What Is an Image?
There have been times when the question "What is an image?" was a matter of some urgency. In eighth- and ninth-century Byzantium, for instance, your answer would have immediately identified you as a partisan in the struggle between emperor and patriarch, as a radical iconoclast seeking to purify the church of idolatry, or a conservative iconophile seeking to preserve traditional liturgical practices. The conflict over the nature and use of icons, on the surface a dispute about fine points in religious ritual and the meaning of symbols, was actually, as Jaroslav Pelikan points out, "a social movement in disguise" that "used doctrinal vocabulary to rationalize an essentially political conflict." In mid-seventeenth-century England the connection between social movements, political causes, and the nature of imagery was, by contrast, quite undisguised. It is perhaps only a slight exaggeration to say that the English Civil War was fought over the issue of images, and not just the question of statues and other material symbols in religious ritual but less tangible matters such as the "idol" of monarchy and, beyond that, the "idols of the mind" that Reformation thinkers sought to purge in themselves and others.
If the stakes seem a bit lower in asking what images are today, it is not because they have lost their power over us, and certainly not because their nature is now clearly understood. It is a commonplace of modern cultural criticism that images have a power in our world undreamed of by the ancient idolaters. And it seems equally evident that the question of the nature of imagery has been second only to the problem of language in the evolution of modern criticism. If linguistics has its Saussure and Chomsky, iconology has its Panofsky and Gombrich. But the presence of these great synthesizers should not be taken as a sign that the riddles of language or imagery are finally about to be solved. The situation is precisely the reverse: language and imagery are no longer what they promised to be for critics and philosophers of the Enlightenment—perfect, transparent media through which reality may be represented to the understanding. For modern criticism, language and imagery have become enigmas, problems to be explained, prison-houses which lock the understanding away from the world. The commonplace of modern studies of images, in fact, is that they must be understood as a kind of language; instead of providing a transparent window on the world, images are now regarded as the sort of sign that presents a deceptive appearance of naturalness and transparence concealing an opaque, distorting, arbitrary mechanism of representation, a process of ideological mystification.
My purpose in this chapter is neither to advance the theoretical understanding of the image nor to add yet another critique of modern idolatry to the growing collection of iconoclastic polemics. My aim is rather to survey some of what Wittgenstein would call the "language games" that we play with the notion of images, and to suggest some questions about the historical forms of life that sustain those games. I don't propose, therefore, to produce a new or better definition of the essential nature of images, or even to examine any specific pictures or works of art. My procedure instead will be to examine some of the ways we use the word "image" in a number of institutionalized discourses—particularly literary criticism, art history, theology, and philosophy—and to criticize the ways each of these disciplines makes use of notions of imagery borrowed from its neighbors. My aim is to open up for inquiry the ways our "theoretical" understanding of imagery grounds itself in social and cultural practices, and in a history fundamental to our understanding not only of what images are but of what human nature is or might become. Images are not just a particular kind of sign, but something like an actor on the historical stage, a presence or character endowed with legendary status, a history that parallels and participates in the stories we tell ourselves about our own evolution from creatures "made in the image" of a creator, to creatures who make themselves and their world in their own image.
The Family of Images
Two things must immediately strike the notice of anyone who tries to take a general view of the phenomena called by the name of imagery. The first is simply the wide variety of things that go by this name. We speak of pictures, statues, optical illusions, maps, diagrams, dreams, hallucinations, spectacles, projections, poems, patterns, memories, and even ideas as images, and the sheer diversity of this list would seem to make any systematic, unified understanding impossible. The second thing that may strike us is that the calling of all these things by the name of "image" does not necessarily mean that they all have something in common. It might be better to begin by thinking of images as a far-flung family which has migrated in time and space and undergone profound mutations in the process.
If images are a family, however, it may be possible to construct some sense of their genealogy. If we begin by looking, not for some universal definition of the term, but at those places where images have differentiated themselves from one another on the basis of boundaries between different institutional discourses, we come up with a family tree something like the following:
Each branch of this family tree designates a type of imagery that is central to the discourse of some intellectual discipline: mental imagery belongs to psychology and epistemology; optical imagery to physics; graphic, sculptural, and architectural imagery to the art historian; verbal imagery to the literary critic; perceptual images occupy a kind of border region where physiologists, neurologists, psychologists, art historians, and students of optics find themselves collaborating with philosophers and literary critics. This is the region occupied by a number of strange creatures that haunt the border between physical and psychological accounts of imagery: the "species" or "sensible forms" which (according to Aristotle) emanate from objects and imprint themselves on the wax like receptacles of our senses like a signet ring; the fantasmata, which are revived versions of those impressions called up by the imagination in the absence of the objects that originally stimulated them; "sense data" or "percepts" which play a roughly analogous role in modern psychology; and finally, those "appearances" which (in common parlance) intrude between ourselves and reality, and which we so often refer to as "images"—from the image projected by a skilled actor, to those created for products and personages by experts in advertising and propaganda.
The history of optical theory abounds with these intermediate agencies that stand between us and the objects we perceive. Sometimes, as in the Platonic doctrine of "visual fire" and the atomistic theory eidola or simulacra, they are understood as material emanations from objects, subtle but nevertheless substantial images propagated by objects and forcibly impressing themselves on our senses. Sometimes the species are regarded as merely formal entities, without substance, propagated through an immaterial medium. And some theories even describe the transmission as moving in the other direction, from our eyes to the objects. Roger Bacon provides a good synthesis of the common assumptions of ancient optical theory:
Every efficient cause acts through its own power, which it exercises on the adjacent matter, as the light [lux] of the sun exercises its power on the air (which power is light [lumen] diffused through the whole world from the solar light [lux]). And this power is called "likeness," "image," and "species" and is designated by many other names.... This species produces every action in the world, for it acts on sense, on the intellect, and on all matter of the world for the generation of things.
It should be clear from Bacon's account that the image is not simply a particular kind of sign but a fundamental principle of what Michel Foucault would call "the order of things." The image is the general notion, ramified in various specific similitudes (convenientia, aemulatio, analogy, sympathy) that holds the world together with "figures of knowledge." Presiding over all the special cases of imagery, therefore, I locate a parent concept, the notion of the image "as such," the phenomenon whose appropriate institutional discourses are philosophy and theology.
Now each of these disciplines has produced a vast literature on the function of images in its own domain, a situation that tends to intimidate anyone who tries to take an overview of the problem. There are encouraging precedents in work that brings together different disciplines concerned with imagery, such as Gombrich's studies of pictorial imagery in terms of perception and optics, or Jean Hagstrum's inquiries into the sister arts of poetry and painting. In general, however, accounts of any one kind of image tend to relegate the others to the status of an unexamined "background" to the main subject. If there is a unified study of imagery, a coherent iconology, it threatens to behave, as Panofsky warned, "not like ethnology as opposed to ethnography, but like astrology as opposed to astrography." Discussions of poetic imagery generally rely on a theory of the mental image improvised out of the shreds of seventeenth-century notions of the mind; discussions of mental imagery depend in turn upon rather limited acquaintance with graphic imagery, often proceeding on the questionable assumption that there are certain kinds of images (photographs, mirror images) that provide a direct, unmediated copy of what they represent; optical analyses of mirror images resolutely ignore the question of what sort of creature is capable of using a mirror; and discussions of graphic images tend to be insulated by the parochialism of art history from excessive contact with the broader issues of theory or intellectual history. It would seem useful, therefore, to attempt an overview of the image that scrutinizes the boundary lines we draw between different kinds of images, and criticizes the assumptions which each of these disciplines makes about the nature of images in neighboring fields.
We clearly cannot talk about all these topics at once, so the next question is where to start. The general rule is to begin with the basic, obvious facts and to work from there into the dubious or problematic. We might start, then, by asking which members of the family of images are called by that name in a strict, proper, or literal sense, and which kinds involve some extended, figurative, or improper use of the term. It is hard to resist the conclusion that the image "proper" is the sort of thing we found on the left side of our tree-diagram, the graphic or optical representations we see displayed in an objective, publicly shareable space. We might want to argue about the status of certain special cases and ask whether abstract, nonrepresentational paintings, ornamental or structural designs, diagrams and graphs are properly understood as images. But whatever borderline cases we might wish to consider, it seems fair to say that we have a rough idea about what images are in the literal sense of the word. And along with this rough idea goes a sense that other uses of the word are figurative and improper.
The mental and verbal images on the right side of our diagram, for instance, would seem to be images only in some doubtful, metaphoric sense. People may report experiencing images in their heads while reading or dreaming, but we have only their word for this; there is no way (so the argument goes) to check up on this objectively. And even if we trust the reports of mental imagery, it seems clear that they must be different from real, material pictures. Mental images don't seem to be stable and permanent the way real images are, and they vary from one person to the next: if I say "green," some listeners may see green in their mind's eye, but some may see a word, or nothing at all. And mental images don't seem to be exclusively visual the way real pictures are; they involve all the senses. Verbal imagery, moreover, can involve all the senses, or it may involve no sensory component at all, sometimes suggesting nothing more than a recurrent abstract idea like justice or grace or evil. It is no wonder that literary scholars get very nervous when people start taking the notion of verbal imagery too literally. And it is hardly surprising that one of the main thrusts of modern psychology and philosophy has been to discredit the notions of both mental and verbal imagery.
Eventually I will argue that all three of these commonplace contrasts between images "proper" and their illegitimate offspring are suspect. That is, I hope to show that, contrary to common belief, images "proper" are not stable, static, or permanent in any metaphysical sense; they are not perceived in the same way by viewers any more than are dream images; and they are not exclusively visual in any important way, but involve multisensory apprehension and interpretation. Real, proper images have more in common with their bastard children than they might like to admit. But for the moment let us take these proprieties at face value, and examine the genealogy of those illegitimate notions, images in the mind and images in language.
The Mental Image: A Wittgensteinian Critique
Now for the thinking soul images take the place of direct perceptions; and when it asserts or denies that they are good or bad, it avoids or pursues them. Hence the soul never thinks without a mental image. Aristotle, De Anima 111.7.431a
A notion with the entrenched authority of three hundred years of institutionalized research and speculation behind it is not going to give up without a struggle. Mental imagery has been a central feature of theories of the mind at least since Aristotle's De Anima, and it continues to be a cornerstone of psychoanalysis, experimental studies of perception, and popular folk-beliefs about the mind. The status of mental representation in general, and the mental image in particular, has been one of the main battlegrounds of modern theories of the mind. A good index of the strengths on both sides of this issue is the fact that the most formidable critic of mental imagery in our time developed a "picture theory" of meaning as the keystone of his early work, and then spent the rest of his life fighting against the influence of his own theory, trying to expel the notion of mental imagery along with all its metaphysical baggage.
Wittgenstein's way of attacking mental imagery is not, however, the direct strategy of denying the existence of such images. He freely concedes that we may have mental images associated with thought or speech, insisting only that these images should not be thought of as private, metaphysical, immaterial entities any more than real images are. Wittgenstein's tactic is to demystify the mental image by bringing it right out in the open where we can see it: "Mental images of colours, shapes, sounds, etc., etc., which play a role in communication by means of language we put in the same category with patches of color actually seen, sounds heard." It is a bit hard, however, to see how we can put mental and physical images "in the same category." We certainly can't do it by cutting open someone's head to compare mental pictures with the ones on our walls. A better strategy, and more in the Wittgensteinian spirit, would be to examine the ways we put those images "into our heads" in the first place by trying to picture the sort of world in which this move would make sense. I offer the figure on the next page as just such a picture.
The figure should be read as a palimpsest displaying three overlapping relationships: (1) between a real object (the candle on the left) and a reflected, projected, or depicted image of that object; (2) between a real object and a mental image in a mind conceived (as in Aristotle, Hobbes, Locke, or Hume) as a mirror, camera obscura, or a surface for drawing or printing; (3) between a material image and a mental one. (It may help here to imagine the diagram as three overlapping transparencies, the first showing just the two candles, the left one real, the right one an image; the second adding the humanhead to show the mental introjection of the depicted or reflected candle; the third adding the frame around the "real" candle to make it mirror the imaginary status of the candle on the right. I assume, for simplicity, that all optical inversions have been rectified.) What the diagram displays as a whole is the matrix of analogies (particularly ocular metaphors) that govern representational theories of the mind. In particular it shows how the classic divisions of Western metaphysics (mind-matter, subject-object) translate into a model of representation, the relation between visual images and the objects they stand for. Consciousness itself is understood as an activity of pictorial production, reproduction, and representation governed by mechanisms such as lenses, receptive surfaces, and agencies for printing, impressing, or leaving traces on these surfaces.