In recent years the field of generative social science has grown among epidemiologists, sociologists, economists and archaeologists as they seek to grow systems from the bottom up, by which they can explore problems and predict outcomes. Programmes such as NetLogo allow the researcher to design simulations in which they can build a virtual world inhabited by heterogeneous agents interacting with each other and their environment. For the archaeologist, we are faced with hypothesising past people’s behaviours and activities in landscapes which no longer exist as they were, and such a tool can help us to tease out the outcomes of our hypotheses. This paper will discuss a simple model in which agents attempt to negotiate a wetland landscape by assigning behaviours in which the agents respond not only to the environment they encounter, but also to the behaviour of previous agents. A well-designed model should agree with the physical evidence of trackways through our wetlands, after which they could be applied to landscapes for which we lack archaeological evidence. The real value of such a method however is not necessarily in the output. The process of programming behaviour forces the researcher to assign realistic decision-making to the agents which is a vital component of our interpretation of the past. By being cognisant of the individual’s role in the past, we can start to explain the archaeological landscape as an emergent phenomenon.

Location

Norcroft Centre, University of Bradford