But never the same poem.

One hundred and fifty years ago it was impossible to take a course in the novel. Novels were softheaded, not suitable for serious study. Fifty years ago you couldn't take a course in film. Movies as they were called were softheaded, not suitable for serious study. Ten years ago you couldn't study the comic book. Today you can't major in video games. But you can apply to a school where you'll be able to. DigiPen, an accredited college, will open next year in Seattle offering a four-year degree in creating video games. Administrators expect 10,000 applicants to compete for 100 slots.

The push of how we think — especially a mindset that puts genres and media in discourse — and the pull of writing technology help bring word and image together. Using the web as a glorified slide projector is probably its crudest beginning. For example, using "real time interactive simulation," i.e., the video-game skills students at DigiPen will develop, future readers of Perseus will be able to walk through a virtual Acropolis, either the archeological site that exists today, or a reconstruction similar to the prototypes of Pompeii that have been created.

But the technical feasibility to incorporate images or sound into a book doesn't provide an aesthetic reason for doing so.

"No not that!" Kafka screamed when his publisher suggested an illustration of the bug in Metamorphosis. "Anything but that!" — a writer's typical reaction to that which is always a limitation: embodiment — the kind of limitation first loosened by that early revolution in writing technology, the alphabet. By freeing the word from the corpse of the mimetic pictogram, a greater degree of abstraction became possible and with it a more fluid literature.

Is Kafka's bug the Other in Vienna? The terminally ill among the healthy? All and more of the above? Or is it the blob from the pen of some sci-fi illustrator? That is, if all the techno jargon is stripped from Wired magazine, the championing of Virtual Reality sounds a lot like the championing of plain old realism, the kind that Kafka (and the alphabet) work against, while the VR experience itself sounds a lot like what used to be called getting lost in a book — an enduring, if low tech, virtual reality device whose experience is easily shattered by attention to its body.

Which is to say, what really horrified Kafka was Simonides, in toga and laurel, claiming that a poem was a speaking picture and a picture a mute poem: an attitude that assumes there are no more differences between these two languages than between Greek and Latin. As we demonstrate each time we try to describe a picture or picture a description, each language is cognate with representation itself, but there is also a gulf between them that is wider than the Aegean. What this difference is is probably unanswerable, given the fact that thinkers since Aristotle have had as much success doing so as they have had in answering all of the Big Questions. Still it helps to recall some of the touch points writers since Aristotle have returned to. Namely, one says, the other is seen. One is spatial — a flat land, its sister temporal. Time in Flat Land mirrors the eternal snows, the moment fixed forever on Keats's Grecian Urn while Narrative time is an arrow transversing a space which exists only in the mind of the reader.

Accordingly, each has its own rhetoric. "The King died and then the Queen died of grief," is according to E.M. Forster, a plot reduced to its simplest elements: a sequence with a causal relationship. More generally we can say that a plot, which moves through time, is an arrangement of events for literary effect. Do I tell the tragic ending of my story first in order to enhance the pathos of early and innocent events? (Love Story) Do I withhold the ending as long as possible to bring suspense to a pitch? (Hitchcock) Do I jumble chronology, even the events themselves, to simulate a postmodern culture? ("The Baby Sitter") Or do I write events and chronology according to an arbitrary constraint? (A Void) Or leave them up to the reader, as in many hypertext novels, effectively eliminating an ending?

In composing my visual narrative, similar rhetorical questions are asked: what is the sequence that I want the eye to travel through. But even if the content is the same, it is expressed through a rhetoric of space, of adjacency.

I can use the rhetoric of one to intrude on the domain of the other. I can suggest a movement through time in flat land by photographing, as does Cindy Sherman, a woman (herself) who takes a pose in a scene that suggests a conventional narrative. I can depict diachronic scenes simultaneously as in narrative sculptures like the Trajan's Column or a mural by José Orozco.

Similarily, ekphrasis, painting in words, is the counterpart to narrative painting. But on the page, the word painting is no more a painting than the narrative painting is a narrative. Or rather one is the other in the way that Magritte's pipe is a pipe; for at one level, as WJT Mitchell points out, the figurative, it is. And, of course, sequential images such as in movies and comics have brought into being that oxymoron, the "non-verbal narrative."

Still, in terms of a literature that is meant to be read, the ways that images are not narratives seem to yield the most fruitful reasons for incorporating images into a narrative strategy. The differences between them also reveals an inherent source of tension. Specifically, to literary narrative, here taken to be a written representation that moves through time, the fixed body of an eternal moment is a corpse. A reader has to stop reading to look at form, even the form of the words he or she is reading — and to a character in a story, a pause in reading is like a little death. The problem is compounded if the reader stops reading in Latin to reread the same material over in Greek.

Of course, imagery can go beyond illustration or decoration. So it seems as if the attempt to marry reading and looking is not as hopeless as Kafka thought.

Rather, a more useful dynamic in terms of image and narrative is closer to the mechanics of a pun. The more successful image/texts, that is the narratives whose success depends as much on words as embodiment seem to assume an awareness of difference similar to that of pun; what makes a pun work is the ability of the viewer/reader to retain the original, such as Othello, in his or her mind while taking in the pseudo-copy, such as O-Jello; the success of the text depends on the reader's ability to note the difference between the two. For example, a clown in As You Like It looks at a clock and muses;

"Thus we may see how the world wags; tis but an hour ago since it was nine; and after one hour more twill be eleven; and so from hour to hour, we ripe and ripe; and then from hour to hour we rot and rot; and thereby hangs a tale."

The point is lost on us unless we realize that Shakespeare's original audience would have heard this: (338k)
When we understand that Shakespeare's contemporaries pronounced "hour" as we say "whore," the conflation of sex, mortality and time snaps into clarity. The wagging of a clock's pendulum becomes an icon for the passage of time as well as cycles of life and death, erection and flaccidity — Sex and Death as Freud might put it, or more generally, boundless desire limited by finitude — but only if we are able to hold both senses in a gestalt-like pattern at once.

Likewise, the image/text, the text that incorporates images as part of its rhetoric, opens up because of the gap between the multiple interpretations suggested by its literary and visual components. An image can reference a thousand and one words, even while it remains mute: that is, the image can link the narrative to associations, to discourse outside that which is said — old news to creators of narrative in theater, opera and film. Thus, Peter Sellers casts the don of Mozart's Don Giovani as an African American and the opera about murder and debauchery, now set in white, upper-class New York, is by this simple visual gesture charged with racial issues completely absent in the original. Here, as in the image/text, it is perhaps of as much importance that the associations created by the narrative mosaic foregrounds the time and place of its own creation: a way of seeing that embodies an epistemology.