Taking life philosophically.
10 November 2010
Having recently taken up digital photography as a more serious hobby, I have been thinking about what it really means to take photographs and what (if any) kind of post-processing is acceptable. So far I cannot claim to be familiar with the relevant theoretical work. However, I have given my ideas some thought, and I think they are defensible.
To me it is banal to think that a photograph is intrinsically subjective. In my opinion, the fact that a photograph only presents a view chosen by the photographer according to the settings chosen by him does not seem enough to justify the belief that there is no “objective” photograph. (Let us ignore the potential difficulties of trying to define what an objective photograph actually means.)
Just like a scientist setting up a experiment or a journalist preparing a news story, a photographer can follow certain rules and principles when taking photographs to ensure a truthful representation of reality and not just his subjective view. In fact, I think almost all of these rules tell what you should not do: the idea is that the photographer only takes the photograph of something that is already there. Basically, if it is there, and you photograph it, your photograph is objective. You might not represent the “whole” truth, which is important too, but you are representing one part of the truth.
Telling only a part of the truth can be claimed to be a possible way of lying, but this to me seems exaggerated. Rather, as far as I can see photography allows for a certain amount of photographic “rhetoric” that is based on the knowledge the photographer has about the way people interpret photographs. For example, a photographer may choose to misrepresent a generally sunny city by taking photographs only when it happens to be raining, because he knows that the viewer often expects a series of city photographs to represent the city in more or less its typical state. But if the photographer makes no implicit or explicit claims about how typical his photographs are, he is not lying. The viewer that is taken in is only being too uncritical of the photographs. Undoubtedly there are myriad other possible scenarios of playing with the expectations of the audience.
However, although I believe that the photographer can be objective, I am not entirely sure to what extent pure photographic representation is possible. By this I mean that from the light that enters the camera to the final print or image on a screen no adjustments or manipulation take place. Obviously, since photography involves technology, some engineering is necessary, and all cameras are engineered to produce a certain kind of photograph.
There are several technologies involved in producing the image, and they are designed according to conscious choices by several people. Such technologies existed already before the transition from film to digital cameras, but as far as I know, photographic film design is more limited in its possibilities than digital image processing. However, photographers have always manipulated their photographs. What amount of manipulation is permissible has always been a difficult question with many different answers.
Photographic processing can be divided into systematic and ad hoc processing. By systematic processing I mean that any camera is designed to produce certain kinds of images systematically through certain mechanical or digital operations that remain consistent. As long as these operations consistently produce a natural, realistic image, this kind of processing to me seems innocent.
By ad hoc processing I mean generally the adjustments that the photographer makes whenever he needs to adjust something to improve the quality of one particular photograph. This can be completely acceptable. For example, often a photographer has to use exposure compensation in his camera to produce a correctly exposed photograph. However, it is one thing to do this while taking the photograph, and a completely different thing to retouch the image, to adjust the brightness and the contrast of the photograph (either digitally or in the darkroom) after you have already taken it. In the first case you are merely adjusting the camera settings to produce a better photograph, and in the second case you are manipulating an already existing photograph. In the end, I think a photographer has to be wary most of all of ad hoc post-processing (or retouching).
Personally, I am hesitant to do anything to my photographs after I have taken them, even if they have obvious technical problems such as a mistaken exposure or boring composition. With the amount of thought and research that has gone into developing the modern digital camera, I see it as a serious flaw in the abilities of the photographer if he has routinely to correct his images by retouching. Yet, even the best photographer makes mistakes. Probably some manipulation has to be considered permissible in the end, but only to complement the abilities of the photographer, not to be a substitute for them.
To get an idea of what kind of manipulation is within acceptable limits, it is useful to think about what a photograph really is in broadly scientific terms: it is a permanent image produced by a camera allowing light to affect a photosensitive surface. A camera on the other hand is a device designed to produce photographs with no exaggerated departures from reality, in the sense that they are not vastly different from what the eye can see.
These definitions rule out certain things. For example, if you remove something from a photograph or add something to it with software, it follows that the result is not just a manipulated photograph but not even a photograph. Obviously whatever you do in image manipulation software is something else than light affecting a photosensitive surface. If you manipulate a photograph, what you are doing is producing an image with a photograph as source material, but in the process you are transforming it into something else. It is still an image, and perhaps even an image of something, but it is not a photograph of it. It might still be art.
The interesting question is what counts as adding and removing, or changing. If you accidentally change the color of one random pixel in ten million, arguably you are not changing enough to turn the image into something else than a photograph, since such a change is something that a slightly defective camera might do the photograph as well. However, if you add or remove a person, then the image is no longer a photograph. Most things fall between these two examples.
Cropping is a common post-processing technique, and it seems quite innocent: after all you are only cutting out a part of a photograph. However, you are still removing something, and depending on the photograph, it may be something relatively unimportant or something crucial. However, in this case the question is easy: a cropped photograph is no longer a photograph, but a part or a “crop” of a photograph. This means it is something less than a photograph, but still something not completely unlike a photograph.
There are probably no simple principles for deciding what kinds of operations are permitted and what operations will be ruled out. For some operations it is more difficult to set hard rules, as they would be allowed in certain situations but not in others.
Even if you feel that some adjustments are necessary every once in a while, I think it is important at least to have an uneasy feeling about manipulation in general, like Harry Benson says: “If I manipulated the photos, I would feel that everything I did was fake.” If you feel like this, you will not permit yourself excessive manipulation, and you will pay more attention to detail when you are actually out taking the photographs.