Last week, we did attended in the DATeCH2014 conference and Digitisation Days in Madrid. As we (the staff of DPKL) had a focus on the more technical issues, our fellow from Kirjallisuuspankki (The Bank of Finnish Literature), Anna Biström, shares her views on the event from a researcher’s point of view.
Home again since a few days it’s time to reflect: What can a literary scholar gain from participating in the Digitisation days and Datech conference, May 19-20. Well, except travelling to beautiful Madrid and being in the same room as Mario Vargas Llosa (yes it’s true!), who was giving a talk in connection with the Succeed prize ceremony.
My hope was to learn a little bit more about the field and perhaps get some useful ideas as a beginner in a project for digitizing Finnish literature and providing information about authors and literary works on the internet (Kirjallisuuspankki-Litteraturbanken). The conference met my hopes to get an overview of the field today, especially during the panel session on digitization of cultural heritage (possibly) as “modern utopia” the second day of the conference. This session gave a short introduction to the central questions to be discussed: should we focus on quantity or rather on quality in digitization, who should be financing this work, which are the needs of the users and how can the users learn to “translate” the aims of their research in ways that can be met by the possibilities of the digital age?
Overall, the conference was also a very good experience although I felt pretty exhausted after about the first two sessions on the first day, while I had been trying so hard to understand at least something about the technical aspects which were in focus. Even though I’m in this job because I love working with literature (rather than machines) and I see the technical solutions more as means to an end than something interesting in itself, I believe it is useful to try to learn a little bit more about the technical side. However, it is also frustrating when you try to understand but just can’t, simply because of the strange terminology.
Then, I figured one should perhaps begin by relating to the technical language the way one relates to abstract poetry or to any other strange language… Instead of trying too hard to understand, be open, be curious, be grateful for every word that at least has some kind of meaning to you, and perhaps gradually start seeing a pattern, recurrent words you realize you have to pay attention to and check up: “ground truth”, “text mining”, “crowdsourcing”… As Frank Zappa, quoted in the beginning of the conference put it: “A mind is like a parachute. It doesn’t work if it is not open.”
Many of the presentations were about the possibilities and challenges in digitization today – indeed from a rather technical point of view which is of course valuable in itself, however I was mostly interested in the point of view of the user. What can the researcher (and why not also the teacher, the editor, the reviewer, the reader…) actually do with the digitized material especially in the field of literature? What’s the use of these technical solutions? However, for me the conference was also a bit of an eye opener in this respect.
Of course digitized material can be useful in many ways, even to a literary scholar using “traditional” methods. A researcher can benefit simply from the fast and easy access to (sometimes rare) material and to information about authors, works etc. on the internet, the possibility to search for a word or a name in literary works… In addition however, digitization is challenging us to develop new methods, new ways of thinking about and analyzing literature.
I think literary scholars in general (myself included) often feel some hesitation towards these methods, partly because this approach is so far from what we have been used to doing with our material in the recent decades (deep reflection on the single works and texts, the heritage of close reading and new criticism still affecting us), and partly because the experience from quantitative studies on literary texts in the past (before the digital era) has not always been that inspiring. In my own more specific field of interest (rock lyrics) there have been studies showing that certain themes (love and relationships) are more common than others. But how interesting (or surprising) is that? What can we do with this knowledge? And aren’t we losing track of the complexity and the uniqueness of the literary work in such studies?
But maybe it’s time to reconsider and give it a chance? Exciting, as well as a bit frightening is the thought of using “TopicZoom” (English version) in literary research, a method for automated assignment of topics in OCR’d historical texts (presented by Klaus Schulz, Christoph Ringlstetter and Florian Fink). With the help of this method, researchers (although they admit it is not a 100 % reliable way to decide what a text is about) have been able to define the topics in material, even create a hierarchy between more and less important topics and link texts with similar topics to one another, in an effort to organize large collections of material. By the way, the German version of TopicZoom was said to work better than the English one.
Especially interesting from the point of view of literary studies was a poster presented by Dimitrios Kokkinakis about a project he and his colleagues from the University of Gothenburg have been working on, using digital methods to discover gender patterns (how characters of same or different genders interact with one other in dialogue) in Swedish fiction (information in English). At least this is very interesting as a supplement to our own reading and interpretation. As it suggests, with the help of digitized material we might be able to extend the study to a much larger material than you could possibly read yourself, to test and perhaps generalize your own discoveries with the help of computers.
During coffee breaks and enjoying tapas-style lunches and conference dinner I also ran into other literary scholars at the conference. It was a relief to see that there were others who have (for some time already) been thinking about the same critical questions I have relatively recently encountered myself (and even some who willingly admitted that they didn’t understand all the technical aspects!). Also, it is interesting to hear about other projects concerning literature, and get some ideas from people who understand where I’m coming from.
I was delighted to have a chance to talk for instance with Mats Malm who represents Litteraturbanken in Sweden, a project which has been an inspiration for the project I’m working for, as well as Kerry Kilner, the director for AustLit, a similar project concerning Australian literature (which has been running for more than 10 years already). Hopefully, I will get a chance to learn even more from the experiences in these projects in the future. Kerry Kilner mentioned Franco Moretti, founder of the Stanford Literary Lab and an influential name in digital literary studies. Moretti is also the father of the concept of “distant reading” (as opposed to “close reading”), a method which apparently can change the researchers conception of a whole genre or period in literary history. However this is interesting, I’m not willing to go as far as Moretti who has declared that we should stop reading books and that distant reading should replace (not support) deep, individual analysis in literary studies. Of course, I’m not the only one who feels this skepticism, see for instance.
That brings us to a crucial question; How good a reader is the computer? The computer can be a much faster “reader” than a human being, but of course it (still) lacks in sophistication, in the ability to see nuances and specificity. But it was stated during the conference that we must be open to the fact that we might not even be able to imagine the possibilities digitization will open up in the future. Still, one crucial difference between the human reader and the machine is the reading experience, that humans can enjoy reading, be emotionally or intellectually moved, inspired and challenged… Therefore, I don’t imagine machines can ever replace us, the real readers, for whom literature is written.