Description
The project “Word Embeddings for Aramaic Dialects” combines the powerful tools of machine learning with the classical field of textual criticism. The goal of the project is to use machine learning techniques to train a machine in differentiating between various dialects of Aramaic. This can be used in turn for critical study of the initial composition and early influences on these texts. We began our project with the two Talmudim, which provide us with large textualcorpora, required for the training of the algorithm. The training was highly successful, and we were surprised to find that the algorithm was already consistently able to correctly identify the influence of foreign dialects within a text. Experiments using the algorithm provided us with new and unexpected critical insights concerning the compilation and subsequent transmission
history of the text. Currently we test the algorithm on the approximately 100 scrolls written in Aramaic, some of which reflect significant Babylonian affinities. If indeed a Babylonian influence will be detected on the dialect of even some of the passages in these scrolls, it may contribute to the ongoing scholarly debate about the compositional process of these texts and on the question of migration of knowledge from Babylonia to Qumran.
| Period | 3 Mar 2022 |
|---|---|
| Event title | The Seventeenth International Orion Symposium: (Con)textual Perspectives on the Dead Sea Scrolls |
| Event type | Conference |
| Conference number | 17 |
| Location | Jerusalem, IsraelShow on map |
| Degree of Recognition | International |