IN MUSIC, I seem perversely to prefer my Parsifal without a libretto. So I find myself in total rapport with those who might prefer their Milton without a map. Besides, it’s humbling enough for an artist to see museum-goers spending more time reading the curatorial captions next to an image than looking at the image itself; do I really want to add to their literary burden?

It would appear I do. The cloud of facts which float around a piece—the sources, references, thematic directions—provides the charged mass necessary to cook up the creative storm. In my case the cloud can become pretty dense. And picking through the layers when I revisit an image’s formulation, I can relive the sense of adventure I felt originally when the haphazard and diverse elements slowly fell into coherence and emotional sense. There is a certain satisfaction in retracing the path where things came from. Where they go to—the wholeness of the piece and a feeling for its total equilibrium—I am happy to say is entirely the province of the viewer.

For me, annotation also serves one other purpose. Even the hard facts that surround my images now and then seem to change in significance during the course of time: and their change in turn may be reflected in the meaning of the immutable image itself. The documentation then can become a benchmark against which to measure the ebb and flow of taste. In 1976, for instance, in the first catalog raisonné of my prints, I described rather fully the development of one of my images, Daylilies. Now, twenty years later, it is poignant to look at that essay and see how much the developments I describe in it have continued to evolve—and how no interpretation, no inventory of ingredients, ever agrees to be final.

Daylilies began as a drawing (The Theft) for a show in Yugoslavia. I was working form a 1909 photograph of a holiday crowd watching a Zeppelin dirigible—which, however, since it seemed so quaintly nostalgic, I left out. Unfortunately, with nothing to watch, the crowd no longer had any purpose and seemed to be simply, bleakly, drifting away.To bring back some focus to the piece and develop the drawings’s middle spaces, I turned to photographs from the twenties and thirties by the Hungarian André Kertesz. But by now the feeling that everyone was isolated and waiting aimlessly was not easily cured. I thought that perhaps suggesting the space of a picture gallery might work: with its different pictorial planes it seemed a good metaphor for memory and one that evoked not only European culture but commented on both Kertesz’s occupation and my own. In homage to this concept, I added the shadow of Kertesz taking a photograph.

Suddenly the image came alive as the boy, who is one of its central figures, seems now to turn as if startled by the photographer recording him. Other shadows have joined Kertesz, and the crowd has found its raison d’être by becoming a painting. What brought the drawing to life for me was that it now incorporated a variety of planes—both literal pictorial planes and metaphorical temporal ones in which events are happening in time as well as space. These planes could be made to refer one to another, even to conduct a sort of dialogue. In my own work, the only images that interest me are those where the fact of surface, the illusion of space, and the poignancy of time coexist, intersect, and are of equal importance. Since I sent the drawing off to Yugoslavia with a sense that I would never see it again, I had photographed it.

At this point it occurred to me that enlarging the photograph onto high-contrast film could be the basis for a further development of the imagery. My usual method of preparing an image was to draw with ink on Mylar. And when I made an enlargement of the photograph onto a sheet of transparent Kodalith, the granular character of the enlargement was so like that of the ink-on-Mylar drawings that I saw it would be natural to make a collage of the two. I set the Kodalith onto a clear Mylar sheet and began to draw both over and around them to see what might evolve. Soon the suggestion of pictures-within-pictures already present in the original image began to suggest pictures-outside-pictures, and I found myself led to a much deeper dimension.

The image had become—literally—darker, and this led me into a more somber metaphorical landscape. Among other things, I had added the images of my own children, a reference to Dürer’s Melencolia I (1514), and a quotation from Eadweard Muybridge. This image of affliction, a child walking on all fours, had also been used by Francis Bacon, and I felt it strongly enough to give it its own space, away from the complexity of the rest of the plate. The picture gallery of holiday crowd, startled boy and Kertesz shadows had all become subordinate to a man, a cat and a few daylilies in the foreground. The cat and the man were borrowed from photographs by Thomas Eakins. On the wall next to them is a portrait of a girl. I had no idea who she was, but caught by her delicacy and remoteness, I had clipped her image from an advertisement in the New York Times. The windows behind her contain the hint of a darkening sky. “The daylilies come from a garden at the side of my studio,” I wrote at the end of my essay. “They are late summer flowers, and seem to grow wild.” In the course of time, however, I discovered that the particular eponymous flower that I had chosen to draw was in fact not a daylily at all. Also, most daylilies do not bloom in late summer, but in June and July. They do still grow, at least, and in greater profusion than ever, outside my studio.

Meanwhile, Yugoslavia is no more, my children are grown; the unknown, delicately remote young woman of the portrait wrote to me after having recognized herself in my print and appeared, still beautiful, at a show in Los Angeles. And by now I am beginning to look upon Daylilies as a phantom from the past.