The Use of Bayesian Statistics for 14C Dates of Chronologically Ordered Samples: A Critical Analysis

Abstract

Bayesian mathematics provides a tool for combining radiocarbon dating results on findings from an archaeological context with independent archaeological information such as the chronological order, which may be inferred from stratigraphy. The goal is to arrive at both a more precise and a more accurate date. However, by means of simulated measurements we will show that specific assumptions about prior probabilities—implemented in calibration programs and not evident to the user—may create artifacts. This may result in dates with higher precision but lower accuracy, and which are no longer in agreement with the true ages of the findings.