Digital Hermeneutics, Medieval Texts, and Urban History

This article presents an inquiry into the use of natural language processing (NLP) methods to enrich, rather than replace, hermeneutical workflows in historical research. Making use of digital technologies in the form of existing tools and custom computational processing, it advocates an approach that fosters deep text interpretation by historical scholars with the aim of incrementally addressing and expanding the range of research questions asked about a particular theme, using a particular textual corpus. In general, this paper argues that success hinges on the possibility of incrementally and systematically unlocking new data for hermeneutical knowledge acquisition and integration without compromising the role of the human historical researcher and their core scholarly analysis methods. This entails that, just like in a traditional manual, close reading effort, the scholar should retain maximum control of research activity and strategy. Our main finding is that the digital hermeneutical method applied in the described work provides relevant results for a ‘gude compt and rekning’ (the late medieval concept of ‘good account’ described in this article) of the conceptual structure of our domain. The general conclusion we draw from this is that NLP brings possibilities for more focused and fine-grained qualitative text analysis in this domain, while allowing easy access to a global perspective on the texts under study. We contend that a combination of tailored quantitative and qualitative text analysis methods can be integrated into a flexible research workflow, which empowers the hermeneutical work of humanities researchers

Comparative historical research · Digital humanities · Hermeneutics · Qualitative research · Workflow · Computational and Text Analysis Methods · Digital Humanities and Scholarship · Sentiment Analysis and Opinion Mining