The data of semantic description

Paradigmatic meaning

The application of componential analysis, and other forms of investigation, by anthropologists to folk terminologies has given us new perspectives for further semantic analysis. However, if less structured areas of the lexicon of a language are to be attacked, the restriction that a domain of vocabulary be clearly delimitable before analysis can proceed must be lifted, and it must be considered possible to enter the continuous semantic system at any arbitrary point. It is as a working hypothesis that items so extracted from the system are treated as though they formed a naturally isolated subset. Their definitions would be made more precise as more items in the system are studied and contrasted with those already extracted.

Like componential analysis, then, the present approach is structural in the paradigmatic sense of the word. The meanings of the items in a language are presented as standing in opposition to one another within the system of the language and as being distinguished by discrete semantic components acting as the distinctive features. This is not meant to imply a definition of the term meaning, which has often been defined so broadly as to make anything but anecdotal semantic investigation seem impossible, nor are all other theories denigrated. Rather, one approach is offered with which an empirical investigator can hopefully work. It restricts itself to minimal definitions (definitive meaning, defining or criterial attributes). A minimal definition of the meaning of an item will be a statement of the semantic components necessary and sufficient to distinguish the meaning paradigmatically from the meanings of all other items in the language.

¹ This article was originally solicited as a condensation of Bendix 1966. It is perhaps inevitable that it has become somewhat more than that. Besides editorial and substantive reformulations, some account has been taken of developments since February 1965, when the original monograph was completed, and in its present form it could probably fit as suitably. After it was submitted, Fillmore (1969) appeared with a good critique of Bendix (1966), but his points could no longer be answered here.