| This dissertation studies three facets of head-final relative clauses in Chinese---their derivation, processing, and acquisition. The unifying theme is to demonstrate universal aspects of languages with different surface structures. The structure of Chinese relative clauses is particularly relevant because their head-final property provides a good case for cross-linguistic comparison. Extensive research work has been done on head-initial relative clauses in English. However, the derivation, processing, and acquisition patterns of head-final relative clauses in Chinese are not well understood. The findings in this study not only support the hypothesis that the underlying mechanisms of the grammar and performance are universally the same for different languages, but also help to understand and clarify issues that cannot be learned from studies on English.; First, I discuss the derivation of relative clauses in Chinese. Head-initial relative clauses in English are believed to be derived via movement because relativization obeys syntactic islands (conditions on movement). In Chinese, certain cases in which complex NP island violations are acceptable have made previous researchers propose a non-movement analysis for the formation of relative clauses. In this thesis, I propose a movement analysis to account for the seeming island violation cases: When the extraction out of complex NP islands is possible, there is actually no island violation, because the predicates allow so-called double subject constructions, and the element being moved is a major subject located outside of the island. If this is correct, it not only argues against the non-movement analysis, but it also shows that all relative clauses in Chinese are derived via movement, obeying syntactic island constraints.; Secondly, I investigate the issue of incrementality and structural prediction in the on-line sentence processing of head-final relative clauses in Chinese. Lastly, I discuss the acquisition pattern of Chinese relative clauses.; This research is significant in that it offers both arguments and experimental results to show that the derivation, processing, and acquisition of head-final relative clauses in Chinese are the same as head-initial relative clauses in English, despite their different surface structures. This thesis supports the claim that the fundamental mechanisms of human grammar and language performance are universally the same. (Abstract shortened by UMI.)... |
