From Late-Antique Text to 21st Century Literature Database: Babylonian Talmud Stories as a Case Study

Using the Babylonian Talmud stories as a case study, this poster illustrates both the potential and difficulties of transferring a rich literary corpus from late antiquity to the digital age. It describes a project in its advanced stages that aims to compile a comprehensive computerized database of all BT stories across genres and subgenres—regardless of their perceived quality, completeness, or predictability. The poster presents a database built over five years of research and development, outlining the conceptual challenges and opportunities encountered when collecting, sorting, structuring, and preparing such unstructured literary data. The poster highlights and problematizes three key issues: Corpus creation methodology; Annotation philosophy; and modeling theory. The project's approach prioritizes human manual annotation over automatic methods; uses machine learning and artificial intelligence modestly; commits to capturing nearly 100% of narrated materials rather than settling for "good enough" automatic extraction; grounds its methodology in poetics and classic narratological questions; and systematically validates computational findings against traditional scholarship.