DiDip at the International Medieval Congress 2026: Charters and Temporalities

This was the one thought that kept looping in our heads the entire trip.

TL;DR: If you need to get to Leeds (or pretty much any regional UK city), budget an extra day for the journey. TL;DR over.

Everyone on the team took a different route to Leeds, but somehow almost all of us ended up with our own travel horror story. Tamás and Giuseppe got their first reality check at Graz Airport. Thanks to the UK’s post-Schengen status, online check-in was no longer an option, which completely defeated the purpose of flying from Graz’s tiny, normally super-convenient airport. Instead of breezing through, we had to be there more than an hour earlier. The beautiful Austrian and English countryside almost made up for it (to be fair, it actually did)… until our plane performed a go-around just before landing because Gatwick was “busy.” That was enough to make us miss our train—despite not having booked an unrealistically tight connection. So instead of heading straight to Leeds, we took an unplanned scenic detour via King’s Cross before finally making it. The trip home wasn’t exactly smoother either. Only three of our trains got cancelled, which, by that point, felt like business as usual. Everyone was convinced Johannes had cracked the code by flying from Bratislava, where there’s a direct flight to Leeds. Flying from Vienna? Clearly too obvious. But the return journey had other plans: his flight decided to make a little detour via Kefalonia before eventually reaching Bratislava. On the bright side, he had a five-hour layover—just enough time to head into town and enjoy some authentic Greek food, much to Anguelos’ delight. Then there’s Georg. He’s clearly operating on a different difficulty setting. Eurostar had two hours delay. But, hey, he had his bicycle with him and a buffer of one day for a presentation in North Yorkshire Archive in Northallerton. So yes, the journey itself was, frankly, dreadful. Next time we’ll probably pack hiking gear just in case.

Leeds, on the other hand—and IMC in particular—was fantastic, as always. Great organisation, excellent sessions, a lovely campus, and an event that runs like a well-oiled machine. Every single time.

Once reunited, the group attended the International Medieval Congress 2026, held at the University of Leeds from 6 to 9 July. For DiDip, the event brought together almost the entire project group, along with collaborators from its wider research network. Since its first edition in 1994, the IMC has developed into one of Europe’s largest events for medieval studies, bringing together researchers from a wide range of disciplines and scholarly traditions. This year, the theme was the “Temporalities” and the different ways in which time was perceived, organized, experienced, represented, and manipulated in medieval societies, as well as the methods through which scholars now reconstruct those temporalities. This theme offered a particularly productive framework for the DiDip team. Charters are instruments designed to give actions a durable form: they record dates and obligations, preserve legal memory, reuse earlier models, project rights into the future, and change through the work of institutions and individual writers. Their temporal dimension is therefore not confined to the date written in the document.

From digital to distant diplomacy and across time

We presented three interconnected sessions under the shared title “From Digital to Distant Diplomatics”. The panels were organized and coordinated by Tamás Kovács. Georg Vogeler co-organized the first two, while Antonella Ambrosio co-organized the third. These sessions were sponsored by the Department of Digital Humanities at the University of Graz, and in particular by the ERC project “From Digital to Distant Diplomatics”. The program translated the broad theme of temporalities into three complementary perspectives. The first session approached charters through historical and diplomatic analysis; the second examined diachronic language through computational methods; the third turned to their visual and material features. Together, traditional diplomatics, Natural Language Processing, and Computer Vision addressed different layers of the same historical problem: how documentary practices preserve continuity while registering change.

Charters as chronicles of change

The first session, “Charters as Chronicles of Change,” examined documentary language as historical evidence across different institutional settings.

Roberto Lamponi–you remember him from his stay in Graz–opened the session with notarial writing practices in Fermo and its territory in the March of Ancona during the fourteenth and fifteenth centuries. Drawing on notarial registers as well as selected public and institutional documents, he examined how notaries used standardized expressions acquired through their professional training and adapted them to different legal and institutional needs. His case study placed traditional and innovative elements within the same documentary practice: formulas transmitted professional knowledge, but remained flexible enough to respond to different acts and contexts.

The tension between learned formulas and changing documentary needs reappeared at a different institutional scale in Daniel Luger’s paper on the papal chancery. He presented a new project at the Institut für Österreichische Geschichtsforschung in Vienna devoted to editing the registers of Pope Honorius III. By comparing them with the registers of his predecessor, Innocent III, the project asks how the formulas of papal charters changed and how the registers themselves evolved. It also provides a case study for applying HTR, text-reuse analysis, and stylometry within distant diplomatics and for integrating these methods into the wider study of papal chancery culture.

Žarko Vujošević then shifted the question from institutional change to the intellectual circles that shaped the formulaic language of Serbian royal charters. His study drew on an XML corpus of around 200 Serbian-language documents issued between 1282 and 1459, with particular attention to approximately 100 charters addressed to ecclesiastical recipients. He concentrated on formulas of historical, political, and ideological relevance, especially in the arenga, narratio, and sanctio, where quasi-liturgical language and biblical references articulated the “political theology” of Serbian rulers. XQuery for Keyword-in-Context analysis, and FLAME connected characteristic expressions, such as references to the ruler of the Greek lands, the “first emperor,” or Saints Simeon and Sava as heavenly avengers, with particular documentary circles. The results point to the role of high representatives of the Serbian Church, especially members of the Hilandar circle, in composing charters for their monastery and other ecclesiastical recipients. This connection between formulaic choices and particular documentary circles prepared the ground for the final paper, which tested the relationship between textual patterns and identifiable scribal hands.

Johannes Laroche concluded the session with a study of the charters issued by the abbots and convent of Lambach between 1325 and 1499. Using ANNIE and FLAME alongside diplomatic and palaeographical analysis, he traced changes within a corpus whose basic documentary structure remained remarkably stable. The results identified a gradual compression of several clauses between 1415 and 1425, coinciding with a change of documentary hand. Textual similarity also correlated strongly with palaeographically identified hands. Formulaic language, therefore, appeared neither as a fixed institutional template nor as unrestricted individual composition. The stable framework left room for what Laroche described as “agency within the formula”: individual writers could shape documentary language at a micro-level while preserving its recognizable diplomatic structure. Taken together, the papers showed that continuity and change are not opposites in documentary culture. A formula could preserve institutional memory, adapt to a new legal or political setting, or be shortened and reorganized through everyday writing practice. Its historical value lies precisely in the relation between what remains stable and what changes.

Decoding diachronic language

The first session raised an immediate methodological question: how can the relationship between stable forms and small variations be identified when the corpus contains hundreds or thousands of documents? The second session, “Decoding Diachronic Language,” addressed this problem by shifting from historical interpretation to computational analysis of large, heterogeneous textual corpora.

Tamás Kovács addressed the problem at its most basic level: a diplomatic formula cannot be defined by literal repetition alone. His study applied FLAME’s Leave-N-Out n-gram method with subword tokenization to more than 2,000 documents dating from 1214 to 1512, including charters from the Benedictine Abbey of Pannonhalma and selected papal material. Adaptive normalization and automatic thresholding helped the system recognize relationships despite differences in spelling, word order, inflection, abbreviation, transcription, and HTR output. The analysis identified formulas shared across the corpora and significant instances of biblical reuse, revealing intertextual connections between religious and legal discourse. In this perspective, variation is not simply noise to be removed: it can be the diplomatic signal itself.

Taking this computational approach over two centuries, Noé Leroy examined the temporal development of formulaic language in European episcopal chanceries. His presentation tested the method on 869 lemmatized Latin acts issued by the prince-bishops of Liège between 1145 and 1345. It distinguished three levels of formulaicity, from the macro-structure of a document to recurrent intra-clausal sequences and more idiosyncratic “micro-formulae.” A decreasing-intersection algorithm identified independent shared segments, which were evaluated according to length, position, rarity, lexical richness, and co-occurrence. Together with stylometric comparison, this approach can identify both the persistence of internal chancery styles and their gradual transformation, while distinguishing local developments from tendencies shared across regions.

Where Leroy–you remember him as well–used formulaic proximity to study chancery styles over time, Franziska Decker and Giuseppe Consolo expanded the analysis to the social actors involved in notarial production. They addressed the construction and analysis of a corpus of fourteenth- and fifteenth-century notarial charters preserved in Monasterium.net. Their workflow combined metadata filtering and lexical searches with Computer Vision for detecting notarial signs based on Anguelos’ object detection work and ANNIE for semantic annotation. The object-detection stage identified 20,103 Monasterium entries containing notarial signs, 11,173 of them dating from 1300 to 1500, but only a small portion currently includes usable transcription data. The resulting corpus is intended to support named-entity recognition, formula analysis, and the reconstruction of networks linking issuers, recipients, and notaries. This workflow connects two levels that are often studied separately: the external signs through which a document presents itself as notarial, and the internal language through which legal action, documentary roles, and social relations are formulated. By treating specific clauses as potential sociological markers of legal identity, the analysis can test how the social context of documentary agents influenced formulaic content. It also shows why textual patterns cannot be separated entirely from the visual and material features through which notarial authority was expressed.

The shape of time

The final session, “The Shape of Time,” took up precisely this connection between textual analysis and documentary form. With three papers devoted to monastic archives from Southern Italy, the discussion was brought back to the document as a visual and material artifact.

Antonella Ambrosio–you remember her from her talk on microareas in Graz–examined a dossier of acts and manuscripts attributed to two “phantom notaries” from late medieval Naples, Ruggero Pappansogna and Dionisio di Sarno. Palaeographical and diplomatic analysis of more than twenty examples, fifteen of them illuminated, indicates that these texts are later reworkings, closer to the beginning of the sixteenth century than to their declared chronology around 1423. Watermarks and decoration support this later dating. Heraldic inserts, ornamental motifs, and miniated initials gave these rough yet meaningful visual objects a precise communicative function, connected with pastoral-visit inventories and noble strategies involving major Neapolitan ecclesiastical institutions. HTR can support the human interpretation and critical edition of such complex material.

If Ambrosio used visual clues to uncover a deliberately reworked chronology, Maria Pia Tiseo–yes, you remember her as well!–asked how graphic features could register slower processes of change. Her study drew on a corpus of 362 notarial charters from the monastery of Santa Maria della Grotta, dating from 1251 to 1499. The experimental analysis focused on thirty documents processed with Transkribus and eScriptorium and asked whether recurrent HTR errors could be treated not merely as transcription failures but as indirect evidence of graphic features. Seven selected features partially confirmed established palaeographical groups while also revealing more specific clusters connected to hands, local practices, graphic chains, and abbreviation systems. Despite the limited test sample, the method supplements palaeographical interpretation and encourages critical reflection on the categories used to describe continuity, breaks, and local rhythms in documentary writing.

The concluding paper shifted from graphic change to administrative rhythms while retaining the manuscript page itself as a data source. Marco Ciampo and Giuseppe Consolo analyzed two registers from the female Dominican monastery of SS. Pietro e Sebastiano in Naples. Register 1395 II records hundreds of notarial acts concerning leases, emphyteusis, sales, and rents between 1473 and 1495; the Quaterno de Polise (register 1400) records collections between 1475 and 1496. Comparing contractual obligations with later payments revealed a dual economic temporality. Liturgical feasts structured prescribed deadlines above all, the Assumption, the principal deadline for monetary payments, while actual collections followed a more distributed rhythm. A Transkribus and PAGE XML pilot linked these temporal patterns to the layout of individual entries, allowing obligations and payments to be compared as both data and material documentary units.

Three sessions, one research problem

What linked the three sessions was not a single method, but a common question: how can we trace change in documentary practices without losing sight of the institutions and individuals that produced the documents? The answers came from different directions, from the study of formulaic language and scribal hands to HTR, Computer Vision, and the analysis of administrative records. In each case, computational methods worked best when they remained closely connected to diplomatic, paleographical, and archival interpretation. The sessions, therefore, offered a useful overview of the research currently being carried out within DiDip and its wider network. Just as importantly, they created a space in which different corpora, methods, and scholarly traditions could be discussed together. Getting to Leeds had not been easy, but the conversations made the journey worthwhile.

Session Overview

Series organizer: Tamás Kovács, University of Graz
Co-organizers: Georg Vogeler, University of Graz (Sessions I–II); Antonella Ambrosio, University of Naples Federico II (Session III)

From Digital to Distant Diplomatics, I: Charters as Chronicles of Change
Moderator: Tamás Kovács

• 1535-a – Roberto Lamponi: Notarial Formularies in the Late Medieval March of Ancona, 14th–15th Centuries
• 1535-b – Daniel Luger: Continuity and Change in Papal Chancery Practice: Editing the Registers of Pope Honorius III
• 1535-c – Žarko Vujošević: Formulaic Language of the Royal Charters: Mirroring the Intellectual Elite of the Serbian Middle Ages, 14th–15th Centuries
• 1535-d – Johannes Laroche: Formulaicity and Change: Standardised Language in the Charters of Lambach Abbey, 14th–15th Centuries

From Digital to Distant Diplomatics, II: Decoding Diachronic Language
Moderator: Žarko Vujošević

• 1635-a – Tamás Kovács: Formulaic Language Analysis in Medieval Expressions
• 1635-b – Noé Leroy: Tracing the Evolution of Formulaic Language in European Episcopal Chanceries
• 1635-c – Franziska Decker and Giuseppe Consolo: Notarial Charters in Context: Computational Analysis of Agents, Networks, and Formulas in the Late Middle Ages

From Digital to Distant Diplomatics, III: The Shape of Time
Moderator: Daniel Luger; Respondent: Maria Pia Tiseo

• 1735-a – Antonella Ambrosio: Documents and Visual Clues of Temporal Manipulation: Two “Phantom Notaries” in Late Medieval Naples, 14th–15th Centuries
• 1735-b – Maria Pia Tiseo: HTR and Script Changes in Late Medieval Notarial Charters from Southern Italy: The Case of St Maria della Grotta Monastery
• 1735-c – Marco Ciampo and Giuseppe Consolo: Liturgical Deadlines, Economic Rhythms: Visual-Computational Study of Two Monastic Registers – Naples, 1473–1496

---

---