Monthly Archives: April 2025

Hätte ich das mal eher gewusst … mit Wouter Kreuze

Can you describe your project in three sentences and explain the digital component of your work?

In early modern times, handwritten newsletters circulating throughout Europe had the same structure; beginning with headlines, introducing where and when the news originated followed by the contents of the news. These discussed a wide arrange of subjects in short paragraphs, mostly covering political, diplomatic and military issues. My project facilitates the collection of this material leveraging existing technologies (namely, the software kraken) for automatic lay-out and text recognition. It enables the structural comparison between different collections enhancing our understanding of early modern information fluxes. To better understand how newsletters circulated, I decided to include collections from different countries, namely Germany, Italy and the Netherlands. These were often collected by either merchants, or people involved in political administration. The comparison between different collections helps explaining the creation and translation of newsletters.

Added value: What was only possible or what advantages did you have because you worked digitally?

Early modern readers of news usually received multiple newsletters a week. I included several collections of newsletters to give a proper assessment of the news system as a whole. To analyse trends over time, I focused on a range of years (1598-1604). Effectively, that means that my body of archival material runs in the thousands of documents. There is no other way of doing this within one lifetime, if not with digital methods.

Once the data is collected, the question of analysis begins; the automatically segmented documents are digitally analysed. Simple methodologies can provide a solution, such as statistics about where and when the newsletters were published. The digital realm, however, also offers more advanced inroads. For instance, I have applied methodologies for the automatic retrieval of textual similarity. That way, we can easily understand which readers received the same newsletters. Usually, newsletters were not undersigned, and thus anonymous. As the news compilers also freely copied from each other, it is often not possible to establish clear authorship. In my opinion, the best way of make sense of the news system is therefore to study their adaptation and dissemination.

Workflow for the digitisation of an early modern newsletters (also known as an ‘avviso’). We start out with plain photographs of the newsletters. In the first step, the program recognizes the headlines (blue)  and news items (pink). In the second step, these fields are automatically transcribed. The third step describes how the newsletters automatically recognize. The images are created with help of the software kraken/eScriptorium, using material from the Fuggerzeitungen, held by Österreichische Nationalbibliothek.

Had you worked with digital methods before?

My first extensive experience with digital humanities came during my PhD at University College Cork. Looking back, I believe I was primarily hired for my palaeographic and linguistic skills, though the doctoral degree itself was within a DH program. While I had some prior exposure to digital methods, it was nothing compared to what was required for the degree. It was a steep learning curve, but it has allowed me to work on projects like these involving Handwritten Text Recognition and visualizations. Otherwise, I would not have been able to undertake these.

What triggered your decision to embark on a digital project?

Before commencing my doctoral journey, I had obtained degrees in History and Italian. These programs were very inspirational, but far more built around traditional methods. While digital methods were not completely omitted, it was not as integral as in the PhD program. 

My dissertation was written within the framework of a bigger project, called EURONEWS. This aimed to be a digital project. As I was in a DH program, the development of a digital angle was required for my degree. Applying digital methods has greatly enriched my toolbox as a historian. For instance, it helped me to manage large quantities of source material, order my records and create visualizations.

Digital projects are known for a steep learning curve. What are five tips you would like to give to your younger self? What do you wish you had known, what would you have done differently, what turned out to be eye-opening for you?

  1. Don’t be afraid to experiment There are many different digital tools available, and sometimes it takes a while to understand what you really need. But perhaps even more important, your understanding of your sources and data greatly improves while working with them.
  2. Learn step by step according to your necessities There is a lot to learn, but you do not need to master it all at once. When I worked on my dissertation project, I went step by step. The basics I learned from a python course I found online. From there, I moved from one problem to the next by googling the challenge I was tackling, usually leading me to Stack overflow. For instance, initially, I was simply curious to see which words were most frequent in my dataset. To complete this in a relatively straightforward task, I wrote a simple python script. Back then, that was already a challenge, as I had no prior experience coding. Having mastered this, I started counting other things. From there, I became interested in creating visualizations, a slightly more complicated task.
  3. Things that appear too advanced at first, become self-evident over time When you learn a natural language, it can initially appear an insurmountable task. Once attaining fluency, however, it’s almost unimaginable that one day you were not able to understand utterances in that language. With coding languages and other digital skills, it is no different.
  4. Don’t try to reinvent the wheel Most of the digital tasks you want to undertake as a digital humanist have already been conducted by others. Therefore, you can build on the work done before you.
  5. Allow yourself to take a break Sometimes a little distance is what gives you clarity. Meet your friends, take a stroll, go for a swim. Sometimes the answers you are looking for come to you when you set yourself at your task again with renewed powers.

Wouter Kreuze obtained his degree in Digital Humanities from University College Cork, Ireland with a dissertation based on the newsletter collections of the Fugger and Medici. At the German Historical Institute, he developed a new workflow for the digitization of handwritten newsletters. Subsequently, he continued this project with a fellowship from the Folger Library. Currently, he is involved in the MALCONTENTS project at the University of Limerick, focusing on social deviancy among the professoriate of Early Modern universities in Germanophone areas. (ORCID-ID: https://orcid.org/0009-0003-4039-7495)

Die Blogreihe „Hätte ich das mal eher gewusst … Was Historiker:innen und historisch arbeitende ihrem jüngeren Ich raten würden“ wird betreut durch das GHI Washington in Kooperation mit NFDI4Memory / Task Area Data Culture, dem Verband der Historiker und Historikerinnen Deutschlands (VHD) und der Digital History Professur der Humboldt-Universität zu Berlin. Wir freuen uns über alle, die offen über ihre Erfahrungen mit digitalen Projekten im Geschichtsbereich sprechen möchten – und Tipps geben, was sie auf jeden Fall beim nächsten Mal anders machen würden.

Sie haben / Du hast auch Interesse, einen Beitrag für die Reihe zu schreiben? Gerne! Christiane Weber (VHD/NFDI4Memory) [weber@historikerverband.de] und Katharina Hering (GHI) [hering@ghi-dc.org] freuen sich auf eine Email.