Publication date: Available online 17 May 2017Source:Journal of Archaeological Science
Author(s): Eric E. Jones
Over 30 years ago, Kintigh and Ammerman (1982) outlined and applied a heuristic approach to spatial archaeology that balanced quantitative analyses and culturally and historically contextualized archaeology. The theoretical and methodological messages were that we need to do more than “eyeball” spatial patterns, we need to apply the proper analyses based on the characteristics of our datasets, and we need to ensure that our models, quantitative analyses, and resulting interpretations are based in the proper cultural and historical contexts. My goal in this paper is to examine how two of the concepts in this approach, significance and context, apply to a modern spatial archaeology that heavily utilizes geospatial computing tools. Although these tools help to solve several concerns that existed in the field 30 years ago, they can also cause others, such as mistaking autocorrelation for correlation or confusion about which of the multitude of available analytical tools is appropriate for particular questions and datasets. In this paper, I present a simplified version of the methodology I have used to address these concerns. I use archaeological, historical, and GIS-modeled data to compare the regional patterning of hierarchical and egalitarian societies in southeastern North America to examine why hierarchical sociopolitical organizations may have arose where they did. I end with a critical review of this approach and a discussion of how such research can be improved moving forward.