Named Entity Recognition (NER) has long posed challenges in the digitisation of historical documents. Recent systems such as spaCy (Mattingly 2021), together with growing interest in the potential for Large Language Models to perform historically informed NER, suggest new opportunities. Yet discussions at the recent workshop on the coding of medieval charters (held at GWZO Leipzig, 4–5 December) made clear that entity identification, while essential, is far from sufficient for historically meaningful enrichment. Beyond recognising historical actors in charter texts, it is critical to identify the roles they played within the specific “legal acts” articulated by those documents.

Participants from a range of charter-based research projects, primarily focused on Eastern Europe, broadly agreed that the Factoid model provides a robust conceptual and formal foundation for expressing more than the mere presence of historical persons (see, for example, Vogeler 2025). The authors were invited to contribute to the workshop (Hammond and Bradley 2025) because of our involvement in the People of Medieval Scotland (PoMS, 2010–18) project, which developed an extensive representation of the content and significance of Scottish medieval charters based on a prosopographically oriented factoid model.

The study of legal charters—diplomatics—has traditionally emphasised the document as a whole, aimed at establishing its form, authenticity, and characteristics (Herde 2022; Williams 2005). In the digital era this document-centred orientation has evolved into a metadata-centric approach to data modelling. Basic NER aligns with this approach: with it one can readily produce a document-level statement that certain individuals “appear in” a charter and sometimes even that they hold particular roles. Difficulties arise, however, when a single charter narrates multiple legal acts with different participants and role assignments. In a property exchange, for example, two individuals may each simultaneously act as both grantor and beneficiary. These roles cannot be meaningfully attached as undifferentiated metadata about the charter; they must instead be attached to the specific legal acts to which they belong. This raises the question of what data structures are adequate for such complexity. A new complex “level” of data was needed within the metadata oriented interests of diplomatics.

PoMS approached charters from a more explicitly prosopographical perspective. The factoid model (Bradley 2018) enabled multiple types of person-centric statements to be recorded from a single source. In PoMS, each legal act within a charter was represented as a “transaction” factoid with additional factoid types capturing such as human relationships, offices and titles. The model was also extended to include diplomatic assertions such as document type (charter, diploma, brieve, chirograph, letters patent, etc.), the kinds of legal acts conveyed (Gift, Confirmation, Renewal, etc.), and the tangible and spiritual benefits associated with those acts. At the Leipzig workshop, the extensible and multi-layered structure of PoMS—integrating prosopographical and diplomatic statements within a unified factoid framework—emerged as a compelling model for charter analysis. Although the factoid model was applied at Leipzig to a specific type of historic documents, it has been demonstrated to work well with other kinds of historic documents in projects such as the Prosopography of Anglo-Saxon England (PASE 2005-16) among others.

This poster will review the Factoid model and will demonstrate how PoMS extended it to accommodate both prosopographical and diplomatic dimensions of medieval charters (including some new thinking that has arisen since the publication of Bradley et al 2019). It will illustrate how the model represents the collection of legal acts within charters and how such structuring supports more accurate historical interpretation. Finally, we would like to use this opportunity to advance ongoing work connecting the factoid model to TEI or CEI representations of charters, building on Bradley and Jakacki (2021), considering the implications of this integration for future digital diplomatic and other historical scholarship.