Texts into databases : the evolving field of new-style prosopography

In this article we discuss some issues that arise when a highly structural approach is taken to the development of prosopographies, based on our experience of three such projects. Databases in prosopography have traditionally been used to help a researcher analyse a large set of database-like historical materials—the database (Bulst, N. 1989, Prosopography and the Computer: Problems and Possibilities. In Mawdsley, E., Morgan, N., Richmond, L. and Trainor, R. (eds), History and Computing III. Manchester: Manchester University Press, pp. 12–18), acts as an intermediate aim, not a final one. For our prosopographical projects, the final prosopography is a highly structured database—not a set of articles derived, perhaps in part, from it. We have found that a highly structured model works well for our prosopographies when we use the database not to model the relevant content of the source materials (which, because many of the texts are historical narrative sources, are of course not well served by relational structures), but to model some aspects of how the prosopographer thinks about their materials, and their task. The article describes how the presentation of a prosopography in this way benefits the end user as well, allowing access via many potential indices and searches. Finally, it begins an exploration of how a user of our databases might interpret what the database tells them