The project develops state-of-the-art technology for Assyriologists to handle large amounts of textual data for sociolect recognition and analysis. Without automated methods, such research is possible only for a few senior Assyriologists having absorbed thousands of texts. However, the continuous publication of Cuneiform tablets makes it impossible for a scholar to read all of them. The multilingual data (Sumerian, Emesal, and Akkadian) is analyzed with language technological methods from a linguistic perspective. For Sumerology, a subfield of Assyriology, the project enables understanding the nature of Emesal in a completely new way. For language technology, dealing with the linguistic context of Cuneiform texts provides an opportunity to develop methods for detecting sociolects in small fragmented corpora using recent methods in neural computing. For linguistics, it is an interesting test case regarding language varieties in a uniquely long timeframe.

2021

2025