|Title:||The Clause Structure of Malagasy: A Minimalist approach||Add Dissertation|
|Author:||Matt Pearson||Update Dissertation|
|Email:||click here to access email|
|Institution:||University of California, Los Angeles, Department of Linguistics|
|Abstract:||This thesis explores the clause structure and word order of Malagasy within the framework of Chomsky’s (1995) Minimalist Program and Kayne’s (1994) Antisymmetry Theory. In particular, I focus on the status of the clause-final external argument (EA), conventionally analyzed as a nominative case-marked subject. I consider two major questions about this constituent: What hierarchical position does the EA occupy in the clause structure, and why does it surface in a right-peripheral linear position, following the predicate?
With regard to its syntactic status, I argue that the EA is not a subject, but a topic, similar in its distribution to clause-initial topics in verb-second languages like Icelandic. I propose that EAs undergo A′-movement to the specifier of a TopP (topic phrase) projection, located above tense and below the position of the complementizer. Concerning word order, I show that the right-peripheral position of the EA can be derived via leftward movement of the predicate phrase over the EA in SpecTopP, in a manner consistent with Kayne’s Linear Correspondence Axiom. I suggest that predicate-fronting is triggered by the same lexical requirements responsible for T-to-C raising in Icelandic and other languages. The difference is that in Malagasy, unlike in Icelandic, T0 does not constitute an independent morphological word, and so it cannot be moved without causing the derivation to crash at PF. Since T0-movement is unavailable, TP-movement is employed instead. Malagasy may thus be regarded as the phrasal-movement analogue of a verb-second language.
The thesis is divided into four chapters. In chapter 1 I summarize my analysis and discuss my theoretical assumptions. In chapter 2 I give an overview of Malagasy word order, clause structure, and morphology. I also offer a tentative treatment of the Malagasy voicing system, which I equate with wh-agreement in Chamorro and other languages. In chapter 3 I present evidence from reconstruction and locality effects to show that the EA position behaves as an A′-position rather than a case position, strongly suggesting that the EA is a topic-like element rather than a subject. I also provide an alternative analysis of the well-known wh-extraction restriction in Malagasy. Finally in chapter 4 I discuss my XP-movement analysis of EA-final word order. I cite evidence in favor of this analysis from two domains, speech-act particle placement and word order in embedded clauses.