Please use this identifier to cite or link to this item:
|Title:||Origins of Phrase Structure|
|Authors:||Stowell, Timothy Angus|
|Abstract:||At a descriptive level, it is a trivial observation that each speaker of a human language knows that words in sentences are organized into classes of hierarchically-defined phrases, each with distinctive clusters of properties pertaining to internal structure and external distribution. The significant empirical question for the theory of phrase structure concerns the form in which this knowledge is represented in the mind. Wiyhin the scientific tradition of generative grammar, it has commonly been assumed that a large part of this knowledge is encoded into the formulae of context-free rewrite rules belonging to the Categorial component of the Base. These Categorial rules are supposed to define the idiosyncratic properties of the phrases of each syntactic category: noun phrases, verb phrases, adjective phrases, etc. Although this rule system has proved to be a valuable heuristic tool for identifying and formulating various properties of phrase structure that must be accounted for by theory of grammar, it suffers from serious problems of explanatory adequacy if it is understood as a hypothesis about the actual structure of linguistic knowledge in the mind. The major claim of this thesis is that the component of Categorial rules does not exist, and that its major empirical effects can be deduced from other components of grammar. In particular, the assignment of syntactic Case is assumed to observe a strict condition of adjacency, which plays an important role in determining the linear arrangement of certain combinations of subcategorized complements. This condition interacts with a principle that prevents certain syntactic categories from being assigned Case to derive a number of complex properties associated with a variety of clausal complement structures. The elimination of the Categorial component and the adoption of the adjacency condition on Case assignmen forces quite radical departures from previous assumptions about several syntactic constructions. In some cases, it is necessary to reinterpret certain constituents that have traditionally been analyzed as independent phrases as actually being incorporated within the structure of a lexical head by rules of word-formation. This has interesting consequences for the theory of the interaction between the word-formation component of the grammar and the hierarchical phrase structure configurations defined by the category-neutral X-bar system. In addition, the extended component of word-formation forms the basis for an account of the distribution of certain marked constructions involving Reanalysis rules in various languages. The principles of phrase structure and Case assignment also interact in complex ways with the assignment of thematic roles to arguments. A formalization of thematic role assignment is developed, providing the basis for a possible explanation for the apparent grammatical equivalence of superficially distinct structures of proper government of empty categories. The theory of thematic structure proposed here allows for a restrictive theory of the encoding of strict subcategorization requirements, and leads to a revision in the syntactic analysis of the categorial identity and X-bar structure of various types of clauses.|
|Appears in Collections:||Dissertations (restricted access)|
Files in This Item:
|stowell.pdf||31.12 MB||Adobe PDF||View/Open Request a copy|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.