Understanding the distribution of archaeological objects
How representative is the data collected by the Portable Antiquities Scheme for understanding the spatial distribution of artefact types and human activities in the past? A geo-statistical investigation
Project leader: Katie Robbins
Department: Portable Antiquities and
start: January 2009
End date: December 2011
Other British Museum
Dan Pett (Portable Antiquities and Treasure)
Graeme Earl, Universityof Southampton (http://www.soton.ac.uk/)
Sally Worrell, Institue of Archaeology (www.ucl.ac.uk/archaeology)
Chris Lloyd, Universityof Belfast (http://www.qub.ac.uk/)
Arts and Humanities Research Council (Collaborative Doctoral Award)
The British Museum
The Portable Antiquities Scheme (PAS) identifies and records finds made by members of the public in England and Wales, including metal detectorists and amateur archaeologists. Information on the artefacts is held in the online Portable Antiquities Database (PAD) which is available through http://www.finds.org.uk/. This is a fast-growing resource that now holds information on over 390,000 finds ranging in date from the Palaeolithic to the early Modern era.
Although the PAD is increasingly being used by academic researchers for studies of particular artefact classes and the identification of previously unknown archaeological sites, there has so far been little detailed research on the nature of the data itself. In particular, there is an urgent need to understand the factors that influence the geographical distribution of the data and the relationships between collection practice, artefact type and space.
This project is asking, how representative of actual past distributions of material is the data? It is also aiming to identify the factors that define the bias inherent in the record. These factors may be geographical (for example, whether land is covered with forest, water or crops), legal (for example, whether people are allowed to search on the land) or individual (for example, how thoroughly a particular person searches a piece of land).
The research will focus on three counties; Hampshire, the Isle of Wight and a third, as yet unspecified, county. Within these counties, detailed case studies will highlight the effect of different types of bias on the data that is held on the PAD.
The primary aim of this doctorate is to analyse how representative the data held in the PAD is of actual past distributions of human activity.
The doctorate will consider how we can approach and overcome the physical and social biases that are inherent in the data collected by the PAS, and that therefore effect the spatial distribution of the data. The PAS data will be combined with other datasets (land-ownership and land-use for example) in a Geographic Information System (GIS), and through the use of spatial statistical techniques, the issues surrounding the distributions of data in the study areas will be investigated.