Just as in any form of fine art, there is the idea and then there is the eye. You have an idea for an image that you want to create and, if you are lucky, you have the eye for a good picture. This is true whether you are a painter or a photographer. The tools and the method are secondary.
I can take photographs, but for me they are not art in themselves, because my artistic ideas are more towards the abstract. In fact, with modern cameras anyone can take photographs, but they are seldom art. And this is how I feel about much of fine art photography. So much of it is beautiful reportage that its difficult to stand out from the crowd.
Seeing Burtynsky at The Photographer's Gallery got me thinking about my own use of the camera in fine art. His images are beautiful reportage. Mine are attempting something more abstract. To create an impression rather than report a situation. I am trying to create surreal images uing a camera.
In some respect, all photographs are surreal. Early fine art photography was usually monochrome, even when colour photography was readily available. This is because monochrome enhances the surreal effect. See a beggar on the streets of Paris or New York (often seen in fine art photography) and the black and white image makes the effect more artistic because it focuses the attention on the subject rather than their circumstances.
Even when the subject matter is abstract (e.g. part of a steam engine) then the use of monochrome enhances the surreal effect and hence the artistic qualification.
But nowadays it's more likely that the photographer will work on the image some time after taking it. Even if the idea was pursued en plein air (and the photograph was taken to pursue an artistic idea) the photographer is freed from achieving their objective in the field, because they know they can improve the image back at the lab. [Of course, some of this post-processing could be achieved in the days of film by dodging and burning-in and other darkroom techniques, but nothing like what is trivially possible with image processing software today]
So to the modern photographer the camera becomes more like the pencil and sketchbook that the painter would take out into the field. Not least because the photographer, aiming at a particular idea, will probably take many "sketches". Digital images cost nothing. Taking twenty or so shots of the same landscape with different exposures and camera positions will give the artist something to work with back at the computer.
One image will surely come closest to the artists "idea" (assuming they have the "eye" for a picture). This image will usually benefit from image processing (cropping, sharpening, selective lightening etc.). If the photographer is seeking to reproduce the reality that they saw when taking the photograph, then this enhancement of the image, surreal though it still is, is more than acceptable if the dramatic impression that is achieved is fine art in its emotional impact. Think, glowering black thunderclouds over a golden sunlit meadow. An 18th century landscape with 21st century tools.
But suppose the idea the artist has is more abstract or more surreal in that the image they seek would never be considered real. Then the same computer-based post-processing could achieve that just as well. That is how I find myself using a camera today. I look for particular images that pursue a particular artistic idea. For example, recently I have been looking at churches that have the same proportions at Mondrian's church at Domburg. I've found a few. Some you can't get far enough away from to get a decent photograph. Some you find on days when the lighting isn't good. But over time you collect enough stock images that you have one or more to work from.
You could of course just use the photograph, as many artists do, to transcribe into an actual sketch on paper and work from there with paint.
What I want to achieve is to automatically turn the photograph into the (in my case abstract and/or surreal) image I had in mind when seeking the photograph in the first place. I want to determine the systematic means of turning the photograph into the artistic impression that was my original idea. I want to use those same systematic means to process many images, to achieve the systems effect (that I have described in an earlier post) of serialisation. For that I have to use image processing software. And while it is usually sufficient to use the direct manipulation interface that these packages provide, I will often resort to scripting the manipulation. Scripting achieves something important for me. It allows me to readily look at many variations of an idea, simply by changing some parameters.
As often as not the idea I am pursuing is based on the study of a particular artist (usually, but not exclusively, abstract). I want to achieve the kind of effect I believe they were trying to achieve with paint. Why not use paint? Because it's been done. More pointedly, why not just use the built-in filters in the image processing package? I could do. I do. But sometimes they don't quite achieve what I want. So I resort to scripting. My hope is that I achieve something unique, something that couldn't be achieved without scripting. I do have some confidence that that is the case, since my scripts will usually encode some carefully created geometry.
Why start from photographs? Well, I don't always. Some of my images are purely geometric. But photographs give me an additional resource on which to apply geometric transformations and I earnestly believe that I can achieve some unique images by my particular way of combining photography and geometry.
Of course, I can only leave it to trust that my artist's eye is the match for my artist's idea.