【作者】 刘乃实；

【导师】 熊学亮；

【作者基本信息】 复旦大学 ， 英语语言文学， 2005， 博士

【摘要】 幽默研究的历史源远流长,甚至可追溯到现代文明的萌芽时期。作为一种可资研究的学科题材,幽默现象受到了来自于传统语文学、文学、社会学、心理学、语言学和文化研究等领域的广泛关注。在传统的幽默研究著作中,代表性的成果主要是乖讹理论、优越理论和释放理论。 在现代意义上真正定位于语言学框架内的言语幽默研究则以Raskin(1985)的幽默语义脚本理论(SSTH)和Attardo&Raskin’s(1991)的言语幽默一般理论(GTVH)为典型代表。SSTH借助来自心理研究的“脚本”这一概念,发展了原有的乖讹与解讹理论,并将脚本的重合与对立确立为文本产生幽默性的充分和必要条件。这也是语义理论第一次系统地应用于幽默研究。随后的GTVH是对SSTH的进一步修正和发展,其核心观点是幽默认知系统中的六个知识资源(KR):脚本对立(SO)、逻辑机制(LM)、情景(SI)、目标(TA)、叙事策略(NS)和语言(LA)。这套机制有助于刻画不同笑话间的相似程度,并拓展了幽默研究的范围。 但SSTH和GTVH都未能充分揭示幽默解读的本质,而这正是本文的立足点。具体来说,本项研究依然在语言学的范围内展开论述,宗旨是对幽默文本的本质要素及其解读进行形式化的处理。在研究方法上,采用的是Attardo等人(2002)引入幽默研究的图论工具。不过先前的研究还处于从脚本中心模式到集合论与图论模式的过渡阶段,本文构建的模型——笑话的简式图论模型(RGMJ)——则更充分地揭示了如何将图论工具应用于幽默研究。 在一定意义上,RGMJ是对Attardo等人(2002)前期成果的发展,但在形式化的细节和应用的范围上都取得了突破。本文的主体部分就是关于对RGMJ的具体阐述和应用,必须明确的是,RGMJ的研究对象不是一般意义上的言语幽默,而是篇幅较短、较容易操作的笑话。 RGMJ的整体框架是依照离散数学中关于图的标准要求建构起来的。其组成部分包括顶点、边和子图。其中顶点并不是图的基本元素,它们有自身的内部结构并表达为一组属性集。属性指的则是笑话中直接导致幽默意义的要素,包括相反相成的属性对(配对属性)和提供幽默文本成立基础的常项属性。属性以上的结构有顶点、边和子图:顶点用文本中明示或暗示的实体标示;边用来连接顶点,刻画的是笑话解读中的推理机制;子图则是图结构的子集,对应的是幽默意义建构中涉及的心理解读空间。RGMJ的范式图是一个标准的图结构,包括两对配对属性、四个顶点、四条边和两个子图。 在阐明图结构的属性、顶点、边和子图这些必要部件的基础上,本文作者进更多还原

【Abstract】 The research of humor has a long-dated history, which has attracted much academic interest in the field of philology, literature, sociology, psychology, linguistics, and culture studies. The traditional works on humor have converged in the representative schools of incongruity, superiority, and release theories.The more recent linguistic studies of verbal humor are represented by Raskin’s (1985) Semantic Script Theory of Humor (SSTH) and Attardo & Raskin’s (1991) General Theory of Verbal Humor (GTVH). Drawing on the notion of script, SSTH develops on the incongruity-resolution theory by identifying the mechanisms of script overlap and script oppositeness to account for the necessary and sufficient conditions for a text to be funny. It is also a systematic application of semantic theory to the study of humor. As a revised version of SSTH, GTVH extends the research purview to the general body of verbal humor by proposing a hierarchy of six Knowledge Resources (KRs): script opposition (SO), logical mechanism (LM), situation (SI), target (TA), narrative strategy (NS) and language (LA). A careful distinction of them and their relations serves to formulate the degree of joke similarity.But neither SSTH nor GTVH can adequately address the issue of humor interpretation, a challenge that is taken by this paper. In practice, the current research follows the linguistic tradition of humor representation and attempts to formalize the essential features of humorous texts and their interpretations. Methodologically, it adopts the useful tool introduced by Attardo et al. (2002) into the study of humor—graph theory. While Attardo et al. represent a transitional effort from the script-based tradition to a set-theoretic or graph-theoretic account of verbal humor, the model built in this work—a Reduced Graph Model of Jokes (RGMJ)—has more adequately revealed to what extent graph theory can be incorporated into humor research.In this sense RGMJ is an outgrowth of Attardo et al. (2002)’s precursory graph-theoretic account on the basis of SSTH and GTVH, but is much refined in the formal details and much strengthened in its applicability. It is thus to the exposition and application of RGMJ that the main body of this dissertation is devoted. But unlike GTVH, the model of RGMJ does not seek to represent verbal humor in general. Rather it is restricted to dealing with the more manageable jokes.The general architecture of RGMJ is designed according to the canonical sense of更多还原