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

Reviewed by:
  • Agreement and head movement: Clitics, incorporation
  • M. Rita Manzini
Agreement and head movement: Clitics, incorporation, and defective goals. By Ian Roberts. (Linguistic inquiry monograph 59.) Cambridge, MA: MIT Press, 2010. Pp. 290. ISBN 9780262514323. $30.

This book opens and closes with the discussion of a passage by Chomsky (2001:37–38) to the effect that head movement is not part of the narrow syntax. In answer to this, Roberts presents a strong and often convincing case in favor of the classical notion of syntactic head movement, providing a key reference for anybody wanting to either uphold that notion or argue against it.

Chomsky’s main empirical argument is that semantic effects of head raising are ‘slight or nonexistent’. In Ch. 1, ‘LF-effects of head movement’, R provides a compelling rebuttal of the argument, based on well-known minimal pairs such as English Hasn’t anybody answered? versus *Anybody hasn’t answered. One could try mimicking the effects of syntactic head movement via VP-remnant movement. But as R correctly points out, this would imply complicating the c-command definition of scope that allows the negative polarity item to be licensed by the negation. Ch. 2, ‘Head movement and pied-piping’, addresses another objection raised by Chomsky, concerning the difficulty of providing formal grounds for the (non)application of pied-piping. The solution that Chomsky suggests is essentially to render syntactic pied-piping obligatory. R preliminarily notices that the obligatoriness of pied-piping could be enforced by the A-over-A constraint, whose classical formulation requires any operation targeting A to target the maximal phrase of category A. But of course if A lacks internal structure (i.e. it is both minimal and maximal), then head movement is allowed; R argues that this is what happens in cliticization.

V-movement is a different matter. Adopting work by Andrea Rackowski and Norvin Richards, R assumes that movement of a goal α is blocked only if there is a distinct goal β that c-commands it. Since a maximal projection never c-commands its head nor vice versa (because of their containment relation), movement of a head is always possible from within its maximal projection (though instances of A-over-A ‘at a distance’ are still blocked, irrelevantly for the present discussion).

The details of cliticization are dealt with in Ch. 3, ‘Cliticization’, which by itself represents about half of the book. For R, cliticization consists of the movement of the clitic from the complement [End Page 212] position where it is first merged to a position adjoined to v, namely the Marantzian category defining the word phase (adjunction is to the left in compliance with Richard Kayne’s linear correspondence axiom (LCA)). The reason why the clitic moves to the edge of the v-phase is that the interpretable ϕ-features of the clitic value the uninterpretable ϕ-features of v. Following proposals of Cardinaletti and Starke (1999), the categorial status of the clitic is characterized as ϕP, since ‘clitic pronouns consist only of the inflectional part of the structure of a pronoun’ (56). It turns out therefore that clitics ‘do not have a label distinct from their host’ (55). This ‘defective’ nature of clitic goals means that in the case of cliticization ‘Agree and Move/Internal Merge are formally indistinguishable’ (60, author’s emphasis). In Agree generally ‘the goal does not constitute an identical copy of the copied features bundle. But, precisely in the case of incorporation, this is what happens. For this reason we see the PF-effect of movement, with the ϕ-features realized on the probe and the copy deleted’ (61). This also allows R to dispose of one of the strongest objections raised by Chomsky (2001) against syntactic head movement, namely the fact that the adjoined landing site does not c-command the trace. I quote: ‘If cliticization is triggered by Agree, then it is unclear that any c-command relation is required beyond that between the probe and the goal, postulated independently of movement’ (59).

While R’s preoccupation is mostly with the formal aspects of the theory, the actual implementation of his ideas, in the empirical domains he chooses, requires a certain amount of substantive assumptions. Thus...

pdf

Share