Cognitive Film

27/07/2019

The Hawks and the Doves

To understand an aspect of cognitive film criticism it might be useful first to quote several statements from its practitioners and then say a little about where it sees itself coming from. Carl Plantinga says, “It is only by understanding folk psychology in relation to scientific psychology and philosophical analysis that we can more fully understand its workings. This is Bordwell’s approach, and is what I take to be the approach of many film cognitive film theorists and philosophers”. ('Folk Psychology for Film Critics and Scholars') These would include Noel Carroll, Paisley Livingston and Gregory Currie, as well as David Bordwell and Plantinga himself, and less directly Stephen Prince and Geoff King. Speaking of a functional approach to film form, Carroll says, in contrast to what he calls descriptive or structuralist approaches to film analysis, he wants to show how a constellation of choices is made, as opposed to the other methods “which provide no inkling of the impulse behind the form of the film.” ('Film Form: An Argument for a Functional Theory of Style in the Individual Film)' Wondering whether film can practise philosophy, Livingston reckons, “in short, inquiries into film's epistemic values can be a rational strategy insofar as they provide a useful component to the overarching project of philosophical pedagogy and research.” ('Theses on Cinema as Philosophy') What these and numerous other remarks made by such writers show is the need for reasoned arguments, coherent motives and factual evidence. This is in keeping with the cognitive film project which Carroll offers, saying that “in the eighties, an approach to film theorizing, labelled cognitivism, began to take shape as an alternative to psychoanalysis. Cognitivism is not a unified theory. Its name derives from its tendency to look for alternative answers to many of the questions addressed by or raised by psychoanalytic film theories, especially with respect to film reception, in terms of cognitive and rational process rather than irrational or unconscious ones.” (‘Prospects for Film Theory’)

The aim of films cognitivists is what they call “middle-level” research, or “piecemeal theory.” As David Bordwell says, “to be specific: Middle-level research programs have shown that an argument can be at once conceptually powerful and based in evidence without appeal to theoretical bricolage or association of ideas.” (‘Film Studies and Grand Theory’) What interests us here is who of the cognitivists happen to be doing this mid-level work and who isn’t. We find in some instances what we will call the empirical (where evidence is in abundance) and in others the exemplary (where it is often all but absent). Writers like Carroll, Plantinga and Livingston tend to focus on the abstract with the occasional filmic example, while David Bordwell, Stephen Prince and Geoff King are much more inclined to attend to the films themselves, working their way through numerous scenes to try and understand aspects of film form and function. If we are much more sympathetic to the latter than the former it rests not least on the consistency of their project. Prince etc. want to see how films play fair to their ideas, and this requires a great deal of empirical analysis in the process. Speaking of the essays in the bible of film cognitivism, Post-Theory, Bordwell says that what he wants to do is “to commit oneself to letting a variety of middle-level theories compete in the field; to accept the constraints of theorizing rather than paraphrasing chunks of Theory – these decisions means that the essays in this volume do not commit themselves to a single doctrine.” (‘Film Studies and Grand Theory’) That statement might require a psychoanalyst to disentangle it as anyone going through the volume looking for alternative voices will find pretty much all of them fall under the cognitive rubric. There is no place in the book for anyone who would be inclined to support psychoanalytic film theory, structuralism or semiotics. When the continental gang get a mention, it is dismissively. As Bordwell says “to this day, contesting orthodoxy often comes down to picking different Parisians to back. A 1993 book that denounces psychoanalytic film theory as a “religious cult” and “utterly bankrupt” goes on to explain: ‘Rejecting Freud and Lacan, I draw instead upon a variety of theoretical sources: Benjamin, Bataille, Blanchot, Foucault, Deleuze, and Guattari.’ The maitres a pense bump into one another in the pages of film books far more than on the Boulevard St Michel.” Bordwell offers an amusing joke but not quite an accurate account. As Dudley Andrew says, “specifically, Christian Metz formulated his Grande Syntagmatique in the atmosphere of Paris’s Ecole Pratique des Hautes Etudes where he daily encountered Roland Barthes, A. J. Greimas, Gerard Genette, Tzvetan Todorov and Claude Bremond.” ('Concepts in Film Theory') Our point isn’t to be pedantic, just to say that for all the cognitivist’s interest in evidence, they aren’t afraid of the odd unsubstantiated polemical thrust of their own. Livingston once referred to Gilles Deleuze as an “irrational interdisciplinarian”, while Carroll questioned, in Post-Theory, academics who use Freud, Levi-Strauss, and Lacan to further links between such thinkers and certain films saying “probably anything can be made to say anything else once interpretative protocols get as loose as they are in criticism departments nowadays.” This is the sort of irritated remark one might expect someone to make who is waiting in a restaurant for their main course to arrive; it is hardly the rigour cognitivism seems to demand.

So what is is that film cognitivism can give us that other approaches to cinema cannot and perhaps a better way of phrasing the question is to ignore cognitivism as such, looking instead at those who manage to arrive at useful terminology out of following an approach that would loosely fall under the cognitive. Whether it is Bordwell’s The Way Hollywood Tells It, Geoff King’s New Hollywood, or Stephen Prince’s essays ‘True Lies’, ‘Graphic Violence in the Cinema’, ‘The Aesthetics of Slow Motion in the Films of Sam Peckinpah” and ‘Psychoanalytic Film Theory and the Problem of the Missing Spectator’ these are all fine examples of mid-level research. In the latter essay, in Post-Theory, Prince notes: “our problem today (the mid-nineties) in film studies is that theories of spectatorship fly well beyond the data, and in ways that pay too little attention to the evidence we do have about how people watch and interpret film and television.” Prince is asking for a bit more evidence and a little less ambition as he himself often provides the type of work the cognitivists so admire but often themselves do not practice. Prince is in this sense an admirable empiricist, exploring for example in 'True Lies' the issue of computer-generated imagery in film. Prince talks about the problem of indexically based notions of cinema that sees film’s essence in the chemical process even as cinema increasingly moves towards the digital. Are digital films no longer cinema on the basis that they are no longer chemical products? Prince thinks they still are and uses what he calls a correspondence based model to understand how to view digital images in film. The correspondence model is where there is a correspondence “between selected features of the cinematic display and a viewer’s real-world visual and social experience.” This is a useful way of shifting the emphasis from the way films are made to the way they are received. If they are no longer made in the way they were, they are at least received in a similar manner. If anything, it could be argued that we respond more ‘correspondingly’ to a recent film than say a forties one. When someone is killed in classic Hollywood, they fall over and die; in more recent, and perhaps especially digitally-oriented films, we see the horrors a gun can inflict on someone. We have lost the chemical link with reality, but ‘gained’ in other ways in what would seem an ever more vivid comprehension of the violent deed. The point Prince makes, however, is that we can notice with the correspondence model, “at the level of social experience, the evidence indicates that viewers draw from a common stock of moral constructs and interpersonal cues and perceptions when evaluating both people in real life and represented characters in the media. Socially derived assumptions about motive, intent and proper role-based behavior are employed when responding to real and media-based personalities and behavior.” ('True Lies') How far we might wish to take such a correspondence theory is another thing: not least because many people do not have experiences that resemble the violent behaviour evident in many films, and if they did they might be inclined to react rather differently than a viewer just watching an entertainment. Also, many great films by Bunuel, Fellini, Bergman, Tarkovsky and Dreyer draw on metaphysical problems that would have no ready real-world equivalent. But these issues needn’t concern Prince because he offers a pragmatic, piece-meal approach that wants to comprehend digital imaging, "to see how digital imaging technologies are rapidly transforming nearly all phases of contemporary film production.”

If we were to privilege too completely an indexical idea of film, then how come viewers are still no less able to respond correspondingly to the digitised image that has nothing to do with film as a chemical process? This is where the correspondence theory has its uses, and Prince utilises it well for the specific problem he addresses. As he says “a perceptually realistic image is one which structurally corresponds to the viewer’s audiovisual experience of three-dimensional space. Perceptually realistic images correspond to this experience because filmmakers build them to do so.” Prince concludes by saying “the correspondence-based approach to cinematic representation developed here, perceptual realism, the accurate replication of valid 3-D cues, becomes not only the glue cementing digitally created and live-action environment but also the foundation upon which the uniquely transformational functions of cinema exist.” To understand cinema today it is hard to ignore the digital that is transforming it, but that doesn’t mean that earlier, indexically-oriented theorists (like Bazin, Cavell and Barthes in very different ways) are wrong, or naive; more that a new problem arises out of art and technology, and a new model is required to understand it. The problem is a decidedly concrete one, and the empirical is utilised to make sense of it as Prince looks at Forrest Gump, Jurassic Park, Terminator 2 and others to explore digital cinema. This doesn’t mean the indexical approach needs to be ignored or that it has been superseded. Prince’s argument is very useful for the films he is addressing, but anything from Schindler’s List to pornography might still want to address the connection between the world and the image as source. When we see numerous emaciated extras going into the showers we might wonder where Spielberg found such thin people, whether they deliberately lost weight for the role, or whether they have suffered long-term malnourishment. When in a porn film a woman is being screwed by six men who slap her and push her around, are we likely to respond differently than if we were to see it as an obviously digitised piece of pornography? Correspondingly, no, perhaps, but, indexically, surely yes. We sense in Prince’s argument not that he has a thesis to push but a problem to resolve and thus the piece need only be as developed as the conundrum that presents itself – evident in the paradox of the title where we are sometimes in the face of images that we cannot know the validity of as he gives as an example of invisible CGI the crowds in Forrest Gump, while the dinosaurs are obviously visible examples of computer-generated imagery in Jurassic Park. This would seem to us a very fine example of the middle-level theory cognitivists wish to practise but which we believe is too often absent.

A good example of this absence is Noel Carroll’s work. Though in the footnotes of 'True Lies' Prince paraphrases Carroll’s claim that film needs to do more of this mid-level research, Carroll himself is the opposite of such a practitioner. In his essay on mood in art, at one stage Carroll talks about Collingwood and other expressionist theorists as he lists the way his thesis differs from expressionist models. There is something homogenising in Carroll’s take here as he lists four elements of the expressionists’ argument that he disagrees with, including “where they speak of emotions, we are expanding the canvas to include moods”, and that “expression theorists prize the clarification of highly individualized, virtually unique affective states, ones so elusive that ordinary language barely has words for them, whereas we see no reason to maintain that the clarification of these states is, in principle, superior to the clarification of more generic or common ones" ('Art and Mood') Carroll name-checks Collingwood, yet makes only a passing mention of the Romantics, Tolstoy and Langer, and quotes none of the expressionists at all. Do all expression theorists agree on the four points Carroll addresses; would they all see themselves as expressionist theorists in the first instance?

It remains a moot point as Carroll pursues his argument over an idea of mood that would seem outside the ready availability of emotion. While emotion for Carroll is object-directed and temporally brief, mood is more abstract and often lingering. There is no mention in the piece of Kierkegaard’s differentiation between fear (which is tangible) and anxiety which is not, Jean Baudrillard’s suggestion in Fragments that classic films like The Best Years of Our Lives explore feeling while contemporary cinema deals in emotions, or a piece by Brian Henderson called 'Tense, Mood, and Voice in Film'. The latter can be found in an anthology which also has Stephen Prince’s 'True Lies' essay – and also one by Carroll himself: in Film Quarterly, Forty Years - A Selection. Carroll seems to us a cognitivist with his eyes wide shut, someone with an agenda to pursue rather than someone seeking nuances that he wishes to bring forth: so much of the work grinds out a point rather than brings things to light. His arguments often rely on ready-made assumptions that nobody would disagree with prefaced by 'of course' or 'obviously', and by adverbs that are rather more contentious. “Of course the kind of counter-examples involving objectless affective mental states that are pertinent to this discussion are not only negative or dark and dysphoric ones like depression and anxiety.” “Of course moods are not merely the affairs of the intellect.” “Obviously also the feeling tones that attach to mood states are usually but not always themselves pleasurable or unpleasurable.” We arrive at the end of a Carroll article bombarded by the properties of reason and devoid of the wherewithal to understand why the article has been written at all. What problem does he find that needs to be addressed by his intervention? If he had explored an aesthetic problem rather than arrived at a logical solution to a non-problem we might have understood the point and purpose. Why for example are some genres reliant more on mood than others? The noir seems much more so than the Western; would mood be the appropriate word for a horror film even if it is usually infused with a consistent sense of dread, albeit one that is, finally, alleviated?

The mid-level research so admired by Carroll is nevertheless not often practised by the man himself: he wants to make the broadest of generalisations about mood in art without wondering whether it might be a useful term in some aesthetic contexts but of less use elsewhere. The main thrust of the piece – the objectlessness of mood – is a common question in phenomenology and existentialism, but Carroll pays no attention to this at all. Whether it is Sartre discussing Pierre’s absence in ‘The Encounter with Nothingness' or the passages on mood in Heidegger's Being and Time, philosophers have long attended to what mood happens to be in the presence of what is absent, yet Carroll’s scientific bent would seem to deny entertaining such possibilities. Instead of acknowledging cognitivist defeat in the face of what goes beyond ready perception, he wants to find a way of incorporating it into cognitive faculties. If he wished to do so as an act of mid-level research that would be fine, but since he wants to go beyond the specific into the very general, to ignore so much work on the subject by previous philosophers is perverse, perhaps arrogant. Maybe Carroll would justify it by saying such thinkers fall outside the cognitive discipline, but that would be to turn academic enquiry into a parlour game: to suggest that certain thought is within the rules and other thinking procedures outside of it. As Carroll says “the philosophy of art has been a beneficiary of advances in the philosophy of mind and cognitive science”, presumably meaning that his purpose is to illustrate how this happens to be so, and anything outside the field may or may not be useful but that is not his concern. Certainly, the references he provides are cognitively influenced as he mentions in the footnotes essays and books like ‘Towards a Computational Theory of Mood’, The Biopsychology of Mood, and A Theory of Moods and Their Place in Our Science of the Mind. Caroll concludes his essay on mood by indicating there are at least two ways in which mood is created in art. One is by a spillover effect that leaves a residue of mood beyond the feeling; the other is somatic feeling states. In other words, film can induce mood through the nature of situations that are representational and that will create a mood out of the emotional state the film generates. Here, Carroll mentions how Jurassic Park throws us into a scene by showing someone mauled. In the somatic instance, music might have no representational dimension but it can nevertheless create moods: by different tempos, sound levels etc.

We may have noticed that both Prince and Carroll mention Spielberg’s dinosaur film, but where for Caroll it is merely an example as readily to hand as a Shakespeare sonnet or a Beethoven sonata, Prince is interested in the film’s specificity. It is not exemplary; it is empirical. The problem comes out of the film; the film isn’t thrown into an argument – one example amongst many thousands of others. When Prince says “the animators who created the herd of gallimumus that chases actors Sam Neill and two children in Jurassic Park were careful to animate the twenty-four gallis so they would look like they might collide and were reacting to that possibility” he adds “they had to simulate the collision responses in the creatures’ behaviours, as if they were corporeal beings subject to Newtonian space”, Prince here addresses the singularity of Jurassic Park – why the film is of importance to the cinema as an art form and as an entertainment. For Carroll, Jurrasic Park is no more than another example, grist to the mill of argument that continues oblivious to nuance and difference. Carroll manages to be teleological without being historical, saying “I am presuming that what can be claimed for science may be claimed eventually for film theory.” (‘Prospects for Film Theory’) He indicates that philosophy of art has benefited from cognitivism, as science can now begin to explain aesthetics because we have an effective theory of mind. But Carroll attends not at all to how the aesthetic objects change in the course of history. Someone oblivious to the history of aesthetics (in a kind of Carrollian mind experiment) might wonder if Jurassic Park and a Shakespeare sonnet were made and written in the same year.

Are we too harsh on Carroll? One hopes not, though Carroll is happy to be harsh when others lack the necessary cognitive faculties. Looking at Barthes's work in Carroll’s The Philosophy of Horror, Carroll says “this immensely, though typically, turgid passage has many problems” as the empty adverb comes out in force. In ‘Nonfiction Film and Postmodernist Skepticism’ he attacks various theorists of documentary for their “facile deconstructions of the boundaries between fiction and non-fiction that conclude nonfiction film is just like any other kind of fiction.” At least in this piece where he takes on Michael Renov, Bill Nichols, Brian Winston and others, he deals with specifics. If no individual attack is evident in the essay on 'Prospects for Film Theory' it rests on the general nature of the claims. In one of the sections called “biases against truth”, he refutes post-modern skepticist arguments but gives no examples of the argument he is countering except to say: “I have actually attended a film conference where one of these sophisticated, postmodern film Theorists announced that he and his confreres were skeptical about the ideas of truth and falsity.” Again the adverb is doing a lot of the work. If one condemns Carroll it is to try and defend a position that allows for the speculative and the suggestive, for a mode of discussion that isn’t only about a narrow notion of reasoned argument but one asking for a broad range of possible enquiries. Reading Carroll is the opposite of thinking outside the box; it can feel like thinking inside a prison cell.

Like Prince, Geoff King would seem to be a cognitive pragmatist, happy to utilise theories of mind but not beholden to the claims made in their name; willing (like Prince and unlike Carroll) to quote Barthes without denouncing him. Invoking Barthes’ essay The Death of the Author, King says “...the point is to emphasize the extent to which all texts draw on multitudes of pre-established meanings and devices that are not all determined, controlled or limited by the creation of any individual author.” (New Hollywood Cinema) But his main interest is to look at what recent Hollywood filmmaking is, just as Prince was interested in examining how digital imaging was impacting on cinema and why certain arguments invoking the index weren’t always so useful. In one chapter on the blockbuster, King believes critics have exaggerated how much big-budget contemporary Hollywood cinema has half ditched storytelling. While for Fred Pfiel we are now in a post-classical era of Hollywood storytelling, King breaks down why he thinks this isn't the case as he examines films like Speed and Mission: Impossible. We may still feel Pfiel has a point, but we might have to refine our argument and eschew his as we try to find out why we believe Hollywood films from the modern era are quite different from classical models. For Pfeil, the difference resides in “both a particular kind of economy and a particular psychic regime” as Pfeil associates it with a "Fordist version of capitalism” and “a version of patriarchy and male sexuality rooted in Freud’s notion of the Oedipus Complex”. Here, the two are closely linked in "a deeply rooted and organic affiliation" between "the accumulation of unspent dramatic or suspenseful elements" throughout the ascending curve of action... "discharged most completely by the story’s climax – and the preferred rhythms of saving and spending, of repression and release, inscribed into the operation of Oedipal masculinity.” King reckons “his argument is interesting and provocative”, but understandably adds “also subject to a number of qualifications." King doesn’t so much dismiss the argument as attend to the specifics of it, and countering it when he believes Pfeil has exaggerated his case to the detriment of the evidence in front of our eyes. King acknowledges there seems to be a hyperbole to action films in more recent years that might suggest a lack of certainty in approaches to masculinity, and which is over-compensated for in the ‘built’ bodies of male stars for example. But he believes that while “spectacle is important, sometimes. In no cases, however, do we find a complete abandonment of narrative components familiar from the studio or ‘classical’ era.” This is a perfect example of middle-level research as King goes on to show why narrative does matter even in a film like Roland Emmerich’s Godzilla.

Both New Hollywood Cinema and also King's American Independent Cinema are excellent examples of mid-level work with the latter looking at independent cinema in the States from various angles, including the industry, the narrative, the form and genre. King also wonders how independent is an independent film in the US as he says: “how exactly any individual title is marked as sufficiently different from the Hollywood mainstream to qualify as independent is subject to numerous variations explored in detail in this book.” (American Independent Cinema) An example of this independence would be Jon Jost’s All the Vermeer’s of New York. “The unconscious familiarity of continuity editing, and the effortless access it appears to grant viewers of events on-screen, makes it one of the most potent domains in which to create alternative and disjunctive effects.” Jost counters this as King notes “we see one of the central characters, Ann (Emmanuel Chaulet), coming through a door. The door opens towards the camera and she walks in, and then past the camera out of the frame, Cut to a door opening again and the action is repeated, twice, and it becomes clear from the background that it is the same door each time. The action is repeated”. King says, “in the kind of loop-sequence associated with more abstract works of the avante-garde.” How is a film independent, King asks, and doesn’t just logically argue for the components that make it so, offering an abstract ideal of the independent film, but very concretely looks at examples and sees if they would seem to justify the term independent in various manifestations. Perhaps some might insist that King isn’t much of a cognitivist at all, yet as New Hollywood Cinema’s opening chapter is prefaced with a remark by the cognitively-inclined Murray Smith, and Bordwell is commonly quoted, King would perhaps see himself as sympathetic to, even affiliated with cognitivism, without being central to the movement. Yet ironically he practises the mid-level research it so praises better than almost anyone else.

Paisley Livingston would seem to be quite the opposite, a cognitivist who will argue till he is as blue in the face as Carroll while insisting film can and cannot do certain things, and will use the minimal amount of empirical research to put alongside the maximum amount of narrow reasoning. If piece-meal work was the aim of cognitivism, Livingston practices instead wholesale generalization. In Theses on Cinema as Philosophy he explains at the beginning what he will do. “In the first section of this essay, I briefly identify key components of a type of bold thesis about film’s exclusive contribution to philosophy. In section II, I outline what I take to be a ruinous dilemma facing advocates of such a thesis.” Livingston then informs us that of course a philosopher giving a talk that happens to be filmed would be doing philosophy, but the philosophy would be the philosopher talking, not the filmmaker filming. Alternatively film could with its own means indicate thinking (as the Kuleshov effect suggests, where a blank face is met by a cut to different objects that indicate a change of expression on that blank face as we assume the character is lustful, hungry or grieving), but this wouldn’t itself be philosophy at work.

Yet we might wonder if Livingston simplifies his case and offers a paucity of examples partly to prove his point rather than suggest the nuance available in philosophic thinking. In the first instance, where a philosopher gives a talk, if the filmmaker just films it we wouldn't assume any philosophy in the filmmaking. But what about a film such as Lacan Speaks, where the director Francoise Wolff manages to capture in the course of the film aspects of Lacan’s philosophy through the procedures the film offers, and the observations it makes? Rather than merely filming Lacan’s talk to students, Wolff edits to emphasise the blocks, the hesitancies, the interruptions in the symbolic order that is vital to Lacan’s theories. At one moment a student interrupts Lacan’s speech, spills water all over the desk and indicates he wants to remove some of Lacan’s power. This is the law of the father as documented action. Is it philosophy? It is much closer to it in its formal endeavour than the made-up example Livingston gives. Equally, while the Kuleshov effect suggests immediate thinking rather than philosophical reflection, what about a film like Bresson’s The Devil Probably, or Antonioni’s The Eclipse, where the shots don’t allow for such immediacy of mental response? In Antonioni’s film, the central character early on looks out of the window and at a mushroom-like building. We cannot at all know what is on her mind in the links between the two shots as we can in the Kuleshov effect. In The Devil, Probably, we have images of the Seine as we see partially a boat passing through the shot. We then cut to the central character who, in a Kuleshovian cut would suggest that he wants to kill himself, but the cuts don’t allow for such certitude as we wonder what is on the mind of the character rather than assuming to know what that happens to be.

This may or may not be philosophical, but we might be inclined in both instances to say that the characters are thinking philosophically as the figure in the Kuleshov experiment is not, just as many characters in an action film working out exactly what they need to do in a car chase or to take out a villain with a sniper rifle, are not philosophising either. Our question here isn’t finally whether a film can do philosophy or not (in some absolute sense of the term, with filmmakers equivalent to Kant or Wittgenstein,) it is more to argue with the way that Livingston insists it is incapable of doing so without giving specific examples that would allow the argument a subtlety of perspective. We sense Livingston wants to win the argument rather than open up the debate, evident when he says “in short, inquiries into films’ epistemic values can be a rational strategy insofar as they provide a useful complement to the overarching project of philosophical pedagogy and research.” For Livingston, film can paraphrase or it can offer thought experiments, but it cannot do philosophy; it can only illustrate it. One might have a philosophical problem like Descartes’ evil genius, or Plato’s cave, and we might use a film that exemplifies the problem. We could have an example like a train driver deciding whether or not he or she should change points: if they do, they actively kill one person; if they don’t they are not responsible directly for the death of one but indirectly for the death of a much larger number. A film might be useful in exploring just such a dilemma. Yet this would be Livingston’s abstract yen: his interest in what we might call philosophical problems rather than philosophical realities. Is film just too damn concrete for such an exploration?

However, one reason cinema has been so useful to contemporary philosophy is because it can access far more of that philosophical ‘reality’. This is not the same as saying that film resembles life; just that the dilemma in film form can be much more varied and affective than the thought experiment. A decent thought experiment will complicate the initial dilemma in the train example (what if the one person is sympathetic and pregnant), the group unsympathetic and criminal. But film can add numerous layers to that initial twist. What if one of the criminals is not quite so unsympathetic and happens to be the father of the unborn child; what if he is played by a star who has been vital to the film and whose motivations we are following, while the sympathetic woman has hardly appeared in the film at all? Even quite average films address such dilemmas, and good ones complicate them. A film like Konchalovsky’s Runaway Train at one moment has just such a ‘thought experiment’: someone has to decide whether to kill three people (all of them film stars!) or let the train run on and result in a massive chemical explosion.

Basileios Kroustallis offers a different take on the thought experiment than Livingston as he looks at Happy-Go-Lucky, musing over what constitutes an instance of it in philosophy and how it applies to film. “Traditionally a thought experiment is meant to challenge our intuitions or common sense assumptions. Descartes introduced the possibility of an evil demon that systematically misleads ordinary human subjects to experience a world like ours, a proposition that challenged belief in the veracity of perception and reasoning.” (Film-Philosophy) Kroustallis gives other examples as he shows film can sometimes illustrate how flat-footed philosophy can be. We might wish for example to screen various films in a philosophy of thought experiments class, including examples by Locke and Putnam, as well as Nietzsche’s eternal return and John Searle’s Chinese Room. For Kroustallis the vital dimension of a thought experiment is that it counters received wisdom, but on the most basic level it is a philosophical what if… It is philosophy occupying the faculties of the imagination as readily as that of logic, though one reason why Kroustallis thinks many films fail as thought experiments is because they may possess imagination but they lack rigour. Thus the Steve Martin film, All of Me, doesn’t hold to its own logic and fails the test. But Happy-Go-Lucky and Pickpocket pass it as he believes they argue a contrary claim with consistent reasoning procedures. Thus what makes a film a work of philosophy for Kroustallis is the imaginative faculty meeting consistent argumentation as we might then see, in turn, that anything from Plato’s Symposium to Kierkegaard’s Either/Or constitute aesthetic works as readily as philosophical ones. In such a case we can observe that films work not only as examples or a paraphrasing of a philosophical problem. Such a position allows a place for cinematic works to pass for philosophical texts. Showing a film in class wouldn't just be to illustrate a problem but to show how a film might go further than conventional philosophy in exploring it. We might even find that the philosophical problem as dramatic form works with a high degree of variables but that the film has multiplied them further, leading to more nuanced thinking of the problem that requires more than just narrow reasoning faculties. Such a case would not be a parenthesis but an extension. Is the film ‘doing’ philosophy? Perhaps, perhaps not, but it is more than just an illustration of it.

Livingston does furnish his theory with examples, but they seem to us exemplary rather than empirical, concerned with making a point rather than addressing the singularity of the works under discussion. They are dragons to be slayed in pursuit of philosophical purity. Of the Seventh Seal, he says his point in addressing aspects of the mise en scene and its possible exploration of cruel Christendom versus loving Paulinism is that these are what the philosopher would bring to bear in the work: it isn’t present as such within it. Even if a film does go far into its own discursive possibilities, such as Resnais’s My American Uncle, Livingston reckons that though the film challenges and exemplifies the figure at its centre, behavioural scientist Henri Laborit, “it is likely however that the latter does not take us into the cutting edge of current debates” but who is to say that is what philosophy has to do? Livingston wants to prove his point that film cannot practice philosophy and will not so much comprehend the film under discussion as utilise its inevitable failure from a philosophical point of view. What happens is the narrowness of his philosophical perspective meets with the narrow readings of particular films that cannot provide the philosophical rigour he insists upon. Kroustallis more interestingly suggests that some do have this rigour and indicates it is evident as though experiment in such films as Happy-Go-Lucky. Mike Leigh's film proposes a variation on Aristotle’s notion of happiness, and explores whether it is the nature of events that make us happy or our disposition, while All of Me, which takes an aspect of Locke’s thought experiment about the same soul in a different body, lacks Leigh's thoroughness. Kroustallis’s approach allows for an engagement with films and their differences. Livingston’s for an argument that needn’t do this as the aesthetic object gets lost in pedantic, even logocentric reasoning: that philosophy must be worded.

Though David Bordwell’s work preceded advances in cognitvism, and whose earlier essays and books were heavily influenced by the Russian formalists (and indeed Roland Barthes), co-editing Post -Theory with Noel Carroll, Bordwell nailed his colours to the mast as he drove a few other nails into the coffin of psychoanalytic criticism. This is what he more broadly calls "subject-position theory”, with its emphasis on how “through film technology, through narrative structure, through ‘enunciative’ processes, and through particular sorts of representations (for example, those of women), cinema constructs subject positions as defined by ideology and the social formation.” ('Film Studies and Grand Theory') Bordwell’s introduction is wide-ranging and informative, even if he can’t resist along the way a few digs: “many film scholars find it more congenial to read the latest translation of a French master (or to turn to the latest Routledge precis) than to engage in research or reflect on questions for themselves.” Bordwell's need to take on the enemy becomes intrusive when writing a book on style like Figures Traced in Light that devotes several pages to attacking Slavoj Zizek – a thinker who might be no master when it comes to breaking down the elements of mise-en-scene, but has a philosophical grasp that can make Bordwell seem out of his depth or, more precisely, swimming in a much shallower pool than the Slovenian thinker. He sees the philosopher resistant to trans-cultural universals and illustrates why he believes Zizek happens to be wrong-headed on such a question. “I take narrative to be a diffuse and complex phenomenon, with many crisscrossing features spread across the world’s storytelling traditions.” Bordwell’s job is often to show the evidence of similarities across various cultures: how cultural specifics may change, but narrative approaches and formal styles can be quite similar. A key example of this in practice is Bordwell’s other essay in Post-Theory, ‘Convention, Construction and Cinematic Vision’, where he looks at the shot/counter shot across various types of cinema. He uses scenes from Straub/Huillet’s Class Relations, Lang’s Metropolis, Chahine’s The Land, and Spike Lee’s School Daze as he explores ‘contingent universals’. Bordwell makes some valid points and as always is attentive to the specifics of form as he shows how differently the shot-counter shot can be utilised. But when in Figures Traced in Light he attacks Zizek for the philosopher’s resistance to trans-cultural universals we might also say that Bordwell’s weakness is his inability to address singularities. Who might we prefer to read on making sense of Tarkovsky’s work: Zizek working on the borderline of ready sense as he searches out a work of importance beyond its formal procedures, or Bordwell who will be able to tell you the average length of a Tarkovsky shot? It isn’t that Bordwell shouldn’t criticise Zizek and contradict assertively some of the Slovenian’s claims; it is that in doing so he reveals too narrowly his own logical presuppositions and argumentative modes of address. In this Livingston and Bordwell aren’t too dissimilar: this is what an argument is; anything that one deems outside this form of argumentation is not philosophy (Livingston would claim) or happens to be nonsense, as Bordwell wonders how Zizek can claim that certain Japanese works are confusing when they have “goal-oriented protagonists, causal consequence, and a psychology of beliefs and desires.”

The works will, of course, be perceived as confusing for other reasons, and with Zizek a philosopher interested in the irrational procedures that get underneath sense, Bordwell’s claims, while not inaccurate, are pedantically diagnostic rather than enquiringly open. Rather like Livingston’s insistence that Deleuze is an irrational interdisciplinarian, so for Bordwell Zizek is someone who “drifts from his point in asking about epistemic conditions for reality”, offers objections “that are ambiguously stated”, and that his “objections are framed idiosyncratically.” There is a Bordwell-knows-best approach here that Zizek would say is part of the problem: that Bordwell frames the nature of the debate within certain parameters and regards nonsensical, contradictory or obscure anything that falls outside these claims. Zizek is we don’t doubt a provocateur, whose love for paradox and deliberate contentiousness on occasion begs argument, and yet the deep point Zizek makes is that Nietzschean question of what someone wants. It is all very well arguing and counter-arguing as both parties believe they are in possession of the truth, but what do they want in pursuit of their claims? Zizek addresses this when saying in the context of the apparent modesty of the cognitivists, “when...I present my intervention in a debate as a ‘modest contribution’, I again imply the arrogant position of enunciation from which I can afford such a deprecating self-designation. For this very reason the only proper way to counter such statements is to take them more literally than they were meant: ‘Actually, what you’re saying is just a modest opinion?'” Zizek goes on to say that when Post-Theory insists on a clear theoretical classification and gradual generalisation based on careful empirical research, one should bear in mind that this apparently modest position “involves a much more immediate position of enunciation of the Post-Theorist himself/herself as the observer exempted from the object of his study.” (The Fright of Real Tears) We might come away from the debate believing that Bordwell is superficially right and that Zizek is profoundly wrong, but how often retrospectively is the profoundly wrong much more useful than the superficially correct? We might think of famous essays by Tolstoy on Shakespeare, Eliot on Hamlet, Achebe on Heart of Darkness, and read them shaking our head at the claims rather than nodding our head in agreement, but they are so robustly wrong-headed, coming from such a place of enquiry, that they become vital to debate. The point is the deep is more important than the shallow, and while the empirical knows its limits and tends to do very good work on the objects under discussion, when Bordwell, Livingston and others try to make grander claims through a position that Zizek would see as suspect modesty, we notice how shallow the argument can really seem. When we look at Post-Theory we may notice how ungrounded the book happens to be. It knows what it is attacking as it has numerous references to Lacan, Althusser, Barthes and Baudry, but not aware of its own presuppositional base: loosely the analytic philosophy that might include early Wittgenstein, Russell, Ayer, Ryle and others. Thus they attack what would be seen as the continental tradition, but don’t say enough about what they are grounded in. There might have been an awful lot wrong with psychoanalytic film theory drawing on Freud and Lacan, politically oriented structuralist film thought relying on Marx, Saussure and Althusser, but they at least acknowledged their presuppositional base. Cognitivist film thinkers might seem to make a lot of sense, but sometimes better the grounded 'non-sense' of Baudry, Barthes, Comolli, Mulvey and others, than the shallow common sense of the cognitivists.

Unless of course, this sense yields useful results. If the proof of the pudding is in the eating, it helps if we believe a dish has been made. This is where the cognitivists can justify their position empirically and often very well, as we find in Prince, King and also Bordwell when he attends to the films themselves. But what is interesting is that the best work in film that utilises elements of cognitivism tends to concern Hollywood cinema. Whether dealing with American film and both digitisation and violence (as in Prince’s case), blockbuster cinema in King’s, or changes in American film style in Bordwell’s (The Way Hollywood Tells it), all three manage to do very useful, even important research. If this was part of the modesty of the claim, we would have few problems, but James Peterson says in ‘Is a Cognitive Approach to Avant-Garde Cinema Perverse?’ cognitivism needn’t limit itself to the relatively uncontroversial idea that it comments on the Hollywood mainstream. It has things to say about other areas of film too. As Peterson reckons: “let me refine the notion that avant-garde film viewing must usefully be thought of as a kind of problem-solving. We can move beyond issues of basic perception by considering a common feature of avant-garde film viewing – one that usually passes without comment: viewers initially have difficulty comprehending avant-garde films, but they learn to make sense of them.” Bordwell suggests the same with films by Hou Hsiao-hsien and Theo Angelopoulos by prefacing Figures Traced in Light with a comment by Gyorgy Ligeti: “I create neither for the audience nor for myself. I create for solutions.” No matter the problem, there will be a solution – whether it is Hollywood, avant-garde or the ‘art film’. But it is one thing to understand how something is done, quite another to know how one feels about what has been done. If cognitivism has its uses in the Hollywood context, it rests on the close affiliation between what is done and how we are expected to feel. If a filmmaker cuts from a child getting beaten to another looking on concerned but scared, we do not think about the connection between the shots, nor is there affective stalling. But if as in Nagisa Oshima’s Shonen the director cuts from a child to another child being beaten without cutting back and forth between the beating and the onlooker, and if the film avoids showing a concerned look on the onlooker’s face, the form and feeling don’t so easily conjoin. As with our examples from The Devil, Probably and The Eclipse the shots do not suggest the resolution of a problem but the intractability of it. A cognitivist might show that these films counter the traditional Kuleshovian effect, but it cannot explain the relationship between the shots in such films: the shots deny the ready possibility. Watching ‘art films’ and avant-garde works we will indeed learn to know what not to expect, but we cannot so easily understand what we feel as a consequence of the eschewal.

This is why a writer like Richard Misek utilises Heidegger to explore time in its various manifestations, as if well aware that there is a speculative aspect to the problem that demands less a solution than a meditation. As he breaks down boredom into various Heideggarian categories, Misek notes that if “a film kills time, then we do not feel bored while watching it, though we may still afterwards look back on the experience and be left feeling empty as a result.” ('Dead Time: Cinema, Heidegger and Boredom') As he looks at various films including I Am Legend, Once Upon A Time in the West, The Omega Man and The Wind Will Carry Us, he sees how The Omega Man remake I Am Legend has no truck with the boredom the original initially presents as we see Charlton Heston’s character as a solitary man. I am Legend instead shows a chase sequence, a flashback with a chase sequence inside it, and a lot of general mayhem. It kills boredom without a doubt, but its avoidance of boredom would seem to deny the possibility of us feeling the character’s. It must play by the rules of constant engagement which Omega Man slightly eschews and that The Wind Will Carry Us very much counters. In Abbas Kiarostami's film, the central character is involved in making a documentary and must kill time while he waits for a woman to pass away. His frustration is immense but in time it gives way to a different type of interest as he becomes involved in the villager’s lives. His boredom is urban, we might say, but his curiosity rural. By the end of the film the character has found time rather than killed it, and the viewer likewise: “the film allows us to attune ourselves to the villagers’ temporality.” Yet this is speculation we offer rather than a solution. We cannot say we discover rural time as we discover who the murderer happens to be in a Hollywood outing. The viewer watching the latter can attend to the various cognitive cues thrown out; the term cue would seem too assertive a word to indicate what Kiarostami is doing. By framing his analysis through Heidegger’s complex notion of boredom, Misek offers ‘deep’ speculation rather than a series of useful but shallow observations. The more the film penetrates our own speculative faculties, the less we might be inclined to think cognitivism has to say on the subject: the subject of the film and the subject that is the viewer.

This is not at all to deny Bordwell’s importance as a critic of the image, but we believe he is at his best focusing on films that play well to the problem/solution model rather than those that defy it. When he discusses formal and technological changes in modern Hollywood filmmaking in The Way Hollywood Tells It, he provides vigorous terminology and very precise analyses. One of the most significant is the idea of ‘intensified continuities’: how American films have speeded up in recent years while still relying on continuity models. Bordwell is both historically precise and analytically coherent as he looks at how Hollywood has become faster and wonders what terms can be utilised to understand this increased speed. He looks for example at ASLs – average shot lengths over a period of many years. (A term and a method borrowed from Barry Salt.) While “between 1930 to 1960, most feature films contained between three hundred and seven hundred shots, so the average shot length (ASL) hovered between 8 and 11 seconds.” By the seventies, the ASLs averaged between “5 and 8 seconds.” Bordwell says “between 1961 and 1999, I can find only one film with an ASL of less than 2 seconds (Dark City, 1988, 1.8 seconds), but in the 2000s there’s at least one every year” as he mentions Requiem for a Dream, Pirates of the Caribbean, Moulin Rouge. He also sees that fast cutting has impacted on other techniques too. “Vehicles whiz through the foreground, breaking our line of sight. Whiplash pans and jerky reframings present two glimpses linked by a blur, Rack focusing (changing focus between foreground and background) can shift a shot’s composition as crisply as a cut can.”

Many a critic and commentator has talked about Hollywood films getting faster, but nobody has more completely pinned down precisely what that means. Bordwell defines what he sees as the four main elements of this cinematic dromology, of which average shot length is one. Another is ‘going to extremes’. With filmmakers offering many different lens lengths within a given work, sometimes within a given shot. Bordwell quotes a conversation between Oliver Stone and his cinematographer on Salvador. Stone asked whether it would be possible to cut a long lens with a wide-angle lens, and Richardson said, of course, he could. Here filmmakers ‘push perspective’ in ever more elaborate ways, far removed from the use of more or less a single lens length in the course of the film, as deployed by Bresson, Uzo and John Ford. A third is what Bordwell calls ‘closer and closer’ where many more ‘singles’ are used. In numerous recent films, it isn’t enough to get in close, you have to crank up the tension with extreme close-ups that remove everything else from the surrounding space except the tight close-up. This would be quite different from the use of close-up in Bergman or Leone (whose usage would, of course, be distinct from the other), but is now a commonplace, and perhaps even a visual cliché rather than a cinematic convention. Its purpose is often to work up tension rather than define space. As Bordwell notes, while the original Thomas Crown Affair shows the central; character leaving the meeting, in the remake a character exits his shot in a tight close-up. The fourth and final aspect is ‘the prowling camera’. Here the camera is likely to “prowl even if nothing else budges”. Here we have creeping zooms and push-ins. In a classic Hollywood film, the camera would push towards a character in a moment of surprise or revelation; now it is a common device to suggest dramatic possibilities even if there is no drama in the situation. In recent cinema, “even ordinary scenes are heightened to compel attention and to sharpen emotional resonance.”

It would be hard to imagine a better book on contemporary Hollywood, though a critic more sympathetic to the sort of theoretical work Bordwell has more generally half-dismissed might wonder how a term we have briefly thrown in like dromology would give a broader context to Bordwell’s specifically cinematic analysis. Paul Virilio, in works like The Information Bomb, Open Sky and Negative Horizon, addresses the question in myriad manifestations; Bordwell holds to the technical and formal elements in film. Yet it is as a consequence a perfect example of the mid-level research the cognitivists profess to believe in. Along with Prince and King’s work, Bordwell’s The Way Hollywood Tells It, gives us a very clear sense of what quantitative filmic analysis can contribute. It doesn’t get caught up in the logic of its argument, but searches out the empirical nature of its subject. If cognitivism wants to justify its project, it must surely be more Bacon than Descartes: more interested in pursuing the specifics of empirical research than backing itself out of the logical conundrums it finds itself in. In the latter claims are made and then supported by reasoning procedures to the detriment of concrete examples. We might laugh a little when Bordwell uses an algorithm to choose the films he analyses in The History of Film Style but this is at least empirical at its base.

Which leads us to Gregory Currie, who proposes to do away with the term cinematic illusion as he utilises aspects of analytic philosophy to pick apart what he sees are the three main arguments behind the thesis. The first that he disagrees with is that the medium is transparent: “film is transparent in that we see “through” it to the real world, as we see through a window or a lens.” The second “we might call perceptual realism, since it says that film is, or can be, realistic in its recreation of the experience of the real world. This doctrine has been asserted, again by Bazin, in connection with the long-take deep-focus style.” “Finally, there is the claim that film is realistic in its capacity to engender in the viewer an illusion of the reality and presentness of fictional characters and events portrayed. Let us call this view, which seems to be held by studio publicity writers as well as by the sternest Marxist critics of the Hollywood film, illusionism.”(‘Film Reality and Illusion') There is not a single reference to a film in Currie’s article as we are very far away from mid-level research indeed. There are however equational remarks to further the argument. “A representational R is realistic in its representation of feature F for creatures of Kind C if and only if 1) R represents something as having F; (ii) Cs have a certain perceptual capacity P to recognize instances of F; (iii) Cs recognize that R represents something as having F by deploying capacity P.” With no films darkening the argumentative door, Currie calls on pure reasoning. Let us not argue with his logic and instead say no more than what usually makes writing on film brilliant is not whether or not it is especially logical but whether it is insightful. If one were to claim that Martin Scorsese made Taxi Driver. and Taxi Driver is a film, therefore Scorsese is a film director,  our reasoning might be impeccable but the insight of no value whatsoever. When Manny Farber notes of the film that a “chief mechanism of the script is power: how people either fit or don’t fit into the givens of their status, and the power they get from being socially snug” (Negative Space), we might disagree, but a statement has been made. So often, what we want from a critical response to a work of art is not the unequivocal statement but the assertive remark. Perhaps for many a cognitivist, this would pass for irrational, but that would say more about their binary need for opposition, and an Aristotelian approach to non-contradiction than it might about the critic’s response.

This is why we have a word like dialetheia: something that is true and false at the same time. But often the cognitivists work with reasoning procedures that do not accommodate the paradoxical and perhaps consequently have little interest in a perspectivism that can: one for example that allows film to be illusionistic and realistic simultaneously: that we can admire the quality of Jack Nicholson’s acting, link it to other roles as we observe his persona, and at the same time get very emotional when the chief in One Flew Over the Cuckoo’s Nest suffocates his character R. P. McMurphy with a pillow. There is no point arguing that if we are admiring the acting we are aware of the performance and are therefore not focusing on the character and subsequently we can’t be moved. We are moved by the character and we are impressed by the performance. One may take place in the register of the diegesis and the other beyond it, but this causes us no problem, perhaps because we are perspectivisng constantly, not only in what we usually call real life but even the extreme moments of it. James Wood in How Fiction Works gives a couple of examples of apparently irrelevant details that give texture to the work and suggests a writer's capacity for detail. In George Orwell’s 'A Hanging', a man about to go to the gallows dodges a puddle. In War and Peace, a character about to be shot by a firing squad loosens his blindfold, finding it too tight. One is a factual account; the other fictional, but what they both indicate is that perspective allows for the apparently paradoxical: why would a man about to die care to dodge a puddle? Because the feet don’t want to get wet just as the man does not want to die? Perspectivism allows us to see beyond surface contradiction, and if the argument is weaker as a consequence of the empirical, then the argument will just have to deal with it. But if the argument is the thing, better not muddy it with examples that might allow for the law of non-contradiction to be undermined. When Currie says “now film watchers do not behave like people who really believe or even suspect, they are in the presence of axe murders, world-destroying monsters, or nuclear explosions, which is what the films they see frequently represent” we may notice that, like Livingston, Currie falls into what we might call the fallacy of adverbial form: the use of an adverb that is being asked to do rather more work than it can possibly justify. When he says film watchers do not behave like people who really believe...” we might say that some viewers believe more than others, and all viewers probably believe a little that they are in danger from the axe murderer. If we go back to our example of the man about to be hanged, from the foot’s point of view it doesn’t want to get wet. Equally, in the example of the axe murderer, the viewer’s nerves don’t want to be attacked. Part of the viewer might be sitting there well aware that there is no real danger, but another part of them very much believe in that danger. They don’t believe in it enough to run out of the cinema, but they believe in it enough to scream in their seats. The problem we often find in cognitivism is that we have an ideal viewer in an abstract film and the fallacy of adverbial form doing a lot of the heavy-lifting. If, as we have noted, Prince says one of the problems with psychoanalytic film criticism is that too little attention has been given to actual viewers then it doesn’t help when Currie removes the films as well. As he argues against illusionism in film there is no specific film and no specific spectators but there is the odd, telling adverb. Semiotically and psychoanalyticaly inclined “film theories have simply presupposed that film is in some way or other an illusionistic medium.” he says. He also reckons that: “there are a number of other apparent motions which are normally classed as merely apparent”.

Currie’s argument throughout is no doubt from a certain perspective sophisticated as he addresses the difference between primary and secondary qualities, of what he sees as the distinction between cognitive illusion and perceptual illusion, and why he believes it is possible to have sympathy with the first of the three arguments he discusses in terms of cinematic illusionism and very little with the third. But for all the logical strength of his position, there is an empirical and affective weakness: as though cinema is something he has heard a little bit about, and wondered what would happen if he laid an argument on top of it. There is no sense in which he has engaged with cinema as an art form, and the same problem arises in his essay on 'The Passion of Anna', which does at least throw in a few film examples. What is odd about this essay we only mention in passing is that while Currie suggests most of the elements in Bergman’s film are depictive, there are a couple of moments which aren’t – most especially the concluding shot where the central characters dissolves into the film’s pixels. This is true as far as it goes, but rather than examining how this works more generally in films as he could have given examples as disparate as Michael Snow’s Wavelength and Monte Hellman’s Two-Lane Blacktop, he gets lost in the argument that few people would dispute, rather than work with the filmic and affective properties of such a claim. At one moment he says “we should avoid disputes over ways of classing shots together when all we have is two equally good systems, fit for distinct purposes. But I think I detect a substantive difference between us here. [George] Wilson argues for his classification on the grounds that shots containing hallucinatory material represent “internal properties of a character's perceptual state,” and therefore deserve to be classed alongside shots that indicate the character's drink-impaired perception by means of blurred focus.” We might wish he just got on with classifying the non-depictive as the category he acknowledges, rather than arguing with others about what they believe. Part of the problem with this type of exemplary cognitivism is that you don’t have an argument unless you are in an argument. There has to be someone you are contradicting, countering and so on. No doubt this is seen as rigour, but we might be inclined to see it as a certain type of affective prevarication: that it helps the writer avoid dealing with filmic examples and the subtle emotional differences therein.

At one moment in the essay on illusionism, Currie says “we hawks are capable of ontological generosity” and we might be inclined to think of cognitivist hawks and cognitive doves. The doves are the empirically focused who have little axe to grind but empirical data to search out and will use cognitivism if and when it proves useful. But the hawks are much more inclined to be squaring up, ready to do battle for the cognitivist cause. Bordwell, of course, is a man who likes a good scrap too, but anyone spending so much time calculating numerous examples of average shot lengths can only see cognitive aggro as an extracurricular activity. Others go in with boots; a point that hasn’t been missed by some of the practitioners. As Carl Plantinga says there were hard feelings from various sources after Carroll and Bordwell’s empirical polemical attacks in Post-Theory. Plantinga says he hopes that “the conversation will continue to turn away from personalities to the discussion and the debate of ideas.” We hope our own attempt to look sceptically at cognitivism in film has resisted most of the way ad hominem attacks and unfair appraisals of the arguments. No doubt, too little attention has been given to specific and complicated points, but probably no less so than many of the claims made against semiotics, structuralism and psychoanalysis in the cognitivists' work. Our purpose has been to rescue an aspect of the discipline that can be useful for those involved in the aesthetic questions cinema raises, and to remain wary of abstractions that seem to drift in from analytic fields that say nothing about specific artworks. If art is the arena of singularities, of creating the perceptually and experientially new, then we need to be careful when it comes to acknowledging a system that is much more inclined to talk up norms and schemas.

© Tony McKibbin

Tony McKibbin Tony McKibbin

Cognitive Film

The Hawks and the Doves

To understand an aspect of cognitive film criticism it might be useful first to quote several statements from its practitioners and then say a little about where it sees itself coming from. Carl Plantinga says, "It is only by understanding folk psychology in relation to scientific psychology and philosophical analysis that we can more fully understand its workings. This is Bordwell's approach, and is what I take to be the approach of many film cognitive film theorists and philosophers". ('Folk Psychology for Film Critics and Scholars') These would include Noel Carroll, Paisley Livingston and Gregory Currie, as well as David Bordwell and Plantinga himself, and less directly Stephen Prince and Geoff King. Speaking of a functional approach to film form, Carroll says, in contrast to what he calls descriptive or structuralist approaches to film analysis, he wants to show how a constellation of choices is made, as opposed to the other methods "which provide no inkling of the impulse behind the form of the film." ('Film Form: An Argument for a Functional Theory of Style in the Individual Film)' Wondering whether film can practise philosophy, Livingston reckons, "in short, inquiries into film's epistemic values can be a rational strategy insofar as they provide a useful component to the overarching project of philosophical pedagogy and research." ('Theses on Cinema as Philosophy') What these and numerous other remarks made by such writers show is the need for reasoned arguments, coherent motives and factual evidence. This is in keeping with the cognitive film project which Carroll offers, saying that "in the eighties, an approach to film theorizing, labelled cognitivism, began to take shape as an alternative to psychoanalysis. Cognitivism is not a unified theory. Its name derives from its tendency to look for alternative answers to many of the questions addressed by or raised by psychoanalytic film theories, especially with respect to film reception, in terms of cognitive and rational process rather than irrational or unconscious ones." ('Prospects for Film Theory')

The aim of films cognitivists is what they call "middle-level" research, or "piecemeal theory." As David Bordwell says, "to be specific: Middle-level research programs have shown that an argument can be at once conceptually powerful and based in evidence without appeal to theoretical bricolage or association of ideas." ('Film Studies and Grand Theory') What interests us here is who of the cognitivists happen to be doing this mid-level work and who isn't. We find in some instances what we will call the empirical (where evidence is in abundance) and in others the exemplary (where it is often all but absent). Writers like Carroll, Plantinga and Livingston tend to focus on the abstract with the occasional filmic example, while David Bordwell, Stephen Prince and Geoff King are much more inclined to attend to the films themselves, working their way through numerous scenes to try and understand aspects of film form and function. If we are much more sympathetic to the latter than the former it rests not least on the consistency of their project. Prince etc. want to see how films play fair to their ideas, and this requires a great deal of empirical analysis in the process. Speaking of the essays in the bible of film cognitivism, Post-Theory, Bordwell says that what he wants to do is "to commit oneself to letting a variety of middle-level theories compete in the field; to accept the constraints of theorizing rather than paraphrasing chunks of Theory - these decisions means that the essays in this volume do not commit themselves to a single doctrine." ('Film Studies and Grand Theory') That statement might require a psychoanalyst to disentangle it as anyone going through the volume looking for alternative voices will find pretty much all of them fall under the cognitive rubric. There is no place in the book for anyone who would be inclined to support psychoanalytic film theory, structuralism or semiotics. When the continental gang get a mention, it is dismissively. As Bordwell says "to this day, contesting orthodoxy often comes down to picking different Parisians to back. A 1993 book that denounces psychoanalytic film theory as a "religious cult" and "utterly bankrupt" goes on to explain: 'Rejecting Freud and Lacan, I draw instead upon a variety of theoretical sources: Benjamin, Bataille, Blanchot, Foucault, Deleuze, and Guattari.' The maitres a pense bump into one another in the pages of film books far more than on the Boulevard St Michel." Bordwell offers an amusing joke but not quite an accurate account. As Dudley Andrew says, "specifically, Christian Metz formulated his Grande Syntagmatique in the atmosphere of Paris's Ecole Pratique des Hautes Etudes where he daily encountered Roland Barthes, A. J. Greimas, Gerard Genette, Tzvetan Todorov and Claude Bremond." ('Concepts in Film Theory') Our point isn't to be pedantic, just to say that for all the cognitivist's interest in evidence, they aren't afraid of the odd unsubstantiated polemical thrust of their own. Livingston once referred to Gilles Deleuze as an "irrational interdisciplinarian", while Carroll questioned, in Post-Theory, academics who use Freud, Levi-Strauss, and Lacan to further links between such thinkers and certain films saying "probably anything can be made to say anything else once interpretative protocols get as loose as they are in criticism departments nowadays." This is the sort of irritated remark one might expect someone to make who is waiting in a restaurant for their main course to arrive; it is hardly the rigour cognitivism seems to demand.

So what is is that film cognitivism can give us that other approaches to cinema cannot and perhaps a better way of phrasing the question is to ignore cognitivism as such, looking instead at those who manage to arrive at useful terminology out of following an approach that would loosely fall under the cognitive. Whether it is Bordwell's The Way Hollywood Tells It, Geoff King's New Hollywood, or Stephen Prince's essays 'True Lies', 'Graphic Violence in the Cinema', 'The Aesthetics of Slow Motion in the Films of Sam Peckinpah" and 'Psychoanalytic Film Theory and the Problem of the Missing Spectator' these are all fine examples of mid-level research. In the latter essay, in Post-Theory, Prince notes: "our problem today (the mid-nineties) in film studies is that theories of spectatorship fly well beyond the data, and in ways that pay too little attention to the evidence we do have about how people watch and interpret film and television." Prince is asking for a bit more evidence and a little less ambition as he himself often provides the type of work the cognitivists so admire but often themselves do not practice. Prince is in this sense an admirable empiricist, exploring for example in 'True Lies' the issue of computer-generated imagery in film. Prince talks about the problem of indexically based notions of cinema that sees film's essence in the chemical process even as cinema increasingly moves towards the digital. Are digital films no longer cinema on the basis that they are no longer chemical products? Prince thinks they still are and uses what he calls a correspondence based model to understand how to view digital images in film. The correspondence model is where there is a correspondence "between selected features of the cinematic display and a viewer's real-world visual and social experience." This is a useful way of shifting the emphasis from the way films are made to the way they are received. If they are no longer made in the way they were, they are at least received in a similar manner. If anything, it could be argued that we respond more 'correspondingly' to a recent film than say a forties one. When someone is killed in classic Hollywood, they fall over and die; in more recent, and perhaps especially digitally-oriented films, we see the horrors a gun can inflict on someone. We have lost the chemical link with reality, but 'gained' in other ways in what would seem an ever more vivid comprehension of the violent deed. The point Prince makes, however, is that we can notice with the correspondence model, "at the level of social experience, the evidence indicates that viewers draw from a common stock of moral constructs and interpersonal cues and perceptions when evaluating both people in real life and represented characters in the media. Socially derived assumptions about motive, intent and proper role-based behavior are employed when responding to real and media-based personalities and behavior." ('True Lies') How far we might wish to take such a correspondence theory is another thing: not least because many people do not have experiences that resemble the violent behaviour evident in many films, and if they did they might be inclined to react rather differently than a viewer just watching an entertainment. Also, many great films by Bunuel, Fellini, Bergman, Tarkovsky and Dreyer draw on metaphysical problems that would have no ready real-world equivalent. But these issues needn't concern Prince because he offers a pragmatic, piece-meal approach that wants to comprehend digital imaging, to see how digital imaging technologies are rapidly transforming nearly all phases of contemporary film production."

If we were to privilege too completely an indexical idea of film, then how come viewers are still no less able to respond correspondingly to the digitised image that has nothing to do with film as a chemical process? This is where the correspondence theory has its uses, and Prince utilises it well for the specific problem he addresses. As he says "a perceptually realistic image is one which structurally corresponds to the viewer's audiovisual experience of three-dimensional space. Perceptually realistic images correspond to this experience because filmmakers build them to do so." Prince concludes by saying "the correspondence-based approach to cinematic representation developed here, perceptual realism, the accurate replication of valid 3-D cues, becomes not only the glue cementing digitally created and live-action environment but also the foundation upon which the uniquely transformational functions of cinema exist." To understand cinema today it is hard to ignore the digital that is transforming it, but that doesn't mean that earlier, indexically-oriented theorists (like Bazin, Cavell and Barthes in very different ways) are wrong, or naive; more that a new problem arises out of art and technology, and a new model is required to understand it. The problem is a decidedly concrete one, and the empirical is utilised to make sense of it as Prince looks at Forrest Gump, Jurassic Park, Terminator 2 and others to explore digital cinema. This doesn't mean the indexical approach needs to be ignored or that it has been superseded. Prince's argument is very useful for the films he is addressing, but anything from Schindler's List to pornography might still want to address the connection between the world and the image as source. When we see numerous emaciated extras going into the showers we might wonder where Spielberg found such thin people, whether they deliberately lost weight for the role, or whether they have suffered long-term malnourishment. When in a porn film a woman is being screwed by six men who slap her and push her around, are we likely to respond differently than if we were to see it as an obviously digitised piece of pornography? Correspondingly, no, perhaps, but, indexically, surely yes. We sense in Prince's argument not that he has a thesis to push but a problem to resolve and thus the piece need only be as developed as the conundrum that presents itself - evident in the paradox of the title where we are sometimes in the face of images that we cannot know the validity of as he gives as an example of invisible CGI the crowds in Forrest Gump, while the dinosaurs are obviously visible examples of computer-generated imagery in Jurassic Park. This would seem to us a very fine example of the middle-level theory cognitivists wish to practise but which we believe is too often absent.

A good example of this absence is Noel Carroll's work. Though in the footnotes of 'True Lies' Prince paraphrases Carroll's claim that film needs to do more of this mid-level research, Carroll himself is the opposite of such a practitioner. In his essay on mood in art, at one stage Carroll talks about Collingwood and other expressionist theorists as he lists the way his thesis differs from expressionist models. There is something homogenising in Carroll's take here as he lists four elements of the expressionists' argument that he disagrees with, including "where they speak of emotions, we are expanding the canvas to include moods", and that "expression theorists prize the clarification of highly individualized, virtually unique affective states, ones so elusive that ordinary language barely has words for them, whereas we see no reason to maintain that the clarification of these states is, in principle, superior to the clarification of more generic or common ones ('Art and Mood') Carroll name-checks Collingwood, yet makes only a passing mention of the Romantics, Tolstoy and Langer, and quotes none of the expressionists at all. Do all expression theorists agree on the four points Carroll addresses; would they all see themselves as expressionist theorists in the first instance?

It remains a moot point as Carroll pursues his argument over an idea of mood that would seem outside the ready availability of emotion. While emotion for Carroll is object-directed and temporally brief, mood is more abstract and often lingering. There is no mention in the piece of Kierkegaard's differentiation between fear (which is tangible) and anxiety which is not, Jean Baudrillard's suggestion in Fragments that classic films like The Best Years of Our Lives explore feeling while contemporary cinema deals in emotions, or a piece by Brian Henderson called 'Tense, Mood, and Voice in Film'. The latter can be found in an anthology which also has Stephen Prince's 'True Lies' essay - and also one by Carroll himself: in Film Quarterly, Forty Years - A Selection. Carroll seems to us a cognitivist with his eyes wide shut, someone with an agenda to pursue rather than someone seeking nuances that he wishes to bring forth: so much of the work grinds out a point rather than brings things to light. His arguments often rely on ready-made assumptions that nobody would disagree with prefaced by 'of course' or 'obviously', and by adverbs that are rather more contentious. "Of course the kind of counter-examples involving objectless affective mental states that are pertinent to this discussion are not only negative or dark and dysphoric ones like depression and anxiety." "Of course moods are not merely the affairs of the intellect." "Obviously also the feeling tones that attach to mood states are usually but not always themselves pleasurable or unpleasurable." We arrive at the end of a Carroll article bombarded by the properties of reason and devoid of the wherewithal to understand why the article has been written at all. What problem does he find that needs to be addressed by his intervention? If he had explored an aesthetic problem rather than arrived at a logical solution to a non-problem we might have understood the point and purpose. Why for example are some genres reliant more on mood than others? The noir seems much more so than the Western; would mood be the appropriate word for a horror film even if it is usually infused with a consistent sense of dread, albeit one that is, finally, alleviated?

The mid-level research so admired by Carroll is nevertheless not often practised by the man himself: he wants to make the broadest of generalisations about mood in art without wondering whether it might be a useful term in some aesthetic contexts but of less use elsewhere. The main thrust of the piece - the objectlessness of mood - is a common question in phenomenology and existentialism, but Carroll pays no attention to this at all. Whether it is Sartre discussing Pierre's absence in 'The Encounter with Nothingness' or the passages on mood in Heidegger's Being and Time, philosophers have long attended to what mood happens to be in the presence of what is absent, yet Carroll's scientific bent would seem to deny entertaining such possibilities. Instead of acknowledging cognitivist defeat in the face of what goes beyond ready perception, he wants to find a way of incorporating it into cognitive faculties. If he wished to do so as an act of mid-level research that would be fine, but since he wants to go beyond the specific into the very general, to ignore so much work on the subject by previous philosophers is perverse, perhaps arrogant. Maybe Carroll would justify it by saying such thinkers fall outside the cognitive discipline, but that would be to turn academic enquiry into a parlour game: to suggest that certain thought is within the rules and other thinking procedures outside of it. As Carroll says "the philosophy of art has been a beneficiary of advances in the philosophy of mind and cognitive science", presumably meaning that his purpose is to illustrate how this happens to be so, and anything outside the field may or may not be useful but that is not his concern. Certainly, the references he provides are cognitively influenced as he mentions in the footnotes essays and books like 'Towards a Computational Theory of Mood', The Biopsychology of Mood, and A Theory of Moods and Their Place in Our Science of the Mind. Caroll concludes his essay on mood by indicating there are at least two ways in which mood is created in art. One is by a spillover effect that leaves a residue of mood beyond the feeling; the other is somatic feeling states. In other words, film can induce mood through the nature of situations that are representational and that will create a mood out of the emotional state the film generates. Here, Carroll mentions how Jurassic Park throws us into a scene by showing someone mauled. In the somatic instance, music might have no representational dimension but it can nevertheless create moods: by different tempos, sound levels etc.

We may have noticed that both Prince and Carroll mention Spielberg's dinosaur film, but where for Caroll it is merely an example as readily to hand as a Shakespeare sonnet or a Beethoven sonata, Prince is interested in the film's specificity. It is not exemplary; it is empirical. The problem comes out of the film; the film isn't thrown into an argument - one example amongst many thousands of others. When Prince says "the animators who created the herd of gallimumus that chases actors Sam Neill and two children in Jurassic Park were careful to animate the twenty-four gallis so they would look like they might collide and were reacting to that possibility" he adds "they had to simulate the collision responses in the creatures' behaviours, as if they were corporeal beings subject to Newtonian space", Prince here addresses the singularity of Jurassic Park - why the film is of importance to the cinema as an art form and as an entertainment. For Carroll, Jurrasic Park is no more than another example, grist to the mill of argument that continues oblivious to nuance and difference. Carroll manages to be teleological without being historical, saying "I am presuming that what can be claimed for science may be claimed eventually for film theory." ('Prospects for Film Theory') He indicates that philosophy of art has benefited from cognitivism, as science can now begin to explain aesthetics because we have an effective theory of mind. But Carroll attends not at all to how the aesthetic objects change in the course of history. Someone oblivious to the history of aesthetics (in a kind of Carrollian mind experiment) might wonder if Jurassic Park and a Shakespeare sonnet were made and written in the same year.

Are we too harsh on Carroll? One hopes not, though Carroll is happy to be harsh when others lack the necessary cognitive faculties. Looking at Barthes's work in Carroll's The Philosophy of Horror, Carroll says "this immensely, though typically, turgid passage has many problems" as the empty adverb comes out in force. In 'Nonfiction Film and Postmodernist Skepticism' he attacks various theorists of documentary for their "facile deconstructions of the boundaries between fiction and non-fiction that conclude nonfiction film is just like any other kind of fiction." At least in this piece where he takes on Michael Renov, Bill Nichols, Brian Winston and others, he deals with specifics. If no individual attack is evident in the essay on 'Prospects for Film Theory' it rests on the general nature of the claims. In one of the sections called "biases against truth", he refutes post-modern skepticist arguments but gives no examples of the argument he is countering except to say: "I have actually attended a film conference where one of these sophisticated, postmodern film Theorists announced that he and his confreres were skeptical about the ideas of truth and falsity." Again the adverb is doing a lot of the work. If one condemns Carroll it is to try and defend a position that allows for the speculative and the suggestive, for a mode of discussion that isn't only about a narrow notion of reasoned argument but one asking for a broad range of possible enquiries. Reading Carroll is the opposite of thinking outside the box; it can feel like thinking inside a prison cell.

Like Prince, Geoff King would seem to be a cognitive pragmatist, happy to utilise theories of mind but not beholden to the claims made in their name; willing (like Prince and unlike Carroll) to quote Barthes without denouncing him. Invoking Barthes' essay The Death of the Author, King says "...the point is to emphasize the extent to which all texts draw on multitudes of pre-established meanings and devices that are not all determined, controlled or limited by the creation of any individual author." (New Hollywood Cinema) But his main interest is to look at what recent Hollywood filmmaking is, just as Prince was interested in examining how digital imaging was impacting on cinema and why certain arguments invoking the index weren't always so useful. In one chapter on the blockbuster, King believes critics have exaggerated how much big-budget contemporary Hollywood cinema has half ditched storytelling. While for Fred Pfiel we are now in a post-classical era of Hollywood storytelling, King breaks down why he thinks this isn't the case as he examines films like Speed and Mission: Impossible. We may still feel Pfiel has a point, but we might have to refine our argument and eschew his as we try to find out why we believe Hollywood films from the modern era are quite different from classical models. For Pfeil, the difference resides in "both a particular kind of economy and a particular psychic regime" as Pfeil associates it with a Fordist version of capitalism" and "a version of patriarchy and male sexuality rooted in Freud's notion of the Oedipus Complex". Here, the two are closely linked in a deeply rooted and organic affiliation between the accumulation of unspent dramatic or suspenseful elements throughout the ascending curve of action... discharged most completely by the story's climax - and the preferred rhythms of saving and spending, of repression and release, inscribed into the operation of Oedipal masculinity." King reckons "his argument is interesting and provocative", but understandably adds "also subject to a number of qualifications. King doesn't so much dismiss the argument as attend to the specifics of it, and countering it when he believes Pfeil has exaggerated his case to the detriment of the evidence in front of our eyes. King acknowledges there seems to be a hyperbole to action films in more recent years that might suggest a lack of certainty in approaches to masculinity, and which is over-compensated for in the 'built' bodies of male stars for example. But he believes that while "spectacle is important, sometimes. In no cases, however, do we find a complete abandonment of narrative components familiar from the studio or 'classical' era." This is a perfect example of middle-level research as King goes on to show why narrative does matter even in a film like Roland Emmerich's Godzilla.

Both New Hollywood Cinema and also King's American Independent Cinema are excellent examples of mid-level work with the latter looking at independent cinema in the States from various angles, including the industry, the narrative, the form and genre. King also wonders how independent is an independent film in the US as he says: "how exactly any individual title is marked as sufficiently different from the Hollywood mainstream to qualify as independent is subject to numerous variations explored in detail in this book." (American Independent Cinema) An example of this independence would be Jon Jost's All the Vermeer's of New York. "The unconscious familiarity of continuity editing, and the effortless access it appears to grant viewers of events on-screen, makes it one of the most potent domains in which to create alternative and disjunctive effects." Jost counters this as King notes "we see one of the central characters, Ann (Emmanuel Chaulet), coming through a door. The door opens towards the camera and she walks in, and then past the camera out of the frame, Cut to a door opening again and the action is repeated, twice, and it becomes clear from the background that it is the same door each time. The action is repeated". King says, "in the kind of loop-sequence associated with more abstract works of the avante-garde." How is a film independent, King asks, and doesn't just logically argue for the components that make it so, offering an abstract ideal of the independent film, but very concretely looks at examples and sees if they would seem to justify the term independent in various manifestations. Perhaps some might insist that King isn't much of a cognitivist at all, yet as New Hollywood Cinema's opening chapter is prefaced with a remark by the cognitively-inclined Murray Smith, and Bordwell is commonly quoted, King would perhaps see himself as sympathetic to, even affiliated with cognitivism, without being central to the movement. Yet ironically he practises the mid-level research it so praises better than almost anyone else.

Paisley Livingston would seem to be quite the opposite, a cognitivist who will argue till he is as blue in the face as Carroll while insisting film can and cannot do certain things, and will use the minimal amount of empirical research to put alongside the maximum amount of narrow reasoning. If piece-meal work was the aim of cognitivism, Livingston practices instead wholesale generalization. In Theses on Cinema as Philosophy he explains at the beginning what he will do. "In the first section of this essay, I briefly identify key components of a type of bold thesis about film's exclusive contribution to philosophy. In section II, I outline what I take to be a ruinous dilemma facing advocates of such a thesis." Livingston then informs us that of course a philosopher giving a talk that happens to be filmed would be doing philosophy, but the philosophy would be the philosopher talking, not the filmmaker filming. Alternatively film could with its own means indicate thinking (as the Kuleshov effect suggests, where a blank face is met by a cut to different objects that indicate a change of expression on that blank face as we assume the character is lustful, hungry or grieving), but this wouldn't itself be philosophy at work.

Yet we might wonder if Livingston simplifies his case and offers a paucity of examples partly to prove his point rather than suggest the nuance available in philosophic thinking. In the first instance, where a philosopher gives a talk, if the filmmaker just films it we wouldn't assume any philosophy in the filmmaking. But what about a film such as Lacan Speaks, where the director Francoise Wolff manages to capture in the course of the film aspects of Lacan's philosophy through the procedures the film offers, and the observations it makes? Rather than merely filming Lacan's talk to students, Wolff edits to emphasise the blocks, the hesitancies, the interruptions in the symbolic order that is vital to Lacan's theories. At one moment a student interrupts Lacan's speech, spills water all over the desk and indicates he wants to remove some of Lacan's power. This is the law of the father as documented action. Is it philosophy? It is much closer to it in its formal endeavour than the made-up example Livingston gives. Equally, while the Kuleshov effect suggests immediate thinking rather than philosophical reflection, what about a film like Bresson's The Devil Probably, or Antonioni's The Eclipse, where the shots don't allow for such immediacy of mental response? In Antonioni's film, the central character early on looks out of the window and at a mushroom-like building. We cannot at all know what is on her mind in the links between the two shots as we can in the Kuleshov effect. In The Devil, Probably, we have images of the Seine as we see partially a boat passing through the shot. We then cut to the central character who, in a Kuleshovian cut would suggest that he wants to kill himself, but the cuts don't allow for such certitude as we wonder what is on the mind of the character rather than assuming to know what that happens to be.

This may or may not be philosophical, but we might be inclined in both instances to say that the characters are thinking philosophically as the figure in the Kuleshov experiment is not, just as many characters in an action film working out exactly what they need to do in a car chase or to take out a villain with a sniper rifle, are not philosophising either. Our question here isn't finally whether a film can do philosophy or not (in some absolute sense of the term, with filmmakers equivalent to Kant or Wittgenstein,) it is more to argue with the way that Livingston insists it is incapable of doing so without giving specific examples that would allow the argument a subtlety of perspective. We sense Livingston wants to win the argument rather than open up the debate, evident when he says "in short, inquiries into films' epistemic values can be a rational strategy insofar as they provide a useful complement to the overarching project of philosophical pedagogy and research." For Livingston, film can paraphrase or it can offer thought experiments, but it cannot do philosophy; it can only illustrate it. One might have a philosophical problem like Descartes' evil genius, or Plato's cave, and we might use a film that exemplifies the problem. We could have an example like a train driver deciding whether or not he or she should change points: if they do, they actively kill one person; if they don't they are not responsible directly for the death of one but indirectly for the death of a much larger number. A film might be useful in exploring just such a dilemma. Yet this would be Livingston's abstract yen: his interest in what we might call philosophical problems rather than philosophical realities. Is film just too damn concrete for such an exploration?

However, one reason cinema has been so useful to contemporary philosophy is because it can access far more of that philosophical 'reality'. This is not the same as saying that film resembles life; just that the dilemma in film form can be much more varied and affective than the thought experiment. A decent thought experiment will complicate the initial dilemma in the train example (what if the one person is sympathetic and pregnant), the group unsympathetic and criminal. But film can add numerous layers to that initial twist. What if one of the criminals is not quite so unsympathetic and happens to be the father of the unborn child; what if he is played by a star who has been vital to the film and whose motivations we are following, while the sympathetic woman has hardly appeared in the film at all? Even quite average films address such dilemmas, and good ones complicate them. A film like Konchalovsky's Runaway Train at one moment has just such a 'thought experiment': someone has to decide whether to kill three people (all of them film stars!) or let the train run on and result in a massive chemical explosion.

Basileios Kroustallis offers a different take on the thought experiment than Livingston as he looks at Happy-Go-Lucky, musing over what constitutes an instance of it in philosophy and how it applies to film. "Traditionally a thought experiment is meant to challenge our intuitions or common sense assumptions. Descartes introduced the possibility of an evil demon that systematically misleads ordinary human subjects to experience a world like ours, a proposition that challenged belief in the veracity of perception and reasoning." (Film-Philosophy) Kroustallis gives other examples as he shows film can sometimes illustrate how flat-footed philosophy can be. We might wish for example to screen various films in a philosophy of thought experiments class, including examples by Locke and Putnam, as well as Nietzsche's eternal return and John Searle's Chinese Room. For Kroustallis the vital dimension of a thought experiment is that it counters received wisdom, but on the most basic level it is a philosophical what if... It is philosophy occupying the faculties of the imagination as readily as that of logic, though one reason why Kroustallis thinks many films fail as thought experiments is because they may possess imagination but they lack rigour. Thus the Steve Martin film, All of Me, doesn't hold to its own logic and fails the test. But Happy-Go-Lucky and Pickpocket pass it as he believes they argue a contrary claim with consistent reasoning procedures. Thus what makes a film a work of philosophy for Kroustallis is the imaginative faculty meeting consistent argumentation as we might then see, in turn, that anything from Plato's Symposium to Kierkegaard's Either/Or constitute aesthetic works as readily as philosophical ones. In such a case we can observe that films work not only as examples or a paraphrasing of a philosophical problem. Such a position allows a place for cinematic works to pass for philosophical texts. Showing a film in class wouldn't just be to illustrate a problem but to show how a film might go further than conventional philosophy in exploring it. We might even find that the philosophical problem as dramatic form works with a high degree of variables but that the film has multiplied them further, leading to more nuanced thinking of the problem that requires more than just narrow reasoning faculties. Such a case would not be a parenthesis but an extension. Is the film 'doing' philosophy? Perhaps, perhaps not, but it is more than just an illustration of it.

Livingston does furnish his theory with examples, but they seem to us exemplary rather than empirical, concerned with making a point rather than addressing the singularity of the works under discussion. They are dragons to be slayed in pursuit of philosophical purity. Of the Seventh Seal, he says his point in addressing aspects of the mise en scene and its possible exploration of cruel Christendom versus loving Paulinism is that these are what the philosopher would bring to bear in the work: it isn't present as such within it. Even if a film does go far into its own discursive possibilities, such as Resnais's My American Uncle, Livingston reckons that though the film challenges and exemplifies the figure at its centre, behavioural scientist Henri Laborit, "it is likely however that the latter does not take us into the cutting edge of current debates" but who is to say that is what philosophy has to do? Livingston wants to prove his point that film cannot practice philosophy and will not so much comprehend the film under discussion as utilise its inevitable failure from a philosophical point of view. What happens is the narrowness of his philosophical perspective meets with the narrow readings of particular films that cannot provide the philosophical rigour he insists upon. Kroustallis more interestingly suggests that some do have this rigour and indicates it is evident as though experiment in such films as Happy-Go-Lucky. Mike Leigh's film proposes a variation on Aristotle's notion of happiness, and explores whether it is the nature of events that make us happy or our disposition, while All of Me, which takes an aspect of Locke's thought experiment about the same soul in a different body, lacks Leigh's thoroughness. Kroustallis's approach allows for an engagement with films and their differences. Livingston's for an argument that needn't do this as the aesthetic object gets lost in pedantic, even logocentric reasoning: that philosophy must be worded.

Though David Bordwell's work preceded advances in cognitvism, and whose earlier essays and books were heavily influenced by the Russian formalists (and indeed Roland Barthes), co-editing Post -Theory with Noel Carroll, Bordwell nailed his colours to the mast as he drove a few other nails into the coffin of psychoanalytic criticism. This is what he more broadly calls subject-position theory", with its emphasis on how "through film technology, through narrative structure, through 'enunciative' processes, and through particular sorts of representations (for example, those of women), cinema constructs subject positions as defined by ideology and the social formation." ('Film Studies and Grand Theory') Bordwell's introduction is wide-ranging and informative, even if he can't resist along the way a few digs: "many film scholars find it more congenial to read the latest translation of a French master (or to turn to the latest Routledge precis) than to engage in research or reflect on questions for themselves." Bordwell's need to take on the enemy becomes intrusive when writing a book on style like Figures Traced in Light that devotes several pages to attacking Slavoj Zizek - a thinker who might be no master when it comes to breaking down the elements of mise-en-scene, but has a philosophical grasp that can make Bordwell seem out of his depth or, more precisely, swimming in a much shallower pool than the Slovenian thinker. He sees the philosopher resistant to trans-cultural universals and illustrates why he believes Zizek happens to be wrong-headed on such a question. "I take narrative to be a diffuse and complex phenomenon, with many crisscrossing features spread across the world's storytelling traditions." Bordwell's job is often to show the evidence of similarities across various cultures: how cultural specifics may change, but narrative approaches and formal styles can be quite similar. A key example of this in practice is Bordwell's other essay in Post-Theory, 'Convention, Construction and Cinematic Vision', where he looks at the shot/counter shot across various types of cinema. He uses scenes from Straub/Huillet's Class Relations, Lang's Metropolis, Chahine's The Land, and Spike Lee's School Daze as he explores 'contingent universals'. Bordwell makes some valid points and as always is attentive to the specifics of form as he shows how differently the shot-counter shot can be utilised. But when in Figures Traced in Light he attacks Zizek for the philosopher's resistance to trans-cultural universals we might also say that Bordwell's weakness is his inability to address singularities. Who might we prefer to read on making sense of Tarkovsky's work: Zizek working on the borderline of ready sense as he searches out a work of importance beyond its formal procedures, or Bordwell who will be able to tell you the average length of a Tarkovsky shot? It isn't that Bordwell shouldn't criticise Zizek and contradict assertively some of the Slovenian's claims; it is that in doing so he reveals too narrowly his own logical presuppositions and argumentative modes of address. In this Livingston and Bordwell aren't too dissimilar: this is what an argument is; anything that one deems outside this form of argumentation is not philosophy (Livingston would claim) or happens to be nonsense, as Bordwell wonders how Zizek can claim that certain Japanese works are confusing when they have "goal-oriented protagonists, causal consequence, and a psychology of beliefs and desires."

The works will, of course, be perceived as confusing for other reasons, and with Zizek a philosopher interested in the irrational procedures that get underneath sense, Bordwell's claims, while not inaccurate, are pedantically diagnostic rather than enquiringly open. Rather like Livingston's insistence that Deleuze is an irrational interdisciplinarian, so for Bordwell Zizek is someone who "drifts from his point in asking about epistemic conditions for reality", offers objections "that are ambiguously stated", and that his "objections are framed idiosyncratically." There is a Bordwell-knows-best approach here that Zizek would say is part of the problem: that Bordwell frames the nature of the debate within certain parameters and regards nonsensical, contradictory or obscure anything that falls outside these claims. Zizek is we don't doubt a provocateur, whose love for paradox and deliberate contentiousness on occasion begs argument, and yet the deep point Zizek makes is that Nietzschean question of what someone wants. It is all very well arguing and counter-arguing as both parties believe they are in possession of the truth, but what do they want in pursuit of their claims? Zizek addresses this when saying in the context of the apparent modesty of the cognitivists, "when...I present my intervention in a debate as a 'modest contribution', I again imply the arrogant position of enunciation from which I can afford such a deprecating self-designation. For this very reason the only proper way to counter such statements is to take them more literally than they were meant: 'Actually, what you're saying is just a modest opinion?'" Zizek goes on to say that when Post-Theory insists on a clear theoretical classification and gradual generalisation based on careful empirical research, one should bear in mind that this apparently modest position "involves a much more immediate position of enunciation of the Post-Theorist himself/herself as the observer exempted from the object of his study." (The Fright of Real Tears) We might come away from the debate believing that Bordwell is superficially right and that Zizek is profoundly wrong, but how often retrospectively is the profoundly wrong much more useful than the superficially correct? We might think of famous essays by Tolstoy on Shakespeare, Eliot on Hamlet, Achebe on Heart of Darkness, and read them shaking our head at the claims rather than nodding our head in agreement, but they are so robustly wrong-headed, coming from such a place of enquiry, that they become vital to debate. The point is the deep is more important than the shallow, and while the empirical knows its limits and tends to do very good work on the objects under discussion, when Bordwell, Livingston and others try to make grander claims through a position that Zizek would see as suspect modesty, we notice how shallow the argument can really seem. When we look at Post-Theory we may notice how ungrounded the book happens to be. It knows what it is attacking as it has numerous references to Lacan, Althusser, Barthes and Baudry, but not aware of its own presuppositional base: loosely the analytic philosophy that might include early Wittgenstein, Russell, Ayer, Ryle and others. Thus they attack what would be seen as the continental tradition, but don't say enough about what they are grounded in. There might have been an awful lot wrong with psychoanalytic film theory drawing on Freud and Lacan, politically oriented structuralist film thought relying on Marx, Saussure and Althusser, but they at least acknowledged their presuppositional base. Cognitivist film thinkers might seem to make a lot of sense, but sometimes better the grounded 'non-sense' of Baudry, Barthes, Comolli, Mulvey and others, than the shallow common sense of the cognitivists.

Unless of course, this sense yields useful results. If the proof of the pudding is in the eating, it helps if we believe a dish has been made. This is where the cognitivists can justify their position empirically and often very well, as we find in Prince, King and also Bordwell when he attends to the films themselves. But what is interesting is that the best work in film that utilises elements of cognitivism tends to concern Hollywood cinema. Whether dealing with American film and both digitisation and violence (as in Prince's case), blockbuster cinema in King's, or changes in American film style in Bordwell's (The Way Hollywood Tells it), all three manage to do very useful, even important research. If this was part of the modesty of the claim, we would have few problems, but James Peterson says in 'Is a Cognitive Approach to Avant-Garde Cinema Perverse?' cognitivism needn't limit itself to the relatively uncontroversial idea that it comments on the Hollywood mainstream. It has things to say about other areas of film too. As Peterson reckons: "let me refine the notion that avant-garde film viewing must usefully be thought of as a kind of problem-solving. We can move beyond issues of basic perception by considering a common feature of avant-garde film viewing - one that usually passes without comment: viewers initially have difficulty comprehending avant-garde films, but they learn to make sense of them." Bordwell suggests the same with films by Hou Hsiao-hsien and Theo Angelopoulos by prefacing Figures Traced in Light with a comment by Gyorgy Ligeti: "I create neither for the audience nor for myself. I create for solutions." No matter the problem, there will be a solution - whether it is Hollywood, avant-garde or the 'art film'. But it is one thing to understand how something is done, quite another to know how one feels about what has been done. If cognitivism has its uses in the Hollywood context, it rests on the close affiliation between what is done and how we are expected to feel. If a filmmaker cuts from a child getting beaten to another looking on concerned but scared, we do not think about the connection between the shots, nor is there affective stalling. But if as in Nagisa Oshima's Shonen the director cuts from a child to another child being beaten without cutting back and forth between the beating and the onlooker, and if the film avoids showing a concerned look on the onlooker's face, the form and feeling don't so easily conjoin. As with our examples from The Devil, Probably and The Eclipse the shots do not suggest the resolution of a problem but the intractability of it. A cognitivist might show that these films counter the traditional Kuleshovian effect, but it cannot explain the relationship between the shots in such films: the shots deny the ready possibility. Watching 'art films' and avant-garde works we will indeed learn to know what not to expect, but we cannot so easily understand what we feel as a consequence of the eschewal.

This is why a writer like Richard Misek utilises Heidegger to explore time in its various manifestations, as if well aware that there is a speculative aspect to the problem that demands less a solution than a meditation. As he breaks down boredom into various Heideggarian categories, Misek notes that if "a film kills time, then we do not feel bored while watching it, though we may still afterwards look back on the experience and be left feeling empty as a result." ('Dead Time: Cinema, Heidegger and Boredom') As he looks at various films including I Am Legend, Once Upon A Time in the West, The Omega Man and The Wind Will Carry Us, he sees how The Omega Man remake I Am Legend has no truck with the boredom the original initially presents as we see Charlton Heston's character as a solitary man. I am Legend instead shows a chase sequence, a flashback with a chase sequence inside it, and a lot of general mayhem. It kills boredom without a doubt, but its avoidance of boredom would seem to deny the possibility of us feeling the character's. It must play by the rules of constant engagement which Omega Man slightly eschews and that The Wind Will Carry Us very much counters. In Abbas Kiarostami's film, the central character is involved in making a documentary and must kill time while he waits for a woman to pass away. His frustration is immense but in time it gives way to a different type of interest as he becomes involved in the villager's lives. His boredom is urban, we might say, but his curiosity rural. By the end of the film the character has found time rather than killed it, and the viewer likewise: "the film allows us to attune ourselves to the villagers' temporality." Yet this is speculation we offer rather than a solution. We cannot say we discover rural time as we discover who the murderer happens to be in a Hollywood outing. The viewer watching the latter can attend to the various cognitive cues thrown out; the term cue would seem too assertive a word to indicate what Kiarostami is doing. By framing his analysis through Heidegger's complex notion of boredom, Misek offers 'deep' speculation rather than a series of useful but shallow observations. The more the film penetrates our own speculative faculties, the less we might be inclined to think cognitivism has to say on the subject: the subject of the film and the subject that is the viewer.

This is not at all to deny Bordwell's importance as a critic of the image, but we believe he is at his best focusing on films that play well to the problem/solution model rather than those that defy it. When he discusses formal and technological changes in modern Hollywood filmmaking in The Way Hollywood Tells It, he provides vigorous terminology and very precise analyses. One of the most significant is the idea of 'intensified continuities': how American films have speeded up in recent years while still relying on continuity models. Bordwell is both historically precise and analytically coherent as he looks at how Hollywood has become faster and wonders what terms can be utilised to understand this increased speed. He looks for example at ASLs - average shot lengths over a period of many years. (A term and a method borrowed from Barry Salt.) While "between 1930 to 1960, most feature films contained between three hundred and seven hundred shots, so the average shot length (ASL) hovered between 8 and 11 seconds." By the seventies, the ASLs averaged between "5 and 8 seconds." Bordwell says "between 1961 and 1999, I can find only one film with an ASL of less than 2 seconds (Dark City, 1988, 1.8 seconds), but in the 2000s there's at least one every year" as he mentions Requiem for a Dream, Pirates of the Caribbean, Moulin Rouge. He also sees that fast cutting has impacted on other techniques too. "Vehicles whiz through the foreground, breaking our line of sight. Whiplash pans and jerky reframings present two glimpses linked by a blur, Rack focusing (changing focus between foreground and background) can shift a shot's composition as crisply as a cut can."

Many a critic and commentator has talked about Hollywood films getting faster, but nobody has more completely pinned down precisely what that means. Bordwell defines what he sees as the four main elements of this cinematic dromology, of which average shot length is one. Another is 'going to extremes'. With filmmakers offering many different lens lengths within a given work, sometimes within a given shot. Bordwell quotes a conversation between Oliver Stone and his cinematographer on Salvador. Stone asked whether it would be possible to cut a long lens with a wide-angle lens, and Richardson said, of course, he could. Here filmmakers 'push perspective' in ever more elaborate ways, far removed from the use of more or less a single lens length in the course of the film, as deployed by Bresson, Uzo and John Ford. A third is what Bordwell calls 'closer and closer' where many more 'singles' are used. In numerous recent films, it isn't enough to get in close, you have to crank up the tension with extreme close-ups that remove everything else from the surrounding space except the tight close-up. This would be quite different from the use of close-up in Bergman or Leone (whose usage would, of course, be distinct from the other), but is now a commonplace, and perhaps even a visual clich rather than a cinematic convention. Its purpose is often to work up tension rather than define space. As Bordwell notes, while the original Thomas Crown Affair shows the central; character leaving the meeting, in the remake a character exits his shot in a tight close-up. The fourth and final aspect is 'the prowling camera'. Here the camera is likely to "prowl even if nothing else budges". Here we have creeping zooms and push-ins. In a classic Hollywood film, the camera would push towards a character in a moment of surprise or revelation; now it is a common device to suggest dramatic possibilities even if there is no drama in the situation. In recent cinema, "even ordinary scenes are heightened to compel attention and to sharpen emotional resonance."

It would be hard to imagine a better book on contemporary Hollywood, though a critic more sympathetic to the sort of theoretical work Bordwell has more generally half-dismissed might wonder how a term we have briefly thrown in like dromology would give a broader context to Bordwell's specifically cinematic analysis. Paul Virilio, in works like The Information Bomb, Open Sky and Negative Horizon, addresses the question in myriad manifestations; Bordwell holds to the technical and formal elements in film. Yet it is as a consequence a perfect example of the mid-level research the cognitivists profess to believe in. Along with Prince and King's work, Bordwell's The Way Hollywood Tells It, gives us a very clear sense of what quantitative filmic analysis can contribute. It doesn't get caught up in the logic of its argument, but searches out the empirical nature of its subject. If cognitivism wants to justify its project, it must surely be more Bacon than Descartes: more interested in pursuing the specifics of empirical research than backing itself out of the logical conundrums it finds itself in. In the latter claims are made and then supported by reasoning procedures to the detriment of concrete examples. We might laugh a little when Bordwell uses an algorithm to choose the films he analyses in The History of Film Style but this is at least empirical at its base.

Which leads us to Gregory Currie, who proposes to do away with the term cinematic illusion as he utilises aspects of analytic philosophy to pick apart what he sees are the three main arguments behind the thesis. The first that he disagrees with is that the medium is transparent: "film is transparent in that we see "through" it to the real world, as we see through a window or a lens." The second "we might call perceptual realism, since it says that film is, or can be, realistic in its recreation of the experience of the real world. This doctrine has been asserted, again by Bazin, in connection with the long-take deep-focus style." "Finally, there is the claim that film is realistic in its capacity to engender in the viewer an illusion of the reality and presentness of fictional characters and events portrayed. Let us call this view, which seems to be held by studio publicity writers as well as by the sternest Marxist critics of the Hollywood film, illusionism."('Film Reality and Illusion') There is not a single reference to a film in Currie's article as we are very far away from mid-level research indeed. There are however equational remarks to further the argument. "A representational R is realistic in its representation of feature F for creatures of Kind C if and only if 1) R represents something as having F; (ii) Cs have a certain perceptual capacity P to recognize instances of F; (iii) Cs recognize that R represents something as having F by deploying capacity P." With no films darkening the argumentative door, Currie calls on pure reasoning. Let us not argue with his logic and instead say no more than what usually makes writing on film brilliant is not whether or not it is especially logical but whether it is insightful. If one were to claim that Martin Scorsese made Taxi Driver. and Taxi Driver is a film, therefore Scorsese is a film director, our reasoning might be impeccable but the insight of no value whatsoever. When Manny Farber notes of the film that a "chief mechanism of the script is power: how people either fit or don't fit into the givens of their status, and the power they get from being socially snug" (Negative Space), we might disagree, but a statement has been made. So often, what we want from a critical response to a work of art is not the unequivocal statement but the assertive remark. Perhaps for many a cognitivist, this would pass for irrational, but that would say more about their binary need for opposition, and an Aristotelian approach to non-contradiction than it might about the critic's response.

This is why we have a word like dialetheia: something that is true and false at the same time. But often the cognitivists work with reasoning procedures that do not accommodate the paradoxical and perhaps consequently have little interest in a perspectivism that can: one for example that allows film to be illusionistic and realistic simultaneously: that we can admire the quality of Jack Nicholson's acting, link it to other roles as we observe his persona, and at the same time get very emotional when the chief in One Flew Over the Cuckoo's Nest suffocates his character R. P. McMurphy with a pillow. There is no point arguing that if we are admiring the acting we are aware of the performance and are therefore not focusing on the character and subsequently we can't be moved. We are moved by the character and we are impressed by the performance. One may take place in the register of the diegesis and the other beyond it, but this causes us no problem, perhaps because we are perspectivisng constantly, not only in what we usually call real life but even the extreme moments of it. James Wood in How Fiction Works gives a couple of examples of apparently irrelevant details that give texture to the work and suggests a writer's capacity for detail. In George Orwell's 'A Hanging', a man about to go to the gallows dodges a puddle. In War and Peace, a character about to be shot by a firing squad loosens his blindfold, finding it too tight. One is a factual account; the other fictional, but what they both indicate is that perspective allows for the apparently paradoxical: why would a man about to die care to dodge a puddle? Because the feet don't want to get wet just as the man does not want to die? Perspectivism allows us to see beyond surface contradiction, and if the argument is weaker as a consequence of the empirical, then the argument will just have to deal with it. But if the argument is the thing, better not muddy it with examples that might allow for the law of non-contradiction to be undermined. When Currie says "now film watchers do not behave like people who really believe or even suspect, they are in the presence of axe murders, world-destroying monsters, or nuclear explosions, which is what the films they see frequently represent" we may notice that, like Livingston, Currie falls into what we might call the fallacy of adverbial form: the use of an adverb that is being asked to do rather more work than it can possibly justify. When he says film watchers do not behave like people who really believe..." we might say that some viewers believe more than others, and all viewers probably believe a little that they are in danger from the axe murderer. If we go back to our example of the man about to be hanged, from the foot's point of view it doesn't want to get wet. Equally, in the example of the axe murderer, the viewer's nerves don't want to be attacked. Part of the viewer might be sitting there well aware that there is no real danger, but another part of them very much believe in that danger. They don't believe in it enough to run out of the cinema, but they believe in it enough to scream in their seats. The problem we often find in cognitivism is that we have an ideal viewer in an abstract film and the fallacy of adverbial form doing a lot of the heavy-lifting. If, as we have noted, Prince says one of the problems with psychoanalytic film criticism is that too little attention has been given to actual viewers then it doesn't help when Currie removes the films as well. As he argues against illusionism in film there is no specific film and no specific spectators but there is the odd, telling adverb. Semiotically and psychoanalyticaly inclined "film theories have simply presupposed that film is in some way or other an illusionistic medium." he says. He also reckons that: "there are a number of other apparent motions which are normally classed as merely apparent".

Currie's argument throughout is no doubt from a certain perspective sophisticated as he addresses the difference between primary and secondary qualities, of what he sees as the distinction between cognitive illusion and perceptual illusion, and why he believes it is possible to have sympathy with the first of the three arguments he discusses in terms of cinematic illusionism and very little with the third. But for all the logical strength of his position, there is an empirical and affective weakness: as though cinema is something he has heard a little bit about, and wondered what would happen if he laid an argument on top of it. There is no sense in which he has engaged with cinema as an art form, and the same problem arises in his essay on 'The Passion of Anna', which does at least throw in a few film examples. What is odd about this essay we only mention in passing is that while Currie suggests most of the elements in Bergman's film are depictive, there are a couple of moments which aren't - most especially the concluding shot where the central characters dissolves into the film's pixels. This is true as far as it goes, but rather than examining how this works more generally in films as he could have given examples as disparate as Michael Snow's Wavelength and Monte Hellman's Two-Lane Blacktop, he gets lost in the argument that few people would dispute, rather than work with the filmic and affective properties of such a claim. At one moment he says "we should avoid disputes over ways of classing shots together when all we have is two equally good systems, fit for distinct purposes. But I think I detect a substantive difference between us here. [George] Wilson argues for his classification on the grounds that shots containing hallucinatory material represent "internal properties of a character's perceptual state," and therefore deserve to be classed alongside shots that indicate the character's drink-impaired perception by means of blurred focus." We might wish he just got on with classifying the non-depictive as the category he acknowledges, rather than arguing with others about what they believe. Part of the problem with this type of exemplary cognitivism is that you don't have an argument unless you are in an argument. There has to be someone you are contradicting, countering and so on. No doubt this is seen as rigour, but we might be inclined to see it as a certain type of affective prevarication: that it helps the writer avoid dealing with filmic examples and the subtle emotional differences therein.

At one moment in the essay on illusionism, Currie says "we hawks are capable of ontological generosity" and we might be inclined to think of cognitivist hawks and cognitive doves. The doves are the empirically focused who have little axe to grind but empirical data to search out and will use cognitivism if and when it proves useful. But the hawks are much more inclined to be squaring up, ready to do battle for the cognitivist cause. Bordwell, of course, is a man who likes a good scrap too, but anyone spending so much time calculating numerous examples of average shot lengths can only see cognitive aggro as an extracurricular activity. Others go in with boots; a point that hasn't been missed by some of the practitioners. As Carl Plantinga says there were hard feelings from various sources after Carroll and Bordwell's empirical polemical attacks in Post-Theory. Plantinga says he hopes that "the conversation will continue to turn away from personalities to the discussion and the debate of ideas." We hope our own attempt to look sceptically at cognitivism in film has resisted most of the way ad hominem attacks and unfair appraisals of the arguments. No doubt, too little attention has been given to specific and complicated points, but probably no less so than many of the claims made against semiotics, structuralism and psychoanalysis in the cognitivists' work. Our purpose has been to rescue an aspect of the discipline that can be useful for those involved in the aesthetic questions cinema raises, and to remain wary of abstractions that seem to drift in from analytic fields that say nothing about specific artworks. If art is the arena of singularities, of creating the perceptually and experientially new, then we need to be careful when it comes to acknowledging a system that is much more inclined to talk up norms and schemas.


© Tony McKibbin