64th RAI in Innsbruck 2018

During the nineteenth century, the professional practice of ancient Near Eastern studies was implicitly or explicitly related in some way to the imperial activities of Great Britain, France and the German empire in the “Middle East”—a geopolitical term in itself also charged with cultural connotations and situational perceptions. While nowadays much of the former imperial and colonial attitudes have been shaken off from ANE scholarship—especially since 1960s and 1970s, in the aftermath of the post-colonial processes in Southwest Asia and the rest of the so-called Third World—there are still conceptual remnants in the field of the previous Western appropriation of the Middle East as a cognitive map and its most ancient past, both in material and symbolic ways. In effect, the key issue in this situation is the importance played by ANE intellectual heritage as claimed especially by the modern Western tradition and how such a heritage has influenced trends in modern ANE studies. This paper calls for establishing a serious and comprehensive study of the conditions by which institutional knowledge about the ANE is manufactured as an international discourse (through universities, academic meetings, research projects, media and popular culture, etc.), touching upon issues of national memories, cultural heritage and religious identities, as well as past and current politics in the Middle East. Furthering such an approach may indeed contribute, firstly, to grasp a clearer understanding of the concept of ANE intellectual heritage in the modern West, and secondly, to provide current ANE scholarship with critical epistemologies, with a scholarly self-awareness precisely of how knowledge is produced and reproduced, where this knowledge is made and located, for what purposes, and which are the potential political implications—in the face of the current political situation in the Middle East—of such a research.

The archive of Nūršamaš is an Old Babylonian archive from the region of the Diyala, exhumed from a location that is not exactly known and presented by an antiquities dealer to the National Museum of Iraq in Baghdad. Most of its 121 documents are loan contracts of silver, barley and emmer. In these documents, there are around 400 different people who are identified by their names. They appear as witnesses and borrowers, besides Nūršamaš, who is always the lender. Some individuals are additionally identified by the name of their fathers or by their profession.
In the 1960s, Fauzi Reschid described eleven groups of witnesses that appeared together in different documents from this archive. The immediate historiographical implication of this finding is that the archive may carry traces of the actual relationships among the people registered in the documents. Based on the analysis of patterns of recurring groups of personal names, Reschid’s description matches the well-known fact that in most communities there are people that prefer to do things together.
I enlarge this analysis by introducing computer-dependent methods. I propose a partition of the whole community into sub-communities through the application of a graph algorithm for what is known as class modularity. Informally speaking, a sub-community is a set of people that in general establish relationships among themselves rather than with people from outside the sub-community.
The first consequence of this method is that each group identified by Reschid, with a very few individuals excepted, is entirely contained in a sub-community, showing that sub-communities extend the original concept of recurrent groups.
This procedure permits furthermore the identification of relationships that would be difficult to detect by traditional methods. Specifically, I will deal with individuals that linked the different sub-communities, that is to say, individuals that may have had greater circulation in the larger context. Preliminary results indicate that a number of such individuals were identified by their professions in the documents, providing thus a concrete instance where being recognized as someone with a profession is a factor of social circulation.
Finally, the presentation will also include the relevant information about the computer assisted methods that were employed, as well as considerations on their strengths and limitations.

This paper presents the work of MTAAC, an international collaborative project involving Assyriologists, Computational Linguists and Computer Scientists from Toronto, Frankfurt and UCLA. The project is funded for two years (2017–2019) by SSHRC (Canada), DFG (Germany) and the NEH (USA) through the Trans-Atlantic Platform Digging into Data Challenge, a program that supports research projects that explore and apply new “big data” sources and methodologies to address questions in the social sciences and humanities. MTAAC is developing methods and tools for the automated analysis and machine translation of cuneiform texts in transliteration, using Ur III Sumerian documents as a test corpus. These documents were chosen because of the relatively high degree of standardization of their contents, which makes them particularly suitable as a test case for the application of machine translation and automated analysis. The project uses Linked Open Data to formalize and make available the results of the automated data extraction, and its working method, code, and results are all being made available in open access on the web. This ensures that our working method can be replicated and modified as necessary, to facilitate the application of machine translation to other ancient language corpora. At this halfway point in the project’s duration, the paper reviews our progress to date as well as the remaining challenges and future prospects.

Digital humanities approaches, specifically social network analysis and language technology, have thus far been fruitfully applied to studies of social networks and lexical data in Mesopotamian studies. Such a quantitative perspective not only brings up new ways to understand and visualize the data, but also offers a firm basis for the traditional philological and qualitative research methods. One of the goals of the Digital Humanities team of the Center of Excellence in Ancient Near Eastern Empires at the University of Helsinki is to further develop and apply digital humanities approaches to Mesopotamian data and to provide the Assyriological community with open-source tools that may be used for quantitative analysis of the Akkadian text material.
The aim of this paper is to present some of our preliminary work with applying these approaches to the 5,000 lemmatized Neo-Assyrian texts of the Open Richly Annotated Cuneiform Corpus (Oracc). Earlier case studies have focused on Akkadian nouns and verbs and have demonstrated the effectiveness of language technological approaches to creating semantic fields for terms. This paper continues this work through studying the use of divine names to test the potential of these methods on a category that is well-attested but distinct from nominal or verbal forms in both form and character. By compiling the lexemes and other gods closely associated with divine names, we are able to extract the collocates, and thus, to identify the prototypical semantic fields for the divinities attested in the texts.
In this paper, we focus on Pointwise Mutual Information (PMI), a statistical measure of association generally used in information theory and computational linguistics, which has already yielded promising results on the semantic analysis of Akkadian nouns and verbs, such as sisû “horse” and verbs of seeing. In applying this method to the attestation of divine names in the Neo-Assyrian text corpus, we are able to more efficiently and precisely examine the semantic fields of the deities and their connections to each other, allowing us to establish and visualize the connections between the divinities of the Neo-Assyrian pantheon and to better conceptualize their characteristics.