可比语境下关系代词取舍的多因素分析  

A comparable context-based multifactorial analysis of relativizer omission by Chinese EFL learners and native speakers of English

作　　者：房印杰 梁茂成 FANG Yinjie;LIANG Maocheng(School of Humanities,Beijing University of Posts and Telecommunications,Beijing 100876,China;School of Foreign Languages,Beihang University,Beijing 100083,China)

机构地区：[1]北京邮电大学人文学院外语系,北京市100876 [2]北京航空航天大学外国语学院,北京市100083

出　　处：《外语教学与研究》2019年第3期435-446,F0003,共13页Foreign Language Teaching and Research

基　　金：教育部人文社会科学研究青年基金项目“基于使用的中国学生英语关系从句认知统计模型研究”(17YJC740018)的阶段性成果

摘　　要：本文以中国英语学习者和英语本族语者的关系代词取舍为研究对象.通过双回归模型分析,发现:1)关系代词取舍并非随机选择,而是受到与认知加工负载相关因素的制约;2)与英语本族语者相比,中国英语学习者过度使用显性关系代词,但对隐性关系代词学习较好;3)中国英语学习者与英语本族语者的组间差异体现为关系代词取舍制约语境的典型性,在典型语境下,二者差异较小;在非典型语境下,二者差异增大。研究表明,基于“可比语境”的双回归分析能够准确剥离关系代词取舍中的显著制约因素,提取组间差异。A double regression analysis of relativizer omission by Chinese EFL learners and native speakers of English reveals that a)relativizer omission is not random but constrained by a variety of cognitive processing related factors;b)Chinese EFL learners,compared to native speakers of English,exhibit an overuse of explicit relativizer and a better command of implicit relativizer;c)the in ter-group discrepancy by Chinese EFL learners and native speakers of English is closely related to ideal constraining conditions on relativizer omission with a closer similarity under ideal conditions and a greater discrepancy under non-ideal conditions.It is demonstrated that the double regression analysis guided by"comparable context"plays a significant role in teasing apart salient constraining factors and retrieving in ter-group discrepancy in relativizer omission.

关 键 词：可比语境 关系代词取舍 过度概括 概率性预测 

分 类 号：H319.3[语言文字—英语]