Ressources numériques en sciences humaines et sociales OpenEdition Nos plateformes OpenEdition Books OpenEdition Journals Hypothèses Calenda Bibliothèques OpenEdition Freemium Suivez-nous

After Leipzig: a participants’ overview on the First Franco-German Summer School on Digital Humanities in Area studies

Par Longhai Gao (Nanjing University/University of Munich), Nguyen Le Thuy Tien (École pratiques des Hautes Études and École nationale des Chartes), Heldiza Sánchez (University of Wroclaw/University of Leipzig)

1. Introduction

From July 7th to 11th 2025, the Leipzig Research Centre Global Dynamics (ReCentGlobe) and the French consortium DISTAM (DIgital STudies Africa, Asia, Middle East) jointly hosted the inaugural Franco-German Summer School on Digital Humanities and Area Studies at the University of Leipzig. The school embraced diversified research themes with an emphasis on interaction. Four core areas were on focus: working with text using handwritten text recognition (HTR) and optical character recognition (OCR); data visualization; spatialization and its digital representation via geo-information systems (GIS) and related methods; and open data and data sharing. These themes collectively covered almost all domains of digital humanities research. Additionally, presentations of specific projects showcased distinct cross-regional, cross-temporal, and interdisciplinary perspectives. Attendance was highly international, encompassing over 50 young scholars from Germany, France, Italy, China, Vietnam, Britain, Slovenia, among other countries. Their research covered an array of languages, including Latin, German, French, Arabic, Armenian, Chinese and various African languages. Guided visits to the Bibliotheca Albertina and its Manuscript Centre, as well as the Grassi Museum of Ethnology, offered complementary firsthand experiences of digital humanities cultural practices in Leipzig, at a time when global demands for the repatriation of non-Western artefacts complexify the work of institutions in charge of tangible heritage.

The didactic part of the event featured three main activities : Keynote Speeches followed by Q&A, Roundtable Discussions, and Parallel Workshops. Willing trainees were, furthermore, offered the possibility to present their research in a poster session.

2. Keynote Speeches and Roundtable Discussions

Keynote Speeches were delivered on the basis of individual or team projects, with a stronger emphasis on practice. Each talk presented a relatively mature case, demonstrating the entire workflow of implementing a digital humanities project and thus providing valuable references for participants. These presentations not only outlined the specific objectives, methodologies, and technical frameworks of the projects but also highlighted the practical challenges encountered during implementation and the corresponding solutions. By showcasing such well-developed cases, they offered participants concrete models to draw upon, helping them better understand how to translate theoretical concepts into actionable research practices in the field of digital humanities.

Daniel Kinitz (Saxon Academy of Sciences and Humanities), in his lecture entitled “Data-driven intellectual history? The challenges of building a research platform for Arabic manuscript cultures”, introduced the Bibliotheca Arabica project, which has established a large-scale database of Arabic manuscripts aiming to provide new perspectives for traditional Arabic studies. Through the specific case of “Commentarial Relations,” the project explores the complex relationships (political, cultural, religious, etc.) between manuscript commentaries and elaborates on the current challenges encountered, such as digitization costs, policy changes, and ethical and legal issues.

Shih-Pei Chen (Max Planck Institute for the History of Science) presented “Treating a genre as a database: a digital research methodology for studying Chinese local gazetteers”, in which she introduced “Local Gazetteers Research Tools (LoGaRT)”—a digital humanities research tool designed for searching, analyzing, and collecting digitized data from Chinese local gazetteers. Using concrete cases, she demonstrated how large-scale gazetteers reveal the underlying knowledge structures and their evolution over time.

Cosima Wagner (Freie Universität Berlin & NFDI4Memory) delivered “Multilinguality and data access: an area studies librarian’s perspective”, which addressed the cultural and technical biases toward English in digital tools and infrastructure, and proposed a multilingual and sustainable knowledge management solution.

Eiloghosa Obobaifo & Iwinosa Oyakhire (Institute for Digital Heritage) focused on the Digital Benin project in their presentation “Digital Benin: a digital platform connecting the displaced royal artefacts from Benin Kingdom.” This platform aggregates digital records of excavated artifacts looted from the Kingdom of Benin (now southern Nigeria). By utilizing platform data and integrating oral histories and local knowledge systems, the project aims to develop a more nuanced understanding of the artifacts’ origins and cultural significance.

Eva Ommert (Leipzig University & NFDI4Memory) delivered “How to think of your research in terms of data management and FAIR,” offering suggestions and insights for digital humanities researchers from the perspectives of research data management, data organization, data documentation and metadata, legal and ethical issues, and archiving and reusing research data.

In addition, the summer school hosted two roundtable forums, focusing on the digital humanities turn in area studies, along with its applications and challenges. The first forum, titled “The digital turn in area studies – Franco-German perspectives in the age of artificial intelligence,” featured discussions among Mercedes Volait (CNRS & DISTAM), Matthias Middell (Leipzig University & NFDI4Memory), Anne Lipp (University Library Leipzig), and Sandra Aube (CNRS), where experts shared their views on the digital turn in area studies. The second roundtable forum centered on “Artificial intelligence, open access, and ethical dimensions,” with Chahan Vidal-Gorène (École nationale des chartes & Calfa), Torsten Wollina (De Gruyter/Brill), Mareike König (German Historical Institute, Paris), and Cosima Wagner (Freie Universität Berlin & NFDI4Memory) and highlighted in particular the importance for scholars to promote Open Science in the field of academic publishing, in order to gain autonomy from commercial ventures.

3. Parallel Workshops

3.1 Working with Text (from Longhai Gao’s perspective)

The “Working with Text” parallel track consisted of four workshops, where instructors delivered introductions to the working principles of automatic text recognition; practical operations including automatic recognition, extraction, proper storage, and critical editing of text data; construction and training of corpora; and application of corpora in relevant research.

Marie Bizais-Lillig (University of Strasbourg & DISTAM), in her workshop “Introduction: How to Structure a Corpus of Text,” focused on the formats and methods for text editing. After standardizing the text, she provided solutions for structuring, documenting, and enriching text editing formats.

Chahan Vidal-Gorène (Ecole des Chartes & Calfa) presented “Detecting and Deciphering Damaged Medieval Armenian Inscriptions Using YOLO and Vision Transformers.” Using Armenian ancient inscriptions and art historical corpus research as case studies, the workshop introduced the basic principles of image annotation and trained a new model for image detection, which facilitated subsequent research. This model training method can be extended to the recognition and processing of any type of corpus, making it one of the most widely used approaches in the current field of digital humanities.


Noëmie Lucas (University of Edinburgh) and Chahan Vidal-Gorène (École nationale des chartes & Calfa) co-led “Automated Text Recognition: OCR and HTR Solutions for Non-Latin Languages,” which focused on automatic text recognition for non-Latin languages and explored the construction of solutions using existing OCR/HTR methods. The workshop not only offered an introductory guide to these methods but also discussed the challenges, difficulties, recent advancements, and new perspectives in applying OCR to non-Latin languages.


Finally, Noëmie Lucas (University of Edinburgh) conducted “Exploring Digital Methods for Arabic Texts and Manuscripts: Handwritten Text Recognition (HTR) and Text Re-use Analysis.” Focusing on Arabic manuscripts, the workshop delved into the technical details of OCR/HTR and project implementation. Additionally, it demonstrated text re-use analysis and explored how to deeply integrate research questions with digital humanities methods.


In contrast with other similar initiatives, such as the 2022 Digital Humanities Summer Workshop co-hosted by Peking University and Harvard University, the Franco-German Summer School featured more international themes and a broader geographical scope of research. It not only included text analysis of major European languages such as English, German, French, and Latin but also covered texts from African countries, Armenia, and the Arab world. Moreover, it incorporated emerging content on large AI language models, applying them to model training developed by researchers themselves to enhance the accuracy of text or image recognition. In comparison, the project practices at the Peking University workshop were mostly related to Chinese history and philosophy, with technical emphasis on R-based text analysis, resulting in a relatively narrow research scope. The Franco-German Summer School enabled participants to gain insight into the entire workflow of digital humanities projects—from automatic text recognition, extraction, and storage to corpus and large model training, and further to critical text editing, research, and corpus application. It also provided a deeper understanding of diverse research pathways across different projects, which is crucial for expanding academic horizons. For participants, this summer school not only offered opportunities to learn more digital tools and methods amid the rapid evolution of digital technologies, thereby forming a clearer and more concrete understanding of the overall process of digital humanities research projects, but also strengthened the resolve to complete doctoral dissertations using digital methods through interactions with instructors and peers.

3.2. Data visualization (from Nguyen Le Thuy Tien’s perspective)

The data visualization workshops included introduction sessions about map conceptualization and development or geographical representations for Digital Humanities research projects, along with practical workshops to get used to some new tools commonly employed in network data analysis and visualization, such as Nodegoat and Gephi, both widely used in Digital Humanities projects.


The “Introduction to Reflective Data Visualisation” workshop led by Jana Moser (Leibniz Institute for Regional Geography) invited to revisit one’s approach to mapping and data visualization. Indeed, Moser clarified that focus would not be on technical aspects such as GIS tools, but rather on conceptual thinking and critical reflection. She emphasized that there is no single “correct” map— and that different maps serve different purposes, depending on the intended message and research question. 
Maps are frequently viewed as purely technical tools for precisely representing location. However, this workshop opened one’s eyes to the many more possibilities offered by mapping: to tell stories, reflect historical or political contexts, and even reveal underlying biases. Moser presented several examples, such as propaganda or distorted maps, due to political borders shifting, and explained how colors and scale can shape the viewer’s perception. As a matter of fact, she recommended using neutral backgrounds such as light gray to ensure that data remains the focal point. She also warned against modern basemaps tools, like OpenStreetMap, to use for historical data, as contemporary features may distort the temporal accuracy of the visualization.
 Moser also introduced alternatives to conventional territorial maps, such as dot maps, proportional symbol maps, and cartograms, which can more effectively convey information such as population or election results. She referenced Jacques Bertin’s theory of visual variables — such as size, color, texture, and shape — and explained how these should be aligned with the type of data being visualized (nominal, ordinal, or quantitative). For example, using a color gradient for nominal data like profession may imply a false sense of hierarchy.
 Another aspect of the workshop was the debate among participants, some of whom presented their own visualizations for critique. Moser’s feedback helped highlight the importance of understanding principles such as map projection, data classification, and color usage — all of which significantly affect how results are interpreted. The workshop inspired one to consider building interactive maps for future research, or even creating story maps that combine narrative text, imagery, and geolocation to convey more engaging stories.


In the  workshops about Nodegoat (led by Kathleen Schlütter, Universität Leipzig & 4Memory) and Gephi (led by Roman Kuhn, Berlin State Library) participants were first introduced to a web-based platform tailored for data modeling, management, and visualization, along with spatial and temporal dimensions. Nodegoat is particularly powerful for its ability to customize data structures, assign geographic coordinates as well as time markers to entities, and visualize them through dynamic maps and interactive timelines. It also supports collaborative work, with options to share research projects privately or publicly, and can easily embed charts or maps into websites.
 During the hands-on session, participants were guided through each step — from accessing Nodegoat, creating a data model, importing CSV files, down to data visualization on maps and network diagrams. This experience made one realizes the potential of Nodegoat not only for organizing research data but also for presenting it as a narrative.


In the Gephi workshop, participants  gained a better understanding of network analysis within the context of humanities and social sciences. Roman Kuhn presented fundamental concepts such as nodes, edges, modularity, and centrality measures like degree and PageRank. He also emphasized that visualizations are never entirely neutral — choosing the wrong layout or metric can lead to misleading interpretations, especially when dealing with historical or textual data.
The practical component allowed participants to explore layout algorithms such as ForceAtlas2 and GeoLayout to construct and analyze networks. Particularly impressive was the Voltaire’s letter network case study, where identifying communities, intermediary nodes, and influential actors sparked ideas for one’s research, such as  the relationships between Buddhist texts and scholarly networks in colonial Indochina.

3.3. Spatialization (from Heldiza Sánchez’s perspective)

In terms of spatialization, the first workshop held was “Introduction to Spatial Humanities I” led by Dr. Thomas Efer (Leipzig University). The workshop covered spatial humanities while implying that digital workflows also go beyond geographical information systems. This allowed trainees to discuss further the attributes of interest that one is given when combining dynamism and ambiguity, and what constitute the elements of space, time, and place. In this sense, spatial humanities redefine research by reinserting and enabling one to craft questions about how geography, human behavior, and cultural development coexist in the same space.
 Participants were then introduced to various applications that use and put into practice the above, from those that provide locations from a historical perspective to others that provide coordinates from text. In addition to these reference systems, which can sometimes involve a “deep map,” combined with a breviary exercise where the class collaborated on a visual representation of networks, it was possible to reveal networks based on the following: human experience and the assignment of categories based on temporality.


Another workshop in the spatialization track was “Geolocating place names in texts and rasters with Recogito,” delivered by Yanis Mokri (Paris 1 University and DISTAM). Recogito differs from traditional forms of text processing because it allows georeferencing and enrichment using place names from resources such as digitized texts and bitmaps.
The activity on this occasion consisted in uploading a dataset, familiarizing oneself with the tools, identification, and downloading the document. This tool represents an advance as it has control features and is compatible for collaborations with access level designations for specific users, contributing to the identification and annotation of locations and people in a text. Since Recogito was a funded project for few years, it is no longer updated. It was replaced by a new sustainable platform developed by members of the first team called Recogito Studio.

4. General assessment

A general assessment among participants was that exposure to such an array of methods and tools shifted significantly their perspective on digital humanities. Other formats for workshops could have been conceived, such as the grouping of trainees
 into teams of 4–5 members, with each team collaborating to complete and present an academic project under the guidance of instructors, focusing on a specific research question. Such collaborative hands-on approach to specific projects could have maximized the integration of theory and practice.
But globally, the Summer School led participants to recognize the field of digital humanities as a space where conventional and digital practices converge, and enabled to create and innovate new ways and alternatives to visualize, study, analyze, critique, as well as generate research and information beyond the limitations imposed by contemporary constraints. The applications provided enabled trainees to construe, design, and interpret data in a manner that is not only easier to observe and classify, but also implies a sense of responsibility regarding information management, a commitment to data sharing, and an opportunity to evaluate and assess how knowledge can be made inclusive.


OpenEdition vous propose de citer ce billet de la manière suivante :
Mokri Yanis (29 septembre 2025). After Leipzig: a participants’ overview on the First Franco-German Summer School on Digital Humanities in Area studies. Carnet hypothèses du consortium Huma-Num DISTAM (DIgital STudies Africa, Asia, Middle East). Consulté le 11 février 2026 à l’adresse https://doi.org/10.58079/14rqf


Vous aimerez aussi...

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *

This site uses Akismet to reduce spam. Learn how your comment data is processed.