In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • The grammar of Q: Q-particles,wh-movement, and pied-piping
  • Sandra Chung
The grammar of Q: Q-particles, wh-movement, and pied-piping. By Seth Cable. (Oxford studies in comparative syntax.) Oxford: Oxford University Press, 2010. Pp. xiv, 249. ISBN 9780195392272. $49.95.

Why do constituent questions in many languages of the world begin with an interrogative word? The standard generative answer is that interrogative words have a special feature that causes them to move to the left edge. The thesis of Seth Cable’s carefully constructed book is that this piece of received wisdom is incorrect. Instead, the syntax of constituent questions revolves around the properties of a novel category, Q, which is distinct from the complementizer or force head that causes the sentence to be interpreted as a question. In C’s system, Q bears the special feature usually associated with interrogative words. Q can take the interrogative word, or a phrase containing the interrogative word, as its complement; Q’s maximal projection, QP, is the category that undergoes movement to the left periphery. Evidence for this proposal comes from Tlingit, an endangered Na-Dene language in which Q is realized overtly. The first third of the book develops an analysis of Q and QP-movement in Tlingit, based on corpus data and data from interviews with native speakers. The rest of the book generalizes the analysis to all languages with (apparent) wh-movement of interrogative words, notably including English. The proposal is that in all such languages, what has actually moved leftward is QP—a constituent that includes the interrogative word. One significant theoretical consequence of the proposal is that it promises to eliminate the vexed phenomenon of pied-piping, by reanalyzing all cases of pied-piping as QP-movement when Q’s complement is a phrase properly containing the interrogative word.

Ch. 1, ‘Introduction’, briefly summarizes the generative view of wh-movement in constituent questions, introduces Tlingit facts that point to a new approach, and sketches the broader consequences of this approach.

Ch. 2, ‘Wh-fronting and Q-movement in Tlingit’, is devoted to the syntax and semantics of constituent questions in Tlingit. These questions have an interrogative word at their left periphery—a multipurpose indefinite that can also occur under the scope of negation. The interrogative word, or some phrase containing it, is followed by the particle , which has the same distribution as particles in Sinhala and Japanese that have been analyzed as Q-particles. C proposes that is a realization of Q that takes the interrogative word, or a phrase containing the interrogative word, as complement. He further proposes that what moves to the left periphery in Tlingit is not just the interrogative word, but QP.

Importantly, when the interrogative word is a determiner, a possessor, or the complement of a postposition, cannot immediately follow it, but instead occurs right after the larger DP, possessive DP, or PP. The generalization, C argues, is that QP cannot intervene hierarchically between a functional head and its complement. This QP-intervention condition is explained by assuming that functional heads always c-select their complements, which do not include QP. (In contrast, lexical heads s-select their complements.) The chapter concludes with a semantics for Tlingit constituent questions that appeals to focus and employs two of the interpretive strategies that have been proposed for indefinites. In this semantics, Tlingit wh-indefinites denote choice functions; denotes a variable over choice functions that combines with the focus-semantic value, [End Page 188] but not the normal-semantic value, of its sister. The variable introduced by is ultimately bound by the question operator introduced by the force head at the top of the structure.

Ch. 3, ‘Applications to wh-in-situ languages’, extends this analysis to constituent questions in Sinhala, Japanese, and Korean. Sinhala shows the same patterning of Q, wh-indefinites, and QP-intervention as Tlingit, so it is given the same analysis. Japanese and Korean differ from Tlingit in permitting Q to occur at the right edge of a constituent question, and in not showing QP-intervention effects. These differences argue that in Japanese and Korean, Q does not take...

pdf

Share