At last September’s MoMA gathering where I spoke about digital criticism and such, I had the pleasure of spending time talking with Alison Nordström, whom I’ve always admired. We were chit-chatting about one of those exasperating moments of censorship, when she suddenly asked me, “Why do people hate photographs so much?”
I confess that question has been on my mind ever since. One needn’t look far to see evidence of photographic suspicion and hatred in the gallery/library/archive/museum sector, certainly. But it doesn’t stop there. The hatred of photographs permeates our culture in even the most banal ways.
The latest example came last month when I was asked to color correct two portraits that were intended for an online staff page. After correcting them, I received the reply that my pictures had been nixed. Reason? Because the organization wanted staff pictures that looked “natural.”
I hadn’t used a healing brush, or filter, or even a selection tool. So what was unnatural? Going to the organization’s website, I looked at the staff page and quickly noticed that every single picture was underexposed. Not one of them had a significant highlight. Most of them were not sharp. Indeed I had to squint multiple times to see various individual’s facial features. Most of the pictures had a shocking color cast, either yellow-orange or cold blue.
Translation: a “natural” photograph is a poor photograph, straight out of a phone.
My contention is that such photographs are not “natural” at all, and my argument is simple. If any non-color blind human had been in the position of the camera at the time these photographs were taken what would they have seen? They would not have seen anything like these pictures. Scientists have already proven this repeatedly in color modeling.
The human eye/brain would have immediately corrected the color casts and perceived whites as white.
The human eye/brain would have immediately corrected the contrast ratio, and adjusted shadows and highlights to fall within human visual range.
The Bayer array these cameras use for their recording of information automatically softens/blurs a picture, so a human eye/brain would have seen each of them as sharper. (In fact, most cameras include a sharpening algorithm precisely because of this.)
If the human eye is the measure of what humans consider natural — and it is — these pictures are not natural. Quite the opposite. Yet here I was, being told that the image that came from thoughtlessly pushing a button and accepting whatever a brainless machine cranked out was natural, and that any attempt to make an image appear something like a human being would actually have seen was unnatural.
This sort of nonsense is as old as Elizabeth, Lady Eastlake’s 1857 essay on photography, in which the camera is relegated “to give evidence of facts, as minutely and as impartially as, to our shame, only an unreasoning machine can give.” But the “evidence” of a camera is bodiless. Being as William Ivins says “without syntax,” that “evidence” has no context of its own, and thus the picture tends to run away.
This picture offers an example of the problem.
It’s one of those pictures in which a notable landmark appears, but relatively small in comparison with the rest of the subject matter. I’ve been here myself, so I can positively identify the subject matter: this is the landscape around Neuschwanstein Castle. If you’ve seen The Great Escape, Chitty Chitty Bang Bang or any of a dozen other films, you’ve seen the castle. It’s one of the top tourist sites in Europe. If memory serves, this is likely taken from the palace access road, looking south-southwest.
Without the castle in the frame, about 5/8th of the picture would be gray negative space where the fog has set in. The remainder would be an landscape of indistinct trees, lost in shadow. On the right, the tallest tree stands out because it is leafless. The rest blend together. In the lower left of the frame is a trapezoidal shape that I take to be the window of an automobile (a VW camper van would be the cliché assumption).
It’s difficult to date the picture. It’s in color. Kind of. The color is well faded. The verso isn’t helpful, either.
The three-line greeting “This Paper Manufactured by Kodak” places the print in the 1970s-1980s, but that’s pretty much all it tells one. The foliage suggests that the photograph was taken in autumn. Beyond that, there are very few clues because the subject matter exists in still time. This photograph depicts a moment that could just as easily be a hundred million other moments.
What then drove the anonymous photographer here to make this photo right there, right then?
I think it’s because the photographer saw something that no photograph could convey. This picture was almost certainly not the one in mind. The processing lab has done the photographer no favors either, but the real problem is that when asked to portray what the human eye actually saw here, the camera is like the law in Dickens’ Oliver Twist: a ass, a idiot. Consider the scene in words:
The van drove along the country road. The nearest city was miles behind them, and they’d had to drive slowly out here because visibility was limited by the thick hanging fog. It still hung in the distance, but the mist began to lift and one could now see the forest below. Where it had been all grey for the past two hours, now the landscape reflected the colors of autumn: red, orange, magenta, yellow, evergreen. As they rounded the bend, the passenger called “Stop!” to the driver. Pulling off the road, the car stopped, and the passenger jumped out. There it was: just visible through the fog in the distance was Mad King Ludwig’s castle. They’d only seen it in movies, and the cheap replica at Disneyland. But there it was: stark, white, immense, dominating the landscape.
Now this is a bit of speculative fiction, to be sure, but the description fits the facts of the picture. Given that little bit of prose, one will likely conclude that the original picture here does it absolutely no justice. Had normally-sighted persons read this story then seen the original picture, they’d be severely disappointed. Had they themselves been at that viewpoint in this scene, they would have seen something more like this:
This is a modest color correction. It restores the shadows, which the human eye would have seen, and some of the color variation, while adding a bit of sharpening. It’s far from a fantastic image, but this comes from an amateur photographer with a 35mm camera who wasn’t much concerned with art. The correction is a whole lot closer to what a human would see, and from there it becomes easy to surmise — or at least to imagine — what a human may have thought.
Oddly, the photographer turned the camera to portrait instead of landscape ratio. Usually portrait ratio emphasizes size, especially height, but that does not happen here. So why use it? I have no idea, especially as I think this picture works much better as a landscape.
Whatever the rationale, one thing is clear: the camera did not reason similarly. The camera did not choose a frame. It did not choose portrait or landscape. It did not see deep into the shadows with the high contrast ratio of the human eye. It did not sharply focus itself on the most interesting object. It failed to be impressed that it was photographing Neuschwanstein Castle for the very first time and so did not make the castle larger in the frame. Instead, being a truly dumb machine, it spit forth a most unnatural-looking picture that no human being would accept as what they saw there.
Poll one hundred people and you will get one hundred people saying that the corrected picture is more natural. Because it is. It is more akin to how human beings see. Those who academically argue that the camera is “correct” are barking up the wrong parapet. Their argument is that a colorimeter measured the ambient light at the scene and it proved that shadows were bluish, and that the trees were dull in color, and the sky was much brighter than the ground, et cetera et cetera et cetera. But the argument is foolish. The colorimeter is not the audience for photographs; human beings are. And human beings visualize the Abney effect, the Helmholtz-Kohlrausch effect, the Bezold-Brücke hue shift, and a dozen other perceptual phenomena. Remember what Humpty Dumpty replied to Alice:
“The question is,” said Alice, “whether you can make words mean so many different things.”
“The question is,” said Humpty Dumpty, “which is to be master—that’s all.”
In my anthropocentric world, the human is always master. Human beings use colorimeters. The colorimeter does not use human beings. To imagine it the other way around is, when it isn’t silly, utterly barbaric.
On the technical level, one corrects a picture to look more like what a human would see if one’s eye were in the position of the camera. But the technical level alone isn’t enough. The corrected picture is a damn sight better than the original — every human being will agree. It looks more natural, which is to say more like human vision. The thing that the corrected picture cannot do, however, is provide the missing ingredient of human memory. No picture could.
Photographs elude fixed human context. No amount of visual “evidence” will answer a viewer what the anonymous photographer above actually saw. Even less likely is finding visual “evidence” that will match, word for word, the stories the photographer would tell, either about the picture itself, or using the picture as a prompt. The photograph will not behave like a nice art-historical artifact that reveals all its secrets with just enough scholarship. Instead, it floats like a wave-particle in some visual equivalent of the luminiferous ether, waiting to attach itself to any context that is handy.
And that, I think, is the beginning.
After years of pondering it, I think I have begun finally to form an answer to Ms. Nordström’s question.
Why do people hate photographs? Because they are like the A.I. machines of the 19th Century.
Created by humans for human purposes, photographs have proven to be too useful. They are so useful that they have infiltrated nearly every human endeavor from art to zoology. They appear everywhere. They reproduce effortlessly. They accept any outfit, repeat any statement. And they obey no rules of proper conduct as they do so. The same photograph can be the source of maternal pleasure, evidence of someone’s history, or child pornography, and the maker of the image has no say in the matter. They are autonomous artifacts that, once created, cannot be controlled.
Just as people fear that A.I. machines will take on lives of their own and put them out of jobs before ultimately exterminating them, they fear that photographs too will take on lives of their own, and destroy civilization from within. (Consider the NEA flaps of the 90s.) But rather than take a growth mindset toward them, most humans have thrown in the towel. Pictures are bad. Pictures happen to us without our consent. Pictures are authoritative — and natural — and there’s nothing we can do. The Photographic Singularity is inevitable.
No wonder people are frightened. And out of fear, hatred.
Omar Willey was born at St. Frances Cabrini Hospital in Seattle and grew up near Lucky Market on Beacon Avenue. He believes Seattle is the greatest city on Earth and came to this conclusion by travelling much of the Earth. He is a junior member of Lesser Seattle and, as an oboist, does not blow his own trumpet. Contact him at omar [at] seattlestar [dot] net