UC Santa Cruz

In this paper, I investigate the syntax of a reduced why-question that I call Why-VP; these are clauses like "Why take Structure of Japanese?" These questions are reduced in a very particular way; they are missing an overt subject and the verb never shows tense marking as is typical in why-questions, such as "why was John taking Structure of Japanese?" Why-VP has received very little attention in the literature even though they are incredibly frequent in production. Over the course of the last year, several naturally occurring examples drawn together with their discourse context have accumulated, specifically from the New York Times portion of the Gigaword corpus. These were found completely by accident when, as part of the Santa Cruz Ellipsis Group, we began annotating root sluices. Despite the fact that they are reduced, and that they were accidentally captured by a parser as ellipsis, the corpus of naturally occurring examples lead me to conclude that these really do not fit the standard profile of ellipsis as they really do not rely on an antecedent. As such, I argue that Why-VP are free-standing and intact clauses, best understood as infinitival and as having a covert modal, thus contributing to our understanding of the clause-building mechanisms available in the language which make reduced clauses such as Why-VP possible. I present evidence that 'why' in Why-VP is best analyzed as a head, which selects for the covert modal hosted by a silent infinitival T, which hosts the structural subject and also selects the v/VP; so while Why-VP questions seem reduced, they actually carry elements typical of the extended projection