Digital humanists who work with relatively small corpora to do digital textual analysis must wrestle with the difficulty of drawing conclusions based on statistical analysis that prefers much larger datasets. Indeed, some methods are too data-hungry to yield reliable, generalizable results. However, I argue that the difficulty and careful design needed to execute digital textual analysis on small corpora can indeed provide insights into them. In this talk, I will discuss a project that involves comparing the role of an embodied Devil figure in two genres of the early modern English popular press: witchcraft accusation pamphlets and murder pamphlets. I contend that closely pairing digital methods with close reading and attention to historical and rhetorical contexts allows us to draw conclusions about the development of the figure of the Devil over the period of the late sixteenth century to early eighteenth century.

This talk will be online (Zoom).

This talk is part of Love Data Week—see the entire lineup at https://lovedatapgh.io/!