.:: CAA 2010 -Fusion of Cultures- ::. Computer Applications in Archaeology, Granada (Spain) 6-9 April 2010

Round table

Since at least the early 1960s, researchers have sought to provide mathematical models to represent the movement of past peoples and ideas. Initially such models focused on "cultural" diffusion, where technology was assumed to move in space and time without significant population movement. Quite quickly, however, focus moved to demic flow. Such models were originally used to model population spread but, it soon became clear that if they can be used to model the movement of people, they must also have some utility in modelling the movement of cultural, genetic, and linguistic traits as they spread along with the people who "carry" them.

As soon as demic and cultural diffusion are linked in this way, there is potential for considerable sophistication in the modelling, but also for considerable confusion. The difficulty is that many cultural changes have been seen as coming as "packages" and so archaeologists have, for example, labelled particular types of stone tools, houses and agriculture as "Neolithic". If we do this, we are grouping together technologies and/or ideas that may or may not have spread in similar ways. This has lead to a very fuzzy and confusing picture of the spread of the Neolithic in Europe. Increasingly, however, archaeologists are aware that seeking to model the arrival of the "Neolithic" is not a sufficiently focused goal. It is now clear that we need to be much more specific and think, for example, about modelling the arrival of a particular gene (pool) or the arrival of cereal agriculture.

With this realisation has come increasingly sophisticated mathematical and statistical models. If such models are to become routine in archaeological research, however, we need to adopt a rigorous and very structured approach to their use. First, we have to define precisely what it is we wish to model. Second we have to develop appropriate models for representing the spread of the phenomenon of interest. Third we need to assemble data that reliably relate to that phenomenon. Fourth we have to develop methods for fitting the model that we have chosen to the available data, so that we can estimate crucial parameters like the rate of spread and the location of any hiatuses in space and time. Finally, we need a way to convey both the most likely values of the key parameters and their associated uncertainties.

In this discussion, we will focus on a selection of modern mathematical and statistical models. We will seek to draw out the strengths and weaknesses of each, to encourage discussants to focus on examples from their own research, and to identify research issues which require pressing attention if such models are to make a real difference to modern archaeological practice.

The discussants who have agreed to take part (schedules and funding permitting) are as follows.

• Prof David Anderson (University of Tennessee, Knoxville - This e-mail address is being protected from spambots. You need JavaScript enabled to view it ) is an archaeologist with a long-term interest in paleoindian colonization of the Americas; recently he has also been working on the link between climate change and cultural dynamics.
• Prof Caitlin Buck (University of Sheffield, UK - This e-mail address is being protected from spambots. You need JavaScript enabled to view it ) is an archaeostatistician who is currently working on the development Bayesian statistical models for the spread of cereal agriculture in Europe and applying them to a large database of radiocarbon-dated cereal grains.
• Dr Mark Collard (Simon Fraser University Canada - This e-mail address is being protected from spambots. You need JavaScript enabled to view it ) is Canada Research Chair in Human Evolutionary Studies and principal investigator of the Centre for the Evolution of Cultural Diversity at University College London. He is using methods and theory from evolutionary biology to investigate archaeologically- and ethnographically-documented patterns of material culture variation.
• Prof James Conolly (Trent University, Canada - This e-mail address is being protected from spambots. You need JavaScript enabled to view it ) is an archaeologist with particular interest in human palaeoecology and human adaptation; mobility and colonisation processes; origins and spread of farming and farmers; geographical information systems; spatial and analytical statistics; computer modelling of population dynamics and cultural change.
• Prof Keith Dobney (University of Aberdeen, UK - This e-mail address is being protected from spambots. You need JavaScript enabled to view it ) is a palaeoecologist whose principal research themes include the origins of agriculture, the domestication of animals, human and animal dispersal, diet and health, palaeopathology and palaeoeconomics.
• Dr Vincent Macaulay (University of Glasgow, UK - This e-mail address is being protected from spambots. You need JavaScript enabled to view it ) is a statistician who works on statistical inference of prehistoric demography on the basis of DNA sequence data, with a particular focus on mitochondrial DNA variation.
• Dr Graeme Sarson (Newcastle University, UK - This e-mail address is being protected from spambots. You need JavaScript enabled to view it ) is an applied mathematician who develops tailored models for representing movement of people in particular landscapes, for example, along river networks and has used these models to find evidence for multiple sources of the European Neolithic.

  Wednesday April 7th, 16.30h, Room Andalucia III