Kreps, Christian; (1998) Extraction, movement and dependency theory. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text U643824.pdf Download (9MB) | Preview

Abstract

My aim in this thesis is to explore the potential of a dependency-based theory of syntax to account for extraction phenomena in English. Dependency grammars differ from constituency-based theories in expressing syntactic structure through direct relations between words rather than through their participation in phrase structure. However, whereas the majority of syntactic theories employing a phrase structure formalism have relatively well-developed accounts of extraction, dependency-based accounts have, in comparison, been virtually non-existent. This is surprising given that extraction phenomena constitute a key body of data, and an important test of a theory's capacity to account for long-distance syntactic relations. In order to account for these data theories have often suggested complex analyses involving multiple levels of representation and syntactic movement. However, in this thesis I hope to show that a monostratal theory of dependency with no such process of syntactic movement can provide a plausible and adequate account of extraction, which may compete on an equal footing with more complex approaches advanced in constituency-based theories. After presenting a wide range of extraction data in the first chapter, in the second and third chapters I compare and contrast accounts of these data formulated within Principles and Parameters Theory, representing the constituency-oriented linguistic mainstream, and Word Grammar, the only dependency theory to offer any account of extraction phenomena. The fourth chapter will outline an alternative theory of dependency syntax, which I will refer to as Licensing Grammar (LG). Although LG takes Word Grammar as its starting point, it incorporates many significant differences from this and other theories, notably in its reformulation of dependency in terms of licensing. The fifth chapter then examines how LG, when integrated with a relatively simple parsing system, can serve as the basis of an effective account of extraction data.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as CSVJSONXML

Archive Staff Only

View Item