A few responses to the comments (more than 100 of them so far, many of them excellent) for "Digital and its Discontents." (We set an attendance record for a Sunday yesterday, with 14,900 page views and another 1,800 people reading the post through feeds. Sundays are usually pretty quiet around here; that was more like a decent weekday. Particular thanks to our friend Colin Jago for his kind mention over at the Auspicious Dragon Photostream.)
First of all, sorry about the title. I was just being alliterative; it's a weakness. Digital apologists need not be so all-fired defensive about their chosen medium, but mea culpa on that one.
Secondly, I want to thank the comment writers who began their comments with the admission that they hadn't read the post. I think comments should be reserved for people who have read the post—that's what comments are for—so I didn't allow those comments, but it was stout of them to cop to it. Salut.
Third, one person said that the post was "a lot fancy talk that just boils down to the Film vs. Digital debate." Wrong. I am not arguing film over digital or digital over film, and I said so several times, although the confusion may be partially due to my ill-chosen title. I can count on two hands the number of rolls of film I've shot in the past year, and anyway, anything anyone wants to do is fine with me, whether it be film or digital or anywhere in between. That doesn't mean we can't talk about the ways in which they're different, of course.
Fourth, the term is indeed "digital photography," sanctioned by the highest authority: usage. I dislike a lot of terms, but nobody put me in charge of it so it's not my call. (I notice that this fall there's going to be a TV show in the U.S. about a main character who goes on journeys, and the show's called Journeyman, to my great disgust. That will effectively put a stake through the heart of the real meaning of that fine old Chaucerian-era word. A pox on their nomenclaturally-numbskulled show; I hope it fails.)
Fifth, several people objected to my interpretation of the etymology of "photography" and reminded us all that photography means "light-writing" or "writing with light." People, I had to be pedantic with y'all, but for anything to be written with light, it stands to reason that something needs to be written. Nothing's written on a sensor. A sensor just passes along a signal, which then gets passed along again many more times. Reader Julian Barkway put it far better than I can:
Physically, a negative or a slide is a direct record of the light that struck the recording material at the time of exposure. Molecules have been immutably changed. A physical event has occurred.
Now consider the digital sensor: light strikes the photosites and triggers an electric current to flow (so far so good) but, and here is the crucial difference, this current does not cause a direct and permanent physical change in the recording medium. Instead, it is interpreted by various software algorithms and the result is a stream of voltage fluctuations which do not directly equate to the light that struck the sensor. Only by the ingenuity of the computer algorithm can this stream of voltages be interpreted as an image. In this sense, a digital image has no existence outside of the context used to create it. However, an image recorded on film is an intrinsic and immutable record of the event being recorded.
Light on a sensor is more akin to...I dunno, maybe American Sign Language than it is to "writing," at least until it's "transcribed" into a digital file, which can be done in innumerable forms at innumerable stages of the process. Then again, a digital sensor works more like our eyes do. Nothing gets "written" on our eyes either, though, except perhaps in the case of retinal burn when Poindexter stares at the sun.
Sixth, I have to admit that, after I so carefully encouraged people to look at the differences between film and digital, I am amazed and somewhat appalled that so many people, some so stridently, have tried to assert that film pictures can be manipulated too and therefore there's no difference.
It's not that film photographs can't be changed or manipulated, as I was careful (I thought) to acknowledge. But the point is that because a negative is a directly-written record that has a physical existence of its own, it's more difficult to manipulate it, and the manipulations are easier to detect in the finished work.
Here's a brief list of just a few of the changes I've made to digital pictures: changed the image geometry (as Ctein kindly demonstrated with "The Moon and Venus"); added extra foliage to trees; substituted one color for another; made people thinner; enlarged a cloud; replaced one head with another in a group portrait; straightened curved edges; replaced closed eyes with opened ones; added shadows and reflections; cloned out objects that were present in the scene; added objects into the picture that weren't present in the scene; moved objects from one place to another; recolored gray hair; replaced an ear with the same ear from the other side of the person's head, only flipped (actually this is something our graphic design wizard did to a picture of Avedon's, no less, but that tale will have to await another time to be told); corrected all manner of imperfections in pictures, from warts to weeds to litter; added sky, and extended the very frame of a picture. This is but a sampling, and all done just with my fairly limited Photoshop skills.
Now imagine that you have an 8x10 negative of a meadow with a barbed wire fence in the foreground. In the background are mountains and a nice sky. You're making a contact print on silver paper.
Remove the fence, please.
In a purely film/darkroom context, you're severely constrained with what falsehoods you can introduce willfully to the picture. It's not so much that they can't be done, but that the bar against them is set much higher: they're more difficult to pull off and much more likely to leave a detectable trace to the eye of an expert viewer. (You should walk around a museum exhibit of silver prints with me and let me point out the printer's tricks. I was a custom exhibition printer for years; we'd be there all day. I could show you where Edward Weston picked black pinholes off his prints, or where Frederick Evans drew in pencil on his negatives to correct an emulsion defect.)
With digital files, I can make all sorts of additions and deletions to the factual content of the image so easily that it's almost play. If you replace a telephone wire against a blue sky with cloned blue sky in a digital picture, there's no way I or anyone else can tell from looking at the print that there was once a telephone wire there. But do you think you can pull off even that relatively simple alteration in a silver print without a guy like me being able to see it? Maybe you can, but I like my chances.
So, how are you going to remove that barbed wire fence from that contact print, without going back and reshooting or giving yourself a day's worth of retouching work? How are you going to make it so a guy like me can't see that something funny is going on when I look at the print? The point isn't that it can't be done; the point is just that it's a lot tougher to do and harder to hide.
So just because you can burn and dodge a double-negative print, that makes digital and traditional the same, does it, rather than points along a continuum spaced robustly far apart? Right.
P.S. Several commenters have said that to get rid of the fence, I'd scan the image and remove the fence in Photoshop...clearly, this exposes another weakness in my presentation of this argument. I thought it was obvious from the first that I'm looking at this from a theoretical perspective, and that what I'm comparing is today's digital imaging as it's commonly practiced by the kinds of people who read this blog, and film-based optical/chemical photography as it existed before the invention of digital imaging. Was that never clear to people? If it wasn't, I'm sorry. It makes no sense to me to compare digital and partial digital, so of course what I meant was a completely analogue workflow, including the necessary steps to make the film picture into a print that people can actually look at. (The same workflow consideration has to be extended to digital, because the untranslated RAW image is not recognizable as a "picture" either.)
At the very least, though, this entire discussion has taught me a lot about how to make a more persuasive argument next time, and suggested ways that I might present it more clearly.