The Flow

From Deep-Learning to Digital Analysis and their Role in the Humanities Creating, Evaluating, and Critiquing Workflows for Historical Corpora

Historical research increasingly makes use of digitization, benefiting from advancements in Handwritten Text Recognition and Natural Language Processing. This is where The Flow comes in and aims to promote the use of digital methods by developing a workflow that can be used by historians without expertise in information science and coding.

The Flow, running from 2023 to 2026, strives to create standardized digital workflows using existing technology, facilitating easier digital work with premodern historical sources.

The Flow is a joint project of the DH departments of the Universities of Bern and Bielefeld and the Research Centre for Hanse and Baltic History in Lübeck. The project is funded by the Swiss National Science Foundation (SNF) and the German Research Foundation (DFG).

The Project

The Flow - Advancing Historical Text Recognition and Analysis

Digitization is now an integral part of historical research. Historians and the humanities in general have much to gain from advances in machine-learning approaches that can transform text processing capabilities. However, their full potential remains largely untapped within the humanities. The tools and workflows to make use of new technologies are often only available to those who have a deep understanding of information science. Our project aims to bridge this gap by facilitating the widespread adoption and critical use of machine-learning technologies.

The Flow seeks to develop user-friendly digital workflows, making these powerful tools accessible to researchers beyond the realm of data scientists and coding experts.The project “The Flow” will develop more standardized digital workflows based on existing technology, making it easier for researchers to work with historical sources digitally. Through semi-automated processes and workflows established in the project, historians will be able to study longer time periods and gain a layered reading-based understanding of larger corpora of pre-modern manuscripts.

Sub-Projects

Four subprojects contribute to the project’s overall goal by analysing legal and administrative sources spanning different periods and regions, including England (13-14th centuries), the northern European Hanse area (14th-17th centuries), Switzerland (16-18th centuries), and Ethiopia (19th century).

The sources will be studied with digital methods, namely the use of (newly created) Handwritten Text Recognition and Natural Language Processing models for historical languages. Through a praxeological and institutional framework, our project tackles complex research questions surrounding societal processes, the practice of law, and its impact on everyday life.

Bernese Tower Books

Digital Humanities, University of Bern

Court Rolls

Digital History, Bielefeld University

Hanserecesse

Research Centre for Hanse and Baltic History, European Hansemuseum Lübeck

Activites

Our past and upcoming activities

  • 12 - 13 Juni 2025

    Paper Presentation “Go with the Flow: Wie Historiker:innen serielles Schriftgut digital auswerten (können)”

    Angela Huang & Silke Schwandt

    Workshop “Künstliche Intelligenz und historische (Justiz-)-Forschung”, Justus-Liebig-Universität Gießen

  • 20 September 2024

    Paper Presentation “Faktizität und historiographische Autorität von Machine Learning Output”

    Dominic Weber

    Digital History and Citizen Science Conference, Martin-Luther-Universität Halle-Wittenberg

  • 12 September 2024

    Paper Presentation “On the Historiographic Authority of Machine Learning Systems”

    Dominic Weber

    Digital History Switzerland 2024, University of Basel

  • 26 - 28 August 2024

    Summer School “A digital Workflow for Historical Corpora - from HTR to NER”

    The Flow Team

    Online / Research Centre for Hanse and Baltic History, European Hansemuseum Lübeck

  • 24 April 2024

    Paper Presentation “Archive im digitalen Zeitalter: Wie (halb)automatisierte Handschriftenerkennung unseren Zugang zu Quellen verändern kann”

    Angela Huang & Vivien Popken

    Schleswig-Holsteinischer Archivtag 2024, Nordkolleg Rendsburg

  • 22 April 2024

    Poster Presentation “Go with the Flow. Towards Reusable Workflows”

    Dominic Weber & Jonas Widmer

    Phil.-hist. Forschungstag 2024, University of Bern

  • 26 February 2024

    Workshop “not opaque flow - Workflows zur Aufbereitung und Auswertung historischer Dokumente”

    The Flow Team, Patrick Jentsch & Inga Kirschnick (both nopaque)

    DHd2024 Conference, University of Passau

  • 28 - 30 August 2023

    Workshop “The Flow - Kick-off”

    The Flow Team

    University of Bern

The Team

Meet the team behind The Flow

tobias.hodel@unibe.ch

Tobias is tenure track assistant professor in digital humanities at the University of Bern since 2019. He researches and teaches machine learning methods in and for the humanities. This includes the automated recognition of historical manuscripts, the extraction of information and the development of specific language models. Hodel holds a doctorate in history and leads research projects at the University of Bern, including on the tower books of the city of Bern from the early modern period, chat systems for university didactics in the 21st century and the historical telephone directories of Switzerland.

alhuang@fgho.eu

Angela leads the Research Centre for Hanse and Baltic History at the European Hansemuseum in Lübeck. Her research has long focused on the history of the German Hanse. With the Lübeck sub-project, she hopes to be able to provide other (Hanseatic) historians with digital tools and methods for their work.

silke.schwandt@uni-bielefeld.de

Silke is professor of Digital History at Bielefeld University. Her research focus lies on quantitative text analysis with a specialization in medieval history. She also works on the impact of digitality on the methodologies and theories of history as a humanities discipline.

christopher.kuhlmann@uni-bielefeld.de

Christopher has been part of The Flow since July 2023 with the aim to pursue his PhD in it. His main research focus is on ‚essoins‘ - excuses for not appearing in court in English common law. He analyses court rolls of the 13th and 14th centuries with a focus on group influences and affiliations, as well as intersectionality, using customised HTR and NLP models.

ilange@fgho.eu

Inga is in charge of the Lübeck sub-project on The Flow. During her studies, she already worked extensively with manuscript sources from the pre-modern period. In the project, she is interested in analysing the Hanserezesse in detail as one of the most important sources of Hanse history. By using digital methods, she hopes to conduct a long-term study on the development of the German Hanse for the first time.

dominic.weber@unibe.ch

Dominic is part of The Flow since July 2023. In his PhD project he works with the Bernese Tower Books (interrogation protocols from the early modern period). He employs various machine learning applications for handwritten text recognition, information extraction and document clustering. Simultaneously, he theorises the epistemological consequences of conducting historical research based on data generated by machines and humans alike.

dana.meyer@uni-bielefeld.de

Dana studies Cognitve Informatics at Bielefeld University. She has been part of The Flow project since July 2023. She is interested in machine learning in the context of natural language processing.

jonas.widmer@unibe.ch

Jonas has been working for the Digital Humanities at the University of Bern since February 2021. As a data scientist he supports The Flow since July 2023, with focus on Natural Language Processing and Handwritten Text Recognition. He is developing and providing digital services/tools for the work with digital sources.

bholterman@fgho.eu

Bart is a historian of the late Middle Ages and early modern period, with a specialization in economic history of the Hanseatic region and the application of digital methods. Employed at the FGHO since June 2024, he will assist in transcribing the Hanserezesse for training HTR models.

anna.funk@unibe.ch

Anna is a student assistant for the Digital Humanities in Bern and has been working on The Flow since July 2023. She is responsible for data generation and pre-processing. Based on the source corpus of The Flow - the interrogation records of the Bernese Tower Books - she investigated witch trials in early modern Bern in her bachelor’s thesis.

jneubauer@fgho.eu

Judith has been a project assistance at the FGHO since April 2023. In her studies, her focus has been on the history oft the North and Baltic Sea region in the Middle Ages. In The Flow, she is assisting with the transcription and further editing of the Hanserezesse.

melvin.wilde@uni-bielefeld.de

Melvin works for Bielefeld University as a student assistant in The Flow. There he mostly handles the technical side of historical work such as annotations, guidelines, transcriptions and data management. He specializes in medieval history with a focus on gender/ masculinity studies.

Alumni

Former team members of The Flow. Thank you for your contribution <3

Myrjam Hostettler (Student Assistant, University of Bern)
Vivien Popken (Research Assistant, Research Centre for Hanse and Baltic History, European Hansemuseum Lübeck)