Düring was the first of four CDHI Postdoctoral Fellows in 2014. As a CDHI Postdoctoral Fellow, he held an appointment in the Department of History where he is teaching Digital History: Trends, Challenges and the Future of the Historical Method.
During his time at UNC-Chapel Hill, he worked on Machine-based Extraction of Relations in Text (MERIT), a research project developed together with Antal van den Bosch of Radboud University Nijmegen. MERIT utilizes tools developed in computational linguistics for a multi-perspective analysis of eyewitness accounts of end of the Second World War in the Arnhem/Nijmegen region and for the comparative analysis of memories of these events. He is also collaborating with colleagues in the Digital Innovation Lab on a project that uses slave narratives to analyze covert networks associated with the Underground Railroad.
Marten Düring studied cultural history at the Universities of Augsburg and Manchester with a focus on the history of the Second World War. He also has an interest in interdisciplinary and computational research methods in the humanities and history in particular. Previously, he consulted with the Centre Virtuel de la Connaissance sur l’Europe (CVCE) on a network-based tool for the analysis of digitized primary sources. He also worked on event extraction from Dutch and US newspapers together with a team of Computational linguists at Radboud University Nijmegen. In July 2012, he successfully defended his PhD thesis which introduces a relational perspective to the analysis of help for persecuted Jews during the Holocaust. To date, this is the first formalized analysis of relations between helpers and recipients of help, and among the first projects which apply formal network methods in Contemporary History.
Over the last few years, Düring has worked to bring together people and resources around the use of network analysis in the study of history. In particular, he has organized a workshop series on the topic and established Historical Network Research, an extension and outgrowth of the workshop series that bundles information on network analysis in the historical disciplines.
And now, a few words from Marten:
What first brought you to the digital humanities, and how has it influenced your research and teaching?
For my PhD project I worked with social network analysis tools and was intrigued by their ability to reveal patterns in historical sources I would not have discovered otherwise and more generally, to offer a new view on the materials I was studying. After I submitted my thesis in 2012 I knew that I wanted to continue doing historical research using software tools and the DH world turned out to be a great place to do just this.
I tremendously enjoy the openness, excitement and creativity within the DH but I am also aware of the limitations, buzz and some unfulfillable promises. The experience of both led me to become more and more interested in building bridges between computational and hermeneutic approaches to text.
Tell us more about the digital history course you are teaching this semester. What are you most excited to share with your students?
The course is organized around the themes of online communication, source digitization and software-based analysis. All three have changed significantly and will change even more in the years to come. As historians we are no longer limited to producing monographs every few years, which are available at high prices and targeted at a select few. Social media, blogs, open and collaborative writing (see e.g. The Historian’s Macroscope: Big Digital History), open access publishing and multi-layered websites (e.g. Freedom’s Ring: King’s “I Have a Dream” Speech) explore new ways for historians to write history and for audiences to engage with it following their own interests. Digital research platforms such as ePistolarium offer promising ways to link primary sources to analytical tools which require little technical expertise but allow us to ask very different questions.
Digital History is still very young. There is some awareness of what could be possible and ideas about how to do it but little practical experience. This also means that at this point there are far more open questions than answers. Again, I think that the combination of challenges and opportunities is what excites me about the course.
Your dissertation used network analysis to uncover covert support networks during the Holocaust. Now that your dissertation is done–what’s next? What projects are you currently working on?
For the past two years I have worked on a number of smaller projects which combined tools developed in computational linguistics and historical primary sources. What fascinates me is the power of these tools to process huge amounts of texts and to independently identify patterns beyond the keywords we feed them.
At the Digital Innovation Lab I am working together with my colleagues Pam Lach and Annie Chen on a well-known collection of some 250 slave narratives hosted by the Documenting the American South archive. First we want to find out to what extent we can automatically identify different topics in these narratives as well as spot mentions of slaves, slaveholders and those who helped slaves in general. Secondly we ask whether we can cross-reference information on individuals with sources in other digitized archives and Southern newspapers which are currently being digitized as part of an arrangement with UNC and Ancestry.com. What I share with the DIL’s approach to DH is the intention to simplify software as much as possible and thereby making it accessible to a wider range of users inside and outside the academy.
What is a day of DH like for you? (They are all so different, so feel free to focus on a particular day from the last week or so!)
I guess not so different from every other postdoc’s: juggling research time, readings, meetings, paper deadlines and teaching. Probably the only real difference is that I also spend some time programming to process my sources.
Do you have a favorite DH project or two? Tell us about them, and why they are your favorites.
It’s probably ePistolarium a research platform on the Dutch Republic of Letters which I have already mentioned. In many ways I think it is a good model for what Digital History should be about: it is open for anyone to approach it with his/her own questions, it lets you choose whether you want to work with basic keyword and faceted search or with more advanced geographical or social network visualizations. And it makes sure that there is always a direct link between software analysis and the underlying primary sources.
You’ve come a long way to join us here in North Carolina. Any favorite things about NC so far—DH or otherwise? What are you most excited about exploring while you are here?
Regular sunshine ranks very high on the list of favorites and I look forward to more travels to the seaside and the mountains – and finding a good mountain bike trail.
Düring is currently a researcher in the Digital Europe Integration Studies (DEIS) unit at the University of Luxembourg.