【作者】 刘颖；

【导师】 盛玉麒；

【作者基本信息】 山东大学 ， 语言学及应用语言学， 2009， 博士

【摘要】 乔姆斯基的内在天赋论建立在刺激贫乏论(the poverty of stimulus argument)基础上,刺激贫乏是指儿童所接受的语言输入和最终获得的语言能力之间是不对称的。如果刺激贫乏论不能成立,将从根本上动摇语言天赋论的根基,从而也会动摇生成语法的根基。为探究儿童所接受的语言输入是否贫乏,是否包含了儿童习得语言所需要的相关信息,儿童的语言能力是否能从语言输入中获得,本文以儿童早期(两岁半以前)语言发展作为研究对象,在细致描写该时期儿童所处的语言环境及其理解与表达状况的基础上,将儿童的理解与表达、儿童的语言输出与所接受的输入加以对照,深入探讨了后天语言环境对儿童习得语言的影响和作用,为揭示儿童语言能力的获得提供了事实依据。本研究属于个案追踪研究,同时采取了语料库语言学方法和实验测定法,三种方法都建立在语言事实的基础上,得出的结论相互补充,相互印证。以录音材料为基础的自建语料库的规模为36万字。本文包括以下主要内容:第一章为绪论,首先说明了本文的研究缘起和术语使用情况,然后概括介绍了国内外儿童语言研究和汉语儿童早期语言研究的历史与现状,指出儿童语言研究的不充分之处及方法上存在的问题,最后对在研究过程中涉及到的主要理论、采用的研究方法及研究难点进行了介绍,其中语料库语言学和统计分析方法建立在对数据的准确、客观分析基础上,在研究手段上具有科学性和创新性。第二章介绍语料库建设情况,包括语料采集、录音转写、文本处理、分词校对四个方面。该语料库的建立为之后的词频、句频统计以及成人与儿童语言的对照提供了切实的数据基础,语料库的建设过程以及一些特别问题的处理方式也为儿童口语语料库的建设提供了借鉴。第三章对语前阶段儿童听到的成人话语进行了全面描写,通过从词、短语和句法结构三个层面进行分析,发现语前阶段的成人话语中包含了丰富多样的语言知识,儿童在语言出现之前就已接受了这些语言知识的刺激。第四章是对语料库中成人与儿童语料的计量分析。首先以词句数量为基础,分别统计成人话语输入量、成人与儿童话语句长、儿童话语类型。结果显示,儿童早期所接受的语言输入量是巨大的,绝非天赋论所说的“贫乏”;儿向语言(Child Directed Speech)的句长与儿童的语言水平密切相关;儿童自发性话语句长是语言水平的重要标志;模仿行为在独词句和双词句阶段作用明显。然后从输入和输出的角度出发对比成人和儿童的词汇量、高频词、高频句式以及首次听到和首次说出的时间。结果显示,输入词汇量远远大于输出词汇量,成人和儿童话语中越高频的词重合度越大,输入频率高的词,儿童使用频率也高,输入频率低的词,儿童使用频率也低。独词句是输入和输出中出现频率最高的句式,前十个高频句式的重合度较大。由此得出结论,成人话语输入对儿童言语输出的影响作用明显,输入频率高低与儿童习得快慢之间呈正相关。第五章是对儿童早期理解状况的详细描写,包括对非言语行为和言语行为的理解两方面。首先从信息传递的角度重新定义体态语,并指出体态语对儿童理解成人话语有促进作用。其次阐明儿童理解成人话语的特点及理解的判定问题,指出儿童理解的动态性,提出判断儿童是否理解成人话语的标准、方法、手段、原则,使儿童理解状况的描写具有可操作性。再次,分类描写儿童理解的语言单位,一类是参照词类标准划分的名词性、动词性、形容词性、代词性、数量性语言单位,一类是从句式、句类角度选取的领属结构、“把”字句、疑问句、祈使句,还有一类是体现汉语独特性的语序、声调。第六章儿童的表达状况,首先描写了儿童运用体态和简单语音等进行交际的非言语表达行为,然后在严格界定习得为自发说出的基础上分析了儿童最早习得的50个“词”的类别构成情况及其深层原因,发现名词和动词在儿童早期词汇中呈强势,但到底是名词多还是动词多是受诸多因素影响的,如语言类型、前导性话语、活动场景等等。最后,描写了儿童表达领属结构、“把”字句、问句的状况,并与儿童理解相对照,发现儿童表达的格式、语义类型都在语言输入范围内,甚至绝大部分都在儿童理解的范围内。结语部分阐明了本文的研究结论,概括了创新点及意义所在,并指出研究中存在的不足以及后续研究状况。本研究将儿童的理解能力纳入儿童语言研究范围,延伸了儿童语言研究的触角,拓宽了儿童语言研究的视野;从理解与表达、输入和输出相对照的角度研究儿童语言习得,得出的结论更有说服力;语料库语言学和统计分析方法的使用使结论建立在大规模语言事实的基础上。本研究能够为揭示后天语言环境对儿童语言发展的作用,进而揭示儿童语言能力的获得,提供事实依据;为汉语儿童母语习得研究提供个案补充;为汉语作为第二语言教学提供借鉴。更多还原

【Abstract】 Noam Chomsky’s Mentalism Innate Theory is based on the poverty of stimulus argument. The poverty of stimulus means that the language input and the language competence are unequal. If the poverty of stimulus argument is untenable, it will fundamentally shake the foundation of Mentalism Innate Theory and thus shake the foundation of Generative Grammar. But is the language input scarce actually? Does it include the necessary information for language acquisition? Can children acquire the language competence because of the language input? In order to probe these issues the dissertation takes the language development of early childhood (younger than the age of two and a half years ago) as the object of its research. On the one hand, this dissertation describes the language input and the child understanding and expression; on the other hand, the dissertation compares the understanding with expression, compares the language output with input. The ultimate aim is to reveal the impact of postnatal language input on language acquisition.This research is one kind of the case follow-up study. At the same time it adopts corpus linguistics method and experiment method. These three methods are all based on the language facts. The conclusions complement and confirm mutually. The corpus which is based on the recordings includes 360,000 characters.This dissertation includes the following main elements:Chapter one: preface. Firstly, we illustrate the origin and terminology, and then overview the history and current situation of the study on child language in China and abroad and Chinese language study on early childhood. Secondly, we point out the inadequate parts and the problems in child language research. Finally, we introduce the main theories and methods and difficulties in this research. The corpus linguistics method and the statistical analysis method are based on the accurate, objective data analysis, so they are scientific and innovative.Chapter two: corpus construction. This chapter focuses on the language data collection, recording to writing, text processing, segmentation proofreading. The establishment of the corpus provides a real foundation for word frequency and sentence-frequency statistics as well as the comparison between the adult language and child language. Corpus construction process and the means to deal with some special problems can provide a reference for the construction of child spoken language.In chapter three, we describe the adult language in Pre-language stage thoroughly. We find that the adult language includes a variety of linguistic knowledge by means of the analysis of words, phrases and syntax. The child has already accepted these stimuli of language knowledge.In chapter four, we compute and analyze the adult and child language in the corpus. Firstly, we calculate the quantity of the adult’s utterances, MLU(the Mean Length of Utterances) of the adult and the child and types of the child utterances. The result shows that the early childhood language input are enormous. It is not poverty actually. The MLU of CDS (Child Directed Speech) is closely related to child language level. The Child’s spontaneous utterance is an important indicator of language proficiency. Imitation happens more often at single-word and double-words stage. Secondly, we compare the adult’s vocabulary with the child’s vocabulary, compare the adult’s high-frequency words with the child’s high-frequency words, compare the adult’s high-frequency sentence patterns with the child’s high-frequency sentence patterns, compare the first time to hear with the first time to utter. The results show that the number of input vocabulary is much more than the output. The more high-frequency of the adult’s words and the child’s, the coincidence degree becomes greater. The vocabularies of high frequency appear in the child’s usage as well as the adult’s usage. Also, the vocabularies of low frequency appear in the child’s usage as well as the adult’s usage. Single-word sentence is the sentence of the highest frequency sentence of input and output. The coincidence degree of the preceding ten sentence patterns is quite great. According to these results we draw a conclusion that the adult language input has an obvious effect on child language output, and there is a positive correlation between the input frequency and child language acquisition. In chapter five, we describe the understanding of early childhood, including the understanding of speech act and non-speech act. First of all, from the perspective of the transmission of information we redefine body language, and point out that body language can promote child’s understanding of the adult’s utterances. Secondly, we put forwards our views about the characteristics and the judgment of child’s understanding and point out the dynamic feature of child’s understanding, and point out the criteria, methods, means, and principles of child’s understanding. Thus we can describe the situation of child’s understanding operatively. Thirdly, we describe the language understanding units. One category is noun unit, verb unit, adjective unit, pronoun unit and numeral unit which are selected by reference to words. The second category is superior-subordinate structure, "Ba" construction, interrogative sentence, and imperative sentence by reference to sentence pattern and sentence mood. The third category is word order and tones which reflect the uniqueness of Chinese.In chapter six, we describe the expression of the child. Firstly, we describe the non-speech act using simple gesture and voice, and then analyze the first 50 "words" after strictly defining acquisition as the act of speaking spontaneously. We find that nouns and verbs are dominating vocabularies in early childhood, and whether nouns are more than verbs or not influenced by the factors such as language type, preceding discourse, and activity scene and so on. Finally, we describe the expression of superior-subordinate structure, "Ba" construction, and interrogative sentence and compare these with the understanding. We find that the structures and semantic types output are in the scope of language input and even the vast majority of output is included by understanding range.The conclusion part clarifies the views of this research, and gives a summary of innovations and significance of the research, and points out the deficiency of this dissertation and the project of further research.This study takes the language understanding of the child into child language research scope, extending the reach of child language study, and broadening the vision of child language research. This study compares child understanding with child expression, compares language input with output, and this makes the conclusions more convincing. The conclusions are based on large-scale language materials by the way of corpus linguistics method and statistical analysis method.This study can provide the facts and basis for revealing the effect of postnatal language environment on child language development, and then revealing the language acquisition. It provides supplementary cases for the study on language acquisition of Chinese children. It provides reference for teaching Chinese as a second language.更多还原