I recently returned from the DH 2014 conference in Lausanne. I went to give my brief (my brief brief) for getting serious about social science methods in digital humanities under the title “Let DH Be Sociological!” The conference offered plenty to think about on this theme. Also I got accused of wanting to “dumb down” an entire field of study…so that’s got me thinking too.
I won’t rehash the whole of my short talk in this post—in fact, a longer [sic] version is to be found online in the abstract, with the few significant changes I made visible in the final set of slides. Basically, I think we should situate quantitative methods in DH (which are currently going under names like “digital methods,” “distant reading,” and “macroanalysis”) in the context of one of the large-scale transformations of literary study since 1970 or so, its steadily growing and now dominant concern with the relation between the cultural, the social, and the political (let’s call it the cultural turn for short, though I don’t mean to identify the transformation of literary scholarship with the roughly contemporary historiographical shift of the same name). This turn is common knowledge, but it’s kind of fun to count it out, as I tried to do in the talk.
One of the major challenges of the cultural turn has been the dubious relation between the handful of aesthetically exceptional texts literary scholars have focused their energies on and the large-scale social-historical transformations which have come to be the most important interpretive contexts for those texts. Do these texts tell us, as clues or symptoms, everything we can learn about the systematic relations between society and literature, or, for that matter, about the systematic development of literature considered just as a body of texts?
What disciplines are concerned with systematically analyzing, using quantitative techniques, the relationships between patterns of culture and patterns in society? The social sciences: sociology, anthropology, political science… Not coincidentally, these are also disciplines in which the relation between quantitative and qualitative methods has been the subject of rich, lengthy, century-long debates. My own particular example was the way digital humanists making use of topic modeling are finding themselves facing the classic challenges of content analysis. How to reduce the complexity of a large body of texts in a reliable and informative way, how to scale up interpretation, how to validate a content scheme…The social-scientific methodological debates are now our debates. And quantitative and mixed-methods digital literary studies are in a position to radically expand and enrich the scope of the sociology of literature.
I saw some wonderful presentations at DH that were doing just that. Elizabeth Dillon and Julia Flanders presented a fascinating talk on planning text markup in light of research questions for the Early Caribbean Digital Archive. With the right markup, an archive can help to answer questions about cultural systems that link race, economy, and genre, or that allow us to map imagined and physical geographic spaces, or that let us trace changes in representations over time. And all of this not in a few hand-picked examples but across the archive. A team from the Stanford Lit Lab gave a fantastic presentation on reconstituting the eighteenth century British literary field from generic labels on title pages. Tim Tangherlini and David Mimno demonstrated how an algorithmic classifier of a collection of Danish folktales could yield remarkable anthropological conclusions about both story themes and story-tellers (not to mention the story-collector, too). Graham Sack showed some results from an ambitious attempt to apply agent-based modeling to the literary marketplace. Natalie Houston showed some techniques for probing the field of Victorian poetry, which promise to allow us to see large-scale trends in versification choices. Also from the annals of field-mapping was Carolina Ferrer’s short paper on applying bibliometrics to the MLA bibliography to understand canon-formation in the Americas (a topic which is naturally close to my heart). I missed the live version of Lauren Klein and Jacob Eisenberg’s discussion of how to analyze the evolution of topics in a 19th-century U.S. newspaper archive, but fortunately they have placed their talk online.
Such work breaks down the boundary between social science and “the humanities” (particularly the literary humanities; as someone at my panel pointed out, this boundary is less marked for historians). And well it should: this is just how we will answer some of our most interesting and urgent questions about the historical and contemporary role of culture. But we might also ask: what does that boundary protect?
It would be more accurate (and would help to explain the animus, I think) to say that such methods threaten to profane the digital humanities.
The digital as such is perfectly compatible with the sacredness of the humanistic text. The prominence of digital hypereditions of canonical authors (say: the Whitman Archive) makes the point. So, however, does the ambivalence with which many “distant readers” discuss their methods: there appears to be a strong temptation to frame such methods as a mere supplement to “close reading,” which remains the ultimate goal. Let us take a symptomatic example. In his DH 2014 keynote lecture, Bruno Latour repeatedly invoked the idea that the real value of the digital for the humanities would be to create a better close reading (a term he used repeatedly). The elaborate technical and institutional apparatus he created for his own Inquiry into Modes of Existence allows a remarkable “distant” view of the many readers and interpretations of his work, a few glimpses of which he gave. But he insisted repeatedly that the ultimate goal was to ensure a “close,” that is faithful, interpretation of his argument. Indeed, I could hardly ask for a better image of the sacerdotal than the video clips Latour showed us of the seminars on his book, in which the serious and credentialed readers voiced their views in the presence of the author.
This is not the only methodological route the digital affords us. Though it might preserve the sacred text, to be touched only by the sacred method (close reading) exercised by those who have been purified (the humanist, individuum ineffabile), it also might help us into an expanding universe of dirty methods, with their bags of words, their noisy classifiers, their obviously reductive models, their coding schemes (markup) planned in advance for counting. It might displace questions of value with questions about the systems that produce value. Instead of the priestly “reading” it might shift our attention to the study of readers and their readings. And so on… The more this transformation happens, the more humanistic research draws closer to social science in method as well as in object.
And here is the reason for going around saying “let DH be sociological, exclamation point”: these other methods can and do produce knowledge about subjects that matter to the community of scholars. Digital or not, quantitative or not—these oppositions are actually secondary. The boundary between sacred humanities and profane social sciences is a barrier to the production of knowledge. Let’s have profanity all around.