To Make a Short Story Long: The Preparation of Talmudic Stories for Distant Reading

• In Event: Every Tag and Tag: Annotating Hard-to-Note Phenomena in Jewish Studies

Abstract

Rabbinical stories are usually very short, even miniature, and this is one of their fundamental characteristics. In all genres of Talmudic literature, most stories consist of either just a few words or a few sentences. Even when longer literary forms exist in this literature, they are usually composed by chaining together short stories into a broader narrative. This is another aspect of the Rabbinic preference for the dramatic mode over the epic one, as Jonah Frankel has already pointed out.
Considering that the challenge of distant reading was introduced at its very beginning (by Franco Moretti) in regards to novels originating from all cultures - that is, long texts of disparate origins - it should be noted that the challenge facing distant reading in rabbinical stories is quite different: these stories are not very long, and originate from one (or two sub-) cultures. By simply reviewing hundreds or even thousands of short stories, an experienced reader can easily gain general and solid insight into the literary nature of the material, its main poetic characteristics, and so on.
Nonetheless, it does not suffice; a comprehensive, systematic, and truly thorough analysis of all the stories - with a desire to view them as a whole, as a system, and not as a mere collection of individual stories - requires adding layers of information to each, which could later be explored. Thus, the layers are supposed to turn the corpus into small-smart-data (as described by Christoph Schoch), which will subsequently enable an un-dogmatic reading (inspired by Jan Christoph Meister' approach for computational literary studies). This is one of the main goals of the project described here, which is funded by the ISF.
This paper, therefore, aims to discuss the process of collecting all the Babylonian Talmud stories - complete or fragmented, fine or seemingly unfinished - and the process of enriching them with metadata that is partly objective and partly subjective. It will describe some of the challenges we faced, as well as some of the methodological and conceptual decisions we made; among them, and the most important one, is the decision not to rely solely on artificial intelligence, but on human reasoning.

Authors

• Itay Marienberg, Ben Gurion University of the Negev

• Avraham Yoskovich, Ben Gurion University of the Negev