Friday, August 11, 2006

Don't Believe What You See in the Papers

Don't Believe What You See in the Papers
The untrustworthiness of news photography.
By Jim Lewis

(Reprinted without permission from Slate)

There's an old joke about a Southern preacher who's asked by a skeptical congregant if he really believes in infant baptism. "Believe in it?" the preacher replies. "Why, I've seen it done!"

I thought of the preacher when I heard the latest in photojournalism's long line of mini-scandals, this one involving a Lebanese freelancer named Adnan Hajj who was working in Beirut. Hajj altered at least two photographs: In one he cloned a plume of smoke rising from buildings that Israeli planes had bombed; in another he altered the image of an Israeli F-16 to make it look like it was dropping more ordnance than it was. Both pictures were bought by Reuters, which sent them out on its photo service. When the forgeries were pointed out, the agency pulled the pictures, dismissed the photographer, and issued a statement asserting that such fakery had no place in the news business.

It may not, but that doesn't mean it doesn't happen regularly. Two days ago, the AP got caught sending out a crudely—and nonsensically—altered photo of an Alaskan oil pipeline worker; last month, the Charlotte Observer fired a photographer for changing the color of the sky in a picture of a firefighters; the same week, the Spanish-language edition of the Miami Herald acknowledged that a picture of prostitutes in Havana had been cobbled together from two different shots; in 2003 the Los Angeles Times sacked a photographer for combining two pictures from Iraq, taken moments apart, into one. In fact, it's beginning to look as if every major institution that prints photos has printed doctored or manipulated photos: Time and Newsweek, the New York Times and USA Today, Harvard University and Science magazine, and the 2004 Bush campaign. (There's a good rogue's gallery here.) Some of these were quite serious attempts to mislead the public, and some were relatively trivial, but all of them undermine the public's trust in the reality of photographs. And so much the better, because that trust is badly misplaced.

Hajj's altered F-16 photoWhat, after all, do we believe when we believe that a photograph is true? That it mimics what we would see with our own eyes, if we were standing where the camera was placed? But a camera sees quite differently: For one thing, to take only the most obvious features, photos are rectangular, whereas the human eye's visual field is an ovoid blob. Moreover, "normal" vision is roughly equivalent to what you get from a 35 mm camera lens set somewhere between 42 mm and 50 mm zoom. Anything longer than that shows details no human eye could see; anything shorter shows an unnaturally broad vista. And cameras are notoriously crude when it comes to dynamic range: Highlights get blasted and dark areas become muddy. Can we countenance using Photoshop to draw otherwise invisible details out of shadows (a correction it can make almost magically, with two mouse clicks), so a photo better approximates what an eyewitness would have perceived? How far can we take this? What about using zoom and contrast enhancement, Blowup-style, to reveal elements that would be lost to the naked eye? Should that be prohibited? As far as I can tell, the news world's answer is, "Maybe. Sometimes." Then maybe we should ban photos taken through microscopes. And so you see how quickly this criterion fails.

Perhaps, instead, we should judge a news photograph as a collection of purported facts about the world that is accurate if its claims are true and inaccurate if they're not. But photographers make editorial decisions all the time: where to point the camera, of course, but also how to frame the shot, whether to crop and if so what, how long a shutter speed to use; and all of these can affect the facts a picture presents, without falsifying the image. And anyway, which facts are relevant? Last week the Times ran an impressive graphic showing before and after satellite photographs of a bombed Beirut neighborhood, but the "before" picture was in blooming color and the 'after' picture was in black and white. Many people would say that the black-and-white shot was more "realistic" because monochrome seems to offer a kind of sobriety. But of course color is closer to the facts, and even color is unreliable. The tones captured by Fujichrome are noticeably different than those captured by Kodachrome, but it's hard to argue that one is more accurate than the other.

To make matters more complicated, news photographs are made by more hands than the photographer's. Editors at home will sometimes crop a picture, or clean it up, and they'll often flesh out captions, which can radically change what we think we're looking at. Hajj's photo of the Israeli F-16 bore a caption that said the jet was dropping missiles; in fact they were flares, but who could know that just by looking? What you see when you contemplate a news photo is what you're told to see. And sometimes it's what you're allowed to see: When Ronald Reagan visited Bitburg cemetery in 1985, photographers were forced to shoot from vantage points that prevented them from getting both the president and the gravesites of the Nazi Waffen SS who were buried there in a single frame. Were the pictures that came out of that event "true"? They certainly weren't fake, but if you were a photo editor, would you have run them? Inasmuch as photojournalism is meant to impart information, the Bitburg pictures were as misleading as Hajj's clumsily darkened smoke plumes. Imposture can take many forms.

None of this would be especially pressing if we didn't still believe that pictures offer up a different kind of truth than prose. But photographs are supposed to be self-validating, to be, in some profound sense, proof. Believe it? Hell, I've seen it done! It's a myth, of course, and it always was, but it was an easy one to believe, especially when film was the medium. Old-fashioned analog cameras were relatively straightforward machines, and the images they made existed in a simple causal relationship to the scenes they were pointed at. Fakery was generally perpetrated after the fact, by airbrushing or collaging negatives. Most news organizations still pay lip service to that understanding of photography by forbidding any post-processing more radical than removing dust spots and some dodging and burning.

But such rules are little more than atavistic superstitions; these days they make about as much sense as trying to minimize bias by forbidding reporters to use adjectives. The switch from film to digital has made the distinction between What the Camera Saw and What the Photographer Did almost entirely moot. Even the cheapest chip-based pocket camera lets you set white balance, color effects, aspect ratio, and a dozen other parameters and automatically interpolates pixels based on its best guess as to what came through the lens; and such tweaking isn't tampering, because the image doesn't exist until these decisions are made. The photographer for the Charlotte Observer was fired for adjusting the color of the sky in his picture so that it more closely matched what he had seen. But if he'd had a Sunset Scene Mode on his camera, post-processing wouldn't have been necessary, and he probably wouldn't have gotten into trouble, though it's hard to see what the difference really amounts to.

Needless to say, news photographers shouldn't doctor photographs any more than reporters should make up quotes. But "doctoring" is a slippery concept, and photographic truth is an illusion. Realism is a special effect like any other, and the sooner we realize as much, the better off we'll be; the decrees of photo editors—no post-processing!—only serve to shore up a faith in photographic evidence that was never justified to begin with. Someday we will approach each photograph we look at with the condign skepticism we bring to each story we read. In the meantime, these useful scandals remind us that we're complacent and credulous, and that photography is rife with paradoxes, which can't be solved with hand-waving and apologies.

No comments: