Opiniatrety

If I only considered my own view of the arguments, I would give a credence of 1 to the Elga view

I should point out that this isn't actually true of me; I think that this view surely isn't right, but I don't have much of an argument why not.

“Suppose p is such a view; so long as you don't believe p, the arguments for p are compelling, but if p is true then belief in p is unacceptable.”

I wonder whether there ever could be such a p---that is, a p that meets both of the following conditions:

1) So long as you don’t believe p, the arguments for p are compelling.

2) If p is true, then belief in p is unacceptable.*

It seems to me that any p that meets condition (2) will thereby not meet condition (1). Why? Since (S) is true, if (2) is true, then no one should accept p. But if no one should accept p, then the arguments for p cannot be compelling. Thus, if (2) is true, then (1) is false.

OK, so my argument assumes what you’re trying to disprove---namely, (S). But doesn’t this show that you’ve begged the question by simply stipulating that p meets conditions (1) and (2)?

* I should add that I think you’ve slightly misrepresented (S) in condition (2). (S) talks about what is acceptable if (S) is accepted (as true) not if (S) is true. But I think our arguments go through just the same.

Hi Dustin,
First of all, I think you're right about condition (2); I slide back and forth a little there.

As for the main point, what I have in mind is a situation like the following:

There is what looks like a good argument for p.
Accepting p makes it irrational to accept p.
But that's the ONLY reason I have not to accept that argument.

This seems at least possible; at least, for limited creatures such as ourselves who may not have found the flaw in the argument yet.

If so, suppose we reject p on grounds of S. What should we take to be the flaw in the argument for p? It can't be whatever makes it irrational to accept p once I've accepted p; since I don't accept p, this doesn't come into play. It seems that it has to be the self-undermining nature of the conclusion itself. But this isn't a very satisfying explanation of the flaw in the original argument. It seems to me that the argument that p still stands as an attractive nuisance until it itself is undermined.

Things get even worse if not-p has the same problem, as I believe it does in this case. After all, it's not a priori that CEF undermines itself (it depends on the existence of CEF dissenters); and for some people it is (a posteriori_ true that when they evaluate the arguments about disagreement for themselves, as anti-CEF dictates, they find CEF extremely convincing. What are they to do? If they take the anti-CEF position, it dictates that they should evaluate the argument for themselves, which would mean accepting CEF; and there seems to be no more reason to disregard the contingent fact that they find the CEF arguments convincing than that others do not. [Perhaps this means they should suspend judgment.] The situation is somewhat like Andy Egan's counterexamples to causal decision theory, where B is better given that you choose A and A is better given that you choose B.

Now, the logical conclusion to draw may be that there's some yet undiscovered reason you shouldn't be convinced by the argument, and that is in fact what I think about the CEF position. (Although here my ignorance of the reason is probably due to not knowing the issue well enough, so I'm not even a peer.) Still, the S argument doesn't seem to me to provide a reason in itself to reject p; it may indicate the existence of a reason and leave us with more work to determine what that reason is.

Hi Matt,

"[the S argument] may indicate the existence of a reason [to reject p] and leave us with more work to determine what that reason is."

But if we have a reason to believe that there is a reason to reject p, don't we thereby have a reason to reject p?

Good question. It's my understanding that something similar is at the heart of the CEF arguments: If we have evidence that there is undefeated evidence for q, don't we thereby have evidence for q? If so, then the acceptance of q by our peer gives us evidence for q, since we have reason to believe that the peer has evidence for q, and we aren't aware of any potential defeaters that the peer isn't also aware of.

You'll note that I haven't answered the question. The situation may be that of the mathematician who's provided an existence proof but would like to actually display the witness; then the mathematician has more work to do, but we (if we aren't intuitionists) wouldn't say that she lacks reason to believe the existence claim.

Still, the person I've described seems to be in a worse predicament than the mathematician. She finds the argument for p convincing even though she has reason to think (so we assume) that there is a flaw in it. This may be more like a mathematician who has an existence proof but has checked all the possible witnesses and found them wanting. Then she knows that she's made a mistake somewhere; but it's not obvious what the proper response is.

[Also, I don't think the last paragraph of my previous comment necessarily follows from the previous paragraph, so the CEF defender might want to disclaim the whole thing.]

"I wouldn't belong to any club that would have me." -- G. Marx, Dialectical Materialism for Dummies

But how do you know that the other person is as smart and well-informed as you? If they believe not-p, aren't they dummies?

Sometimes clearly not -- if they're in a better position to know, you should adjust your arguments to them; if you're in a better position, you should ignore their opinion. But we're thinking of cases where they have the same credentials, you know them well enough to know that they're not generally batty, and (most important) you're both looking at the same evidence. So it seems presumptuous to say that they're dummies just because you disagree on this question.