What do pictures want?

Interview with W. J. T. Mitchell

Authors: Asbjørn Grønstad and Øyvind Vågnes

From the on-line magazine Image & Narrative, November 2006

• For this specific issue of Image and Narrative, it would be very interesting to hear your thoughts about the iconoclasm of a number of the grand narratives of cultural theory in recent years, not least against the background of your idea of "living images" in What Do Pictures Want? But let us begin with "the object," and return to the more sweeping claims of theory later. In What Do Pictures Want? you describe critical practice as a way of responding to a "resonant" object, and this made us think about Mieke Bal's description of the object that "talks back" in her Travelling Concepts of the Humanities (8-10). Bal calls for a "qualified return to ‘close reading' that has gone out of style" (10); in What Do Pictures Want?, you suggest that answers to the central questions of visuality "must be sought in the specific, concrete images that most conspicuously embody the anxiety over image-making and image-smashing in our time." We'd like you to comment on this, but perhaps you first could talk a little bit about what you in Picture Theory call the "metapicture," since that conceptualization made us think about Bal's notion of a "thinking" object in the first place?

— In Picture Theory I tried to distinguish three different kinds of metapictures: First, the picture that explicitly reflects on, or "doubles" itself, as in so many drawings by Saul Steinberg, in which the production of the picture we are seeing re-appears inside the picture. This is most routinely and literally seen in the effect of the "mise en abime," the Quaker Oats box that contains a picture of the Quaker Oats box, that contains yet another picture of a Quaker Oats box, and so on, to infinity. (Technically, I gather, the term first appeared in reference to heraldry, where the division of a coat of arms into increasingly diminutive sectors containing other coats of arms traces the evolution of a genealogy). Second, the picture that contains another picture of a different kind, and thus re-frames or recontextualizes the inner picture as "nested" inside of a larger, outer picture. Third, the picture that is framed, not inside another picture, but within a discourse that reflects on it as an exemplar of "picturality" as such. This third meaning implies, of course, that any picture whatsoever (a simple line-drawing of a face, a multi-stable image like the Duck-Rabbit, Velasquez's Las Meninas) can become a metapicture, a picture that is used to reflect on the nature of pictures.

The ever-present potentiality of the metapicture has several implications for the rest of your question. First, it suggests that any picture is at least potentially a kind of vortex or "black hole" that can "suck in" the consciousness of a beholder, and at the same time (and for the same reason) "spew out" an infinite series of reflections. This is not just a matter of the infinite or indefinite spatial depth that is suggested the moment a surface is marked and thus opened as a space for perception and reflective thought. It is also right there on the surface, in the infinity of aspects that a line or color or blurred erasure can provoke. As William Blake puts it, infinity is located in the "Definite & Determinate Identity" of the "bounding line," and not just the endless, empty space of perspective or the void of the unmarked space, the blankness or chaos of potential out of which images emerge. (Think here of Leonardo's advice to painters to look at the random splashes of mud left on plaster walls by passing carts, and to meditate on the forms of figures and landscapes that seem to emerge from them; or Nelson Goodman's notion of the "density" and "repleteness" of analog symbol systems).

Of course this infinity of potential aspects in a picture is rarely experienced. Most images pass by and through us so quickly that we scarcely notice them. They are fast food for the eyes, and mostly junk food. But some of them demand more attention, and even the trivial or overlooked ones have this potential waiting to be tapped. The approach I am proposing with the metapicture is thus quite compatible with Mieke Bal's appeal for a return to the "close reading" of images (though I'm sure she would want to interrogate the model of reading itself and raise the question of what we mean by reading, and whether the image is perhaps always opening up a threshold of the unreadable and even the indecipherable). My general pedagogical aim is to slow down the reception of the image, to encourage prolonged contemplation, second and third looks, reversals of perceptual fields such as figure/ground and surface/depth, and the Foucauldian strategy of suspending the rule of the "proper name" and nominative discourse over the image, as in his treatment of Las Meninas. I urge this practice, not (as is sometimes feared) because I have a magical or mystical view of images, but because I am seeking a clear-sighted analysis of the nature of pictures, one that is willing to explore its object with rigorous phenomenological or psychoanalytic or semiotic or socio-historical modes of interpretation. But I do not see any of these modes of analysis as a uniquely privileged metalanguage for the understanding of pictures. And the aim of the metapicture is to create a critical space in which images could function, not simply as illustrations or "examples" of the power of this or that method, but as "cases" that to some extent (generally unknown in advance) that might transform or deconstruct the method that is brought to them. The widest implication of the metapicture is that pictures might themselves be sites of theoretical discourse, not merely passive objects awaiting explanation by some non-pictorial (or iconoclastic) master-discourse. In relation to the domesticating tendencies of semiotics, for instance, with its taxonomies of signs and sign-functions, I like to think of the image as the "wild sign," the signifying entity that has the potential to explode signification, to open up the realm of nonsense, madness, randomness, anarchy, and even "nature" itself in the midst of the cultural labyrinth of second nature that human beings create around themselves. In What Do Pictures Want? I put this in terms of the following analogy (roughly paraphrased): "when it comes to images, then, we are in something like the position of savages who do not know where babies come from. We literally do not know where images come from, or where they go when (or even if) they die."

The metapicture, then, is also a figure that helps to explain the often-observed uncanniness of images, their ghostliness or spectrality, their tendency to look back at the beholder, or seemingly to respond to the presence of the beholder, to "want something" from the beholder. I don't think we can properly understand images without some reckoning with vitalism and animism. And I do not mean by this some kind of regressive return to primitive thought, but (as Levi-Strauss so often insisted) a taking account of the persistence of the "savage mind" at the dialectical heart of whatever we mean by the modern. I would also want to urge that we not see this exclusively in anthropomorphic terms, as if the vitalistic or animated character of the signs and symbols we create around us could be exhaustively described in terms of personification or prosopopoeia. Certainly, the conceit of the "desiring picture" or the "animated icon" may involve an analogy with human attributes, but the features of vitality, animation, and desire (at minimum, appetite) also permeate downward, into the animal and vegetable kingdoms. This is why, in What Do Pictures Want? I want to stress the non- or inhuman desires of images, and explore the neglected concept of totemism (with its emphasis on natural iconographies—plants, animals, and even minerals, including fossils, of course), in addition to the more familiar and anthropocentric concepts of fetishism and idolatry. My aim in What Do Pictures Want? is thus not to project personhood onto pictures, but to engage with what I call "the lives and loves" of images. So, while I like very much Mieke Bal's concept of "art that thinks," I don't want to begin with the assumption that it always thinks like us. The principles of vitalism and animism require that we also take account of what are sometimes called "lower" forms of consciousness—mere sentience, for instance, or sensuous awareness, responsiveness, as well as forms of memory and desire. What we call thinking (in images or in living things) goes deeper than philosophical reflection or self-consciousness. Animals remember. And most of human consciousness is pre- or unconscious. The nervous system is not the only system in our bodies that can learn. There is also the immune system, which learns to recognize and deal with an staggeringly large number of alien organisms in the life of any individual, and which works through a mechanism of copying, mimesis, and reproduction of antibodies that are symmetrical "twins" of the antigens they combat.

• Do you think of yours and Bal's alternative as symptomatic in any way for how things are turning around, with the increase of interdisciplinary work being done in the humanities?

— I hope they are more than symptomatic. My aim is to be diagnostic and (even more challenging) to create prognoses or interventionist strategies both in pedagogy and research. From the standpoint of disciplinarity, this means something more than the familiar invocation of "interdisciplinarity," which in my view is a bit too safe and predictable (I've argued this elsewhere in an essay entitled "Interdisciplinarity and Visual Culture"). I prefer a notion of image science and visual culture as sites of what I want to call "indisciplinarity," moments of breakage, failure, or deconstruction of existing disciplinary structures accompanied by the emergence of new formations (to some extent this is probably a reflection of my long-standing attraction to anarchist theories of knowledge, the sort pioneered by Paul Feyerabend). It is clear, to begin with, that images do not belong exclusively to any single discipline—not semiotics, or art history, or media studies, or even cultural studies (if it is a discipline). Their study compels us to be interdisciplinary at a bare minimum, just as paleontology requires that its researchers be geologists, biologists, anatomists, and artists.