Who and What Do Who and What Range Over Cross-Linguistically?

Dayal’s (1996) account of the presuppositions of wh-questions makes faulty predictions for languages which draw number distinctions in the domain of simplex wh-expressions: Dayal predicts that a singular wh-expression should always give rise to a Uniqueness Presupposition; the Anti-Singleton Inference associated with its plural counterpart is expected to be parasitic on the uniqueness presupposition. We provide new data from Spanish and Hungarian, where simplex wh-expressions inflect for number. We claim that singular simplex wh-expressions do not give rise to a Uniqueness Presupposition, but plural simplex wh-expressions nonetheless give rise to an Anti-Singleton Inference. We provide an analysis of these facts that is consistent with Dayal’s account of constituent questions, by assigning simplex wh-expressions a type-flexible denotation.