Nonconceptual content and the space of reasons

Philosophical Review

Volumn 109, Issue 4, 2000, Pages 483-523

  Heck, Richard G ^a  

^a HARVARD UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0041982127     PISSN: 00318108     EISSN: 15581470     Source Type: Journal    
DOI: 10.1215/00318108-109-4-483     Document Type: Article

References (47)

1. Gareth Evans, The Varieties of Reference (Oxford: Oxford University Press, 1982). See, in particular, sections 5.1, 5.2, and 7.4. Further references are in the text, marked by ' VR and a page number.
2. John McDowell, Mind and World (Cambridge: Harvard University Press, 1994), chap. 3 and part 2 of the afterword. Further references are in the text, marked by 'MW' and a page number.
3. Unless, of course, one held that all content is nonconceptual-though, then, the claim so stated would not register any difference between belief and perception.
4. Jim Pryor suggested to me that it might be better to speak of "conceptually-constituted content" and "nonconceptually- constituted content." This seems reasonable. But, as the terminology used in the paper is well established, I am afraid we are stuck with it.
5. I suspect that the state view is indefensible-even incoherent, if coupled with the claim that the contents of beliefs are conceptual-but I shall not argue this point here.
6. See Robert Stalnaker, "What Might Nonconceptual Content Be?" in Concepts, ed. Enrique Villanueva (Atascadero, Calif: Ridgeview, 1998), 339-52, in which he argues, pace McDowell, that the contents of beliefs and perceptions are both nonconceptual. Stalnaker understands very well that, once the nature of conceptual content has been properly understood, the claim that beliefs have conceptual content can be seen to be controversial; the claim that perceptions do, even more so.
7. Of course, it is an open question whether McDowell understands this thesis as Evans does: But McDowell certainly does not use the term 'conceptual content' in such a way that the claim that beliefs have conceptual content is trivial. Failure to grasp this point will make certain aspects of Mind and World hard to understand.
8. I think Evans actually meant more by the claim that beliefs have conceptual content. In particular, someone who took the contents of beliefs to be sets of possible worlds could agree with what has been said so far- could profess to accept the Generality Constraint-so long as she conceived of grasp of a concept as an abstraction from an ability to entertain certain sorts of Thoughts (and of concepts themselves as abstractions from a class of Thoughts). For Evans, though, possession of the cognitive capacities that constitutes grasp of concepts is supposed in some sense to be prior to, or independent of, one's capacity to entertain such Thoughts as these: Possession of such capacities-grasp of certain concepts-is supposed to figure in an explanation of one's ability to entertain the Thoughts one can. One might say, then, that, for Evans, Thoughts are conceptual in so far as they are intrinsically conceptually articulated.
9. This sort of issue was discussed, some time ago, in connection with the question of what justifies the attribution of knowledge of a structured, axiomatic theory of meaning to a competent speaker. See Gareth Evans, "Semantic Theory and Tacit Knowledge," in his Collected Papers (Oxford: Oxford University Press, Clarendon Press, 1985), 322-42;
10. Martin Davies, "Tacit Knowledge and Semantic Theory: Can a Five per cent Difference Matter?" Mind96 (1987): 441-62;
11. Crispin Wright, "Theories of Meaning and Speakers' Knowledge," in his Realism, Meaning, and Truth (Oxford: Blackwell, 1986), 204-38.
12. For remarks in a similar spirit, see M. G. F. Martin, "Perception, Concepts, and Memory," Philosophical Review 101 (1992): 745-63, at 758-59.
13. McDowell is concerned to emphasize that my demonstrative concept need not die with my experience: This is the reason for the parentheticals. But, having inserted them here, I shall omit them below, as it will be enough for us to focus on the situation that gives rise to the concept in the first place.
14. If to say that it appears as if that part of my desk is that color is to give a completely faithful report of (part of) the content of my experience, and if perceptual content is conceptual, I could not have such an experience unless I had the demonstrative concept that color.
15. One could have such information without having such a concept, if the state view were correct. (This is one sort of reason it might seem appealing.) But I mean to be setting that view aside here.
16. Independently, maybe, even of your understanding what I have said, since, on Evans's view, you do not understand what I have said if you are not, yourself, in a position to perceive the relevant part of my desk (or have some other informational link to it), though you do know what kind of thing I have said.
17. There are cases in which the demonstrative need not refer to the color of the desk. If there is a transparent film of color covering the desk, the demonstrative 'that color' could refer to the color of the film, in which case my utterance of 'That part of the desk is that color' need not be true. But I am interested here in what we might call "pure" cases of misperception, cases in which I simply misperceive the color of the desk, not because of some external interference, such as films of color, funny lighting, etc. I find, for example, that it is frequently the case that, if I shut one eye, colors take on a slightly greenish tint, whereas, if I shut the other, they take on a comparatively reddish tint. That is the kind of thing I have in mind. (Such "pure" misperception can occur in film cases, too, of course: If I misperceive the color of the film, no "external" color need be available to determine the content of my experience.)
18. Just as Evans would insist that I cannot have a demonstrative concept of the little green man who appears to be sitting on top of my monitor, if there is no such man there, perhaps he would also insist I can have no demonstrative concept of the color the relevant part of my desk appears to me to be, if I am misperceiving its color. But the intuition is very strong that if my experience presents my desk as being a certain shade of brown, I can form a concept of that particular shade. It will not do simply to deny it.
19. Perhaps the beliefs I form on the basis of my experience are, by and large, true. But I see no reason to think my total perceptual state is usually veridical. My total perceptual state almost always represents the world as containing something that is ringing at a particular frequency, which there rarely is, and my visual experience misrepresents depth with an astonishing regularity. And because I wear glasses, the periphery of my vision almost always represents the world in a most peculiar fashion. (It is also worth remembering that illusions, like the Müller-Lyer illusion, highlight quite ordinary features of perceptual experience. That is why they are interesting.)
20. Of course, the question what to do about the richness of nonveridical perception would remain open: But perhaps that would not trouble McDowell.
21. Note that this is different from saying that my belief has the following content: That, if it is true, then snow is white, and, if it is false, then snow is pink. We certainly can, and I certainly do, have beliefs with that sort of content.
22. Some theorists of perception have denied this claim, namely, those who think that perceptual knowledge is to be justified by an inference to the best explanation: The existence of an external world causing them is supposed to be the best explanation of our having the sense-data we do (see Frank Jackson, Perception: A Representative Theory (New York: Cambridge University Press, 1977)). I take it that McDowell's objection to such views would be one to be mentioned in note 21 below, namely, that they leave it obscure how we could so much as have concepts of external objects and properties. But I do not intend to evaluate this claim here.
23. McDowell presents Davidson's view as an alternative to the sense-datum theory, which is certainly how Davidson himself conceives it. See Donald Davidson, "A Coherence Theory of Truth and Knowledge," in Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson, ed. E. LePore (Oxford: Blackwell, 1986), 307-19. But it seems to me that if the objection McDowell brings against Davidson works at all, it works against this view, too: see MW, 13-8.
24. One thing that is distinctive of McDowell's treatment of these issues is his connecting them with questions about how, if the sense-datum theory were correct, we could so much as have beliefs, not just about our sense- data, but about the world itself-let alone how these beliefs might be justified: How, McDowell wants to ask, could we so much as have a concept of a square object if that concept never made contact with square objects? indeed, with anything outside our minds? But I shall not discuss this aspect of his presentation in any detail, since I do not think it bears directly upon the concerns of this paper.
25. McDowell himself describes Evans's view as being, for just this reason, "deceptively innocent" (MW, 53) and admits that his charge that it is a version of the Myth of the Given "can seem difficult to sustain" (MW, 162).
26. I have heard it objected that an organism will only be able to make use of the information experience provides insofar as it can conceptualize it, where the conclusion is supposed to be that the content of the experience-the content it has for the organism-is limited by the concepts the organism possesses. But the inference from premise to conclusion is questionable (it ignores the distinction between concepts one has independent of the experience and experience-dependent concepts that it might make available) and the premise is independently doubtful (see note 27 and the text to which it is attached).
27. See A Study of Concepts (Cambridge: MIT Press, 1992).
28. See note 31 and the text to which it is attached for some further remarks on this issue.
29. Although the comparison with testimony can be misleading if not treated with care-it tends to suggest the representational theory considered in the last section-we might say that perception is more like a little voice saying, "There is a desk in front of you." (See VR, 122-23.)
30. This suggests that perceptual experience can influence action otherwise than by giving rise to beliefs. (Evans might have had something like this in mind when he wrote of connections between the informational system and action that are "more primitive" than its connections to belief.)
31. I use this term, rather than 'representational', because the term 'representational content' is in such common use. (The term has also been used by John Searle: See his Intentionality: An Essay in the Philosophy of Mind (Cambridge: Cambridge University Press, 1983).) Indeed, when people speak of perception as having representational content, this seems to me to run together the two aspects of perceptual experience I am trying to distinguish. One does not easily speak of intentions as having representational content.
32. Of course, it would be nice to know more about what it means to say that perceptual states are "assertive." Unfortunately, I do not know how to explain this. Nor, however, do I know how to explain what is involved in a belief's being assertive: What exactly does it mean to say that beliefs purport to represent how the world is? I think an answer to this question could be converted into an answer for the case of perception, too: At least, that is the point of my relying upon this analogy here.
33. The occurrence of this word should set off no alarms: When people deny that perceptual knowledge is inferentially justified, what they mean, typically, is that it is not inferred from any prior beliefs about my experiential states. The view I am developing can acknowledge this point: If it is right to say that perceptual beliefs are inferred, they are inferred from prior presentational attitudes about the world.
34. Although I have no special interest in retaining the word 'inferred', it seems to me to be useful. In making an inference, although there is a sense in which I must recognize that the premise entails the conclusion, if the inference is to yield knowledge, my recognizing this need not, and in general cannot (on pain of Lewis Carroll problems), involve my having the belief that it does so. So room has to be made, even here, for my recognizing the existence of such a semantic relation, without my forming a belief that it obtains.
35. I would certainly understand a request to be told more about the nature of this transition-and I wish I had something positive to say about it, that is, something to say about conceptualization, which is what the translation between types of representational content involves. But I do not see any reason for skepticism that there is something to be said here: The problem is not one that should be underestimated, but Peacocke has done much to explain in what a grasp of certain basic, observational concepts might consist and how possession of them is related to the capacity to form beliefs on the basis of perceptual experience. (See A Study of Concepts, especially §1.2, and 79-80; compare VR, 229.) McDowell discusses and criticizes certain of Peacocke's proposals, but his objections mostly concern Peacocke's right to speak of a subject's reasons (see MW, 162-63). It is that right, of course, that I am defending.
36. Evans's claim that their content will systematically depend upon that of my perceptual states is too weak: I take this just to be a slip.
37. So, on Evans's view, there is something right in the idea that statements of the form 'It appears to X as if p' mean that X has prima facie justification for believing that p. I do not think we should take that as an analysis of judgments of appearance, though, since I think this analysis would miss the fact that such judgments do make reference to the underlying perceptual states themselves. See note 38, and the text to which it is attached, for more on this issue.
38. There is no reason Evans needs to insist that judgments about how things appear are incorrigible, especially certain, or what have you: So his view need not inherit any of the problems often thought to attach to views that allow real judgments to have such properties.
39. Of course, one might also exercise other conceptual capacities in evaluating these reasons-one might as well bring one's full cognitive capabilities to bear-and one typically will. But this point is of no help to McDowell either.
40. A diagnosis of McDowell's reflections not unlike mine is given by Alex Byrne: See his "Spin Control: Comment on John McDowell's Mind and World," Philosophical Issues 7 (1995): 261-73, at 267-68. In his "Reply to Gibson, Byrne, and Brandom," 283-300, at 299ff., McDowell responds in a way that suggests he takes Byrne's diagnosis to rest upon the assumption I am here disowning.
41. It seems to me that one can truly say that it appears to me as if p, even if I lack the concepts required if I am to form that judgment, so long as I would so judge, if I did have those concepts (cf. Wright, "Human Nature?" 244-45.) But it does not seem to me that it matters for our purposes what view one takes here.
42. I hesitate to say that it is for me to judge that I am in a state that gives me reason to hold the belief that p, because it is not clear to me that Evans's account of judgments of appearance should be taken as an analysis of them. It is not, in particular, clear to me that, in order to be able to make judgments of appearance, a creature has to have such concepts as that of a reason, has to be able to think about how she would judge under certain circumstances, and so on and so forth. Compare note 33.
43. This difficulty may be caused by a linguistic tangle: We speak of beliefs as being reasons for holding other beliefs; but, if I infer q from p, my reason is not that I believe that p, but just that p, where that is something I believe. So it is okay to say that my belief is my reason, so long as it is clear that my reason is what I believe, not the fact that I believe (though, of course, what I believe can only be a reason for me so long as I do believe). And similarly, if we say that my perceptual state is my reason, it is not that I am in that state (that is, the fact that it appears to me that p) that is my reason, but how things appear.
44. The idea that there is such a connection is so deeply embedded in that tradition that-in the form of a thesis that an account of thought can only be given via an account of language-it was once enshrined by Michael Dummett as "the fundamental principle of analytic philosophy." See his "Language and Communication," in his Seas of Language (New York: Oxford University Press, 1993), 166-87, at 171.
45. For some interesting, and related, reflections, see Alexander George, "Has Dummett Over-salted His Frege? Remarks on the Conveyability of Thought," in Language, Thought, and Logic: Essays in Honour of Michael Dummett, ed. R. Heck (Oxford: Oxford University Press, 1997), 35-69.
46. One that includes "experiences" among the things that must cohere. Compare Wright, "Human Nature?" 241-42.
47. Whose tongue lashes out at any moving black thing, and not just at flies.