This is a follow up, of sorts, to the earlier post on a puzzle about epistemic modals. Like that post it owes a lot to conversations with Andy Egan and John Hawthorne, in this case the proximate cause being a phone chat with Andy last night. (In other words, don’t give me credit for much of this, even though it’s on my site!)

First puzzle. Intuitively, a might be F is logically weaker than a is F. So let’s assume we have a case where the first is true and the second false. In such a case I might have the true belief. For example, right now I believe that it might be raining in Dublin, and that it might be not raining in Dublin, and both of those beliefs are true (at least according to a plausible picture of how epistemic modals might work). Since God is omniscient, whatever true beliefs I have you’d expect He would have too, and indeed they would constitute knowledge for Him. But if a is not F, and hence He knows that it is, then He can’t believe that a might be F.

Extra half puzzle. Intuitively I can believe a might be F when a is not F, and I think intuition says that can be a true belief. But can I ever know that a might be Fwhen a is in fact not F? On orthodox analyses of epistemic modals, you would think so, but in fact this is rather counterintuitive. And I think it doesn’t get any better when you take the stress off know.

Second puzzle. I’m chasing Batman and Robin around a factory. Robin has been keeping count and noticed that I fired all six bullets in my gun. But Batman lost count. Finally I corner Batman and have him at gunpoint. After going through the usual rituals (“I know what you’re thinking…”, etc) I convince Batman to surrender. Seeing Batman surrender robs Robin of what little confidence he had left, so he takes off. I think in that case (1) and (2) are true, but (3) is not.

(1) Batman surrendered because I might have had a bullet left.
(2) Robin fled because Batman surrendered.
(3) Robin fled because I might have had a bullet left.

Is this a counterexample to the transitivity of ‘because’, or is there an equivocation involved?

There may be some more ‘big news’ posts about epistemic modals soon, so if you’re interested, watch this space.