Archaeological Simulation and the Testing Paradigm

Simulations in archaeological research make use of a wide range of techniques and methods. As a general rule, they tend toward early adoption of new computational approaches (such as agent-based modeling and multi-agent simulation), evaluative statistical tools (such as sensitivity or uncertainty analyses), and borrow heavily from similar applications in other disciplines; particularly ones that emphasize the dynamic (or dynamical) group behavior of biological organisms. As these approaches become more sophisticated our theoretical bases for testing, and ultimately refuting or validating their implications are falling behind. Largely this is due to the nature of the assumptions we make about the testing of archaeological hypotheses in general, and our expectations about causality, not merely the oversimplification of past human decision-making by the simulations themselves. We have very sophisticated computational devices for simulating the past in many unique and interesting ways, but there are some basic underlying issues that clearly indicate the problems with archaeological simulation are primarily theoretical and not mathematical.