“What has digital technology changed in our relationship to research?”

This post is the English translation of a statement given at the Journée de la recherche en Lettres at UNIL on March 12, 2021, in response to the question: “how has digital technology changed (or not) your relationship with research?”

I first studied computational linguistics, English and applied linguistics in Erlangen. The research projects I conducted during my studies (in seminars and for my dissertation) were therefore already “digital” and also “cutting edge.” In my master’s thesis, I worked on improving information retrieval through morphological analysis. The decomposition of compound words, for example, can significantly improve search results for languages like German.

Later I worked on e-learning; in Magdeburg we were simultaneously supporting computer science education and investigating innovative forms of teaching. On the one hand, we were exploring new ways of distributing “content,” which, like information retrieval, is part of document engineering. That’s what my PhD thesis is about. On the other hand, we were studying how to automatically verify programming exercises and provide useful and meaningful feedback.

My interest in document engineering then led me to the Law Sources Foundation in Zurich. Here, too, it was a mixture of concrete tasks—making editions available and searchable online—and research into the automatic analysis of these old, multilingual texts, as well as support for editors.

At the Leibniz Institute for European History in Mainz, I then worked on digital research infrastructures, which required an understanding of historical research issues, but which also raised the question of what digital historical research could be if we thought beyond databases and digitization of sources.

Thus, over the past twenty years or so, I have witnessed and, in some cases, contributed to the evolution of several fields that are now involved in what is called the digital humanities. One thing is certain: twenty years ago, one could only dream of the technical possibilities and data that are available to any researcher today.

It is from this mix of experiences and interests that my current research interests were born. For at some point, the question arises whether it is only about processing more data faster, whether the role of digital in the humanities is limited to digitization, whether “digital humanities” only means using “tools” or doing quantitative analyses and visualizing them nicely.

While individual research projects in the digital humanities are often impressive, I observe that digital humanities as a field remains largely under-theorized. This begins with the question of what “the field” actually is. This may seem counterintuitive, but debates about the definition of the digital humanities result from a lack of theoretical reflection rather than overproblematization; this is underscored by the fact that conclusions most often boil down to “digital humanities is undefinable.” I believe, however, that only a precise definition will allow us to examine the epistemological questions that the “digital” raises with respect to humanities research.

My definition of digital humanities focuses on the construction of computational models. This definition has several important properties:

  1. The construction of computational models is a truly distinctive feature that marks an epistemic shift rather than a superficial feature, such as “using digital tools.”
  2. The construction of computational models as a specific type of formal models places the digital humanities in a tradition whose beginning is well before the development of the modern computer, starting, among others, with Llull and Leibniz.
  3. Building computational models requires formalization, but this does not imply quantification. The digital humanities are not quantitative per se. As Gilles-Gaston Granger (1960/1967, p. 113) has remarked, formalization in the humanities “consiste dans le passage de l’astructuré au structural, plutôt que dans une quantification.”1 The digital humanities are, to a certain extent, mathematical, but the word “mathematical” must be understood as referring to mathematics as the “science of structures” rather than as the science of numbers and space.

In retrospect, my research has always been “digital.” Computational linguistics, e-learning, document engineering, digital humanities; the labels vary. I find myself, as Silvio Ceccato said, “un tecnico fra i filosofi(Ceccato, 1964).


Ceccato, S. (1964). Un tecnico fra i filosofi. Padova: Marsilio.
Granger, G.-G. (1967). Pensée formelle et sciences de l’homme (Nouvelle éd. augmentée d’une préface). Paris: Aubier-Montaigne. (Ouvrage original publié en 1960).
Granger, G.-G. (1983). Formal thought and the sciences of man. Dordrecht: Reidel. (With the author’s Postface to the English edition (1982)).

  1. “It seems then that a scientific elaboration of qualitative notions consists in the transition from the astructured to the structural, rather than in a quantification.” (Granger, 1983, p. 91)↩︎

Image credits: Thomas Schanz, CC BY-SA 3.0, via Wikimedia Commons

Author: Michael Piotrowski

Computational linguist, computer scientist, professor of digital humanities at the University of Lausanne.

Leave a Reply

Your email address will not be published.

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

Search OpenEdition Search

You will be redirected to OpenEdition Search