In this talk I will introduce examples of "at-risk" textual archives from the ancient Near East. By exploring two ancient sites which have yielded large bodies of textual and material culture through illicit excavations, I will describe the latent archival organization found in the hoards of cuneiform tablets. Then, I will describe the computational methods that I and my colleagues have used to restore these archival boundaries within their respective corpora. These methods incorporate a combination of network analysis and ML classification models together with close reading for supervision. The result is a series of interactive color-coded network graphs, available for scholarly collaboration.
Adam Anderson is a Postdoctoral Fellow in the Digital Humanities. His work brings together the fields of archaeology and computational linguistics to quantify the social and economic landscapes emerging during the late third to early second millennia in the ancient Near East. Collaborating with BPS @BerkeleyProsop to visualize early Assyrian and Sumerian social networks, Anderson's research focuses on tracking the flow people and the exchange of commodities in Bronze Age societies (2100-1800 B.C.). His dissertation, "The Old Assyrian Social Network," combines classical philological methods with natural language processing and social network analysis, to disambiguate the actors, cliques and groups found in a text corpus of 6,000 cuneiform tablets. His work shows how networks of internally related archives provide a means of mapping the overlapping data sets from ancient texts and modern archaeological records, to explain the hierarchical roles and positions of individuals and groups within a society.
This talk is part of the Ancient History & Mediterranean Archaeology Colloquium: Digital Humanities and the Ancient World.