【作者】 田慧芳；

【导师】 解秀琴；

【作者基本信息】 西南交通大学 ， 外国语言学及应用语言学， 2006， 硕士

【摘要】 对比性语篇标识语是标识语言片段之间对比关系的语言表达式。尽管对比性语篇标识语是语言学学术论文中常用的语言形式之一，但段首句中的对比性语篇标识语还需进一步探讨。本文基于语料库技术，量化统计一组公认的对比性语篇标识语在语言学学术论文段首句中的分布特点及功能。 对比性语篇标识语根据其在段首句中所处的位置分为段首句句首、句中和句末三类，其功能根据Sehiffrin的功能观分为：语义结构和人际功能；根据对比源的位置分为：段首句句内对比和段际对比。Schiffrin认为对比性语篇标识语有基本语义，表示对比关系，体现于联接对象之间的语义和人际功能中。Fraser认为，对比性语篇标识语没有语义内容，只是标识以此引导的后一语言片段是对前一语言片段的否定或对比。本研究忽略前后语言片段否定与对比之间的细微差异而表示为对比关系，并赋Fraser的后一语言片段为对比本体，前一语言片段为对比对象，即对比源。故，Fraser所谓前后语言片段之间的对比关系则转化为对比本体和对比对象之间的对比关系。当对比对象位于段首句内时，对比性语篇标识语标识段首句句内对比；当对比对象位于前一段或前几段落时，对比性语篇标识语标识段际对比。 本研究属实证性研究范畴，采用语料库语言学技术，集定性、定量分析为一体。定性分析以量化数据为基础，量化数据为定性分析提供佐证。根据本文研究，得出结论如下： Ⅰ．对比性语篇标识语(简称标识语)在语言学学术论文段首句中的分布特点： 1．分布不均：句中最多，占总体的68.99％，其次为句首，30.14％，句末最少，仅0.87％。当显著性水平接近0时，这种差异在统计学上具有显著性； 2．每一标识语在段首句各位置上的使用趋势是不均等的，如but，常用于句首、句中，鲜见于句末。 3．段首句句首出现词频最高的前五个标识语(由高到低)分别有but、although、however、yet和contrast；句中的则为but、rather、更多还原

【Abstract】 Contrastive Discourse Markers (herein DMs) refer to lexical expressions that signal contrastive relations between linguistic segments in a written discourse. The units of analysis for the thesis are the occurrences of selected putative contrastive DMs co-occurring in the paragraph-initial sentences (herein PIS) of linguistic papers. The departure point of the current research is to identify and characterize the physical positions of occurrences in question into the PIS-initials, -medials, and -finals, and then populate their frequencies in the context. After that, their functions are reclassified by hand into idea structures and interpersonal functions; intra-PIS contrast displayers and paragraphic contrast displayers.The distributional features of occurrences in question are theoretically based on the observations and other researches. When an occurrence introduces a PIS, it is categorized into the PIS-initial group; when an occurrence ends a PIS, the PIS-final group; otherwise the PIS-medial group.The functional theoretical orientations in the thesis are mainly directed to two aspects: Schiffrin’s (1987) functional model and positions of contrast counterparts which are based on Fraser’s contrast or denial displayer model (1998, 1996c) of contrastive DMs. Both are conducted independently. Among them, Fraser’s the "following" segment linked by a contrastive DM is termed as a contrast body, the "preceding" segment conjoined as a contrast counterpart saying the source of contrast. Hence the contrast relationship between the "following" segment and the "preceding" is converted into a parallel relationship between the contrast body and the counterpart.The current research is of empirical in nature. The method of corpus techniques is adopted, with the qualitative method and the quantitative method. In light of statistical results, the findings of the thesis are summarized as follows:i.Positional distribution of contrastive DMs in the PISs in our data:1.Contrastive DMs are not evenly distributed in the PISs, with the PIS-medials more than the PIS-initials, and more than the PIS-finals.更多还原