Skip to content

Why a painting is as good as a photo on a passport

by Rebecca Roache

Fredrik Saker, a Swedish artist, is in the news this week for having successfully applied for a driving licence using a photograph not of himself, but of a self-portrait painting. It is interesting to consider, in the light of this, what is so special about photographs. Why do agencies that issue documents featuring images of their bearers – like driving licences and passports – require applicants to submit photographs? Is there any good reason not also to permit self-portrait paintings, drawings, or any other sort of artistic creation? 

The philosopher Robert Hopkins compares photographs to other pictorial representations in a paper published this month.¹ He argues that the important thing about photographs is that they are factive: they relay facts about their objects, in the sense that they ‘capture how things are (or were)’. Had it been our eyes, rather than a camera, that were directed towards the object of a photograph at the moment it was taken, we would have seen whatever is depicted in the photograph, as it is depicted. On the other hand, when looking at a painting or at another type of ‘handmade’ picture, we cannot (without further information) assume even that the painting depicts a real scene, let alone that it depicts the scene accurately. Even handmade pictures that can be taken to depict reality – such as diagrams in a technical manual or Ordnance Survey maps – are not truly factive, Hopkins tells us, since the way in which they are created introduces a potential for human error, and therefore for the picture to fail to depict its object accurately. Handmade pictures, then, are not factive. Further, whilst photographs can sometimes be deceptive – we can use Photoshop to change aspects of a digital photograph, and we can use clever positioning to make small, close-up objects look large and far away, for example – we view these as deviations from the normal practice of photography, thanks to the existence of recognised and established norms governing the creation and viewing of photographs.

Photographs, then, being factive, enjoy a special epistemic status; and this plausibly explains why they are the preferred method of capturing people’s likenesses on passports and driving licences. However, whilst photographs as a mode of pictorial representation might be factive, there are good reasons to deny that the images on our passports and driving licences are factive in the sense presumably required by those who issue and rely on those documents (that is, in the sense that they convey the right sort of facts about the holders of those documents). Recall Hopkins’ remark that handmade pictures are not factive because the way in which they are created introduces a potential for human error. Specifically, he tells us, ‘information from the object portrayed can only reach the viewer via the artist’s take on the world’. Whilst the method in which photographs are created is not vulnerable to human error in this way, the process by which our likenesses come to appear on our passports and driving licences is. Consider that, when applying for a first passport, applicants are required to have a professional person countersign the photograph to certify that it is a good likeness of the applicant. This introduces a potential for human error in a number of ways. First, judgments about what counts as a ‘good likeness’ are unavoidably subjective. Second, such judgments are vulnerable to distortion as a result of social factors: passport applicants must have known their countersignatories for a number of years, and we might anticipate that such long-standing friends may feel under pressure to avoid creating problems by refusing to certify a photograph on the ground that it is not a good likeness (especially, perhaps, if the photograph is an unusually flattering one of the applicant). Third, some people look unrecognisably different in public to how they look when they wake up in the morning as a result of make-up and hair styling; as a result, without further guidance about how to make judgments about what constitutes a ‘good likeness’, we might wonder how the term can have any reliable application at all in such cases. Fourth, it would be possible to fool even an honest and well-intentioned countersignatory into certifying a photograph that does not depict the applicant. This process is further complicated by the fact that, when applying for passport renewals, applicants are required to return their expired passport along with new photographs which do not need to be countersigned (unless the applicant ‘can’t be recognised from the photo in [their] existing passport’ – something that, again, depends upon a subjective judgment). Presumably, then, whether the new photographs are acceptable is decided by staff at the Identity and Passport Service, based on a comparison between the old and the new photographs. As a result, the selection of images for subsequent passports is mediated by not one but two subjective human judgments, the second of which is made by a person who need never have seen the applicant in person. (Since one may update one’s driving licence photo using information already held by the government about one’s passport, the vulnerabilities in the process for selecting passport photos also apply to the process for selecting driving licence photos.) This means that, just as, in the case of handmade pictures, ‘information from the object portrayed can only reach the viewer via the artist’s take on the world’; pictorial information about the holder of a passport can only reach the immigration official via the countersignatory’s (and perhaps also the passport worker’s) take on the world. It is not difficult to imagine how one could manage to acquire a passport featuring, among other things, a photograph of an identical twin, a lookalike, an unrecognisable earlier version of oneself, a friend who bears only a passing resemblance to oneself, or – as in Fredrik Saker’s case – a painting.

Whilst Hopkins’ account helps us understand why photographs are the preferred means of pictorially representing the holders of passports and driving licences on those documents, it also helps us see why handmade pictures would do just as well. For whatever privileged epistemic status photographs may enjoy over other means of pictorial representation is compromised by the process of photograph selection and approval used in passport and driving licence applications. As a result, there seems no obvious case against using handmade pictures (or, rather, photographs of them) in those documents. If the Identity and Passport Service is willing to rely on human judgments about whether photographs are good likenesses of the people they depict, why not also accept handmade pictures judged by the same people to be good likenesses?


¹ Hopkins, R. 2012: ‘Factive Pictorial Experience: What’s Special about Photographs?’ Noûs, 46/4: 709-731 (journal copy available from the publisher here, pre-publication version available here).

Share on

3 Comment on this post

  1. “Is there any good reason not to request, instead, self-portrait paintings, drawings, or any other sort of artistic creation?”

    Is this question just misleading? Having scanned the article beneath, there seems to be nothing that directly challenges the appropriate answer which is that, unless you’re Chuck Close, they are easier to take; to vet for appropriateness; and have a higher level of accuracy than most paintings. I’m sure any opposing argument will just point out the obvious: they are an imperfect method of identification. Startling find.

  2. JC, thanks for your comment. I should not have set up handmade pictures in opposition to photographs in my opening paragraph, and will adjust my wording accordingly. As for your other points, I think they are dealt with in the remainder of the blog: photographs do have the properties you mention, but the process by which they are selected and approved for documents undermines the special status they enjoy above other forms of pictorial representation.

Comments are closed.