Brett Sherman: Open Questions & Epistemic Necessity
Wednesday, February 24, 12-1 pm, Gambrell Hall 006
If I am standing outside in the rain and I am asked over the phone what the weather is like, it would be odd for me to say, "It must be raining." Using epistemic 'must' here seems to suggest that I am in a weaker epistemic position than I would be if I had simply said, "It is raining." This presents a puzzle for standard approaches to modal semantics, according to which modal claims---claims about what could or must be the case---are analyzed in terms of existential or universal quantification over a set of possible worlds. A proposition is predicted to be epistemically necessary on the standard approach just in case it is true in every world consistent with what is known. And when I am standing outside in the rain, every possible world consistent with what I know is one in which it is raining. Yet, it is inappropriate to utter "It must be raining" in such a situation.
One response to this problem is to build into the semantics of epistemic 'must' an evidential restriction that rules out direct sources of evidence. Another response is to give a pragmatic explanation for the infelicity. I argue that we can explain the data semantically, without positing any arbitrary evidential restrictions, by embracing an alternative conception of modality, one that does not identify possibilities with possible worlds. The alternative conception is grounded in the idea that there is a conceptual connection between possibilities and open questions: to be a possibility is to be an answer to an open question.