of possibility. In simple terms, a film would not be able to present images, scenes, or stories if Dasein did not possess the potency for the projective, illuminating character that renders these elements meaningful. In this light, the Heideggerian model works in the opposite direction of the  Ibid., 138 [142]ff.  Mark A. Wrathall, “Heiddegger on Human Understanding,” in The Cambridge Companion to Heidegger’s Being and Time, ed. Mark A.  Wrathall (Cambridge: Cambridge University Press, 2013), 180ff. 89  Heidegger, Being and Time , 144–45 [149–50]. 87 88



Deleuzian model, in which films reveal reality to us. Rather, in the Heideggerian casting, it is we who disclose reality in and through films because it is we as Daseins who exist projectingly. For an example, consider the opening scene of Sergio Leone’s film Once Upon a Time in the West (1968). This scene unfolds over the course of several minutes, during which the opening credits flash over images of three unfamiliar men sitting idly at a frontier railroad stop. In a certain sense, by virtue of its merely formal character, one cannot gather much from this scene, at least until the action ensues with the arrival of the assassin played by Charles Bronson. Nonetheless, my reason for highlighting this opening scene in which not much happens is precisely for purpose of calling out the projective, interpretive level of understanding upon which its viewing is predicated. To view this scene and experience meaning in it entails interpreting the setting as a rail stop, interpreting the men’s idle sitting as waiting for something, and interpreting the very mood as one of boredom. On a broader level, the viewer interprets this scene, disclosed via the projection of understanding, as one from which further events will follow. In other words, the scene is disclosed in terms of its possibilities. And in the other direction, on a more microlevel, these interpretations occur over and above the more fundamental and basic projection involved in simply appropriating the film’s images as, for example, a source from which a narrative will ensue. Finally, the existential Heidegger calls “discourse” (Rede) is in one way less important than the preceding two just outlined, but in another way much more so. Heidegger defines discourse as the articulation of intelligibility.90 This to say, discourse characterizes the disclosures through which the structuration of things can be expressed. In very simple terms, discourse represents the phenomenological character of experience in which one is naturally receptive to discerning language and able to respond in kind. A crucial qualifier is that discourse is not reducible to language; instead, it is the condition for the disclosures in which language occurs.91 Language has long figured into theoretical discussions of the film image and its meaning, with some holding that film images and shots contain their own kind of words or semiotics, and, likewise, that film’s use of editing, cutting, and montage comprises a kind of syntax. For present purposes, I suggest that Heidegger’s existential of discourse both underwrites the possibility for any notion of film as language, and on the flip side, the very possibility for images to be discursive at all. When Heidegger describes discourse as the articulation of intelligibility, this notion is more broadly indicative of Dasein’s capacity to experience intelligibility and to articulate it in accordance with its disclosure. In a manner of speaking, discourse is already there, before being expressed in words, symbols, or gestures, by virtue of Dasein’s very character of being-there and bringing things to meaningful presence. In summary, I suggest that this phenomenology is really at the root of the theoretical sensibility that cinematic images are able to express anything  Ibid., 155 [161].  Ibid.

90 91



like language through their own power. I would also suggest that Dasein’s existential discursivity is the condition for what contemporary scholars of film as philosophy such as Stephen Mulhall and Robert Sinnerbrink (and farther back, Bazin and Cavell) conceive to be the discursive, philosophical power of cinematic images and narratives.92 In Heideggerian terms, the crucial distinction is that viewing Dasein is discourse, with the result that its disclosures occur in a discursive, intelligible light.

Merleau-Ponty, Sobchack, and the Embodied Film Viewer In the last section of this chapter, I wish to venture briefly into the phenomenology of Maurice Merleau-Ponty (1908–1961) as read through the writings of the contemporary theorist Vivian Sobchack. While it is true that of the three principal figures of phenomenology I have taken up in this chapter, Merleau-­ Ponty’s writings have the most to say about film, Sobchack’s influential adoption of Merleau-Ponty for the philosophy of film considerably advances what insights the latter was able to achieve. More than this, I will suggest in what follows that Sobchack’s cinematic adoption of Merleau-Ponty is decisive for reckoning with the question of film’s ontology as discussed in the earlier sections, especially as this is centered in the viewer experience.93 Before I take up Sobchack, a few words about Merleau-Ponty’s relevance to this subject. Merleau-Ponty’s contribution to phenomenology in the legacy of Husserl and Heidegger centers in the insight that human existence is not constituted simply in a Cartesian intentional consciousness but, equally so, that human existence is founded in the body. The intentional, world-oriented being of the human subject is phenomenologically given as embodied.94 Consciousness is a function of the sensible materiality of the world in which it finds itself. Because perception of the world and the things in it occurs through the senses, this entails that perception (and ultimately consciousness) is always an expression of the sensible. In this light, consciousness is existentially underwritten by a synaesthetic enfoldment into world. For Merleau-Ponty, the physical, material basis of conscious embodiment is therefore irreducible, as are the forms or Gestalts in which perceptual experience occurs. In general, although he significantly revises cornerstones of the phenomenological program, Merleau-Ponty shares with Husserl and Heidegger the position that, at once, subjectivity arrives in a ready-made world, and subjectivity and world are discursively bound to one another. 92  See Mulhall, On Film, Preface and Preface to Second Edition; Robert Sinnerbrink, Cinematic Ethics (Abingdon: Routledge, 2016), Ch. 6. Mulhall and Sinnerbrink both draw heavily on Cavell in particular. 93  In what follows, I cite Soback’s most recent synopsis of these views: Vivian Sobchack, “The Active Eye (Revisited),” Studia Phaenomenologica, XVI, 2016: 63–90. 94  Herbert Spiegelberg, The Phenomenological Movement, Vol. II (The Hague: Springer, 1972), 522.



Also continuous with Husserl and Heidegger is Merleau-Ponty’s observation that the bidirectional character of embodied consciousness—of one’s own consciousness and the things with which one engages—extends into the topological, mobile aspect of human existence. The subject’s engagement with the things of its world is always a function of one’s embodied finitude. Perception is at once receptive and creative. While one can spatially move from this place to that place (because consciousness entails consciousness of one’s own bodily motility), one can never achieve a complete view of something, even by walking around it and examining every side. Unique to Merleau-Ponty’s view here, however, is that this impossibility of total perception comprises an ontological feature of the sensible as such. The world simply reveals itself as ambiguous, both visible and invisible, with embodied subjectivity implanted in both realms.95 This last also speaks to Merleau-Ponty’s own published writings on cinema, most notably his remarks in the brief text of a 1945 lecture entitled “The Film and the New Psychology.” In this text, films are described as “Gestalts,” ready-formed conscious states that are irreducible to raw sense material.96 In this light, Merleau-Ponty understands the images of film to enable a kind of surrogate consciousness that demonstrates the bond—not the divide—between mind and world by virtue of rendering intentional experience visible and communicative from the first-person perspective.97 For similar reasons, films for Merleau-Ponty possess the capacity to exhibit some of the “nascent” dimensions according to which ordinary perception receives meaning.98 This is to say, film images enable the communication of pre-sensory, prerational disclosures of meaning from filmmaker to viewer, in the same way Merleau-Ponty describes the work of painters such as Cezanne whose images convey the genetic moments underlying everyday perception of their subjects.99 A starting point for Sobchack’s adoption of Merleau-Ponty in service of the philosophy of film is the latter’s observation that the perception of motion is inextricably linked with embodied intentionality.100 To perceive motion, as in film’s moving pictures, requires a mental directedness on part of the viewer. One needs to be able to intend the object in motion, where it has come from and where it is going, because the perception of motion transcends a mere passive staring at a scene. And this perception likewise requires implicit bodily motility; to perceive something in motion entails that I am able to comport my body accordingly. I need to be able to adjust my bodily stance, including head, shoulders, and spine. I also need to be able to move my eyes, which in turn implicates the required functionality of pupil and iris. In brief, the very intelligibility of motion at all seems to presuppose an embodied, conscious state. As  Spiegelberg, The Phenomenological Movement, Vol. II, 548.  Merleau-Ponty, “The Film and the New Psychology,” 54. 97  Ibid., 55, 58; Ferencz-Flatz and Hanich, “What is Film Phenomenology?” 15, 28. 98  Yacavone, “Film and the Phenomenology of Art,” 167–69. 99  Merleau-Ponty, “The Film and the New Psychology,” 57–58. 100  Sobchack, “The Active Eye (Revisited),” 65. Much of my summary to follow is also informed by Sobchack’s longer work, The Address of the Eye. 95 96



Sobchack holds, this feature of motion is decisive for both the motion of pictures that occurs in film and the motion in the pictures of film. In other words, this phenomenology involves not just the automatic character of film as a photoplay of static images run through a projector at 24 frames per second. This phenomenology also entails accounting for the intentional consciousness of the viewer who perceives things that move on screen. At the heart of Sobchack’s development of this viewer-centered phenomenology is the thesis that the film has a body, viz. that, phenomenologically speaking, the ontology of films is constituted in the communication of intentional, embodied consciousness from the filmmaker to the film viewer. In other words, a film’s ontology has its seat in the person-to-person sharing of embodied perception. Sobchack makes her case for this thesis by highlighting four specific, overlapping ways in which cinematic movement avails itself of and discloses an embodied intentionality. First is the observation that the film camera’s view, which we take up when we watch a film, is always directional. In traditional lingo, the film camera has a “subject”—it is directed toward a definite something, which it “looks at.” The camera’s view is not a blank stare conscious of nothing. In addition, the perception of movement or the possibility of movement implicates the camera view as one of a subject that is spatially situated. A filmed image or shot always occurs from a specific place, a place which is variously evidenced by what actually makes it into the shot.101 For instance, a moving, aerial view of a city entails that the camera is in a high-up place such as an airplane. Or, camera angles placed very close to the ground might communicate the embodied perception of a child. The second means by which cinematic movement conveys an embodied conscious state is the optical movement enabled by the camera, most notably zoom and focus, but also features such as color change and slow motion. To consider focus, for instance, this feature of films conveys the aspect of embodied consciousness in which the subject does not simply view this or that object point-blank; the subject focuses on aspects and parts of an object that command attention. To focus on one object in everyday visual experience entails a precise direction in which eye and mind work together; it involves a deliberate fading of the remainder of one’s field of vision. Sobchack’s point is that the mechanical capacity of focus employed in a film camera is communicative of an embodied, conscious state of focused vision.102 Third, Sobchack highlights that camera movement demonstrates responsiveness to what is seen through the camera, in the same fashion that embodied consciousness is not merely mentally aware of its surroundings but always visually and spatially engaged with them as well.103 From the dawn of the movable camera, the camera view has always been able to move, via unseen equipment such as the crane, dolly, or shoulder mount, with the result that film is often expressive of an embodied conscious view. In today’s film and television world,  Ibid., 69.  Ibid., 69. 103  Ibid., 70. 101 102



this feature of camera movement often appears in the rapid pan of the camera to reveal characters or objects out of the frame, in order to show their reaction to something occurring in the scene. Another contemporary example occurs in the practice of a film camera following behind an actor, going where the actor goes, though they may appear to be exiting the field of the shot. Finally, the fourth feature Sobchack highlights in film’s movement as expressive of embodied intentionality is its realization and expression of physical finitude. In other words, the film camera’s occasional inability to move or to go where one would like is reflective of the finite, physically confined character of the conscious embodied state. As Sobchack puts it, visually perceptive motility becomes visibly expressive mobility.104 This feature of films conveys the phenomenological character of embodied consciousness by which one sometimes is unable to go where one pleases, or where the object of one’s intentionality can only present itself insofar as it exceeds one’s ability to view. Several of the films of Michael Haneke make use of this capacity. In Haneke’s Cache (2005), several views from static camera shots, filmed surveillance-style, assert a finite character through the camera’s very inertia. These shots reveal their surveilling aspect by virtue of the camera’s lack of movement. And consequently, an expressive aspect is borne by the surveilling shots’ ability to capture unwitting subjects (most notably, the main character Georges Laurent (Daniel Auteuil)  and his innocent son) precisely through the camera’s static, hidden placement. This assertive character afforded by the camera view’s inertia communicates a specific, embodied state of intentionality, in this case the intentional state of a voyeur who can see others from a hidden point of view. In sum, Sobchack’s most significant claim for bringing a phenomenological approach to understanding film is an ontological one, namely that film’s essence lay in the communicative power of embodied consciousness. Or, as she puts it, film comprises a “viewing view” (the embodied intentionality of the filmmaker, captured through the camera) functioning as a “viewed view” (the former, as experienced by the viewer, through the projected film image).105 Film allows one’s embodied point of view, one’s perception, and one’s intentionality to be shared from one person to another. To circle back to where this chapter started, Sobchack’s thesis provides an ontology of film—an account of what film is—by means of a phenomenological account. What makes her account phenomenological? Her account is framed around the descriptive first-person experience of film-viewing, subject to a phenomenological reduction that traces film viewership back to its source, namely the embodied view of the filmmaker. In terms of what we gleaned from Husserl and Heidegger earlier, then, Sobchack’s account is something of an apex to assessing film phenomenologically, insofar as it gives specific focus to the essence of the medium itself as this is disclosed in first-person experience. In view of this conclusion, I suggest that Sobchack’s work offers a genuine challenge to theories of film ontology that omit the first-­ person, descriptive dimension of film-viewing.  Ibid., 70–71.  Sobchack, The Address of the Eye, 3–14, 23–24, 56–57.

104 105



Bibliography Andrew, Dudley. 1985. The Neglected Tradition of Phenomenology in Film Theory. In Movies and Methods, ed. Bill Nichols, vol. 2, 625–632. Berkeley: University of California Press. Bazin, Andre. 1967. The Ontology of the Photographic Image. In What Is Cinema? ed. Hugh Gray, vol. 1, 9–16. Berkeley: University of California Press. Bordwell, David. 1996. Contemporary Film Studies and the Vicissitudes of Grand Theory. In Post-Theory: Reconstructing Film Studies, ed. David Bordwell and Noël Carroll, 3–36. Madison: University of Wisconsin Press. Brough, John. 1992. Some Husserlian Comments on Depiction and Art. American Catholic Philosophical Quarterly LXVI (2): 241–259. ———. 2010. Showing and Seeing: Film as Phenomenology. In Art and Phenomenology, ed. Joseph Parry, 192–213. Abingdon: Routledge. Carroll, Noël. 1996. Prospects for Film Theory: A Personal Assessment. In Post-Theory: Reconstructing Film Studies, ed. David Bordwell and Noël Carroll, 37–68. Madison: University of Wisconsin Press. ———. 2008. The Philosophy of Motion Pictures. Malden: Blackwell. Casebier, Allan. 1991. Film and Phenomenology: Toward a Realist Theory of Cinematic Representation. Cambridge: Cambridge University Press. Cavell, Stanley. 1979. The World Viewed. Revised ed. Cambridge: Harvard University Press. Chamarette, Jenny. 2012. Phenomenology and the Future of Film. Basingstoke: Palgrave Macmillan. Ferencz-Flatz, Christian, and Julian Hanich. 2016. Editor’s Introduction: What Is Film Phenomenology? Studia Phaenomenologica XVI: 11–61. Heidegger, Martin. 2001. The Thing. In Poetry, Language, Thought, 161–83. Trans. Albert Hofstadter. New York: Harper Perennial. ———. 2002. The Age of the World-Picture. In Off the Beaten Track, ed. Julian Young and Kenneth Haynes, 57–72. Cambridge: Cambridge University Press. ———. 2003. The Question Concerning Technology. In Basic Writings, Revised and Expanded, ed. David Farrell Krell, 2nd ed., 307–341. New York: Harper. ———. 2010. Being and Time. Trans. Joan Stambaugh. Revised edn. Albany: SUNY Press. Hopkins, Robert. 2015. The Real Challenge to Photography (as Communicative Representational Art). Journal of the American Philosophical Association 1 (2): 329–348. Husserl, Edmund. 1962. Ideas: General Introduction to Pure Phenomenology. Trans. Boyce Gibson. New York: Collier Books. ———. 2000. Logical Investigations. Trans. J.N.  Findlay. Atlantic Highlands: Humanities Books. ———. 2005. Phantasy, Image Consciousness and Memory (1898–1925). Trans. John B. Brough. Dordrecht: Springer. Loht, Shawn. 2017. Phenomenology of Film: A Heideggerian Account of the Film Experience. Lanham: Lexington Books. Merleau-Ponty, Maurice. 1964. The Film and the New Psychology. In Sense and Non-­ Sense, 48–62. Trans. Hubert L.  Dreyfus, and Patricia Allen Dreyfus. Evanston: Northwestern University Press.



———. 1988. The Sensible World and the World of Expression. In In Praise of Philosophy and Other Essays, 71–79. Trans. John Wild, James Edie, and John O’Neill. Evanston: Northwestern University Press. ———. 2002. Phenomenology of Perception. Trans. Colin Smith. Abingdon: Routledge. Mion, Regina-Nino. 2016. Husserl and Cinematographic Depictive Images: The Conflict Between the Actor and Character. Studia Phaenomenologica XVI: 269–293. Price, Brian. 2008. Heidegger and Cinema. In European Film Theory, ed. Temenuga Trifonova. New York: Routledge. Raviv, Orna. 2016. The Cinematic Point of View: Thinking Film with Merleau-Ponty. Studia Phaenomenologica XVI: 163–183. Rozzoni, Claudio. 2016. Cinema Consciousness: Elements of a Husserlian Approach to Film Image. Studia Phaenomenologica XVI: 295–324. Scruton, Roger. 1981. Photography and Representation. Critical Inquiry 7 (3): 577–603. Sheehan, Thomas. 2017. Heidegger and the Right Heideggerians: Phenomenology Vs. Crypto-Metaphysics. Kronos VI: 78–90. Shum, Peter. 2015. The Evolution of Husserl’s Concept of Imagination. Husserl Studies 31: 213–236. Sinnerbrink, Robert. 2011. New Philosophies of Film: Thinking Images. New  York: Continuum. ———. 2016. Cinematic Ethics. Abingdon: Routledge. Sobchack, Vivian. 1992. The Address of the Eye: A Phenomenology of Film Experience. Princeton: Princeton University Press. ———. 2004. Carnal Thoughts: Embodiment and Moving Image Culture. Berkeley: University of California Press. ———. 2008. Phenomenology. In The Routledge Companion to Philosophy and Film, ed. Paisley Livingston and Carl Plantinga, 435–445. Abingdon: Routledge. ———. 2016. The Active Eye (Revisited). Studia Phaenomenologica XVI: 63–90. Sokolowski, Robert. 2000. Introduction to Phenomenology. Cambridge: Cambridge University Press. ———. 2010. Husserl on First Philosophy. In Philosophy, Phenomenology, Sciences: Essays in Commemoration of Edmund Husserl, ed. Carlo Ierna, Hanne Jacobs, and Filip Mattens, 3–30. Dordrecht: Springer. Sorfa, David. 2014. Phenomenology and Film. In The Routledge Encyclopedia of Film Theory, ed. E. Branigan and W. Buckland, 353–358. Abingdon: Routledge. Spiegelberg, Herbert. 1965. The Phenomenological Movement. 2nd ed. The Hague: Martinus Nijhoff. Vaughan, Hunter. 2012. Where Film Meets Philosophy: Godard, Resnais, and Experiments in Cinematic Thinking. New York: Columbia University Press. Walton, Kendall. 1984. Transparent Pictures: On the Nature of Photographic Realism. Critical Inquiry 11 (2): 246–277. Wrathall, Mark A. 2013. Heidegger on Human Understanding. In The Cambridge Companion to Heidegger’s Being and Time, ed. Mark A.  Wrathall, 177–200. Cambridge: Cambridge University Press. Yacavone, Daniel. 2014. Film Worlds: A Philosophical Aesthetics of Cinema. New York: Columbia University Press. ———. 2016. Film and the Phenomenology of Art: Reappraising Merleau-Ponty on Cinema as Form, Medium, and Expression. New Literary History 47: 159–186. Zahavi, Dan. 2003. Husserl’s Phenomenology. Stanford: Stanford University Press.


Ideology and Experience: The Legacy of Critical Theory Espen Hammer

The term “Critical Theory” has come to denote a number of different yet interrelated approaches in the humanities, ranging from gender and queer studies to postcolonial theory, cultural studies, various forms of Marxist and post-Marxist ideology critique, as well as poststructuralist and psychoanalytic modes of critical engagement in general. In this broad sense Critical Theory has exerted a profound influence upon film studies. Its narrow and original meaning, however, stems from the collaborative work of a group of Marxist philosophers and social scientists initially based in Germany but later operating out of New York: the so-called Frankfurt School. According to Max Horkheimer, one of the early directors of the later so illustrious Institute for Social Research (Institut für Sozialforschung), the purpose of Critical Theory was to unite philosophers and social scientists in the effort to understand and theorize modern society with a view to uncovering its emancipatory potentials. The research would be self-reflective, questioning itself and its conceptual foundations, yet it would also be collaborative, bringing theoretical perspectives and empirical evidence together. While attentive to the dangers of social regression (and with fascism being its most important topic), critical theorists would seek to understand the ambivalent nature of rationalized, social life—its effective causes as well its ideological self-representation. It would deal with social and political life, but also with psychic life and culture, including mass culture. Indeed, the researchers associated with the Institute for Social Research were pioneers when it came to analyzing and theorizing the new media of photography and film.

E. Hammer (*) Temple University, Philadelphia, PA, USA © The Author(s) 2019 N. Carroll et al. (eds.), The Palgrave Handbook of the Philosophy of Film and Motion Pictures,




The focus in this chapter is restricted to the Critical Theory provided by the Frankfurt School. Its aim is to interpret and discuss its contribution to film studies, emphasizing in particular its understanding of cinema and the cinematic image. I offer an overview of the Frankfurt School’s theoretical commitments. I then turn briefly to the accounts offered by two peripheral yet important associates of the Frankfurt School, namely Walter Benjamin and Siegfried Kracauer. Both saw film as harboring an emancipatory, progressive potential. I also discuss Adorno and his conception, developed together with Max Horkheimer in Dialectic of Enlightenment, of the culture industry. I argue that while the account of film on offer in the culture industry theory is deeply critical, associating Hollywood studio productions with propaganda and ideology, Adorno later formed a more positive view of film, arguing that it aspires to be a genuine art. While a number of subsequent theorists of film, including Alexander Kluge and Gertrud Koch, have identified with the early Frankfurt School’s project of critically considering film in its social contexts of production and reception, it would not be incorrect to point out that cinema became less of a preoccupation as the Frankfurt School re-established itself in Frankfurt after the war. As Jürgen Habermas and his associates in the 1960s started to form what later became known as second-generation Frankfurt School theorizing, the attention shifted from social and cultural studies, as well as ideology critique, to the more abstract activities of providing an account of rationality and a social theory geared toward combining interpretive and system-functionalistic perspectives. In its classical form, Frankfurt-style Critical Theory, which came into being in the early 1930s, seems to have come to an end around 1970.

Historical Background and Theoretical Assumptions As Rolf Wiggershaus, Martin Jay, and others have documented, the history of the Frankfurt School is extraordinarily complex, comprising the activities and interactions of a significant number of major thinkers, many of whose names eventually became household names of the humanities and social sciences.1 It was formed as an academic institution of the University of Frankfurt in 1923, predominantly oriented toward scholarly studies of Marx’s writings as well as research on the history of socialism and the labor movement, economic history, and the history and criticism of political economy. When Max Horkheimer acceded to the Directorship in 1930, the Institute changed its course by incorporating a more philosophical attitude while also intensifying its empirical research. In the period leading up to its forced exile in 1933 due to the establishment of Nazi rule and including the eventual re-establishment of the 1  Rolf Wiggershaus, The Frankfurt School: Its History, Theories, and Political Significance, trans. Michael Robertson (Cambridge, Mass.: The MIT Press, 1998); Martin Jay, The Dialectical Imagination: A History of the Frankfurt School and the