Seriation Method and Its Evaluation

Abstract

Seriation as a scaling technique produces a formal arrangement of units, the significance of which must be inferred. Arrangement per se is a statistical matter, while the inference of significance is archaeological method. Here seriation as an archaeological method for inferring relative chronology is reviewed in terms of its assumptions and the conditions under which it is applicable. From this examination it is concluded that seriations may be inferred to be chronologies when and only when: (1) the comparisons are conducted using historical classes; (2) the units ordered are of comparable duration; (3) the units ordered are from the same cultural tradition; and (4) when the order is repeated through several independent seriations. The means of assessing whether or not a given seriation meets these conditions is considered in detail. Within specifiable limits seriations can be inferred to be chronologies, but these limits are more restricted than generally appreciated.