Remarks I read (much too quickly) at the MLA roundtable organized by Jim English. Feed of Twitter response to the panel here. Some of this drawn from the scale piece, some from remarks on close reading in On Literary Worlds, and some from thoughts leading into the new book, now tentatively titled What Kind of Information is Literature?
I want to begin by arguing that the current state of affairs with respect to “data” and “literature,” itself a mirror of the entire structure that organizes the cultural relationship between the digital humanities and literary criticism, is bad for proponents on both sides. I mean in the most general possible way, but here I want to focus especially on the antagonism between data-based analysis of literary texts, which has been called “distant reading,” and the more historically traditional reading practice of focusing on small units of meaning, which we call, pretty loosely, “close reading.”
The first thing to say is that distant reading is not really distant, and close reading is not just close. No reading practice ever maintains itself as one “distance” from a text; rather what we call a reading practice is among other things a pattern of system of habitual distances and relations among those distances. So “close reading” is not always close; rather it pairs a certain kind of analysis of relatively small pieces of text with very powerful analytic tools—the tools of New Criticism, but also of psychoanalysis, deconstruction, new historicism, and so on—that leverage those small pieces of text into structures that are more “distant” from the text than is, say, the sentence or the phoneme. As the farthest level of distance these readings manage to make claims about some of the largest possible conceptual structures in human society, namely the nature of being (or beauty), the organization of the unconscious, the ethics of language, or the totality of an era. On the way they almost inevitably pass through other levels of what we might think of as “distance” from the text, in which they both use (as tools) and make claims about (interpretively) things like subgenres (sonnets, science fiction), genres (poetry, the novel), modes (epic, lyric), and so on.
Close reading is not, I say again, close; it is an arrangement of closeness and distance that behaves as though its epistemological fundamentals took place entirely at the level of “closeness”; whereas in fact as in any system these fundamentals operate as part of a larger pattern. You could say the same for “distant” reading.
What I want to propose is that modes of reading contain buried theories of what kind of information literature is. And I want to suggest that literature is a both a very particular kind of information but also, that this particularity constitutes not a difference in kind from other kinds of information but rather one of degree. I am willing to make this argument both ontologically and pragmatically, but for now since I have no real time I will simply say that pragmatically if we could think of literature as information—the same way we have learned to think of the codex book as a medium instead of the thing that media were against—then we would be on our way to getting rid of the somewhat stupid antagonism between the digital/data-oriented/distant model and the older analaog/close reading model, which would free us all up to do a wider variety of work, and to think our old categories through in interesting ways via new ideas.
I’m going to close by rewriting a couple sentences of the anthropologist Terence Turner’s, which are part of an argument he’s making about the role metaphor plays in the production of social life, in which he (like me) wants to emphasize the contextual, contingent, and essentially degree-oriented differences among various tropic and social practices, against those who want to make those differences differences of kind. I only mention that I’m rewriting the sentences because I don’t want to be accused of plagiarism:
“It is essential to understand the structural continuity of the step from information to literature and back again—in other words, to grasp the nonuniqueness of literature an absolute structural sense—in order to appreciate the nature and importance of literature’s relative specificity and distinctive role in the construction, and continual reconstruction, of new or distinct contexts of cultural meaning and subjective consciousness. That the difference literature makes to the history of information is not fixed and qualitative, but pragmatic and, as it were, quantitative, does not imply that such a dimension of difference does not exist, only that it is a relative, fluid, and quantitative matter.” (Turner 129)
More specifically, I want to ask what happens if we think of literature as the site for the storage l of information, and if we think of literary criticism, then, following that model, as a series of efforts to retrieve that information. We then come back to the question that organizes this roundtable, “What is Data?,” and begin with an answer, not in the form “Data is X or Y or Z,” which I suppose is the usual way to answer such questions, but rather by saying, “Literature (among other things) is data (among other things),” which leads to the question, “What kind of data/information is literature?” … which is a question I would like to answer.