【作者】 管玮；

【导师】 马博森；

【作者基本信息】 浙江大学 ， 英语语言文学， 2014， 博士

【摘要】 转述是人类日常交际中一种常见且重要的语言现象。只要人们告诉他们的听众,某人在他时某个场景所言、所写和所想的内容,他们就是在转述。缺少转述,语言的交际功能将大打折扣。但是已有的语言学领域的转述研究却很少关注在汉语自然会话中,说话人是如何使用这一语言策略的。基于以上原因,本文在对汉语自然会话中的转述现象进行界定和描述的基础上,构建了一个研究汉语自然会话中转述现象的三维分析框架。该框架包括转述方式、转述内容和转述声源三个分析维度。在转述方式维度,我们提出分析转述方式的两种途径：引入方式和转述视点,分别考察转述者如何引入转述信息以及采用何种视点表达转述信息。在转述内容维度,分转述内容类型和转述行为信号两个方面,考察转述者向受话人传递何种类型的转述信息,以及他们在传递此类信息时采用的转述行为信号。在转述声源维度,我们首先把转述信息的声源分为生命体和非生命体两大声源子类,然后将这两个子类进一步分类,从而考察转述者向受话人转述的信息来自何人。我们希望通过这个三维分析框架立体呈现转述行为的动态性、系统性和选择性。之后我们根据该三维分析框架,借助语料库语言学方法,对自建对比语料库中儿童和成人两类人群的转述策略进行对比研究。我们首先制订了详细的标注方案,然后对两类人群的转述策略进行了标注,量化对比分析主要从五个方面展开：转述现象总体对比、转述方式对比、转述内容对比、转述声源对比和综合交叉对比。在量化分析的基础上,我们还探讨了导致两类人群转述策略异同的原因。我们的研究显示：1.在总体使用上,儿童和成人在日常会话中均使用了相当数量的转述,证明转述作为一种会话策略对人类交际的重要性。不同的是,成人的转述使用频率明显高于儿童,这很可能是儿童和成人叙事能力的差异造成的。2.在转述方式上,两类人群的相似点主要是：1)儿童选择的方式类型在涵盖范围上接近成人。2)常规转述均为两类人群主要的引入方式。3)被述者视点均为两类人群主要的转述视点。不同在于：1)儿童对主观构拟式转述、混合视点的使用频率明显低于成人,这可能是儿童在抽象思维能力和词汇能力上的欠缺造成的。2)儿童对零转述使用高于成人,这与儿童和父母之间问答形式的交流方式有直接关系。3.在转述内容上,儿童和成人在转述内容类型、一些常见转述行为信号(主要是转述动词)类型的使用上呈现出很大相似性。差异主要在于：1)成人使用的转述行为信号在语法种类和语义内容上更具多样性,这说明儿童在语法、语义水平层面的发展与成人存在的差距。2)儿童更多地转述单独的半言语信息如叹词,这可能与他们偏爱借助模仿表演进行叙事的方式有关。4.在转述声源上,两类人群的共同点主要是：1)儿童使用的转述信息声源的类型在涵盖范围上接近于成人。2)定指人物声源和自我声源是两类人群转述最多的两种声源类型。主要区别在于：1)成人对不定指人物的转述频率显著高于儿童,重要原因是成人借助不定指人物声源实现更多的交际目的,如避免提及被述者身份,为自己观点提供有力证明等,这些都是儿童没有的。2)儿童对艺术作品人物的转述频率高于成人,这与他们的日常生活经历有关。基于以上的研究发现,我们认为,7-8岁儿童已经掌握了主要的转述策略能力,但其使用的多样性以及熟练程度与成人相比仍存在差距。因此,本文的研究成果也在一定程度上促进了我们对儿童在转述策略上的语言能力发展规律的认识。更多还原

【Abstract】 The use of represented discourse is a universal and significant language phenomenon in human communication. People are representing discourse, when conveying to their receivers what was or might be said, written or thought, either by themselves or by others in a displaced ground. Without represented discourse, language would be sadly limited in its potential as a means of communication. Little of linguistic literature, however, explicitly investigates how Chinese interactants manage representation as a language strategy in spontaneous conversation.This dissertation defines, describes the use of represented discourse in Chinese naturally-occurring conversation and then constructs a dynamic three-dimensional framework for analyzing it. The framework consists of three analytic dimensions:the analysis of representing ways; the analysis of represented content and the analysis of source of voices of represented discourse.When we turn to the first dimension, two approaches, namely introducing ways and perspective-taking, are adopted for exploring the ways which the interactants choose to represent discourse. The two approaches address the questions of what representing signals representers have available to illustrate that they are going to represent discourse and what perspectives they may assume when demonstrating the represented discourse.When turning to the second dimension, we explore two avenues for investigation:the content types of represented discourse and the categories of representing behaviour signals. The first avenue attempts to distinguish the content categories of discourse that the representers could convey to their receivers, and the second aims to elucidate the grammatical categories of representing behaviour signals they may utilize in accordance with the content types.The last dimension construes who might be presented as the source of the discourse being represented. We first distinguish two main classes of the source of voices of represented discourse, namely animate source of voices and inanimate source of voices and further divide them into several sub-classes. The three-dimensional framework attempts to uncover the intrinsic dynamism, systematization and selectivity in the process of rep resenting discourse.Based on the proposed analytic framework, the dissertation makes a corpus-based contrastive study between children and adults in terms of their use of represented discourse. Being careful to design the annotation scheme by taking the proposed framework as a guide and annotate the represented discourse in the corpus data, we go on to analyze the annotation results, comparing the representing strategies employed by both children and adults by using a quantitative method of comparison. The comparative studies between the two groups are carried out in terms of the overall usage of representing strategies, the usage of the representing ways, the represented content and the source of voices of represented discourse, and in terms of a synthetic comparison. After the quantitative comparison, the dissertation then tries to provide accounts for the similarities and differences between the two groups.Our study demonstrates that:1. With regard to the overall usage, the results indicate that represented discourse is significant as a conversational strategy in human communication in that it is prevalent in the spontaneous conversation of both children and adults. But the data also demonstrate that adults use represented discourse much more frequently than children do plausibly due to the differences in terms of narrative abilities between the two groups.2. The wide range of choices of representing ways possessed by children is close to that possessed by adults. And in each group, the prototypical representing signal is the most frequent form used to introduce represented discourse, since it is the clearest for the receivers to recognize the representation. Moreover, the representee’s perspective helps the representers avoid changing the referential expressions in the represented information, explaining the fact that it is the first option for both groups when demonstrating the represented segment. The two groups, however, behave differently in that:1)such ways as subjectively constructed representation and representation with mixed perspectives are significantly more likely to be employed by adults than by children, which probably has to do with children’s constrained imagination and lexical knowledge;2) children tend to make use of zero mode of representation more than adults do, which can presumably be attributed to the Question-Answer interaction model between the children and their parents.3. In terms of the content types and some common representing behaviour signals, the two groups share a similar range of choices. But the representing behaviour signals adopted by adults manifest themselves with much more grammatical and semantic diversities than those by children. The fact that individual interjections as represented content occur more in the speech of children supports the claim that children prefer to mimic in the process of narrating.4. With respect to the source of voices of represented discourse, the two groups share a vast majority of the types of source of voices. Moreover, specified others and self are the first and second most-common types of source of voices for both children and adults. However, adults show a higher frequency to represent discourse from unspecified others than children and the explanation probably lies essentially in their varied manipulative purposes which never lead children to do so. Children represent more discourse from story or cartoon characters than adults do, which can probably be attributed to their daily experiences.The above results lead us to propose that children aged7to8have shown the general ability to represent discourse, but that there are still differences between them and adults in terms of diversity and frequency of the use. This study, therefore, has taken one small step toward our understanding of the linguistic development of children’s representing ability.更多还原