January 14, 2014 – 5:04 pm
Before I give you the listicle, I’m going to make you skim through some summary. I’m mean like that. (Back off, man. I’m an academic.)
I go to the Modern Language Association annual convention these days for the digital panels — and to hobnob with the smart people on them, of course, many of whom I know already, but many of whom I don’t know, even “just” online. If I had to name just one thing I got from MLA this year, it’s that digital humanities is no longer the next big thing — it’s beginning to be just an ordinary thing. In other words, I felt that there was a lot less defensiveness about digital methods in the study of literature this year.
Granted, most of the papers dealing with digital methods are still located on panels themed around digital methods, as I think you can tell by Mark Sample’s helpful annual listing, but there was some promising intermingling between digital and traditional methods papers on panels such as “Diversifying the Victorian Verse Archives.” (Note: I was on my way to that panel when a minor emergency came up that took an hour or so to resolve; I was very sorry indeed to have to miss it.) All three of those papers basically teach us a bit more about Victorian songs and their relation to Victorian poetry than we used to know, but only two of them explicitly mention the creation or sophisticated use of digital archives as a major component of the research. And Brian Croxall and I both had the same idea in forming our panels: to concentrate on the results rather than the methods. Both his Association for Computers and the Humanities panel “Beyond the Digital: Pattern Recognition and Interpretation” and the panel I put together titled “Things My Computer Taught Me About Poems” tried (with fair success) to do less description of and argument for digital approaches while giving more concrete examples of the new insights into language and literature these approaches have given.
But I have to admit that the single best paper I heard that took what we might call this “more interpretation, less demonstration” approach was Mark Algee-Hewitt’s, of the Stanford Literary Lab. In a panel titled “Making Sense of Big Data,” Algee-Hewitt seemed all insight, though granted what he had insight into was a particular well-known literary theory of Mikhail Bakhtin’s rather than a literary text or corpus. In The Dialogic Imagination Bakhtin famously argues that novels get their energy from “heteroglossia,” or what we might call “polyvocality.” Novels have so many voices, so many registers of diction: all the characters and their dialogue, the narrator in various moods and modes. Lyric poems, by supposed contrast, generally have one voice: that of the poet. Algee-Hewitt recounted an absorbing tale of getting results from his analysis of the comprehensive Eighteenth-Century Collections Online database (ECCO) that first supported and then challenged Bakhtin’s theory: ultimately, at least if you consider these texts at the semantic level of individual words (what individual words mean, that is), Bakhtin seems to be wrong — poems are actually far more polyvocal, less “self-similar” in Algee-Hewitt’s term, than novels and non-fiction. Theorists versus data analysts! I love it. Can’t wait to hear more. I’m sure the debate over that particular issue of the heteroglossia of novels and poetry, if it branches out, will indeed circle back to method — one questioner raised the issue of whether semantic word analysis really matches Bakhtin’s idea of “heteroglossia” — but at least it won’t be an argument about whether such data analysis is legitimate: only how to do it as well as possible. As a side note from that paper, I was impressed by the very fact that the Stanford Literary Lab has developed a simple and effective algorithm to tell novels apart from nonfiction apart from poetry and poetic drama from the full-text data: I think Algee-Hewitt said it had about a 95% accuracy rate. I can see that being very useful for someone else’s project. Clearly I need to keep a closer eye on that Stanford Literary Lab.
Here’s the listicle for you, then — The 7 Best Links to Digital Poetry Projects from MLA:
- Princeton Prosody Archive, Meredith Martin et al. (database not live yet)
- Steepletop Library – The Books of Edna St. Vincent Millay, Amanda French (me!)
- Poem Viewer, Julie Lein and Katharine Coles et al.
- Songs of the Victorians, Joanna Swafford
- Stanford Literary Lab Projects, Mark Algee-Hewitt et al.
- eMOP: The Early Modern OCR Project, Laura Mandell et al.
- I ♥ E-Poetry, Short-Form Scholarship on Born-Digital Poetry and Poetics, Léonardo Flores et al.
There’s a lot more that happened at MLA that I didn’t see that I could have posted, I’m sure, and there’s a lot more that I saw and liked that I can’t easily link to, but that’ll do for now. (I’ve always wanted to dip a toe in the listicle biz.) And if you’re interested, here are a couple of Storify stories from the two panels I presented on — they often have helpful summaries and reactions and commentary from the twittering audience: