Abstract

This research deals with the study of current pastoral seasonal settlement patterns in the uplands of Val di Fiemme (Trentino province), in order to create a quantitative locational model for predicting archaeological pastoral site locations. In fact, few archaeological sites related to pastoral economy are known in the Alps, and this lack of data affects the interpretation of the ancient pastoral strategies. A predictive model could be useful to identify new sites and to optimize archaeological surveys in mountain environments. However, inductive predictive modelling is considered a field with many unresolved theoretical problems. Ethnoarchaeology of pastoralism seems to be a good method to provide a behavioural framework for predictive modelling, and the interaction with quantitative approaches may be worthwhile to improve the ethnoarchaeological methods and theory as well. The ethnoarcheological research is divided into two parts: a “desk ethnoarchaeology”, the spatial analysis of the relationship between current pastoral/dairying sites (malghe) and mountain environment of Val di Fiemme, in order to create an inductive predictive model; and a “field ethnoarchaeology”, the study of the relationship between modern shepherds/dairymen and environment, in order to understand their locational strategies. The interaction between the first and the second part has enabled the interpretation of the settlement pattern of modern malghe in Val di Fiemme. It has hence been assumed that the model can predict the location of modern and ancient dairying sites, as the malghe are mainly related to milking, milk-processing and cheese storing activities. The final step has been the archaeological evaluation. The predictive model doesn’t predict the location of Mesolithic hunting sites in the uplands of Val di Fiemme. It predicts instead the location of some dry-stone enclosures in the upland valleys of Ortisé (Mezzana, Val di Sole, TN), while it doesn’t predict the location of rock-shelters in the same area. It has therefore been assumed that enclosures were linked to dairying economy and rock-shelters to simple (“dry”) pastoralism. Further qualitative tests of the model have been carried out in different areas of the Alpine arc. These results allow two different application of the model: a “predictive” application, aimed at finding new pastoral sites in the uplands, and an “interpretative” application, aimed at discriminating hunting sites and simple pastoral sites from dairying sites. Furthermore, this model has suggested that the interaction between predictive modelling and ethnoarchaeology is useful to tackle the theoretical and methodological problems of these fields of research.