Paper on unspecific antecedents
A new paper (draft) on counterfactuals with unspecific antecedents, to appear in a festschrift for Al Hájek. The paper discusses a range of phenomena related to the "Simplification of Disjunctive Antecedents". I argue that they can't be explained by a chance-based account of counterfactuals, as Hájek has suggested. Instead, I hint at an RSA-type explanation. I also suggest that this explanation might somewhat weaken the case for counterfactual skepticism.
I regret how much time I have spent on this topic. I first noticed it in 2006, and thought I had a nice explanation. When I posted it on the blog, Kai von Fintel kindly pointed me towards some literature. A little later, Paolo Santorio suggested that my explanation resembles the one in Klinedinst (2007). This seemed right, but I had in mind a more pragmatic implementation. I eventually wrote up my proposal in Schwarz (2021). Although my original interest was sparked by conditionals, that paper focuses on possibility modals, and only briefly mentions how the account might be extended to conditionals. When I got the invitation to write something for Al's festschrift, I thought I could spell out the application to conditionals, and compare it with Al's account. But I couldn't really make it work. So I ended up defending a more orthodox derivation based on Kratzer and Shimoyama (2002).
A minor question: On page 6, you writes, "For a more controlled example, imagine I’m about to draw a ball from an urn. You bet that I’ll draw a red ball. The ball I draw is green. In this context, (8a) appears to imply (8b). (8) a. If I had drawn a red ball or a yellow ball, you would have won. b. ↝ If I had drawn a yellow ball, you would have won. This apparent entailment remains intact..."
But how does this entailment remain intact, given the fact that you bet that I will draw a red ball? If that is your bet, then drawing a yellow ball wouldn't make you win.
It could be argued that 8b is vacuously true if the antecedent is false in all possible scenarios. But it does not seem the case that the antecedent is necessarily false. It must be false in view of the actual content in the urn, but it could be true physically or metaphysically.