Font Size: a A A

Clausal peripheries and resumptives: A cross-linguistic study of topic-comment structures

Posted on:2008-06-28Degree:Ph.DType:Thesis
University:Georgetown UniversityCandidate:Kuong, Io-KeiFull Text:PDF
GTID:2445390005468241Subject:Language
Abstract/Summary:
This thesis is a study of the syntax of sentence topics from a cross-linguistic perspective. Several phenomena involving the clausal periphery are investigated, including topicalization, focus-movement, Hanging Topic Left Dislocation, (Clitic) Left Dislocation, and Frame-Setting Topics. Adopting a feature-based minimalist approach (Chomsky 1995, 2000, 2001, 2004), I have revised the fixed-order Split-CP hierarchy developed in Rizzi (1997) and Beninca (2001) to account for data from English, Mandarin Chinese, Portuguese, and Italian. The Split-CP domain is composed of Disc(ourse)P, ForceP, TopP, FocP, and FinP.; Topicalization in English, Mandarin, and Portuguese is shown to be a heterogeneous construction, confirming Gundel's (1974) and Reinhart's (1982) claim. Topicalization can be a topic marking operation or an instance of focus-movement in English, Mandarin, and Portuguese. In Italian, topicalization can only be focus-movement. Focus-movement targets [Spec,FocP]. This analysis is extended to Negative Preposing in English. Topicalization of sentence topic targets [Spec,TopP], with the topicalized constituent either being base-generated in or moving to Spec. The previously understudied fact that in nonfinite clauses topicalization and Negative Preposing are unavailable, but WH-movement is possible is explained by what I refer to as Clause Structure Minimization, according to which FocP and functional projections above it are not projected in nonfinite clauses. It is argued that sentential subjects and adjunct clauses also have a reduced clausal periphery.; Three kinds of sentence topics are studied and I have found that they do not occur randomly in overt syntax. To explain why Hanging Topics precede Displaced Topics, I follow Beninca's (2001) suggestion that Hanging Topics are base-generated in [Spec,DiscP], higher than [Spec,TopP] targeted by Displaced Topics. Frame-Setting Topics (FST) are island-insensitive and precede Hanging Topics and Displaced Topics. I claim that FST is generated in a recursive DiscP, and that its precedence over Hanging Topics follows from the scopal nature of the former.; An interesting correlation is found between resumptives in topic-comment structures and syntactic island constraints. The presence of resumptives inside the comment is obligatory if and only if topic extraction would result in island constraint violations. Hence, resumptives can salvage an otherwise illicit derivation.
Keywords/Search Tags:Topic, Resumptives, Clausal, Spec
Related items