摘要
本文以转换-生成语法和二语习得的认知理论为基础,从多源语料库提取there be句型为研究对象,通过标注、统计和自然语言处理模型,对英语本族语者和中国二语学习者输出文本的意义、句法饱和度进行对比分析,结论如下:在基本结构层面,本族语和二语意义、句法饱和度构建无显著差异;在深层结构层面,前者倾向于使用高信息度的集合矩阵论元,后者倾向于低信息度的线性论元,其意义、句法饱和度显著低于前者;二者意义、句法饱和度差异源自深层认知构建和内化过程的不同,关联性认知和广域的概念化推动意义、句法的饱和,但二语意义、句法饱和度的增加会以句法错误率的上升为代价。
Based on the theories of transformational-generative grammar and cognitive linguistics of SLA,taking there be existentials from multi-sourced corpora as target structures and with the methods of tagging,structural statistics and natural language processing models,this study examined the differences between native English speakers and SL learners in the construction of semantic-syntactic saturation(SSS)in the output texts.The findings are as follows:there was no significant difference between native English speakers and SL learners in the SSS construction of the main frameworks while the SSS of predicate-arguments of SL learners was significantly lower than that of native speakers;native English speakers were inclined to build high information loaded and embedded arguments while SL learners tended to build low information loaded and linear arguments.The differences in SSS between them were caused by the divergence in deep-rooted cognitive relationship recognition and internalization process since widely context-correlated relationship recognition and conceptualization were the driving forces of SSS,but the development of SSS of SL learners would be at the cost of syntactic mistakes.
作者
秦琴
丁研
QIN Qin;DING Yan(School of Languages and Communication Studies,Beijing Jiaotong University,Beijing 100044,China)
出处
《山东外语教学》
北大核心
2021年第2期67-78,共12页
Shandong Foreign Language Teaching
基金
教育部人文社科基金青年项目“二语动词结构意义-句法加工图式研究:基于大数据分析技术”(项目编号:15YJC740061)的阶段性成果。
关键词
意义-句法饱和度
转换-生成语法
二语习得
semantic-syntactic saturation
transformational-generative grammar
second language acquisition
