News

In a recent research seminar, scholars from various Danish universities and beyond gathered to explore innovative approaches to bridge the gaps and biases in historical archives. The seminar featured Folgert Karsdorp, who presented his work on applying statistical methods originally used in ecology to humanities research. These methods, known as “Unseen Species Models,” offer a valuable tool for estimating missing historical data and understanding the factors behind those gaps.

Karsdorp showcased the versatility of these models through three case studies. The first case delved into the realm of European medieval literature where “Unseen Species Models” were employed to estimate the number of lost books within a specific genre, shedding light on the extent of losses across different geographic regions.

Next, the model was used to estimate the number of sailors employed by the Dutch East India Company, even though their contracts didn’t survive. Using the same statistical framework, they estimated the size of the workforce, providing valuable insights into the maritime history of the era.

The third case study ventured into historical criminology, utilising a generalisation of the unseen species model. This case analysed police registers from the Amigo prison in Brussels (1879-1880) to estimate the “dark number” of unapprehended perpetrators and uncover the demographic composition of this elusive group.

The seminar didn’t stop at theoretical discussions. In the second half of the event, participants had a hands-on opportunity to apply these ecological methods themselves. They explored a new dataset, the correspondence of Charles Darwin, to estimate the number of lost Darwin letters. This practical exercise sparked interesting observations and potential research collaborations among attendees.

The workshop brought together scholars from various Danish universities, including KU, AU, and AAU, along with participants from outside of academia, and the focused event provided an invaluable opportunity to delve deeper into a single method and topic, fostering a better understanding of its strengths and limitations than could be achieved at a standard conference. 

“It was great spending half a day focusing on one method, one topic, and going a bit more in-depth. People also had interesting comments, and I left the workshop with a better idea of the method’s strengths and limitations than if I saw this at a conference,” says Jan Kostkan, PhD Student at Center for Humanities Computing, Aarhus University.