No abstract

A central goal of this paper is to present a new account of improper-movement phenomena based on ChomskyÕs (2007, 2008) phase-based derivational approach. We claim that improper movement is excluded by virtue of Agree failure between a moving element and a finite T as a consequence of ''feature-splitting'' Internal Merge, which we argue is the most (or at least a very) natural implementation of ChomskyÕs /-feature-inheritance system and RichardsÕs (2007) value-transfer-simultaneity analysis. This analysis has a number of empirical and theoretical consequences: (i) regarding the explanation of A¢-opacity/-transparency intervention effects (Rezac 2003, Carstens 2005); (ii) the possible elimination, or reduction in scope, of the Activity Condition; and (iii) the possible characterization of A/A¢-position types solely in terms of categorial features. Moreover, we propose that (iv) the ban on improper movement is, in fact, not universal but is morphologically parameterized (at least) between English and the Bantu language Kilega.

No abstract

No abstract

This chapter seeks to shed new light on the formal properties of Merge. Chomsky's (2007, 2008) feature-inheritance system makes it possible that two heads such as T and C simultaneously attract a single element. Obata and Epstein (2008, 2011) propose a new kind of structure building operation called Feature-Splitting Internal Merge (FSIM): which splits a single element into two syntactic objects. FSIM gives a new derivational, agreement-based, phi-featural account of improper movement phenomena, some of which converge. The FSIM hypothesis is applied to other types of structure building with the proposal that as there are three kinds of T/C there are also three kinds of v/V (including one licensing null-Case). One consequence is that tough-constructions are, in fact, derivable by a kind of “proper” (i.e. convergent) improper movement derivation. In addition, the analysis implies that Case on a tough-subject is revised" in the course of the derivation.