Sat - May 29, 2004

Bobaljik & Wurmbrand: Domains of agreement



The domain of agreement (with Jonathan Bobaljik)
Formerly "Relativized phases"
Available here: http://wurmbrand.uconn.edu/research/papers.html

Abstract:
In certain types of infinitival complementation constructions in three quite dissimilar languages (German, Japanese and Itelmen) expected interpretations curiously, but systematically, fail to arise. The missing interpretations are precisely those that would be expected if Agree—the establishment of licensing relations without movement—were possible; this is shown by comparison to minimally different constructions that establish both the existence of the Agree operation and the independent possibility of the interpretations in question. The account we are led to suggests that locality domains are not absolute but are relativized in two ways: firstly, Agree and A-movement respect different (if overlapping) locality conditions, and secondly, whether or not a given projection constitutes a domain boundary depends partly on its syntactic context. At the core of the paper is the proposed generalization that A-movement is forced, and cannot reconstruct, exactly when a DP originates in a lower agreement domain than its licensor.

Posted at 12:59 PM  |  Category:   |   |   |   |  E-mail me | 


©