ISSN 1466-4615

Volume 3 Number 27, June 1999



Robert Hopkins

Pictures and Film; Philosophy and the Empirical Disciplines

A Reply to Dean





Jeffrey T. Dean

Getting a Good View of Depiction

_Film-Philosophy_, vol. 3 no. 26, June 1999


I am grateful to Jeffrey Dean for his helpful and insightful review. For the most part, his grasp of my ambitions in _Picture, Image and Experience_, and of my attempts to achieve them, is perfect. There is one point at which perhaps I failed to make my position clear to him, and I will attempt to clarify this in what follows. I will then say a little in response to one of his more substantive criticisms. I close by asking how the book's claims might apply to film.



Dean notes that I describe (in pages 15-17) four features of our experience of pictures, seeing-in. However, I am at pains to point out that these features do not provide an adequate characterization of that experience (17). Some of these features are too vague to be of much help as they stand, and as a set they are exhibited by other experiences too. The most fundamental claim of the book is intended to fill this lacuna. It is a claim about what constitutes seeing-in, a claim intended to capture its phenomenology -- that seeing-in is experiencing resemblance in outline shape.


I am worried that this is not how Dean sees the contours of my account. For when he introduces the notion of outline shape, it is as an answer to the following question: 'in virtue of *what* are we said to experience resemblance between a depiction and its object?' On at least one natural reading, this is not a question I seek to answer at all. I aim to capture the phenomenology of seeing-in, by construing it as the experience of resemblance in outline shape. It is a further question what features an object must have to sustain the experience so constituted. And it is an essential part of my treatment of the problem posed by misrepresenting pictures (chapter five) that, while a decent account of depiction must answer the constitutive question, it need not be more than consistent with the answer to the question about how that experience is sustained (112-3). The task of providing that answer falls instead to the empirical disciplines of psychology, neuropsychology, and art history.


The substantive criticism I would like to address also introduces the theme of the boundary between philosophy and other disciplines. Dean says some complimentary things about my account of visual imagining, but is clearly disappointed that I do not attempt to integrate my views with results in cognitive psychology. I find it difficult to get clear about the various issues here, and the relations between those disciplines which hope to contribute to them. So what I have to say will be very tentative. But I hope nonetheless to describe the interrelations in such a way as to take the edge off Dean's disappointment.


My goal in the last chapter of the book was to get clearer about the phenomenology of visualizing, and thereby to gain some understanding of what visualizing is. As regards the first part of this goal, appeal to the empirical disciplines, in the form in which we now know them, seems to me to be of necessarily limited use. Not that such disciplines can never illuminate phenomenology. On the contrary, there are concrete instances of their having done so. But of any discovery at the level of processing, there is always a farther question: what, if any, phenomenological dimension there is to the facts thus uncovered. This farther question needs answering by attention to the phenomenon -- here visualizing -- *as experienced*. So, as far as phenomenology goes, the contribution of the empirical disciplines, at least in their current state of development, is heuristic. They provide clues as to where to look for interesting aspects of phenomenology. Given this, while I might indeed have used those disciplines to spur my investigations, I have not failed in a duty in not doing so. It is legitimate to ignore one set of clues, provided the set one is using are proving fruitful. For clues are not data, and a good investigation is not obliged to consider all of the former, as it is with respect to the latter.


This defence, such as it is, leaves untouched the other part of my stated goal, to explore what constitutes visualizing. And here I must concede that phenomenological approaches are vulnerable. If some mental event lacked just one aspect of the phenomenology I ascribe to visualizing, would it necessarily not count as an instance of this last? It is hard to feel confident that the answers to such questions always go my way. [1] However, while this raises questions about quite what my project can be, it does not obviously lead to the triumph of the empirical disciplines. For there is at least as open a question about the significance of their results for constitutive inquiries. Perhaps when we visualize we deploy some of the same cognitive resources as when we see, but is overlap in that respect, or even just overlap to a similar degree, necessary if any mental state, of any creature, however structured, is to count as visual imagining?



How far do my claims about pictures tell us anything about cinematic representation, and to what extent can they be extended to illuminate cinema as an art? In part, these questions reduce to another, viz. how far is cinema a pictorial art? For, naturally enough, insofar as what we appreciate in cinema is extra-pictorial, an account of pictures will have nothing to contribute to our understanding of those elements.


It is obvious that there are important aspects to cinema which are not pictorial -- literary, musical, and theatrical elements. But it is equally obvious that cinema is, in at least some sense, fundamentally pictorial. Standardly, the medium of communication for cinema is either entirely or in key part pictorial representation. It is so even where the content thus conveyed is hard to make out (consider the magical aerial sequence of the great waterfalls in Wong Kar-Wai's _Happy Together_), or where, even though there is representation, there is no *figurative* representation, as in the extended psychedelic sequence towards the close of _2001: A Space Odyssey_. [2] Cases in which pictorial representation plays little or no role at all are comparatively rare -- the most extreme example is Derek Jarman's _Blue_. For the rest, although the relative importance of linguistic (the spoken word) and pictorial content shifts from case to case (and although there are many different ways in which either of these might be parasitic upon, or usefully inform, the other), it is the presence of pictorial content, with an important role in the content of the whole, which makes a work recognizably cinematic -- as opposed, say, to a piece of radio drama.


This core pictorial element in cinema is something to which my account applies with relative ease. My account of depiction, as Dean notes, is roughly that P depicts O iff some part of P is experienced as resembling O in outline shape, and some standard of correctness applies, be it intentionally or causally grounded, by which it is appropriate to see P in that way. If I am to experience cinematic marks as resembling something else in this way, it is necessary (i) that I see the marks *as marks* -- I could not experience resemblance if the resembling item did not feature in my perception of my environment. But it is also necessary (ii) that I see the marks as organized in a certain way, i.e. as resembling the depicted item in outline shape. The first condition here is readily met. The marks on the cinema screen, although changing almost permanently and having fewer visible properties than the marks on a canvas, are no more transparent to me than the oil swirls left by the artist's brush. This is the fundamental reason why cinema is not an illusionistic art. It does not fool us into taking what it represents to be present because it does not engender experiences which are the phenomenological twins of those the represented scenes would support. And it does not engender such twins because when we see the lights falling on a flat screen, we see them, in part, as just that.


Condition (ii) is no more problematic for cinematic images than for still photographs, and considerably less problematic than for some sorts of picture. The cinematic process projects three-dimensional objects onto the screen in such a way as to preserve outline shape. What preserves outline shape is at least likely to sustain experience of resemblance in outline shape (though, as noted above, the question of what does so sustain is an empirical one, and I very much doubt that actual resemblance is either necessary or sufficient for the experience of it). Of course, cinematic images move as other pictures do not. But if it is plausible that we see outline shapes at all (chapter four), it is plausible that we see changes in outline shape. If so, we have all the resources we need to experience the moving images as resembling in outline shape the changing objects they represent.


This leaves the matter of standards of correctness. Here the case of cinema is undoubtedly complicated in interesting ways, compared with most still pictures. In the book I argue that neither the causal nor the intentional way of grounding a standard of correctness can be reduced to the other (chapter four, section one). This leaves it open, however, that certain pictures might draw on both. I suspect this is true of cinematic images. The situation will be further complicated if some intentional standard of correctness is established by exploiting some prior causally-grounded standard. Thus far these sketchy comments are intended to apply to classical cinema. The theoretical terrain is ruptured further once we turn our attention to newer cinematic technologies, such as the use of cartoon or computer-generated imagery as part of the whole image. However, while the theoretical issues here are intricate, I see no grounds for anxiety that they might threaten my basic framework. The work which needs doing can be done within that structure. Futile to throw away the tool because the material requires its careful use.


I hope these comments, sketchy as they are, make it at least somewhat plausible that my account of depiction fits the cinematic case. Let me conclude by asking what the consequences of this might be for the aesthetics of film.


These consequences are limited, but not negligible. I will note two. As Dean sees clearly, my central goal is to make good sense of the idea that pictures are *visual* representations. Pictures sustain a special kind of visual experience, and their most important features, in terms of what they can and must represent, and what is required to make sense of them, flow from this fact. That visual experience is not to be understood illusionistically. Seeing something in a picture is quite different, phenomenologically, from seeing it in the flesh. But one way of understanding the drift of my position is to see it as offering a better account of the experience, which nonetheless earns us the right to many of the conclusions a thoughtful illusionist might draw. Elsewhere, I have defended one of these -- the idea that pictures might, inter alia, offer us the very same visual satisfactions as the scenes they represent. [3] Those satisfactions might be aesthetic, erotic, or of yet other kinds. The range will depend in key part on what sort of satisfactions are offered by looking at things in the flesh. The first consequence of applying my view to cinema is that it draws attention to the possibility that much of that range of satisfactions will be on offer in looking at films too. This seems to me to account for some of the visceral appeal of film (though not, of course, of film as opposed to other sorts of picturing). It lets our sensibility engage with the represented scenes in many of the ways we could were we (visually) to encounter those scenes face-to-face. [4]


The second consequence is more theoretical, and more cautionary. I have said that cinema is fundamentally pictorial, and that pictures are fundamentally visual. Both claims to basicness need careful unpacking. But, however they are unpacked, it seems likely that other phenomena, of significant aesthetic interest, might ride on the back of the visual core of cinematic representation. The right way to think about such parasitism is, in the first instance, to model it on the case of seeing in the flesh. Just as some item I see on a daily basis might come to embody for me some aspect of my life or world, so some character or object seen in the course of a film might, *in a closely analogous way*, come to embody some idea the piece seeks to explore or deride. And if one approaches matters by this path, talk of *symbolising* and analogies with linguistic meaning are likely to be postponed until they become pressing. This may be at odds with the practice of many of those who write about film, especially those whose interests are more theoretical.


Of course, the parallel with face-to-face seeing cannot be perfect. In the cinematic case, issues of intention enter, and do so early. And this might alone seem to justify talk of symbolising, and to do so at a stage almost as early as anyone hoped. But is the mere manipulation of one's thoughts, experiences, and affective responses by another enough to justify the linguistic analogy? I am not sure. One goal of the book was to explore one kind of representation to show just how different, in some ways, it is from language. In the light of the conclusions thus reached, it seems to me that intention and representation are seen to be features of a genus only one species of which has been taken by some to be paradigmatic of the whole. Bearing this possibility in mind may not only help suggest different ways to conceive of cinema's most important features, but also encourage a different sense of quite which features are the important ones.


University of Birmingham, England





1. I am grateful to David Bell, and other members of the Sheffield visiting speaker seminar, for bringing this home to me.


2. For the distinction between figurative and non-figurative representation, and the difference between the latter and true abstraction, see Richard Wollheim _Painting as an Art_ (Thames and Hudson, 1987), chapter 2, section B.


3. See my 'Pictures and Beauty', _Proceedings of the Aristotelian Society_, vol. XCVIII part 2, 1997, pp. 177-94.


4. In 'Pictures and Beauty' I concede that my view of depiction does not explain how this engagement is possible. For all that, the view at least renders it comprehensible that such engagement might occur as, say, a *semantic* conception of picturing, a la Goodman, could not.



Copyright © _Film-Philosophy_ 1999


Robert Hopkins, 'Pictures and Film; Philosophy and the Empirical Disciplines: A Reply to Dean',  _Film-Philosophy_, vol. 3 no. 27, June 1999 <>.




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