Perception: Objects and Contents (McDowell Lecture 1)
I’m astonishingly busy at the moment. I have some serious writing commitments, and I haven’t had any real time to rest in between travelling over the last week or two. I owe so many people so many things, and they’re coming, so please accept my apologies if you’re waiting on anything. I really should write up some thoughts on Markus Gabriel’s lecture ‘The Meaning of Existence and the Contingency of Sense’, which I attended at CRMEP yesterday. However, I’ve just come back from the first of McDowell’s two Edgington lectures on the topic of perception, and I promised to write up a summary of them for a few people who weren’t there, so I took notes during it. This means all I have to do is dump these here, rather than having to write up something from scratch, thankfully.
These notes are a bit terse, but they’ve got the crucial points, and they also come with a certain parsing of what I take McDowell’s position, and a few potential challenges to it at the end.
1. Perception places our surroundings in view (at least visual perception). Visual perception is the major topic of this lecture.
2. MacBeth does not ‘see’ a dagger before him, because it doesn’t place his surroundings in view. This is because his visual experience does not (nor can it) provide knowledge. Uses of the verb ‘see’ whose purpose is to talk about non-disclosive visual experience are perfectly fine, but they must be rigorously distinguished from those whose purpose is to talk about disclosive experience. ‘Sight’ in the sense to be used here (placing our surroundings in view), things are visually present to us in a way that can produce knowledge.
3. Naive realism is supposed to be in tension with the fact that our visual experiences can ‘seem’ to provide us with knowledge (or place our surroundings in view) when they do not. McDowell thinks this tension is an illusion. The point of these lectures is to reveal this illusion.
4. There is a plausible position held by some that any possible descriptions of the content of our experiences by means of ‘that’ clauses will never exhaust this content. However, regardless of the truth of this thesis (call it the excess thesis), it does not prevent such descriptions from being true. They can be true without exhausting the possible questions we could ask about the content of the experience.
5. There is a further sense of ‘see’ that we must exclude here, namely, the sense in which we can correctly say that we ‘see’ a red rectangle even when our experience does not furnish us with this content, such as when we ‘see’ a red rectangle in a darkened room in which its colour is indeterminate. This sense of ‘see’ is dangerous insofar as it can tempt us to let the experiential content we’re interested in fall out of the picture entirely.
6. One justification of the idea that there is a tension in naive realism is indirect realism, or the idea that we principally encounter representations of our environment and its features, rather than the environment and its features themselves. What this shows is that we have to be very careful in deploying the notion of representation in talking about the nature of experiential content.
Q1. Transparency: Metzinger has theorised that the transparency of our experience, the self-evidence of the presence of a pig, as opposed to the evidence of its presence we find when we see pig-tracks, is a structural feature of our brains qua representational systems, and thus should be studied empirically rather than philosophically. This is about how we think about direct evidence as opposed to indirect evidence, in phenomenological terms. Whether we’re interested in introspective or extrospective phenomenology.
7. There are two premises that seem inconsistent with naive realism when taken in conjunction:-
i) If we attribute representational content to experience, then we’re committed to the idea that its subjective character is connected to the way in which it brings the environment into view in some appropriate way.
ii) An experience can represent things as being a certain way in a subject’s environment when they aren’t that way in the subject’s environment.
8. This seems to commit us to the idea that there could be an experience with a subjective character (e.g., transparency) that indicated the environment was brought into view, without it actually bringing it into view. This seems inconsistent with naive realism insofar as naive realism is supposed to be wedded to this particular subjective character.
9. It is of course optional to use the term ‘representation’ in describing perceptual content, but there is a particular way of using this term that reveals why this argument doesn’t work. What the argument misses is that in the case of an experience that is a perceiving (or seeing) the representing that is described as doing (in terms of its transparent subjective character) is purely a matter of disclosing, or bringing into view. This is an externalist characterisation, rather than an internalist characterisation of representation. It is about representational success, rather than representational purport.
Q2. Functionalism: It seems as if the point here is that the representational character of experiential content is understood in functional terms, but these functional terms are not purely internal ones regarding the functional economy of the subject’s internal states qua causal system, but are external ones regarding the functional economy of the subject’s internal states in their relation to its environment qua situated causal system. The point is that the explanation of the representational character of certain states should take the form of a functional externalism, rather than functional internalism.
10. On this view, the attribution of representational content to experience is consistent with naive realism.
11. Philosophers often present the crucial issue with perception as its veridicality, but it is important to realise that there can be veridical experiences that do not actually reveal the environment (perceptual Gettier-style cases). These experiences are veridical but nonetheless defective. This defect is something which is specific in functionally externalist terms. The subjective character is not something that the subjects themselves have privileged access to.
12. In short, it can seem to us that our surroundings are in view, even when they are not. We can be wrong about whether our surroundings are in view.
Q3. Subjective Character: Does it really make sense to say that the states one is talking about in such an externalist functional account are really subjective? Subjectivity is usually associated with internalism, and objectivity with externalism, and this characterisation of experience cuts across this traditional divide in way that potentially distorts the traditional understanding of the terms. Is there a useful non-traditional concept of subjectivity here? It might be that the disagreement between McDowell and Metzinger is thus entirely linguistic, namely, a disagreement about precisely what we can call ‘subjective’.
13. The internal aspect of the subjective character of transparency/revelation does not exhaust the subjective character entirely, which also has an external aspect. It is the fact that the representational content of experience has both a internal dimension (purport) and an external dimension (success) that is supposed to differentiate McDowell from an indirect realist like Metzinger on the one hand, and an experiential nihilist like Brandom.
Q4. Between a Rock and a Hard Place: McDowell would respond to Metzinger that perception has an external component as well as an internal component, but Brandom will ask why, once we recognise there is an external component we even need the internal component anymore. The real problem for McDowell seems to be that he is interested in telling a story about the subjective character of perceptual content in terms of representational purport and representational success, but he seems to want to explain each representational dimension in a different register. He wants to describe success in functionally external terms, but he doesn’t want to describe purport in functional terms at all, but in introspective phenomenological terms, and this seems to pull the very unity of the notion of representation apart here.