论文部分内容阅读
通过统计或计算方法来实现高层次语义表征的量化和质化处理是现有语篇研究的发展趋势之一。现有非在线语篇研究大多以“命题数量”和语篇“表层表征”为基础来实现语篇表征的量化处理,具有一定的局限性。本文以认知语义学框架下的“认知常量”和“语篇表征”理论为基础,以命题密度计算为基本原则,系统探究了语篇理解过程中从微观表征到宏观表征等不同级别语义表征的量化处理方法,尤其重点阐述了“因果单元”和“因果链”的量化方法。此方法将语篇局部连贯和整体连贯纳入量化范畴,其结果更吻合认知主体的语篇理解过程。
It is one of the trends in the existing discourse research to realize the quantification and qualitative processing of high-level semantic representation by means of statistics or calculation. Most of the existing non-online discourse studies have quantified discourse representation based on the number of “propositions” and discourse “superficial characterization”, which has some limitations. Based on the theory of “cognitive constant” and “discourse representation” in the framework of cognitive semantics, this dissertation takes the proposition density calculation as the basic principle and explores systematically the process from microscopic representation to macroscopic representation And other different levels of semantic representation of the quantitative treatment methods, with particular emphasis on “causal unit ” and “causal chain ” quantification method. This method integrates the textual coherence and overall coherence into the quantitative category, and the result is more in line with the cognitive process of discourse comprehension.