I think the exchanges between Zena and me, which Janine has also joined recently, have become somewhat tangled.  Looking back over my previous handout, and thinking about Zena and Janine’s recent responses, I have come to think that there may be a simpler way of blocking Boghossian’s argument, which may also have the advantage of getting pretty quickly to the heart of the matter.  So I’m going to give it a fresh try now.  (I am hoping that, at least indirectly, what I say will be responsive to Zena's and Janine's responses.)

    Let me start by quoting something I say toward the end of a previous handout (entitled “On Boghossian’s Argument Re What Externalists Know Apriori”): (1) “Accepting the DBH line means accepting that my (“apriori”) knowledge of externalism plus my “apriori” knowledge of my thought’s content does not after all imply that I can have apriori knowledge of the existence of what fixes my thought’s content.”  I then added: (2) “Even as I know what I am thinking – say, that water is wet, and even as I realize that my thinking that thought requires the existence of water around, I am not thereby in a position to ‘deduce’ that water exists in my environs.” (Emphasis added now.)  Regarding the latter claim, Zena might rightly press: Why is it that knowing apriori that I am thinking a WATER-thought, and knowing apriori that having such a thought entails the existence of water (through knowing apriori the truth of externalism) doesn’t allow me to know apriori that I am indeed in a water environment?
    This makes me think now that moving as I did from (1), which I still think is right, to (2), was incautious.  Doing so may indeed play directly into Boghossian’s hands, in that it involves the same illicit move made by Boghossian.  Let me explain.

    The illicit move involves casting the whole argument (and the debate over it) in terms of what externalism says (or would say) about particular thought-contents.  I have conceded that we may know the truth of externalism apriori, in Boghossian’s loose sense of the notion.  But what I should not have conceded, and what is required for Boghossian’s argument to go through, is the claim that we can know apriori what satisfies the externalist’s condition in particular cases.  For Boghossian’s argument to go through, he needs to have it that I can know apriori that If I am thinking that water is wet, then water exists in my environment (claim A in the ‘base argument’ displayed in my previous handout).  But A, I submit, is not something I can know apriori, even in Bog’s loosest sense.  I was going along with Boghossian and Bill ’s idea that A is entailed by externalism.  But it should be clear that externalism alone cannot entail anything as specific as A.  The metaphysical thesis of externalism says something like this:

Many of our concepts exhibit dependence on natural and/or social environment of the thinker.  Which concept a thinker has or uses depends (in part) on certain facts about her natural/social environment.

     Now, the DBH line comes in to point out that such an investigation is not necessary for knowledge of one’s own present thoughts.  Understanding the DBH line should allow us to see that the aposteriori knowledge that is necessary to ground claims about the particular externalist conditions necessary for possession of particular concepts (such as WATER) is not a necessary prerequisite for self-ascriptions of thoughts with particular contents.  This is because, in the normal case, ascription to oneself of a present thought does not require ascertaining that one has this or that concept – it does not require determining that one has met the necessary conditions for possessing the concept.  It only requires redeploying the concept in the self-ascription of the thought.  (See handout on externalism and self-knowledge – this is the idea that the content of the self-ascription, e.g., “I am thinking that water is wet” ‘locks onto’ the content of the first order thought “Water is wet”.)  So the DBH line assures us that the truth of externalism is consistent with the possibility of apriori knowledge of premise B in the ‘base (A-B-C) argument’<1>.

     To sum up: Although B is knowable apriori, and B does concern knowledge of particular thought contents, this fact cannot be fed into a reductio argument that aims to show that, by externalism, we should have apriori knowledge of the existence of particular substances.  This is because such a reductio requires presuming that we can know apriori which conditions must obtain for the exercise of particular concepts (specifically, the ones deployed in the self-ascription displayed in step B).  Yet, even if we grant that theses like externalism are establishable by apriori philosophical reflection, rather than (in part) by an investigation of the world, it should not be granted that we can establish without such an investigation the specific ‘possession conditions’ that would be appropriate for particular concepts, given externalism.  Quite the contrary; given what externalism maintains about the individuation of concepts, it would seem that knowing such conditions requires empirical knowledge of the world (which we do have in the case of water, but lack in many other cases).  But knowing A requires precisely knowing the condition for having WATER-thoughts.

     A final comment:  in the preceding paragraph, I was taking (A)’s consequent to involve reference to the existence of the substance water (i.e., H2O, as opposed to XYZ).  Boghossian might concede that knowledge of the actual ‘possession conditions’ for WATER does require empirical investigation, but insist that there is some less ‘robust’ condition for WATER which we can derive apriori from the truth of externalism.  I’m not sure how to ‘thin out’ the condition.<2>   (Something like: “If I am thinking Water is wet, then there must be a liquid with the right underlying structure in my environment”?) But let me say this: to preserve the validity of the A-B-C argument, whatever thin sense of “water” is used in (A) would have to be used throughout the argument.  And if we can find such a sense, the idea that we know apriori of the existence of water in that thin sense (given externalism) may not seem so alarming.<3>   (That is, on the (dubious) assumption that the truth of externalism requires no empirical knowledge of the world, it doesn’t seem implausible that we can know apriori that there is an external world with various substances/things whose nature constrains our thoughts.  But that is a far cry from claiming that externalism is committed to the possibility of apriori knowledge of the existence of water (as opposed to twater) in our environment (or else to the claim that we don’t know in a non-inferential way what we’re presently thinking).

     I think I’ll stop now.
 

1  That argument, recall, was:

(A) If I am thinking that water is wet, then water exists in my environment.
(B) I am thinking that water is wet.
(C) Water exists in my environment.

3  If I understood her correctly, I think Janine was proposing some such thin sense for the apriori knowability of B.  But I think externalists want to salvage a stronger sense of privileged self-knowledge.  What I know when I know that I am thinking that water is wet is not merely that I am thinking that some liquid with such&such features is wet.  I know that I am thinking that water is wet.  (Never mind if this makes no sense.)