- The philosopher, Roger Scruton, argues that photographs cannot convey artistic intention because they can only reflect what is given within a scene.
- Paintings however can convey artistic intention – they can represent things other than what is being depicted, that is, they can be ‘seen as’ something else.
- I question Scruton’s sharp distinction between ‘seeing as’ (ascribable to paintings)and merely ‘seeing’ (ascribable to photographs). I argue that these capacities are simply different stages of the same process of perception/cognition.
- I therefore reject Scruton’s argument that photographs cannot convey artistic intention.
- However, I conclude that the model of perception as described in ‘representational terms’, that is in terms of perception as minds having a representation of how things are in terms of mental content, does not help us to understand the nature of aesthetic enjoyment, or indeed anything else.
In 1981 the English philosopher, Roger Scruton, wrote one of the foundational texts in philosophical aesthetics – ‘Photography and Representation’ (Critical Inquiry 7:3; Spring 1981). Scruton argued that photographs cannot be artworks because, unlike paintings, they cannot ‘represent’, that is they cannot ‘sensuously embody … idea(s)’ in ways that give rise to aesthetic satisfaction (Scruton 2008).
In this article, I will set out Scruton’s argument in more detail and then go on to question the key point of his argument, which is that a painting can depict something ‘as’ something melancholic, or as joyful or as macabre and so on. A photograph however is tied to the visual scene it depicts. It cannot depict it in any other way than that state of affairs presented to the lens.
Perhaps the best way of setting out his argument is by taking an example, say: Edward Hopper’s ‘Nighthawks’ (Figure 1).
Figure 1. Nighthawks, 1942 by Edward Hopper; © The Art Institute of Chicago
There are several features that we can set out that make up what Scruton means by ‘seeing as’, but I will confine myself to two of them. Firstly, in ‘Nighthawks’ we are not tied to believing that such and such a situation is being depicted in any type of way since we look at the painting knowing that it comes from Hopper’s imagination and therefore not necessarily from a historic scene. The representation is not tied causally to a state of affairs. Scruton places considerable emphasis on the ability of paintings to be fictive – ‘unless it were possible to represent imaginary things, representation could hardly be very important to us’ (Scruton 2008, p.146).
Secondly, Scruton emphasises that ‘Seeing as’ is like thought in that it has intentionality. In the absence of Scruton setting out clearly what he means by ‘intentionality’, one can adduce from his arguments that it is based on the notion of mental (or intentional) content: to have intentionality is to have conceptual content. Here we appeal to the idea of representation: a state has content insofar as it ‘represents something to be a certain way’. For Scruton therefore, it seems that representational content is conceptual, a hallmark of ‘Seeing-as’.
In ‘Nighthawks’, ‘the four anonymous and uncommunicative night owls seem as separate and remote from the viewer as they are from one another’ (Chicago 2013). Hopper acknowledged that in Nighthawks ‘(I was) unconsciously, probably, painting the loneliness of a large city’. In this painting, therefore we see the juxtaposition of an empty dark street with a group of people talking in a brightly lit bar. We also see the people as guarded: hats are on, shoulders are hunched; and we as viewers feel disconnected from the scene, separated on the outside looking in by virtue of the people being set apart on the other side of a window, distanced by their closed postures. We see the painting as a whole as reflecting Hopper’s thought about the alienation of urban living. In short, we see content that represents the scene to be a certain way.
In summary, Hopper’s painting shows all the supposed hallmarks that Scruton attributes to ‘Seeing as’. ‘The element of thought involved in ‘Seeing as’ lies clearly in the field of imagination: it is thought that goes beyond what is believed or inwardly asserted, and beyond what is strictly given in perception’ (Scruton 1998, p.112, my emphasis).
The corollary of this argument is that ‘Seeing’, as opposed to ‘Seeing as’, does not display these hallmarks. Imagine that we had walked by as Hopper sat at his easel in front of Phillies Bar and looked onto the same scene. How would we have seen this scene?
This is a difficult question to answer as we have now already been educated into a way of seeing it. Scruton would need to argue in the following way: If we simply looked at the scene as we walked by, we would see the scene in an unintentional i.e. ‘causal’ way. We would not see the scene as something other than just the objects depicted. Our seeing would be doxastically tied to the objects we believed we were seeing and we would not tend to give it much further thought unless there were a reason for doing so. Paraphrasing Wittgenstein but using more common terminology, we would look at the scene but not see it (Wittgenstein 1945, §211).
Suppose however we took a photograph in such a way as to approximate some features of the painting. Suppose we had walked past Phillies Bar and taken a photograph of the scene using expired Kodachrome film to approximate the colours that Hopper had achieved. Why does Scruton deny to us the same thoughts and feelings that we experience when we look at the painting, the same endpoint? Scruton’s answer would be that he would not deny to us these thoughts and feelings, (indeed, how could he?). Scruton would say that the photograph would be a photograph of a representation, not a representation in itself. The photographer has no degrees of freedom in the way in which she can change the pattern of light striking the film surface at the moment of shutter release and therefore cannot affect the pattern of light in a way that accords with her thoughts. The use of expired Kodachrome is but a painterly effect. The composition of the photograph, i.e. its design features, is the making of a representation prior to the taking of the photograph. Scruton holds that an ideal photograph stands in a causal relation to its subject, that is, its endpoint is a ‘copy’ of an appearance. Photography, the act of picture-taking, provides a clue to what an ideal photograph is: it lies in the ‘take’ part of ‘picture-taking’. One cannot take what is not available to be taken. Picture-taking depends on what is present and resemblance to subject is its defining characteristic. ‘Taking’ is analogous to ‘Seeing’. ‘Seeing’ is the taking of what is visible. Like mirrors, photographs simply reflect what is seen, and ‘Seeing’ just takes the visible.
To summarise: Scruton distinguishes between painting’s ‘Seeing as’ and photography’s mere ‘Seeing’, based on an assumed relationship between ‘Seeing’ and the visible. The underlying distinction is between interpreting, which is a form of knowing, and seeing, for ‘Seeing as’ (interpreting) is to understand a picture in a certain way, that is in the way an artist intends a picture to be seen. ‘Seeing as’ is a knowing that is not doxastically tied to the objects seen.
Photographers however do not have the fine-ness of control to inflect in a photograph a way of seeing that bears these characteristics. ‘Seeing’ is missing at least one if not all (it is not clear) of the characteristics of ‘seeing as’, effectively disqualifying it as capable of carrying aesthetic experience. For Scruton ‘Seeing as’ is an all-or-nothing phenomenon, just as is seeing a duck or a rabbit in the oft-quoted example of a Gestalt phenomenon.
One way to challenge Scruton is by questioning his understanding of the relationship between ‘Seeing’ and the visible. Central to his argument is that when we look at a painting we see certain features set out by the artist that guide us to seeing the painting in a certain light. We have already examined an example of this. ‘Seeing as’ is similar to thinking in that it employs concepts. ‘Seeing’ however mirrors what a camera does. Like a recording device, it simply reflects back what is already there. But can the divide between ‘seeing as’ and ‘seeing’ be so sharp? Just what is the difference between cognition (aka ‘seeing as’) and perception (aka ‘seeing’)?
At this let us imagine an alternative conception through a series of assertions and ask ourselves whether it is so improbable.
Assertion 1: ‘Seeing’ is penetrated by cognition and is conceptual ‘all the way down’. A Kantian view supports the notion that visual perception is representational in that it is shaped by concepts. Objects in our visual field are individuated through concepts, otherwise we would not recognise those objects, we would not recognise the X in ‘seeing X’ or the X in ‘seeing X as Y’. The objects seen are contents of experience but need not be brought into thought in order for us to be able to navigate the world. ‘Seeing X’ reflects the representational aspect of visual perception that presents the world to the subject as an objective world about which we can form true or false judgments.
There is also a negative argument here. If visual perception was not conceptual ‘all the way down’, one might ask how thinking could engage with our senses, how ‘Spontaneity’ could accord with ‘Sensibility’ for one would expect some common substrate to provide the friction for this accord (McDowell 1996).
Further, empirical studies suggest that visual processing can be segmented into stages, each taking different but linked functional roles. Taking in a scene involves increasing semantic organisation the longer a scene is given attention (a point that supports the view of a non-categorical difference between ‘Seeing’ and ‘Seeing as’).
The point to be emphasised here is that the scientific account is consistent with an account of ‘Seeing’ that is conceptual from a very early stage and that a sharp divide between ‘Seeing’ and ‘Seeing as’ cannot be easily discerned. It is also consistent with the idea that aspect perception forms an integral part of early stage parsing.
It would be tempting to ‘fit’ this description to a philosophical account. Pre-attentional processing is akin to a phenomenal awareness, (‘sheer receptivity’; ‘non-conceptual content’). With increasing attention, concepts come into play selecting behaviourally relevant objects for attention (‘spontaneity’; ‘conceptual content). We conceive a linear process in which objects that come under concepts are exhausted under those concepts such that no residue is left. But this would be far too simplistic: such a conception would not do justice to the ‘what it is like’ quality of an ‘aesthetic experience’ and further, ‘sheer receptivity’ would fall to be criticised under the ‘myth of the Given’.
The consequence of the forgoing is that it is but a short step from accepting the conceptual nature of ‘Seeing’ to regarding ‘Seeing as’ in terms of a ‘Seeing’ that is suffused with thinking, the conceptual nature of ‘Seeing’ allowing it to have sufficient isomorphic structural content to be apprehended by thought or the imagination through concepts.
Assertion 2: ‘Seeing is very different from thinking’ – ‘Seeing’ has a non-propositional component, which provides the phenomenal content that underpins the process of ‘picturing’.
Seeing a pink ice-cube, say, is different to thinking a pink ice-cube. In the words of Sellars, ‘… there is all the difference in the world between seeing something to be a pink ice-cube and merely thinking something to be a pink ice-cube … over and above its propositional character … [perceptual] thinking has an additional character by virtue of which it is a seeing as contrasted with a mere thinking’ (Castaneda, Calderon and Sellars, 1975). The pinkness of the ice-cube has an immediacy and presence that is different from the way it is in thought. If we subtract from ‘seeing pinkness’ the thought of pinkness, we are left with a remainder that is the phenomenal content of seeing. The conceptual apprehension of ‘pinkness’ underdetermines what it is. This phenomenal content is non-conceptual and therefore not available to scrutiny through thinking (for to do so would be to conceptualise it). We can only intuit it as the result of transcendentally framing an inquiry into the conditions that must pertain for us to have this immediate experience of pinkness that fades when we leave its presence for its half-life is measured in seconds.
There is a further point to be made here. To distinguish ‘lookings’ from ‘mere thinkings’ (Sellars, Rorty and Brandon 1997) we exercise concepts about phenomenal consciousness and in so doing we become practised, and possibly skilful at attending to the nature of this content (Coates 2007). Just as we can become practised at seeing intentions in paintings, we can also become practised at looking at photographs in ways that reveal novel relationships between objects.
Taking assertions (I) and (II) together, the distinction between ‘Seeing as’ and ‘Seeing’ finally comes into focus. The contents present in ‘Seeing as’ and ‘Seeing’ are well accounted for as conceptual content. Seeing the cube as ice is very like seeing it while believing in its iciness. In this regard, seeing is like thinking. Some of the visual content in seeing is present in the mode of being thought. However, there is also an essential difference since not everything in a visual experience is present in this way: ‘We see no only that the cube is pink, and see it as pink, we see the very pinkness of the object…’ (Sellars 1982, p.89).
Assertion 3: ‘Seeing’ guides thinking and imagining by virtue of non-conceptual content in the form of aspect perception. Scruton identifies aspect perception with thinking but the image I want to get across is that aspect perception is a condition of ‘Seeing’.
We see references to the notion of the phenomena of gestalt forms in Kant , but it is to Christian von Ehrenfelsin and Edgar Rubin that we owe a more elaborate conception. Wertheimer, Koffka and Köhler significantly developed the notion and more recently Arnheim (Arnheim 2006) pioneered its application to Art. The notion of aspect perception in terms of gestalt forms was also recognised by Wittgenstein.
For my purposes two aspects of Gestalten will serve to make the necessary connections with the form of ‘organisational aspect perception’ that is relevant here: the first is the notion of basic navigational gestalt forms and the second is the Principle of Prägnanz, that is: the principle that the organisation of any structure in nature or in perception, will be as ‘good’ as the prevailing conditions allow, the qualifier ‘good’ being sensitive to the context of the particular environment in question .
The basic gestalt form ‘Figure/Ground’ provides a perceptual organisation through which we individuate objects such that objects stand out from their backgrounds. The figure emerges from the ground. Merleau-Ponty’s discussion of Cezanne’s paintings (Johnson 1994, p.65) in terms of the conditions of perception that give rise to the experience of his paintings is noteworthy for its dependence on gestalt forms. Deleuze discusses Bacon in similar terms (Deleuze and Smith 2004, p. 34).
The principles are not restricted to paintings however. As convincingly argued by Lopes the design features of pictures commit one to a way of seeing, and therefore also denies to one another way of seeing. Pictorial content is aspectually structured – ‘pictures … are essentially selective, because every picture is explicitly noncommittal in some respect’ (Lopes 1996, p.125).
Wittgenstein lists examples that purportedly demonstrate the phenomenon of aspect perception, the most notable example for present purposes being ‘organisational aspect perception’: the occasion in which one can see a row of four equidistant dots either as two groups of dots or as two dots in the middle bracketed by a dot on each side:
‘…How does one teach a child … “Now take these three dots together!” or “Now these belong together”? Clearly “taking together” and “belonging together” must have had another meaning for him than that of seeing on this way or that. – And this is a remark about concepts, not about teaching methods’ (Wittgenstein 1945, Part 2 §220-221).
This particular example of aspect perception is notable because I contend it is an aspect perception that is situated at the level of ‘Seeing’, which affords an opportunity to rethink the relationship between ‘Seeing’ and ‘Seeing as’ in different terms other than the mutually exclusive classes posited by Scruton.
I have already argued that ‘Seeing’ is cognitively penetrated and therefore Wittgenstein’s reflection that aspect perception is concept-laden, that is seeing something as X presupposes a mastery of the concept X, is no threat to my proposition. Cognitive penetration suggests that to ‘see X’ is to ‘see X as X’ for as we have already noted ‘seeing X’ requires that we ‘see X as X’ in a process of individuation and recognition. In this regard ‘seeing X as Y’ is just a special case of ‘seeing X as X’. ‘Seeing X as X’ therefore is a form of aspect perception in the same way that Scruton reserves for ‘seeing X as Y’. For example, it would be possible when confronted with an X, for a person to see X as a Y, as in a hallucination. ‘Seeing X as X’ is a form of aspect perception but one where the X remains unasserted. In Merleau-Ponty’s words ‘Perception hides itself from itself… it is of the essence of [awareness] to forget its own phenomena thus enabling things to be constituted’ (Merleau-Ponty, 1962 p.58). Aspect perception is intrinsic to ‘Seeing X as X’ for X to be recognisable as X.
Now why need I posit a non-conceptual aspect to perception? Theories of perception that recognise non-representational phenomenal properties need to provide a plausible account of the role such properties play in the perception of physical objects. The short answer is that it helps us to understand what things are like. When I admire Caponigro’s photograph ‘Running White Deer’ (Figure 2):
Figure 2. © Paul Caponigro, ‘Running White Deer, County Wicklow, Ireland 1967
I can imagine the loping movement of the deer, the sound of light hooves drumming the grass, the smell of the pine resin of the forest. It is as if each of these senses is made present to us in an attenuated way. The visual parts of this perception are partly made present through gestalt forms.
However, Köhler held that this ‘organization’ of the visual field through gestalt forms is a ‘sensory fact’, implying direct contact with sense-data or, to use McDowell’s phrase, a ‘bare existence’ in the constitution of an experience. Wittgenstein was resistant to this idea, as am I. Such a conception tips us into the ‘myth of the Given’, which can be avoided either by taking a McDowellian approach of denying any role for non-conceptual content or by taking one that offers a version of experience that affords a role for both conceptual and non-conceptual content but not in a way that equates perception with sense-data. Making sense of gestalt phenomena requires that we adopt the latter approach.
The issue comes to a head in thinking about organisational aspect perception. As discussed by Malcolm Budd (Budd 1987), in order for organisational aspect perception to be aspectual, our way of experiencing the structure of a scene would need to be able to change without the structure of the scene actually changing. In Wittgenstein’s terms ‘if you put the ‘organization’ of a visual impression on a level with colours and shapes, you are proceeding from the idea of the visual impression as an inner object. Of course, this makes this object into a chimera; a queerly shifting construction’ (Wittgenstein 1945). The point Wittgenstein makes is that for there to be aspect change at organisation level on a par with colour and shape (held to be sensory by Köhler) whilst the organisational properties of a scene remain unchanged, as they must for it to be aspectual change, then we must posit a further visual feature other than colour or shape. Hence, we are tempted to credit a visual impression about organisation with sensory rather than with intellectual processes.
So, we are taken to a potential difficulty: on the one hand aspect perception is based on changes to sensory entities such as colour and shape. On the other hand, aspect perception is based on changes to intellectual entities like thoughts. Such an inconsistency creates doubt over the nature of organisational aspect change. Such doubt is resolved however when we posit a two-component account of perception where the phenomenal non-conceptual component of perception remains constant during an aspect change whilst the representational conceptually-mediated element changes during aspect perception. In this model ‘Seeing’ is rich in the former component but still shaped by concepts, whilst ‘Seeing-as’ is rich in both components.
Assertion 4: “‘Picturing’ is what happens when ‘Seeing’ and thinking and imagining together form representations of the kind that sensuously embody ideas”.
Bringing assertions (1) to (3) together, I have asserted that to ‘See X’ is to ‘See X as X’ engaging conceptual capacities, that ‘Seeing X’ is different to ‘Thinking X’ by virtue of its phenomenal content, that ‘Seeing X’ is to ‘See X as X’ partly by virtue of aspect perception at both conceptual and non-conceptual levels.
Picturing is a process with increasing regulatory complexity and normativity as we place more attention on a visual scene, but also crucially one that brings together phenomenal aspects that enable us to navigate a scene well. ‘Picturing’ brings together into one representation ‘Seeing’ and thinking and imagining, ‘Seeing’ with its a non-representational, non-conceptual phenomenal aspect. The phenomenal aspect guides the forming of a relationship between a subject and world but not in a way about which we can form true or false judgments since it remains bodily situated. It presents itself by reference to a way of moving in the world like a form of tacit understanding.
The final assertion (5) is ‘In navigating well, we aesthetically enjoy the world’. Evans points out ‘what is involved in a subject’s hearing a sound as coming from such and such a position in space … we do not have to think or calculate which way to turn our heads’ (Gunther 2003, p.50). Heidegger expresses this sense as well as anyone has in his discussion of ‘readiness-to-hand’ in which ‘things never show themselves proximally as they are for themselves, so as to add up to a sum of realia and fill up a room. What we encounter as closest to us … is the room and we encounter it not as something “between four walls” in a geometrical spatial sense, but as equipment for residing’ (Heidegger 1962, p.98 §68 and §69).
The way we navigate the world contributes to how we regard objects aesthetically. When we look at paintings or photographs we navigate routes through them in much the same way that we navigate ourselves through a busy scene. The intelligible world stands out as foreground by virtue of the plenitude of a background. Gestalt forms provide the intuitive structures that guide us and the substrate upon which certain representations can be both sensed and felt. Here, a parallel with Kant’s ‘sensus interior’ is difficult to avoid. Feeling sensations and simultaneously apprehending ‘an attentiveness to the state of the subject’ (Kant, Dowdell and Rudnick 1996, §15, 32, VII, 153) is an idea that echoes the notion of ‘affordance’.
Before I summarise the picture presented in Part 2, let me illustrate organisational forms of aspect perception in street photographs to give some credence to my argument.
Starting with Cartier-Bresson’s photograph of race-goers in County Tipperary (Figure 3):
Figure 3 – © Henri Cartier-Bresson/Magnum Photos, Courses de chevaux,
Thurles, Tipperary County, Munster, Irlande, 1952.
Several gestalt organisational principles act together in this photograph, ’Similarity’, ‘Proximity’ and ‘Common Fate’: we see the race-goers as similar in virtue of their dress and their proximity to each other. We also see them as members of one group thereby sharing a common fate. By dint of the forms of organisation that Cartier-Bresson saw, we ‘see as’ – we tacitly understand the idea being expressed of group membership and we feel its impact by virtue of being excluded from membership as a result of facing the opposite way. This is a bodily sensitivity as we navigate through the photograph as if navigating through a real scene before us.
Kertész’s simple image, ‘Le Balcon, Martinique’, (Figure 4), works on the simple rule of thirds and quarters with a disrupting diagonal.
Figure 4 – © André Kertész, ‘Le Balcon, Martinique’. 1972
But central to the image is the form of a man through a translucent frosted window that works on the principle of figure/ground. Even if the sky had been cloudless and therefore mono-tonal thereby rendering almost an abstract image, the pictorial information would have been sufficient to enable us to recognise that we are looking onto a sea-scape. The form of a figure looking at a horizon is such an archetypal one in our bodily awareness that it predisposes us to navigate it in a certain way.
Kertész’s photograph (Figure 5 – ‘Carl Shurz Park, New York’), illustrates the form of organisation involving gestalt principles of ‘figure/ground’, the ‘good curve’, and ‘closure’. In this example, Kertész sets up a tension by placing the man in a coat at the edge of the frame looking beyond the frame onto a scene we cannot see. This lack of closure is counter-poised by a curve facing into the frame to the point of finality. We balance the lack of closure on the left of the image with the strong closure to the upper right. In fact, most people will navigate the picture by finding closure from the curve before attending to the man in a coat.
Figure 5 – © André Kertész ‘Karl Shurz Park’, New York
The point being made here is that perception is constrained to seek order even when it is difficult to find. This need seems to be culturally conditioned through learning and there is a sense therefore that competence in the photographer and in the viewer is necessary. The Gestalt principle of ‘Closure’ demonstrates this in particular:
Figure 6 – © André Kertész ‘Stairs’, 1955.
Here in Figure 6, Kertész times his shutter release in a way that allows the perceiver to complete a form, albeit in an eccentric way.
From these examples, we see the Gestalt Principle of Prägnanz as situating perception within a dispositional frame of reference. Simple and clear forms are noticed in preference to complex and unclear ones. Perceptual navigation is by the easiest route but the sense that I want to convey here is that it is still subject to constraint, a notion common in Japanese aesthetics and art (Aviman 2014). In the experience of a picture or a scene, in navigating a way through one, we look for the path of least resistance. Many street photographs depict, for example, the constraints suggested by the boundaries between pavements and roads, kerbs or the view through a window, or down a dimly lit passage.
A few examples are sufficient to illustrate the point. In Kertész’s ‘Place Gambetta, (Figure 7), the gestalt form of similarity underpins the structure of the scene but we also have the principle of Prägnanz at work through uniform regularity and the diagonal boundary between road and pavement.
Figure 7 – André Kertész, Place Gambetta, 1928
Josef Sudek’s photograph (Figure 8) works by virtue of the framing effect of the tunnel entrance, illustrating the Prägnanz principle of moving from dark to light. There is a sense that the figure is optimally situated in the frame from the perspective of the viewer. All other placements would have created more tension.
Figure 8 – by Josef Sudek, © Anna Fárová The Sonja Bullaty and Angelo
Lomeo Collection of Josef Sudek Photographs
Let me now summarise If we regard the relationship between ‘Seeing’ and the visible as being one that partly determines the thinking and the imagination that come into play when we perceive a painting or a photograph, and we view ‘Seeing as’ as ‘Seeing’ that is suffused with thinking, rather than being categorically different, then we start to better understand how photographs represent. The examples of street photography above all illustrate the skill of the photographer in ‘framing’ an instant moment in such a way as to give licence to gestalt forms in eliciting aesthetic experiences. The distinction that Scruton makes between ‘Seeing as’ and ‘Seeing’ is better conceived in terms of a conceptual/phenomenal axis rather than the complex five point criteria he posits.
I have argued that ‘Seeing as’ is a special case of ‘Seeing’, and not categorically different in the manner set out by Scruton. ‘Seeing’ is categorically different to thinking in that it has a non-conceptual phenomenal component that is ‘guided’ by aspect perception. I have introduced the notion of ‘picturing’ in terms of representations that predispose us to navigating a scene in a particular way. Picturing is a blend of ‘Seeing’, thinking and imagining: ‘Seeing’ contributing a role for organisational aspect perception in navigation; thinking contributing a role in self-reflexion; and imagining contributing a role in fictional competence, all combining towards an aesthetic experience. The extent to which we are able to navigate a painting or a photograph, or for that matter any scene, with ease, that is, in accordance with some form of bodily expectation through the dispositional effects of Prägnanz, contributes to our sense of aesthetic enjoyment. The disposition to navigate through a scene informs how a picture is to come together as a picture.
Paintings and photographs are both forms of picturing and can be understood as tokens of a representational type, each with distinctive modes of representational expression particular to their respective media, but both sharing a common denominator: ‘Seeing’. Scruton’s distinctions between paintings and photographs, between ‘Seeing as’ and ‘Seeing’ are not substantive in metaphysical terms nor effective in explaining the sensuous embodiment of an idea, which must be posited in terms that make sense from a first-person perspective of knowing ‘what it is like’ to see the ‘pinkiness’ of an ice-cube. The relationship between thinking (‘Seeing as’) and perception (‘Seeing’), if they are different things, is complicated and we will go wrong if we try to draw hard boundaries between them.
The examples from street photography illustrate how a photographer can shape the way a viewer navigates a picture through the act of framing. An idea is expressed by virtue of how the objects in a frame relate to each other and to objects outside the frame. A crucial aspect of street photography is the ‘ephemeral connection between unrelated things’. Alex Webb’s photograph in Figure 9 illustrates this well. The arrangement of the figures within the frame is an ephemeral pattern unseen by all except by Webb at the moment of shutter release. Scruton’s admonition that this picture is a photograph of a representation is to miss the point that ‘Seeing’ by its nature is representational in his own terms.
Figure 9 © Alex Webb/Magnum Photos, ‘La Calle’, Nuevo Laredo, Mexico 1996.
In terms of Q we do not need to see photographs in terms of the objects that they reflect. The extent to which a viewer navigates a photograph with ease, perhaps more especially a complex image, contributes to the aesthetic pleasure derived from the photograph. A photographer has control over this through framing. I have illustrated that photographs, by virtue of organisational aspect perception, can represent thoughts in a manner that Scruton reserves for paintings. It now remains for me o examine possible objections to this proposal.
Possible objection to my assertions
Objection: ‘Seeing as’ is categorially different to ‘Seeing’.
I contend that ‘Seeing as’ and ‘Seeing’ are not different in virtue of the conditions [A to F] set out by Scruton but in virtue of ‘Seeing’ being conceptual ‘all the way down’ and phenomenal ‘all the way up’ [I and II] and that any difference lies in the degree to which ‘Seeing’ has been suffused with thinking and imagination. Scruton attacks the notion that all “Seeing’ is ‘Seeing as’. ‘Seeing as’, positioned as a branch of imagination, goes beyond the content of present sensory experience, beyond the beliefs we form when we ordinarily see something. ‘Seeing’ is not volitional in the same way that seeing an aspect can be as a result of a decision to do so. So, when through a ‘trick of thought’ we enforce a change of aspect and make a swaying tree seem as a man waving his arms, this does not establish the point. Seeing is logically related to belief. However, ‘Seeing as’ involves the entertaining of an unasserted proposition and therefore if we are to believe that aspect perception is involved in ‘Seeing’, then all ‘Seeing’ entails the entertaining of unasserted propositions. From this Scruton draws a tongue-in-cheek reductio ad absurdum: ‘The consequence must be that animals, which can have beliefs but no other kind of thought, cannot see’ (Scruton 1998, p.115).
Seeing can involve the entertaining of thoughts but not necessarily so, as Scruton acknowledges. When we see a swaying tree as a waving man, we think of the tree as something else. The imaginative thought becomes embodied in an image of the tree. There is no change of aspect, ‘only the dawning of an aspect where previously there was none’ (Scruton 1998, p.115). The cause of the view that all seeing involves ‘Seeing as’ lies in the vagueness of the locution ‘to see X as Y’. Sometimes, as in the example of the tree, we actually mean ‘we took X for Y’, that is, we believed X to be Y. This locution is analogous to ‘Seeing’ as it involves belief.
But at what point, we may ask, does the entertainment of thoughts presage aspect perception rather than ‘taking X for Y’? At what point in the process of picturing do we put aside a belief in what we see and entertain in imagination alternative thoughts? For Scruton, the answer would need to be ‘all at once’! One either ‘Sees’ or one ‘Sees as’. The former does not develop into the latter because for the former, aspects are not a possibility, for if they were, then we would be ‘Seeing as’.
This seems an odd argument. It would make more sense to acknowledge a continuum of possibilities between ‘Seeing’ and ‘Seeing as’ such that the latter is but a case of the former, but more suffused with thinking or imagination. Before we can see X as Y we would need to have established X, which means that we would need to see X as X. Seeing X as X is antecedent to seeing X as Y. If we saw X as Y immediately we would just be seeing Y. Our imagination must be able to make a connection between X and Y and therefore move from the thought of X to the thought of Y. So, X must pre-date Y. Of course, this is trivial, but it does point to major doubts over Scruton’s insistence of a sharp divide between ‘Seeing’ and ‘Seeing as’ and makes his position difficult to accept.
Objection: The proposal does not take account of the fact that photographs are fictionally incompetent.
Scruton lays great store in the fictive qualities of paintings, which he sees as being associated with ‘Seeing as’. Due to their causal nature, photographs cannot depict a state of affairs other than that presented to the camera. How would my proposal allow for this? Quite simply, Scruton’s notion of what constitutes a fiction is too narrow to account for all fictional devices. The fact that ‘Seeing’ is conceptual ‘all the way down’ gives it the ground for aspectual seeing.
Given its importance to his argument some detailed attention is necessary. As mentioned, we can discern three separate senses of fiction in Scruton’s essay: fiction as in the involvement of the imagination, fiction as in ‘truth conditions’ that give a work meaning and finally fiction as in the indifference to truth conditions in maintaining an aesthetic attitude. Taking each in turn:
In the first sense, put simply one cannot take a photograph of an imaginary scene. A photograph is a ‘good idea of how something looked’ based on the relationship of resemblance (Scruton 2008, p.149). At first blush this might seem to be a reasonable proposition. A photographic film consists of an emulsion made from silver halide crystals which change into metallic silver when light falls onto them in proportion to the amount of light available. The latent image created and subsequently developed with chemicals is contingent on the reflection of light from objects in the scene. If an object is not there, then there will be no recorded image of it unless we resort to the painterly effects of double exposure or post processing.
But what about the converse situation? It is perfectly possible to take a photograph of a scene and yet record a state of affairs different to the one perceived at the moment of shutter release. The following photographs illustrate such possibilities:
Figure 10 – © Tony Cearns,’ Shadows after shadows’, Liverpool.
In Figure 10, I released the camera shutter whilst people walked between me and the wall. However, the resulting photograph does not capture those people. The effect of the long exposure used to capture the scene in relation to the speed of movement of the people walking through it means that moving people do not register in the latent image. One might argue that the causal way that film records light has resulted in an image that does not depict the scene as perceived at the time.
Australian street photographer, Trent Parke, is well known for his arresting images. In this photograph (Figure 11) the specular highlights of the person caught in the bright sunlight are enough to burn out the person as if to leave a cardboard cut-out for the position that the person occupied in the scene. This effect has been created purely by the causal interaction between light and photo-sensitive film.
Figure 11 – © Trent Parke ‘White Man’ 2001.
A causal process has recorded a state of affairs in a different way to that presented to the photographer’s eye due to the differences between retinal and silver halide sensitivity.
Another example is given by Robert Hopkins (Hopkins 1998) who cites a photograph of a black dog appearing as a green dog due to the colour shift effect of over-exposure. Examples of these types of effect abound in photography and are often seen as anomalies by photo-editors to be corrected by post processing. The effects however are volitional rather than accidental.
So, we ask: in what way are these photographs not fictive? The state of affairs depicted in a photograph are quite different to the state of affairs presented to the photographer. Scruton might respond as follows: in the first example (Figure 9) the photographic process has been manipulated in such a way as to prevent the latent image of people from forming. The analogy with painting would be to paint over a scene with a correction. In the case of the long exposure photograph, the latent images of people are formed temporarily but replaced by the longer-standing, and therefore ultimately more reactant, latent images of immobile background objects, making the latent images of people so faint as to render them undetectable under normal viewing circumstances.
In the second (Figure 11) and third examples Scruton would argue that such photographs are either photographs of representations or have been embellished with ‘painterly effects’ and are not ideal-typical. Over-exposure in these cases is an artefact of the photographic process and not the normal result of an ideal photograph. The photograph by Trent Parke is a photograph of a representation not a representational photograph. Ideal photographs can only come from ideal cameras used in an ideal fashion. The camera cannot represent something, but only point to something; ‘a gesturing finger would do as well’ (Scruton 2008 p. 152). Scruton would need to resort to his methodological device of the ideal-type in defence of his argument. We could reply that restricting our analysis to what is only allowed within a standard set up by ideal photographs misses the point about photography, but then we would be playing into his hands, as we want to demonstrate that ‘ideal photographs’ can represent, as argued in Part 2.
Scruton sets up an ideal photograph as one which simply reflects like a mirror. Indeed, one might think that the ideal ‘ideal photograph’ would be one that involved no human agency such as we might find in a still from a street CCTV. We do not however impose such conditions on ‘ideal paintings’, for in that case to establish a fair comparison, an ideal painting would be one painted without human agency, say by an automotive robotic paint spray gun gone out of control. But then we would lose a key hallmark that Scruton identifies with paintings, that of artistic intentionality. Hence to preserve some symmetry across the ideal-types that enables it to do useful work, we must rule out the possibility of an ideal photograph as one not involving a competent photographer intending to express an idea. As already argued, the skill of a street photographer is to catch a frame which encapsulates the idea that she wants to convey. ‘A gesturing finger would do as well’ but only at the end of a skilled photographer’s hand.
The second sense of fiction concerns truth conditions. Scruton concludes ‘There seems no justification … for thinking of representation in terms of reference’, in the way the term is employed in linguistic scholarship (Scruton 2008, p.145). The theory he denies would say that we can read the meaning of a picture like a text, and therefore understand the artist’s thought that gave rise to it, in an analogous way to grasping the meaning of a sentence when we read it.
Scruton finds no use for the notion of ‘understanding’ as in to grasp pictorial meaning expressed through a putative grammar of pictures. Pictorial understanding cannot be secured through rule-following – ‘(It is) not secured either by rules or by conventions but seems to be, on the contrary, a natural function of the normal eye’ (Scruton 2008, p.145). Whereas we can concede the point that visual perception does not have a syntactical structure similar to well-formed sentences, it does not follow that ‘pictorial’ understanding does not follow certain constraints, that there is not a normative dimension to perception, that cognition does not go ‘all the way down’ in Brandom’s phrase (Brandom 1994), as I argued in Part 2. To set up ‘mere’ (my inflection) rule-following in such a way suggests a diminished notion of what it is to follow a rule. Rules suggest a certain normative compulsion, not just a heuristic process for obtaining semantic outputs given some syntactical inputs. We should not dismiss the possibility that rules that determine what a perception is ‘of’ are somehow connected with rules that form our dispositions to behave and talk in certain ways, for certainly that is the force of my argument intended by Part 2. We can agree that obtaining visual literacy is not like learning a language but it is like learning a something, a skill, a way of navigating the world. Pictures may lack syntactic devices for making propositions in a direct sense, in Sol Worth’s phrase ‘pictures can’t say ain’t’ (Worth 1981), but they are of forms which act through conceptual content and also through non-conceptual expressions that give rise to dispositions, both constituents of the way we structure our experiences.
Putting aside his misplaced views of rule following, we can support Scruton in his view that an account of reference will not give a full account of understanding but will support the invocation of reference in the truth or falsity of a painting. Truth is not a property of the painting, in the way that it is the property of a sentence, but rather it characterises the thoughts that animate our perception when we see a painting with understanding. Scruton identifies these features in terms of what we see in paintings, a relationship between what we see in the painting and the world, that is ‘the intentional objects of sight’ corresponding to the nature of the subject. Paraphrasing, understanding paintings involves understanding thoughts which can be true or false. But by extension therefore, as photographs do not embody thoughts for Scruton, photographs cannot invoke references to truth or falsity as possibilities. But this is surely wrong.
I have argued that a key aspect of street photography lies in the framing of a scene. The act of framing is often to strip away the context in which a scene is situated, a context that positions the objects within a veridical relationship. Shorn from its context a photograph becomes ambiguous as to what is being represented and this aspect naturally invokes a reference to the truth or falsity of a scene. An example is given by the following photograph by Tony Ray-Jones (Figure 12):
Figure 12 – Tony Ray-Jones, (title withheld) 1967.
Without a title, shorn from its context, we are left to ponder the circumstances that gave rise to the scene in a process of thoughtful inquiry, invoking possibilities of interpretation and therefore references to truth or falsity. Paraphrasing Wittgenstein ‘much can be gathered’ from the thought that is embodied in this photograph.
The final sense of fiction concerns aesthetic attitude. ‘An aesthetic appreciation remains… indifferent to the truth of its object. A person who has an aesthetic interest in the Odyssey is not concerned with the literal truth of the narrative’ (Scruton 2008, p.146). Representing imaginary things is what makes representation important to us and gives us an aesthetic interest. The unstated corollary here is that photographs cannot hold an aesthetic interest by virtue of them being fictionally incompetent. The transparency of photographs to the objects that they depict disqualifies photographs from aesthetic appreciation since, apparently, a photograph cannot be indifferent to the truth of its object.
Here Scruton is mistaken in at least two senses. Firstly, in what sense does he mean that aesthetic interest always remains indifferent to the truth of its object? What would be the point of reading fiction or poetry if it did not make true statements about the world? Homer did not tell a story about Odysseus’s wanderings and home-coming to Ithaca in the absence of a view about the consequences of physical absence to relationships. We may not be concerned with the literal truth of the narrative but we are interested in the significance of the story as a whole. As Barthes recognised (Barthes 1993) a distinction needs to be made between connotation and denotation, between significance and meaning. In Fregean terms the literal story is fictional in a denotational context, that is the characters and events of the story have no ‘referents’ (although this itself is debatable) but do have ‘senses’. The story as a whole can be seen on a different level by virtue of the way it is told. The aesthetic interest in part comes from the significance of this sense to an appropriately receptive and trained reader. Literature concerns itself with universals and such abstract notions are as much a part of the world as are objects. Likewise, the argument applies to photographs.
Secondly an understanding of the context of a representation, say a poem, surely enhances our aesthetic enjoyment of it. The ending of Virgil’s First Georgic (Dewar 1988) portrays a horrifying image of civil war as a chariot careering out of control. But our aesthetic appreciation is surely enhanced when we understand the identity of the charioteer. Is it Man or Madness or Mars or some other abstract entity? Asking makes a difference. Without this understanding of context, it would be difficult to make any sense of the poem’s significance. Hence, we cannot agree with Peter Lamarque for example when he says that ‘truth is not an artistic value’ (Kieran 2006, p.127). Lamarque is right to hold that ‘artistic achievement is to move beyond the subject to something more universal’ (Kieran 2006, p. 131) but the point is that moving from the particular to a universal does invoke a truth.
To summarise the position relating to the apparent fictional incompetence of photographs: Scruton places great weight on the fictive possibilities of paintings in their ability to represent. In the first sense photographs, being causal, cannot depict a state of affairs different to the one presented to an observer. I have demonstrated some doubt over this assertion. In the second sense, pictures do not follow rules; but this seems questionable; and photographs cannot invoke references to truth or falsity as possibilities, but as we see in the Tony Ray-Jones example, they can. Finally, the transparency of photographs to the objects that they depict disqualifies photographs from an aesthetic appreciation since a photograph cannot be indifferent to the truth of its object. I argue that indifference to the truth of an object is not a necessary condition for aesthetic enjoyment. Indeed, the opposite may be the case.
Objection: Organisational aspect perception is insufficiently robust a notion to explain its inclusion in ‘picturing’.
A third objection might question the validity or nature of organisational aspect perception to justify its inclusion in a concept of picturing.
The difficulty of assessing the empirical evidence for gestalt phenomena is made difficult by the historical antipathy felt by many Gestaltists for the scientific enterprise, particularly its method (Madden 1952). In addition, the wide range of problems tackled by Gestaltists, ranging from psychology, psychotherapy, epistemology, metaphysics to value theory creates a problem of defining a central kernel for consideration. Perhaps the best hope for normalising an empirical foundation for it lies with its ideas being subsumed under scientific vision research programmes as shown by recent trends. Indeed, a leading authority on the experimental side was able to conclude recently that ‘the field of research on perceptual grouping and figure–ground organisation is thriving, and progress has been tremendous’ (Wagemans et al., 2012).
This is not the place for a balanced assessment of the explanatory claims of gestalt principles. Supporting citations are easy to harvest but playing the tit for tat game of garnering citation support against counter claims will hardly settle the matter. The best appeal I can make is to common sense. Would it not be odd if humans, for whom the visual sense dominates perception, did not evolve to like looking at what we are good at seeing? If natural selection works as generally thought, then we become good at seeing objects and changes to states of affairs that are significant to our biological fitness. Hence pattern recognition, figurative assessment and other gestalt forms that render local visual processing possible are likely to be prevalent in evolutionary adaptations. We therefore should expect gestalt phenomena as responses to regularities in the environment so that irregularities can be more quickly attended to. We would actually be surprised if no such phenomena could be found.
From a philosophical position Gestalt phenomena have not attracted much concentrated attention. The most worked out exposition remains Merleau-Ponty’s treatment (Merleau-Ponty 1962) but other notable associations can be found in Whitehead (Desmet 2015), Husserl (Ierna 2009, Brentano (Macnamara and Boudewijnse 1995) and as already mentioned Wittgenstein in response to Köhler. Holism plays an important part in Spinoza’s philosophy but not in a gestaltist sense.
Further objections, of course, can be levelled at my proposed account in Part 2. For example, the notion that ‘concepts go all the way down’ is a hotly contested idea. Similarly, the role of non-conceptual content is a subject that has attracted a huge literature and is much debated. The notion of ‘bodily knowledge’ is one that finds much disfavour, particularly in analytical philosophy circles. These are vast topics and cannot realistically be defended in depth in the confines of a short dissertation. I must simply leave them to one side.
In Part 3, I have examined a number of objections to my contention that photographs do represent. I have defended my view that ‘Seeing as’ is categorically the same as ‘Seeing’. I have set out the notion that fiction is important to how we look at photographs. I have countered a technical objection to organisational aspect perception and a potential charge of the ‘myth of the Given’ by defending a two-component account of visual experience. I have examined the empirical evidence for gestalt phenomena and concluded that a balanced assessment is difficult to construct but that we should accept a role for gestalt phenomena on teleological grounds.
At the heart of Scruton’s argument is the idea that the appreciation of Art is founded on imagination and in particular on our ability to understand a thought embodied in a work of art. When we change the aspect under which we look at a picture, seeing it now as a duck, now as a rabbit, what changes? Not the picture, for that stays the same. What changes is the way we look at it; we see it differently and, if we are sufficiently trained, in the way intended by the artist. For Scruton, photographs are too causally tied to their depicted objects to enable us to change the aspect under which we look at them, too much determined by chemistry to enable the artist to inflect a thought upon an image. Hence photographs cannot be appreciated as works of art.
Scruton’s poses the following question: Can photographs direct our thoughts to pre-established (intended) objects in such a way that a subject’s feelings are expressed towards what is symbolised, or must we see photographs only in terms of the objects that they reflect?
I conclude that we need not see photographs only in terms of the objects that they reflect. The fictive device of framing that enables the seeing of ‘ephemeral connections between unrelated things’ serves to convey ideas conceived by the photographer. These are consequences of ‘Seeing’.
In this paper, I have argued that the key to Scruton’s thesis, ‘Seeing as’, is based on an untenable conception of picturing. I have shown that his argument is flawed because the sharp distinction he draws between ‘Seeing as’ and ‘Seeing’ puts him in an untenable philosophical position: that ‘Seeing’ is inert.
In opposition to Scruton I have defended the idea that it is ‘Seeing’ with its phenomenal content that guides (‘all the way up’) the mode of understanding by which we judge a visual artwork through conceptual shapings. For Scruton ‘Seeing’ is causal. I contend that it is more than this. Visual perception is not exhausted by conceptual understanding but rather invokes a surplus of sense. The mode of understanding that arises from ‘Seeing’ is partly articulated through bodily dispositions that stem from navigating a picture through its perceptual Gestalt forms. As Kant recognised, what is apprehended when we picture something is not just a perceptual shape, but a purposiveness ‘… (the) transcendental principle which represents a purposiveness in nature … in the form of a thing’ (Kant and Walker 2007, Intro viii, 31). The way we navigate through photographs, ‘picturing’, provides an important basis for the aesthetic enjoyment of visual artworks.
A photographer’s skill helps a viewer to navigate a photograph in such a way as to understand a conveyed thought and, in the process of navigating through a picture, enjoy it aesthetically.
Inevitably however, represenational reflections about the nature of ‘Seeing’ seem to fall short, as Sellars recognised: ‘I believe … sheer phenomenology or conceptual analysis takes us part of the way, but finally lets us down. How far does it take us? Only to the point of assuring us that Something, somehow a cube of pink in physical space, is present in the perception other than as merely believed in’ Sellars 1982, p.89).
‘There is more to seeing than meets the eye’, but an understanding of this surplus of sense is elusive. Our perception of an object relies on features of that object, its qualities and relations and perspectival structures. What we perceive the object as is a matter of conceptual content. But asking where one ends and the other begins is already to put a foot wrong if one holds that they are co-dependent. What seems clear is that the boundary between them is indistinct and, as a consequence, the concept of ‘Seeing as’ cannot be used to discriminate picture-taking from picture-making.
Arnheim, R. (2006). ‘Art as perceptual experience’, Oxford: Oxford University Press.
Aviman, G. (2014). ‘Zen Paintings in Edo Japan (1600-1868), Playfulness and Freedom in the Artwork of Hakuin Ekaku and Sengai Gibon’, New York: Routledge.
Barthes, Roland. (1993). ‘Camera lucida: reflections on photography’, London: Vintage.
Brandom, R. B. (1994). ‘Making it explicit: reasoning, representing, and discursive commitment,’ Cambridge, (Mass.): Harvard University Press.
Budd, Malcolm, (1987). ‘Wittgenstein on Seeing Aspects’, Mind: 1.
Butt, R. (1969). ‘Kant’s Schemata and Semantical Rules’, Open Court Press.
Castaneda Calderon, Héctor Neri, and Wilfrid Sellars. (1975). ‘Action, knowledge and reality: critical studies in honor of Wilfrid Sellars’, Indianapolis: Bobbs-Merrill.
Chang, D., et al. (2007). ‘The gestalt principles of similarity and proximity apply to both the haptic and visual grouping of elements’. Conferences in Research and Practice in Information Technology Series, 2007, 64:79-86.
Chicago, The Art Institute of. (2013). ‘The Essential Guide’, accessed 23 August 2017
Coates, Paul. (2007). ‘The metaphysics of perception: Wilfrid Sellars, perceptual consciousness and critical realism’; New York: Routledge.
Davies, David. (2008). ‘How Photographs ‘Signify’: Cartier-Bresson’s ‘Reply’ to Scruton.’ in, Photography and Philosophy: Essays on the Pencil of Nature’; Malden (MA): Blackwell Publishing.
De Vries, Willem, A. ‘Wilfred Sellars’, (2005) published by Acumen; (2014) Abingdon: Routledge.
Deleuze, G. and D. W. Smith (2004). ‘Francis Bacon: the logic of sensation’, London: Continuum.
Desmet, R. (2015). ‘The Gestalt Whitehead.’ Process Studies 44(2):190-223.
Dewar, M. (1988). ‘Octavian and Orestes in the Finale of the First Georgic’; The Classical Quarterly, 35 (2) pp 563-565, Cambridge University Press.
Dreyfus, Hubert L. 2002. ‘Intelligence without Representation–Merleau-Ponty’s Critique of Mental Representation: The Relevance of Phenomenology to Scientific Explanation’, Phenomenology and the Cognitive Sciences, 1: 367-83.
Fisette, Denis, and Guillaume Fréchette. ‘Themes from Brentano’, Editions Rodopi, electronic book.
Forti, B. (2015). ‘What Are the Limits of Gestalt Theory?’ Gestalt Theory 37(2): 161 – 188.
Goodman, Nelson. (1976). ‘Languages of art : an approach to a theory of symbols’ Indianapolis: Hackett.
Gregor, Mary, J. (1983). ‘Baumgarten’s “Aesthetica”‘, The Review of Metaphysics: 357.
Gunther, Y. H. (2003). ‘Essays on non-conceptual content’, Cambridge (Mass.) MIT Press.
Heidegger, M. (1962). ‘Being and time’, Malden, MA; Blackwell.
Hopkins, Robert. (1998). ‘Picture, Image and Experience: A Philoposphical Inquiry’ Cambridge: Cambridge University Press.
Ierna, C. (2009). ‘Husserl et Stumpf sur la Gestalt et la fusion.’ Philosophiques 36(2): 489-510.
Johnson, G. A. (1994). ‘The Merleau-Ponty Aesthetics Reader’, NorthWestern University Press.
Kant, I. and N. Walker (2007). ‘Critique of judgement’, [electronic book], Oxford; Oxford University Press.
Kant, Immanuel, Allen W. Wood, and Paul Guyer (2000). ‘Critique of pure reason’ Cambridge: Cambridge University Press.
Kant, Immanuel, Victor Lyle Dowdell, and Hans H. Rudnick. (1996). ‘Anthropology from a pragmatic point of view’; Carbondale: Southern Illinois University Press.
Kennedy, J. M. (1974). ‘A psychology of picture perception’, San Francisco: Jossey-Bass.
Kenny, A., et al. (2010). ‘Mind, method, and morality’. [electronic book]: essays in honour of Anthony Kenny, Oxford: Oxford University Press.
Kieran, M. (2006). ‘Contemporary debates in aesthetics and the philosophy of art’, Oxford: Blackwell.
Lehar, S. ‘Gestalt References.’ From http://cnsalumni.bu.edu/~slehar/references/gestalt-refs.html – accessed 16 July 2017.
Lopes, Dominic McIver. (1996). ‘Understanding pictures’, Oxford: Clarendon Press.
Lopes, Dominic McIver. 2003. ‘The Aesthetics of Photographic Transparency’, Mind: A Quarterly Review of Philosophy, 112: 433-48.
Maclean, N. (2001). ‘A River Runs through It and Other Stories’, University of Chicago Press.
Macnamara, J. and G.-J. Boudewijnse (1995). ‘Brentano’s Influence on Ehrenfels’s Theory of Perpetual Gestalts.’ Journal for the Theory of Social Behaviour 25(4): 401-417.
Madden, E. H. (1952). ‘The Philosophy of Science in Gestalt Theory.’ Philosophy of Science 19(3): 228-238.
Makkreel, R. (1990). ‘Imagination and Interpretation in Kant: The Hermeneutical Import of the Critique of Judgment’, University of Chicago.
McDowell, J. H. (1996). ‘Mind and world: with a new introduction’, London: Harvard University Press.
Merleau-Ponty, M. (1962). ‘Phenomenology of perception’, London: Routledge & Kegan Paul.
Merleau-Ponty, Maurice, and Claude Lefort. (1968). ‘The visible and the invisible’, followed by Working notes (Evanston (Ill.): Northwestern University Press.
Peacocke, Christopher. (1983). ‘Sense and content: experience, thought, and their relations’, Oxford: Clarendon Press.
Pinna, B. (2010). ‘New Gestalt Principles of Perceptual Organization: An Extension from Grouping to Shape and Meaning.’ Gestalt Theory 32(1): 11 – 78.
Raftopoulos, Athanassios. (2009). ‘Cognition and perception’. [electronic book] : how do psychology and neural science inform philosophy? Cambridge, (Mass.) : MIT Press.
Sabar, S. (2013). ‘What’s a Gestalt?’ Gestalt Review 17(1): 6-34.
Savedoff, Barbara E. (1992). ‘Transforming Images: Photographs of Representations’, The Journal of Aesthetics and Art Criticism: 93.
Sachs, Carl. B. (2014) ‘Intentionaliy and the Myth of the Given’, London; Pickering and Chatto.
Scruton, R. (1998). ‘Art and imagination: a study in the philosophy of mind’, South Bend, Ind.: St. Augustine’s Press.
Scruton, R. (2008). ‘Photography and Representation’. Photography and Philosophy: Essays on the Pencil of Nature. Malden (MA): Blackwell Publishing.
Sellars, Wilfrid; Richard Rorty; and Robert B. Brandom. (1997). ‘Empiricism and the philosophy of mind’ Cambridge, (Mass.): Harvard University Press.
Sellars, Wilfed. 1982. ‘Sensa or Sensings: Reflections on the Ontology of Perception’, Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition: 83.
Smith, N. H. (2002). ‘Reading McDowell: on Mind and world’, London: Routledge.
Smith, S. M. and S. Sliwinski (2017). ‘Photography and the Optical Unconscious’. Durham: Duke University Press Books.
Steven M. Silverstein, a. and a. Peter J. Uhlhaas (2004). ‘Gestalt Psychology: The Forgotten Paradigm in Abnormal Psychology.’ The American Journal of Psychology (2): 259.
Swanson, L. R. (2016). ‘The predictive processing paradigm has roots in Kant.’ Frontiers in Systems Neuroscience 10.
Todes, S. (2001). ‘Body and world’, Cambridge, (Mass.): MIT Press.
Van Hiel, A. and I. Mervielde (2003). ‘The Need for Closure and the Spontaneous Use of Complex and Simple Cognitive Structures.’ Journal of Social Psychology 143(5): 559-568.
Wagemans, J. ‘Gestalt Vision.’ from http://www.gestaltrevision.be/en/ – accessed 20 August 2017.
Wagemans, Johan, Jacob Feldman, Sergei Gepshtein, Ruth Kimchi, James R. Pomerantz,
Peter A. van der Helm, and Cees van Leeuwen. 2012. ‘A century of Gestalt psychology in visual perception: II. Conceptual and theoretical foundations’, Psychological Bulletin, 138: 1218-52.
Walden, Scott. (2008). ‘Photography and philosophy: essays on the pencil of nature’ Malden, (MA.): Wiley-Blackwell.
Ward, D. ‘Gestalt Structure and Phenomenology.’ From http://www.blogs.hss.ed.ac.uk/gestalt-structure-phenomenology/2017/04/20/cfp-july-conference, – accessed 23 July 2017.
Wittgenstein, L. (1945). ‘Philosophical Investigations’, Oxford: Wiley-Blackwell.
Worth, Sol. (1981). ‘Studying Visual Communication’; University of Pennsylvania.
 The crux of Scruton’s argument runs like this:
- A picture can only be a work of art by virtue of its representing: it is brought to light by a competent beholder seeing it as a symbol of something thought
- There is a set of conditions, Z, that distinguishes ‘Seeing as’ from mere ‘Seeing’. Z is a set of necessary and sufficient conditions for a picture to be a work of art. It is also a set of necessary conditions for a work of art to be experienced aesthetically
- Photographic depiction necessarily entails resemblance to an object in virtue of the causal counterfactual dependence of a photograph upon its source
- Therefore, a picture such as an ideal photograph, as opposed to a painting, cannot be ‘seen as’ a symbol of something thought.
- Therefore, an ideal photograph cannot be a work of art and cannot be experienced aesthetically as a work of art.
 One might think that the meaning of the term ‘intentionality’ would be made clear. Scruton begins ‘… ideal paintings stand in a certain “intentional” relation to a subject’ to which he appends a footnote citing Brentano’s ‘Psychology from an Empirical Standpoint’. From the reference to Brentano we might be forgiven for adducing the meaning of the term to be that ostensibly set out by Brentano, i.e. the mind as having the property of ‘directedness’ or ‘about-ness’ although this is very much an oversimplification of Brentano’s concept.
 For a summary of some neural research into attention in visual processing see (Raftopoulos 2009)
 The initial parsing of a scene occurs very quickly, within 15 milliseconds, before the stage of object recognition and categorisation. This pre-attentional stage occurs locally without access to higher levels of cognition and is presumably more primitive. Parsing involves short-lived representations of proto-objects and discrete spatio-temporal entities. This sensitivity to pattern suggests some local ability to differentiate non-cognitively and therefore suggests access by the visual system to information at local level, that is removed from higher cognition. Up to 70 milliseconds later with increasing attention comes object individuation and object recognition. Individuation precedes recognition, which requires access to higher cognitive faculties like memory.
 In the Critique of Judgment, Kant talks about aesthetic judgment as going ‘beyond the concept and even beyond the intuition of the object and add to that intuition as predicate, something that is not a cognition, viz., a feeling of pleasure or pain’ (Kant and Walker 2007 §36). Kant is not referring here to the possibility of a synthesis without concepts but rather with the reflective nature of apprehension. Categories remain relevant to aesthetic judgements but not in the manner of a synthesis resulting in cognitive judgments – See for example (Makkreel 1990).
 I follow here the position held by Sellars. My understanding of Sellar’s position is very much influenced by de Vries (de Vries 2005).
 The literature on gestalt phenomena is extensive – for a good introduction see (Sabar 2013). We see references to the notion of the phenomena of gestalt forms in Kant – for example: ‘aesthetic apprehension of beautiful form does not belong with … discrete … representations which must be synthesised, but with an indeterminate sense of a whole’ (Kant and Walker 2007 – p. 4) and more directly ‘every form of the objects of sense … is either figure (gestalt) or play (spiel)’ (Kant and Walker 2007 – §14, 61, V, 225). We also see similar views in Baumgarten: ‘…aesthetics is more than supplying sensory content for the logic of concepts; it develops its own logic for the complete determination of the singular – see (Gregor 1983). Another interesting parallel is Kant’s rule for generating spatial forms (Monograms) – (Kant, Wood and Guyer 2000; A142/B181). However, we run the danger of misinterpretation if we read too much into these statements.
 See (Kennedy 1974) for an interesting insight into the history of Gestalt theory.
 Wittgenstein seems to have found it difficult to distinguish aspect perception from thought – ‘and that’s why the lighting up of an aspect seems half visual experience, half thought’ (Wittgenstein 1945 Part 2 §140). The assertion through a thought is required otherwise we would not be able to navigate from the X in ‘seeing X as Y’ to the Y. Navigating from X to Y suggests a requirement for thought.
 Wittgenstein’s discussion about aspect perception seems quite muddled. As Schroeder has convincingly argued (Kenny et al. 2010 pp. 352 – 371), his discussion of aspect perception used aspect perception in two different contexts, dispositional and episodic, and was primarily interested in its episodic sense, ‘what lights up here lasts only as long as I am occupied with the observed object in a particular way’ (Wittgenstein 1945 Part 2 § 237). Scruton’s self-acknowledged reliance on Wittgenstein’s analysis also draws on an episodic sense of aspect perception, taking aspect perception to be episodic and therefore special rather than (apparently) ubiquitous like ‘Seeing’).
 See (Peacocke 1983) for a full account of this approach.
 The work of J.J. Gibson (Gibson,1979) is instructive with regards to how we navigate the world. My reading of his notion of ‘affordance’ is that we navigate the world by comparing expected perceptual results with actual perceptual results and we ‘manage’ the difference between the two, that is we postulate ‘X as Y’ but we find that X is Z and our perceptual state navigates Y-Z. Affordance is the state of Y=Z or Y being close to Z. For an interesting article on the predictive processing framework and its relation to Kant’s insights see (Swanson 2016). Gibson also talks about the importance of the horizon in pictorial informational structures.
 We may agree with Merleau-Ponty that ‘the “structure world” with its double moments of sedimentation and spontaneity (lie) at the centre of consciousness’ (Merleau-Ponty, 1962, p.132) without having to accept his overstatement: ‘the Gestalt contains the key to the problem of the mind’ (Merleau-Ponty and Lefort 1968, p.192), nor his rejection of representationalism.
 I use street photography to illustrate the Gestalten because of all photographic genres it could be argued that street photography is the most ‘causal’ in Scruton’s terms. The photographer is unable to arrange the objects in a photograph in a predefined intentional way; instead she must convey an idea through noticing a pre-existing but short-lived arrangement in the moment.
 This reminds us of Kant’s notion of schemata in which objective meaning is made possible by schemata by restricting understanding as in selecting the type of empirical concepts eligible to be applied to objects, and realising understanding through anticipating the possible objects of experience, see (Butt 1969).
 See for example (Van Hiel and Mervielde 2003).
 Perhaps the closest notions to it in recent literature are J.J Gibson’s ‘affordance’ (Gibson 1979), Brandom’s reliable differential responsive dispositions (RDRDs) (Brandom 1994), Merleau-Ponty’s notion of ‘maximal grip’ (Merleau-Ponty 1962) and Todes’ ‘absorbed coping’ (Todes 2001).
 I do not want to imply that gestalt phenomena are sufficient for picturing – undoubtedly many other factors will be at work – but the retrieval of basic perceptual components such as figure/ground, shape similarity, motion, closure and so on give rise to weak representations which precede richer representations of the same objects when we attend to them and indeed, and here is the point, governs the kind of attention that we utilise in arriving at richer representations.
 See Joel Meyerowitz: http://ifearbrooklyn.com/2013/10/29/joel-meyerowitz-on-ephemeral-connections-2012/
 But see (Pinna 2010) for the working up of an idea of a rich pictorial language
 Goodman’s theory of depiction seems to stumble in respect of his denotation theory (Goodman 1976) as it does not seem to explain how dog-pictures are not seen as cat-pictures. The idea that pictures are ‘symbolically dense’ runs counter to the position I am taking here. However, he is right to set out that photographs are representational. Scruton’s argument against Goodman is unconvincing in this respect. Lack of space prevents me from pursuing this line of thought here.
 (Virgil. G. 1.511-14): ‘Saevit toto Mars impius orbe,
ut cum carceribus sese effudere quadrigae,
addunt in spatia, et frustra retinacula tendens
fertur equis auriga neque audit currus habenas’.
 See (Wagemans et al. 2012) for extensive scientific research citations and (Ward) for a collection of papers. Also see (Lehar) for an extensive bibliography.
 Note for example (Chang, Nesbitt et al. 2007, (Pinna 2010), (Forti 2015) and (Steven M. Silverstein and Peter J. Uhlhaas 2004)