Loosely apropos Ed Finn’s panel at DH on Pynchon, Matt Jockers and I were trying to guess the most-published-upon author in English. I figured Shakespeare, he suggested Joyce. This morning I ran a couple of quick queries on the MLA database and came up with the following:
Shakespeare Joyce 2008+ 716 151 2004+ 3826 937 1999+ 8159 2135 All (1923+) 35489 9315
There are some details to explain, but the take-away point is that Shakespeare seems to be the object of about four times more scholarship than Joyce.
The details: These are raw result counts for the subject queries “Shakespeare William” and “Joyce James,” both of which are defined subject headings in MLA. The counts are total matching items of all types (journal articles, refereed journal articles, books, chapters, and other) published from the listed year to the present. I didn’t make any attempt to distinguish major from minor works (e.g., books from articles), nor single-subject studies from multi-subject ones. This is obviously pretty non-rigorous, but it was good enough to satisfy my passing curiosity.
This is interesting and at least a little unexpected to me. I figured Shakespeare would be in the lead, especially over the full history of criticism, but I thought things would be much closer, especially in recent years. I wonder if part of the gap might be explained by a higher likelihood of talking about Shakespeare in any given English renaissance context than about Joyce in any given modernist one?
Work Product 
As a quantitatively minded humanist, I can’t help but wonder if we really ought to do a little norming based on the significant head start that the Bard got in terms of building up his fan base? I’ve posted an alternative analysis over on my blog at https://www.stanford.edu/~mjockers/cgi-bin/drupal/node/29
Ah, fair enough, but I want my beer, so … since we’re talking about academic critical interest, maybe we should normalize the counts not over the full time since each author’s death, but only over the period during which there has been such a thing as academic-industrial literary criticism. Say 1880, roughly the period when the German research university was imported to the States?
If we do that, Shakespeare has 128 years to build a fan base, Joyce 67; that about halves the Bard’s lead, but it’s still a 2-to-1 advantage (and growing, apparently). Beer is mine — at least until Matt J. comes up with the next round of number crunching. :)
I’m not surprised. I recently taught a class on early modern English drama and tried to get the students interested — with mixed results — in stuff written by the other guys (not many girls in that group). Then I did searches in JSTOR on plays like ‘The tragedy of Hoffman’ or “Lust’s dominion”. I was surprised to find that in each case (and some others) there were a few substantial treatments of such plays half a century ago but very little since then. From which I draw the tentative conclusion that there is an inverse relationship between the rhetoric of canon busting or diversity and the dimensions of the canon(s) that are actually studied.
It would be quite interesting to compile a frequency-based list of works that have been referred to more than twice in JSTOR articles over the past century. On would need to look at the results with a keen awareness of the fact that the number of writers has increased greatly. And one might want to develop some concept like ‘range of reference’ that can be applied either to individuals, scholarly communities, or generations. In such an inquiry one would certainly discover that the focus of interest has changed over time. It is much less certain whether the range of reference has expanded.
Yes, I’d say that so long as close reading remains our dominant working method, the canon can’t go away, nor can its total size increase much—we’re just as finite as we’ve always been. So as we add new stuff on the contemporary end and discover new works worth considering from earlier periods, something has to go. No surprise, perhaps, that among that something is early modern drama outside Shakespeare. In a way this is an instance of the same trend that’s long been driving the humanities away from the center of the university curriculum; even being generous and saying that no one wants to stop studying literature, history, etc., the hours and years and dollars devoted to chemistry, say, (once a marginal pursuit that made up no necessary part of an educated person’s formation) have to come from somewhere in the existing devotion-budget.
In other matters, I love the idea of tracking trends in scholarship over the last century—you’ve given me a great project. I wonder how much of this could be done simply with JSTOR’s existing cataloging (re: subject headings, etc.) and how much might require data mining along the lines of named entity extraction (usually pretty good on names and titles, I believe). I should have a look at the details of their Data for Research program.
Incidentally, Martin, did you submit a proposal for Digging into Data?
Hi Matt (and Matt), I’ve been RSS-remiss and just stumbled across this debate. As someone pointed out in response to my own MLA-mining, there are tons of possibilities here for more complex analysis. This probably matters more for contemporary authors, but just breaking things down by work as well as author could be quite interesting (i.e. do particular Shakespeare plays go in and out of fashion?).
I’m not deeply familiar with the limits of the MLA bibliography but there must be ragged edges–sources that were added later than others, journals that arrived and disappeared, etc. One important factor here must the number of Shakespeare-only or tightly subject-specific journals.
I’m wondering if there’s a way to get access to the MLA bibliography on a CD or in some kind of complete database form…
Good to hear from you, Ed. I don’t know about getting access to the MLA bib, but I’d bet that your library might (know, that is). Hard to see why you couldn’t get it, though nothing amazes these days when it comes to locked-down data access.