The following contains some of my loosely organized thoughts about image manipulation, digital and otherwise. I do not expect these ramblings to answer to any profound questions, but I hope they can open a few avenues for thought.
Some seem pathologically concerned about any sort of image manipulation. They wonder how much is too much, as if they might be arrested by the Image Police if they make that final tweak. Just when I thought I had seen every conceivable comment on the topic, I read how one poor fellow worried that basic adjustments might make his images look "better than they really are". Many of these writings hold the opinion that film was somehow the standard for image faithfulness, while any and all digital manipulations are somehow deceptive. I feel these notions are critically flawed.
Many years ago I read a book called "The Print" by Ansel Adams. Although it was written before digital imaging existed, its most fundamental lessons are as valuable now as they ever were. In the book Adams states "...the negative is similar to a musician's score, and the print to the performance of that score". Whether we work with negatives, positives, or RAW digital files, the image as captured realizes only part of its potential. Realizing its full potential requires refinement on many levels for many reasons, some technical and some artistic.
There are a few artists who still use analog darkroom methods for printing images captured on film. For most other uses images on film are nearly always scanned to create digital files. Those files are subject to the same manipulations as any other digital image file. Further, many digital manipulations are simply digital versions of darkroom techniques that are nearly as old as photography itself. Digital image manipulations range from color balance and level adjustments to the removal of objects from a photograph, to the addition of elements that were not present when the photograph was captured. In between are techniques that can improve the quality of nearly any printed image. Their usefulness of each depends on the image in question. For discussion purposes I have sorted the myriad of available digital manipulations into three groups. I call them adjustments, subtractive manipulations, and additive manipulations.
This grouping contains items like brightness, contrast, color balance, color saturation, and curves that among other things help us map the larger dynamic range in a camera's RAW file to a smaller one like that of an LCD display or a print. Nearly any image ever captured on film or digital media can benefit by the manipulations contained in this group. Without these most basic manipulations, creating fine prints is usually a hopeless exercise. For instance, many images look dull or flat because they use only a portion of the available dynamic range. Mapping the brightest and darkest pixels to white and black respectively brings life to these images, making them look more like a we would perceive the original scene. In the black and white darkroom the same thing is accomplished by matching the contrast grade of the the paper to the negative, using different developers, and varying development time of the exposed paper. Things get more complex with color media, but the principals are the same.
Cloning usually replaces some pixels in an image, usually with other pixels from the same image. Cloning is often used to remove spots caused by dust on scanned film or a camera’s image sensor. With film scans, cloning is also often used to remove scratches present on the film. Similarly, it can also remove a distracting twig on the edge of an image, power lines cutting across a sky, or even major image components like roads. Accomplishing this in a darkroom is difficult, although scratches in the base layer of film are often “removed” by smearing a light coating of oil on the film to fill the scratch, rendering it invisible on the print. The technique has obvious drawbacks, but is better than the alternative of having visible scratches in the printed image or retouching the final print by hand.
These manipulations typically involve copying an object from one image and pasting it into another. Additive manipulations are obvious unless the light direction, exposure, and scale in both images are exactly the same. These manipulations can put a full moon in the sky of an image taken on a moonless night or add a wolf that didn’t exist in the original scene, howling at a moon that was not originally there either. Many UFO and other “trick” photos have been made in darkrooms, but as is typical, the process is much easier when done on a computer.
Now that we know what the fundamental groups of manipulations do, it should be fairly obvious that manipulations in the "adjustments" group can help overcome limitations in photographic processes. Scenes in real life radiate a tremendous range of light, but photographic prints can only reflect a little of it. Relatively speaking, there are feeble limits in all of our photographic processes. Adjustments like those described can help us more accurately fit a piece of reality onto a piece of paper. But, like subtractive and additive manipulations, these adjustments can also be used to distort reality to varying degrees. That is the matter we take up next.
What is Reality?
Even those who decry any form of image manipulation alter their photographic representations of reality in various ways, whether they realize it or not. They expand or compress perspective by the choice of lens focal length. They represent a world full of color in shades of gray by choosing black and white films, or wildly and inaccurately exaggerate colors by choosing films like Fuji Velvia. Lens mounted filters transport them to other worlds in terms of reality alteration, and even the act of representing a three dimensional world on a two dimensional piece of paper distorts reality. The question we must each ask ourselves is whether the goal of our photography to accurately document reality, or whether there is something more involved. For instance, in "The Print" Ansel Adams writes “Photography is more than a medium for communication of reality, it is a creative art”.
Film, the Paragon of Truth?
People love unreal colors. Kodachrome 25 was one of the most color-accurate films ever made. When Fuji came out with Velvia 50, Kodak engineers laughed among themselves at its wild inaccuracy and crayon-like colors. Of course we now know Fuji had the last laugh. Professionals stopped using Kodachrome because Velvia images nearly jumped off of editor’s light tables by comparison, and were therefore more often picked for publication. People loved the way images on Velvia film look, but no real sunset was ever as red, and no real grass was ever as green as their portrayals on this film. To think digitally boosting saturation or emphasizing certain colors is somehow worse than slapping a roll of Velvia film into a camera is completely delusional.
Our brains automatically white balance everything we see. This is similar to, but much better than, the automatic white balance on digital still and video cameras. Even the most neutral daylight balanced films display colors accurately only when the incident light has a color temperature of 5500 degrees K. This occurs outdoors during only a few hours of each day, and even that depends on geographic latitude. Films also show shadows and objects photographed in open shade with a blue cast. Things never appear this way to our eyes. Obtaining very accurate color in prints from film always requires that adjustments be made, and often they are significant.
Shifty Shifts, Tilts, Fisheyes, and Telephotos
Shift and tilt capabilities available in view cameras and specialized 35mm format lenses are often used to create images of objects as they would appear from locations where a person cannot physically stand, with perspectives that no one could ever really see. Fisheye lenses have a completely warped view of the real world and super-telephotos compress perspective so much that entire mountain ranges appear as paper cutouts. Normal and rectilinear wide angle lenses make parallel lines appear to converge when the lens axis is not perpendicular to the objects being photographed.
Our brains do lots of amazing tricks when it comes to seeing things, often acting as a filter capable of removing nearly anything, based on our particular experience and bias. At least initially, our brains often filter out power lines or a piece of litter by a roadside because we are focused on a beautiful scene in the distance. A person who has spent their life collecting litter may well see it first, though, and notice the distant scene afterward. Likewise, a lineman may well see the power lines first. Everyone will immediately notice power lines and litter in a photograph, though. It is interesting to note that while our minds often subtract objects from our vision, only disturbed minds add objects to scenes. Perhaps this is why additive manipulations are the most troubling to the vast majority of people.
Are all lies equal, or is the exaggerated color of Velvia film a little lie while cloning out an errant stick is a big lie? In both cases the image does not accurately represent the original scene. If color exaggeration is acceptable, how much is acceptable? Are cartoon colored people and trees better than cloning out a stick? It is obvious that digital image manipulation is philosophically no different than any other kind of image manipulation; whether it is by choice of film, in-camera technique, lens mounted filter, darkroom technique. It should also be obvious that film holds no monopoly on truth, and regarding color is more often wrong than right. It may not all be a matter of inaccuracy, lies, and betrayal, though. Perhaps we like Velvia’s color exaggerations because our minds tend to see and remember spectacular scenes with more vivid colors than those that actually existed. In a way, it shows us what we saw more accurately than it shows us what was really there. That might explain some try to get "Velvia color" from their digital cameras, which are typically far more color accurate than the film.
If it is acceptable to render colors the way we "saw" or remember them, should the same standards not apply to the scene itself? As discussed earlier, when we witness an amazing sunset and later think about what we saw, we do not usually remember superfluous items like an old soda can that was also in our field of view. Should we remember that soda can with our images, or does it serve only to detract from what we are trying to convey? Should it be cloned out of the image so the image can represent what we saw and felt, or should it remain in the image to simply document the fact that we viewed a beautiful sunset over an old soda can? What about the equally superfluous and distracting tree branch that could not be avoided and pokes into the edge of the frame?
Artistic License and Personal Bias
A photographer can do whatever he or she likes with his or her photographs, unless they are journalistic, documentary, or are otherwise represented as completely factual. For instance, in doing nature photography over the last 35 years I have seen vast areas consumed by development. It even happens inside national parks to make more room for cars and people. In an otherwise unspoiled place I have no more regard photographically for something like a road or sign than I do any other man made trash that degrades the landscape. I have no problem removing it, if it helps the image show the beauty we once had, but have since destroyed for greed or convenience. As Joni Mitchell sang in 1970:
"Don't it always seem to go
That you don't know what you've got
Till it's gone
They paved paradise
And put up a parking lot"
It's Only a Matter of Goals
Worry and consternation regarding image manipulation is largely a waste of time. Each of us simply needs to take the time to define our own goals and then work accordingly. Do you want to document fact or express your personal vision? On one end of the spectrum, scientific images include calibrated references for size and color, and purposely avoid unique personal influence by the photographer. On the other end of the spectrum, artistic photography exists to convey the photographer’s view of the world to others. You can seldom, if ever, have it both ways in a single image. The vast majority of us, including many who apparently do not realize it, live somewhere between scientific accuracy and purely fanciful art. Each of us need only find a place where we are personally comfortable. It’s that simple, and it always has been.