Text Mining Tafsir: Compilation and Preliminary Explorations of a Curated Corpus of 80 Qurʾanic Commentaries

The aim of this article is threefold: First, we introduce a curated corpus of 80 Arabic Tafsir ( tafsīr ) works that has been compiled as part of a project investigating the development of Qurʾanic WAY metaphors. Second, we present systematic quantitative and qualitative characteristics of the Tafsir corpus and parts of its metadata. Finally, to demonstrate the potential of computer-driven analyses of the Tafsir corpus, the article includes a clustering of selected subchapters and sentences from the Tafsir texts that discuss early middle-Meccan Qurʾanic verses ( āyāt ) in which the term ṣirāṭ (“road”) appears. We thereby hope to showcase how advanced computational methods can serve as a heuristic tool to examine the Arabic texts, identify text passages worth investigating, assist with research questions, or even generate new research questions by using distant-reading approaches.