Ted Underwood on Intellectual Genealogies: Distant Reading is Social-Science, Not Digital Humanities [#DH]

Ted Underwood, “A Genealogy of Distant Reading”, DHQ Vol. 11, No. 2, 2017:

Abstract: It has recently become common to describe all empirical approaches to literature as subfields of digital humanities. This essay argues that distant reading has a largely distinct genealogy stretching back many decades before the advent of the internet – a genealogy that is not for the most part centrally concerned with computers. It would be better to understand this field as a conversation between literary studies and social science, inititated by scholars like Raymond Williams and Janice Radway, and moving slowly toward an explicitly experimental method. Candor about the social-scientific dimension of distant reading is needed now, in order to refocus a research agenda that can drift into diffuse exploration of digital tools. Clarity on this topic might also reduce miscommunication between distant readers and digital humanists.

Rather than attempt to summarize it myself, I’ll present a set of tweets by Alan Liu, starting with this:

Made 41 annotations in my copy of @tedunderwood's new, crucial DHQ piece, "A Genealogy of Distant Reading" https://t.co/CIuXj7uAdH.

— Alan Liu (@alanyliu) July 10, 2017

Liu continues with a long series of tweets, which I’ll present as quotes without the Twitter format. Along the way I will present brief comments of my own, thus inserting my own concerns into the argument.

Liu begins:

Here are my top 13 quotes--a kind of thirteen ways of looking at distant reading, cited by paragraph number.

(As it were: "Among twenty snowy mountains of texts, The only moving thing Was the eye of the distant reader"):

¶5: "The questions posed by distant readers were originally framed by scholars (like Raymond Williams and Janice Radway) who worked on the boundary between literary history and social science."

¶10: "these projects … pose broad historical questions about literature, and answer them by studying samples of social or textual evidence. I want to highlight the underlying project of experimenting on samples, and the premise that samples … have to be constructed"

¶21 "The crucial underlying similarity between [Radway & Moretti's] works, which has made both of them durably productive models for other scholars, is simply the decision to organize critical inquiry as an experiment."

¶22 "Distant reading is a historical science, and it will need to draw on something like Carol Cleland’s definition of scientific method, which embraces not only future-oriented interventions, but any systematic test that seeks 'to protect … from misleading confirmations.'"

¶22 "Literary historians who use numbers will have to somehow combine rigor with simplicity, and prune back a thicket of fiddly details that would be fatal to our reason for caring about the subject."

¶24 "I try not to join any debate about the representativeness of different samples until I have seen some evidence that the debate makes a difference to the historical question under discussion…. [S]amples are provisional, purpose-built things. They are not canons. It makes no sense to argue about their representativeness in the abstract, before a question is defined."

¶27 "Instead of interpreting distant reading as a normative argument about the discipline, it would be better to judge it simply by asking whether the blind spot it identified is turning out to contain anything interesting."

¶28 "Consensus about new evidence emerges very slowly: inventing an air-pump doesn’t immediately convince readers that vacuums exist…. But at this point, there is no doubt in my mind that literary scholarship turned out to have a blind spot. Many important patterns in literary history are still poorly understood, because they weren’t easily grasped at the scale of individual reading."

I have proposed the description and analysis of form as another blind spot. This, of course, is a blind spot at the level of the individual text. I assume, furthermore, that forms evolve over time. The only way to understand that is to examine the formal properties of an appropriate sample of texts.

Note also that I talk of “description and analysis” rather than “reading”. Reading seeks meaning. Formal analysis does not.

Finally, thinking about literary process in computational terms is quite different from programming a computer to do distant reading, or anything else for that matter. For example, my teacher, David Hays, directed RAND’s machine translation project in the 1950s and 1960s, but he didn’t program the computers. Thinking about language computationally is one thing; programming a computer (to execute computational procedures on text) is another.

¶34 [Mark Olsen's 1993 "Signs, Symbols, &Discourses"]"originally pulled me toward distant reading.... I still find it a prescient argument"

¶34 "… computers still contribute relatively little to our understanding of individual texts and authors. But computational methods now matter deeply for literary history, because they can be applied to large digital libraries, guided by a theoretical framework that tells us how to pose meaningful questions on a social scale."

Of course I have long advocated that we think about the process of reading and comprehending a text (literary and otherwise), as being irreducibly computational in some respect. This does not require us to use computers, though the time may come when he can use computers to simulate the reading process.

The formal features of texts are dominated by computational considerations.

¶36 "The best distant readers do in practice approach their projects as experiments…. But the experimental structure of our research is not always foregrounded when we write it up for publication…. It can be more effective to pretend that your work grew in a casually discursive, thesis-driven way, and then happened to be illustrated with some scatterplots you had lying around."

¶41 "Distant reading … is not primarily concerned with technology at all: it centers on a social-scientific approach to the literary past."

Note that when I talk of literary form as computational in kind, I am not interested in digital technology. I am interested in the human mind. Note also that I focused in on this paragraph in my own reading of Underwood:

I like this: pic.twitter.com/oezvWY0sZz

— Bill Benzon (@bbenzon) June 30, 2017

Returning to Liu's tweet stream:

¶43 "I think [distance reading] is ready to move past exploration. Large-scale literary history could now reorganize itself around clear research questions and rigorously advance our knowledge of the past."

Yes.

Also, literary historians might want to start thinking about literary history in terms of an evolutionary process, for evolutionary processes are about interactions among large populations of objects, such as texts and people, over long periods of time.

* * * * *

I’ve written (it seems like) endlessly about the topics in my comments. Two places to begin:

Sharing Experience: Computation, Form, and Meaning in the Work of Literature, Working Paper, 2016, 21 pp. https://www.academia.edu/28764246/Sharing_Experience_Computation_Form_and_Meaning_in_the_Work_of_Literature

Literary Morphology: Nine Propositions in a Naturalist Theory of Form, PsyArt: An Online Journal for the Psychological Study of the Arts, 2006, article 060608. http://www.psyartjournal.com/article/show/l_benzon-literary_morphology_nine_propositions_in

Downloadable PDF: https://www.academia.edu/235110/Literary_Morphology_Nine_Propositions_in_a_Naturalist_Theory_of_Form

The first uses Shakespeare’s Sonnet 129 and Barack Obama’s Eulogy for Clementa Pinckney as examples. The second addresses history and is my most comprehensive theoretical statement.