On a Possible Interpretation of Componential Analysis in Archeology

Abstract

Listen

Translate article

A. The more obvious remark is that the analysis deals primarily with terminological systems, or more correctly with sets of terms. The question whether such sets constitute a system has to remain open until the analysis is completed. The terms are selected on the basis of their reference to supposedly well-accepted categories of denotata: kinsmen, in the more frequent case,' or plants,2 physiography,3 personality attributes,4 etc. The resulting sets are in this respect similar to lexical domains or semantic fields in linguistics, and there is indeed a good case for considering that componential analysis should be related for that reason to linguistics, even if it aims at the design of structural models the interpretation of which belongs to anthropology. More specifically, by equating componential with semantic analysis or sememics to account for its structural goals, one perhaps conveys a better idea of its analogies with the object and methods of linguistics, both traditional5 and contemporary,6 and of its differences from other methods of formal analysis in anthropology, which imply no such reference to the naming, in natural language, of the phenomena under study. B. Once a given set of terms has been selected, the next step is to match it against a number of analytical features (usually named in the observer's language), which are taken to be, or at least to include, the meaningful components of a possible structural model for all lexical units within the set. C. The models derived through componential analysis are thus the product of a study in the interaction of two sets of data, one linguistic, the other nonlinguistic. This observation is undoubtedly trivial, but it will help us to establish the limits of that method, when considered within the framework of archeology. Before doing so, let us add a few comments on the three points that have