Reviewed by: Movement and silence Andrew Carnie Movement and silence. By Richard S. Kayne. Oxford: Oxford University Press, 2005. Pp. xv, 376. ISBN 019517917X. $39.95. In the early to mid-1990s, as the minimalist program (MP) began to bloom, comparative syntax came into its own as a tool for investigating and explaining typological variation within [End Page 411] a universalist parametric approach to language. In particular, the investigation of microparametric variation tied to specific lexical items or even features became a central methodology within the framework. It would be hard to find anyone more influential in such studies than Richard Kayne. Movement and silence is a collection of K’s papers about parametric variation written between 2000 and 2005. All but one of these papers have appeared elsewhere, but some are in more difficult-to-find venues, so this volume is a welcome and accessible collocation of these works. This is K’s third such collection (Kayne 1984, 2000 are older collections of articles); this one stands out, however, as the results reported in each paper are intricately tied to one another. With the exception of two papers, which are independently important, the book reads more like a tightly argued monograph than a mere collection of papers. Running through the papers are two interrelated hypotheses, hinted at cleverly in the title of the book. First, we have the consequences of K’s now-famous antisymmetry hypothesis. In particular, the book focuses on the movement operations (often massive and very abstract) that are necessary to account for word-order variation in the face of the claim that the underlying order of constituents is universally specifier-head-complement (S-H-C). Other orders require movement, often of remnant constituents, to the specifiers of a variety of functional (and occasionally lexical) projections. Because of the underlying assumptions of MP, variation among languages must be tied to the availability of these projections. This gives rise to the second hypothesis at work in the papers (but made most explicitly in the appendix to Ch. 8 and in Ch. 12), the principle of decompositionality: universal grammar imposes a maximum of one interpretable syntactic feature per lexical item. A consequence of this principle is that the range and number of categories is far richer than is typically posited. This result is compatible both with the Italian cartographic project (see, for example, Rizzi 1997, 2004, Cinque 1999, and Belletti 2004) and with the view that K and others (e.g. Koopman & Szabolcsi 2000) adopt where surface strings are derived through massive remnant movement. The quantity and extent of the movement posited will surprise most syntacticians. Take, for example, the derivation of DP-internal prepositional phrases. Consider the derivation of the phrase admiring a picture of John given in Ch. 71 (138). 1. admiring [John a picture] → merger of K-of K-of admiring [John a picture] → movement of John to spec, K-of Johni K-of admiring [ti a picture] → merger of of of Johni K-of admiring [ti a picture] → movement of VP to spec, of [admiring [ti a picture]]j of Johni K-of tj . Since everything moves out of its base position and results in a structure that seems to violate our basic understanding of constituency (note that a picture of John is not a constituent in this structure), it is not surprising that such derivations will be largely met with skepticism from researchers outside the antisymmetric paradigm. However, K (1998) convincingly argued that the constituency seen in the last line of 1 accounts for preposition stranding in English. One finds that similarly complex derivations are necessary in K’s system to account for basic word order in Japanese (Ch. 92), stylistic inversion in Romance (Ch. 13), ‘even/if’ clauses in Irish (Ch. 24), Heavy NP shift in English (Ch. 115), causatives in French (Ch. 56), and a wide variety of other cases. [End Page 412] A similar level of abstractness is found in the other major theme in the book: the role of silent words in the syntax. K argues for a variety of null words in English, which at...