Questions and Answers in Montague Grammar

This paper is built on Bennett (1979) (henceforth QMG). QMG was drafted by Bennett to be a joint paper by the two of us; we held a number of discussions on the issues raised therein, but none of the changes on which we agreed or the differences which we wished to mention were committed to paper before his untimely death. It is my ultimate plan to bring out a jointly authored version of QMG containing both agreements and differences. The central portion of this paper, Section 3, on quantification into questions, represents an area of agreement: Section 3 is motivation of and minor alteration of some of the constructions suggested in QMG. The rest of the paper divides up as follows. Section 1 is a brief discussion of the role of Montague grammar (Montague, 1974) in a theory or system of natural language processing (by man or machine). Section 2 discusses various aspects of the concept of “answer.” Section 3, as mentioned above, builds on QMG to give a theory of quantifiers whose scope is wide with respect to question-words. Section 4 discusses one question word, which, in the singular, indicating how QMG could be altered to agree with the suggestions made in Section 3. Section 5 defends the view that different question-words should be assumed to behave differently — it is argued that the contrary assumption is false.