The project “Shakespeare His Contemporaries” will make a systematic effort to harness the energy and imagination of undergraduates as editors and explorers of old plays in new forms. It will begin on Monday June 24 2013, when Hannah Bredar, Madeline Burg, Melina Yeh, Nayoon Ahn, and Nicole Sheriko start an eight-week curation marathon during which they will collaboratively fix many of the most common errors of transcription in some 600 non-Shakespearean plays from the EEBO-TCP project. They are Northwestern undergraduates whose work is generously supported by summer research grants from the Weinberg College of Arts and Sciences. Their work will contribute to a much expanded and improved version of the Early Modern Drama Wordhoard site, and I hope that much of it will also find its ways into the TCP source texts. They will learn that in any philological endeavour even the simplest acts very quickly raise fundamental questions of evidence and certainty, that both God and the devil are in the details, and that at the foundation of an edited text you are more likely to find shifting sands than bed rock (adapted from Stephen Greenblatt’s “the palace of the normal is built on the shifting sands of the aberrant”).
Nicole Sheriko, who will be a senior next year, will also create a prototype of a Young Scholar edition of Fair Em, the Miller’s Daughter of Manchester, an anonymous play from around 1590, whose bitter-sweet reflections on desire call to mind Yeomen of the Guard (or the other way round). A Young Scholar edition takes a TCP text as its point of departure, turns it into a proper diplomatic edition, and adds metadata and commentary that enhance the query potential of a digital text in the context of a larger corpus. More about the details of this in another blog. I hope there will be many such editions.
Some readers of this blog will recognize the title of the project as a portmanteau combining Ian Kott’s Shakespeare Our Contemporary with Early Modern titles like Catiline His Conspiracy or Purchas His Pilgrimage. The title has an edge to it: Shakespeare’s contemporaries deserve more attention than they are getting, whether we think of them in Karl Marx’s phrase as witnesses to the “wishes and struggles of the age”, or as guides to a more nuanced understanding of the ways in which Shakespeare was and was not “of his age,” or finally as interesting objects in their own right.
The MLA bibliography for 2012 contains ~64,000 entries compared with some 15,000 entries in 1962. We also live in an age that has been suspicious of the highly canonical, celebrates the margins, and likes to decenter or open up things. You would think, then, that both in absolute and relative terms, non-Shakespearean plays receive a lot more attention than they have in the past. But this is not the case. Instead we have more scholars chasing the same three dozen Shakespearean and at best two dozen non-Shakespearean plays. Consider Titus Andronicus and Lust’s Dominion, both of them plays about a lascivious queen and an ambitious Moor. There are 2915 references to Shakespeare’s play in Jstor, compared with 133 for Lust’s Dominion. And Lust’s Dominion does relatively well because it is about a hot topic. If you compile the bibliography for a minor Elizabethan play, you will often find that it is very short and that the more substantive treatments are quite old. Fair Em has received only the most cursory attention in the past two decades. The highly canonical is alive and well or, to put it a little more cynically, scholars like to attach themselves to celebrities, celebrities are ‘positional goods’, and you can’t have more than a few dozen at a time. You can see why professional scholars stay away from minor plays, unless they explicitly deal with hot topics. A play may interest them, but how will an entry about it look on a c.v.? This is not a problem for bright and ambitious undergraduates: anything done with intelligence and imagination will look good on their resumé. And while from a bibliographical perspective Early Modern plays have been meticulously catalogued their textual world is full of regions where young explorers may find much to discover.
If Early Modern plays continue to be read at all, the young reader’s first encounter with them will increasingly be through some mobile device. That is an easy prediction to make: in 2015 EEBO-TCP texts will begin to move into the public domain, 25,000 in that year alone and another 45,000 over the following five years. It is not much of an exaggeration to say that by 2020 every book from the English-speaking world published before 1700 will be available as an epub. This digital migration of 227 years of Early Modern print culture changes the documentary infrastructure of text-centric disciplines. Beyond bringing more books to more readers digital texts can be enriched with levels of metadata that support new forms of corpus-wide exploration. You need to do a lot “to” texts before you can do these new things “with” them, but once you have done them, there are oceans of opportunity.
In the wider context of this digital migration “Shakespeare His Contemporaries” is a project that uses the “digital turn” to take advantage of features of Early Modern drama that make it particularly suitable for corpus-wide analysis. Highly conventionalized metadata are built into the genre with its hierarchically ordered system of explicitly marked speeches, entrances, exits, scenes, and acts. The length of most plays stays within the bounds of what Shakespeare called “the two hour traffic of our stage.” A digitally structured corpus will in most cases follow explicit textual directions and will produce segments that are comparable in size and type. Early Modern drama is certainly heterogeneous, but differences exist within well-defined bounds. The aggregation of plays into a corpus is therefore a fruitful and unproblematical step. You can read the genre as if it were a single play or “fragments of a grand confession” as Goethe called his own work in Dichtung und Wahrheit. Much Shakespearean criticism has proceeded on such a model. Why not extend it to the genre as a whole?
In a corpus of digitally encoded plays, where every speech is tagged as an event associated with a particular character, you can create a cast list of cast lists. And if every character is explicitly classified in terms of sex, age, and social status you have something like census data for the lives and communicative acts of several thousand dramatic characters over a century and a half–a digital turn to the Ciceronian definition of comedy as imitatio vitae, speculum consuetudinis, imago veritatis. Who speaks and at what length to whom? Such census data tell us much about what writers and audiences like or felt compelled to imagine. Cicero and Marx: drama as the “image of the struggles and wishes of the age.” It remains to be seen whether the application of social network analysis will deeply alter our view of the genre and its change over time. But at the least it will add nuance and firmness to our understanding of it.
The “digital turn” is also a “linguistic turn”: digital corpora, enriched with linguistic metadata, make it easy to look at microscopic details of verbal texture across millions of words. The corpus becomes a network of intertextual relations in which changes or differences in the distribution of words or repeated phrases let you contextualize particular textual objects by author, topic, style, genre, or period. The papers of the Literary Lab at Stanford have demonstrated the power of such inquiries. When Charles Lamb described James Shirley as “the last of a great race, all of whom spoke nearly the same language and had a set of moral feelings and notions in common” he pointed to qualities of the genre that digitally assisted analysis can modify or corroborate in illuminating ways with new kinds of evidence.
Is this not what scholars of an earlier age and with decades of reading experience have always done? Of course it is, and in the understanding of such scholars there may well be nuances and intimacies that will forever elude the machine. But their memory was not perfect , and they relied on dictionaries, concordances, and other print-based “metadata” that were incomplete, biased, and slow. Bright undergraduates lack the experience of those scholars, but the metadata built into a digital corpus give them resources that are more complete, more flexible, and much faster to manipulate. The drastic reduction in the time cost of look-ups gives them a competitive advantage over scholars of earlier generations. If they shuttle systematically between “close” and “distant” reading it may take them less time to acquire what Hilda Hulme, a scholar of an earlier generation, praised as the “disciplined imagination” that combines “the sober ant-like industry of the professional scholar” with the “grasshopper swiftness of the crossword puzzle addict (as well as the relaxed alertness of the confidence trickster)” (Explorations in Shakespare’s Language, 1962, p. 40).
So much for the wider prospects of what you can “do with” a digital corpus once the right things have been “done to” it in the form of systematically applied digital metadata. The “doing to” combines algorithmic procedures with human review and intervention. The first, quite humble, and often tricky task is to clean up the texts and correct incompletely or incorrectly transcribed words. It is a useful thing to do in a double sense: useful to readers of the cleaned-up texts, and useful to the students who as potential future editors learn much about the fog of uncertainty near the textual ground. More about this in the next blog.