Identifying Ancient Mesoamerican Fortifications with a Bayesian Network Model

Summary

This is an abstract from the "Theorizing Warfare: Global Perspectives on Defense and Fortification" session, at the 90th annual meeting of the Society for American Archaeology.

Identifying fortifications in the archaeological record can lead to inferences of important elements of past human behavior and social evolution. However, identifying ancient fortifications can often be difficult and contentious. One solution is to avoid a binary approach to identification and instead utilize a Bayesian approach with a probabilistic model which quantifies the confidence that a specific site is a fortification. To test this approach, I reassessed the identification of fortified and defensible sites in the Valley of Oaxaca in southern Mexico. Such sites can be indicators of warfare, boundaries, and changes in political systems, such as the emergence, expansion, and collapse of the state. Correctly identifying them is essential. Building on pioneering work by Steve Kowalewski, J. Michael Elam, and others, this Bayesian approach confirmed many of the original identifications of fortified and defensible sites, while questioning others. This approach can be fruitful for archaeologists attempting to identify fortifications on landscapes in any region of the world.

This Resource is Part of the Following Collections

• Society for American Archaeology 90th Annual Meeting, Denver, CO (2025)   •
• Theorizing Warfare: Global Perspectives on Defense and Fortification

Keywords

Record Identifiers

Abstract Id(s): 50724