Image and Word – W. J. T. Mitchell

The recognition that pictorial images are inevitably conventional and contaminated by language need not cast us into an abyss of infinitely regressive signifiers. What it does imply for the study of art is simply that something like the Renaissance notion of ut pictura poesis and the sisterhood of the arts is always with us. The dialectic of word and image seems to be a constant in the fabric of signs that a culture weaves around itself. What varies is the precise nature of the weave, the relation of warp and woof. The history of culture is in part the story of a protracted struggle for dominance between pictorial and linguistic signs, each claiming for itself certain proprietary rights on a ‘nature’ to which only it has access. At some moments this struggle seems to settle into a relationship of free exchange along open borders; at other times (as in Lessing’s Laocoon) the borders are closed and a separate peace is declared. Among the most interesting and complex versions of this struggle is what might be called the relationship of subversion, in which language or imagery looks into its own heart and finds lurking there its opposite number. One version of this relation has haunted the philosophy of language since the rise of empiricism, the suspicion that beneath words, beneath ideas, the ultimate reference in the mind is the image, the impression of outward experience printed, painted, or reflected in the surface of consciousness. It was this subversive image that Wittgenstein sought to expel from language, which the behaviorists sought to purge from psychology, and which contemporary art-theorists have sought to cast out of pictorial representation itself. The modern pictorial image, like the ancient notion of ‘likeness,’ is at last revealed to be linguistic in its inner workings.

Why do we have this compulsion to conceive of the relation between words and images in political terms, as a struggle for territory, a contest of rival ideologies? … a short answer may be provided here: the relationship between words and images reflects, within the realm of representation, signification, and communication, the relations we posit between symbols and the world, signs and their meanings. We imagine the gulf between words and images to be as wide as the one between words and things, between (in the largest sense) culture and nature. The image is the sign that pretends not to be a sign, masquerading as (or, for the believer, actually achieving) natural immediacy and presence. The word is its ‘other,’ the artificial, arbitrary production of human will that disrupts natural presence by introducing unnatural elements into the world – time, consciousness, history, and the alienating intervention of symbolic mediation. Versions of this gap reappear in the distinctions we apply to each type of sign in its own turn. There is the natural, mimetic image, which looks like or ‘captures’ what it represents, and its pictorial rival, the artificial, expressive image which cannot ‘look like’ what it represents because that thing can only be conveyed in words. There is the word which is a natural image of what it means (as in onomatopoeia) and the word as arbitrary signifier. And there is the split in written language between ‘natural’ writing by pictures of objects, and the arbitrary signs of hieroglyphics and the phonetic alphabet.

What are we to make of this contest between the interests of verbal and pictorial representation? I propose that we historicize it, and treat it, not as a matter for peaceful settlement under the terms of some all-embracing theory of signs, but as a struggle that carries the fundamental contradictions of our culture into the heart of theoretical discourse itself. The point, then, is not to heal the split between words and images, but to see what interests and powers it serves. This view can only be had, of course, from a standpoint which begins with skepticism about the adequacy of any particular theory of the relation of words and images, but which also preserves an intuitive conviction that there is some difference that is fundamental. It seems to me that Lessing, for instance, is absolutely right insofar as he regards poetry and painting as radically different modes of representation, but that his ‘mistake’ (which theory still participates in) is the reification of this difference in terms of analogous oppositions like nature and culture, space and time.

What sorts of analogies would be less reified, less mystifying, more appropriate as a basis for historical criticism of the word-image difference? One model might be the relation between two different languages that have a long history of interaction and mutual translation. This analogy is, of course, far from perfect. It immediately loads the case in favor of language, and it minimizes the difficulties in making connections between words and images. We know how to connect English and French literature more precisely than we do English literature and English painting. The other analogy which offers itself is the relationship between algebra and geometry, the one working by arbitrary phonetic signs read progressively, the other displaying equally arbitrary figures in space. The attraction of this analogy is that it looks rather like the relation of word and image in an illustrated text, and the relation between the two modes is a complex one of mutual translation, interpretation, illustration, and embellishment. The problem with the analogy is that it is too perfect: it seems to hold out an impossible ideal of systematic, rule-governed translation between word and image. Sometimes an impossible ideal can be useful, however, so long as we recognize its impossibility. The advantage of the mathematical model is that it suggests the interpretive and representational complementarity of word and image, the way in which the understanding of one seems inevitably to appeal to the other.

In the modern era the main direction of this appeal would seem to be from the image, conceived as a manifest, surface content or ‘material/ to the word, conceived as the latent, hidden meaning lying behind the pictorial surface. In The Interpretation of Dreams Freud comments on ‘the incapacity of dreams’ to express logical, verbal connections and latent dream-thoughts by comparing ‘the psychical material out of which dreams are made’ to the material of visual art:

The plastic arts of painting and sculpture labour, indeed, under a similar limitation as compared with poetry, which can make use of speech; and here once again the reason for their incapacity lies in the nature of the material which these two forms of art manipulate in their effort to express something. Before painting became acquainted with the laws of expression by which it is governed, it made attempts to get over this handicap. In ancient paintings small labels were hung from the mouths of the persons represented, containing in written characters the speeches which the artist despaired of representing pictorially.

For Freud, psychoanalysis is a science of the ‘laws of expression’ that govern the interpretation of the mute image. Whether that image is projected in dreams or in the scenes of everyday life, analysis provides the method for extracting the hidden verbal message from the misleading and inarticulate pictorial surface. But we have to remind ourselves that there is a countertradition which conceives of interpretation as going in just the opposite direction, from a verbal surface to the ‘vision’ that lies behind it, from the proposition to the ‘picture in logical space’ that gives it sense, from the linear recitation of the text to the ‘structures’ or ‘forms’ that control its order. The recognition that these ‘pictures’ which Wittgenstein found residing in language are no more natural, automatic, or necessary than any other sorts of images we produce may put us in a position to make use of them in a less mystified way. Chief among these uses would be, on the one hand, a renewed respect for the eloquence of images and, on the other hand, a renewed faith in the perspicuousness of language, a sense that discourse does project worlds and states of affairs that can be pictured concretely and tested against other representations. Perhaps the redemption of the imagination lies in accepting the fact that we create much of our world out of the dialogue between verbal and pictorial representations, and that our task is not to renounce this dialogue in favor of a direct assault on nature but to see that nature alreadv informs both sides of the conversation.

 J. T. Mitchell – Iconology, Chicago and London, 1986