Eva Elisabeth Houth Vrangbæk: Text Mining Augustine: A Case Study of ‘God' and ‘Will’ in De Civitate Dei

This short paper will through a computerized analysis of the themes of ‘God’ and ‘will’ in De civitate dei communicate how text mining as a method of text analysis can be applied on Augustine studies. As for now, text mining is rarely used within patristic studies, but the methods are rapidly gaining terrain within the traditional fields of humanities. I will in my paper show an example of how these methods can be operated in patristic studies1) In the first part of the communication, I will give a brief presentation of the field of text mining focusing on the use in the humanities, its merits and its limitations. 2) In the second and main part, I will demonstrate four central analytical models for analyzing the themes of ‘God’ and ‘will’ in De civitate Dei using text mining and display how this example can be used as a starting point to work with larger portions of Augustine’s writings. One of the key benefits of text mining is that we can analyze vast corpora of texts (Big Data). In this paper I will use a single - but comprehensive and varied - work, to demonstrate the methodological benefits. I will show how one advantageously can work both quantitatively and qualitatively with texts using e.g. sentiment analysis, word-clustering, word-frequency and topic modelling. 3) In the third and last part, I will give a brief evaluation of how these computerized methods can be fruitful to and complement traditional analogous examinations of Augustine’s texts.