Does It Work? Where Theory and Technology Collude

1Leave a comment on block 1 3

1. Laura Mandell
Director, Initiative for Digital Humanities, Media, and Culture
Texas A&M University

2Leave a comment on block 2 0

2. –ideas included in a forthcoming book, Breaking the Book, to be published in Blackwell’s Manifesto Series.

3Leave a comment on block 3 0

3. It’s right I think in an age when the modern university system is threatened from what Iain Pears calls “management takeover” to worry about possible modes of collusion underlying the new instantiation of the two cultures war, digital technology versus the Humanities. That’s what I’m going to do here today in order to discover where the oxymoronically-titled digital humanities and high theory currently collude in the negative sense and meet in the positive: at stake is knowing what we know, most of here being both technologist and humanist, empiricist and theorist, tool builder and critical reader. The digital humanities are currently in a fishbowl, as Lisa Spiro suggests, or panopticon, as Melissa Terras argues: the world is watching, and what we do matters for the future of the humanities as well as the university (Liu, “Report”). Today I’m interested in thinking about not what the Humanities and Digital Humanities are, but in how they are, and more specifically, where they break.

4Leave a comment on block 4 0

4. One of the most important documents in what has come to be called Ordinary Language Philosophy is J. L. Austin’s essay, “A Plea for Excuses.” The essay makes a plea for by showing us how important it is to examine the excuses people make. He offers a parable of the two donkeys. Someone, an “I” in the essay, presumably Austin himself, sees two donkeys in a field and shoots one of them. The donkey that falls turns out to be your donkey. In one case, he thought he was shooting his own donkey but shot yours instead. In the other case, he was aiming for his own donkey but missed and shot yours instead. He goes to your door and says, “Hey old chap, dreadfully sorry, but I just shot your donkey by . . . .” And then Austin gives us two choices: by mistake or by accident? Austin’s point is that, as speakers of British or North American English, we know exactly which word to choose in each instance.

5Leave a comment on block 5 0

5. Let’s imagine a scene of instruction instead: a mother sits with her 6-year-old boy. He says, “I ate that cookie by accident.” She says, “No honey, if you had stepped on the cookie, that would have been an accident, but you picked it up and ate it, deliberately.” “But Mommy, it was a mistake.” “No,” she replies, “you knew that cookie was your sister’s because you already ate yours—it was not a mistake.” We must wonder why Austin is out shooting donkeys, but leaving that aside, the parable is grounded in a very modern insight—Freud’s on a good day—that it is when things break, when something goes wrong and people are making excuses for it, that’s when we can best see how they work. Just from the fallen donkeys and eaten cookies, we know what models of agency, responsibility, subjectivity, and ethics reside in our language itself. Just because you speak the language doesn’t mean you have to subscribe to those beliefs, but, especially if you don’t, you need to know what these models are. “It was a mistake.” “What do you mean it was a mistake? I saw you aim and pull the trigger!” “Well, you see, I was trying to shoot my own donkey—I mistook your donkey for mine.” OR “Well, you see, I was trying to shoot my own donkey when yours walked in front of mine—it was an accident.”

6Leave a comment on block 6 1

6. The shared agreements harbored in ordinary language have been called common sense—to Edmund Burke, they proffer up the wisdom of all past generations—and ideology—we live in the “prison house of language” to quote both Fredric Jameson and William Wordsworth a la fois. Ludwig Wittgenstein’s point however is that they are neither wisdom nor mystification but fact: they are the system of agreements shared by people who share a form of life because they share a language (or, given that he was writing in German, share language games). Our shared assumptions are revealed, Austin maintains, by excuses, when something goes wrong. Breakage.

7Leave a comment on block 7 0

7. I’m going to take this notion of breakage further in two ways. First I’m going to look at cultural criticism, offspring of high theory, parent of the new media analysis found especially in California and Durham, North Carolina. Second, I’m going to look at two instances of text-focused humanities computing, authorship attribution and text encoding and transformation using TEI and XSLT, something many of you are here to learn. In both cases, cultural criticism and textual DH, I’m going to ask, do they work? in what ways are they broken? I’ve shifted slightly Steve Ramsay’s notion that digital humanities needs to be about making. There will be a new humanities, a set of disciplines that will go forward to compose the new academy, whatever that turns out to be. Those need not necessarily make things but they need to work. I don’t mean work as in 9 to 5, but work as in un-neurotically able to love and work. They need to be able to work to be left standing after the revolution comes, which is not as it turns out communist but managerial.

8Leave a comment on block 8 0

8. Sounds dire, but of course things are dire at the moment. I was just last June hired to direct an emerging Digital Humanities Center, the Initiative for Digital Humanities, Media, and Culture, at Texas A&M University. Maura Ives and Amy Earhart wrote the whitepaper that garnished the funding to launch this center. Patrick Svensson recently quoted it in his essay “Beyond the Big Tent,” and it is soon to be published by Earhart and Ives, and is available here. It successfully made a case to the Texas A&M System to invest in the Digital Humanities broadly defined. The Humanities being given major funding, as the digital humanities, should be cause for celebration, and it is, but it is crucial to see the big picture. Since I have been at Texas—I arrived after the white paper, as a result of it—there have been two other major calls for initiatives: one was for interdisciplinary teaching that makes undergraduate and graduate students into co-researchers with faculty, another was a call for a case to be made by various colleges to bring in research fellows at the level of nobel-prize winners, with huge stipends attached. The IDHMC has been awarded $100,000 to co-teach with professors in the Visualization Department a course that spans several years, during which time we will build a Humanities Visualization Space in the new lab of our new building. Additionally, our selection for a fellow was approved by the college and then by the system itself. We seem to be the golden calf. But I sympathize intensely here with how Melissa Terras felt when giving her plenary speech at King’s College London for the DH 2012 Conference just after King’s had fired its only palaeographer—the blog you see on the screen discusses it. This professor, they said, the only one who could teach this discipline at King’s, was too expensive to keep.

9Leave a comment on block 9 0

9. Similarly, though not quite as drastic, the Texas A&M English Department has had funding cuts to its graduate program that are quite severe: no more fellowships for MA students, no more postdoctoral funding to give our Ph.D. students teaching positions while they look for jobs. In fact, the RFP for the teaching-research award that we received was called the “Strategic Reallocation Plan”: the university is strategically re-allocating funds—that is, taking them away from existing streams and channeling them into new ones. That it continues to bet on digital humanities is good—we may in fact be able to get a Ph.D. program through the approval process when others are being cut. But the fact of the RFP shows, in my view, that the Texas A&M managers are asking professors to compete for strategic reallocations of funding as a way to see how the university might be restructured from the ground up. We need to know what’s working in the humanities disciplines in order to promote it, champion it, and argue for it as the academy is being re-structured. I’ll lay my cards on the table here: promoting the liberal arts by saying that they teach critical thinking skills has not worked and it will continue not to work. I want to explain why.

10Leave a comment on block 10 0

10. Alan Liu recently issued a state-of-the-field report in the February 2012 issue of the Arts and Humanities in Higher Education journal, via the title implicitly comparing his own essay which might be called “the postindustrial condition” to Jean-Francois Lyotard’s The Postmodern Condition: a Report on Knowledge (1984). Here Liu reiterates themes he has chimed before: the need for the “‘text-based’ digital humanities” to take up the cultural criticism that informs the “new media studies side” of the field (12), and the need for the digital humanities in general (now including public humanities as well) to save the humanities—he doesn’t quite put it that way, he says exactly this: without taking up cultural critique,

11Leave a comment on block 11 0

11. The digital humanities are not yet prepared to accept their likely future responsibility to represent—both by critiquing and by advocating—the state of the humanities at large in their changing relation to higher education and the postindustrial state. (28).

12Leave a comment on block 12 0

12. Many of us will remember Liu’s impassioned plea to that effect, delivered before this essay was written, at the Centernet Luncheon of DH2011 which was hosted by Stanford University. But there is a new theme here in the state-of-the-field essay as well: Liu admonishes the more theoretical side of DH, media studies, for not paying sufficient attention to the condition of their own possibility, viz. the academy as it now stands, before any strategic restructuring which is surely designed to take away the little that theory now has. The high theory people will tell you that the university wants to shut it down because it is politically left-leaning, and that may be true. But cultural criticism, the very theory that fostered new media studies in its current assault on empire and global imperialism—that’s broken.

13Leave a comment on block 13 0

13. Postmodern theory rightly in my view critiqued the aestheticism that gave us totalizing discourse ensconcing canonical literary works in an aura that only the self-proclaimed priests of high culture such as Harold Bloom could adequately help people understand. Lyotard in particular attacked that totalizing narrative, along with other such narratives. But then postmodern cultural critique too often slipped into what David Simpson calls “parodic literary history” (SubStance; Praxis): we literary critics are demystifiers par excellence, and the past is full of those bad deluded ancestors. Bloom in response said, you—all you postmodern schools of cultural criticism, Marxism, feminism, etc.—all of you are like lemmings following each other off a cliff only to be dashed against the rocks by the incoming waves of the sea. I disagree with Bloom’s canonizing ethos, don’t get me wrong, but he is right in this: we cannot defend teaching a canon or history simply as historical blindness and ideological delusion exemplified; there is no reason to know the past if it is only that. [1]

14Leave a comment on block 14 0

14. I’m about to echo and of course re-interpret Julia Flanders’s excellent question after Liu’s luncheon talk: we don’t want to save the humanities as they were constructed, do we? In my interpretation and elaboration of this comment: how do we defend teaching cultural history in a way that works, that is, not as cultural critics currently do by purveying it as an exercise in critical thinking, honing one’s skills at demystification, but also not as the Harold Blooms would like us to do, without recidivism into Priesthood? What will work if sheer critique does not?

15Leave a comment on block 15 0

15. Paradoxically, one of the descriptions of reading that impresses me the most as what literary history should do comes from Thomas Kuhn in discussing how he came to realize that there were paradigms and paradigm shifts in the first place, the basis for his book The Structure of Scientific Revolutions. He was reading Aristotle’s Physics, and he asked himself, how could someone as smart as Aristotle say things that were so patently wrong about the physics of objects?[2] How could Aristotle be so stupid? He suddenly realized, though, that the underlying assumptions about matter, the system in which Aristotle operated was completely different than ours: in the context of that system, what Aristotle said made sense—Aristotle thought of motion as a quality of changeableness in matter, not as something affecting it from the outside. In order to understand Aristotle’s meaning and reconstruct that historical moment of shared agreements in which Aristotle operated, agreements too obvious to him to articulate directly, or even only half-consciously assumed, Kuhn had to believe that Aristotle knew something he didn’t, that Aristotle’s broken capacity to communicate wasn’t Aristotle’s fault but something about the difference between ancient Greece and now. To dive from there into Aristotle worship à la Bloom is no better (but I have to say no worse) than quoting Marx or Judith Butler or Derrida as unassailable authorities. They each have their systems, of course, and each system becomes comprehensible as we explore breaks in our understanding: in their cases, breaks made for philosophical reasons, in Aristotle’s, a philosophical break caused by history. Giving each of them the benefit of the doubt, we ask, “What form of life would it take for this statement to make sense?” Kuhn gets at Aristotle’s, at the Aristotelian paradigm. (If you want to see the form of life imagined by the disciplinary jargon of Anglo-American analytic philosophy, look at the poor half-persons yelling “slab” at each other in the opening parable of Wittgenstein’s Philosophical Investigations.)

16Leave a comment on block 16 0

16. Absent the canon, and absent the bourgeois system that privileged a person’s knowledge of high culture, what motivates reading intensively the writings of a dead person? Rather than constructing a parodic history of praise and blame, the best cultural historians conduct a kind of self-analysis as they write, usually—or let me say, most visible to me—in the form of the strategic and creative use of anachronism. Just to give a quick example, Deidre Lynch examines how the triple-decker novel, the romance, has contributed historically to the way we conceive love, those shared beliefs hidden in the dead metaphors used to speak about it but nonetheless part of our shared agreements. She performs this analysis partly by using the anachronistic phrase, in her analysis of triple deckers, “going steady.”[3] Reading a novel in three volumes is rather like going steady with it, and that’s more than just a coincidence—that’s the legacy, the impact, of our cultural history on our forms of life. What could be more important than studying how history has shaped one’s life?

17Leave a comment on block 17 0

17. Many of the reasons we have for studying the past, for our intellectual loves, are disciplinary: it isn’t just students who read because the book was assigned and there would be a quiz. Sometimes we read specific authors because we are marching in lockstep to a disciplinary drum, a metaphor from Liu’s collected essays Local Transcendence.

18Leave a comment on block 18 0

18. How is the author functioning in the case of parodic literary history, in canonizing, and in disciplinary lockstep? As a mere function rather than a person deserving as much respect as one wants oneself, idealized or demonized, the author cannot do his or her work. Searching for the author’s meaning means finding his or her and one’s own historical moment by teasing out the author’s shared agreements and of course immediately comparing and contrasting them with our own. Does the concept of the author work in digital attribution studies, of the sort practiced by Ian Lancashire, John Burrows, and our very own David Hoover, teaching this week? Hoover has found that, although authors do not have “an invariable wordprint” when texts are analyzed digitally using Burrows’s Delta, the algorithm’s failures “are all failures to attribute rather than failures of attribution,” suggesting that, though alterable when an author chooses, he or she does indeed have a style.

19Leave a comment on block 19 0

19. In disciplinary protocols surrounding authorship digital textual studies and theory have never seemed more opposed. Ian Lancashire discovers Agatha Christie’s progressive dementia when it seems as if, for Roland Barthes, no Agatha Christie exists at all. And yet I think Roland Barthes’s “Death of the Author” essay has been grievously misunderstood, by cultural critics particularly. I remember teaching it in a graduate class, an Introduction to Literary Theory. Christopher Coake, then a creative writing MA, looked at me plaintively: how can he say the author is dead? But the author qua person was not Barthes’s target. Barthes was attacking literary critics for invoking the author as a means for interpreting the text ONCE AND FOR ALL, reducing its polysemy, putting to rest definitively any questions it raised. Walter Jackson Bate knew that Samuel Johnson took tea at 4 p.m., precisely, every day, and also that “just representations of general nature” mean “realistic descriptions,” end of story. Christopher Coake’s debut novel You Came Back was positively reviewed by Slate Magazine: who could believe, the reviewer wondered, that a writer who isn’t even a parent could write so movingly and effectively about a father’s grief over the death of a son?[4] The fact is that he can, and so knowing when Johnson took tea is not a legitimate interpretive move. But in “The Death of the Author” essay, Barthes actually says “Death!” death not only to the author as an instrument for foreclosing interpretations but also all his “Hypostases”: Society, History, the Psyche—and that includes global capitalism and even postindustrial society. I’m a feminist who has written feminist literary history, but there is a difference between seeing Sexism capital S as the cause and explanation for what a text says and reading a text to see precisely in it how gender is figured and what it’s up to.

20Leave a comment on block 20 0

20. There is a point here at which disciplinary lockstep must stop. The literary critic has to say, for instance, the canon’s formation is built partly by the history of Shakespeare’s reception, but no, Christopher Coake, you are not dead. It has to look ordinary people in the eye and give over its excessive criticalness. In ordinary language, being “critical” means being unduly negative, and the discipline of new media studies too often channels Orwell’s 1984 in its discussions of global capitalism. But do such critics live on Mellon, Carnegie and Guggenheim funding? Is their critique funded by the money of robber barons? Liu would say, I believe, that there is a failure here on the part of cultural criticism, new media studies included, to recognize that high theory is a luxury affordable only when capitalism or some other surplus producing economic system is working. Whenever I hear a colleague defending the old humanities dispensation that produced cultural studies based upon the fact that they inspire critical thinking, I remember Blake’s poem, “The Ale House.” In it a little boy is admonished to go to church instead of the pub, and he answers, “I know where I’m treated well.” Critical thinking means in ordinary language being unnecessarily negative: it is a PR failure, as well as an intellectual failure to take into account the conditions of production. It’s dominance now, I would suggest, is the result of an excessive and unexamined lockstep disciplinarity.

21Leave a comment on block 21 0

21. So does digital textual studies collude in disciplinarity gone mad? Well, we’re out of time now. No, just kidding—I will wrap up here shortly, but I do have something to say about that. I told my daughter you would all throw tomatoes, and she said that tomatoes are good for your complexion, if a little hard on the wardrobe, so it’s all good.

22Leave a comment on block 22 0

22. My argument based on Ordinary Language Philosophy is that excessive disciplinarity is always wrong-headed, always a failure to take account of ordinary agreements that are neither affirmable or rejectable, after all; they are just the water we swim in, and it is not possible for us to live some other life than the life we live. That doesn’t mean, “don’t think”—the whole point of this talk has been precisely to advocate looking for what breaks as a way of figuring out how things work. Trouble-shooting reality, as it were.

23Leave a comment on block 23 0

23. I’ll take one small instance of TEI encoding. The premise of TEI is that it is a semantic markup system, a purely descriptive encoding of texts that cares not at all for how the encoded documents are made manifest to people on the other end of production—it is not concerned with what texts look like when presented to people for reading. This separation of underlying code and interface is crucial: we depend upon it insofar as html will certainly not be the typical way that texts are made manifest in the near future. If we encode texts descriptively rather than for look in one kind of code serving one particular medium at a certain stage of technological development, then no matter what new technologies come down the pike, we can use XSLT to transform our documents into that new format. That is a tremendous comfort.

24Leave a comment on block 24 0

24. But if TEI is the church, HTML is the ale house, and I want to give one example of an instance when I think the church might learn from the alehouse.

25Leave a comment on block 25 0

25. In encoding Lynda Pratt’s edition of Robert Southey’s Letters, we came up with a method for how to deal with references to people and places, a method that is indeed allowed by the TEI guidelines—these documents validate using a TEI schema—but that many people would call tag abuse. Whenever Southey mentioned people’s names, the editors sitting in the pub wanted the names to be linked to a biographical document describing all the people in Southey’s social circle, another HTML page. But of course some people mentioned in Southey’s letters such as King George or Aristotle are not members of his circle. For those people, the editors wanted note superscripts that people could click on and be taken to notes filled with historical information. So for those mentions of people, we used note tags. For Southey’s friends, family, and acquaintances, we used in TEI the ref element with the target attribute containing a the name in a form that it could be used to generate via the XSLT a link to the name anchor in a separate bibliography document. That way of coding, completely pub- or presentation-dependent, has been incredibly useful semantically: distinguishing who Southey actually interacted with or lived with habitually, at least vicariously, through friends or news, from figureheads who occasionally enter his mind, that’s an incredibly useful distinction, one that cannot be made by machines. (I’m even having trouble formalizing it in linguistic description, so much more forgiving than code.) If I told you, an avid reader of all Southey’s works and letters, to put members of his social circle in a pile here, and people he just mentions in a pile there, you might have trouble with Godwin or Milton, but you would sort things pretty well. Here, precisely, is an instance of what Johanna Drucker is talking about when she says that research is a matter of organizing one’s workspace.

26Leave a comment on block 26 0

26. In his article “Transcendental Data: Toward a Cultural History and Aesthetics of the New Encoded Discourse,” Alan Liu argues that code has managerial implications and side-effects.[5] One could imagine an academic sweatshop replacing Lynda Pratt with a coder and presenter, each working in separate offices from each other, neither having anything to do with one another. Fordism of a digital kind. What’s the difference between the sweatshop and the academy of expert researchers that we want in our future? When you use a dynamic graph to visualize relationships in the Southey letters, you can see that David Garrick doesn’t pop up as Southey’s friend. A machine might be able to weed out Shakespeare and Milton, based on their death dates, but no machine, only a human expert, will know that Southey is just mentioning something he heard or read, that he is not at all engaged with Garrick in any way. Tag abuse—a stumble in the disciplinary march, motivated by intellectual rigor—argues for the necessity of humans in the coding process, experts, even, and ultimately, as many, many versions of this one little instance when human experts are needed build up, collectively arguing for the necessity of the humanistic rigor rather than sweatshops following orders—the kind of rigor that we wish to preserve. So we digital humanists can learn from cultural criticism not to repeat its mistakes, not to exclude the ordinary from view—here, the pub, the interface—not to discipline ourselves into insignificance. I have one last thing to say about Liu’s article. There is an appendix containing some XSLT code. I tried it, and it works. Thank you.

27Leave a comment on block 27 0

27. [1] In an excellent and more nuanced essay, John Guillory describes the anti-realist tendency in literary history, which I see as underlying this ethos as the “spontaneous philosophy” of our discipline (“The Sokal Affair and the History of Criticism,” Critical Inquiry 28.2 (2002): 470-508, p. 475).

28Leave a comment on block 28 1

28. [2] Thomas Kuhn, “What Are Scientific Revolutions?” in The Probablistic Revolution: Volume I: Ideas in History, eds. Lorenz Kruger, Lorraine J. Daston, and Michael Heidelberger (Cambridge, MA: MIT Press, 1987), excerpted http://www.units.muohio.edu/technologyandhumanities/kuhn.htm

29Leave a comment on block 29 0

29. [3] Deidre Lynch, “On Going Steady with Novels,” The Eighteenth Century: Theory and Interpretation 50: 2-3 (2009): 207-220.

30Leave a comment on block 30 0

30. [4] http://www.slate.com/articles/arts/books/2012/06/christopher_coake_s_you_came_back_reviewed_.html

31Leave a comment on block 31 0

31. [5] Alan Liu, “Transcendental Data: Toward a Cultural History and Aesthetics of the New Encoded Discourse,” Critical Inquiry 31.1 (2004): 49-84; reprinted in Local Transcendence: Essays on Postmodern Historicism and the Database (Chicago: Univ. of Chicago Press, 2008), 209-238.

page 5