What was wrong with propositional content?

Currence, the lucky devil, got to attend another McDowell lecture this weekend. He's written up a short summary post and linked to the handout for the weekend's lectures. Supposedly, McDowell has found cause to revise his views. I have no idea what the cause is, or what is being revised. (Edit: I have received a full version of the paper since writing this post. Hopefully it will clear things up.) A few comments:

Remark 4 on the handout:

(ii) The content of an experience would need to include everything the experience enables its subject to know noninferentially.
4. Against assumption (ii): when an experience enables me to know something noninferentially in a way that depends on my having a recognitional capacity, the recognitional capacity need not enter into the content of the experience. Rather, the experience makes something present to me, which (over and above having it present to me) I recognize as belonging to a certain kind.

Why shouldn't we take facets of the world we recognize to be part of the content of an experience in which we recognize those facets? I'd been taking it that the "content" of an experience was just a variety of suggestions to the effect "that things are thus-and-so", and I don't see why this should exclude things which I need to exercise recognitional capacities to notice.

"The recognitional capacity need not enter into the content of the experience" -- well of course the capacity needn't be a content of the experience! If I recognize that a certain fruit is an apricot, then I should suppose that the content of the relevant experience is something like "That's an apricot", not "That's an apricot, and I have the ability to recognize apricots." But it seems that my capacity to recognize apricots does have to be already active in my experience of an apricot, if I am to experience the apricot as an apricot. The ability to recognize and discriminate between various types of fruit is a prerequisite for my being able to have any experience whose content includes something to the effect of "There's an apricot in front of me." Without my recognitional capacities being active within experience, I don't see how experience doesn't end up as "mute intuitions", bare presences which don't tie into my conceptual web at all.

Similarly, if my recognition that I'm looking at an apricot isn't inferential, then why should we say that my recognizing it is something "over and above having it present to me"? The "having it present" in this case is: there's an apricot. There's not a something-which-is-present-to-me which I then (as on the inferential model) go on to see as an apricot (upon reflecting that this is the sort of way in which apricots appear); what I see is the apricot, and I see it as it is. Being able to recognize apricots without inference lets me simply see apricots, when they are in my vicinity and I have a clear view etc. They are not visible in a way which is "over and above" being present to me; the way they are present to me is: Visibly.

Suffice to say, I don't have the foggiest idea what remark 4/assumption ii are going on about. I'm especially not sure what part of McDowell's older positions is under attack, here. If anything I'm saying in this post is something which McDowell has disavowed, then that's news to me.

Point 7 refers to "subsection 4" -- assuming that this is the '4' from above, point 7 makes a distinction between knowledge which makes use of my recognitional capacities and knowledge which is merely a case of adding the "I think" to some bit of the manifold given to me in intuition. But I don't know why attaching the "I think" to some bit of the intuitive manifold given to me shouldn't be regarded as involving recognitional capacities. I "re-cognize" that some bit of the manifold can be cognized in this way -- can be articulated in this way. (This is an Hegelian point, about the relationship between erkennung and anerkennung -- cognition and recognition. What I can recognize is what is there for me to recognize. And when I cognize rightly, when my judgement agrees with its object, then in my judgement I recognize how things are.)

This is the case even when I pick out a bit of the manifold by using a demonstrative. When I pick out this with a demonstrative (and an ostensive gesture or whatever) I mean this as opposed to that (the various things I'm not picking out, in this instance). I recognize that this is not that, I recognize that my demonstrative gesturing picks out one/some thing(s) and not others, I recognize that what is given to me is capable of being carved up in this way.

One possibility does occur to me: In "Mind and World" McDowell speaks of experience as having "propositional content"; in his post-M&W works, he frequently clarifies how he means this, since the presentation in M&W is a bit dense & hurried. Experience "contains claims", said Sellars; an experience gives a subject suggestions as to how to judge things to be. In veridical experience, when we take it as veridical, we take in how things are; the experience suggests to us that things are thus-and-so, and if we do not discount the experience, then we take it that things are thus-and-so. In this way, the world makes itself known to us in experience.

Now, if one stuck purely to "Mind and World", one might end up with a different view of things: Experience has "propositional content", and this means that what is given is (so to speak) sentence-shaped. "The World" is to "Experience" as a speaker is to his assertions -- experience is made out to be a series of interjections by The World that This. Is. So. and we subjects merely trust The World or don't, as we see fit. This is pretty clearly an insufficiently nuanced sort of view; experience isn't entirely like listening to a speaker making assertions, though both can be informative. Getting clearer on the difference by making a distinction between the way in which experiences have "content" and the way in which sentences have "content" seems to be a decent enough route to take. But then I have no idea what position McDowell is saying Travis has lead him to revise; a great deal of his post-M&W works has been devoted to clarifying how he sees the role of intuitions, but I've never gotten the impression that his views were actually changing; he was merely revising his presentation, making it clearer how he meant for things to be taken.

As a final WTF-type response to the notion that McDowell is revising his older views, here: Section 9 on the handout says that an advantage of McDowell's (new) view of experience as having intuitional content (as opposed to propositional content) is that it lets us steer a middle-route between Davidson's view of experience and the Myth of the Given. Surely that achievement was supposed to have been already achieved with "Mind and World." If the distinction between intuitional content and propositional content is supposed to be addressing a new problem, then it can't be that one; if it's supposed to be addressing an old problem, then this would indicate that McDowell is unsatisfied with his older views. But if he is unsatisfied with some of his previously-held opinions, then I don't know where or when he held them, or what they were.