Cognition

There is a continuing debate about whether semantic analyses, especially componential analyses in anthropology, can be models of the cognition of speakers. Cognitive or psychological reality, however, belongs to a level of abstraction referring to individual behavior, whereas a semantic analysis of a language is on the higher level of abstraction of the superorganic or cultural emergent, of which language is a part. Like the investigator, a native speaker, during enculturation, constructs his model or set of rules on the basis of observed behavior, a model which may or may not accord with that of another speaker. The investigator uses certain methods in his analysis. For example, in accounting for various uses of a form, he may, from a principle of economy, avoid unnecessary polysemy (or homonymy) by trying to generalize to core meanings or lowest common semantic denominators commensurate with paradigmatic distinctiveness. To the extent that his analytic methods represent those used by native speakers, his process of analysis may be said to be a model of a more abstract level of speaker processes, but his resulting description may or may not be congruent with the set of rules actually arrived at by any one native speaker.