Digital Humanities is not Merely Data
In his recent Los Angeles Review of Books essay “Literature is Not Data: Against Digital Humanities,” Stephen Marche extends a dogmatic finger to stem the groundswell of digital humanities (DH). I expect that soon many HASTAC users will gainsay Marche’s argument that literature should not become data and that digitization leads to the “desacralization” of texts—and perhaps many English professors will nod in agreement with him. I can contribute to this conversation by showing how my activities in the University of Maine Digital Humanities Lab and my colleagues’ work serve as a counter-example to Marche’s commentary.
Aren't intellectuals weary of simplistic “against x” pieces? I came to work in DH because digital humanists encouraged my development in archival research and practical criticism––and did not denigrate my interests. Whereas Marche seems to be exclusionary, digital humanists have a long tradition of collaboration: every digital project with which I’ve been involved is multi-faceted, complicated, and in need of content and design experts with different skill sets. Marche repeatedly and vaguely claims that “literature becomes data” without demonstrating any knowledge of what most digital humanists do.
Which brings me to what I do. I use databases, visual aids, and annotation tools to supplement, not replace, the reading experience and scholarly research (all of which, yes, consist of a lot of data, but the final products are user friendly). I am interested in building databases and critical editions—as Stephen Ramsay said in “On Building”—that improve upon on old print models by making more information available in a visually appealing way. One of UMaine DH Lab’s affiliated projects, Melville’s Marginalia Online, shows precisely why Marche is misguided.
Suppose you are reading Melville’s Billy-Budd, Sailor, in a literature course, and you want to know what Melville was reading while he was composing the novella. Suppose further that you wonder what, if anything, Melville’s reading during Billy-Budd revealed about his religious attitudes in his later life. You can go to Melville’s Marginalia Online, search through the online catalog, and see a bibliographic record (which would be more complete than any existing print editions) of every book Melville was known to have owned, borrowed, and consulted during those years. You would find some interesting starting points for serious scholarship—for one, that Melville bought seven volumes by the pessimistic German philosopher Arthur Schopenhauer in the last year of his life. (He also read a lot of Balzac.)
Has literature become data? Not really. And, in any case, the argument is far from over. To expand on my last example, the Hayford-Sealts edition of Billy-Budd, one of my prized corporeal objects, is in my hands, crinkled and heavily marked, and Melville’s Marginalia is simultaneously open on a tab on my laptop screen.
My colleagues at the Melville Electronic Library (MEL) are similarly working to complement the reading experience. One of their major undertakings is a virtual whaling ship that will help readers make better sense of Melville’s physical surroundings in his whaling stories (you know the big one, but there are others with different kinds of whale ships). I have annotated two poems that are difficult to print with notes because they are so heavily indebted to newspaper and literary sources. Digitization permits me to make more annotations available to scholars and students, and, like the Melville's Marginalia database, it lacks all of the hindering symbology that excluded readers from print editions. Other scholars at MEL are working further with the TextLab markup software, which will provide “revision narratives” for people interested in Melville's revisions, and how his words have been re-interpreted in culture (this constitutes a "fluid text theory," which suggests, like Marche, that "literature is terminally incomplete"). Others are developing a timeline software that will make better sense of Melville's life, work, travels, and the historical context in which he found himself.
These examples prove that DH is still an emerging discipline (or sub-discipline) and that it is multi-faceted and not easily categorized. DH researchers might have different implementation goals, but we are all interested in building reliable tools for humanists to improve their research, which will make the human record more accessible. We are empiricists; it is our duty to present the evidence as plainly as possible. We emphasize archival and textual research and experiment with the digital tools that can enhance our understanding of that research. None of which entails the death of the book and the archive—in fact, it enhances both. We embrace each other’s work, because it improves knowledge and access. This is a good in itself.
Marche is right to point out some of the losses of DH, though his argument does so at the expense of the gains. His fears that we will cease to visit archives and museums, and lose the sacred sense of reading a codex, discards the baby with the bathwater. Big data is not equivalent to DH. Marche's spiritual arguments become self-defeatingly circular, but if the choice is between seeing a digital artifact or not seeing it at all because one cannot afford to travel to the archive’s repository, who would not choose the former? Digitization cannot give us all the answers. Oftentimes in my own research, I used a digital resource to guide my research in preparation for a visit to an archive. In the course of my doctoral research, I went to some trouble to see a collection of letters from Paul Bowles to a famous correspondent, only to find out that they were totally useless for the immediate concerns of my project (after all, a finding aid doesn’t tell you whether a manuscript will be useful). But digital research can refine the process of archival research––and I would venture to say that nine times out of ten, even if you’ve seen digital copies of manuscripts, you will still visit archives to proof your work.
Marche is also right to point out that many professors and writers are actually very conservative types; but what does conservatism necessarily have to do with DH? Is not his “sacred text” argument more conservative? He lacks valid conservative arguments, e.g. that data mining and digital projects are useless if the power goes out or the Internet is disabled.
Yet Marche seems to be more afraid of e-books than DH. For every seemingly useless DH project and tasteless data mining operation, there are many more peer-reviewed articles that only serve the interests of professionalized humanities professors and, sometimes, their graduate students. The digital humanists I know are concerned with a bigger audience, because the humanities have been in crisis for some time, and they can only be saved by giving them a larger audience, not a smaller one. Self-enclosed, jargon-ridden discourse (with words like “desecralization”) and elite access will stunt the growth of the discipline. Marche knows all of this, for he notes how humanists have spent decades complicating, rather than solving, problems (he also quotes Kingsley Amis’s wonderful bit of wisdom that academia only cares about “the pseudo-light it threw on non-problems”). Here we are in complete agreement about the problem, even if not about the solution.