(Part I) Location analyses of archaeological site and settlement distributions at the regional level have been undertaken for at least a half century. In regions throughout the globe archaeologists have analysed these distributions for relationships, associations or correlations with aspects of the physical, and sometimes social, environments of regions. A host of approaches and methods have been employed ranging from goodness of fit tests for categorical and quantitative data, two-sample tests between groups indicating site presence and absence, regressions between aggregated site counts and regional variables, and many other techniques. Much related are archaeological location modelling (ALM) studies that frequently combine findings from location analyses using multivariate discriminant functions, logistic regressions and other methods to model the empirical pattern of settlement distributions. Such studies often suggest the significance of predictor variables related to positive archaeological responses. Moreover, through GIS the resultant functions may be mapped region-wide to portray graphically the “essence” of a spatial pattern of land use or settlement through a region in the form of archaeological likelihood. Recently, agent-based modelling approaches have also generated new insights into site and settlement distributions across landscapes. If we accept that a goal of archaeology is building theory that helps explain the past (as well as present and future), then the development of theories of location choice or “settlement theory” must surely form a central one. Many significant relationships with environment (whether physical or social) have been demonstrated in countless archaeological location analysis and modelling projects. The dictum “theory begins with facts” implies that findings from these studies should yield necessary information from which theory might be generated, assuming pattern and replicability. Yet, a frequent complaint about ALM in particular is the general lack of theoretical insights it has offered. With the volume of results in settlement location analysis and modelling through the decades surely the time is ripe to consider whether such efforts are warranted with respect to theory and the nature of theoretical insights that might be possible? This session includes papers that (1) describe methods that demonstrate linkages or associations between sites, settlements, or other uses of landscapes and features of a region, and (2) are able to form generalizations describing regularities in location or other behaviours that offer theoretical insights or lead to outright development of theories of location.
