As a result of what he terms the ‘virus of manipulation’, Andy Grundberg (1990) laments that even some of photography’s icons have been subjected to alteration, and cites the example of a Eugene Smith image of Albert Schweitzer that was later revealed to be a composite of two negatives (the original images not available since Smith developed all his prints in his own laboratory before submitting them to Time):
Apparently the silhouetted hand and saw handle were added to cover up a fogged part of the frame. Ironically, this was the very image used on the cover of his monograph “Let Truth be the Prejudice.” Grundberg notes that the majority of us are aware that we can no longer take camera images at face value, so why do we continue to do so?
Barthes observed that image manipulation, or what he termed “trick effects… intervene without warning in the plane of denotation; they utilize the special credibility of the photograph” and thus convey greatly connoted messages under “the ‘objective’ mask of denotation” (1977, p 21).
On the subject of fakery, digital forensics laboratory Fourandsix Technologies have an online gallery of forged imagery dating right back to the American Civil War:
The first example is a 1864 composite of three images that purportedly showed General Ulysses Grant posing in front of his troops at City Point, Va. and the website goes on to provide many more examples, particularly of images used in the political arena. As Chief Technology Officer and co-founder of Fourandsix Hany Farid says: “Photography lost its innocence many years ago” (Farid n.d., p 4). In a specific case where Farid’s expertise was called upon to investigate accusations of fakery, he expressed concerns about the tendency among photographers (as well as competition organizers) for the practice of ‘instagramming‘ photographs in a bid to make them appear more dramatic, or at the very least, conforming more to current consumer tastes.
The image in question was this winner of World Press Photo of the year 2012, which had been very obviously manipulated to give it a stylized look, but the judges were ok with that; problems arose when a blogger accused the photographer Paul Hansen of making a composite. Employing digital forensics technology, Farid and his co-workers proved that the image was not a composite, but some remained unconvinced and the discussion continues. The image really does look too good to be true, almost like it was staged and studio-lit. the public seem to doubt images that look very clean and well composed, while believing blurred, grainy or out-of-focus images to reflect the truth:
“A beautiful shot is apt to lie, while a bad shot “is a guarantee of authenticity,” one that loses in attractiveness but gains in truth” (Trinh 1991, p 61).
As Farid observed in an article for the Scientific American, the days when the manipulation of photographs involved lengthy and painstaking darkroom procedures (and thus reserved for politicians) are long gone:
“anyone with a computer can readily produce fakes that can be very hard to detect. Barely a month goes by without some newly uncovered fraudulent image making it into the news” (Farid 2008, p 66).
That same article outlines five common methods used in digital forensics to identify fakes, some of which can be detected by the naked eye (differences in lighting, shape and position of eyes and the specular highlights they contain) while others rely on computer technology and algorithms to detect tampering (cloning, camera fingerprint). Generally, as Farid admits, fakes are spotted by alert individuals who spot flaws in the images, while digital forensics are used to authenticate images for legal purposes and media outlets.
Another website that looks at image manipulation is 10b photography. Since they offer image manipulation services to professional photographers, they begin with a section about the ethics of minimal manipulation, demonstrating that even the likes of Salgado or Koudelka use minimal manipulation post production to burn in detail that would otherwise be blown out in the highlighted areas:
Their point is that as long as the essential information in the picture remains unchanged, such manipulation is not just forgivable, it is standard practice! They cite the case of Klavs Bo Christenson, whose images of Haiti were disqualified from the Danish POY competition for being overly Photoshopped:
I do agree that the images have been extremely worked, but the problem is that who is to judge what is or is not too much? Surely the competition rules should have set out what was acceptable or not. We are so used to seeing saturated and high contrast images in the media and elsewhere, it seems almost ridiculous not to apply some sort of image enhancing filter or software. From the 10b website:
Some very obvious saturation, contrast and gamma tweaking, as well as dodging to emphasise the figures. To my mind the image looks awful! The argument could also be made that black and white photographs should not be considered at all, since conversion to B&W completely changes the colours recorded (after all, contemporary digital cameras record RAW files in colour, not in B&W).
It is a debate that I began looking at in Level 2, and does not seem to be any closer to any real kind of solution – particularly with the digital age, the widespread employment of Photoshop, and vigilant blogosphere detectives who are constantly on the lookout for and revealing shameless and deliberately misleading image manipulation in the media!