Authoritative Practices and Collective Validation: Wikidata within the Collaborative Digital Edition of the Greek Anthology

Maxime Guénette, Mathilde Verstraete, Marcello Vitali-Rosati

Confirmation: I have read and agree with the workshop's policy on behalf of myself and my co-authors.

Authors Biographies: Maxime Guénette (0009-0006-2076-1220) is a PhD student in History at the Université de Montréal. His research focuses on religions in the Roman Empire, on the sacralization of space and on ancient globalization. Since 2020, he is editor on the Anthologia Graeca platform, and has been working on the integration of English and French translations of epigrams, as well as keywords to identify historical figures and places, and philological notes. He has recently been involved in the IAL (“Intelligence artificielle littéraire”) project to detect variations in the Greek Anthology’s corpus of epigrams. During his internship at the Canada Research Chair on Digital Textualities in Summer 2023, he also worked on the application of HTR (Handwritten Text Recognition) to the Codex Palatinus graecus 23 using the eScriptorium platform. As a result of this internship, he trained models to recognize ancient Greek in medieval manuscripts such as the Palatine Anthology. His main contributions are [1]Maxime Guénette et al., “HTR Model Palatinus Graecus 23 (Meleagre-NFD-Finetuned),” April 5, 2024;[2]Caron, Émile, et Maxime Guénette. « Éditer la plateforme Anthologia Graeca : vecteur de changements dans la pratique des classics ». Sens public, nᵒ 1722 (2024). http://sens-public.org/articles/1731/;[3]Guénette, Maxime, Mathilde Verstraete, Marcello Vitali-Rosati, et Alix Chagué. « Transcrire un manuscrit en grec ancien : un modèle de reconnaissance automatique pour le codex Palatinus graecus 23 ». Dans Humanistica 2024. Université Moulay Ismaïl, Meknès, 2024. https://hal.science/hal-04563548. Mathilde Verstraete (0000-0003-1642-8610) is a PhD student in digital humanities at the University of Montréal. After obtaining a master’s degree in classical languages and literature at the Catholic University of Louvain (Belgium), she joined the Canada Research Chair on Digital Textualities to coordinate the collaborative digital edition of the Greek Anthology. Under the supervision of Marcello Vitali-Rosati and Elsa Bouchard, her research focuses on digital critical editions and the tools that produce them. Her main contributions are [1] Mathilde Verstraete and Margot Mellet, “Passés et présents anthologiques. Le projet d’édition numérique collaborative de l’Anthologie grecque,” in Communautés et pratiques d’écritures des patrimoines et des mémoires, by Nicolas Sauret and Marta Severo, Presses universitaires de Paris-Nanterre. (Paris, 2024) ; [2] Mathilde Verstraete, “Miscellanea anthologica,” Sens public, no. 1722 (April 15, 2024) ; [3] Maxime Guénette et al., “HTR Model Palatinus Graecus 23 (Meleagre-NFD-Finetuned),” April 5, 2024. Marcello Vitali-Rosati ([e-mail]) is a professor in the Department of French Literature at the Université de Montréal and holds the Canada Research Chair in Digital Textualities. He is developing a philosophical reflection on the challenges of digital technologies: the concept of the virtual, digital identity, notions of author and authority, forms of production, legitimization and circulation of knowledge in the age of the web, and the theory of editorialization – to which he is one of the most active contributors. He is the author of numerous articles and monographs, and also works as a publisher, editing the journal Sens public and co-directing the Parcours Numériques collection at the Presses de l’Université de Montréal. As holder of the Chair in Digital Textualities, he leads several projects in digital humanities, particularly in the field of scholarly publishing. These include the development of platforms for publishing enriched journals and monographs, software for editing scientific articles, and a collaborative publishing platform for the Codex Palatinus 23.

Keywords: Greek Anthology, authority, collaboration, digital philology

TL;DR: This study explores how the Anthologia Graeca project uses Wikidata to manage data, redefining authority and validation in the Humanities while fostering collaboration and contributing to an inclusive and sustainable knowledge framework.

Abstract:

The management and preservation of research data in the Humanities increasingly raises questions about its sustainability, sharing, and validation. In this context, Wikidata constitutes a powerful and collaborative tool. By challenging traditional models where researchers act as both producers and gatekeepers of authority, Wikidata redefines these issues and fosters new paradigms of collaboration.

This paper will explore these dynamics of collaboration and shifting authority through the case study of the collaborative digital edition of the *Greek Anthology* (the AG project, hosted at the Canada Research Chair on Digital Textualities since 2014), implemented on a collaborative platform (<https://anthologiagraeca.org/>) where everyone is invited to participate according to their own knowledge.

Wikidata is used in many ways within the AG project. First, all keywords (place names, authors, metrical forms, literary genres, etc.) used to annotate the platform have a Wikidata identifier or is created accordingly. Indeed, when a user participates in the editing of the corpus and wishes to add a keyword to an epigram, if the keyword does not exist, he or she must create it on Wikidata and then link it to the platform. Second, Wikidata has been used in a more intensive way to address inconsistencies in our list of authors. Like Wikidata, our data model is multilingual. However, the gaps and inconsistencies in Wikidata ---such as missing authors, duplicate entries, and inconsistent information across languages--- were directly mirrored on our platform (<https://anthologiagraeca.org/authors/>). This alignment made it essential to tackle these issues systematically to ensure the accuracy of our data. We started by searching for the names of these authors in various languages (at least in French, English, Italian, Ancient Greek and Latin). We then uploaded this information to Wikidata, and subsequently fetched it back to integrate it into the AG platform. Almost immediately after our data dump on Wikidata, its community quickly reviewed and corrected it to align our contribution with Wikidata's standards and guidelines. This process means we not only retrieved our data but also benefited from the community's improvements.

We are making a conscious strategic choice: rather than positioning ourselves as the sole custodian of authority, we are delegating that responsibility to a wider community. Our presentation invites reflection on the implications of this shift toward distributed authority. How can that shift in authority benefit academic research projects? Is Wikidata's epistemological paradigm coherent with ours? Can we think of a generic epistemological framework to be effectively applied to specific academic endeavors?

Based on the experiments carried out and the choices made as part of the AG project, this presentation will provide practical and conceptual answers to the questions of (distributed) authority, validation and collaboration in the use of Wikidata, opening up prospects for other projects in the Humanities. We suggest that Wikidata is not merely a technical tool but rather a space where methodological and epistemological debates can unfold. By engaging with this dynamic, researchers can enhance their projects while contributing to the creation of a more sustainable, inclusive, and collaborative knowledge base.

Format: Paper (20 minutes presentation)

Submission Number: 19