TY - JOUR
AU - French,, Craig
AB - In The Unity of Perception: Content, Consciousness, Evidence Susanna Schellenberg presents an account of perception wherein perceptual capacities play a central role. On her view, perception is constituted by the employment of perceptual capacities. And such capacities are the basic element in a ‘unified account’ of perception (p. 2). On such an account, different aspects of perception, most notably perceptual content, perceptual consciousness, and the epistemic force of perception, can all be understood by appeal to the perceptual capacities constitutive of perception (p. 2). Schellenberg dubs the view she develops in this mould ‘capacitism’ (p. 2). Perceptual capacities are at the heart of Schellenberg’s view, and she develops her account of them in Chapter 2. According to Schellenberg, perceptual capacities are low-level capacities which function to discriminate and single out particulars in the mind-independent environment (notably, mind-independent objects, events, and property instances). A favourite example of Schellenberg’s is the capacity to single out instances of red from instances of blue (p. 31). On Schellenberg’s analysis, a given perceptual capacity is individuated by the type of a particular it functions to single out (so the capacity to single out instances of red, say, differs from the capacity to single out instances of circularity) (pp. 38–40). And the analysis is an ‘asymmetric counterfactual analysis’ (p. 32). It is counterfactual in that possession of a given perceptual capacity is analysed counterfactually as follows: a subject possesses a given perceptual capacity just in case, ‘if [they] were perceptually related to a particular that the capacity functions to single out, then [they] would be in a position to discriminate and single out that particular’ (p. 40). (Schellenberg specifies this much more precisely, with important qualifications on pp. 40–43.) It is asymmetric in that the employment of a given perceptual capacity in cases where it fulfils its function is metaphysically and explanatorily more basic than the employment of the very same capacity in cases where it fails to fulfil its function (for example, an illusion as of an instance of red) (pp. 46–47). There is a lot more to Schellenberg’s analysis than I’ve captured here, but those are some of the key points. And they highlight how mind–independent particulars are crucial in Schellenberg's conception of perceptual capacities. Let’s now look at some of the key claims that Schellenberg argues for over the course of the book, and how she puts perceptual capacities to work. In Chapter 1, Schellenberg defends the claim that perceptions are partly constituted by the particulars we perceive. In Chapter 3, Schellenberg explains this in terms of the idea that perceived particulars are constitutive of the contents of perceptions. She further develops the notion of perceptual content in Fregean terms: we are to understand perceptual content not as constituted merely by the particular objects, events and properties we perceive (as on some ‘Russellian’ views of content), but by Fregean modes of presentation too, where ‘a mode of presentation is the specific way in which a subject singles out a perceived particular’ (p. 85). She dubs her view ‘Fregean particularlism’ (Chapter 4). Note that these are claims about perceptions and perceptual content but not phenomenal character, which Schellenberg thinks is not constituted by particulars. (There is more on her positive view of character below.) Now, some philosophers may be inclined to reject all this because they are sceptical of the idea that perception has content. But Schellenberg attempts to fend off such scepticism in Chapter 5, ‘In Defence of Perceptual Content’. Not only does she address various objections to the idea that perception has content, but she develops a positive argument for this idea (pp. 114-116). In the course of her discussion of perceptual content, Schellenberg embraces a non-disjunctivist, or common-kind account of perception and its character. On such an account a veridical perception, an illusion, and a hallucination can all have the same phenomenal character (pp. 77–79), accounted for by a ‘metaphysically substantial common element’ (p. 91). She develops this in the Fregean particularist framework, so as to offer a content-based common kind account of veridical perception, illusion, and hallucination. To illustrate this, I’ll focus just on veridical perception and hallucination, and leave aside illusion. Now, given what I’ve just said about Schellenberg’s view of the content of perception (as constituted by particulars), it might seem as though she can only maintain a content-based common-kind account if she holds that, like perceptions, hallucinations have contents which are constituted by particulars (as well as Fregean modes of presentation). But Schellenberg does not hold that hallucinations have contents which are constituted by particulars. How, then, are we to make sense of her view? Suppose that I see a badger scrubbing around beneath a tree. On Schellenberg’s view, the content of my perception comprises a de re mode of presentation of the badger. The mode of presentation is de re in that it is constituted by the badger I see. Consider now the subjectively matching hallucination I have as of a badger scrubbing around beneath a tree. The content of this hallucination, according to Schellenberg, involves a mode of presentation brought about by employing the same perceptual capacities. But it doesn’t involve the badger itself; instead, it involves a gap. It is gappy. So, on Schellenberg’s view, my perception has content of the form <MOP(b)>, whereas the matching hallucination has content of the form <MOP(_)>, where ‘MOP’ names the mode of presentation in question, and ‘b’ names the particular badger, and ‘_’ indicates a gap. (I’m simplifying in my outline here, as on all of Schellenberg’s examples the content is fuller, involving property-instances and modes of presentation of those. See pp. 88-91.) Schellenberg assures us that there is ‘nothing metaphysically spooky about gaps’. Phew! Rather, the ‘gap simply marks the failure to single out a particular’ (p. 89). As it stands, however, it is not clear why this should be a common-kind account. It looks as though the perceptual case and the hallucinatory case differ significantly: perception has object-involving content, whereas hallucination has gappy content. This is true, Schellenberg admits, but only at the level of token content. When it comes to the type of content, there is identity, and this is what makes the account a common-kind account. The type of content common to the two cases is: <MOP[__]>, with <MOP(b)> and <MOP(_)> being different token contents of this type (p. 91). But on Schellenberg’s account, it’s not just that my perception and hallucination have the same content type; they also have the same phenomenal character. A simple option would be to explain phenomenal character in terms of content, and to explain the sameness of phenomenal character in terms of sameness of content. This isn’t quite what Schellenberg holds – at least, it is not the whole story, and it neglects to highlight the fundamental element in her story. For her, there is another, more basic element which we need to add to the explanation: perceptual capacities. The idea is that the experiences have the same character because they have the same content type (so it is a content-based common kind account), but this is because the same perceptual capacities are employed in both cases. The employment of perceptual capacities is what yields perceptual content, and so what ultimately accounts for phenomenal character and the sameness of phenomenal character across these cases. Thus: How should we understand the content types? According to Fregean particularism, a perception, a hallucination, and an illusion with the same phenomenal character share a metaphysically substantial common element: the perceptual capacities employed. Employing perceptual capacities yields a content type that experiential states with the same phenomenal character have in common. (pp. 91–92) Schellenberg argues that employing perceptual capacities constitutes content. Applied to the case at hand, I take it that Schellenberg would hold that employing a perceptual capacity for discriminating and singling out badgers or badger-like creatures is constitutive of my perception having content involving MOP. For MOP is, after all, just the specific way in which I single out the perceived badger. Schellenberg develops her view of phenomenal character, or perceptual consciousness further in Chapter 6 ‘Perceptual Consciousness as Mental Activity’. In that chapter she argues that ‘perceptual consciousness is constituted by a mental activity, namely the mental activity of employing perceptual capacities’ (p. 141). In the course of her discussion of perceptual consciousness Schellenberg nicely summarizes the common-kind aspect of her overall view that we’ve just been focusing on: Consider Hallie who suffers a hallucination as of a white cup on a desk. Like Percy [a subject who perceives a white cup on the desk], she employs the capacity to discriminate and single out white from other colours and she employs the capacity to differentiate and single out cup-shapes from, say, computer-shapes and lamp-shapes. Since she is hallucinating rather than perceiving, and so is not perceptually related to a particular white cup, she employs these capacities baselessly. Yet even though she fails to single out any particular white cup, she is in a phenomenal state that is as of a white cup, in virtue of employing perceptual capacities that purport to single out a white cup. As in the case of perception, employing these perceptual capacities constitutes her phenomenal character. So what perception, hallucination, and illusion have in common is that perceptual capacities are employed that constitute the phenomenal character of the relevant experiential states. (p. 152) So far, then, we can see not only what some of Schellenberg’s main claims are, but capacitism in action. Perceptual capacities are at the heart of her accounts of content and consciousness. What about the other strand of Schellenberg’s work, evidence, or more generally, the epistemology of perception? Schellenberg has a lot to say in the epistemology of perception. For instance, in Chapter 8, she discusses and rejects luminosity (cf. Williamson 2000, Chapter 4). In Chapter 9 she presents a new analysis of perceptual knowledge with perceptual capacities at its heart. On her analysis, assuming that S has evidence sufficient for knowledge (more on which below), ‘S has perceptual knowledge that p if and only if p is true, S employed a capacity to single out what she purports to single out, and S’s mental state has the content it has in virtue of S having successfully employed her capacity to single out what she purports to single out’ (p. 206). In Chapter 10, Schellenberg distinguishes her views in the epistemology of perception from other views, such as knowledge-first approaches, versions of reliabilism, and versions of virtue epistemology. But it strikes me that Schellenberg’s most significant contribution here comes in Chapter 7, on perceptual evidence. In this chapter, Schellenberg argues that perception provides us with two types of evidence: phenomenal evidence and factive evidence. Phenomenal evidence, Schellenberg notes, ‘is determined by how our environment sensorily seems to us when we are experiencing’ (p. 167). Hallucinations too provide us with phenomenal evidence. Factive perceptual evidence, on the other hand, is ‘necessarily determined by the perceived particulars such that the evidence is guaranteed to be an accurate guide to the environment’ (p. 167). Hallucinations do not provide us with such evidence. So, when I see the badger scrubbing around under the tree, I have phenomenal evidence in that it seems to me that a badger is scrubbing around under the tree. The same is true of a matching hallucination. But I also have a stronger kind of evidence in the perceptual case (but not the hallucinatory case), factive perceptual evidence, in virtue of my perceptual link to the particular badger in question. Factive evidence is, on Schellenberg’s view, sufficient for knowledge, whereas phenomenal evidence is not (p. 205). With this account, Schellenberg is able to neatly capture two intuitions: on the one hand the intuition that hallucinations are not epistemically empty—they do provide us with some sort of evidence; on the other hand the intuition that genuine veridical perception provides us with more or better evidence than hallucinations—evidence sufficient for knowledge. In developing this view of perceptual evidence, Schellenberg once again puts perceptual capacities to work. For she argues that both phenomenal and factive evidence ‘have their rational source in the perceptual capacities employed in experience’ (p. 183). For the phenomenal evidence an experience provides is determined by the content type of the experience it involves. And this is constituted by the employment of perceptual capacities. By contrast the factive evidence an experience provides is determined by the token content of the experience. And this is constituted by the employment of perceptual capacities (together with the relevant particulars). Thus, Schellenberg provides a ‘unified account of perceptual evidence’ (p. 167) with perceptual capacities doing the unifying work. She argues for this view in Chapter 7, details its advantages on pp. 185-187, and puts it to work in the subsequent epistemological discussion in Chapters 8 and 9. From this brief overview of some of the main claims that Schellenberg argues for, I hope it is clear that this book is wide-ranging, and that it will appeal to philosophers across many different branches of the philosophy of perception, and epistemologists too. The book is also quite comprehensive: as well as making original contributions, Schellenberg surveys existing contributions and situates her view in relation to others. By way of critical comment, I will limit myself to Schellenberg’s argument in the first chapter for the thesis that our perceptions are partially constituted by the particulars we perceive. Schellenberg argues as follows:   (I) If a subject S perceives particular α, then S discriminates and singles out α (as a consequence of being perceptually related to α).   (II) If S discriminates and singles out α (as a consequence of being perceptually related to α), then S’s perceptual state M brought about by being perceptually related to α is constituted by discriminating and singling out α.  (III) If S’s perceptual state M brought about by being perceptually related to α is constituted by discriminating and singling out α, then S’s perceptual state M brought about by being perceptually related to α is constituted by α. From: I-III: If S perceives α, then S’s perceptual state M brought about by being perceptually related to α is constituted by α. (pp. 24–25) I will focus my critical attention on premise I. Premise I specifies a condition on perceiving a particular: if one perceives a particular, one must discriminate and single it out from ‘its surround’ (p. 25). Schellenberg notes that the ‘necessity in question is metaphysical necessity (not logical or natural necessity)’ (p. 25). And she holds that this condition is ‘modality general’, not restricted to vision (p. 25). What does Schellenberg say in support of this premise? In support of Premise I we can say that it is unclear what it would be to perceive a particular without at the very least discriminating and singling it out from its surround. Consider a perceiver who sees a white cup on a desk. He employs his capacity to discriminate white from other colours and to single out white in his environment. Similarly, he employs his capacity to differentiate and single out cup-shapes from, say, computer shapes and lamp-shapes. Such discriminatory activity allows for scene segmentation, border and edge detection, and region extraction. If there is no discriminatory activity, it is unclear how he could be perceptually aware of the cup. (p. 25) My first comment is that these remarks seem most relevant to supporting the condition as applied to vision and touch. It is not clear how they carry over to, say, audition and olfaction. And, further, it seems to me that reflection on cases of olfaction and audition might put pressure on the condition. Consider, first, the following case. I am in a dreamless sleep. I awake and hear a sound: that of a sustained single note played on an electronic organ. (The neighbour is recording some experimental music.) I hear this sound for about thirty seconds before falling straight back into the dreamless sleep. I heard the sound but did my hearing the sound involve my discriminating it from its surround? This is not obvious. And one could argue that the sound has no surround. That is, there is nothing else auditorily perceived (or auditorily apparent) to me, either before, after, or during my hearing of the sound, from which I discriminate it. The olfactory case is a simple modification of the auditory case. Suppose that upon waking from my dreamless sleep I smell the scent from a scented candle by the bedside. I get a whiff of it and return to my dreamless sleep. Do I discriminate this smell from its surround? Again, it is not obvious that there is an olfactory surround: there is nothing else olfactorily perceived (or olfactorily apparent to me), either before, after, or during my smelling of the smell, from which I discriminate it. One reply to the auditory case is that when I hear the sound of the organ I do auditorily discriminate it from an aspect of its surround after all. I auditorily discriminate it from the silence I hear elsewhere in the auditory scene. But whether this reply can be developed satisfactorily is far from obvious. First, it relies not only on the controversial idea that we can hear silence, but the more controversial idea that we can hear silence at the same time as hearing sound. Now, even if we can persuade ourselves of the viability of these ideas, we can stipulate that in this case the sound of the organ is very loud and prominent such that it crowds out not only other sounds, but surrounding and concurrent silences. So, once again, we seem to have a case of hearing a sound without discriminating it from its surround. It may be peculiar, but a case as described does seem to be metaphysically possible, which is all that is required to challenge Schellenberg’s condition. And a similar point can be made in response to the equivalent reply to the olfactory case. Do these objections operate with an overly demanding notion of discrimination? Schellenberg’s claim is that we auditorily discriminate particulars by registering differences between them (p. 37), and this is what hearing them amounts to. But does highlighting this help to respond to the points above? It is not obvious that it does. For the case can be re-described as one in which there is not, at any point, any relevant auditory registering of differences in play. In the auditory case we can suppose that no silence is auditorily registered or otherwise brought into the mix, and thus there is no question of registering any differences between the sound and surrounding silence. There is no relevant auditory registering of differences for the hearing of the sound to amount to. (Similarly for the olfactory case.) Though the above is assertive, we can perhaps more helpfully frame the point as an explanatory challenge: how is it possible for me to hear the sound of the organ in this case, given the details of the case, and given the discrimination requirement on hearing? This breaks down into a number of different questions: what do I discriminate the sound from? An aspect of its surround? But it seems to have no surround. So perhaps we can include silence as an aspect of its surround. But then how does silence get into the mix such that I register the difference between it and the sound (such as to constitute my hearing the sound)? And—supposing this can be addressed—why think that this is what my hearing the sound amounts to? (Similar questions arise with respect to the olfactory case.) But what about the condition as applied to vision? Most of Schellenberg’s discussion in the book is about visual perception, so perhaps she would be happy to settle for a version of premise I and the argument restricted to vision. But even then, I think we can challenge premise I. Dretske (1969) endorses a similar condition to Schellenberg’s premise I (as restricted to vision), for he holds that one sees an object only if one visually differentiates it from its surroundings (p. 20). But he presents a challenge to this condition with a case we can call Nose: Touch your nose to a large smooth wall and stare fixedly at the area of the wall in front of you. There is not much doubt about the fact that you see the wall, or at least a portion of it. It is also fairly clear that you do not differentiate it from its immediate surroundings. In this position it has no environment …. (Dretske 1969, p. 26) This presents a challenge to Schellenberg’s premise I, even once restricted to vision, as it describes a possible case in which a subject sees a particular—a wall—without visually discriminating it from its surroundings. Though she doesn’t credit it to Dretske, Schellenberg is well aware of this case and the potential challenge it poses to her view. We’ll come to her own response to it shortly. But first let’s consider Dretske’s response and how it might help Schellenberg. In response to Nose, Dretske simply restricts the visual differentiation condition. He suggests that the condition is applicable just to those cases where surroundings are perceived or apparent (pp. 26-27). Similarly, Schellenberg might restrict her premise I and argument even further so that the condition is only applicable to cases of vision where the surroundings of the particular in question are visually perceived or apparent to one, or else available to the subject so as the subject can register differences (p. 37) between them and the particular. However, I now want to suggest that we can challenge even this condition. Consider the following case, Invisible Frame, adapted from French (2018, pp. 144-145): Imagine a modern art gallery known for its quirky installations. In this gallery one of the installations is a large frame, on one of the gallery walls. The frame is empty; it doesn’t house anything. Thus, when installed, one can see right through the frame’s rectangular gap to the wall it is attached to. Suppose also that the frame is designed so as to visually blend in with the wall behind it. That is, suppose the colour and texture of the frame’s surface matches the colour and texture of the wall’s surface. Suppose also that although the frame is very large, in that it effectively ‘frames’ a large area of the wall, its structural parts (the lengths of material used to construct the rectangular shaped frame) are extremely thin, so that when attached to the wall it barely extends out from the wall. This helps to reduce shadows and other depth cues which might otherwise prevent the visual blending or camouflage effect that the artist is going for. Suppose also that the lighting is carefully designed so as to help bring about this effect. Let’s suppose that viewers don’t just fail to notice the frame; they literally can’t see it. This makes sense if seeing a thing requires it to be discriminated from its surround. Yet the frame still looks some way to such viewers in that it is an element of the scene before them, in their field of vision, which makes a positive contribution to how the overall scene looks to them. (See Dretske 1969, pp. 23–24.) Now consider a subject, Juliet. Suppose that she is viewing the installation. Juliet has no idea that there is a frame there. Suppose she puts her nose right up to the wall so that all that is in her field of vision is the wall. (She doesn’t get any of the frame in her field of vision at that moment.) As noted in relation to Nose, intuitively in such conditions Juliet can see the wall, even though she doesn’t visually discriminate it. But now suppose she gradually moves back so that a bit of the frame enters her field of vision, but she cannot visually discriminate it from the wall. In these circumstances, it seems plausible to suppose that Juliet still doesn’t see the frame, but would we be inclined to think that she now doesn’t see the wall before her? Surely not! But at this moment a bit of the frame enters her field of vision and looks some way to her, so now the wall does have an immediate environment, but she doesn’t visually discriminate the wall from this environment. The problem is that if we restrict Schellenberg’s premise I in the way that Dretske suggests, then we have the result that when Juliet gradually moves away from the wall and a little of the frame comes into her field of vision, she no longer sees the wall. But this is counterintuitive. So Schellenberg is right not to endorse a Dretskean response to Nose. Let’s look, then, to how Schellenberg does respond. First, Schellenberg highlights a response which she recognizes is flawed. She notes that ‘when we stare at an undifferentiated and uniform field of colour the ganzfield effect sets in: after a few minutes, one simply sees black and experiences an apparent sense of blindness due to the lack of structure in one’s environment’ (p. 27). Now, one issue with this response (which Schellenberg doesn’t highlight) is that though it may tell us what actually happens, it doesn’t tell us what must happen as a matter of metaphysical necessity. For all the response says, it is metaphysically possible for one to see a wall that is uniformly coloured and that fills one’s entire field of vision without the ganzfield effect setting in. The issue that Schellenberg herself notes is that the ganzfield effect sets in after a few minutes. So the critical question remains: ‘what is going on before the ganzfield effect sets in?’ (p. 27). Someone resisting Schellenberg’s approach will maintain, in line with the verdicts encouraged by Nose and Invisible Frame, that before the effect sets in, the subjects see the uniformly coloured wall even though they do not visually discriminate it from its surroundings. In contrast, Schellenberg says the following in immediate response to her question about what happens before the effect sets in: The particularist can argue that the subject employs perceptual capacities insofar as she is discriminating the part of the uniformly coloured wall to her right from the part of the wall to her left. While the different parts of the wall have the same colour, they occupy different locations within the subject’s egocentric frame of reference. So she is employing perceptual capacities to discriminate the parts of the uniformly coloured wall within her egocentric frame of reference’. (p. 27) But it is unclear how to interpret Schellenberg’s move here. She suggests that the discrimination condition in her Premise I is met because subjects can discriminate one part of the wall from the other part of the wall. If that is true, then Schellenberg’s Premise I is compatible with subjects seeing parts of the wall in Nose and Invisible Frame. How does this help her respond to the objection? On one interpretation of her response, the claim is that before the effect sets in—in Nose and Invisible Frame—the subjects don’t see the wall, but only certain parts of it. If that is what Schellenberg is claiming, she doesn’t motivate it. For she doesn’t speak to the intuition that the subjects in our cases can see the wall right before their eyes, and not just parts of it. (Note that they might not see the whole of the wall, but that is no bar to them seeing the wall.) On a second interpretation, Schellenberg is claiming that before the effect sets in, the subjects are seeing the wall, but this is compatible with her Premise I because there is appropriate discriminatory activity occurring after all, namely the discrimination of one part of the wall from the other part of the wall. But the problem here is that this move shifts attention away from the wall which is allegedly seen in Nose and Invisible Frame to parts of the wall. Our cases encourage the thought that the wall is seen despite not being visually discriminated from its surroundings. This is counter to Schellenberg’s claim that if a particular α is seen then it must be discriminated from its surroundings. It is not clear how it helps with this to note that when α is seen some of its parts are discriminated from each other. Schellenberg’s initial condition is not that for S to see α S must engage in some discriminatory activity or other, or even some α-related discriminatory activity or other. It is more specifically that S must discriminate α itself from its surroundings. This is not secured in Schellenberg’s reply. The problem cases I’ve presented hinge on the idea that one can perceive a particular without discriminating it from its surroundings. My framing of things in this way derives from Schellenberg’s framing of things in this way in the support that she gives for her premise, and the claim that ‘it is unclear what it would be to perceive a particular without at the very least discriminating and singling it out from its surround’ (p. 25, emphasis added). Yet Schellenberg’s premise doesn’t say that if S perceives particular α, then S discriminates and singles out α from its surroundings. It merely says that if a subject S perceives particular α, then S discriminates and singles out α. But whether this observation helps to respond to the above cases is unclear. For to discriminate something is to discriminate it from something else. And what, if not an aspect of the relevant particular’s surroundings, does the subject discriminate the particular from in these cases, such that they perceive the relevant particular? It is not clear how Schellenberg would answer this question. So, to conclude, though I am sympathetic to the particularist position that Schellenberg advocates, I don’t think that Schellenberg has yet developed a plausible capacitist argument for the view. This doesn’t, of course, support a general scepticism about Schellenberg’s carefully developed capacitist project. The book is packed full of capacitist arguments that I haven’t touched upon here, and which I’m sure will generate much fruitful discussion.* Footnotes * Thanks to Susanna Schellenberg and Anil Gomes for discussion. References Dretske Fred ( 1969 ). Seeing and Knowing . Chicago : University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC French Craig ( 2018 ). ‘Object Seeing and Spatial Perception’. In MacPherson Fiona , Dorsch Fabian (eds.), Phenomenal Presence . Oxford : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Williamson Timothy ( 2000 ). Knowledge and its Limits . Oxford : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © Mind Association 2019 This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - The Unity of Perception: Content, Consciousness, Evidence, by Susanna Schellenberg
JO - Mind
DO - 10.1093/mind/fzz026
DA - 2020-01-01
UR - https://www.deepdyve.com/lp/oxford-university-press/the-unity-of-perception-content-consciousness-evidence-by-susanna-l6qXE7ACtv
SP - 339
VL - 129
IS - 513
DP - DeepDyve
ER -