Against wide-scope free choice

Abstract

Sentences of the form "x may A or x may B" have a reading that implies "x may A" and "x may B". This has led many to conclude that there is a specific problem of wide-scope free choice, namely, the problem of explaining how a disjunction of possibilities ◇A ∨ ◇B can receive a reading that implies its disjuncts. We argue that this conclusion is mistaken: once we consider a broader range of data, it becomes plausible that the relevant reading in fact results from an LF of the form ◇(A ∨ B). This raises the question of how this LF arises compositionally. We propose a solution based on the theory of modal concord of Zeijlstra (2007), which avoids the problems of previous movement-based approaches and explains an interesting contrast observed by Meyer & Sauerland (2017).

EARLY ACCESS