中文摘要：

本文借鉴生成词库论关于物性结构的观念,提出亲属名词的社会结构的概念,从而形成一套对汉语亲属名词进行语义描写的形式化体系。这个体系包括亲属名词的词条说明、社会角色、社会结构（即语义结构）和它所激活的情境网络四个方面。文章指出,亲属名词的词典释义应尽可能反映促动亲属名词语义结构的形成的概念结构。这个概念结构的主体是认知主体识别亲属词的所指时搜索指针所经历的搜索路径。它是一种以各种亲属关系作基底,以称呼角色、受呼角色和中介角色等社会角色作侧面,以受呼角色作焦点的认知图式。亲属名词的词典释义还应尽可能反映出相关社会角色在性别、年龄等属性参数上的取值,基元性亲属名词的释义应尽量避免同义互释或转释。计算机编程实验表明,本体系对于计算机模拟人脑进行亲属关系的模板匹配,从而计算和推导亲属关系具有直接的应用价值。

英文摘要：

Enlightened by the conception of qualia structure from the Generative Lexicon Theory, we propose in this paper the conception of social structure of a kinship term, which leads to a formal description scheme of the semantics of Chinese kinship terms. This scheme contains four parts： 1） explanation for each lexical entry, 2） social roles, 3） a social or semantic structure, and 4） the situation network activated by every kinship term. This paper points out that the paraphrase of a kinship term should reflect as much as possible the conceptual structure which motivates its semantic structure. The conceptual structure is a search path covered by search probe when a cognitive subject identifies the kinship term’ s denotation. It is a cognitive schema that takes a variety of kindred as its base, many a social role like addressor role, addressee role, and bridge role as profile, and addressee role as focus. The paraphrase of kinship terms should also reflect as much as possible the correlative role’ s parameter value on sex, age and so on. For the basic kinship terms, lexicographers should avoid mutual paraphrase or circular paraphrase to the best of their ability. Computer’ s programming test shows that this scheme has direct application value for computers to model human mind’ s computation and inference of kinship relation by means of template matching.