There are three fundamental ways in which we may describe the domain of referential meaning of individual lexical units. We may (1) name the class, usually by contrasting it with other classes, (2) identify the individual members of the class, generally by listing, or (3) describe the distinctive features of the class. The first meaning may be called a "tyрe" (Morris, designatum ) or generalized meaning. The second "definition" is based upon the listing of tokens of the class (Morris' denotatum), and may also be called the particular meaning. The third type of meaning is based upon "sufficient and necessary features of the class", (Morris' significatum), or, in other words, the abstract meaning of the class. For example, we may define chair, first by contrasting the term chair, as the name of a class, from other possibly related classes, e.g. bench, stool, position, post. Secondly, we may define chair by listing the specific tokens of the class, e.g. the chair in the living room, the chair of philosophy, the electric chair, he will chair the meeting. In a sense, such tokens are almost unlimited, for such referents of chair are extensive. Thirdly, we may define chair by describing the sufficient and necessary features which distinguish it from all other objects. For example, chair in at least certain aspects of its meaning is described in The American College Dictionary (1947) as "a seat with a back and legs or other support, often with arms, usually for only one person." The American College Dictionary then goes on to list other meahings based upon tokens, not upon "sufficient and necessary features," for many symbols cover such a wide area of meaning that there is not a single cluster of defining features, but chains of such features.
If we apply the three-way distinction between class, token, and features of the class to the above analysis of certain core words in the English kinship system, we find that the tokens (or denotata) are the individual persons identified by such abbreviations as FaFa, MoFa, FaBoSo, MoFaDa, etc. The classes are the groups identified by such terms as father, grandfather, uncle, etc.. including all the tokens which belong to any such classes (the designata). The distinguishing features of the classes (the significata) are the componential features of generation, sex, and lineality. The significata are, of course, the prerequisites of any class, but it has been noted the meanings of words are not limited to these prerequisites, for there аre also other features, which may be called the probabilities or possibilities of the class. We may say that in certain usages the terms father, dad , daddy, pop, and old man all have the same significata, and that to this extent the denotata are identical. Thus the same referent may be identified by all these terms. But these words certainly have different emotive meanings, i.e. features which as possibilities and probabilities are far more "fluid" and difficult to describe than are the prerequisites involved in the significata.
In order to describe the domains of referential meaning of individual lexical units, whether in terms of type, token, or features, there have traditionally been three kinds of techniques which have yielded the most helpful results:(1) derivational, (2) componential, and (3) distributional. Derivational analysis tends to concentrate on relating tokens to each other within the class. Componential procedures highlight the necessary and sufficient fcatures of the class, and distributional techniques employ the elements of the linguistic context to provide clues to basic divisions of semantic domains.
责任编辑：admin
上一篇：DRYDEN'S IO: A BEAUTIFUL TRANSLATION?
下一篇：VARIOUS SCIENTIFIC APPROACHES TO MEANING
微信公众号搜索“译员”关注我们，每天为您推送翻译理论和技巧，外语学习及翻译招聘信息。