节点文献

英语中的生命度等级研究

A Study of Animacy Hierarchy in English

【作者】 吉洁

【导师】 梁茂成;

【作者基本信息】 北京外国语大学 , 外国语言学及应用语言学, 2014, 博士

【摘要】 生命度作为语言学的重要范畴之一,影响着多种语言现象,具有跨语言的普遍性,引发了语言学各个分支学科的广泛探讨。然而,已有研究未涉及生命度的内涵与本质,对其等级的划分多集中于名词和代词层面,且缺乏系统的分析方法和大规模的语料验证。因此,本研究试图对生命度概念进行深入的理论思考,在其基础上提出生命度等级的分析方法,并使用真实语料对其研究层面进行拓展。本研究认为,生命度具有体验性、原型范畴性、共性与动态性。即,生命度的本质是一种以人为原型的等级范畴。对生命度的认知是一种从人类视角出发的、以人类为中心的范畴化认知。这种认知基于大脑体验、根源于自然进化,具有普遍共性,同时又具有文化和语境中的动态性。本研究进一步归纳出生命度范畴的八个原型特征及其权重,为生命度等级提供了行之有效的分析框架。在上述理论视角下,本研究从横、纵两个维度拓展了生命度的研究范围。在横向维度中,将英语中的生命度等级,从原有的名词和代词,扩展到动词和形容词;并从词语层面,扩展到了搭配和语篇层面。这拓展了生命度的研究宽度。在纵向维度,将生命度从静态等级,扩展到动态等级。这拓展了生命度的研究深度。研究过程全部基于BNC语料库与Crown/CLOB语料库,可靠性与代表性较强。本研究的主要结果如下:1)词语层面存在静态生命度等级。各词类的等级分布不同:名词个数呈“沙漏型”分布,频数呈“蝴蝶型”分布;动词个数呈“橄榄型”分布,频数呈“陀螺型”分布;形容词的个数与频数均呈“金字塔型”分布。并且,不同等级词语所带有的原型特征也不同:词语等级越高,所带特征就越多,特征的权重也越大。2)搭配和语篇层面既存在静态等级,又存在动态等级。生命度等级之所以会发生动态变化,主要是由于转喻、隐喻、限定、及多等级性这四种作用机制。有九种搭配形式经常发生动态变化,它们比其他搭配形式具有更强的动态性。不同语体语篇的生命度等级和动态性也不相同:其中小说语体的生命度最高,动态性最弱;而学术语体的生命度最低,动态性最强。本研究结果可以广泛应用于文本分类、自动解岐、机器翻译、转隐喻识别、语义错误识别、中介语分析等多个领域。

【Abstract】 As one of the most important linguistic categories, animacy has cross-language effects on many linguistic phenomena, which has attracted a lot of discussions from various linguistic disciplines. However, previous studies have not paid much attention to the essence of animacy. The animacy hierarchy proposed therein is limited to nouns and pronouns, and lack of systematic analyses and supportive data. Hence, this research is intended to explore the theoretical issues of animacy, based on which, to propose a feasible analytical method of animacy hierarchy, and then to expand its researching area by using authentic data.This research proposes that, animacy has empirical, prototypical, universal and dynamic properties. Animacy in essence is a hierarchical category with human beings as the prototype. The perception of animacy is human-centered cognition from human’s own perspective. This cognition is embodied and rooted in evolution, therefore has cross-cultural universality. The animacy could be dynamic in different culture or context. This research has further extracted eight prototype features and their weights, which serves as an effective analytical frame for animacy hierarchy.Based on this viewpoint, this research expands the concept of animacy from two dimensions. On the horizontal dimension, the animacy hierarchy is extended from nouns and pronouns to verbs and adjectives, and then from the lexical level onto the collocational and textual level. On the vertical dimension, the animacy hierarchy is extended from static hierarchy to dynamic hierarchy. This research is based on the BNC corpus and Crown/CLOB corpora, which exhibits a high reliability and representativeness.This research yields the follow findings:1) There are static animacy hierarchies on the lexical level. Different word classes have different hierarchical distributions: nouns have the "sandglass" distribution in type and "butterfly" distribution in token; verbs have the "olive" distribution in type and "whipping top" distribution in token; adjectives have the "pyramid" distribution in both type and token. Besides, different hierarchies have different prototype features:the higher on the hierarchy, the more features it bears, and more weights of its features.2) There are both static and dynamic hierarchies on the collocational and textual level. The dynamic property of the animacy hierarchy is due to the following four mechanisms:metonymy, metaphor, determiners, and multi-hierarchic nature. There are nine collocations which have higher dynamicity, therefore more likely to have the dynamic animacy. Different genres have different animacy hierarchy and dynamicity:Fictional texts have the highest animacy and lowest dynamicity, while academic texts have the lowest animacy and highest dynamicity.The findings of this research can be widely applied to the following disciplines: text classification, automatic parsing, machine translation, metonymy/metaphor identifation, sematic error identifation, and interlanguage analysis, etc.

  • 【分类号】H314
  • 【下载频次】250
节点文献中: 

本文链接的文献网络图示:

本文的引文网络