The Inclusion of Ethnographic Data And Controlling for Political Bias Leads to Robust Modeling in Archaeology

Summary

There have been multiple advances in recent approaches to modeling within archaeology. The power of advanced computational techniques including agent-based modeling, Bayesian approaches, etc., have enabled archaeologists to hypothesize and describe complex multi-scalar processes affecting past societies, while paying heed to multiplicity of variable factors. However, while anthropological archaeologists reject models within economics and political science as "data-poor," recent archaeological modeling also stands accused of similar neglect and increasing reliance on assumptions of past behaviors, that are often driven by the social and political training of the modelers. These can affect both the selection of necessary/sufficient data and variables, and the assumptions made regarding the relationships between variables. I suggest that the careful inclusion of ethnographic data while controlling for political and social bias confers robustness to modeling approaches. To showcase this, I combine ethnographic and archaeological data, network dynamics, and engineering models to describe and test the evolution of trading behaviors, as traders respond to changing social and political regulation. Cross-cultural ethnographic data on traders (n=576) and archaeological data from Asian and African port-cities is used to model and test global trade interactions in the Indian Ocean, ca. 1000 -1900 CE.

This Resource is Part of the Following Collections

• Society for American Archaeology 81st Annual Meeting, Orlando, FL (2016)   •
• Computational approaches in the practice of archaeology

Keywords

Spatial Coverage