On 13-14 June the Radboud University Nijmegen and the eHumanities group from KNAW co-organized the eHumanities workshop Soeterbeeck. This very nice mid-sized workshop with under 50 participants staying in the beautiful monastery Soeterbeeck in Ravenstein included very interesting invited-talks as well as a nice postersession where we presented the PoliMedia project. However, the workshop also demonstrated that so far the Digital Humanities is more Digital than Humanities.
Talks; Tools, Tools and Scholars
The opening talk by Sally Wyatt (MU/eHumanities group) gave a very interesting overview of the role of eHumanities (or Digital Humanities) in the scientific community and method, as well as the consequences of performing science increasingly with digital tools. She very nicely demonstrated that digital research is not a separate methodology, but an enhancement of the existing scientific methodologies in practice. However, as tools shape the types of knowledge than can be produced, we might want to ask whether it’s desirable that scholars use tools they don’t completely understand or explain. This might be a consequence of the current status quo of the Digital Humanities though: she stated that on the long term not only will the Humanities benefit from, but will also bring significant contributions to, the Computer and Information Sciences.
However, the rest of the talks mainly focused on what the Computer and Information Sciences have to offer the Humanities. José de Kruif (UU) demonstrated a tool for text mining historical texts, after which relations between people can be visualized. The advantage of this tool was that an overview could be gained of a large corpus of historical texts, showing relations that otherwise would not have been found. Similar themes were in presentations by Karina van Dalen-Oskam (UvA/Huygens) who presented a tool to visualize author-styles and find differences between authors, and Joris van Zundert (Huygens) who presented several tools to perform text-analysis through visualizations. An interesting comment by Van Zundert was that the humanities do not have big data (as compared to CERN), but complex data; data from the humanities can be interpreted in a multitude of ways, dependent of the research questions. This makes Humanities data difficult (and interesting!) for the Computer Sciences, because for each tool a decision needs to be made as to how interpret the data. Marieke van Erp (VU) presented the fascinating Newsreader project in which a semantic web is created of events described in texts. Several of these tools are interesting in the discussion of the desirability of scholars using tools they don’t fully understand, as the algorithms underlying visualisations are often difficult to understand and predict. Christian Gudehus (University of Flensburg) demonstrated a tool for manually coding texts such as transcripts and interviews; interestingly, he mentioned several scholars deemed this tool to technological, something that really surprised me as the entire tool was an environment for manual annotations of text. Lambert Schomaker (RUG) demonstrated a tool Monk with which handwritten texts can be digitized. As OCR simply does not work for handwritten texts, the Monk system performs recognition of words rather than characters (OWR?). This presentation diverted from the other ones as there was more of an interplay between the computer scientists and scholars: this tool requires people to train the system (crowdsourcing) by providing the initial descriptions of words, after which the system can learn to recognize this word.
eHumanities A Technological Push?
The dilemma of the Digital Humanities is that at this point the exploitation of the potential of digital tools appears to remain a technological push, rather than a scholarly demand. The lack of participants from the Humanities faculty of Nijmegen (the organizers!) was striking; having Humanities scholars incorporate digital practices in their existing research , so that the Digital Humanities are indeed an enhancement of the Humanities rather than a separate methodology, remains a future vision.
Nice summary and interesting remarks. I today encountered other questions: e- or internet publishing could change our research questions but how can we publish Open Access and get scientifical credits as well?
Thanks for the reply Gera. There are several Open Access journals, see a list here: http://www.doaj.org/, one of the more known OA publisshers is PLoS, where articles are published Open Access online, but you probably already know those journals 🙂
Hi Max, nice report on the workshop!
I am not sure though whether I agree with your ‘conclusion’. I can’t imagine the researchers who presented, working with the tools for technology only. Surely they use the tools with a specific question from their discipline in mind?
Furthermore in your text you mention implicitly: digital tools give the opportunity to ask new questions. In my view asking new questions is precisely the enhancement of the Humanities you want! And perhaps asking new questions is a scholarly demand also 🙂
Thanks for the reply Marnix!
Several researchers who presented indeed used these tools with specific questions in mind, for example Jose de Kruif and Karina van Dalen-Oskam. But from their presentations,I thought that 1) they are enthousiasts, and represent a minority in their fields, and 2) their presentations were still about the tools and what these tools could do, and not so much driven from the research questions. As Jose de Kruif also mentioned, actually then using the information for Humanities publications is really difficult, as the added value is still under-appreciated.