Files
Abstract
The primary goal of this dissertation is to demonstrate that a close analysis of the syntactic and semantic properties of phrasal idioms, both with respect to some well-known constraints on idioms syntactic mobility and to some lesser-known aspectual co-occurrence restrictions, leads directly to the conclusion that the phonological forms of idioms are not present in narrow syntax. I therefore propose a model of sentence generation in which the conceptual-intentional system is involved in choosing phonologically null but semantically individuated Roots-as-nodes (N-Roots) at the outset of a derivation; these abstract Roots are the units processed by the computational component of the grammar. Cyclically transferred to the interfaces for interpretation and externalization, they are paired at the latter interface with the phonological forms of Roots-as-exponents (E-Roots), based on syntactic and semantic feature mapping. The secondary goal of this dissertation is to demonstrate that from the psycholinguistic perspective, syntactic structure-building in the context of language production (at least) cannot plausibly be modelled as a bottom-up process powered by iterative Merge. Surveying a range of theory-external as well as theory-internal evidence to this effect, I argue instead for top-down structure-building powered by iterative Branch, wherein N-Roots are inserted in functional, discourse, and scopal positions, and head chains which terminate in their theta positions. Although this aspect of the model is provisional, my intention is to lay the groundwork for in-depth consideration of the interaction between the timing of insertion question (i.e., pre-syntactic or post-syntactic insertion) and the directionality of derivations question (i.e., bottom-up or top-down structure-building) in a unified model. I refer to this combined, provisional model as Late Insertion, Top-Down (LIT-D), to differentiate it from models involving other settings of what one might call meta-parameters: Late Insertion, Bottom-Up (LIB-U), as assumed by most advocates of Distributed Morphology; Early Insertion, Top-Down (EIT-D), as assumed by most who have worked on top-down derivations; and Early Insertion, Bottom-Up (EIB-U), as assumed in most other versions of the Minimalist Program.