Reviewed by: Constraints on displacement: A phase-based approach by Gereon Müller Hisatsugu Kitahara Constraints on displacement: A phase-based approach. By Gereon Müller. (Language faculty and beyond 7.) Amsterdam: John Benjamins, 2011. Pp. x, 339. ISBN 9789027208248. $158 (Hb). The central issue Gereon Müller addresses in this monograph is the existence of locality constraints on displacement in human language. Beginning with a careful evaluation of locality constraints based on notions such as simplicity, generality, efficiency, nonredundancy, and minimization of search space, M argues that every major locality constraint proposed in recent decades is incompatible with core minimalist assumptions. Such locality constraints include the (generalized) minimal link condition (where structure-building features are enclosed in bullets, as in [•F•]) and the condition of extraction domain (where proper-government is replaced by complement). 1. (Generalized) minimal link condition ((G)MLC): In a structure α[•F•]… [ … β[F] … γ[F] … ] …, movement to [•F•] can only affect the category bearing the [F] feature that is closer to [•F•]. 2. Condition of extraction domain (CED) a. Movement must not cross a barrier. b. An XP is a barrier iff it is not a complement. Thus, a new approach to the effects of the (G)MLC and CED is called for, and this monograph attempts to see to what extent such effects can be derived from more basic principles within a minimalist framework. The monograph begins with a brief introduction (1–8), which outlines key assumptions and new proposals, and it consists of the following seven chapters. Ch. 1, ‘Locality constraints’ (9–66), and Ch. 2, ‘(G)MLC and CED in minimalist syntax’ (67–118), provide an overview of the development of the (G)MLC and the CED, and argue that these two highly general, widely accepted constraints (as well as other existing minimalist accounts for CED effects) are not just empirically problematic but also conceptually questionable, as they are evaluated in terms of notions such as simplicity, generality, efficiency, nonredundancy, and minimization of search space. They are shown to be incompatible with such core minimalist assumptions in a strictly derivational model of syntax. Ch. 3, ‘On deriving (G)MLC effects from the PIC’ (119–64), and Ch. 4, ‘On deriving CED effects from the PIC’ (165–238), argue that the effects of the (G)MLC and CED are derivable from the phase impenetrability condition (introduced in Chomsky 2000) and the edge feature condition (revised from Chomsky 2000, 2001). 3. Phase impenetrability condition (PIC): The domain of a head X of a phase XP is not accessible to operations outside XP; only X and its edge are accessible to such operations. [End Page 655] 4. Edge feature condition (EFC; revised): The head X of phase XP may be assigned an edge feature before the phase XP is otherwise complete, but only if there is no other way to produce a balanced phase. These two principles are supplemented by the following four assumptions: (i) all phrases are phases, (ii) all syntactic operations are driven by features of lexical items, (iii) operation-inducing features are hierarchically ordered on lexical items (where only features on the top are accessible), and (iv) edge features can be assigned only if the phase head is active (meaning that it bears at least one feature to discharge). Given these assumptions, the PIC demands that successivecyclic movement takes place in a radically local manner (via every phrase edge); and under the EFC, the assignment of an edge feature (required for intermediate movement) is limited to an active head (bearing at least one feature to discharge), and such feature assignment takes place only when there is no other way to produce a balanced phase. Following Heck and Müller (2000, 2003), M takes a phase to be balanced if, for every structure-building feature in the numeration, there is a matching feature that is either part of the workspace of the derivation, or at the edge of the current phase. Given this much, the effects of the (G)MLC and CED are shown to follow from the PIC. First consider the effects of the (G)MLC. Suppose that two items are competing for movement, and one is higher (meaning it is merged later) than...