This paper begins with discussion of a training data set that marks out chains of transmission (isnads) in a 1.5 billion-word corpus of Arabic texts (running ca. 700-20th century). The presenter discusses how she and a team of historians and computer scientists have been working together to train a model to detect isnads (chains of names documenting transmission) across different types of Arabic texts (crucially, not limited to Hadith collections). The historians providing the training data began with an understanding of isnads grounded in decades of work on all genres of Arabic writing. They believed they knew what an isnad was. But weeks working together, discussing parameters for in-text isnad annotation, plus a test study comparing their annotations of a single text, showed them that their definitions remained at odds. The variety of forms and features of isnads were far higher than they expected. Which made the performance of the computer scientist’s model, based on their training data, more remarkable (at present, achieving 85% precision and 81% recall (with training data generation continuing across 2020). The paper’s main goals are three-fold and based on a larger Arabic DH project. First, I will consider the ways that “humanist” researchers can and should inform models guiding our emerging understanding of matters such as isnad identification, text reuse detection, and optical character recognition. The exchanges between computer science and the humanities can be mutually informing. I discuss aspects of project design relevant to gaining the best benefit for both, and also the challenges we have faced. Secondly, I want to show what the case of isnad identification can teach us about book history, specifically. Our work on isnads forces a rethink of many assumptions in current scholarship on book history (including how isnads relate to one another in large works). Finally, I propose that a crucial difference of DH is the way it can bring research to the public. For this, I show a web application that makes use of our project's training data.

Medieval Middle Eastern society expressed itself in a greater number of simultaneous literary traditions than any other portion of the pre-industrial world. Despite this fact, for reasons of subfields’ particular institutional histories, very few scholars can or do approach the Middle East between 800 and 1500 CE in ways that consider the full range of literary evidence available. Instead, the scholarly study of the medieval Middle East has to a surprisingly large degree been balkanized among distinct subfields, each privileging a different literary tradition. The development of Digital Humanities provides medieval Middle Eastern historians an opportunity to overcome scholarly subfield divisions by accessing a fuller range of evidence rather than omitting sources due to linguistic limitations. This talk presents a digital humanities project currently under development which seeks to bridge the subfields’ divides by making visible the ways that medieval Middle Eastern primary sources in different languages discuss the same places, many of the same people, and often the same social practices. In other words, the literary traditions were linguistically but not sociologically separate. The project is developing a multi-lingual index of selected medieval Middle Eastern primary sources in a range of different languages, so that a scholar searching for, say, Fatima or Salah al-Din or Baghdad will find references which span the linguistic spectrum. The project will be extensible, so that additional primary sources can be added progressively. Each reference includes translations where available, so that scholars can access the contents of sources even beyond their linguistic training. This paper has three goals. First, it contrasts the polyglot and multiconfessional culture of the medieval Middle East with the limits of current scholarship, to show the benefits of taking a broader view of available evidence. Secondly, this paper proposes that digital tools, such as the project to be discussed, are useful sites of discovery to enable scholars to develop more deeply informed models of the medieval Middle East. Finally, this paper demonstrates how digital tools permit more flexible and manipulable data visualization than print media, thereby avoiding merely replacing one canon with an alternative. The difference of DH is that multiple alternative frameworks for understanding medieval society can be explored simultaneously without designating a dominant discourse, as was required by the static nature of print. In this way, DH scholarship may more closely approximate the many voices and views simultaneously embedded in our sources for the medieval Middle East.

How did pre-modern Islamic societies develop over time? Historians have been searching for answers to this complex question in massive Arabic chronicles and biographical collections that survive by the dozens. Advances in digital humanities offer an approach that can overcome all usual methodological limitations associated with the study of these massive historical texts and open virtually unlimited research opportunities. This paper will present the main conceptual and methodological aspects of the implementation of a specific and new DH approach. Medieval historians composed their texts by picking and choosing “passages” from their sources, rephrasing them, commenting on them, and, in the end, reassembling them into their own representations of historical reality. In a somewhat similar manner, we can computationally disassemble all surviving sources into such passages. (Due to the use of unique identifiers with all passages we will still be able to reassemble original texts as well as to keep adding other historical texts as they become available.) Using metadata from the original books and such methods as named-entity recognition, we can arrange these passages chronologically and geographically, merging them into what will virtually become the “master chronicle” of Islamic history. Using computational methods for identifying textual similarities, we can then assemble these passages into networks of related historical information. This arrangement will allow for a variety of modes of reading. For example, one will be able to read: “historically”—by moving from one event to another in chronological order; “historiographically”—by exploring how specific events were presented by different Arab historians; “thematically”—by focusing only on events that deal with specific topics. Moving between distant and close reading will be effortless as trends can be graphed and mapped, and any specific passage, which constitutes a given trend, can be read carefully in a traditional manner. By bringing together all available historical texts, the master chronicle will provide the most thorough possible coverage of historical periods and geographical regions. By marshaling all available quantitative evidence, the master chronicle will make the qualitative analysis more substantial as findings will rest on exhaustive textual evidence. The master chronicle will allow us to conduct research in a “cumulative” manner where results are attained at a large scale and with reproducible means, providing a solid foundation for asking novel research questions. Finally, the master chronicle will serve as a robust exploratory environment for studying any research question that can be possibly approached through Arabic historical texts.

There is little granular data for an urban history of Abu Dhabi for the growth years of the end of the twentieth century. Sources are dispersed and have not yet been mass digitized so that their data can be analyzed with contemporary digital historical methods. Our project builds on examples of in spatial humanities demonstrating how city directories are rich sources for urban history (Sutton, NYPL, Milliken) and focuses on an analysis of the diverse populations in the capital city of the UAE. Through studying the “phone book”--itself a highly globalized object with its share of documentary shortcomings--we study temporal and locational patterns of markers of both national belonging and religious identification in some 400K telephone subscriptions across three decades. The results presented in this paper stem from a recent project to digitize a phone book collection, as well as dataset under construction, allowing a data-centered comparison with qualitative scholarship on transience, urban planning and placemaking in the Gulf (Alshestawy, Alamira, Molotch & Ponzini). The difference of digital history, this paper argues, lies in the types of new sources that are being used in Gulf Studies, the scale at which data are collected and analyzed as never before, as well concomitant forms of critical awareness required to carry out such research ethically and effectively.