Dan Cavedon-Taylor

The Limits of Photographic Seeing

Draft only

I've Read This

34 Views
Like

The Limits of Photographic Seeing1
[Draft only]
Dan Cavedon-Taylor Birkbeck College dan.cavedon.taylor@gmail.com

Introduction
Many share the thought that one of the things special about photographic representation is that it is strongly linked, in some way or other, to perception, that seeing a photograph of an object puts one in a perception or perception-like relation to that object. Perhaps the most well-known and provocative claim made about photographs in this respect is that they are, as Kendall Walton puts it, ‘transparent’, where this means that photographs are such that they allow viewers to indirectly see the objects they depict.2 Despite the amount of literature left in its wake, some elements of Walton’s theorising
1

Parts of this paper have been presented to audiences at the open sessions of the 2008 Joint

Sessions of the Aristotelian Society and Mind Association, the 2008 American Society for Aesthetics Annual Conference, MindGrad 2008 and the 2009 British Society of Aesthetics Annual Conference. I am extremely grateful to my commentators: Hemdat Lerman and Patrick Maynard. Thanks also to Paloma Atencia Linares, Diarmuid Costello, Miguel dos Santos, Anil Gomes, Rob Hopkins, John Hyman, Aaron Meskin, Dawn Phillips, Paul Snowdon, Scott Walden and especially Keith Hossack for comments and discussion. This research was made possible by a PhD studentship from the British Society of Aesthetics.
2

Kendall Walton, ‘Transparent Pictures: On the Nature of Photographic Realism’, Critical

Inquiry, vol. 11 (1984), pp. 246 –277.

about photographs have yet to be fully explored; specifically, that Walton’s motivation for introducing transparency is to account for photography’s affect and, relatedly, the theory constructive methodology that drives his work here. After examining transparency and its standard objections in part I, my first aim, in parts II and III, is to explore the aforementioned aspects of photographic transparency. I argue that their neglect has stymied the debate surrounding transparency and that discussion can only move forward once they are made explicit, since they block many of the objections of Walton’s critics. However, I do not claim we should endorse transparency, arguing that the theory constructive methodology it relies on should only be endorsed if there are not other theories that can account for the data transparency seeks to explain. My second aim is to put forward and examine two such theories. The first, which I discuss in part IV, is an error theory which holds that whilst photographs are not transparent, viewers nonetheless (either implicitly or explicitly) hold the theory to be true. Although this error theory accounts for the affect of photographs and is one Walton should find attractive, there are other features of photography, notably the medium’s epistemic status, that it remains silent on. The second theory I examine, in part V, and which I call the photographic displacement theory, holds that photographs allow us to indirectly see that such and such is the case about the objects they depict but without allowing us to see those objects. I argue that photographic displacement is to be preferred to photographic transparency on the basis that it vindicates the affective and epistemic features of photography we want vindicated without forcing us to rethink the limits of seeing and without having to introduce a new sense of the word ‘see.’

I. Walton and his Critics
Kendall Walton famously holds that photographs allow perceivers to indirectly see the objects they are photographs of, just as one might think that seeing an object’s reflection in a mirror allows one to indirectly see the object reflected. Walton signals this by saying of photographs that they are ‘transparent.’ Transparency is a technical term of art in this context. Walton is not claiming that we cannot attend to the vehicular or configurational properties of photographs; Walton does not think photographs are transparent in the ordinary sense of the word. Paintings, by contrast, are alleged by Walton to be non-transparent or ‘opaque’; they do not allow for indirect seeing because their creation is mediated by the image-maker’s intentional mental states. Walton is typically read as offering two arguments in support of the transparency thesis. First there is a slippery-slope argument. Walton suggests that since we see objects by means of eyeglasses, mirrors, microscopes and telescopes, how can we deny we see objects via closed-circuit television? If we are willing to affirm this, then we should similarly say we see objects via delayed broadcasts as well. But if this is so, then what is there to block us from saying that when looking at photographs we indirectly see the objects they depict?3 In response, critics have pointed out a number of reasons for resisting sliding down this slope as far as photographs.4 Recently, Diarmuid Costello
3

Walton, ‘Transparent Pictures’, p. 252. See Edwin Martin, ‘On Seeing Walton’s Great Grand-Father’, Critical Inquiry, vol. 12 (1986),

pp. 796–800; Nigel Warburton, ‘Seeing Through “Seeing Through Photographs” ’, Ratio, vol. 1 (1988), pp. 64—74; Jonathan Friday, ‘Transparency and the Photographic Image’, British Journal of Aesthetics,

and Dawn Phillips have argued that since Walton holds that footprints are transparent this ultimately shows the slippery-slope argument to be ‘beside the point.’5 Footprints presumably lie further along the slope than paintings, which Walton thinks are not transparent. More developed considerations in support of photographic transparency evolve over the course of “Transparent Pictures” where, with the assistance of various thought experiments, Walton attempts to establish that the causal conditions he takes to be necessary (though not sufficient) for seeing an object are satisfied by seeing a photograph, but not a painting, of an object. Unlike someone who is fitted with prosthetic eyes, Helen, who has her brain stimulated by a surgeon in such a way that she has the veridical visual experiences had by the surgeon, is alleged not to see. Here, photographs are claimed by Walton to be analogous with the visual experiences produced by prosthetic eyes, since their causal ancestry is similarly not mediated by anyone’s (i.e. the photographer’s) beliefs about the scene before them. A difference in the scene before the photographer appears to make a difference to the resulting photograph independent of the photographer’s noticing or believing the scene to be different; hence, plausibly, why there is the possibility of accidental depiction in photography. Paintings, by contrast, are alleged to be analogous to the visual experiences had by Helen, since their creation is held hostage to the fortunes of
vol. 36 (1996), pp. 30—42; Noël Carroll, Theorizing the Moving Image, (Cambridge: Cambridge University Press, 1997), and Berys Gaut, ‘Opaque Pictures’, Revue Internationale de Philosophie vol. 62 (2008), pp. 281—396.
5

Diarmuid Costello and Dawn Phillips, ‘Automatism, Causality and Realism: Foundational

Problems in the Philosophy of Photography’, Philosophy Compass, vol. 4 (2009), p. 8.

someone else’s (i.e. the painter’s) beliefs (and presumably their visual experience as well). A difference in the scene before someone painting that scene would make a corresponding difference to their picture only if they noticed the difference; hence, plausibly, why accidental depiction seems to make little sense in the case of painting. The difference between photographs and paintings here is often expressed by Walton and others6 by the claim that photographs are ‘mind independently’ or ‘naturally’ counterfactually dependent upon their subjects, whilst paintings are ‘mind-dependently’ or ‘intentionally’ dependent upon their subjects. Since we are inclined, according to Walton, to say that Helen does not see, we do not see objects by means of seeing paintings of objects. Since we are inclined to say that someone fitted with prosthetic eyes does see, we should say we see objects by means of seeing photographs of objects, with lack of mediation via intentional mental states being what fundamentally marks the difference.7 Against this argument many have claimed that there are certain necessary conditions on seeing an object which fail to be satisfied by seeing photographs of objects.8 For instance, when one sees an object one is in a position to keep track of that object’s spatial and temporal location over time, but photographic pictures are still and they give
6

Notably Gregory Currie, ‘Photography, Painting and Perception’, Journal of Aesthetics and Art

Criticism, vol. 49 (1991), pp. 23—29.
7

Walton, ‘Transparent Pictures’, p.265. See, in addition to those critics already mentioned, Gregory Currie, Image and Mind: Film,

Philosophy and Cognitive Science (Cambridge: Cambridge University Press 1995) as well as Jonathan Cohen and Aaron Meskin, ‘On the Epistemic Value of Photographs’, Journal of Aesthetics and Art Criticism, vol. 62 (2004), pp. 197—210.

us no clue as to their subjects’ spatial location. Moreover, the link between a visual experience and objects seen is one that continues through time and the former is sensitive to changes in the latter. But there is no continuous link between a photograph and the object that it depicts—the photograph is a ‘detached display’9—and for this reason photographs are not sensitive to changes undergone by their subjects.

II. Against Walton’s Critics
How decisive are the objections? We should note that Walton is explicit in holding that slippery-slope considerations only lend an ‘initial plausibility’ to the transparency thesis.10 Recently, in an interview with Hans Maes, he states that ‘I don’t think it is something we should rest too much weight on.’11 With this being the case, granted the slippery-slope fails to establish transparency, it is far from clear that those who wish to resist Walton’s thesis should bother focusing too much of their attention here; critics who do so risk taking Walton’s remarks much more seriously than he is asking us to. In any case, the fact that a particular argument fails to establish some thesis or other is not evidence that the thesis is false. Moreover, as we shall shortly see, this is not Walton’s main argument for transparency.12

Carroll, Theorizing the Moving Image, p. 62-3. Walton, ‘Transparent Pictures’, p. 252. Hans Maes, ‘Aesthetics and Theory Construction: An Interview with Kendall Walton’, (2007),

p. 6. [http://www.kent.ac.uk/arts/hpa/pdfs/Kendall_Walton_interview.pdf]
12

Pace Warburton, ‘Seeing Through “Seeing Through Photographs” ’, p. 67.

Furthermore, granted there are established necessary conditions on seeing an object that are not satisfied by seeing a photograph of an object, this fact should not concern Walton, indeed the replies he has offered over the years suggests it has not,13 since Walton’s methodology is one of positive theory construction rather than conceptual analysis. That is, in claiming that photographs allow us to see their subjects Walton is not attempting to vindicate any particular division or line between seeing and not seeing, but can be interpreted as inviting us to re-draw that line. He writes ‘I say we see through photographs, and see things directly, as well as through mirrors and telescopes, in a (single) literal sense of “see,” without worrying about whether this is a new sense of the word.’14 A question arises, however, as to whether the theory constructive methodology here is well motivated. Theories have data they need to explain. A theory presented without reference to data risks looking ad hoc. So we might worry that photographic transparency fails to pay its way. But Walton does not simply pluck the transparency theory from thin air. It has significant motivation, and not just as an account of the difference between how photograph and paintings represent. If that were all Walton wanted to explain he could rest content with the idea that photographs are mindindependently counterfactually dependent upon their subjects whilst paintings are not.
13

See his ‘Looking Again Through Photographs: A Response to Edwin Martin’, Critical Inquiry,

vol. 12 (1986), pp. 801–808; ‘On Pictures and Photographs: Objections Answered’, in Richard Allen and Murray Smith (eds), Film Theory and Philosophy (Oxford: Oxford University Press, 1997), pp.60 –75; and ‘Postscripts to “Transparent Pictures”: Clarifications and To Do’s’, in his Marvellous Images: On Values and the Arts (Oxford: Oxford University Press, 2008), pp. 110—116.
14

Walton, ‘Postscripts to “Transparent Pictures” ’, p. 111.

Walton’s motivation for introducing the transparency thesis is often overlooked by critics who focus almost exclusively on the slippery-slope and causal conditions argument. When Walton’s motivation is mentioned it is frequently thought that he is attempting to vindicate photography’s epistemological status—that his interest in photographs is in explaining why they are quite so good at furnishing evidence of their subjects in a way that paintings are not.15 Rightly or wrongly, Walton has been keen to distance transparency from the epistemology of photographs. Recently he has written that ‘The transparency of photographs is not essentially connected to any thesis about their epistemological value. In attributing this special kind of realism to them I was not aiming to explain the supposed value of photographic evidence.’16 But if this is so, then what does photographic transparency purport to explain? As the above quote indicates, ‘realism’ plays a key role here. To best answer the question “What motivates photographic transparency?” we need to be aware that “Transparent Pictures” starts with a discussion of what it means to say that photographs are realistic in a way that hand-made pictures are not. Photographic realism, Walton starts by arguing, has nothing to do with the fact that photographs excel at capturing detail and subtleties of shading, say. If this is all we mean by ‘photographic realism’ then it is a realism that is not specific to photographs since they can be visually indistinguishable from paintings.17 So is there nothing unique
15

Commentators who construe Walton’s motivation in these terms include Jonathan Cohen and

Aaron Meskin, ‘On the Epistemic Value of Photographs’ and Scott Walden, ‘Objectivity in Photography’, British Journal of Aesthetics, vol. 45 (2005), pp. 258—272.
16

Walton, ‘Postscripts to “Transparent Pictures” ’, p. 113. Walton, ‘Transparent Pictures’, pp. 247—49.

to photographs that talk of their realism gestures at? No. Walton claims that talk of photography’s realism gestures at the fact that photographs affect us emotionally and can cause us to feel in contact with their subjects in a way that hand-made pictures, even the most vivid and detailed ones, cannot. Walton writes: Photographic pornography is more potent than the painted variety. Published photographs of disaster victims or the private lives of public figures understandably provoke charges of invasion of privacy; similar complaints against the publication of drawings or paintings have less credibility. I expect that most of us will acknowledge that, in general, photographs and paintings (and comparable nonphotographic pictures) affect us very differently.18 As further evidence in support of this claim, Walton compares an etching by Goya that depicts dead bodies piled up on a battlefield with a civil war photograph of a similar scene. Walton claims, rightly it seems to me, that the photograph possesses an ‘immediacy’ that Goya’s etching lacks.19 Now although not wholly explicit about it, I read Walton in the above remarks as offering a third argument for the transparency of photographs, which I shall call ‘the affective argument.’ For I take it that Walton believes transparency explains why we experience a closer, more intimate and heightened connection with reality when seeing photographs than we do when seeing paintings. The idea seems to be that photographs affect us in a way that paintings do not because only photographs are transparent. We are really (though indirectly) seeing dead bodies on the battlefield when we see such
18

Ibid, p. 247. Ibid.

things in a photograph. When we see Goya’s etching, by contrast, perceptual contact with reality terminates at the picture and does not extend beyond it to the depicted bodies themselves and it is precisely for this reason that Goya’s etching is seemingly less shocking. Thus, we are now in a position to appreciate that the affect of photographs, the emotional responses they prompt and their causing us to feel in contact with their subjects, is the datum that Walton introduces the transparency theory in order to explain. Moreover, Walton’s theory constructive methodology invites us to endorse transparency, despite it seeming that photographs fail to meet certain conditions on seeing, because this feature of photography would otherwise go explained. Walton writes that if we were to deny transparency then we would be left in the dark, ‘groping for explanations’ of photography’s affect.20 Walton thus appears to hold that we shouldn’t be concerned if, in endorsing transparency, we end up re-drawing the limits of seeing. For the benefits of doing so outweigh the cost of leaving the affect of photographic pictures unaccounted for. Neglect of the affective argument should strike us as puzzling. For it seems to be the only argument Walton offers in support of transparency that explicitly ties the theory to his central aim in “Transparent Pictures”—explaining the affect of photographs. Moreover, we can see that neglect of this argument and Walton’s theory constructive methodology has arguably stymied the debate surrounding transparency and led it off in directions away from Walton’s original concerns. Insufficient attention to Walton’s methodology has left critics talking at cross purposes with Walton. It seems to me that
20

Walton, ‘Looking Again Through Photographs’, p. 807.

progress on the issue of photographic transparency will only be made if we pay attention to these elements of Walton’s theorising about photography.

III. Affect and Theory Construction
I have so far tried to make explicit certain elements of photographic transparency that have often gone overlooked, and suggested that their proper appreciation blocks standard objections to Walton’s view. I now turn to the task of critically examining these elements to see if they are in good order. Firstly, to say that the affective argument has been wholly neglected would be too strong. Gregory Currie, for instance, agrees with Walton that photographs affect us more than paintings, but objects to Walton’s mobilizing this fact in support of transparency on the basis that ‘while photographs and films may affect us more than paintings do, they surely affect us less than witnessing the offensive or disturbing acts directly would.’21 But Walton should be unmoved by this objection. After all, Walton’s claim is not that photographs allow us to directly see their subjects; rather he believes photographs allow us to see their subjects only indirectly. So Currie’s observation that directly seeing an object affects us more than seeing that object in a photograph, which seems intuitively right, is explained by Walton’s thesis. For according to Walton, it is precisely an indirect type of seeing that is afforded by such pictures, so it is no surprise

Gregory Currie, ‘Visible Traces: Documentary and the Content of Photographs’, Journal of

Aesthetics and Art Criticism, vol. 57 (1999), p. 289. Emphasis in original.

they affect us less than seeing directly. The affective argument stands: If photographs are transparent, this straightforwardly explains their affect. Secondly, a consequence of the idea that photographs are transparent is that certain other phenomena end up transparent as well. It was mentioned earlier, for example, that Walton holds footprints are transparent. For footprints are also mind-independently dependent upon the feet or boots they are prints of; a difference in foot or boot size would make a corresponding difference to the resulting print independent of anyone’s beliefs about the size of that foot or boot. Critics have worried that the same will end up being the case for such things as fossils, footprints, and animatronic gorillas built in such a way that their every movement and appearance is counterfactually dependent upon real gorillas.22 Yet it would seem intuitively wrong to think such phenomena allow for indirect seeing. One thing that’s not been noticed is that brushstrokes turn out to be transparent on Walton’s view too. They are seemingly no less mind-independently counterfactually dependent upon the movement of the painter’s brush than a footprint is on the boot it is a print of.23 Of course Walton will simply defer to his theory constructive methodology and claim that any counter-intuitiveness on the matter of seeing feet or boots by seeing their prints or bones on the basis of seeing their fossils, say, is a price we must pay if we are to explain the affect of photographs (at least his claims about being left ‘groping for explanations’ of photography’s affect without transparency suggests as much). Moreover, the fact that brushstrokes end up transparent
22

The first two are suggested by Martin in ‘On Seeing Walton’s Great Grand-Father’, the third by

Gaut in ‘Opaque Pictures’.
23

The suggestion being that a difference in direction of brushstroke makes a difference to the mark

thereby made on the canvass independent of the artist’s beliefs about the direction of brushstroke.

on Walton’s account is not enough to rob transparency of its power to explain what is special about photographs, since transparency is a claim concerning the subject of a photograph and brush-stokes are seemingly not the subjects of the paintings they produce (which is not to deny that they might figure prominently in, or ‘inflect’, one’s pictorial experience). Moreover, Walton may indeed welcome the fact that brushstrokes are transparent since this can explain why an original painting is more valuable than a reproduction or reprint. The idea might be that we value originals (at least in part) because seeing an original can cause us to feel in contact with the artist who made the brushstrokes, and such affect is underwritten by the marks on the canvass being transparent to the artist’s strokes. If we were to worry that this doesn’t seem to be seeing in the ordinary sense, then we need to remind ourselves that Walton is not concerned with whether he is introducing a new sense of ‘see.’ Indeed, given that it turns out on his view that we see object by means of fossils and footprints, this is exactly what it appears Walton must be read as doing. However, it seems to me that not all is well with the affective argument and the theory constructive methodology that it is driven by. There appears to be some sleight of hand going on here that we ought to be suspicious of. Specifically, given all Walton says about transparency being able to vindicate photography’s affective properties, it does not follow that we’d be left ‘groping for explanations’ of such features if we chose not to endorse photographic transparency and were disinclined to follow Walton in claiming that things such as footprints and fossils similarly allow for indirect seeing. Walton seems here to be sliding from the claim that transparency suffices to explain the affect of photographs, to the claim that transparency is necessary to explain their affect. But not

only does sufficiency fail to establish necessity, but, absent a full investigation of the matter, Walton surely cannot rule it out that a competing theory may explain the affect of photographs just as well, or even better. For, granted that some theory, T1, is sufficient to explain some datum, d, it is open that there is another theory, T2, which explains d just as well and which may not have the costs of T1. Therefore, it seems we should only join with Walton in embracing transparency and its revisionist consequences if we are convinced that transparency is not just sufficient but that it is required to vindicate the affect of photographs, that it is the only viable theory available to us and that we really would be left ‘groping for explanations’ without it. If another theory can explain the affect of photographs just as well and, moreover, if this second theory, unlike transparency, does not force us to re-draw the limits of seeing or introduce a new sense of the word ‘see’, then we should clearly find that theory preferable to transparency. Are there are any such theories? In the rest of this paper I investigate two.

IV. Error Theoretic Transparency
Suppose that whilst photographs are not transparent typical viewers nonetheless believe (either implicitly or explicitly) that they are. Call such a view ‘error theoretic transparency’ since it holds that whilst viewers believe photographs are transparent it simultaneously claims they are in error in doing so. Error theoretic transparency is capable of explaining the affect of photographs. Viewers who believe photographs to be transparent believe photographs allow them to indirectly

see the objects photographed, and believing that one sees an object can have a ‘topdown’ effect upon the phenomenal character of one’s visual experience, making one feel in contact with that object. Suppose, for instance, that a perceiver hallucinates an elephant in front of them, but that they believe (falsely) that they are actually seeing an elephant. Such a perceiver is analogous to the viewers of photographs on error theoretic transparency. Now it seems that a perceiver who takes themselves to be seeing an object, even if they really are not, will nonetheless feel in contact with that object precisely on the basis that they believe themselves to be seeing it. If, by contrast, our hallucinating perceiver dismissed their experience as non-veridical and took themselves not to be seeing an elephant they might, for exactly that reason, feel little affect with respect to the elephant and idly walk on by. The idea that typical viewers believe photographs to be transparent is one that is endorsed by Walton. (Though we should note that viewers believing photographs to be transparent is a further fact over and above the transparency theory; showing that photographs are transparent does not suffice to show that viewers believe photographs to be transparent.) In fact at one point Walton slides between photographic transparency explaining the affect of photographs and viewers’ belief in photographic transparency explaining photography’s affect. Specifically, Walton reflects upon what may happen when we first see Chuck Close’s hyper-realist self-portrait as a photograph, but then come to learn we are actually looking at a painting. Walton writes that ‘The discovery jolts us. Our experience of the picture and our attitude toward it undergo a profound transformation... We feel somehow less “in contact” with Close when we learn that the

portrayal of him is not photographic.’24 This ‘jolt’ is presumably a change in how the picture, now seen as a painting, affects us. Walton then writes that ‘My theory accounts for the jolt. At first we think we are (really) seeing the person portrayed; then we realize we are not.’25 But note that although Walton says here that his theory accounts for jolt, by which he can only mean photographic transparency, he then references belief in photographic transparency. Walton claims that we think we are really seeing Close when we see his self-portrait (wrongly) as a photograph, and that this is what accounts for our feeling of contact with Close. What explains our then feeling distanced from Close, once we learn we are looking at painting, is not the fact that paintings are not transparent, but that we do not believe paintings to be transparent, that we do not believe they allow us to see the objects they depict. The transparency theory itself, our indirectly seeing object via photographs but not paintings, doesn’t enter into Walton’s explanation here of the affective difference between photographs and paintings; rather, our belief that we (indirectly) see reality when we look at photographs, but not paintings, does all the explanatory work.26 Noting this slide between transparency and belief in transparency serves to bring out an interesting dialectical point; specifically, that if one holds that viewers believe photographs are transparent and one is also interested in explaining the affect of photographs, both of which are true of Walton, then one need not also insist that photographs really are transparent. For, granting Walton that belief in transparency
24

Walton, ‘Transparent Pictures’, p. 255. Ibid. The same slide between transparency and belief in transparency explaining the affect of

photographs is made again by Walton in ‘Looking Again Through Photographs’, p. 807.

suffices to explain the affect of photography, which is also what the above reflection upon hallucination seems to show, there is no further explanatory work for transparency itself to perform here; the data transparency is motivated to account for is already explained. And if there is no further work for transparency to perform then we have little reason to endorse the theory with all the revisionist baggage that comes along with it. Since Walton holds viewers believe photographs to be transparent could have remained neutral on the matter of whether photographs really are transparent and still explain the data he wants to explain. In fact Walton could have been an error theoretic transparency theorist if he so wished. He could have denied that transparency is true and still explain the affect of photographs, so long as he holds onto the idea that viewers believe photographs to be transparent. If this is right, then it shows we are not left ‘groping for explanations’ without transparency, but can explain the affect of photographs by referencing viewers’ belief in transparency instead. Although error theoretic transparency explains the affect of photographs without forcing us to revise the limits of seeing or introduce a new sense of the word ‘see’, there are at least three reasons one might be suspicious of the view. Firstly, it is an error theory, and to some error theories are to be regarded with scepticism. One might agree that viewers believe photographs to be transparent, but hold that this in conjunction with the falsity of transparency is unacceptable. This is perhaps a reason for one to prefer Walton’s transparency over error theoretic transparency. How attractive one finds error-theoretic transparency and how well one thinks its costs measure up with those of transparency hinges upon one’s prior commitments as to what makes for a good theory. Error theoretic transparency commits

us to pervasive error; transparency commits us to re-drawing the line between seeing and not seeing and introducing a new sense of ‘see’, all for the benefit of explaining a single feature of photographic pictures. Alternatively, one might have no particular qualms about error theories but be suspicious of the idea that typical viewers believe photographs are transparent (something which we have seen Walton also assents to). For one might think that it is no reasonable part of the folk theory of photographs that they are transparent. For instance, Walton’s critics frequently complain about how counter-intuitive they find the idea that we see objects by means of seeing photographs of objects. This suggests that, for some of us at least, it would be very surprising if it were part of the folk theory of photographs that they are transparent and, moreover, since we are dealing here with folk theory, surprise on this matter might be taken as counting against the hypothesis on offer. Intuitions on this matter appear somewhat divided however. Jonathan Friday, for instance, although he denies that photographs are transparent in Walton’s sense, claims that photographic experience is a matter of regarding photographs as their subject and that when looking at such pictures we regard them, but not paintings, as if they are ‘putting us into perceptual contact with the object depicted.’27 On the face of it at least, this sounds like an endorsement of what I have called error theoretic transparency. Moreover, in a similar vein, Patrick Maynard, who endorses a version of photographic transparency, holds that, as an empirical hypothesis, viewers regard photographs, but

Jonathan Friday, Aesthetics and Photography (Aldershot: Ashgate), p. 59.

not paintings, ‘as manifestations of what they are photographs of.’28 Maynard’s remarks here serve to bring out that whether or not viewers believe photographs to be transparent is not something we can really hope to settle from the arm-chair. So, absent any empirical data on the matter, it is not clear that we can reasonably come down against error theoretic transparency on this matter. Though by that same token this means there is a question hanging over error theoretic transparency as to whether it adequately characterises the folk theory of photographs. Thirdly, one might think there are other features of photography, aside from its affect, which we want explained and which error theoretic transparency is not in a position to explain. Recall that Walton is frequently taken by commentators to be in the business of vindicating photography’s epistemic properties. Although Walton insists commentators have him wrong here, we plausibly do want a theory of photographic representation to explain why photographs are quite so good at furnishing knowledge of their depicta, more so at least than are paintings. Photographic transparency, whether Walton recognises it or not, seems in a prime position to explain this. After all, seeing (whether directly or indirectly) is frequently a way of finding out about reality, one that is arguably more privileged and fundamental than other ways of knowing such as, e.g., by inference, testimony or memory. If photographic pictures alone allow for perceptual access to their subjects whilst paintings, drawing and sketches, say, allow for inferential or possibly testimonial access, then we vindicate photography’s privileged epistemic status amongst picture types. But error theoretic transparency seems unable to explain what is epistemically special about photographs since the only thing it claims about
28

Patrick Maynard, ‘Drawing and Shooting: Causality in Depiction’, Journal of Aesthetics and Art

Criticism, vol. 44 (1988), p.123.

such representation itself is that it is not transparent. Of course the theory might be supplemented with auxiliary hypotheses which explain photography’s epistemic value without reference to transparency, but error theoretic transparency by itself says nothing that can be taken as accounting for photography’s epistemic status. Since the best theories are ones that explain the widest range of phenomena, transparency seems to come off better here than error theoretic transparency in that it can, by itself, explain both what is affectively and epistemically special about photographs. So it is not true that without transparency one is left ‘groping for explanations’ of photography’s affect; for as some of Walton’s own remarks suggest, belief in transparency seems sufficient to account for it as well. To that end, error theoretic transparency is a live option that deserves to be taken seriously; particularly since it explains photography’s affect without the revisionist baggage that comes with transparency. Though where the view seems to measure up less well is that, by itself at least, it remains silent on why photographs are epistemically special amongst picture types.

V. Photographic Displacement
According to photographic displacement photographs allow for displaced seeing, where displaced seeing involves indirectly seeing that such and such is the case about an object on the basis of directly seeing that such and such is the case about a numerically distinct object. On this view photographs allow us to form perceptual beliefs about objects, but without allowing us to see those objects. Photographic displacement holds

that in directly seeing that such and such is the case about a photograph one indirectly sees that such and such is the case about the objects the photograph depicts. To fully appreciate the proposal, firstly distinguish between epistemic and nonepistemic seeing.29 Seeing an object does not entail that one acquires any beliefs about that object, and when this occurs one can talk of seeing objects non-epistemically. Consider, for example, looking in some drawers for your cufflinks at t1, and failing to find them; but later, at t2, on the basis of episodically recalling how things looked to you as you searched, coming to realise the cufflinks were in the second drawer.30 Here you saw the cufflinks at t1 but without believing any content about them; the cufflinks appeared a certain way to you at t1, but you only noticed them later, at t2. When, by contrast, seeing does lead to the forming of beliefs we can say such seeing is epistemic in nature, involving not just the seeing of an object but seeing that such and such is the case about an object, where the ‘that such and such is the case’ specifies the content of the belief acquired about that object on the basis of seeing it.31 Someone who

On this distinction see Quassim Cassam, The Possibility of Knowledge (Oxford: Oxford

University Press, 2007); Frank Jackson, Perception: A Representative Theory (Cambridge: Cambridge University Press, 1997) and especially Fred Dretske, Seeing and Knowing (London: Routledge & Kegan Pau, 1969l).
30

The example is from MGF Martin, ‘Perception, Concepts and Memory’, Philosophical Review,

vol. 101 (1992), p. 749-50.
31

Cassam and Dretske sometimes talk of epistemic seeing in terms of seeing ‘facts’ about objects

instead of seeing that such and such is the case about objects (this may be because that-clauses are sometimes known as factive complements), in a seemingly non-committal way as to the metaphysics of facts. For convenience’s sake I shall sometimes follow suit.

looks in a draw and notices the cufflinks therein sees that the cufflinks are in the drawer, say; they acquire the perceptual belief that the cufflinks are in the drawer.32 Epistemic seeing bifurcates into primary and secondary epistemic seeing. When one sees that such and such is the case about a particular object in a primary way one sees that such and such is the case about an object, and acquires the belief that such and such is the case about it, on the basis of directly seeing the object in question. One can see that the cufflinks are in the draw by seeing the cufflinks, say. But there are cases where one sees that such and such is the case about a particular object indirectly, on the basis of directly seeing that such and such is the case about a numerically distinct object. Call such cases secondary epistemic seeing since they depended upon and entail primary epistemic seeing. Fred Dretske, who was written extensively on this topic, sometimes refers to such episodes of seeing as ‘displaced seeing.’33 Here are some examples of such seeing: I see how much I weigh by looking at the bathroom scale on which I stand. The object I see is the bathroom scale. The fact I learn is a fact about me—that I weigh 170 pounds. This pattern—perceptual object in one place, perceptual fact in another—is familiar. One looks at a gauge on the dashboard in order to see how much gas remains in the gas tank—a fact about an object, a tank located beneath the car, that one does not see… During instrument landing all the important facts are outside, but the instruments and gauges—the objects the pilot
32

One reason for believing in the existence of non-epistemic seeing is that it can help make sense

of how creatures that lack concepts, such as, e.g., infants and some non-human animals, say, are able to see.
33

Fred Dretske, Naturalizing the Mind (Cambridge, MA: MIT Press), p. 41.

actually sees—are inside. In such cases one comes to know that k is F—sees that k is F—by seeing… not k itself, but h. The perceptual fact is displaced from the perceptual object.34 A cook may place a toothpick into her cake to see if it is done… The traffic officer, can, by examining the length of the tire marks produced by a sudden braking action, see whether the driver was exceeding the speed limit. And the way I tell my cigarette lighter is low on fluid is by observing the flame.35 The general pattern of such cases of displaced or secondary epistemic seeing is that one sees that such and such is the case about one object, that the dashboard gauge is all the way to the left, say, and one thereby sees that such and such is the case about a numerically distinct object, that the fuel tank is empty, say, but without actually seeing that second object (in this case the fuel tank). What is required in order for some subject, S, to see that some object, b, is P, on the basis of seeing that another object, c, is Q, according to Dretske, are the following:36

(i) b is P. (ii) S sees that c is Q. (iii) Conditions are such that c would not be Q unless b were P. (iv) S, believing condition (iii) to be satisfied, takes b to be P.
34

Ibid. Dretske, Seeing and Knowing, p. 153. Ibid.

Clause (i) ushers factivity into proceedings. What one can see depends upon how the world is. One cannot see, by the dashboard gauge, that the fuel tank is empty unless this is indeed the case. If the fuel tank is full, then one cannot see, by the dashboard gauge, that the fuel tank is empty. Clause (ii) requires, in regards our current example, that the perceiver sees that the dashboard gauge is all the way to the left. This clause involves primary epistemic seeing; one sees that the dashboard gauge is all the way to the left, and acquires the belief that it is all the way to the left, on the basis of seeing the dashboard gauge. This clause rules it out that someone who sees the fuel tank nonepistemically, someone like the cufflink-hunter, say, can see the state of the fuel tank, which seems intuitively correct. Clause (iii) specifies a conditional; in our present example, that conditions are such that the dashboard gauge would not be all the way to the left unless the fuel tank were empty. If the car battery were flat or if the gauge were malfunctioning, say, clause (iii) would not be satisfied. This is the result we should expect. In such situations one clearly cannot see, by the dashboard gauge at least, what state the fuel tank happens to be in. Clause (iv) requires of the perceiver that they believe the aforementioned conditional holds, and that on that basis they take the fuel tank to be empty. Again, this is the result we want. Someone who sees that the dashboard gauge is all the way to the left but who isn’t aware of their being any connection between this state of the gauge and the state of the fuel tank cannot see, by the gauge, what state the fuel tank is in. Moreover, if someone sees that the dashboard gauge is all the way to the left but falsely believes the link between it and the tank to be malfunctioning, say, then they blind themselves from seeing that the tank is empty.

Significantly, clauses (iii) and (iv) illustrate that episodes of displaced seeing draw not only upon the existence of some counterfactual-supporting link between the state of one object and the state of another object, but also upon the perceiver’s belief that there is such a link. Dretske sometimes calls such belief a ‘connecting belief.’37 The fact that displaced seeing draws upon this belief is not enough to make it the case that beliefs formed about the unseen object in these cases are not really perceptual and are instead inferential, say. The connecting belief, the belief that c would not be Q unless b were P, is a mere background belief. It is something manifested in one’s attitude towards the situation and not something consciously drawn upon as a step in an inferential reasoning process from the fact that c is Q to b’s being P.38 Do photographs satisfy Dretske’s conditions on displaced seeing? The first of Dretske’s clauses, (i), entails that one cannot see, by a photograph, that things are thus and so in the world unless they actually are thus and so. One cannot see, by a photograph, that there are three bullet casings at the crime scene, say, if there are only two present at the scene. Again, what one can be said to see depends (partly) upon how the world is. Clause (ii) involves one’s seeing the photograph in a primary epistemic way, seeing that such and such is the case about the photograph on the basis of seeing the photograph. One cannot see, by the photograph, that there are only two bullet casings at the crime scene unless one sees the photograph. And one cannot see, by the photograph, that there are only two bullet casings at the crime scene, acquire the belief that this is the case, unless one sees the photograph epistemically. I take this to involve not just seeing the
37

Dretske, Naturalizing the Mind, p. 58. Dretske, Seeing and Knowing, p. 158-61.

photograph as a photograph (or, more modestly, as a picture) but more specifically as seeing that the photograph depicts two bullet casings on the basis of one’s pictorial experience of the photograph. One sees two bullet casings in the photograph and thereby sees that the photograph depicts two bullet casings, which, when clauses (iii) and (iv) are satisfied, allows one to see that there were two bullet casings at the scene itself.39 Note that this is not to say that pictorial experience (seeing-in, call it what you like) is necessarily a conceptual affair. One can see objects in photographs, such as bullet-casings, which one has no concept of. The claim is (only) that one needs to have the relevant concepts in order to see, in virtue of one’s pictorial experience of the photograph, that there are two bullet casings at the scene where the relevant concepts are TWO and BULLET CASING. This entails that displaced photographic seeing is not available to creatures that do not possess concepts. This is a result I welcome. Displaced seeing is epistemic seeing and epistemic seeing involves the acquisition of a perceptual belief. We should not expect non concept using creatures to be able to form perceptual beliefs, by any means, because beliefs are built out of concepts. So a non concept using perceiver, who sees the photograph non-epistemically like the cufflink-hunter sees the cufflink, clearly cannot see, by any means, that there are two bullet casings at the crime scene. But this is just as true of someone who sees a photograph as a photograph but who fails to enjoy pictorial experience of the photograph, say, or someone who, lacking the relevant concepts, sees two bullet casings in the photograph without seeing them as two bullet casings. The latter perceiver enjoys pictorial experience of the photograph but without displacedly seeing that there are two bullet casings at the scene since this (in

I remain neutral on what pictorial experience involves.

part) involves forming the belief that there are two casings at the scene and the perceiver in question lacks concepts. Clause (iii) requires that conditions are such that the photograph would not depict two casings at the scene unless this is how many were actually there. Philosophical reflection upon photographic representation, we have already noted, standardly takes is that photographs support such counterfactuals—that had the scene before the camera when the shot was taken differed then the resulting photograph would have correspondingly differed. Had there, contrary to fact, been three bullet casings at the scene, for instance, this is how many the photograph would depict. Note that the same is not true of hand-made pictures. Had there, contrary to fact, been three casings at the scene it does not follow that this is how many would be depicted by a police artist’s sketch, say. A difference in the scene before someone painting may make a difference to the resulting painting, but only assuming that the difference is represented in the artist’s visual experience of the scene and that they notice and believe there to be a difference. Paintings are not intrinsically counterfactual-supporting, but only accidentally so. Clause (iv) requires, firstly, that viewers of photographs believe photographs to be counterfactually dependent upon their subjects and, secondly, that on the basis of believing this to be the case, take it that the photographed object is in the state depicted by the photograph. We frequently acquire beliefs about the world on the basis of seeing photographs (or more specifically, on the basis of our pictorial experience of photographs), so there seems no problem with claiming that we often take it from seeing a photograph which depicts the world as being thus and so that thus and so is indeed how the world is. The more pressing question is whether viewers of photographs

believe such pictures to be counterfactual supporting. It would set the bar too high if we were to require of viewers that they explicitly represent to themselves that photographs are counterfactual-supporting, it is enough that the way viewers typically respond to photographs manifests this attitude. We need to reminded ourselves that this connecting belief is meant to be a mere background belief, and not one that operates as an intermediate step between our seeing that a photograph is in such and such a state and our then reasoning to the world being in such and such a state. The fact that we readily form beliefs about the world on the basis of seeing photographs suggests that our attitude towards photographs carries with it an implicit conception them as counterfactually dependent upon their subjects. Photographs have our confidence. We require them, not hand-made pictures, in passports and on driving licenses to help confirm a person’s identity. We treat them as veracious, as evidence of their subjects. The epistemic practices that photographs are bound up in, the various evidential uses we put them to, suggests that we believe them to co-vary with facts about their subjects. Thus, I suggest, photographs satisfy Dretske’s requirements on displaced seeing. Perceivers who enjoy pictorial experience of photographs can thereby displacedly see that such and such is the case about the objects photographs depict, assuming the aforementioned clauses are met. Photographic displacement entails that photographs furnish perceptual knowledge of their subjects. Assuming that the only candidates for the nature of the knowledge photographs furnish is either testimonial, inferential or perceptual, does this result look

plausible? Testimony looks like it can be ruled out on the basis that it seems to be a process that is mediated by belief, but photographs are mind-independently counterfactually dependent upon their subjects. Inference looks like it can also be ruled out on the basis that, for those of us who are exposed to photographic media on a daily basis, the experience of extracting information from such pictures does not have the phenomenology of an inferential process. If this is right, then the only candidate for photographic knowledge left is perceptual knowledge, and the best theories of photographic representation are ones that can explain how photographic knowledge can be perceptual.40 But how might Walton’s critics, say, who deny photographs are transparent, explain how photographic knowledge can be perceptual? Photographic displacement supplies them with an answer: Photographs allow for displaced seeing. This consideration counts considerably in its favour. Moreover, the view clearly vindicates the peculiar epistemic status of photographs. Paintings, since there is no strong link between them being the way they are and their subjects being any particular way do not allow for displaced seeing, indeed paintings do not require the existence of their subjects. Clause (iii) of displaced seeing is not satisfied here. Moreover, I suggest, we do not typically take it that had the scene been different then the painting would have differed. Paintings do not have our confidence in the way photographs do. Clause (iv) of displaced seeing seems not to be satisfied here either. I suggest that paintings are naturally thought of as testimonies of their subjects.41 Testimonies need not be true, and in this respect they are similar to paintings since,
40

When first coming into contact with photographs it is plausible that some persons may rely on

inferences from the state of the photograph to the state of the world when acquainting themselves with such pictures. But it is not plausible, I suggest, to hold that this is how it is for us.

again, such pictures do not require the existence of their subjects. Most significantly, however, testimonies, like paintings, are intentional processes in that they are mediated by belief. This point has a further significance. Perception is a generative epistemic process in that it is one that enlarges the number of propositions known; perception allows us to acquire knowledge about previously unknown propositions. Photographs, because of their mind-independent counterfactual dependence upon reality, are a generative source of knowledge in this respect. Muybridge’s photographs of animal movement and photographs from the Hubble Telescope have allowed us to learn propositions about our world that no person previously knew to be true. But painting does not seem to be a generative source of knowledge; for, since it is a process mediated by intention, it is difficult to see how one could learn a proposition about a depicted scene which someone else, specifically the artist whose thoughts mediated its production, was not already aware of when painting it. Given the intentional basis of painting, and that it does not allow for accidental depiction, anything learned from such a picture must have already been known by the person who produced it. So whilst photographs are both a generative and transmissive source of knowledge, capable of enlarging both the number of propositions known and (via their reproduction) the number of knowers, paintings only transmit knowledge rather than also generating new knowledge.42 All this suggests that when one sees a painting of an object one is not in a
41

On this point see also Gregory Currie, ‘Visible Traces: Documentary and the Content of

Photographs’, Journal of Aesthetics and Art Criticism, vol. 57 (1999), pp. 285—297.
42

I do not rely on a transmissive theory of testimony in making this point. It is consistent with

what I say that testimony may be a generative source of knowledge. My claim is that for a source of knowledge to be perceptual it must, amongst other things, admit of the possibility of generativity; paintings, I claim, given that they are intentionally mediated, do not meet this condition, ergo they do not

position to see that such and such is the case, form perceptual beliefs, about that object; rather, given the intentional basis of painting, we have good reason to think beliefs acquired from such pictures are testimonial in nature.43 What is perhaps less obvious is how photographic displacement vindicates the affective status of photography. But we do not need to see objects in order to feel in affective contact with them. Following Dretske’s remarks on displaced perception, the photographic displacement theorist holds that photographs are allied with such things as measuring instruments, gauges for example, and also so-called ‘natural signs’ of objects, the marks left on the road by a car for example. There is some precedent for identifying photographs with the latter.44 But seeing signs of an object can prompt significant affective responses with respect to the things they are signs of. Seeing that the enemy has been in our camp by seeing that our flag is defaced and burning, or seeing that a recently departed loved-one had spent their last moments in the front garden by seeing that the grass is half-mown, say, are cases where one does not see the
furnish perceptual knowledge.
43

Though we may, when looking at a painting, say we see that such and such to be the case. But

the mere use of a perceptual verb is no guide to one’s actually seeing, since there are innumerable everyday cases where we say we see that things are thus and so when we are not really talking about seeing. For example, you might say to someone on the phone that you ‘see what they mean.’ This observation also blocks at potential objection to photographic displacement—that we sometimes talk about seeing objects when looking at photographs. But the claim that photographs allow for displaced seeing, whilst paintings do not, is not one concerning, or held hostage to, our use of language.
44

See, e.g., Currie, Image and Mind; Patrick Maynard, The Engine of Visualization: Thinking

Through Photography (Ithaca, NY: Cornell University Press); Susan Sontag, On Photography (London: Penguin, 1997).

enemy or one’s loved-one, but where one sees that such and such is the case about them (that they had been the camp, that they had been mowing the grass), and in a potentially emotionally charged way. Signs are intimately and closely related to the things they are signs of. Moreover, signs reveal things about objects to us, and sometimes the things revealed will have emotional significance for us. This last point is arguably true of testimonies as well. But testimonies do not require the existence or obtaining of what is testified. I can say that I am now in Paris or produce a painting that depicts my being located there, when I am really now in London, say. However, I cannot, now, take a photograph of myself in Paris unless I am actually there. The fact that photographs are counterfactually dependent upon their subjects entails they require the existence of their subjects for their identity, that they have what some call ‘object-involving’ contents.45 But this is not just a reason for having greater confidence in a photograph than a painting (assuming one knows one is looking at a photograph), but for having a stronger emotional responses to it as well. Our implicit belief that paintings, unlike photographs, are not counterfactually dependent upon their subject means there is greater room for doubt as to a painting’s veracity and this is a factor that may plausibly act to quash our emotional reactions to the depicted objects. Significantly, displaced seeing via photographs is a special kind of displaced seeing; where one sees that the world is in a particular state because one sees the photograph is in the very same state. That is, in photographic displaced seeing we frequently come to
45

On object-involving content see especially Gareth Evans, The Varieties of Reference (Oxford:

Oxford University Press, 1982) and John McDowell, ‘Singular Thought and the Extent of Inner Space’, in Phillip Pettit and John McDowell (eds), Subject, Thought, and Context (Oxford: Clarendon Press, 1986), pp. 137-168.

see of an object that it is P not on the basis of seeing that the photograph is Q but because we see that the photograph is also P. That is, photographs afford displaced perception which is of, we might say, a mimetic nature. When we saw that the enemy was in our camp this was not because we saw anything in the camp that looked like the enemy. Such displaced seeing was not of a mimetic nature. But when we see, by the photograph, that the cufflinks are in the drawer this is by our seeing that the photograph depicts the cufflinks as being in the draw; we see in the photograph cufflinks in a draw and thereby see of the cufflinks that are causally responsible for the photograph that they are in the draw. Photographs make facts about objects manifest to us by allowing us to see those very same facts in the photograph. One sees, by the photograph, that Paul’s car is red, that Jake is taller than Kate, that the castle is by the sea, say, because (coupled with Dretske’s clauses and possession of the relevant concepts) one sees in the photograph Paul’s red car, Jake standing taller than Kate, and the castle’s being close to the shore. The fact that displaced seeing via photographs has this mimetic structure may help explain why photographs not only prompt particularly strong emotional responses but why they can also make us feel in close contact with their subjects. This is something that Walton often discusses under the guise of photography’s realism or affect. His explanation of this feature being, of course, that we are seeing the subject of photographs when gazing upon such pictures. But the photographic displacement theorist, by relying on mimetic displaced seeing, can explain this feature too. When we see civil war photographs of dead bodies on a battlefield, although not seeing the bodies themselves, we see that the bodies were strewn on the battlefield in a particular way,

say, on the basis that the bodies we see in the photograph are strewn the same way. A feeling of closeness with these bodies, a feeling of their being made present to us, can then be explained in terms of photographs manifesting to us that such and such is the case about them in a manner that, because of the mimeticity of such seeing, is thereby made particularly vivid to us. By relying on the idea that photographs are signs of objects which are mimetic in nature one can explain why photographs seem to put us in a more intimate relation with their subjects than does, say, seeing a burning flag make the enemy present to us, even though both are cases of displaced seeing. If photographs allow for displaced seeing then a photograph of one’s great-grandfather allows one sees that such and such is the case about him. One can see that he was wearing a particular suit when the photograph was taken, that his hair was parted on the left, that he had a kind look about him, say, and various other facts besides. One acquires perceptual beliefs about one’s great grand-father by seeing his photograph. There are things which, I believe, Walton would not deny. Where the displacement theorist and Walton parts ways, however, is that the displacement theorist denies that in addition to all that you literally see your great-grandfather when looking at his photograph; the displacement theorist denies that your great-grandfather is a constituent of the content of your experience when you look at his photograph. The displacement theorist claims this to be an unnecessary extravagance, given that it is not required in order to vindicate photography’s epistemic and affective status. Note that in endorsing photographic displacement we need not re-draw the lines between seeing and not seeing or introduce a new sense to the word ‘see.’ I suggest that the cases that looked puzzling for Walton, the footprints, fossils, and animatronic

gorillas, are ones we should regard as belonging to the category of displaced seeing. One sees that a fossil is a certain size and can thereby see that the dinosaur bone it is a fossil of is a particular size (likewise for footprints and the boots or feet they are prints of and brushstrokes as well). One sees the state of the animatronic gorillas and can thereby see that the gorillas themselves are in such and such a state. Crucially, one does not see the dinosaur bones and one does not see the gorillas on this view. Moreover, these cases are naturally regarded as ones of mimetic displaced seeing. We see that the dinosaur bones were a particular shape and size by seeing that the fossil is the same shape and size, and we see that the gorillas are feeding themselves by seeing that their animatronic doppelgangers are feeding themselves. Such seeing would, I suggest, because of its mimetic nature, make us feel in close contact with the objects in question. We can thus agree with Walton that: There is a natural kind which includes seeing photographs of things as well as seeing them directly and through mirrors and telescopes, and so forth, but not seeing handmade picture of them.46 But we can deny Walton’s claim that the kind in question is the seeing of objects.47 Instead, we can hold that ‘seeing that’ picks out the kind in question. One can see that the cufflinks are in the drawer by seeing the cufflinks directly, through mirrors and telescopes, and via photographs, but not via hand-made pictures. We could put the difference, if we so wished, as follows: Whilst Walton holds that photographs are transparent to objects, the displacement theorist holds that photographs are transparent
46

Walton, ‘Postscripts to “Transparent Pictures” ’, p. 111. Walton explicitly states his view about the seeing of objects in Ibid, p. 113.

only to facts about objects—transparent, that is, to the states of objects and not those objects themselves.

VI. Conclusion
I have not attempted to show that transparency is false. However, I have suggested we should only endorse the theory if there are no others that can also account for the peculiar epistemic and affective status of photography and that without it we really would be left ‘groping for explanations’, for transparency has severe revisionist consequences which we should avoid if possible. I have suggested that one alternative is to be an error theoretic transparency theorist. Although this view may look attractive to someone who, like Walton, believes that viewers hold photographs to be transparent and is interested in explaining photography’s affect alone, the fact that the view has little to say about photography’s epistemic status, and that there is a question hanging over it whether viewers really do believe photographs to be transparent, appears to count against it. Another alternative, to both transparency and its error theoretic cousin, is to claim that photographs allow for displaced seeing. Seeing via photographs, on this view, is limited to seeing facts about their subjects. This view has the benefits that it explains both photography’s affective and epistemic status, and without the baggage that comes along with transparency; we can explain what needs explaining without re-drawing the limits of seeing or introducing a new sense of the word ‘see’—without having to take a sledge-hammer to the problem. Moreover, the view vindicates the idea that

photographic knowledge is specifically perceptual in nature, which, I have argued, is exactly what we want from a theory of photographic representation. It is difficult to see how a theory could hope to explain photographic knowledge in perceptual terms without claiming that photographs are transparent, but photographic displacement supplies the answer. For the above reasons it is a position that we should find extremely attractive. It has all of the explanatory benefits of transparency, with none of the revisionary costs.

Readers

Recent searches finding this paper

photographic seeing	via Google
photographic seeing	via Google
nigel warburton seeing through seeing through photographs	via Google
photographic seeing	via Google