Critical Edition as Graph: The Chronicle of Matthew of Edessa Online

In the last few years, several of the bodies that govern higher education and research in Switzerland – the swissuniversities council, the Swiss National Science Foundation (SNF), and the Swiss Academy of Humanities and Social Sciences (SAGW) – have placed an increasing focus on the importance of appropriate and sustainable handling research data, and in particular research data in the humanities. As ever more research in the humanities includes collections of digital data, these Swiss bodies are putting a significant investment into building an infrastructure capable of maintaining it. Against this background, the SNF has begun to focus in particular on the scholarly text edition; beginning in 2017, all SNF-funded edition projects are expected to be digital and to contain some strategy for long-term preservation (Swiss National Science Foundation, 2015). One of the purposes of the SNF-funded project “The Chronicle of Matthew of Edessa Online” is to experiment with forms of digital critical edition of texts that can be preserved and maintained in the long term.
The Chronicle is an Armenian-language historical work, written in the early 12
th century, covering nearly two centuries’ worth of the history of the medieval Near East, up to and including the establishment of the Crusader County of Edessa. The text survives in at least 35 manuscript copies, all from 1590 or later. Digital methods have been used since the beginning of the edition work in 2008; this includes full transcriptions of the individual manuscripts in a TEI-compatible markup system, automatic collation of the textual variants, computer-assisted stemmatic analysis, and systematic editorial analysis of the variant texts (Andrews, 2009). Since the work began, however, there has been significant progress in thought about how to represent textual variants in digital space, most particularly with the adoption of the variant graph (Andrews, 2014; Dekker, Hulle, Middell, Neyt, and Zundert, 2014; Jänicke, Geßner, Büchler, and Scheuermann, 2014; Schmidt and Colomb, 2009).

In this poster session we will demonstrate the ways in which, beginning with the variant graph concept, we are implementing all aspects of the edition in a graph database model – the textual collation itself, analysis of the manuscripts and stemmata, the identification of events, timelines, persons, and places that normally occupies much of the commentary of the edition, and the annotations that together with these identifications will comprise the full commentary. Since it is a graph, the data becomes straightforward to transform directly into the Resource Description Framework (RDF) required for compatibility with the SALSAH database provided by the SAGW-funded Data and Service Centre for the Humanities, which is the nascent repository for all humanities research data within Switzerland and guarantees long-term API-addressable accessibility to the data itself (Rosenthaler, Fornaro, and Clivaz, 2014). Moreover, in the construction of the graph model we draw upon relevant sets of standards and tools, such as the TEI Guidelines (http://www.tei-c.org/) for description of the textual features, CIDOC-CRM (http://www.cidoc-crm.org/) to express information about the documents as objects and for notations of time and place in a manner suitable to use with tools such as Topotime (Grossner and Meeks, 2014). We are also participating in the DARIAH-DE Working Group on Digital Annotations in order to ensure that our working methods are methodologically and practically compatible with best practice in humanities research.
The result of the project will be a digital edition whose constructed text, scholarly commentary, and individual manuscript witnesses will be fully accessible through a machine-queryable API. The web user interface will be built entirely on this API; in that way, no intellectual content will be embedded in the interface, only to be eventually lost when or if the interface disappears. The manuscript texts and a version of the edition itself will be available to download in TEI and ePub formats, for print consumption, archiving, or further offline analysis.

Bibliography

Andrews, T. L. (2009). Prolegomena to a Critical Edition of the Chronicle of Matthew of Edessa, with a Discussion of Computer-Aided Methods Used to Edit the Text. University of Oxford, http://ora.ouls.ox.ac.uk/objects/uuid%3A67ea947c-e3fc-4363-a289-c345e61eb2eb.

Andrews, T. L. (2014). Analysis of Variation Significance in Artificial Traditions Using Stemmaweb. Digital Scholarship in the Humanities, doi: 10.1093/llc/fqu072.

Dekker, R. H., Hulle, D. van, Middell, G. et al. (2014). Computer-supported collation of modern manuscripts: CollateX and the Beckett Digital Manuscript Project. Literary and Linguistic Computing, fqu007. 10.1093/llc/fqu007.

Grossner, K. and Meeks, E. (2014). Topotime: Representing historical temporality. Proceedings of DH2014, Lusanne.

Jänicke, S., Geßner, A., Büchler, M. et al. (2014). 5 Design Rules for Visualizing Text Variant Graphs. Presented at the Digital Humanities 2014, Lausanne, http://dharchive.org/paper/DH2014/Paper-652.xml (accessed 27 October 2015).

Rosenthaler, L., Fornaro, P. and Clivaz, C. (2014). Long term preservation of digital information. Presented at the Digital Humanities 2014, Lausanne, http://dharchive.org/paper/DH2014/Paper-68.xml (accessed 27 October 2015).

Schmidt, D. and Colomb, R. (2009). A data structure for representing multi-version texts online. International Journal of Human-Computer Studies, 67: 497–514.

Swiss National Science Foundation. (2014), Call für Editionsprojekte mit Blick auf die Finanzierungsperiode 2017–2010, http://www.snf.ch/SiteCollectionDocuments/call_editionen_phase2_d.pdf (accessed 27 October 2015).