漢語量詞“張”的去范疇化研究
[Abstract]:Quantifier is a special part of speech in Chinese and Tibetan languages, which is active in people's daily communication and literary works with its vividness and vividness. Generally, quantifiers follow numerals to form quantitative structures to modify nouns to express quantitative meanings. However, in some cases, quantifiers appear non-quantifier expression, no longer have the characteristics of typical quantifiers. The theory of de-categorization is a dynamic study of unconventional phenomena in language. The non-classifier expression of classifier is the result of decategorization. Previous researches on quantifiers are mainly static description, but little attention has been paid to the dynamic development of quantifiers. As we all know, languages are constantly changing, so it is not enough to focus on their static features and regular usage. However, for the theory of de-categorization, most scholars apply it to the study of the phenomenon of decategorization of nouns and verbs, and few scholars combine it with the study of quantifiers. Therefore, this paper applies the theory of de-categorization to the study of classifiers, and tries to systematically analyze the syntactic distribution characteristics, functional changes and semantic changes of quantifiers after the phenomenon of de-categorization, in order to have a comprehensive understanding of the dynamic development of quantifiers. The quantifier "Zhang" often appears in literary works and daily communication, and it is representative and convenient to study in the corpus. Therefore, this paper takes the quantifier "Zhang" as the object of study to investigate the phenomenon of de-categorization of the quantifier "Zhang". All the entries with "Zhang" in the corpus are used as the corpus data in this study. The author finds out the non-quantifier usage of quantifiers by processing the corpus and marking the parts of speech, and combines qualitative analysis and quantitative analysis to analyze the syntactic distribution of classifiers, respectively. The functional and semantic aspects of the detailed study. With the help of de-categorization and its related theories, this study finds that the classifier "Zhang" is different in the process of de-categorization. In some cases, the quantifier is completely de-categorized into a new category, while in some cases, the quantifier is still in the middle state of de-categorization, that is, it does not completely lose the original category characteristics but has some attributes of other categories. Secondly, in the syntactic distribution, the prototype syntactic distribution features of the classifier "Zhang" can be expanded to form a new noun with the noun morpheme, or the verb can be modified with the numeral "one" and reduplicated. They can also be followed by prepositions to define verbs and form a new adjective with adjective morphemes. Thirdly, on the functional level, the author analyzes the syntactic function and semantic function. From the perspective of syntactic function, the decategorization quantifier "Zhang" is extended from the original attributive position to the position of affix, adverbial and object, and the attributive function is weakened after the adjective morpheme. In the position of affix, only has the grammatical function. In terms of semantic function, the classifier "Zhang" extends the function of statement and reference from the original function of non-declarative and non-referential. The way in which a verb acts when it has a declarative function similar to a nomenclature. Finally, in the semantic change, the quantifier "Zhang" will have semantic loss, semantic transfer and semantic abstraction generalization. The systematic analysis of syntactic distribution, function and semantics enables readers to have a more comprehensive understanding of the dynamic development of quantifiers and to better grasp the development and changes of language.
【學位授予單位】:貴州大學
【學位級別】:碩士
【學位授予年份】:2017
【分類號】:H146.2
【相似文獻】
相關(guān)期刊論文 前10條
1 宗守云;;范疇化理論與四個世界理論[J];畢節(jié)學院學報;2008年06期
2 于艷;陳全獻;;探究去范疇化的定義[J];科技信息;2009年02期
3 羅建平;;語言非范疇化理論簡述及討論[J];廣東石油化工學院學報;2011年02期
4 賈志高,程杰;隱喻語言中的范疇化[J];山東師大外國語學院學報;2000年02期
5 劉潤清,劉正光;名詞非范疇化的特征[J];語言教學與研究;2004年03期
6 陳建生;楊玉娟;;詞義的范疇化與非范疇化[J];長沙理工大學學報(社會科學版);2006年03期
7 張士奇;;模糊語言與原型范疇化理論[J];科技信息(科學教研);2008年14期
8 王莉;;限制性和非限制性關(guān)系分句使用規(guī)律的次范疇化研究[J];長沙理工大學學報(社會科學版);2010年03期
9 梅晴;;謅議語言的范疇化及其過程[J];讀與寫(教育教學刊);2010年07期
10 齊紅霞;謝勇;;名詞的功能游移是名詞非范疇化的結(jié)果[J];蘭州教育學院學報;2010年05期
相關(guān)會議論文 前10條
1 宗守云;;范疇化理論與四個世界理論[A];走向科學大道的修辭學[C];2010年
2 黃偉紅;;類典型論,范疇化和詞義擴展[A];第四屆全國認知語言學研討會論文摘要匯編[C];2006年
3 陸紅艷;;非范疇化視角下的多義性分析[A];首屆海峽兩岸外語教學與研究學術(shù)研討會暨福建省外國語文學會2011年會論文集[C];2011年
4 韓習武;王金勇;;漢英動詞次范疇化對應關(guān)系的形式描寫[A];黑龍江省計算機學會2009年學術(shù)交流年會論文集[C];2010年
5 韓習武;劉春陽;;漢語動詞次范疇化的形式描寫[A];黑龍江省計算機學會2009年學術(shù)交流年會論文集[C];2010年
6 ;A Boundary based Classifier Combination Method[A];2009中國控制與決策會議論文集(3)[C];2009年
7 程飛雄;李衛(wèi)華;唐峗;;In silPrediction of Cytochrome P450 Inhibition and Inhibitory Promiscuity:Towards an Understanding of Chemical Promiscuity[A];第十一屆全國計算(機)化學學術(shù)會議論文摘要集[C];2011年
8 ;Impact of the Size of Training Set on Text Categorization[A];第三屆學生計算語言學研討會論文集[C];2006年
9 ;A New Measure of Diversity for Classifier fusion Based on Fuzzy Integrals[A];2011年全國電子信息技術(shù)與應用學術(shù)會議論文集[C];2011年
10 Mok Hingtung;;Fault Diagnosis Based on Knowledge Extracted from Neurofuzzy Networks Using Binary and Real-Valued Fault Databases[A];第二十七屆中國控制會議論文集[C];2008年
相關(guān)博士學位論文 前3條
1 陳勇;漢語數(shù)量范疇及其非范疇化研究[D];暨南大學;2011年
2 陳偉;范疇化理論中幾個問題的研究[D];華南理工大學;2014年
3 HAZRAT ALI;基于深度學習模型的說話人識別算法研究[D];北京科技大學;2015年
相關(guān)碩士學位論文 前10條
1 何梅;漢語量詞“張”的去范疇化研究[D];貴州大學;2017年
2 劉昭;《漢俄大詞典》單字條目詞類標注實證研究[D];四川外國語大學;2015年
3 郭建強;范疇化理論在高中英語詞匯教學中的實證研究[D];遼寧師范大學;2015年
4 張海寧;范疇化理論在高中英語詞匯教學中的實證研究[D];沈陽師范大學;2016年
5 沈麗雙;中國日報中名詞非范疇化現(xiàn)象的研究[D];沈陽師范大學;2016年
6 趙璐;《論語》中“仁”的認知語言學研究[D];西南大學;2016年
7 韓秋靜;基于范疇化理論的初中英語詞匯教學研究[D];青海師范大學;2016年
8 張麗晶;基于基本范疇化理論的教學方法對初中生英語詞匯習得影響研究[D];揚州大學;2016年
9 崔璨;基于范疇化理論的高中英語詞匯教學研究[D];延安大學;2016年
10 宋琦;“七A八B”的非范疇化數(shù)詞分析[D];長沙理工大學;2015年
,本文編號:2345133
本文鏈接:http://sikaile.net/wenyilunwen/yuyanyishu/2345133.html