It is a <Marxist> <culturalideology> truth <classformation> universally </classformation> acknowledged</culturalideology> </Marxist>, that a <SexGender> <gender> single man <capital> in possession of a good fortune </capital> </gender> must be <capital> in want of </capital> a <gender> wife </gender> </SexGender>.
All right, it’s a bit preposterous–but perhaps not.
TEI is the attempt to make possible rigorous use of computer technology in the humanities. The initiative stems from a recognition that, like eyes, microfilm, the internet, peer-review, and so forth, computation can?/could?/should? constitute more than a technology in the colloquial sense; they could be–and probably already are–a technology in a Foucauldian sense: a material and discursive formation that comprises a fundamental structure of “the academic self.” The only difference between a scholar who refuses to acknowledge the potential of computation for the humanities and one whose heart leaps at the sign of sideways carrot brackets is that the latter admits the reality that humanities scholars have grafted computers into the scholarly body. Any time I begin research with a cursory search of JSTOR or the library catalogue, I’m making use of XML-encoding (or an equivalent). Unless I know specifically where I want to start, someone, somewhere, encoded the data that leads me to begin my research with Ed Cohen and not D. H. Miller. TEI is an attempt to make sure that someone is someone who thinks like a literary scholar.
Given that argument, whether you buy it or not, I’m just going to ask you, Dear Reader, to accept for the next four-hundred words that quantitative computation of the TEI sort is a worthwhile endeavor. Through that vein, I would like to briefly consider the notion that if we can calculate or compute and interpretation, we should be skeptical of that interpretation. A professor once told me that reason that they [sic] rejected “Theory” was that if one knew what someone’s “Theory” was, one could just as easily tell what that person’s reading of any given text might be. Sure. Fine. Marxists will talk about class; Feminists will talk about women; Deleuzians will talk about assholes and werewolves and lines and bodies.
So, let’s do two things with encoding and theory, as an experiment. First, let’s imagine that it could be an exceptional teaching tool, a way to introduce undergraduates to the field. “Here, undergraduate,” you will say. “Search this text and mark up all the places where someone could do a reading with Theory X.” Or, even better, “Here, undergraduate. Make use of this edition of Middlemarch that a professor has marked up, so that you can see all the interpretive possibilities that a text can provide.”
And secondly, let’s imagine that a whole bunch of us did this, and then we ran all of that data through an even bigger algorithm. We could, I think, legitimately examine the way that we think of texts. Can we chart Mary Ann Evans’s interest in Spinoza through frequency of references? Can we chart the Victorian interest in brain science by compiling that kind of data? Perhaps even more interesting would be the questions we could ask about ourselves: How much (quantitative, word-count-wise) of a text do we overlook in order to make our specific reading on X in text Y? What seem to be the topics about which most scholars write in a given text? What seem to be the topics most scholars see in a given text but on which they don’t publish?
Certainly, all of this data would be quantification of the subjective. There are many reasons why I think class matters in Pride and Prejudice, none of them ontological–but all of them potentially interesting. Moreover, this wouldn’t provide a given reading of a passage; rather, this would constitute a markup in potentia, the possibility of reading. Imagine what we could find out if we stopped panicking about whether literature was data or had data or could be rendered in data and just went for an acknowledged amalgam of sense (numbers) and sensibility (interpretation)?
“Hey,” we might say, “it turns out that, for all we talk about it, The Moonstone really doesn’t quantitatively center around the detective story and Victorian disinterestedness. In fact, it seems to be much more concerned reading books as bodies and bodies as books!”
Well, all right–that’s what it would say if the data included pretty much just my reading. But that’s a different post.
I think this is an interesting idea but, in reality, not as useful as one might think. I feel like, while you say these markups constitute a markup “in potentia,” they also, in some ways, close down interpretations by layering a text with multiple possible “correct” readings. To code theories in this way and apply them to pertinent passages makes interpretation, as you say, an almost quantitative task. While you are offering your potential interpretations, it seems almost if you are applying a formulaic reading to these texts. For instance, in your marked up passage above, you have marked “wife” as open to interpretations of sex and/or sex/gender. While this is layered with considerations of other reading strategies, it also relegates each of those strategies to certain traits– want will concern capital, etc. To do this, I feel, limits possibilities when one (perhaps an undergrad) approaches a text for the first time. To encounter a text marked up as such would tell that undergrad what types of arguments are appropriate to make, robbing them (sic) of the ever-so-useful experience of approaching a text and stumbling through their (sic) own intellectual pathways, allowing them to form in useful directions. That being said, perhaps that undergrad is not the intended audience for a text encoded as such. Perhaps it would prove useful for a graduate student to approach a text like this, able to see which passages (and how many passages) open themselves up to certain readings. (Potential project??)
Why, of the posts available right now, is yours the only one, even if tentatively, enthusiastic about TEI? I would agree that these could be good teaching tools, especially if offered qualified as teaching tools. One might find it advantageous to embed a canonical reading, via semantic markers, into, say Frost’s “Home Burial.” Why this is preferable to looking at criticism in the form it currently takes (here’s U of I’s collection of crit on “Home Burial:” http://www.english.illinois.edu/maps/poets/a_f/frost/burial.htm ), I am unsure. Perhaps ease? The all-in-one-ness of the new text? I wonder if, by including criticism alongside, and literally coded *within* the body of the text, it would make the criticism more a part of the primary text (as perhaps Latour would argue happens anyway) and to possible disadvantages (see Staci’s post).
My question is, what is it, Peter, about your approach to literature in general that makes semantic markup less threatening than it seem to be for others of us? (I doubt “threatening” is the right word, but bear with me) I have the same giddy inclinations to exploit TEI, but they are buried beneath a sort of conservative Luddic (Luddite-esque?) inclination for sameness.
While I agree with Staci, I’m curious to know why this has turned toward discussion of possible pedagogical uses of these encoding schemes as opposed to what new research or interpretations they could allow.
It might be interesting to note that these applications are predominantly pedagogical and tend to avoid options for new research or interpretation. Unless I already posted this and I missed a message about waiting for it to show up.
Because this is attached to my teaching portfolio, I moderate the comments to make sure I don’t get spam on here. Sorry.
I think, Joe, (and by extension, Jesse and Staci), that I find it exciting because I tend to think of theory as occurring on two different levels: the level of formulaic but important interpretation (pedagogy), and imaginative, self-reflexive metatheoretical innovation.
That is to say, when I teach a theory, while I want my students to understand the whys of the theory, it’s often more important that they first learn the hows–which are formulaic. Applying Sedgwick to Middlemarch is going to turn out slight variations, but it’s mostly going to be about the homosocial relationship between Lydgate and Bulstrode, or another couple of men. A good undergraduate reading probably isn’t going to challenge the way we interpret Sedgwick or think of the homosocial (though more power to it if it does); it’s going to do a bang-up job of identifying how Sedgwick works with the text. TEI would be enormously helpful on this front in its capacity to demonstrate or allow students to demonstrate where those readings might take place. Staci, I think then that I agree with you: TEI would close down some potential readings, but I think it would enable more that it would close down until such time that a given reader begins to move beyond the basics of theory.
The way I tend to do my research, on the other hand, reflects more on the theories and my/other theorists’ deployment of those theories. Here, as I outlined above, TEI could help show us how we use theories, which theories we use, etc. I also find that exciting.
So, Jesse, I think it’s turned that way because my throwaway move toward interpretation–tracing Spinoza in Eliot–is less satisfactory (less ludic (not luddic), less virtual potential?) that the pedagogical elements.
After talking about this more in person (sorry other commenters, not digitally followable), I think it’s interesting to explore what different KINDS of pedagogy this type of project could open up (how this might be less valuable to first semester freshmen learning to read actively, but how it could be more valuable to juniors/seniors who know how to read actively and are ready to observe how a text can open itself up to multiple interpretations). I think understanding and analyzing those differences lends itself to interesting research, research that encompasses and advances pedagogical pursuits. Perhaps researching the different ways TEI can help with pedagogy at multiple levels will produce interesting answers in relation to the efficacy of the digital encoding of texts.
While my enthusiasm for TEI at this point is certainly mixed, I really like what you’ve done here to redirect the object of study back onto the critics themselves. Certainly there are pedagogical implications, but the theoretical and research implications are present as well. If we think of marked-up texts as necessarily authorial productions – in that each mark up is the direct product of a single interpreter – then it might be possible to trace theoretical bents to actual praxis in a single critic. This makes me wonder how we as a community of critics might use TEI not as a technological integration but as a self-reflexive tool which uncovers our own ideological predispositions. Would it even be worth it to try this out?