Following the New Generations workshop in Glasgow University’s Medical Humanities Research Centre (MHRC) on December 11th 2014, New Generations Programme member Hieke Huistra writes:
The second day of our Glasgow workshop centred on digital humanities. The University of Glasgow has many projects in this field — for an overview, see the website of Glasgow’s Digital Humanities Network — and the workshop introduced us to some of them. Most of the projects presented focused on one digital humanities practice in particular: digitising historical sources. Laura Stevens spoke about the ongoing digitisation of mental health records, done in cooperation with the Wellcome Library. David Shuttleton and Mark Herraghty showed us the Cullen Project, which creates a publicly accessible digital edition of the consultation letters of eighteenth-century physician William Cullen. Sonny Maley explained how the university’s syphilis collection (approximately 250 books) is opened up to researchers through careful cataloguing and web-based resources.
All speakers stressed that such projects have their own practical problems, often related to the required cooperation between different disciplines (humanities and computer sciences) — a practicality the medical humanities are not unfamiliar with, as we discovered in our previous workshop. But such issues can be overcome (as with many things in life, it helps to talk about it), and then the benefits are clear: suddenly, you can study your sources at home or in your office, which not only saves travel costs and time, but also enables side-by-side comparison of manuscripts from different corners of the globe.
Improving access to sources is one of the simplest examples of how computers change humanities research. But the field of digital humanities encompasses more — in particular, computers can be used to analyse sources, and we saw glimpses of that during the workshop as well. Brian Aitken, Glasgow’s digital humanities research officer, presented the Mapping Metaphor project, which analyses the Historical Thesaurus of English to better understand the development of metaphorical language use. Its visualisations show, for example, how the word ‘healthy’ was used outside the medical domain (e.g., ‘a healthy economy’) and when such metaphorical connections arose. Linguist Marc Alexander argued that a large corpus of digitised texts should be approached as ‘big data’, and demonstrated tools enabling this, in particular Glasgow’s SAMUELS (Semantic Annotation and Mark-Up for Enhancing Lexical Searches). SAMUELS helps researchers to search not simply for words, but for meanings, thus offering a way around the problem that over 60% of English word forms have multiple meanings. It delivers a focused corpus, which can then be analysed further with tools like the IMS Open Corpus Workbench.
All of this is intriguing, but not without methodological problems. Some were addressed in the workshop discussions; many more can be found in digital humanities literature. In his 2013 article ‘Confronting the Digital’, historian Tim Hitchcock warns us that consulting sources on the screen instead of on paper can easily result in a loss of context and a false sense of completeness: we forget about the non-digitized sources or about the parts of the digital sources invisible on the screen, like blank pages that have been skipped in the scanning process. Johanna Drucker, another critical digital humanist, has shown that many data visualisations suggest an objectivity the humanities cannot, and should not, offer. Furthermore, many scholars argue that (statistical) patterns in large amount of texts, the main outcome of digital analysis, are irrelevant to humanities scholars because they lack meaningful interpretation.
How should we evaluate this criticism? What do the new techniques offer, and where do they fail? These questions remain open – it takes more than a one-day workshop to answer them. In particular, it takes hands-on engagement with digital tools: only actually using them can give you a proper grasp of what they can and cannot do. Unfortunately, our Glasgow workshop was more theoretical than practical, but fortunately, there is, as always, the web, which offers plenty of tutorials on working with digital tools. There are general introductions, overviews of tools, and more specific instructions on for example text mining, topic modelling and programming for historians. If you’ve got an afternoon to spare, I recommend taking a look at one or two of these resources. They will help you form an opinion on digital humanities – and they might further your research.