Joel Snyder & Neil Walsh Allen, ‘Photography, Vision, and Representation’, pp.143-169 in:
Critical Inquiry, Vol.2, Autumn 1975.
Is there anything peculiarly “photographic” about photography – something which sets it apart from all other ways of making pictures?
[…] for most of this century the majority of critics and laymen alike have tended to answer these questions in the same way: that photographs and paintings differ in an important way and require different methods in interpretation precisely because photographs and paintings come into being in different ways.
Our purpose here is not to show that these answers are always, in every case, wrong, or that there is never anything to be gained by dif- ferentiating between photography and other visual arts, or that ques- tions about how photographs come into being are never appropriate to any investigation.
But we do want to suggest that they are a very small part of the story and that they have been supported by definitions of the “nature” of photography and the way it works which are misleading at the very least, and are more often quite simply wrong.
Although Emerson’s program of “naturalism” was not universally accepted, the basic notion that photographs were representations, and should be understood and judged as were other kinds of representations, was quite widespread.
The consensus of modern critics is that photographs differ from other sorts of representations in a fundamental way and that special theoretical principles and critical standards are necessary to account for them.
[Citing Stanley Cavell:]
“Photography overcame subjectivity in a way undreamed of by painting, one which does not so much defeat the act of painting as escape it altogether: by automatism, by removing the human agent from the act of reproduction.”
Photographs are not simply different from other kinds of pictorial representation in certain detailed respects; on the contrary, photographs are not really representations at all. They are the practical realization of the general artistic ideal of objectivity and detachment.
Stated in the most general terms, the modern position is that in photography there are certain necessary connections between a photograph and its “real life” original which simply do not (and perhaps cannot) exist in the “traditional” arts. But just what are these guarantees, and how important are they for our understanding of photographs?
We will examine the way one modern theorist, Rudolf Arnheim, has recently formulated answers to these questions.
[…] it should be emphasized that Arnheim is not advancing anything very new or different here. Siegfried Kracauer and Andre Bazin, two critics whom Arnheim cites with approval, have advanced rather similar positions, and writers as diverse as Etienne Gilson, R. G. Collingwood, Stanley Cavell, William Ivins, and E. H. Gombrich have all used photography as a benchmark of “pictorial fact” against which to measure more traditional pictorial media.
Arnheim describes himself as a “media analyst,” not a critic, but despite this disclaimer, we must point out that he makes criticism of photographs difficult if not impossible. And since his views are held by critics and their audience as well, it is not surprising that there is very little intelligent criticism of photography.
We are told that when we look at a photograph we are on “a vacation from artifice” – but should we be on vacation? We are told that a photograph is a “coproduction of nature and man” – but is this coproduction along the lines of Michelangelo and a piece of marble, or a geneticist and breeds of corn, or some other sort of coproduction altogether?
It is odd that modern critics who believe that the photographic process should be the starting point for criticism have had very little to say about what the process is, how it works, and what it does and doesn’t guarantee.
Aside from the simple notion of automatism, two models of how photography works have been used, or at least assumed.
One of these, which we will call the “visual”model, stresses the supposed similar- ity between the camera and the eye as optical systems, and posits that a photograph shows us (or ought to show us) “what we would have seen if we had been there ourselves.”
The other version of how photography works we will call the “mechanical” model. It stresses the necessary and mechanical connections which exist between what we see in a photo- graph and what was in front of the camera. According to this model, a photograph may not show us a scene as we ourselves would have seen it, but it is a reliable index of what was.
Writers on photography have often treated these models as though they were identical, or as though one were contained within the other, but this is not the case, and such assumptions gloss over the basic challenge to any theory which attempts to find the meaning of photographic images by referring to their origins – the challenge of extracting pictorial meaning from the operation of natural laws.
It is difficult to find a term describing the relation between what was in front of the camera and the image which does not predetermine the results of any investigation into that relation. We will use the words “characterization” and “characterize” to describe both individual steps and aspects of the photographic process (a given lens characterizes things in greater or lesser detail than another lens) and also the end result of the process (a photograph is a characterization of something).
A photographer – even a Sunday snapshooter – makes a number of characterizations by his choice of equipment and how he uses it.
He may not consider each one in detail on every occasion, and in a simple snapshot camera (and subsequently, in drug-store processing) these characterizations are “built in” – but they are still there.
The position of the camera effects further characterizations; once again, this holds true whether the position of the camera is carefully planned by the photographer, or whether the camera goes off by accident when dropped, or whether the camera is built into a booth and goes off automatically when people feed coins into a slot.
It is the light reflected by the objects and refracted by the lens which is the agent in the process, not “the physical objects themselves.” These “physical objects” do not have a single “image”-“their image”-but, rather, the camera can manipulate the reflected light to create an infinite number of images.
An image is simply not a property which things naturally possess in addition to possessing size and weight. The image is a crafted, not a natural, thing. It is created out of natural material (light), and it is crafted in accordance with, or at least not in contravention of, “natural” laws. This is not surprising. Nor is it surprising that something in the camera’s field will be represented in the image; but how it will be represented is neither natural nor necessary.
What we have called the “visual”model of the photographic process is another way of trying to flesh out the bare bones of photography’s alleged “intimate involvement” with “physical reality.” No doubt this model originated in, and retains its plausibility because of, the supposed resemblance of the human eye with its lens and retina to the camera with its lens and film. But once this resemblance has been stated, the model fails to establish anything further.
A photograph shows us “what we would have seen” at a certain moment in time, from a certain vantage point if we kept our head immobile and closed one eye and if we saw with the equivalent of a 150-mm or 24-mm lens and if we saw things in Agfacolor or in Tri-X developed in D-76 and printed on Kodabromide #3 paper. By the time all the conditions are added up, the original position has been reversed: instead of saying that the camera shows us what our eyes would see, we are now positing the rather unilluminating proposition that, if our vision worked like photography, then we would see things the way a camera does.
The camera-eye analogy is no more helpful for people investigating human vision than it is for the investigator of photographs. The more the supposed analogy is investigated, the more convincing becomes the conclusion that we do not possess, receive, or even “make” an image of things when we see-that there is nothing corresponding to a photographic image formed in one place which is then inspected or interpreted.
Images are indeed formed on the retina of the eye, but they do not answer functionally to the image at the film plane of a camera. In the living, active eye, there is nothing that can be identified as the retinal image, meaning by that a persisting image that is resolved on one definite topographical portion of the retina.
Rather, the image is kept in constant involuntary motion: the eyeball moves, the image drifts away from the fovea and is “flicked” back, while the drifting movement itself vibrates at up to 150 cycles per second.10 Amidst all this motion, is there one privileged image to set beside a photograph for comparison? At the material level (the level at which arguments about photography are usually pitched), the two processes are simply incommensurate.
The problems of photographing “what we see” are substantial, and the solutions only partial, when “what we see” consists of stationary dry goods. When we turn to the problems of photographing things that move or even might move, the visual model breaks down completely.
[Ref photographing running horses]
To the extent that the mechanical model holds that a camera is a certain sort of extension or expansion of our normal visual experience along certain lines, it seems quite plausible. For instance, when we see a horse “frozen” in mid-gallop in a photograph, we have no reason to doubt that, at a certain moment, the horse “really”assumed that posture. Here we are simply extending and modifying the notion that the camera is an eye.
But the blurred horse we see in a time exposure is another matter. Here we do not assume that the galloping horse ever “really” became a blur at all. We assume, instead, that the horse “really” galloped and that this galloping plus perhaps the movement of the camera and the peculiarities of the film resulted in the horse being characterized as an equine blur. Thus the photograph is not a substitute for vision, not even a modified or extended form of vision, but simply the inevitable outcome of a certain series of events.
No doubt it is this sense of inevitability, this feeling that a photograph is the end result of a series of cause-and-effect operations performed upon “physical reality,” that inclines us to impute a special sort of veracity to photograph – an “authenticity from which painting is barred by birth,” to use Arnheim’s phrase.
No matter how great the photographer’s range of controls, no matter how labyrinthine the path from scene to image, one can always find mechanical connections between the two. The question is whether these mechanical connections are really important to us when we look at and try to understand the final picture.
[Ref the ‘photofinish’ photograph]
We are accustomed, when we see five horses occupying five different positions in a photograph to think that we are looking at a picture of five horses that were all in different places at the same time. In a photofinish, we see five horses that were at the same place at different times. When we look at the nose and tail of a single horse in the picture, we are still looking at things which were recorded as they occupied the same place at different times. As we move from left to right across the picture, we are not looking at distance, but at time.
There is no doubt that the photofinish is an accurate characterization of the finishing order of horses in a race and no doubt that this accuracy derives from the mechanisms of the camera, laws of optics and chemistry, and so on. But the way in which the picture is made has little to do with the way we normally interpret it. The photofinish looks like a snapshot taken at the end of a race, and no amount of knowledge about photofinish cameras can supplant this interpretation with another one.
The mechanical relations which guarantee the validity of the photograph as an index of a certain kind of truth have been almost completely severed from the creation of visual likeness. In infra-red and ultra-violet color photography, visible colors are arbitrarily assigned to invisible bands of the spectrum. In color Schlieren photography used to analyze motions in gases and liquids, colors are arbitrarily assigned to directions, and no surgeon expects to find anything resembling an X-ray when he opens up a body.
In all these cases, the picture is valuable as an index of truth only to the extent that the process by which it was made is stated explicitly, and the pictures can be interpreted accurately only by people who have learned how to interpret them.
The mechanical model, by explaining everything, ends up explaining nothing. In practice, the mechanical workings of the photographic process must constantly be regulated by a set of rules for making “acceptable” pictures, and simple mechanical procedures must be augmented by additional processes to produce a number of different degrees and kinds of acceptability.
If “automatism” and both the visual and mechanical models of photography explain so little of how photography works, why are they advanced?
Of course once a theorist has defined photography as being non-manipulative, nonimaginative, and noninventive (in a literal sense), and has defined “art”as being manipulative, inventive, and imaginative, the distinction between the two becomes relatively clear.
As far as principles of aesthetics go, John Szarkowski has gone further than many other writers by stating explicitly the theory of art that separates photography from “handmade” representations: “most of the literature of art history is based on the assumption that the subject exists independent of, and prior to, the picture.”
When Szarkowski says of a photograph by Harry Callahan that the subject “is not the figure, or the room, or the shape and graphic weight of the light window against the dark ground, but every element within the frame, and their precisely just relationship,” he is hardly revolutionizing the aesthetics of the visual arts.
Now it would be quite correct to point out, á la Gombrich, that there is nothing in photography that corresponds exactly with the schemata used by many artists of other eras as aids in making representations of individual objects.
However, formulas and standardized procedures of representation are certainly not lacking in photography, especially in those kinds of photography often thought to be simple, straightforward “documents.” The passport photograph and the police “mug shot” are each produced by formulas regulating choice of lens, framing, and lighting.
Similar manuals exist to instruct commercial photographers in the methods appropriate for architecture, family groups, silverware, and glassware. In addition, there are the “built-in” formulas of the snap- shot camera, designed for “typical”snapshot subjects and to compensate for the amateur’s problems with focus, exposure, framing, and holding still.
Even in the realm of serious and inventive photography there is no clear-cut break with older traditions of representation. Genres such as portraiture and landscape have been appropriated, expanded, and redefined, and new genres and subgenres have been created. Furthermore, photographers have relied upon conventions and habits of pictorial interpretation (both by confirming conventional expectations and by deliberately frustrating them), have created new conventions of their own, and have borrowed other conventions from the nonvisual arts.
Thus to formulate a set of critical principles for photography based on what is purely or uniquely or essentially photographic is as absurd and unprofitable as would be the adoption in its place of standards taken from a mummified canon of nineteenth-century painting.
[On a photograph of James Dean by Dennis Stock]
We will not try to show that this is a great or even a very good photograph, much less to establish that it is better than, or at least as good as, some comparable painting or drawing. But we will try to show that it is capable of being analyzed as a picture “made and controlled by man.”
There is no doubt that Stock made a number of choices in the course of producing this photograph, but it is difficult to imagine that he calmly evaluated every possible photograph that might have been taken and chose this one as the best.
[Stock] knew what kind of picture he was after: one that would show something of Dean’s character from his reaction to his surroundings. He knew how his audience might react to various arrangements of figures with one another, with other objects, and within the space of the overall picture. He also knew what his camera, lens, and film would do under all sorts of circumstances. To this must be added Stock’s own sensibility, his ideas of what sorts of pictures were worth making.
Considerations of this sort are available to every photographer, although how they are employed in creating photographs seems to vary greatly. Some photographers “previsualize” every detail of the negative and print before tripping the shutter. Others may have a number of nebulous possibilities in mind which take on more specific form as they shoot a number of exposures, and their expectations may take on final form only when they “discover” their picture on a contact sheet.
But certainly there is nothing new here: artists have long sought out favorite bits of countryside, hired favorite models, returned time and again to congenial themes or restricted them- selves to one or two genres. The limitations on visual artists, including photographers, are usually self-imposed or imposed by a lack of invention or a lack of representational schemes and programs for translating ideas into pictures. So if we find some fault with Stock’s picture, we needn’t let him off the hook by saying that Nature hasn’t done her share, or trying to see things as mechanical deposits of light, or reminding ourselves that, after all, it’s only a photograph.
Does this picture have any special status by virtue of the fact that it was made by a camera rather than by hand? One is tempted to say that it does, that it establishes certain facts about James Dean – that, at the very least, he once stood next to the grave of Cal Dean. But even this minimal statement is not incorrigible. We might be challenged to prove that it was indeed James Dean, not a look-alike, or that this is a real grave, not a stage set, or that Cal and James Dean were related.
Of course our knowledge about the real James Dean – that he died young or that he played a character named Cal in East of Eden – may add a good deal of poignancy to this photograph. This sort of thing happens all the time, regardless of medium and even regardless of “the facts.”
[…] the variety of critical approaches (of which we may think some to be valuable and others to be wrong-headed) provides us with a variety of ways to assess the merit or lack of merit of this and other photographs. Just as important, this variety provides us with a number of ways of defining just what this photograph is, both in itself, and as the cause of a variety of effects and as the effect of a variety of causes.
The documentary value of a photograph is not determined solely by Arnheim’s questions of “authenticity,” “correctness,” and “truth.” We can also ask what it means, who made it, for whom was it made, and why it was made in the way it was made. These questions are asked of other “documents,” ranging from Minoan warehouse receipts to great works of art.
The poverty of photographic criticism is well known. It stands out against the richness of photographic production and invention, the widespread use and enjoyment of photographs, and even the popularity of photography as a hobby. To end this poverty we do not need more philosophizing about photographs and reality, or yet another (this time definitive) definition of “photographic seeing,” or yet another distillation of photography’s essence or nature. The tools for making sense of photographs lie at hand, and we can invent more if and when we really need them.