From the archive of gaze: the address to posterity, the notebook preparing the archival entry, the spectacles anticipating the achievement of S. M. Eisenstein, the pistol in the pocket. Subject of the archive: the presidential election of 1908. Source of the image: George Grantham Bain Collection, Library of Congress, http://www.loc.gov/pictures/item/ggb2004002184/. Photoshopped.
And a montage element bridged in from the textbook of comparative anatomy:
What am I reading here? Some history. I can place the document in that genre because it explains and is explained by its date of composition: April 1942. Off the page, history has already taught me that April 1942 has something to do with the United States’ entry into World War II in December 1941, and I accordingly think I understand what the words on the page mean when they say: “Los Angeles, California. The evacuation of the Japanese-Americans from West Coast areas under U.S. Army war emergency order. Japanese-American children waiting for a train to take them and their parents to Owens Valley.”
That timestamped and circumstantial text isn’t only a history, of course. It’s also a literature. Shaped by narrative convention, it belongs to the literary genre of the caption — specifically, the caption to this photograph by Russell Lee in the Library of Congress’s Farm Security Administration / Office of War Information archive at http://www.loc.gov/pictures/item/fsa1998003537/PP/.
Because a caption has an explanatory power over its attached image, its words make the image in some sort verbal. They make it tell a story. From this historical era, for example, there exist similar photographs taken in other theaters of World War II, and our reaction to any of them will be, so to speak, captional. If the caption tells us of a Japanese baby being evacuated from Los Angeles, we’ll react one way; if the caption tells us of a Chinese baby being evacuated from Nanking, we’ll react another. Likewise, as of August 2014 I think most of my academic colleagues would react with sympathy if a caption told them that the image were of a Palestinian baby, but with exasperation if an editor then corrected the caption’s adjective “Palestinian” to read “Israeli.”
It’s been a long time since a news photo could be thought of as intelligible on its own terms, of course. A century ago, not long after Freud taught us how hard it is in principle to know what we’re seeing, Lev Kuleshov demonstrated that in practice we can’t even see the difference between a dead baby and a bowl of soup. In Kuleshov’s experiment, a movie clip — one clip, only — showed an actor going through the physical correlates of emotion. A montage of that clip with some ostensible stimuli of emotion then made clear that any imputed sense of emotion, of emotion about an ostensible stimulus, was demonstrably nothing but an artifact of the montage effect. Juxtaposed with the image of the baby, the actor’s mobile features and heaving chest seemed to mean one thing; juxtaposed with the image of the soup, they seemed to mean something else. We might have thought they expressed feeling, but In themselves they were nothing but mobility and heaving. Whatever emotion we derived from them was an illusion. We were misled by our expectation of a caption to read. But from the belated realization that Kuleshov’s tiny silent movie is captionless there follows a happy ending. To learn that one is free from captions is to learn to be free from other things as well.
Therefore, face to face with an image that has been captioned, I find myself wondering whether I can do something in the captional space above the border where the caption’s words begin. Wondering, I open Photoshop and set about trying to change the non-verbal part of this historical record. Timidly, at the start, I may tell myself and you that I’m only restoring the captioned image, only using modern narrative technique to put an illustrated story — a children’s book, a picture book! — back together. But of course what Photoshop and I are doing to this ensemble of words and non-verbal forms isn’t merely a historiographic revision. Photoshop and I aren’t doing history now; we have subordinated ourselves to a corpus of aesthetic principles that have nothing to do with Los Angeles or trains or 1942. Our project has been taken over by art. So:
And you see: I have not merely restored the record or corrected it. I have (as editors used to say in the days of Thomas Bowdler, he of the verb bowdlerize) improved it. See how much more tragic than Russell Lee’s original my little girl is, hear how much more clearly we can say “Little does she know” about her! How satisfyingly pretty I have made this children’s story!
Because it’s open to the possibility of an aesthetic judgment like that one, my version of Russell Lee’s photograph is no longer quite a historical document. It can now be read without its caption, as if it were in the process of growing distant from the history of events. It’s no longer merely captional. If it isn’t yet art, it may at least be art history. Because the space around the little girl has now been filled with art, her mother has now been barred forever from entering the image frame to reroll her daughter’s cuff. Because art always has a completion function, the caption below this image has now been translated into a dead language and made emotionally unreadable on any terms but art’s. The little girl’s picture can now say only the one thing art ever can say about itself: The End. Waiting for a train which can now never arrive to transport her out of the image, holding the doll which she now will never outgrow, the little girl has become an unravished bride of quietness.