Constraining Operations : A Phase-based Account of Improper Movement and Anaphoric Binding

Abstract

The goal of this dissertation is to offer a novel way of relating locality restrictions on movement and binding dependencies to a common set of syntactic factors. The starting point is the long noted observation that movement operations must apply in a particular order to yield a licit result (Chomsky 1973). The question of how the restrictions on ordering of movement operations might be derived in a principled way has been one of the major concerns of the linguistic theory ever since they were fist noted. This thesis aims at providing a new analysis of the phenomenon, relying on theoretical tools and conceptual advances of the now broadly adopted Minimalist framework of syntactic theory (Chomsky (1995, 2001)). I start by investigating possible feeding/bleeding relations between various movement operations and show, drawing on some recent work on the topic, that the same ordering restrictions can be observed not only in cases involving consecutive movements of the same phrase (captured by the standard formulation of the Ban on Improper Movement), but also in cases involving subextraction from moved phrases and remnant movement. I then argue that the observed ordering restrictions in all three configurations can be derived from the independently needed hierarchy of functional projections, and without recourse to ordering statements and/or representational filters. I show that this can be achieved by relying on the internal featural make-up of the moving phrase, in conjunction with a particular view regarding the timing and manner in which linguistic structure is spelled out. In the second part, I turn to a different empirical domain, that of anaphoric binding, and argue that the same factors are also crucial in regulating the distribution and interpretation of anaphoric relations. I show that the particular assumptions concerning the nature of syntactic computation motivated by the analysis of movement phenomena lead to some novel predictions in the domain of anaphoric binding, which I argue are empirically supported. Though the main focus is on the syntactic aspects of binding relations, particularly on locality restrictions, semantic and discourse properties of binding dependencies are also taken into consideration. The primary data is drawn from English, but other languages, including German, Dutch and Serbian, are also discussed.