Saturday, August 5, 2017

Colberg on Pictures

I do so like it when people actually take a credible swing at "what's it all about then?" sorts of questions in photograph. Colberg has written a piece, The Perfect Imperfect Picture. Don't misunderstand me, he's all wrong, and I am about to talk about why, but I appreciate the effort, and I appreciate that people are thinking about this kind of thing.

First of all, he's more or less right that there is a great deal of variation in what we actually get to look in with a digital photograph, just as there was a certain amount of variation with analog. He does start right in with muddling things up, though. In the analog world he seems to be talking about frame-to-frame variation. This piece of film rendered this scene in one way, the next one rendered this other scene, this other picture in another way. In digital land, he's talking about how differently the same picture is rendered in various contexts.

Later, he revists this idea in a way that makes it clear that the first thing (different renderings of the same picture) is irrelevant, even to Colberg.

Next up, he jumps to a brief vignette on "what is a photograph anyways, where is the, you know, the actual thing itself?" which is a perfectly good question to ask, and one that doesn't have a good answer, but he fetishizes technology for a moment by saying "it must be the code, the code is the thing" which is a refrain I find numbingly tiresome. In the first place, "code" is the program, it's the app, it's the part that does calculation. "Data" is the passive stuff upon which calculations are done. What Colberg is talking about is, technically, "data" not "code", unless his phone encodes pictures as a FORTRAN program that, when executed, writes out a JPEG file. Which I can assure you is not the case.

To say "code" sounds cooler, though, so it's kind of in vogue.

Anyways, it's all irrelevant, because in the second place there's a clear and precise analog between the first generation file (JPEG, DNG, RAW, whatever) and the negative. It's the unique thing produced by the instrument, and if you must answer "what is the actual thing?" then that's as good an answer as any. And it's simply not very interesting (unless you're planning to fetishize technology, which apparently you must) to examine the differences between "negative" and "first generation data file", they're the same for all ontological purposes. And then, just as every print pulled from a negative is a little different, so every rendering of the file is. In digital land, surprise, everything happens a lot faster, a lot more frequently.

After a mysterious trip to the world of gratuitous and wrong-headed slams on Sontag, Colberg gets to to the meat of his thesis.

People, he asserts, like imperfections in their pictures.

The first problem here is that he's approaching this like an engineer and an amateur gearhead photographer. What on earth is an imperfection? He sounds like some douchebag on a forum bitching about "missed focus" or "white balance is wrong."

Colberg is now back to muddling up variations in renders of the same picture with variations in handling of different pictures, but here it is important.

Nobody cares much if colors look a little different on their phone and your screen. This is variation between different renders of the same picture, and most people consider it irrelevant unless the differences are enormous, and then they find it irritating. The important point here is that the renders remain indexical. It is important to note that an indexical representation of a scene, one that corresponds directly to the scene, is not unique. There can be many indexical representations of the same thing. There can be, in short, many photographs of the same thing, each exactly as "true" as the other, all different.

When people apply filters to their phone snaps, to make them look like vintage film, obviously they do care, and they like the look. This is handling different pictures differently, or sometimes making multiple different pictures from the same underlying first generation file. It doesn't actually have anything to do with the things Colberg started with, with this notion of different renderings.

Finally, Colberg gets around to something sensible and interesting.

So, people are smashing up their pictures with filters and whatnot. Somehow, this is not damaging the credibility of pictures (or is it?), and somehow, people seem to want to want the look of glitchy, weird, serendipitous accidents, and that does indeed have to mean something.

Unfortunately, he leaves it right there, just when it's about to get interesting.

He has an opportunity here to draw his two muddled things back together, to talk about something like the evidentiary properties of a photograph. We, collectively, will accept various renders of the same picture, as well as quite a lot of deliberate variation in the form of edits and filters, as evidence. You can smash a picture up fairly brutally, at least in technical terms, and we, your audience, will still treat it as proof that you were at that diner, that party, that beach. If my screen renders your picture with a weird green cast, or if Facebook's servers have made it all soft with overcompression, I still treat it as proof.

There's nothing inherent about digital technologies here, to be sure. Commentators like Colberg and, and well everyone except me as far as I know, are obsessed with the idea that digital photographs are somehow, inherently, in their underlying nature, different. That is simply not true. There are direct analogs with every aspect of analog photography (of course they are, analog photography is the model the digital guys copied, and continue to copy). There are at least two different approaches to Adams's Moonrise over Hernandez photo out there, and many different prints. Nobody, however, denies that Adams was there on that road, at that time.

What is different is the pace, the amount, the numbers. Rather than 4 different prints, all subtly different, we have a thousand, a million, a billion different renders on different screens. We have different interpretations through filters and photoshop, 2, 5, 100, as many as you care to churn out, every one a click away. None of the individual instances is substantively different from what might have occurred in the bad old days of film.

But the sheer quantity is different, is new. It changes things, in ways we don't fully understand. Digital has changed things, it is different.

Commentators and critics like Colberg and Bush are, however, quite wrong about why.


  1. "Nobody cares much if colors look a little different on their phone and your screen."

    Except possibly the photographer and discerning viewers?

    I, for one, very much care about such things, which is why I greatly prefer to view photos in print and book form, not as .jpgs on a monitor. And the same is also true when it comes to displaying my photos.

    This is because I know the photo they the viewer sees in front of them is the same photo I signed-off on and released into the world.

    This creates a connection between the viewer and the photographer's vision as expressed by the photo that does not -- cannot! -- exist otherwise.

    I will admit this is a minor point for many, but I believe it's also a crucial one, because printing and/or publishing a photo effectively fixes it in time and locks it into a final, finished form.

    It represents the photographer's intent and vision at the time it was made and while this may or may not change over time, I believe there is something noble and worthwhile about preserving its original form, if only because the resulting connection between the viewer and the photographer is both more direct and more reliable.

    This is the reason why many collectors focus their efforts on collecting a photographer's vintage prints instead of modern ones, even when they are made by the same person and, quite often, visually inferior as a result of improvements in materials and techniques over time.

    Unfortunately, the distinction between vintage and modern prints will disappear once technology makes it impossible to lock a print into final form the way that making a print or publishing it in a book does now.

    In the future, the only way to determine the vintage of a photo will be to look at the timestamp on its digital file (and even that isn't a reliable indicator, because it, too, can be modified at any time who possession of a copy of the file.)

    And when this happens, it will be a shame, because it will no longer be possible to study how a photographer's vision for their photos evolved and changed over time.

    Or is this just old-school thinking by a boomer-generation photographer that will have no place in our brave new world?

    1. There's a *place* for it, sure, but it's a minority position. There's a place for practically anything, as a minority position.

      Color precision is a bit of a chimera, honestly, and my opinion - an admittedly personal and idiosyncratic one - is that there are very very few photographs which require much color precision.

      Tonality is more important, I think. There certainly are photographs which rely pretty strongly on the overall tonal placement (most of Adams's work disintegrates into an unremarkable mess without strong blacks).

      But the vast majority of photographs out in the world don't need either of these things to function in the way they are intended to function. Colberg's diner snap is a snap, evidence that he was there, evidence of a sunset. As it stands, it doesn't need much of anything in terms of reproduction "accuracy" (whatever that might even mean) to function.