“Humanities for the 21st century.”
Interview: Samuel Schlaefli
The humanities of the 21st century will be computer-based,
believes Gerhard Lauer, Professor of Digital Humanities.
Digitalization is about more than just facilitating access to research
material, he explains – it brings completely new analytical
methods to the humanities.
UNI NOVA: Mr. Lauer, the digital humanities have been described as a revolution in the fi eld. Is this really an apt description?
GERHARD LAUER: It is indeed. Today, we areable to explore culture in unprecedented breadth and depth. Digitalization has done away with many of the boundaries that have traditionally shaped our fi eld of inquiry. In literary studies, we typically work our way through the canon by reading a small number of books very meticulously. But a person can read up to 4,000 books in their lifetime. Thanks to the digitization of source material, we can now in principle treat an entire century’s worth of publications as an object of study. The same goes for collections.
UNI NOVA: As in the case of the British Museum, which is digitizing its entire collection and putting it online?
LAUER: Exactly. Consider, for example, a medieval manuscript by St. Augustine. In order to look at its various parts – which is not even allowed in some cases – researchers used to have to travel to at least Paris, Geneva and St. Petersburg. This is why experts in St. Gallen and Fribourg created the “e-codices” project, a virtual library of manuscripts in which writings located all over the world are digitally pieced together. Now researchers can get to work in a matter of minutes.
UNI NOVA: It’s obviously thrilling that so much culture is being made available online. But do people actually look at it? After all, digitalization is also to blame for a surfeit of information and a general attention deficit in society.
LAUER: They do more than just look at it – they study it in great detail. In the “Perseus” project, researchers at Tufts University in Boston have put practically all of the literary works of antiquity online, as they felt this knowledge was relevant to how we think about democracy. The website
receives some 350,000 hits a month from all over the world. Ten percent of visitors download the original texts in Greek and Latin. Printed editions never reached that kind of audience. Numerous examples have shown that if you make texts digitally available, readers will come. In this sense, digitalization is expanding our concept of culture.
UNI NOVA: So far, we have spoken mainly about new channels for the dissemination of knowledge. Is this enough to denote a revolution in the humanities?
LAUER: No. Providing a gateway to our cultural heritage is just one part of what the fi eld of digital humanities sets out to do. Another, perhaps even more significant aspect of the revolution is that we can now analyze culture using methods that no longer rely solely on traditional reading, listening or observational skiln.
UNI NOVA: You mean machine learning – automated pattern recognition that becomes slightly more effi cient and “intelligent” with each pass.
LAUER: Just like the natural sciences, we now solve problems by formalizing and modeling them. An example of such a problem is understanding the plot of a novel. This is not something that a computer can do. But we can simplify the problem by breaking the novel down into 10,000 lexical units. We then look at whether positive words like “happy”, “wedding”, or “sunshine”, or negative ones like “depressed”, “dark”, or “tired” occur more frequently in each section. This allows us to digitally analyze the plotline and gain an insight into how novels are structured. Computers are unable to read or interpret, but they are good at counting. They can measure the distribution of words, colors or shapes much better than we can.
UNI NOVA: Isn’t it a little bit painful for a literary scholar to see world literature being sliced up and “read” by computers? The humanities of the 21st century will be computer-based, believes Gerhard Lauer, Professor of Digital Humanities. Digitalization is about more than just facilitating access to research material, he explains – it brings completely new analytical methods to the humanities.
LAUER: No – this is something that literary science has always done. Literary history classifi es texts into diff erent eras or movements, and groups them into genres. What the Brothers Grimm did was no different: they collected and compared fairy tales from all over the world in search of common elements, such as motives. We are doing the same, but with the help of computers.
UNI NOVA: What do research and publication look like in computer-assisted humanities projects?
LAUER: Projects of this sort are for the most part still very labor-intensive. Digitally performing a sentiment analysis to determine which texts Robert Musil wrote during World War I while he was a soldier in charge of a propaganda newspaper in Tyrol took us three years, including two weeks of computation time at a supercomputing center. The project resulted in two publications: one version for humanities scholars and another for the computer science nerds describing all the statistical techniques we used. There is no question that the work we do is changing.
UNI NOVA: A notable feature of your latest publications is that they are mostly team eff orts.
LAUER: That’s another thing that has changed. I find myself writing fewer and fewer publications alone. There are usually computer scientists or other humanities scholars involved as well. The various disciplines in the digital humanities are heavily reliant on each other – as has been the case in the natural sciences for a long time. The idea that “I am a lone genius and can do everything myself”, although still widespread in our field, is a thing of the past. Collaboration and division of labor will be crucial elements of the humanities in the 21st century.
UNI NOVA: Does digitalization devalue conventional skills in the humanities?
LAUER: Yes, and this is why it is encountering such resistance. We have a modernization problem. Imagine you are a renowned theology professor, and you have a PhD student who is using new-fangled computer-based methods to determine which letters where in fact written by St. Paul – a classic problem in theology, as many letters have been “falsely” attributed to him. You probably don’t fully understand what your PhD student is doing, but you suddenly realize that their computer-based research could make your own research tradition and skills obsolete. That can be disconcerting.
UNI NOVA: Are there any examples of the digital humanities not just making research more effi cient, but actually creating new knowledge?
LAUER: Absolutely. In literary studies, constructivists like Foucault and Barthes have long argued that the author of a text is irrelevant, claiming that authorship is merely a social construct. However, in computer science there are now entire competitions devoted to authorship attribution – the process of using computerbased methods to attribute texts to a particular author. And the analysis shows that authors do in fact have what we might call a style of their own. So we’re actually refuting highly popular constructivist theories.
UNI NOVA: In the past, the humanities and social sciences have had a reputation of viewing digitalization in a critical light, and focusing mainly on its negative effects on society. Are they now being won over by the promise of new opportunities?
LAUER: I would question your premise. What many humanities scholars do – and this is generally as far as their criticism goes – is write articles about how digitalization is destroying our culture and making young people fat and stupid. In the 15 years I have been working in this field, I have seen very little constructive input on the issue from humanities scholars. Who was it that campaigned for open access to ensure equitable distribution of cultural wealth, for instance? Computer scientists. And as for the new methods: in the natural sciences there is a Nobel Prize for them. In the humanities, they are first and foremost an object of criticism.
UNI NOVA: How are the digital humanities aff ecting teaching?
LAUER: Radically. Our students attend introductory lectures in computer science and learn about statistics. Incidentally, our close collaboration with computer sciences has opened up new potential for that field as well: a significant number of women, who are otherwise underrepresented in computer science, are drawn to the combination of culture and computer assisted methods.
UNI NOVA: How well-established are the digital humanities in Switzerland?
LAUER: ETH Lausanne and the University of Lausanne are leading the way. The latter was the fi rst university in Switzerland to off er a degree program in digital humanities. Geneva is currently advertising a vacancy for a professor in the field, and the University of Bern has been looking for one for some time. Digital humanities are a hot topic at a growing number of universities.
UNI NOVA: What is the situation like in the English-speaking world?
LAUER: The best place to do research in digital humanities right now is Nebraska.
UNI NOVA: Nebraska? Not Stanford, MIT or Harvard?
LAUER: No. Several high-caliber humanities scholars left Stanford University to establish a Center for Digital Research in the Humanities at the University of Nebraska, because there they had the freedom to do things they weren’t allowed to before. Even though Stanford is at the heart of Silicon Valley and has a techy reputation, the historians and literary scholars there still have a very traditional understanding of their field. This is why the canon of U.S. literature is being explored fi rst and foremost in Nebraska. Innovation often comes from the periphery, not the center – Max Weber was right about that.
UNI NOVA: Before you came to Basel, you tried to establish the digital humanities at the University of Göttingen, but came up against resistance from the faculty. Aren’t you worried that something similar might happen in Basel?
LAUER: No, in Basel the fi eld’s potential is acknowledged not just by the University president but also by the dean’s office directly responsible for our lab. Despite the very modest means at our disposal, the signs are encouraging. For example, work is underway on a Data Analytic Center where humanities scholars, sociologists, biologists and chemists will work together, as the computer-assisted methods employed by these disciplines often have a lot in common. The difference between analyzing letters or DNA is smaller than you might think.
UNI NOVA: How much interest have your colleagues at the University of Basel shown in the digital humanities?
LAUER: More than we can cater for with our modest resources. We are currently listed as a partner in ten applications for funding under the Swiss National Science Foundation’s Digital Lives scheme – besides five applications of our own.
UNI NOVA: What is your long-term vision for the Digital Humanities Lab in Basel?
LAUER: I hope to carry out computer-assisted analysis and humanities research, ideally in combination with experimental models such as eye tracking. My aspiration is to enable a humanities for the 21st century, at least for certain research issues. Although it is by no means a must for every discipline, I am convinced that this is the way forward for a growing subset
of the humanities.
UNI NOVA: Does that mean that in future, humanities scholars will spend more time sitting at their computers wrestling with probabilities and codes?
LAUER: Of course we are still something of an oddity at this stage – we’re the nerds with eccentric formal ideas. But if things go well, the digital humanities will eventually be incorporated into the various subjects in much the same way as is happening with bioinformatics at the moment. At some point, it will be perfectly natural for a student of art history or German studies to be profi cient in computing and statistical models.
More articles in the current issue of UNI NOVA.