William Bouchard is awarded the Ian Lancashire prize at CSDH

William Bouchard (student member and project coordinator within the laboratory) is the recipient of the 2026 Ian Lancashire Award for a Promising Student, presented by the Canadian Society for Digital Humanities.

The award was presented during the society’s annual conference in Montreal. During the conference, William delivered three presentations. The first, co-authored with Marcello Vitali-Rosati, focused on their ongoing work within the digital edition project of the Greek Anthology. The second, co-authored with Juliette Sokolov, examined the identification and modeling of paraphrases of Les Caractères by La Bruyère in eighteenth-century novels. Finally, William presented an application designed to improve the transparency of editorial choices in critical digital editions of ancient texts.

The jury considered his presentation, entitled “Editing Decisions in the Anthologia Graeca: Making Classical Philology Computable,” to demonstrate a strong grounding in both computational methods and the study of classical texts and scholarly issues.

Abstract

In digital environments, edited texts quickly come to function as stabilized objects: indexed, cited, reused, and sometimes even adopted as authoritative references. This stabilization transforms a chain of interpretive choices into what appears to be a self-evident result, even though digitization multiplies the operations that shape meaning, including segmentation, normalization, encoding, alignment, and visualization. In classical philology, however, this chain of decisions lies at the very heart of scholarly work. Texts transmitted through manuscript traditions are inherently unstable and fragmentary, and while the critical apparatus records variant readings, it does not always make explicit the reasoning behind editorial choices or the linguistic, metrical, stylistic, and stemmatic frameworks that favor one reading over another. The shift to digital environments makes this tension even more pronounced when computational categories solidify what remains gradual, contextual, and argumentative within the editorial workshop.

This paper argues that the fundamental critical unit of a digital edition is not merely the text and its variants, but the editorial decision itself, conceived as an addressable, traceable, and debatable object.

An editorial decision is formulated as a conclusion linked to a stable textual target. It is supported by evidence (witnesses, editorial tradition, linguistic coherence, metrical plausibility, stylistic likelihood), makes explicit a warrant understood as a rule of inference, documents epistemic support, records alternative interpretations, and situates a degree of certainty. This proposal is demonstrated through a prototype edition of the Anthologia Graeca, centered on a small corpus of funerary epigrams selected for the density of their borderline cases.

The project relies on Philographos, a web-based scholarly environment developed for the project and designed to connect each editorial intervention to its accompanying justification file while preserving version histories. The architecture combines two layers: a textual layer serialized in TEI-XML (established text, segmentation, critical apparatus) and an argumentative layer serialized in JSON-LD, in which decisions and their relationships form a graph. Alignment is based on stable identifiers, notably URNs, allowing a precise locus to be targeted, audited, compared, and revised independently of any particular display interface.

This framework makes visible forms of mediation that are often left implicit, enables the mapping of zones of uncertainty, and facilitates the exploration of scholarly disagreement, while acknowledging that part of meaning ultimately resists computation.

Congratulations, William!