Analysis Advance Access published March 17, 2016

CRITICAL NOTICE Perception and Its Objects

Bill Brewer’s Perception and Its Objects is an admirable, thorough and persuasive defence of a version of naı¨ve realism, which Brewer calls ‘the object view’.1 The object view takes perception to be fundamentally a matter of being perceptual related to external mind-independent physical objects. The perceptual relation is characterized as a relation of ‘conscious acquaintance with various mind-independent physical objects from a given spatio-temporal point of view, in a particular sense modality, and in certain specific circumstances of perception (such as lighting conditions in the case of vision).’ (2011: 96). Although I personally lean towards a representational account of perceptual experience, I think Brewer’s view is one of the most convincing alternatives to a representational account. Here, I will not provide a defence of the representational account against the object view. This is something I hope to do in forthcoming work (Seeing and Saying). Instead I will raise five potential challenges for the object view.

1. Is the object view in opposition to all versions of the content view? Brewer appears to consider advocates of what he calls the ‘content view’ among his main opponents. It wasn’t completely clear to me why he took this to be the case. To be sure, much recent literature in the philosophy of perception has been concerned with the question of whether perceptual experience has content. I have recently edited a volume devoted primarily to that question. The significant amount of recent literature on this topic certainly justifies dealing with this question, not least in an extensive defence of a view that appears to be opposed to the content view. The question, though, is whether the content view is indeed opposed to the object view. Philosophers such as Susanna Schellenberg (2014) and Heather Logue (2014) have defended the view that perceptual experience is fundamentally a matter of being perceptually related to an external object and fundamentally a matter of having a representational content. This view would seem compatible with the object view. But there may even be versions of the content view that are simple technical variants of the object view. Consider the following version of the content view. (I hasten to say that I don’t know of anyone who has explicitly defended it but it is nonetheless a view in logical space.) Perceptual experience is fundamentally a matter of being related to a representational content. The representational content is a Russellian proposition that 1

Perception and Its Objects. By Bill Brewer. Oxford University Press, 2013, 216 pp. £32.00 cloth, £16.99.

Analysis Reviews Vol 0 | Number 0 | 2016 | pp. 1–7 doi:10.1093/analys/anw023 ß The Author 2016. Published by Oxford University Press on behalf of The Analysis Trust. All rights reserved. For Permissions, please email: [email protected]

Downloaded from http://analysis.oxfordjournals.org/ at University of Miami - Otto G. Richter Library on March 18, 2016

BERIT BROGAARD

2 | critical notice

consists of external objects and perceptible property instances. It seems to me that the only difference between Brewer’s view and this view is that Brewer takes perception to be a relation to external reality, whereas this version of the content view claims to take perception to be a relation to a proposition. But the proposition, on the view in question, is Russellian and is therefore firmly grounded in external reality. So, it seems to me that the two views are simply technical variants on each other.

Personally, I think that the main problem with many of the standard versions of the content view is that they take perceptual experience to be partially constituted by the proposition that it is related to. To be sure, defenders of the content view are careful to distinguish between the perceptual state and its content. What exactly the vehicle (i.e. the state) is supposed to be over and above the content is not always made clear. The problem with most of these views, in my opinion, is that they don’t specify how the proposition that is supposed to do the representational work comes to have its intentional properties. There is a long debate in philosophy of language about how sets of objects and properties could possibility come to be intentional entities (e.g. propositions). The most recent responses to this problem take a cognitive approach (see e.g. Brogaard 2014). The intentionality of propositions derives from (or in some cases the proposition is said to be identical to) a type of cognitive state, viz. a type of predication that is implicated in all mental states with a mind-world direction of fit, ranging from perceptual states to cognitive affirmations. I think this reply is one of the most convincing ways of dealing with the problem of intentionality. The problem for the content view is that if the intentionality of propositions derives from mental states, then explaining the nature of mental states in terms of intentional entities called ‘propositions’ appears to involve a vicious explanatory circle. One possible response to this problem is to treat at least some mental states as inherently representational. It is a view of this type that I favour. On this sort of view, perceptual states are representational but there is no need to take them to be relations to a content, except in a derivative sense. This raises the question of whether the object view is compatible with this sort of view. There is a sense in which a representational view of this kind is relational. One crucial difference between the view that Brewer defends and a representational view of this kind is that the perceptual relation that is constitutive of perceptual experience in one case is that of presentation and in the other case is that of representation. If Brewer takes his view to be incompatible with this type of representational view, then I would like to know why the perceptual relation is not, or could not be, that of representation, and more generally I would like to know why perceptual experience is not, or could not be, representational.

3. Does the object view collapse into a representational account? On Brewer’s account, a visual illusion in which a mind-independent physical object o, looks F, although o is not actually F, o is the direct object of visual experience from a

Downloaded from http://analysis.oxfordjournals.org/ at University of Miami - Otto G. Richter Library on March 18, 2016

2. Is the object view in opposition to all representational views of perception?

critical notice | 3

Downloaded from http://analysis.oxfordjournals.org/ at University of Miami - Otto G. Richter Library on March 18, 2016

spatiotemporal point of view and in viewing conditions relative to which o has visually relevant similarities with paradigm exemplars of F. Brewer notes that the naı¨ve realist cannot take visually relevant similarities to be identical to the ways that the relevantly similar relata are visually represented as being, ‘or else [naı¨ve realism] clearly collapses into a version of [the representational view]’ (Brewer 2011: 103). I suppose the argument here is that if the naı¨ve realist’s account of illusions depends on the notion of representation, then the representational feature of experience is fundamental. However, it is not clear that Brewer actually is able to avoid this consequence. He says that the visually relevant similarities are similarities in visual processing. As he puts it, ‘visually relevant similarities are identities in such things as the way in which light is reflected and transmitted from the objects in question, and the way in which stimuli are handled by the visual system, given its evolutionary history and our shared training during development’ (2011: 103). Paradigm exemplars are the kinds that are crucially relevant to our possession of concepts of those kinds. As indicated, the notion of visually relevant similarities must be implicitly restricted to a particular type of visual system. This is required, otherwise there may not be any or sufficiently many identities in the visual processes. Likewise, the notion of a paradigm exemplar must be restricted to a specific type of cognitive system possessing particular concepts. If we don’t hold the cognitive system fixed, then illusory experiences may turn out to count as veridical. For example, if a cat is a paradigm exemplar of the concept of cat for you, but a dog is a paradigm exemplar of the concept of cat for me, then your illusory experience in which a dog looks like a cat would have visually relevant similarities with paradigm exemplars of dogs relative to my cognitive system. Finally, the laws of nature must be restricted to those that actually obtain. Otherwise, there could be illusions in which there are relevant similarities in how light is reflected and how visual stimuli are handled by the visual system for F cases, but in which light is transformed into a different kind of light insight the visual system. So, an illusory experience in which a white table looks green might be one that bears visually relevant similarities to paradigm cases of red objects. Brewer’s definition of visually relevant similarities, as formulated, makes no mention of the notion of representation. The problem, though, is that Brewer’s notion of visually relevant similarity appears to collapse into the notion of representation. The notion of visually relevant similarities to paradigm exemplars of F is restricted to a particular type of visual and cognitive system as well as our actual physical laws. So, if visual experience is representational, though not fundamentally representational, then a visual experience bearing visually relevant similarities with paradigm exemplars of F and a visual experience representing F are necessarily co-extensional. Or at least they are necessarily co-extensional, given that Brewer denies that hallucinations are visual experiences. But if they are necessarily co-extensional, then he has failed to give an account of illusions that is logically independent of the notion of representation. So, his account makes the notion of representation essential to our understanding of illusions. Now, this may not be a problem for all types of naı¨ve realism. The naı¨ve realist could, for example, simply bite the bullet and say that veridical experience and illusions are fundamentally distinct types of mental states but it does seem to present a problem for Brewer’s account.

4 | critical notice

4. Is the object view really a form of direct realism?

5. Does non-transparency provide a challenge for the object view? Brewer does not use transparency in any direct argument for the object view but he does point out that transparency does not favour the content view over the object view. I think that is correct. But I wonder whether considerations of transparency may count against both kinds of direct realism. Transparency captures the idea that when we try to introspect, it seems that we look right through the experience only to find external objects and their properties. Moore put it succinctly as follows: The moment we try to fix our attention upon consciousness and to see what, distinctly, it is, it seems to vanish: it seems as if we had before us a mere emptiness. When we try to introspect the sensation of blue, all we can see is the blue: the other element is as if it were diaphanous. (Moore 1903: 41 in 1993 reprint) Moore’s point is that in visual experience it is as if the external scene is simply presented to us. If we try to access features that are internal to experience, it seems as though we access the external object and its attributes. I think transparency, rather than supporting naı¨ve realism, may turn out to be a threat to the view. Brewer holds that the external stimulus exhausts the phenomenology of veridical experience (Brewer 2007). So, if there are cases of veridical experience in which features presented to the subject in the experience are not identical to property instances borne by the external stimulus, then naı¨ve realism would seem to be false. Are there any such cases? I think there are. In Type-1 blindsight, the best-known kind of blindsight, subjects have an ability to discriminate visual stimuli but report having no associated sensory consciousness. In Type-2 blindsight, a lesser-known form of blindsight, subjects typically report not having any direct awareness of external objects (Brogaard 2015). Subjects have reported being aware of ‘something’ or being aware that ‘something is happening’ when

Downloaded from http://analysis.oxfordjournals.org/ at University of Miami - Otto G. Richter Library on March 18, 2016

Brewer brilliantly cashes out the perceptual relation that obtains between perceiver and external reality as a relation to external objects from a certain perceptual perspective in particular circumstances, etc. This is definitely the way to go for the naı¨ve realist. However, one wonders whether this move turns the object view into a form of indirect realism. When I look at trees of the same size taking up different proportions of my visual field, I perceive them as trees of the same size at different distances from me. But my conscious acquaintance with the trees does not seem to be direct. From my perceptual perspective, there is a sense in which the trees appear to take up different proportions of the visual field and hence there is a sense in which they appear to be of different sizes. Is it by being acquainted with these differently sized entities in my visual field that I come to experience them as having the same size? To use some jargon from Pylyshyn (1999), the retinal image of the trees is not what I ultimately see. What I ultimately see is the result of computations based on perceptual principles inherent to the perceptual system. Brewer’s view (‘conscious acquaintance from a given spatio-temporal point of view, etc.’) seems perfectly compatible with this theoretical framework. But then why is Brewer’s view a version of direct realism rather than indirect realism? What makes it direct? Or is the debate about directness versus indirectness simply a verbal dispute?

critical notice | 5

Downloaded from http://analysis.oxfordjournals.org/ at University of Miami - Otto G. Richter Library on March 18, 2016

presented with a fast-moving, high-contrast stimulus in their blind field but they are unable to consciously identify any other characteristics of the stimulus. Other subjects have reported seeing ‘shadows’ or ‘pinpoints of light’. But subjects typically deny seeing the object that is causing the experience in their blind field. Subjects furthermore report seeing attributes they are aware of in their blind field through a veil of perception. Several of Riddoch’s subjects described their experiences of moving stimuli in their blind fields as ‘shadowy’ (Riddoch 1917). Gordon Holmes’s (1918) subject 11 reported awareness of moving white stimuli but described them as seen ‘through a mist’ and as having a ‘dirty grey colour’. Blindsight subject GY has described his experiences of moving objects as that of a normal person with his eyes shut, who looks out of the window and moves his hand in front of his eyes (Barbur et al. 1980) and as that of ‘a black shadow moving on a black background’ (Zeki and Ffytche 1998). More recently three blindsight subjects GN, FB and CG were asked to report on their residual awareness and to draw their experiences in their blind field (Ffytche and Zeki 2011). They were able to draw the features they were aware of but described the stimulus as ‘foggy’ or as a flash seen behind a screen filtering out anything other than the change in light. These characterizations seem to explicitly contradict the idea that the phenomenology of visual experience is such that we see through any internal features and simply see the external objects and their visually perceptible property instances. The reason Type-2 blindsight differs from ordinary visual experience in all of these respects is likely that it is generated by an alternative visual pathway that bypasses V1. Morland et al. (1999) investigated GY’s ability to make luminance matches in his hemianopic field and between both hemifields. They found that GY was able to make matches when the stimuli were presented in the blind field but was unable to establish matches based on luminance when the stimuli were presented in opposing fields. The most likely explanation for this observation is that the perceived luminance of the stimuli in his blind field is derived from direct projections from subcortical areas to extrastriate areas bypassing V1, whereas the perceived luminance of the stimuli in his intact field originates in the normal visual pathway that includes V1. This would make it possible for him to compare stimuli on the basis of luminance when both are presented in the hemianopic field but when the stimuli are presented to opposing fields, the distinct pathways would yield different kinds of percepts, making lawful matching difficult. This suggests that V1 plays a crucial role in generating normal conscious perception of luminance and that normal, conscious luminance perception is required for generating direct conscious awareness of external objects and their property instances. In spite of impoverished luminance awareness, the awareness of stimuli in Type-2 blindsight contributes significantly to the subject’s ability to determine the attributes of the stimulus, making these cases analogous to other cases of veridical vision in that there is a significant correlation between awareness and discrimination. So, the unusual phenomenology and alternative visual pathway underlying the condition do not provide evidence against the veridicality of Type-2 blindsight. It might be held that some of the descriptions of the residual awareness in Type-2 blindsight suggest that Type-2 blindsight is illusory, representing some features of the external stimulus correctly and representing other features incorrectly. For example, the description of a moving stimulus as a black shadow on a black background or the description of seeing the stimulus the way a sighted person with his eyes closed would see a

6 | critical notice

University of Miami, Coral Gables, FL 33124, USA [email protected] References Barbur, J.L., K.H. Ruddock and V.A. Waterfield. 1980. Human visual responses in the absence of the geniculo-calcarine projection. Brain 103:905–28. Brewer, B. 2007. Perception and its objects. Philosophical Studies 132: 87–97. Brewer, B. 2011. Perception and Its Objects. Oxford: Oxford University Press. Brogaard, B. 2014. An empirically-informed cognitive theory of propositions. Canadian Journal of Philosophy 43: 534–57. Brogaard, B. 2015. Type 2 blindsight and the nature of visual experience. Consciousness and Cognition 32:92–103. Brogaard, B. Forthcoming. Seeing and Saying, Manuscript. New York: Oxford University Press. Ffytche, D.H. and S. Zeki. 2011. The primary visual cortex, and feedback to it, are not necessary for conscious vision. Brain 134: 247–57. Holmes, G. 1918. Disturbances of vision caused by cerebral lesions. British Journal of Ophthalmology 2: 353–84. Logue, H. 2014. Experiential content and naı¨ve realism: a reconciliation. In Does Perception Have Content?, ed. B. Brogaard, 220–41. New York: Oxford University Press.

Downloaded from http://analysis.oxfordjournals.org/ at University of Miami - Otto G. Richter Library on March 18, 2016

moving hand might suggest that the subjects misperceive features of the external stimulus. The thought would be that subjects sometimes attribute the wrong features (e.g. black) to the stimulus. However, I think there is considerable reason to resist this conclusion. Subjects with Type-2 blindsight emphasize that they use these metaphors to describe their experiences to ordinary, sighted people. When GY describes a moving stimulus as a black object on a black background, he does not seek to report on a conscious experience of a black object on a black background. In fact, he explicitly denies that he can consciously perceive the colour of a stimulus in his blind field. Although Type-2 blindsight undoubtedly has a very different phenomenology from ordinary experience, there is good reason to think that subjects with Type2 blindsight are aware of features that correspond to attributes of the stimulus. If, indeed, Type-2 blindsight is a form of veridical visual experience that does not have an object-involving phenomenology and is not transparent, doesn’t this give us some reason to resist a view of veridical visual experience that takes the phenomenology of veridical perceptual experience to be exhausted by the external objects and its perceptible property instances? If there are cases of veridical visual experience lacking this type of phenomenology, then how can we uphold the view that visual experience is fundamentally a matter of being perceptually related to an external object? Brewer, I suspect, has solid responses to all of these concerns, and in any event, Brewer’s book is noticeable for its detailed arguments and meticulous analysis. It is no overstatement to say that it is among the best monographs on philosophy of perception published in recent years. It is certainly well worth a thorough read.

critical notice | 7

Downloaded from http://analysis.oxfordjournals.org/ at University of Miami - Otto G. Richter Library on March 18, 2016

Moore, G.E. 1903. The refutation of idealism. In Philosophical Studies, 1–30. Totowa, NJ: Littlefield, Adams & Co. (1965). Morland, A.B., S.R. Jones, A.L. Finlay, E. Deyzac, S. Le and S. Kemp. (1999). Visual perception of motion, luminance and colour in a human hemianope. Brain 122(Pt 6): 1183–98. Pylyshyn, Z.W. 1999. Is vision continuous with cognition? The case for cognitive impenetrability of visual perception. Behavioral and Brain Sciences 22: 341–423. Riddoch, G. 1917. Dissociations of visual perception due to occipital injuries, with especial reference to appreciation of movement. Brain 40: 15–57. Schellenberg, S. 2014. The relational and representational character of perceptual experience. In Does Perception Have Content?, ed. B. Brogaard, 199–219. New York: Oxford University Press. Zeki, S. and D.H. Ffytche. 1998. The Riddoch syndrome: insights into the neurobiology of conscious vision. Brain 121: 25–45.