Week 1: Question 7 Close vs. Distant Reading
  • I want to raise another question that Alan brings up which has not really been discussed. Alan says clearly that that "distant reading" poses a threat to what has been the core value of the (literary) humanities, "close reading." Citing Kate Hayles, he notes that "close reading" since the 1968 debates has assumed "a compensatory role as what remained quintessentially literary"--in other words, even if one isn't reading literature, one is still using literary techniques to read texts.

    So--what do you guys think? Does "distant reading" really pose a threat to the traditional humanities because we're removing the human one step from the reading? Yes, all forms of "distant reading" still need humans to make sense of the output of the data analysis; but nonetheless, it almost seems like the human reader is being replaced by the machine. In other words, is "distant reading" what makes the digital humanities both so exciting--and so threatening--because of its radical difference from "close reading"? Liu goes so far as to say that "distant reading" changes our very object of study;“The relevant text is no longer the New Critical ‘poem [text] itself’ but instead the digital humanities archive, corpus or network”; where graphs start to replace blocks of text quotations.

    So what are your thoughts? Is "distant reading" scary? Why and why not?

  • 11 Comments sorted by
  • In some respects, distant reading is scary because it seems to stray so far from the Leavisite tradition in which so many of us were trained as literary scholars. I think there is also a fear that such an approach turns literature into data and, in doing so, strips it of what distinguishes it from, well, word soup. I'm inclined to think that close reading and distant reading together can be instrumental and instructive methods of analyzing a text. I also think it's not so bad of distant reading destabilizes the fetish of the "novel" or the "poem" a little bit - maybe we could use a little levelling at the level of "What is Literature?" Those of us working in fields beyond the canons of American or British literature in English know how definitions of "literature" have functioned as a tool of exclusion at best and a denial of humanity, citizenship, and legibility (I'm thinking the Phillis Wheatley trial here) at worst.
  • I really like what you're suggesting here, R. While I muse over this further, here is an example of how big data is scary to traditional humanists--Stephen Marche's infamous article, "Literature is not Data": "But there is a deeper problem with the digital humanities in general, a fundamental assumption that runs through all aspects of the methodology and which has not been adequately assessed in its nascent theory. Literature cannot meaningfully be treated as data. The problem is essential rather than superficial: literature is not data. Literature is the opposite of data."
  • I believe the definition of "distant reading" is no less ideological than that of DH itself.

    Taken "literally," distant reading isn't what we usually mean by "reading," a human being parsing a string of text. We say machines "read" texts, but we all know this is at least partly a metaphor. 

    Taken literally, much of what we do during the day is "distant reading," including Google searches, database searches, even RSS feed and Twitter searching. Using and building concordances was distant reading before it had a name. So was using the OED.
    But that's not how it's usually taken--it's taken as a special activity that overlaps with the definition of DH. In this sense, the term itself is meant as an threat to the one activity scholars from Spivak to Harold Bloom could agree upon as the sine qua non of literary studies.

    If we used a more pedestrian name for this activity--which as Heather Froehlich and others including me have pointed out repeatedly, is something very close to corpus linguistics--the frisson that would be lost just is a mark of the ideological work done by the term. "Corpus linguistics" is definitely not exempt from ideological weight in linguistics, but it does not carry the suggestion that "you old folks don't realize your methods are totally irrelevant today" that the term "distant reading" brings along.

    Is "distant reading" actually a threat to close reading or literary studies? I don't know, but I think that like DH, the alternate pedestrian name exercise -- in this case let's call it "corpus textual analysis"--shows that there is something at stake in the terminology, beyond method.

  • David, that's precisely my point. The "narrow" definition of distant reading that you bring up threatens our notion of both the literary object (from the poem/passage) to a larger corpora of data. 

    What for me is disturbing is when I see distant reading being bandied about/applied without enough scrutiny to how/why/what enables the method (to) create its meaning; in other words, the application of tools without enough of an explanation/investigation of the tacit knowledge behind the tool.

    This for me is both disturbing yet fruitful. To intervene directly in this point is to emphasize the "humanities" part of the digital humanities, by examining, questioning and making explicit the ideological implications and biases behind the methodologies used for distant reading. This is why, as Heather Froehlich and Michelle Moravec have explained to me so far, corpus linguistics is especially attractive as it makes the meaning-making of distant reading transparent through showing its analysis of the relationship linguistic elements to one another. 
  • more violent agreement :)
  • I think this is a fascinating quesiton, one I struggle with all the time. I think D's points are correct -- perhaps the distinction between 'close' and 'distance' reading is more ideological than anything else. Might not distance reading (for lack of a better term) supplement our close readings? One of the real benefits of DH and close reading, at the very least, is that is asks us to rethink the meaning and practice of received notions such as close reading.

    Marche's argument doesn't make much sense to me. He writes, "There are nine different versions of Richard III; there are three versions of Hamlet, each with missing sections or added sections." Isn't this exactly the kind of puzzle for which distance reading can offer a unique perspective? Furthermore (and I'm not a Shakespeare scholar so I may be totally wrong here), how much Shakespeare criticism these days is really involved in ye olde close reading and how much is archival work? Is this gesture to protect close reading from data analysis really a defense of our work as literary scholars (retro 80s term) or is it just rejection of an exciting new method of analysis?
  • I agree with you, Adeline and David, and think these are important points. I think it’s easy to overlook all of the work behind
    not only the tool (the analysis) but the “data” itself in this kind of reading.
    There’s the issue of the transparency of the tools themselves, but there’s also
    the question of corpus.

    To what extent do digitized corpora perpetuate existing gaps
    in the archive? How do the processes required to convert books from print to
    screen change the shape of the corpus itself? In large-scale digitization
    initiatives (e.g. Google Books) books that are too small or too big or
    otherwise experiment with form and content are often excluded or distorted by
    the digitization process. Google’s emphasis on OCR takes a large step in the
    direction of defining books in terms of words (that which is translatable into
    machine readable – and therefore quantifiable – text) in a way that elides the
    important history of text and images (as evidence, illustration, or content)

    There’s an increasing need, I think, for a tool literacy which
    includes a better accounting for both the form and content of digitized corpora
    and what, exactly, these corpora are being asked to represent. As you pointed
    out, interrogation of ideological assumptions and biases is something for which
    humanists are well equipped, but the technical nature of the methodology here
    presents a challenge.

  • Paul B has a great point. I don't know that the two can be so easily separated, just as I am uncomfortable with the split between "traditional" humanities (invoking, if I have it right, classical philology) and this "new thing that we do" with the digital. I don't think you can do distant reading without close reading. Going back to postcolonial basics, Said did his work on Romanticism first by closely reading and then by stepping backward and reading at a distance. I think the same might be said of how the Digital Humanities performs the kinds of work that we do with the tools that we have and the materials that they house and that they produce.

    We must closely read the artifact to understand its placement in the archive and to then consider the politics surrounding the archival practices that grant access to the artifact. I prefer to think about more concrete examples--what would the politics of access look like for those seeking to archive French ephemera surrounding the Battle of Algiers? Considering what this would mean, from the resources that digitize the materials and the choices of the tools to do it to the consideration of who would be served or shut out of this archive, requires understanding the objects in the archive in ways that close reading reveals and then understanding the modes of circulation that grant them power or marginalize them from certain spheres of discourse.

    I'm still chewing on this excellent question. After more coffee and more thinking, I'll perhaps post a more pointed response; I could not resist chiming in, however, at this inchoate phase of my thinking because this problem has been one that has intrigued me ever since Old Dominion University began heavily investing in the Digital Humanities. We have a robust distance ed program here and it's supplemented by our decisions to hire digital humanists that work beyond their own individual course offerings to achieve widespread pedagogical influence. It's housed in the departments of English and Communications and we grapple with the manner in which being subsumed beneath the umbrella of the Humanities (and its philological associations!) risks alienating CS, who can be a valuable ally in shaping the form of what we do. The politics of that separation frustrate me.
  • So, I decided to compose a blog post for this question. I've included an excerpt here. If you want to read the rest, please follow here: http://benjaminjdoyle.org/2013/07/04/dhpoco-summer-school-q7/

    Okay, first, I say we do away w/ the anachronistic rule of Literature and Literary Studies
    as culture of taste, sacred text, NIMBY, don't you dare d[ata]file my
    dear Shakespeare! Pshoo! see ya, good bye. If a critique of distant
    reading is always grounded in a rhetoric of sacred art, then I don't really know how to engage that as an open discourse/debate — seems far to
    "figured out" to be worth paying any serious attention to. Just sayin'.
    Are we really still studying literature for the sake of lit(ə)rə-ch-ər
    (exact pronunciations please)? Am I wrong in asking this?

    In considering the question of "close" vs. "distant" reading, my mind
    always returns to Morretti's argument for an understanding of literature
    as necessarily "world literature" — a “planetary” system of "one
    and unequal,” a “problem” that must inform our approach, appreciation,
    and treatment of the (con)texts and forms of culture (http://newleftreview.org/II/1/franco-moretti-conjectures-on-world-literature).
    Morretti’s purpose is most compelling in moments (sometimes subtle)
    that underscore the stakes of thinking of the political aesthetic of
    literature and literary forms in global terms, terms that are for him
    problem-posing, leading toward new categories, new problems, new
    critical methods, (a kind of Butler-esque spirit of a necessary,
    generative “trouble[ing]” of the world of/as a given form in favor of the (non)relation between variations of form(s)).

    The call for a critique of "close" reading in favor of distant reading for
    me, however, is not so much a critique of the practice but of the
    politics informing the practice. I’m less interested in critiquing a
    meaningful playfulness w/ a single text or small body of texts, though
    making big claims from small data can be problematic (yet potentially
    generative, too, leading to provocative / productive debate???). In the
    context of Morretti, I’m less excited (though maybe I shouldn’t be) by
    his effort to produce a more generously scaled discussion of the novel
    (i.e., by adding texts to the count — mere increase can itself become
    problematic) than I am by the political perspective he brings to this
    increase, such that an operation of critique guides his building-up a
    corpus — a concern w/ the asymmetries of political inter actions and
    cultural forms at a global scale: source lit vs target lit, where
    periphery cultural systems are infiltrated, re-formed, or ignored by
    dominant cultures. Like Morrettti, I am more concerned w/
    perspectives/purposes that refuse to acknowledge through their practice
    the value/import of that which remains outside the familiar frame. The
    negation for normalcy sake, yeah, that’s reason enough to be troubled.
    But can big data and its analysis do more to unsettle the normative
    politics at work in traditional practices that exclude others from the count?

    Big data and distant reading demands their own critical methodologies that
    will maintain an always already approach to interrogating the
    absences/gaps/silences in any form of mass representation. I always
    point my colleagues and students interested in the debates over big data
    to Kate Crawford’s critique of the assumed neutrality of data in “The
    Hidden Biases in Big Data” (http://blogs.hbr.org/cs/2013/04/the_hidden_biases_in_big_data.html).
    Crawford’s discussion of those communities absent from the data set, so
    often the very same folks whose socioeconomic, political, raced,
    gendered position situate them as well outside a represented part of the
    given social set, reminds us, in much the same spirit of Alex and
    GO::DH and the partners of DHPOCO, that the continued lack of access to
    technology itself calls for a more deliberate discussion of an
    uncritically situated valuing of technology, and a more careful,
    thoughtful eye toward our understanding of how and/or whether certain
    technologies will help us understand better social issues and their
    possible resolutions.

    ( . . . http://benjaminjdoyle.org/2013/07/04/dhpoco-summer-school-q7/)

  • I applied the concept of close and distant reading at a different level in dealing with objects in physics and ideas in physics in this paper http://www.academia.edu/1866351/Speculative_Reading_Speculative_Physics-the_Ontology_of_the_Large_Hadron_Collider

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Login with Facebook Sign In with Google Sign In with OpenID


In this Discussion