When W.J.T. Mitchell flexes his intellectual muscles, people stop to look. Editor of one of the U.S.'s most prestigious cultural journals, Critical Inquiry, and professor in the art history and literature departments at the University of Chicago, Mitchell began his career in the relatively parochial field of William Blake studies, but went on to become one of the most prominent contemporary cultural theorists. He is widely associated with an approach to studying images known as "visual culture," and at least two of his previous works, Iconology: Image, Text, Ideology (Chicago, University of Chicago Press, 1987) and Picture Theory: Essays on Verbal and Visual Representation (Chicago, University of Chicago Press, 1995) are essential reference points for understanding this movement. Mitchell's project has been nothing less than the creation of a method for understanding, as he puts it on the second page of his Iconology, "the ways that images in the strict or literal sense (pictures, statues, works of art) are related to notions such as mental imagery, verbal or literary imagery, and the concept of man as an image maker and maker of images," an adventure that has taken him into the more intractable realms of contemporary philosophy, and led him to write about everything from post-colonialism to landscape painting to dinosaurs.

Because of all this, Mitchell's just-published book, What Do Pictures Want?, is important to consider, not just for those interested in understanding the scholarly issues that it broaches, but because it represents a whole sensibility towards images. Stylistically, Mitchell's wide-ranging, theory-heavy criticism is what a whole generation of art writers aspire to, admiring the way that the framework of "visual culture" offers the prestige of moving beyond narrow issues of taste to make big, decisive philosophical pronouncements about the state of the world. It also helps that What Do Pictures Want? has a punchy topic: Mitchell's goal is to investigate the way that people impute human emotions and desires to images, and thus his book promises something juicy -- to lay bare all the traces of mysticism lurking beneath supposedly blas contemporary attitudes about art.

Mitchell's essential thoughts on this matter are laid out in the volume's second chapter (the text of which was originally published in a shorter version called "What Do Pictures Really Want?" in October magazine.) Here, he charts the persistence of "primitive" ideas about the magical power of images in modern ideas of fetishism, including Freud's erotically invested objects and Marx's ideas about the way commodities function increasingly as if they had a life of their own both themes that are relevant to the concerns of the hyper-transgressive, hyper-commercialized world of contemporary art.

True, the provocative wackiness of Mitchell's claim that we should, essentially, consider pictures as living things he begins the chapter by almost begging his reader to accuse him of an outright "subjectivization of images, a dubious personification of inanimate objects" (28) is somewhat deflated when one realizes that this is just a fancy path by which to arrive at the commonplace that we live in a world where images have power over people's lives (Mitchell repeatedly references Andre Agassi's statement that "Image is everything" as evidence of our continued "animism.") However, the framework does allow Mitchell to make some nice points, as when he compares the fashionable critical practice of "ideological critique" to the superstition that if you damage the image of something you are harming the real thing: "[T]he critical exposure and demolition of the nefarious power of images is both easy and ineffectual," Mitchell writes. "Pictures are a popular political antagonist because one can take a tough stand on them, and yet, at the end of the day, everything remains pretty much the same" (33).

Unfortunately, Mitchell's investigation into the ways that people treat images as if they were alive is not really sustained over the course of What Do Pictures Want? Partly, this is a matter of an ambiguity about the work's design. He begins the book by claiming that it is an integrated vision of the topic, divided into three sections, each investigating one of what Mitchell claims to be the three fundamental elements of pictures -- "images," "objects" and "media." But in fact, the tripartite scheme seems to have been introduced after the fact to give structure to an otherwise diverse collection of essays -- more than half of the book is composed of material that was created for symposia or catalogues, and thus each new chapter seems to respond to a different, relatively specific concern.

All in all, the theme of the "lives and loves of images" pops up only sporadically or impressionistically in most of the essays. At one point, Mitchell spends a chapter discussing the issues that Beat photographer Robert Frank's book, The Americans, raises about national identity. The elaborate argument focuses on the theme of decapitation, and is encapsulated in Mitchell's discussion of Parade -- Hoboken, New Jersey, a photo that features two women with obscured faces: "What Frank recognizes here is that the heads must be 'cut off,' as it were, to allow the women's bodies to have any chance of becoming 'pupils' in the facade of patriotic spectatorship" (284). It is only at the very end of this long essay that suddenly we are returned to the supposed theme of the book as a whole, with a mention to how Frank abandoned photography for film because, he said, making films was more like "being among friends" than taking photos (293). It is as if this reference was introduced as an afterthought to make the Frank piece fit in with the rest of the material -- and this lack of focus is the rule rather than the exception in the book.

But the disjointed feeling of What Do Pictures Want? is also determined by Mitchell's theoretical approach itself. His method is fantastically macroscopic: "If visual culture is to mean anything," he cautions, "it has to be generalized as the study of all the social practices of human visuality" (349). At this level, of course, everything connects to everything, something that makes natural a disregard for the kinds of art historical categories that might normally serve to structure an argument. Within a single essay, Mitchell will leap between Byzantine icons, Velzquez, an Al Jolson movie poster, the image of Uncle Sam, Chardin and Barbara Kruger. It should be no surprise, then, that the book as a whole skips around a lot, from object to object and issue to issue. Images are swept up from their specific places and re-threaded together by a wild, swirling flow of references to Derrida and Deleuze, McLuhan and Luhmann.

The common and conservative complaint about "theoretical" criticism is that somehow, in all its panting about the "issues" raised by works, it forgets everything that is of interest about art in the first place. But this is not really my feeling. I think a little bit of philosophy can help artists and critics to really think about what they are dealing with I'm just not certain that this is how theory is being used in Mitchell's book. In the course of a chapter-length reflection on the disastrous "Sensation" show at the Brooklyn Museum, Mitchell indulges in a detailed analysis of the Biblical narrative of the golden calf and brings in references to among others -- Walter Benjamin, Clement Greenberg, Jacques Lacan, G.E. Lessing, Karl Marx, Laura Mulvey and himself. But the outcome of all this is the conclusion that "[p]eople are afraid of images. Images make us anxious" (141). Similarly, Mitchell spends more than 20 pages ruminating on the theoretical status of abstract painting after modernism helpfully providing us an image of Lacan's "lozenge" figure to unlock the motif of one painting in order to surprise us with the notion that the art object today "is still important, but perhaps not quite so self-important" (233).

In these instances, are the theoretical references really being used to enrich our understanding of images, or do they serve as a sort of short hand for "sophisticated criticism," spicing up what is really pretty conventional stuff? The names of theorists drop so fast and thick in What Do Pictures Want? that the discussion of them doesn't seem to have much depth, and most of the time what seems to be invoked is some kind of received wisdom about a figure rather than a complex insight about a body of thought. To pick one example from among many, Mitchell's introduction of Lacan as a thinker who is limited to the idea that "[d]esire as longing produces fantasies, evanescent specular images that continuously tease and elude the beholder" (66) is easily refuted by reference to any late work by the great French thinker.

Time and again, what one gets seems to be a fairly pedestrian insight dressed up in elaborate theoretical drag. Despite all the references to "scopic regimes" and Mitchell's tough-talking stance against "unreconstructed modernists," I don't think I could distinguish his discussions of Antony Gormley and Robert Frank from conventional "great man"-style celebrations of genius, if you took away all the references to Judith Butler and Roland Barthes. And Mitchell's chapter on the ways artists have responded to issues raised by genetic engineering which bears the groan-inducing title "The Work of Art in the Age of Biocybernetic Reproduction" is an example of the use of "theory" at its most glib, harvesting the canon of ideas for motifs that are then simply pinned onto art objects to validate them, as if simply parroting the concerns of a particular thinker made an artist interesting. Not every conversation one has about genetic engineering is equally illuminating, so why should every artwork be?

Finally, there's a funny kind of overdetermination at work in this book. Mitchell sets out to produce a theory about how we relate to pictures in a way that makes them seem to take on magical powers, abstracted from their real substances. But this very theory, humorously enough, is the best description of the way Mitchell approaches "theory" itself, reducing various thinkers to conventionalized figures that are invoked ritualistically, as if their names had totemic power to convey depth and seriousness onto his considerations. Ironically, then, Mitchell's own impressive success as a thinker may be the best proof that he is onto something with his core claim that even the most refined corners of contemporary culture are still alive with superstition. Unfortunately, in his own hands, the formulation goes to waste, yielding a strange hybrid art theory that is neither particularly about art nor, at the end of it all, particularly satisfying as theory.