ISSN 1466-4615



Samuel Guttenplan

Analytic Philosophy and Film





_Film Theory and Philosophy_

Edited by Richard Allen and Murray Smith

Oxford: Clarendon Press, 1997

xii + 474 pp.

ISBN 0-19-815921-8



The editors in their Introduction aim, in part, to 'describe and clarify the characteristic methods and strategies of analytic philosophy and the film theory it informs'. In addition, they set out on what could be called a mission of sketch and destroy, the target being a continental tradition which has been the dominant philosophical inspiration for film studies. In principle these two introductory tasks would seem the right ones for a collection setting out to show by example the value of analytic film theory. But, as is admitted at various places, the distinction between analytic and continental philosophy is slippery, and tends to be more historical and institutional than philosophical. So, there is a certain woodenness in the attempt to draw up genuinely intellectual battle lines whilst trying to enlist the reader on their side. Nevertheless, being an outsider to non-analytic philosophy of film, I found much that is both clear and informative. As to whether it was completely fair, I cannot say. Also, perhaps contrary to the editiors' intentions, it left me more rather than less interested in reading more widely in non-analytic philosophy, though with a healthy scepticism about the application of some of these doctrines to film studies.


The picture of the analytic philosopher drawn in the Introduction is a little off-putting. Sober, careful, an analyser of concepts, always ready to catch out mistakes or simple bluster, not an adherent of this or that political or temperamental tendency, the analytic philosopher comes across at times as too much the truth seeker to be true, and as rather less than human. With respect to this character, and having been a witness to many intense, though ultimately shallow, debates in areas such as moral philosophy, philosophy of language and, I am afraid, contemporary philosophy of mind, I am on firmer ground in finding the picture of analytic philosophy misleading. As is now standard, the editors note that analytic philosophy is more 'a style or approach, rather than a doctrine or body of knowledge', but this formula suggests a kind of immunity from fadishness that is probably not possible, and certainly isn't realised in any philosophical tradition. In the case of a good deal of analytic philosophy, the main fad -- though it runs too deep for this word to be wholly appropriate -- is the model of science and scientific investigation. The editors are careful to point out that some of their contributors have noted this, and have reacted against it. But my point is not that scientism is ubiquitous in contemporary analytic philosophy, for it certainly isn't. It is just that its existence and pervasiveness gives the lie to the claim that analytic philosophy is not in any sense doctrinal. (That said, I do admit to this bias, and it seems to me more fruitful to begin the war against scientism from within a tradition that takes science seriously, than to have to attack it from the rather wobbly -- even if interesting -- starting points characteristic of non-analytic philosophy.)


The first group of papers are collected under the heading 'Cinematic Representation', and, though the rubric isn't too misleading, the issue in dispute is a bit more general than it suggests. What is at issue is not how the cinema manages to represent this or that, so much as what kind of experience the viewer has when the house lights dim and the film is projected. Is it that the viewer really does see characters and events, or is under the illusion of seeing them, or only imagines seeing them? These alternatives are not of course either exclusive or exhaustive, though they give one sense of what is at issue. Moreover, one should not confuse these sorts of question with two others. The first of these is the question of whether normal human observers really do see motion in motion pictures. We all know that a film is made up of 'stills' but, from the point of view of film studies, it is unrewarding to spend time worrying whether this particular fact makes cinema viewing illusory. Gregory Currie is quite firm about this in his essay 'The Film Theory that Never Was', and no one who joins the more general debate sees reason to take him up on this particular issue. (Incidentally, whilst I think that we can take it as a datum that human beings just do have the sense of seeing motion in motion pictures -- that this is just a fact about our perceptual processes -- I think that Currie is too quick in his assimilation of this case to that of seeing colours. It is not obvious -- and is, I think, false -- that objects are 'really' colourless, and we just happen to see them as coloured.) The second question concerns the fictional status of film characters and plots. Unlike the question of 'illusory' motion, this is an important issue for any philosophical treatment of film. But it is arguably only peripheral to the debate about what viewers 'see' in a darkened cinema. Understanding the nature of fiction is fundamental for literature, painting and other artistic activities, and is perhaps no less fundamental for the more general project of understanding human thought itself. Therefore, understanding the status of characters and plots holds no special or immediate significance for our understanding of the projected moving image.


Sharing many background assumptions, the fundamental issue that separates Currie and Kendal Walton is whether, in viewing a film, one imagines seeing things happen. Currie describes his view as 'representational realism', and he claims that we see projected images as a pictorial representations of how things are. More specifically, he grounds his notion of pictorial representation on the fact that human beings have certain recognitional capacities -- capacities which allow us, for example, to judge that a horse and a picture of a horse share a certain appearance. When confronted with moving images in the cinema, we can tell what is there represented by deploying the same recognitional capacities that are exercised in non-cinematic perception. I have the ability to see a horse moving slowly across a field, and that ability lies behind my taking certain images projected onto a screen to be of a horse moving slowly across a field. In some contrast to this, Walton, in his paper 'On Pictures and Photographs: Objections Answered', claims that we are not merely leaning on certain antecedent visual capacities when we take appropriate images to represent things; we are, as one is tempted to put it, actually imagining seeing the things themselves.


It should be noted that Currie's contribution is not a detailed defence of his view. In effect, he offers us a sketch of work already published, a sketch he calls a 'nervous manifesto' for film theory, and this includes only a summary of the representational realist position. So, it is a little unfair to scrutinise this aspect of his contribution to the volume too closely. But when he writes: 'What I deny is that the standard mode of audience role-play in film watching consists in playing the role of someone who is seeing those events as they happen', one can begin to understand why Walton thinks that there is less to their disagreement than meets the eye. It is certainly hopeless to think that we are ever going to catch either the notion of imagination or seeing in the net of necessary and sufficient conditions, but there are nonetheless plenty of true things we can say about them. One of them is this: in imagining seeing something one does not have to take oneself, or be taken to be, playing any kind of role. When out for a walk, I look at a wintry field and imagine seeing it filled with ripe wheat, I don't do anything like 'playing the role' of someone seeing such a field. To the extent that I understand what is meant by this, I would think that playing the role of a viewer would minimally require me to think of myself as looking at the field, as pointing my eyes in a field-wards direction and adopting a particular focal perspective. But imagining seeing the field filled with ripe wheat requires none of this. This is because it requires me to think only about the field, not about myself. Of course, as Walton readily admits, when one thinks visually in the way required by imagined seeing, the images one has are bound to offer clues as to perspective on the scene. That is part of what it is to be a visual image. If you imagine seeing the field of wheat, you might come to wonder about your own location and try to guess at it by considering the images that you had entertained. But you don't need to engage in this kind of thinking merely in order to imagine seeing the field of wheat. For this sort of reason, I found myself in agreement with Walton's insistence that imagined seeing does not require the viewer to think of the perspectival (and other) consequences of what is imagined. The fact that in the cinema we often have to imagine seeing things from one and then another point of view, or that we sometimes have to imagine seeing something that no one in the film can see, are not grounds for rejecting his account. And he is certainly right to say that imagining seeing something is different from imagining that I see something.


That said, there is still something worrying about Walton's view. When a viewer confronts an image of a field of wheat in a film it just doesn't seem right to say she is imagining seeing the field in the same sense as I do when looking at some such field in winter. One wants to say that the film has somehow taken the place of imagination. Perhaps the whole issue turns on unreliable intuitions about the word 'imagination', but I think there is something deeper here. I found myself thinking that it would have been better if Walton's view were expressed this way: when we view a film of a wheat field, we don't imagine seeing it, we filmically see it. Of course, if this were accepted, it would not be much of an account. All the work would lie ahead, the work involved in analysing the notion of 'filmically', and distinguishing it from 'imagining'. Moreover, there is no guarantee that there could be an informative analysis. So, though unexciting, perhaps the irreducible use of 'filmically' is unavoidable. (At least that is what I would argue if I had the space here.)


In a similar vein, the thesis of transparency that Walton takes to distinguish photographs from paintings would seem more acceptable if we removed the suggestion that it embodied an explanatory analysis of the difference between the two media. Adapting my earlier remark about film, it is surely natural to say that when we see a photograph of uncle Fred, we photographically see Fred; whereas when we see a painting of Fred, it is at least less natural to describe what we experience by some such verb of qualified seeing. To be sure, we can recognise Fred in the painting, we can see Fred in the painting, but these are not parallel to seeing Fred photographically. (We might say that we can see Fred in the photograph, but this usually implies that his image is somehow indistinct or uncharacteristic. In any case, I am not claiming here that there is a sharp and principled distinction between photographs and paintings; merely that it comes much more naturally to take ourselves to be seeing an object in the case of a photograph.)


In his contribution, 'Looking at Motion Pictures', Richard Allen says things which sound something like the minimalist view canvassed above, namely that we do not illusively, recognitionally or imaginatively see things in films, we just see them (albeit, filmically). But he takes this minimalist realism -- for that is what it is -- to be a consequence of the rejection of the so-called 'causal theory of perception'. In summarising his view, he writes:


'By requiring that we separate our perceptual reports into a mental component and a physical component, the causal theory of perception imposes a conceptual straight-jacket [sic] upon our understanding of what it is to look at a picture: what it is that we see when we look at something must be specifiable in terms of a physical object array that causally affects our senses. This profoundly distorts our understanding of what it is we see when we see depictions, because it entails that we are mistaken in claiming that we do see what is depicted in pictures' (91).


Yet earlier he notes: 'we can recognise that an object of sight is not always a physical object . . .', and I found there to be some tension between these two claims -- a tension that could be resolved by a more sensitive understanding of the kinds of thing that go under the label 'causal theory of perception'. The tension is simply stated: when I see a film depicting a field of wheat, I surely see the field of wheat. This is certainly what the first of the above citations suggests. Yet a field of wheat is surely a physical object (taking this in a plausibly broad sense). The problem that intervenes between Allen's two citations is that he never comes to grips with the different kinds of thing that a causal theory could be. In particular: is the theory a sophisticated philosophical theory of what is going on in perception; is it part of the conceptual analysis of 'see'; or, more simply, is it a story about the physical processes that go on when we do see? There are many reasons, not connected with film, to abandon the causal theory of perception when it is used as a bit of philosophical analysis. But it is innocent when used simply to describe the circumstances that obtain when we are correctly described as seeing things. My minimalist suggestion is that, when we go to the cinema we do just see the characters and events there shown. To be sure, we do not see them as we would in the non-filmic situation, nor as we do in the theatre, but that is unsurprising. Moreover, there is nothing illicit in someone's telling of the obvious physicalist story about what is going on in any of these situations. To repeat, where a causal theory would go wrong would be if it took itself to be part of the analysis of what it was to 'really' see something. On this view, one I think just mistaken, we do not in fact see the field of wheat in the cinema, nor the football players on the television, nor for that matter the colour of the rose. But there is no reason to take the scientific causal theory as undermining the fact that we see colours, and this is surely enough to support the minimalist view that we just do see things -- physical objects in some cases -- in the cinema.


The minimalist account of what is seen in the cinema is, as was noted earlier, unexciting. If we take it that we just do filmically see persons and events, and that we shouldn't expect to analyse (this word in crucial here) 'filmically' in terms of imagination, representation, illusion, or even causally, then all the work of any interesting philosophy of film lies beyond the issue which figures so centrally in the first section of the book. Still, in providing sections on 'Meaning, Authorship and Intention', 'Ideology and Ethics', 'Aesthetics' and 'Emotional Response', this collection is far from disappointing; there is a wide discussion of many of the issues that are perhaps more relevant to a substantive philosophy of film. Indeed, the range and depth of these discussions makes it impossible for a review of this length to do more than touch on a few themes.


One particularly interesting series of papers was that concerning authorship. Whilst no philosopher of film, I am still shocked when I meet someone who goes to the cinema regularly but has no idea of who has directed this or that film. Yet it wouldn't be easy to justify this shock. Indeed, the traditions of literary theory over the past decades -- adding together, as it were, the New Criticism, structuralism, and post-structuralism -- would seem to be on the side of those who just don't care, as we often say, 'whose film' it is. Yet cinema criticism, in contrast to that of the written word, has tended to have a kind of immunity to the charge of the 'intentional fallacy'. Perhaps this is because it is such a relatively recent medium, we haven't yet been faced with generations of interpretation and evaluation. The wildly different readings and estimations of Shakespeare's plays in the 18th, 19th and 20th centuries, as well as the absolute historical distance of the plays from us, makes non-intentionalist, audience-based criticism seem more justified.


Whatever the actual explanation of film's relative immunity, there are certainly good reasons to wonder at it. More than any other contemporary art form, films require the contributions of a very large number of people. Nor are these contributions easily relegated to the mere ancillary. For every case in which a director writes the screenplay, controls the cinematography and moulds the actors slavishly to his or her will, there are hundreds where these crucial functions are in many hands. The pieces by Paisley Livingston and Berys Gaut -- both firmly in the intentionalist camp -- consider these issues and come up with rather different suggestions. In his essay 'Cinematic Authorship' Livingston distinguishes between makers of films and authors. Identifying authorship with the intention to communicate, his claim is, at bottom, that whilst all films have makers, only some have authors. Running through a series of examples, he takes us from an imagined case in which a film has no author, to one in which we could be in no doubt about authorship. Moreover, he is careful to insist that any film criticism must take the facts of this range into account; it must not become so obsessed with the author that it invents (constructs) one when there is no actual author, and neglects the real situation that results in the making of a particular film. Gaut's 'Film Authorship and Collaboration' is more ambitious. He does not recommend searching a film's credits for someone counting as the author, nor resurrecting the idea of a constructed author; instead he defends the idea that films have an implied group of authors. Using his example, it is rather like the way in which a group of jazz musicians co-ordinate their actions to create something that is *their* product, and not the product of any one of them.


This view is appealing, but there is a problem with it; or, rather not so much a problem as an area needing further investigation. Films -- at least good-ish ones -- seem to the viewer as if they are moving in some direction, as if they have an internal structure or point. This is not to say that we should always look for something like a message in a film. I think that Gaut is right to demur from the linguistic model of film interpretation -- and recommend George Wilson's thoughtful piece, 'On Film Narrative and Narrative Meaning', as further support for such hesitation -- but any good film does give one the sense of being in the presence of some kind of intelligent control, and this is no doubt the basis for the search for authorship. Yet however tempting, there is really no need to explain this sense of intelligence in films by appealing to real, constructed or multiple authors. My suggestion is that there is something like a task in a film -- the 'common end' of which Gaut himself speaks -- which can be profitably discussed, and can be the basis for understanding a film, without insisting that responses to that end are due to any real or imaginary individual. The model here might still be a jazz group, but a better one might be the constructors of some medieval church. Famously, in the latter case, we have a typically nameless group of individuals responding to a more or less common end: the construction of a building fit for a certain purpose. Here the purpose, and the physical and historical constraints recognised in achieving it, is what gives the finished product its sense of intelligence. And we can, and typically have to, understand this created product without any real knowledge of its creators. Moreover, if we understand a building in this way and apply it to film, the linguistic model will be less appealing, and will therefore not tempt us so easily into a search for authorship. Utterances with meaning are much less easy to see as the products of a group working to a common purpose, since we tend to think that such utterances simply must be down to some one intelligence. (Resolutions passed by committees might be an exception, but one might reasonably doubt their intelligence, or, more seriously, one might know from experience that the 'utterances' of a committee are in fact usually the products of some one of its members.)


Of course, I do not recommend taking the medieval church analogy too far. (Nor would Gaut have leaned too strongly on his jazz analogy.) On the one hand, we do know a lot about how any film was made, and we should be able to use this knowledge in our understanding of the film. On the other hand, the end to which a film is responding is not fixed as it was for a medieval church. Still, neither of these undermine the point I have been making. For I do not recommend that we ignore what we know about the contributions of actual individuals to a film; merely that we should not canvass these contributions for 'real' authorship. Nor should we use this knowledge to attempt the construction of an imagined author. Indeed -- and this addresses my second point -- our attempts should be aimed at what a viewer would take as the constrained ends of the film (what drives the film). Instead of constructing authors we should aim at constructing (what of course might only be) an imagined framework of purposes intrinsic to a film. (I recognise that talk of ends and purposes is apt to mislead, but I am at a loss for the best way to describe the coherence and pointfulness which is I think apparent to most viewers of reasonable films, even if they cannot articulate it. Also, whilst I include narrative in these ends and purposes, what I have in mind is not confined to narrative. In any case, what I am here only gesturing at is what I think lies behind Gaut's talk of the common ends of a film's authorial group.)


As is I hope obvious, virtually every piece in this collection is bound to set one thinking more, and more deeply, about film. My earlier hesitation about the analytic/continental divide stands, but the quality of this collection should recommend it to persons of any philosophical persuasion, provided only that they care about film and the truth about film.


Birkbeck College, London, England




Samuel Guttenplan, 'Analytic Philosophy and Film',  _Film-Philosophy_, vol. 2 no. 36, November 1998 <>.


Copyright © _Film-Philosophy_ 1998




Save as Plain Text Document...Print...Read...Recycle


Join the Film-Philosophy salon,

and receive the journal articles via email as they are published. here


Film-Philosophy (ISSN 1466-4615)

PO Box 26161, London SW8 4WD, England



Back to the Film-Philosophy homepage