The Human Niche and the Rise of Capitalism: applying Bayesian machine learning to causality modelling in historical social-ecological systems

With the advent of machine learning and big data, new opportunities are emerging for the quest for history's Holy Grail: elucidating causality, and, ultimately, developing a predictive science of the past. In my talk, I will present a new algorithm developed in MPI Geoanthropology, Bayesian Model Selection under Constraints (BMSC), designed to help in exploring and evaluating past causal relationships based on historical, archaeological and palaeoenvironmental data. I will discuss two case studies, both focused on understanding the interplay of environmental (physical, climatic, epidemic, etc.), economic and cultural factors as drivers of human niche construction strategies in Greece (Peloponnese and Macedonia) and Poland (Pomerania).