McDowell
These are some thoughts on McDowell prompted by the lecture on Friday at MIT. They represent at attempt to bring together an understanding of that talk with the other things of his I've read--Mind & World and a handful of essays.
The McDowell talk at MIT on Friday was very interesting. Several of my questions were asked, and his responses were often those I hoped or expected him to give. Not that the talk entirely made sense. The first part was about modest & immodest transcendental arguments, and he claimed to be offering a transcendental argument that was neither. I still don't understand how, though. The main point seemed to be that the skeptic can accept that our experience is either veridical or it isn't: that, for any particular perception, we're either hallucinating, or we're not. (That's the "disjunctivism.") But 1) I don't think the sort of perceptual experience we make claims about is exhausted by that disjunction, and 2) I certainly don't think experience is exhausted by that disjunction. This is the difficulty with Mind & World: he makes it sound as if we're always in the business of making claims about the world that we must justify. And this goes against the Wittgensteinian insight (and to some extent the parallel Austinian insight) that we only need to justify our claims when a question--an ordinary question--of justification arises; and that we only make claims at all for certain purposes and in certain circumstances.
Anyway, it isn't clear what the point of the disjunction is. Unless we say that it doesn't matter which side of it you're on. And that would be similar to what McDowell says (seems to say) in "Values & Secondary Qualities": that, with regard to perception, there is, on one level, no question about whether what I'm perceiving is real or not. After all, I'm really perceiving it. Ultimately, what I would want to accomplish, were I McDowell, would be a "softening" of the facts demanded by the skeptic when he or she demands that we justify our perceptual claims. The degree of softening necessary would depend on the situation: certain sorts of facts are not rightly demanded in certain situations. And one way to accomplish this softening would be to make it clear that we only need to justify a perceptual claim when there's some real question about it's rightness.
Towards the end of the talk, McDowell seemed to be championing ordinary methods of justification (I can "tell a zebra when I see one")...but it wasn't clear whether he meant to challenge the "internal" skeptic (whose skepticism arises and is settled within ordinary notions of justification) or the "external" or "global" skeptic (whose skeptical questions about justification know no bounds). He talked mainly as if he was addressing the "external" skeptic. This, at least, is the way it sounded given his disjunction. The disjunction, "Either I'm seeing a red cube, or I'm having an experience as-of seeing a red cube" assumes that it's responding to an "external" skeptic. (When else would such a statement suggest itself to us?) I think he intends this rather unnatural disjunction to respond to the "external" skeptic by saying that it's only because we have experiences we're willing to call "(veridically) seeing a red cube" that can we make sense of the possibility of "hallucinating a red cube." The response gestures towards particular contexts without actually describing them.
Perhaps that's what produced the impression in me that the upshot of what he was saying was rather like Austin. Granted, Austin (say, in "Other Minds") does describe in more detail ordinary ways of justifying knowledge claims. But to hold those up against the question of the "external" skeptic leaves "external" skepticism unscathed. It's like saying "it's because of this (what we call "telling a zebra") that you can ask your skeptical question." Ordinary methods of justification are a precondition for extraordinary ones. But that doesn't mean the extraordinary questions are always (should always be) in play.
This is a concern for me because, in McDowell's essays on rule-following, he supports the Wittgensteinian line that justifications come to an end--that at some point we reach "bedrock" in a given activity. His criticism of the anti-realist about meaning (Crispin Wright) is that he looks for "bedrock" "lower than it is" in demanding that we characterize our--for lack of better term I'll say "semantic"--agreement in terms that don't demand an insider's view of the language. This criticism seems right to me, albeit overly general. It's this over-generality--also present in the way he addresses the skeptic in the talk on Friday--that I'm trying to put my finger on.
Labels: External World Skepticism, Justification, McDowell, Rule-Following Skepticism
posted by Blakely at
8:07 PM
9 Comments:
Your (1) is a (highly generalized) version of the main point of my post.
However the degree of my disdain for subpersonal facts about perception cannot be overestimated. I'm content to lump them all into "very general facts of nature." But I agree that if they change, the way we talk will change, too.
The question to which my post was meant to point is: When, if ever, do we make disjunctive judgments like this? Only when doing philosophy? If so, then I think McDowell's response to the skeptic is lacking.
This criticism is not perfectly placed yet. But generally, I'm coming to think that just making a transcendental argument to a richer notion of "experience" or "perception" or whatever isn't sufficient if the argument doesn't also place these terms in their normal contexts. Or, what I think may more often happen, replace these terms with groups of terms we'd naturally use to talk about what philosophers might lump under "experience" or "perception."
That was overly dismissive of me.
You seem to be worried about the firmness of McDowell's idea of objectivity: pointing out that maybe if we're hallucinating all the time, the disjunction (and its implication of the possibility of veridicality) may not apply.
That we might be less sure of ourselves in such situations may be true; but the world isn't like that. The possibility that it might be does not, therefore, seem especially threatening.
Or perhaps "objective" would just have a different meaning in such a world. We wouldn't have the notion of degrees of trustworthiness of perceptual claims the way we do now.
I want to agree with you; but "real" skepticism does arise in our normal environment. (Well, what we know as "real" skepticism arises. Your point could be that "external" skepticism would be stable and reasonable, rather than unstable and merely philosophical, if the world were different.)
It is tempting to think of orindary language as a "tacit theory of the world"--but that seems kind of dangerous, too. It gives the activity of constructing theories a kind of primacy it shouldn't have, leading us to call ordinary language a "kind of theory."
Science does not have the task of making our language better. The point I usually make is that in the case that concepts laid out by scientists become part of our ordinary language, then ok. But they don't often do that, nor is it their purpose.
I've been thinking about your examples. I'm not sure I see how the tree root is continuous with the other 4.
The tree root, to me, seems like an uncategorizable experience--one that stands out as uncategorizable; not necessarily one in which we're not using our powers of perception, or our powers of perception are fogged. The disjunction is irrelevant to this experience.
The other 4, however, are more cases of fogged perception than uncategorizable experience. They remind me of examples Professor White uses to get us to wonder whether "external" skepticism couldn't ever be coherent (at least, I think that might have been the purpose). It seems like it wouldn't be coherent here, either--more because of the epistemic limitation that we have to believe something, however fallible we think we are, and whatever our conditions. So maybe the disjunction would fit these cases--; unless you're saying we'd be less likely to entertain beliefs about our perception in them.
I can't get my head completely around this topic, so I am just going to post some scattered thoughts while I take time off from studying for my Mind exam.
I think part of the confusion about what my examples purport to show is due to the ambiguity of "objective experience," which can mean
(1) Experience of objects
or
(2) Veridical experience.
McDowell, insofar as I understand him, is working from the quasi-Kantian tradition where (1) and (2) are actually connected.
In Kant, the structure of the understanding and the intuitions of sense perception work in concert to build concepts of the objective world, which are then structured into judgments, which are what I would consider to be knowledge claims (Did I get the Kant bit right, Blakely/Felipe?)
Now, IMHEI (In-My-Humble-Epistemological-Intuition), it is only when we begin self-consciously to consider the truth of concepts as structured in a proposition that justifcation, including whatever kind of justification we want to employ against the skepticc, begins to be an issue. Perception is, as it were, already there when justification becomes an issue.
Traditionally, the skeptic acknowledges this, but then goes on--through things like the argument from illusion--to show that perception itself is in question and so our inference to knowledge about the external world--as structured in the concepts we employ in our judgments--is completely defeasible.
McDowell's transcendental argument seems to be trying to get rid of this "prop" that a certain kind of skeptic relies on (the idea that all knowledge is inferential). I'm not sure I understand McDowell's argument, but it seems to lean heavily on the idea that (a)experience is perfectly disjunctive and (b) direct percpetion itself, rather judgment, provides us with knowledge of objects.
McDowell seems to have a Sellarsian Kant in mind when he makes his transcendental arguments. I don't know much about Sellars' view of Kant, but I found this online: http://www.ditext.com/sellars/ikte.html.
The point of all of my examples, as well as the link to Noe, was to question whether we need McDowell's argument for the disjunctive character of experience to take away the skeptic's prop. I hope the answer is "no," because I don't think experience is much like that.
Experience is much more fluid; our phenomenology is not as of the world providing direct knowledge (at least prior to our making of judgments). I think that was part of what you were saying in your original post (or least last Friday).
The synthesizing, objectifying, structuring part of the mind--what Kant calls the imagination--works mostly sub-personally. The point of the examples--and in a certain sense, Putnam's original point in his Brain-in-a-vat argument against absolute skepticism--is that we can't get the mind outside of the world. We might imagine us having massively, massively incorrect beliefs, but we cannot put the idea of an objective external world that causes our experiences to rest with any form of skepticism, internal or external.
The other point, which Professor White mentioned in class, is that REAL external skepticism, if it produces no behavioral effects, is meaningless (because meaning is use--the meanning of proposition X is what X causes me to do when I hold it as true).
The counterfactual scenarios show us when or where REAL skepticism--when we REALLY doubt the existence of an "objective world," in the sense I started with--could be MEANINGFUL. We have intuitions about skepticism because we can imagine these places. But, as you said, we do not live in those places, and we cannot even imagine ourselves out of the world (every counterfactual scenario we can imagine has us at least silently taking in the scene).
Back to Chalmers, Jackson, Nagel vs. Dennett and Churchland.
I agree that I don't think we need McDowell's argument about the 'disjunctive' character of experience to remove the 'prop' that all knowledge is inferential.
Your criticism of direct perception as a means to knowledge in this sense is well-placed. We don't just automatically know things just from what we see. (But we can.) So it is likely that McDowell assimilates cases of knowledge to cases in which we know because we just see that things are this or that way. The problem might be that even saying: "we could know things through just seeing how things are OR we could see how things are and turn out to be wrong (if that's the disjunction); therefore we needn't always be skeptical" has questionable force against the skeptic, who needs to see why we shouldn't conflate one way we talk about knowing things with another. I feel like I'm only gesturing in the direction of what needs to be said. Do you by any chance have the pdf of the paper on your computer? (It's been removed from the MIT website.)
On where "REAL" skepticism can be meaningful: I thought White's point was that we could not coherently doubt all of our beliefs at once, and that was the sense in which "external" skepticism is meaningless. (It will produce behavrioral effects: it will lead skeptics to say things like "maybe you're dreaming or maybe you're a brain in a vat.") My hunch is that in those worlds you envisioned, skepticism would be a constant temptation, but would be "internal."
In support of your point, or one of your points: Philosophical Investigations 486:
"Does it follow from the sense-impressions which I get that there is a chair over there?--How can a proposition follow from sense-impressions? Well, does it follow from the propositions which describe the sense-impressions? No.--But don't I infer that a chair is there from sense-impressions? From sense-data?--I make no inference!--and yet somtimes I do. I see a photograph, for example and say 'There must have been a chair over there' or again 'From what I can see here I infer that there must have been a chair over there.' That is an inference, but not one belonging to logic. An inference is a transition to an assertion; and so also to the behavior that corresponds to the assertion. 'I draw the consequences' not only in words, but also in action."
I don't have a PDF, but I do have a hard copy, which you can have.
Some more comments (though I feel awkward following Wittgenstein):
(1) I think it is important, in these conversations, to distinguish between kinds of knowledge. What McDowell proposes, again with the caveat that I may completely misunderstand him, is that we have direct knowledge of states of affairs, events, and objects via direct perception. This is the kind of knowledge that can go into a "that" clause:
(P1) I see that the pencil is yellow (therefore, I know that the pencil is yellow).
(P2) I see that the sky is blue (therefore, I know that the sky is blue).
You don't need skepticism to consider this kind of direct perception problematic. Consider,
(P3) I see that the sun rotates around the earth (therefore, I know that the sun rotates around the earth ).
(P4) I see that God punishes the wicked (therefore, I know that they are wicked).
What ulitmately distinguishes the acts of "seeing" in P1 and P2 from P3 and P4 at the level of phenomenology? We could say that the latter two are more theoretical, which is true, but how far does that get us?
It seems like, if I am going to resist the idea that P3 and P4 are real instances of knowledge of states of affairs or events via direct perception, then I must resist McDowell's idea that P1 and P2 are instances of such (or, I must tell some story about what base level of perception is immune from theoretical over-reach).
Old school empiricists used sense-data to get out of this problem: spatio-temporal relations (indexed to a subject of perception) of colors, lines, massess, shapes, and sounds were the 'foundations' of all perceptual knowledge, even if they were abstracted out of our more basic phenomenology, which is that of perceiving middle-sized dry goods with use and action-values).
So I guess my question to you is, when you say that "we know things just by seeing them," what kind of cases do you have in mind?
(2) Maybe, after we get rid of Mcdowell's idea of direct perception as a kind of access to objective states of affairs that gives us a kind of propositional knowledge, we can invoke another plausible sense of "knowledge" that replaces it. This is would be more like know-how or the possesion of a kind of competence or skill, which is embedded in our acts of perception (Noe also talks about this). Consider:
(P*1) I saw that it was the right moment to throw the bounce pass (therefore, I knew that it was the moment to throw the bounce pass).
(P*2) I saw that it was the right time to break into my piano solo (therefore, I knew that it was time to break into my piano solo).
Maybe these sorts of basic skills can then get elaborated into more robust, "theoretical" kinds of skill.
(P*3) I saw that the baby needed to be adopted and that I was the only person that was in position to adopt it (therefore, I knew that the baby needed to be adopted by me).
(P*4) I saw that "I summon up remembrance of things past" would be the perfect line to follow "When to the sessions of sweet silent thought" (therefore, I knew that . . )
These examples are similar to those in (1), with the exception of the fact that all the knowledge invoked in the examples is deeply value-laden. I don't know what to make of this fact, but I want to resist the idea that the examples only work, if they work at all, because they invoke value-laden descriptions of states of affairs, which are therefore "subjective" and not instances of "real" knowledge. I kind of see it as my whole purpose in doing philosophy to combat this notion.
(3) I think your interpretation of White's argument about the meaningfulness of skepticism is right and better put than mine (basically, the key point is the internal vs. external distinction and putting all your beliefs, a la Descartes, into question). So I guess what my examples purport to show is a situation in which internal skepticism might be justified?
Thoughts from others welcome.
The logical extension of my Wittgenstein quote would be that, in your examples (P1)-(P4), the statements "I see that..." are made for particular reasons, in particular contexts. If they are to make sense, that is, we must imagine them in some context of other. And these will be, in their various ways, contexts of action. This makes them more like your examples (P*1)-(P*4). "Seeing that" God punishes the wicked is not the same kind of thing as "seeing that" it's the right time to do a piano solo.
It isn't clear that we'd be inclined to say, in every case where we have reason to say "I saw that...," we'd also want to talk about knowing (that...). (Maybe we _can_ in all of them, but I don't see why it *follows.*)
Is what constitutes knowing that it's time to start a piano solo the same as what constitutes knowing that a baby needs to be adopted by you? In the cases you give it seems far more likely (as I think is your intention) that the person will just do the thing than that they will ascribe knowledge of its rightness to themselves. But perhaps they can attest to the rightness of the action if asked, or have it ascribed to them by a bystander.
If the situations in which we talk about seeing that... are tied to various contexts (including contexts of action), and those in which we talk about knowing that... likewise, then it seems like the real trouble with wanting to model all knowledge on "seeing that..., therefore knowing that..." is that it overgeneralizes and doesn't capture accurately what happens in real cases.
If McDowell ultimately wants to pursue a virute-ethics model (and we have it on hearsay that he does), he might ought to be sympathetic to your description of knowing as knowing-how rather than knowing-that. (Although he really likes talking about the propositional content of things, so that seems unlikely.) But then I would worry about overgeneralizing in the other direction. (Sometimes we do know that...)
Post a Comment
Subscribe to Post Comments [Atom]
<< Home