The formation of Chinese multi-word constructions: evidence from the processing and production of Chinese native speakers

Chinese multi-word constructions;processing

Zheng, Hang

关键词: Chinese multi-word constructions;    processing;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/89210/ZHENG-THESIS-2015.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

In this thesis, the formations of two multi-word constructions in Chinese—the V-O verb copy constructions (VCC) and the disyllabic root-compound words (DSC)—are re-examined based on the evidence from native speakers’ processing and production. The purpose of this study is to fill an empirical gap from the perspective of theoretical postulations for the two constructions, namely, the Copy Merge Theory of Movement under the generative framework, and the ‘pre-stored templates’ and ‘emergence model’ assumed by the constructionist framework. Different from previous studies, the experimental approach is used as an alternative approach in my thesis. In examining the VCCs, a fill-in-the-blank task elicited native speakers’ responses to two blank conditions of the VCC context; a self-paced reading task compared native speakers’ reaction times to the VCC and non-VCC chunks of controlled frequencies. The overall findings confirm that native speakers perceive the VCCs the same way as they are derived according to the syntactic operations; the results also reveal that there may be a length constraint for the productivity of the ‘pre-stored templates’. In the investigation of the headedness structures of the DSCs, native speakers’ coinages were elicited and compared to the generalizations drawn on the basis of the textual materials. The results reveal that headed forms (as opposed to non-headed forms) are the prevalent structures in Chinese language as well as in Chinese speakers’ mental representations. This finding supports the language acquisition model put forward by emergent grammar, which assumes a correlation between inputs and outputs.