Ceramic style change and neutral evolution: a case study from Neolithic Europe

Following on the work of Dunnell, the evolutionary archaeology school has made a sharp distinction between functional and stylistic variation in archaeological artifacts. Variation is defined as functional if it is affected by selection processes and as stylistic if it is a result of processes of random drift. The argument has been further developed by Neiman (1995), who showed by simulation that processes of cultural mutation and drift could produce the kinds of battleship curves that generally characterize artifact-style frequency distributions through time, and also demonstrated that they could account for patterns of stylistic variation through time in Woodland-period ceramic assemblages from Illinois. In this paper we present a case study of change in the decoration of pottery from early Neolithic Central Europe. We show that the actual diachronic frequency distributions and those expected under the neutral model do not coincide and conclude that in this case the neutral model does not provide an adequate description of change in ceramic decoration. A model involving selection, in the form of a bias in favor of novelty in the later phases of the period studied, seems likely to be more appropriate, and we note the social interpretation of the original investigator of the data. In conclusion, it is suggested that neutral models provide an important heuristic tool but that there is not a radical break between functional and stylistic variation.