The Meaning Group will meet online on Friday, October 16, 3-4pm. Our own Ahmad Jabbar will present his work on
Knowledge-wh and Intermediate Exhaustivity
In this paper, I consider a puzzle about knowledge-wh ascriptions. While (i) intermediate exhaustive (IE) readings for third person knowledge-wh ascriptions exist (Cremers and Chemla (2016)), (ii) they don’t for first person ones (as first noted by Groenendijk and Stokhof (1984)). (i) requires for a theory of question embedding to be able to derive IE readings, and (ii) requires this derivation to be somehow blocked for first person ascriptions. In this paper, I argue against the approaches in the literature, and some possible ones, to solve this puzzle. I conclude that (i) and (ii) cannot be explained by invoking an ambiguity for ‘know’ as in Theiler et al. (2018). Finally, I propose a solution to the puzzle: a relativist semantics for ‘know’ using MacFarlane’s assessment-sensitive framework (2014). [[know]] is construed as a function that takes, inter alia, an information state provided by the context of assessment. The variation in (i) and (ii) is then attributed to the variation in the information state of the agent of the context of assessment. This solution is promising in that not only does this semantics explain the puzzle, it handles much more intricate knowledge-wh ascriptions data. I round off my discussion by considering some objections against my proposal, and considering the possibility of tweaking my semantics to make it trivalent.