Does the manipulation of news and documentary photographs matter, and how should we talk about this issue?
The discussion about the number of images disqualified for manipulation in the 2015 World Press Photo contest has been intense, and the debate will be ongoing. But I’ve now left the Secretary’s seat for this year and have returned to civilian life as an independent writer. As Secretary I had enough to say on the specifics of the contest issue last week, and those contributions are best summarised in this podcast and its associated links.
Now it is time to reflect personally on why manipulation matters. I’ve written a lot about manipulation over the years, but not really addressed up front why it matters. In many respects the reasons for being concerned about manipulation, and the way those reasons are articulated, have not been at the forefront of the recent controversy either.[1. I find this particularly fascinating because of my commitment to an ethos of criticism best articulated by Michel Foucault when he wrote that “practicing criticism is about making facile gestures difficult.“] To keep the big picture in mind, so to speak, we need to focus on the reason and how it is justified.
The first thing to observe is that the question of possible manipulation is far from exhausted by the focus on processing digital image files (though that priority makes perfect sense for a debate ignited by a photo contest). At almost every stage in the photographic process from capture, production, to the publication and circulation of photographic images contains the potential for manipulation. The mere fact of going to place A rather than place B to produce an image involves a choice that might represent reality in a partial manner. How travel to a photographic location was enabled and funded raises a series of questions. Once on location, the composition and framing of scenes necessarily involves choices that shape representations. The editing, selection, tagging, and captioning of images for potential publication adds more layers of decision. Which images are then distributed to media clients for purchase, and how those clients present, sequence and contextualise those images, is another realm of creative choice that shapes the representation of events and issues. As David Levi Strauss has observed, “the truth is that every photograph or digital image is manipulated, aesthetically and politically, when it is made and when it is distributed.”
I agree with Levi-Strauss here, especially as I once wrote that we should regard all photography as staged. That was a deliberate provocation to try and break away from the exhausted, incoherent and unsustainable position that photography could be objective and true, which is a commitment still so strong we see traces of it every time an image’s status is questioned.[2. There’s a lot that needs saying about the notion of objectivity as commonly used in relation to photography, something that will have to draw on the 500 page magnum opus on the topic, Lorraine Daston and Peter Gallison’s Objectivity (Zone Books, 2007), which details the surprising cultural history of this supposedly scientific concept.] Given its manner of production and distribution, all photography is unavoidably and inherently a construction, an interpretation, a claim or a statement. As a result, I don’t see the concern about manipulation in terms of defending truth or securing objectivity, because nobody has ever been able to do that. If we knew what was irresistibly true and objective in any given situation, there would be no need for any debate or questions – we would just acquiesce to the obvious.
The constructed nature of photography is even more of a fact in the digital age. In many ways, while the digitisation of photography has been understood as transformative, I don’t think we have full appreciated quite how radical the change has been. We have conventionally thought of photography as a practice that makes images directly. This is largely the case with analogue processes, which produce observable or latent images on film or other media. In this context, the camera is understood as a picture-making device.
In the digital era, we still think of the camera of a picture-making device, but this needs to change. In the digital era, we need to understand the camera as a data-collection device, a device which, according to Kevin Connor, is “gathering as much data as you can about the scene, and then later using advanced computational techniques to process that data into the final image. That creates a much more slippery definition of an original, because what is defined at the time of capture is not necessarily a fully formed picture.” With this understanding we need to appreciate much photography has become “computational photography.”[3. Seeing digital photography as essentially computational photography is to adopt a broader definition of computational photography, and take it beyond its more common concern with specific digital processes like HDR.]
This should change much of the language in the manipulation debate because no image exists without processing. Sensors record data. It takes processing to produce the first in-camera file as well as the image on the camera’s LCD screen. That means the idea of “post-” processing is redundant, because there is no “pre-” or original image, to work on. Raw files are not negatives, but data files that are the first step in processing which both allow for and require further processing. This makes all the references to the darkroom, and what famous photographers used to do in those spaces, an anachronistic and irrelevant analogy for the question of manipulation today. These conceptual and technological points underscore the fact concerns about manipulation cannot be expressed in terms of truth and objectivity. What, then, can the concern about manipulation be founded upon?
This is where we need to change the conversation about photography – meaning news photography, photojournalism, documentary or editorial photography, however your want to name these visual accounts of our world. The change involves understanding the integrity of the image in relation to its function, rather than its philosophical status as an object. We need to focus on the process of photography rather than just its products, and consider the issue in terms of what images do rather than what images are.[4. There are many sites in which the purpose and process of photography is not properly highlighted. Elizabeth Edwards makes the compelling case that many exhibitions see the default value of photography as “art” even when presenting documentary images. As she writes, “this implies that photography’s ultimate purpose is aesthetic discernment and expression. But I don’t think that this alone communicates the importance or power of photography.”]
Images can have lots of purposes, and there will be many we want to just entertain or please us. For them, we can have more relaxed standards. However, if we want some pictures to be able to function as documents and evidence, we have to ensure certain things so those pictures can function as documents and evidence. Given that the first question asked of a contentious image is always whether it is faked or somehow changed, it is only by drawing a tight line against even materially small changes that we can underwrite the credibility of such images. We have to be able to show, in a variety of fora, that for the pictures we want to be documents and evidence, material content has not been added or subtracted from the data file captured by the camera and later processed. This is a task for image makers, editors and publishers alike, and it will take time to enhance the integrity of the image through this commitment. No doubt many will break it or bend it in practice, but if there is at least a very low tolerance for even materially small changes, it is feasible to imagine norms changing over time so that the trust of readers and viewers is bolstered.
It is obvious that manipulation is a problem for news images, and the changes are significant, when we see photographs like those of North Korean hovercraft or Iranian missiles cloning objects in order to deceive. It might seem less of an issue to some when the alterations seem to be just tidying up the photographs but not changing their overall meaning, as has been argued in relation the Narciso Contreras’ case. Associated Press severed ties with him when he revealed he removed an object from the bottom of corner of one of his news images. The level of manipulation might have been materially small, but it was nonetheless ethically significant, because AP said he had violated their standards for truth and accuracy. Lewis Bush made a fair observation that AP was defending its business model despite the fact Contreras – who outed himself for an aesthetic change that was not by itself intending to deceive the audience – had in effect revealed the constructed nature of imagery. Lewis argues persuasively Contreras’s revelation runs agains “a common sentiment amongst proponents of photojournalism, that the truth of photography must be continually shored up against the erosive waves of untruth.”
This rhetorical commitment to truth is obvious in many of the comments on the New York Times Lens blog that debated the rules and ethics of photojournalism in the digital age. It is also obvious in the more sophisticated claim that what makes photography unique is the indexical relationship between image and subject. Those who asserted truth in various ways were matched by commenters who expressed their doubts about objectivity by asserting that photography was self-evidently subjective. And there was the featured but anonymous photojournalist, who asked:
What is truth? Photography certainly isn’t. Photography is artifice. We can underexpose and overexpose the same image, neither version is “true” or “untrue” — it is just a different interpretation of the world in front of us.
Claiming photography is just artifice or subjective is very different from what I am arguing here. The notion of artifice can just refer to creativity and expression, but it also means a contrivance designed to deceive or trick, so that is not helpful. Subjectivity is equally problematic. It is a concept that depends on objectivity as its other, the reverse side of the same coin. Logically, if we doubt objectivity we bring down the edifice that includes subjectivity too. Subjectivity is also a concept that leaves the impression everyone has a perspective of equal credibility and value. That might be an appropriate claim for art, but it is not a claim that can advance the understanding of contentious events and issues, where there has to be a struggle in terms of those claims that are better supported by evidence than others.
I have said above that my starting position is that “all photography is unavoidably and inherently a construction, an interpretation, a claim or a statement.” That is the basis on which all I think all understandings of photography must be built, but it is an insufficient basis on which to rest. That is why I think we need to accept that, but then proceed to consider images in terms of what they can do rather than what they are. I think, following Ariella Azoulay, we also need to understand an image as being one statement in a larger regime of statements, so that we dispense with the idea that pictures alone can testify to all they show. Images can be powerful documents and evidence, but they require other statements, other information, to be truly effective.
This is where dealing with the pragmatics of photography (what we want images to do) rather than being bound by false commitments to an untenable philosophical ideal (what we think images are) offers a recasting of the manipulation debate. That might seem like a semantic change to some, but I believe recasting the issues in the terms I have outlined – how do we underwrite the credibility of images we want to function as documents and evidence? – offers a completely different way of understanding manipulation. The return to the mystical foundations of truth have not moved any of this discussion forward because we always founder on the obviously vacuous rocks of objectivity or surrender to the hopelessness of asserting subjectivity. Instead, the reframing in terms of the purpose of image I am detailing here suggests we need to maintain the most vigilant line against material changes to data files that compromise processed images we need as documents or evidence, but we do so in a way consistent with the inherently constructed nature of photography.
Of course, much will be needed in order to develop this argument properly. But one consequence of this reframing is that the integrity of the image cannot be underwritten by commitments to the credibility of digital files alone. Given the many points in the practice of photography which can result in manipulation – according to the broader sense of that term discussed above – the debate needs to bring the manipulation of pixels into the broader realm of verification. If an image is just one statement in a larger and wider regime of statements, then we have to deal with the credibility of that regime as a whole. Verification is a concept getting a lot of attention in journalism more widely, especially but not exclusively because of the rise of user generated content, and photojournalism needs to catch up with these issues (some of which I canvassed in section 9 of the Integrity of the Image report). A digital audit trail is an essential part of a broader commitment to open photographic practices that allow images and stories to be verified, and it is possible that a more developed approach to verification for photojournalism could address some of the broader issues of manipulation that cannot be revealed through digital forensics. That is a topic I will explore in the future.
PS: I meant to add this at the time of posting, but John Macpherson’s thoughtful comment below reminded me to include this as an update – I’m not going to debate here anything about the World Press Photo contest experience or rules. As I say at the start of this article, I’m back in civilian life as an independent writer, and interested here in the conceptual issues about how, when and why talking about manipulation matters with regard to images we want to function as documents and evidence in the media economy.