中国英语学习者导句语类（C）习得研究

发布时间：2018-04-04 03:24

  本文选题：句法-形态接口　切入点：参数设置　出处：《北京外国语大学》2016年博士论文

【摘要】：本研究目标为考察特征识别与特征重组复杂度是否为影响中国英语学习者习得导句语类(Comp,亦缩略为C)的两个关键要素。基于不同学者对句法-形态接口广泛的探索,本研究以屈折形态学中的词汇-实例化模型(Stump 2001)为蓝本,借鉴Chomsky(2007b,2014)主张在后句法阶段语音外化句法运算结果的操作,尝试重新定义句法-形态接口,即：在前句法阶段实现抽象句法属性与形态的映射,在后句法阶段将句法运算的结果通过形态-音系运算按个体语言要求显性实例化。以此为指导,本研究在当前最简方案生物语言学框架下重构参数方案,即整合前句法微观词汇参数与后句法语音外化参数。依据这一新的参数整合方案,重新审视生成语法视角下备受关注的“特征重组假说”(Lardiere 2005,2009a, 2009b),提出“特征识别-重组复杂度假说”(Feature Recognition and Reassembly Complexity Hypothesis,FRRCH),探索二语中介语发展的路径,充实二语变迁理论的具体内容。数据收集结合口头提问与书面语法判断,综合利用语音线索与对最小对立对(minimal pairs)的直觉判断,探究学习者习得英语C的过程。209名被试参加英语水平测试,通过筛选,排除不符合分数段要求以及口头提问与书面判断数据不全者,初、中和高级三组被试共计94人的数据进入统计分析。研究发现,被试对英语C中[TNS],[EPP]以及[Decl]和[Q]/[Wh]等的识别成功度与C语类习得呈正向相关关系；并且,被试C语类习得程度呈现从低至高的递增态势。这就表明,当母语为汉语的英语学习者习得英语C时,依据Kayne(2005a,2013)认为语言中存在隐性句法成分的观点,当母语(L1)特征为隐性,二语(L2)相应特征为显性时,较难习得L2句法结构。研究结果似乎证实了FRRCH关于前句法阶段的特征识别与C语类习得关系的预测。此外,被试口头产出一般疑问句中C的正确率显著高于产出特殊疑问句中C的正确率,以及被试对英语内嵌句中T-C移动抑制判断的正确率显著高于T-C错误移动判断的正确率。这就说明,如果前句法阶段相关形式特征识别不成功,特征深度重组复杂度越高,C语类习得难度越大。这也似乎验证了FRRCH有关前句法阶段的特征识别、特征深度重组复杂度与C语类习得关系的预测。在阐释中国英语学习者如何习得英语C时,FRRCH以乎比“特征重组假说”的阐释更加细致,比一般习得理论(如迁移说)提供更加具体的解释。这一发现与当前“接口假说”(Interface Hypothesis)所持的普遍性结论一致：涉及(语言模块之间)接口的属性比仅涉句法结构的属性给L2学习者带来更大的学习负担,总体上更难习得(Montrul 2002,2004；Sorace 2006；White 2011；戴曼纯2014)。这也表明,FRRCH的理论基础——重新界定的句法-形态映射与参数整合方案有其合理性。
[Abstract]:The purpose of this study is to investigate whether the complexity of feature recognition and feature recombination is the two key elements that affect the acquisition of the Chinese EFL learners' guide sentence category / component and also abbreviate to C).Based on the extensive exploration of syntactic and morphological interfaces made by different scholars, this study based on the lexical instantiation model of inflectional morphology (stump 2001), using Chomskysky 2007 bbm2014) for reference, advocated the operation of the results of speech externalization in the post-syntactic phase.This paper attempts to redefine the syntactic-morphological interface, that is, to realize the mapping of abstract syntactic attributes and morphology in the pre-syntactic stage, and to instantiate the results of the syntactic operations through Morphology operations according to the individual language requirements in the post-syntactic stage.Guided by this, the present study reconstructs the parameter scheme under the framework of biolinguistics, namely, the pre-syntactic micro-lexical parameters and the post-syntactic externalization parameters.According to this new parameter integration scheme, this paper reexamines the "feature recombination hypothesis", which has attracted much attention from the perspective of generative grammar, such as Lardiere 2005 / 2009a, 2009b, proposes the feature Recognition and Reassembly Complexity HypothesisFRRCHS, and explores the development path of L2 interlanguage.Enrich the specific content of the second language change theory.Data collection combined oral questions with written grammatical judgment, combined with phonetics cues and intuitive judgments on minimal pairsto explore the process of learners' acquisition of English C. 209 subjects took English proficiency tests.The data of 94 subjects in the initial, middle and advanced groups were included in the statistical analysis, excluding those who did not meet the requirements of the score paragraph and the incomplete data of oral questioning and written judgment.It is found that the degree of success of [TNS], [EPP], [Decl] and [Q] / [Wh] in English C is positively correlated with the acquisition of C, and the acquisition degree of C is increasing from low to high.The results seem to confirm FRRCH's prediction of the relationship between pre-syntactic stage feature recognition and C class acquisition.In addition, the correct rate of C in general interrogative sentence was significantly higher than that in special interrogative sentence, and the correct rate of T-C motion inhibition judgment in English embedded sentence was significantly higher than that in T-C false motion judgment.This shows that if the recognition of related formal features in the pre-syntactic stage is not successful, the higher the complexity of feature recombination is, the more difficult it is to acquire C category.This also seems to verify the prediction of the relationship between the feature recognition of pre-syntactic stage, the complexity of feature deep recombination and the C class acquisition in FRRCH.In explaining how Chinese EFL learners acquire English C, FRRCH is more detailed than the "characteristic recombination hypothesis" and provides more specific explanations than the general acquisition theory (such as transfer theory).This finding is consistent with the universal conclusion of the current Interface hypothesis: attributes involving interfaces (between language modules) impose a greater learning burden on L2 learners than attributes involving only syntactic structures.On the whole, it is more difficult to acquire Montrul 2002 / 2004 / Sorace 2006 / White 2011;This also shows that the theoretical basis of FRRCH-redefined syntactic-morphological mapping and parameter integration scheme is reasonable.
【学位授予单位】：北京外国语大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：H319.3
，

本文编号：1708151