The authors have worked over several years on a software tool to make word counts from an archive of old-spelling early modern English plays and poems. In this article we present the outcome, a computational model for dealing automatically with variant spelling, implemented in an application which we call an ‘Intelligent Archive’. We also reflect on the perspective on Early Modern English, and on the probabilistic aspect of language in general, gained from working through the practical problems which arose in establishing the model.
Literary and Linguistic Computing Vol. 25, Issue 1, p. 37-52