Modelling Temporal Uncertainty in Archaeological Analysis

This paper discusses the problem of temporal uncertainty in archaeological analysis and how it affects archaeological interpretation. A probabilistic method is proposed as a potential solution for modelling and quantifying time when high levels of uncertainty restricts temporal knowledge and scientific datings are unavailable, while Monte Carlo simulation is suggested as a means to formally integrate such knowledge into actual analysis. A case study focusing on counts of prehistoric hunter-gatherer pithouses in Mid-Holocene Japan provides an example of how uncertainty can be problematic and bias the results of the most straightforward archaeological analysis and how the coupling of a probabilistic and simulation-based approach nonetheless offers a useful solution. The discussion that follows also addresses the need for more robust and quantifiable ways to illustrate the chronological flow of our archaeological narratives.