Coordinating questions: The scope puzzle

This paper introduces a new puzzle concerning the interaction between questions on the one hand, and conjunction and disjunction on the other. It shows that a conjunction of two polar interrogative clauses is interpreted so that each conjunct involves a polar question operator and the conjunction takes scope over these, whereas a disjunction of two polar interrogative clauses can only be interpreted as involving a single polar question operator scoping over the disjunction. In other words, two full-fledged polar questions each including their own question operator can be conjoined, but cannot be disjoined. We argue that the source of this contrast is semantic (rather than syntactic, pragmatic, or other), and we formulate two general constraints on question meanings which can each account for it. The first, based on Fox (2018), requires that the resolutions of a question are related in a particular way to the cells of the partition that the question induces on the context set. The second requires that the exhaustive interpretation of a consistent resolution of the question is never inconsistent. We leave open which of these two constraints is to be preferred.