Non-declarative sentences and the theory of descriptions

Abstract

This paper shows that Russell’s theory of descriptions gives the wrong semantics for definite descriptions occurring in questions and imperatives. Depending on how that theory is applied, it either assigns nonsense to perfectly meaningful questions and assertions or it assigns meanings that diverge from the actual semantics of such sentences, even after all pragmatic
and contextual variables are allowed for. Given that Russell’s theory is wrong for questions and assertions, it must be wrong for assertoric statements; for the semantics of ‘the phi’ obviously doesn’t vary depending on whether it occurs in a question or an assertion or a command.