Four Codes for the Description of Artifacts: An Essay in Archeological Technique and Theory

Abstract

IN COMPARATIVE work in archeology, the investigator ought not to have to spend more time in assembling data than in analyzing them. Yet with the vast accumulation of publications in recent years, sheer bibliographical search (involving sources that not even all large libraries possess) demands prodigious effort. Important descriptive details are scattered in the text and in the illustrations, so that the student must return again and again to the source once he has located it. This process is repeated endlessly by different archeologists using the same materials; hence, the total duplication of effort is enormous. Moreover, description is insufficiently standardized both because of the variations in usage between languages and because of different conventions that prevail among archeologists. Points can be settled if illustrations are excellent and abundant, but the cost of publication makes it rare for both of these conditions to be met. Can one resolve-or partially resolve-these difficulties by the construction of standard categories that are relatively culture-free and by the use of mechanical aids such as systems of punch-card indices? This paper suggests certain steps toward the economical presentation and dissemination of artifact description. Actually, the first "code" of description which will be reported here (that for Bronze Age metal tools between the Balkans and the Indus Valley) is already in use and gives good results in practice. Four thousand punch cards, compiled by Jean Deshayes, include essentially all of the data available. The problem, however, goes far beyond the practically useful task of getting information on punch cards. The most basic issues of anthropological typology and categories must inevitably be raised. Here I have drawn-especially in the latter portion of the paper-upon contemporary linguistic theory. Let me begin with a sketch of the code for tools already published in French in greater detail (Gardin 1956; see also Gardin 1955, for some of the theoretical and methodological background). The aim of this and of all the codes is to reduce to small compass in standardized form stores of published and unpublished information. The decks of punch cards should be as publicly available as any book and as widely circulated as significant archeological monographs or general treatises. Having the data on punch cards will make it possible for archeologists to do sorts on an infinite variety of differences and similarities of features and complete artifacts, on constancy and variation in time and in space. My presentation builds upon a series of important technical and theoretical explorations by American archeologists (e.g., Black and Weer 1936; Ehrich 1950; Finkelstein 1937; Ford 1954; Gebhard 1946; Krieger 1944; Mc-