In response to Dorit

 Let me begin with a quote from Dorit:

"So: if you accept the DBH line, I can know ("apriori") that I am thinking that water is wet, without thereby knowing (let alone knowing apriori) that I am in a water environment.  This is so, even though I cannot have a water-thought without being in a water environment."
I think that Dorit plays right into Boghossian's hands, rather than avoiding his clutches. Here's why....

A challenger to externalism once said, "Look, isn't there an epistemological problem? Otto doesn't know whether he's thinking that water is wet, or whether he's thinking that twater is wet."

The DBH line is to say: Otto does know whether he's thinking that water is wet, rather than thinking that twater is wet. He knows simply by having the second-order thought 'I am thinking that water is wet'. He doesn't have to have the thought 'I am thinking that water, rather than twater, is wet' - but his second-order thought serves to rule out the skeptical possibility that, for all he knows, he's thinking about twater. Otto doesn't have to go out and investigate the environment in order to find out whether he's thinking about water. So, as Dorit says,

"if you accept the DBH line, I can know ("apriori") that I am thinking that water is wet"
Let's grant this much to the externalist. But now the problems really start. If Otto can know that he's thinking about water, rather than twater, then it seems that if Otto can run through externalist  thought-experiments, he can know a priori that there is water in his environment. Otto can argue as follows:
(1) I know I'm thinking about water.
(2) In order to be thinking about water, I know (by going through twin-earth thought-experiments) that I am in a watery    environment.
Therefore, I am in a watery environment.
That is, we have arrived at Boghossian's argument, because premises (1) and (2) are known by Otto a priori, so the conclusion is known by Otto a priori. So the next sentence from Dorit is exactly what Boghossian is challenging:
 "without thereby knowing (let alone knowing apriori) that I am in a water environment."
What the DBH route allows you to avoid is that Otto has to find out a posteriori that he's in a water environment before he can know that he's having water-thoughts. It doesn't allow you to rule out that you know a priori that you are in a water environment. In fact, it's taking the DBH route in the first place that leads to the problems Boghossian is talking about. (This is what I was trying to explain right after the break in the class when I presented.)

Suppose the challenger to externalism grants that Dorit is right that Otto doesn't know that he is in a watery environment - after all, perhaps Otto never reasons through the twin-earth thought-experiments. (Strictly speaking, though, this is not relevant. What matters is what Otto can know.) There is still a problem with the next thing Dorit says:

"This is so, even though I cannot have a water-thought without being in a water environment."
How does Dorit know this fact? She knows it through a priori twin-earth thought-experiments. So she seems to have reached, by a priori reasoning, the conclusion that she must be in a water environment! She has given a metaphysical argument about the conditions that must hold if someone is to have a water-thought. Right - but if she reaches a metaphysical conclusion that in order to have water-thoughts there must be water in her environment, then we can make an epistemological claim about her: That she can know that she is in a watery environment, given that she knows she is having water-thoughts!

Sorry this was so long. Thanks for reading, if you got all the way down to here,

                                                                                                                                    Zena.

PS  Incidentally, my view is that we should reject the DBH line. It's too good (and easy) to be true. If externalism is true, there is a skeptical problem - Otto doesn't know that he's having water-thoughts rather than twater-thoughts, once that possibility has been raised. (I'm sympathetic to David Lewis' contextual views on knowledge.) Ordinarily, Otto does know he's having water-thoughts, because the possibility of his being in a twatery environment has not been raised. So his being in a twatery environment is not a relevant alternative situation - just as, ordinarily, the possibility of you being a brain in a vat is not a relevant alternative situation.