A Diet of Worms in the Digital Age

by Charis Chapman

I can’t prove it, but there is no doubt in my mind that textual arguments have been raging in scholarly circles for as long as there has been text to debate. In my mind’s eye I can see them: ancient Sumerian scribes lecturing each other about clay types and wedge depth; early Semitic peoples voting “no” on the vowel; medieval European scribes boldly pushing forward with punctuation, spaces between words, and the lower-case alphabet, and having heated debates on the long-term viability of the capital letter. And then came the printing press! Can you imagine the contention! What bold new vistas were opened up for scholarship! But anyone could publish anything—no matter what the quality! And surely, this spelled the doom of calligraphy.

With the advent of the digital age, scholarly textual debate has simply entered a new phase. At issue here: what is the potential of digitally-powered scholarship, and how can that potential be realized? What approaches should we take in terms of format and methodology?

William H. O’Donnell and Emily A. Thrush (“Designing a Hypertext Edition of a Modern Poem”) discuss the issues involved in designing hypertext editions of literature. Specifically, they refer to the edition of Yeats’s “Lapis Lazuli” that they designed. The main work to be studied, they feel (be it “Lapis Lazuli” or something else), must not be cluttered with visible links. They have devised a method of windowing that anticipates the modern “frames” format of Internet documents, and stress that any attempt at electronic analysis of a work of literature must be intuitively structured, easy to use, and customizable. Their edition of “Lapis Lazuli” appears well-constructed and functional and seems to have fairly broad appeal, but it seems also to be intended more as an aid to understanding the poem than as a tool for scholarly research. This distinction separates this article somewhat from the others considered here, though the basic format could be applied to other projects.

Peter Shillingsburg (“Principles for Electronic Archives, Scholarly Editions, and Tutorials”) frankly admits that what we all secretly want is to have every conceivable kind of information—textual, visual, audio—related to a given topic all sorted for us and available right at our fingertips. This is, of course, impossible, but he feels that electronic editions of scholarly works have the potential to come closer to that ideal than any other medium. He systematically lays out the main problems facing those who create electronic editions and suggests some ways to address them. Much of his discussion relates to technology which is now long outdated (When was the last time anyone used DOS?), but many of the underlying issues are still valid. Editors of scholarly editions must, for example, take into account distribution methods and long-term usability issues. Maybe that hot new technology won’t work in five years; how will you provide for upgrades? He also feels that electronic editions should be open to comment, criticism, appendation, and revision by those who use them—a pretty neat idea, though obviously the process would need to be regulated. He also makes one point worth mentioning: “It is very difficult to determine what to include and how to link. This part of the concept is very much dependent on talent, interests, and resources of the archive compiler” (34).

I wish his colleague, Susan Hockey (“Creating and Using Electronic Editions”), had given more thought to this. Hockey dreams of the day when computers will do all our reading and interpretation for us. Scholars of the future will be little more than compilers of links, and the scholarly articles they publish (electronically, of course) will consist entirely of a few words conceptually linking the links. Much of her article is devoted to suggesting the ways in which the computers of the future will interpret text, and how we can encode it now to facilitate that process. There is nothing wrong with the general idea, she just carries it to extremes. She proposes using on a tagging system called TEI to format text and give information about it. Now, a tag is basically a little bit of text, offset from the main text by a <tag> that tells the computer what to do with the text following that tag. For example, if I wanted to put a word in italics using tagging, I would do it <i>like this.</i> The average user would not be able to see the tags, but would only see the italicized text—like this. Hockey sees a lot of potential here. All kinds of information could be given in the tags, and the computer could be told to manipulate them in any number of ways. To give an idea of what he is talking about, I have tagged a short poem:

<author>Robert Frost</author><title>Nothing Gold Can Stay</title><poem>Nature’s first green is gold,<line>Her hardest hue to hold.<line>Her early leaf’s a flower;<line>But only so an hour.<line>Then leaf subsides to leaf.<line>So Eden sank to grief,<line>So dawn goes down to day.<line>Nothing gold can stay.</poem>

I don’t think these are real TEI tags, but they are similar to what I saw online. Hockey’s point is that these tags might not only be used to tell the computer how to format text, but would also prove useful to the scholar who’s analyzing it. For example, in an electronic collection of poems tagged in this way, you could choose to search or browse by author or title, or by a line of your choice. If you really wanted to, you could quickly and easily compare all the fifth lines of all the poems by authors whose last name begins with “T.” It would also be possible to indicate poem type (e.g., sonnet, villanelle, limerick, blank verse) and rhyme scheme via tags, so that it would be possible to search for and compare all poems that contain four-line stanzas with an ABAB rhyme scheme. You could search for a word and see what it was most frequently rhymed with. The potential here is clearly huge, but tagging is an incredibly laborious and time-consuming process, and it requires more technical expertise than I fancy your average humanities scholar will ever possess. The tags would also only work for English. Completely different tags would have to be devised for other languages. It would therefore take a really organized, well-funded international program to create the kind of sophisticated database Hockey envisions, and given the current state of academia, both in this country and outside it, I really can’t see her ideas being implemented on a large scale.

By contrast with his colleagues, Warwick Gould (“A Diet of Worms: Descriptive Bibliography and the Structure of Digital Editions: The Example of W. B. Yeats”) emphatically does not want his new-fangled technology to do his reading for him. I must confess that this issue was something that bothered me while reading the other articles.

Take Hockey, for example. She envisions a world where most aspects of interpretation are built into the text through tagging—right down to “separating the different meanings of homographs and for lemmatization (putting words under their dictionary headings)” (15). What if a poet’s or novelist’s usage of a word does not fall under a specific dictionary definition? How would a computer make sense of figurative language? Should we make it the computer’s job to make sense of figurative language? Surely these interpretive tasks are better left to the individual reader! Hockey reveals her true colors again when she laments the fact that “[a]nyone can put anything on the Internet, and it is then entirely up to the user of the information to determine whether it has any value.” Because of this, “[t]here is no way of knowing whether a document is accurate other than to read it and check it” (13). The solution: reliable authentication procedures for those creating and editing electronic text. But, authentication or no authentication, are either of these flaws any less a part of print media than of digital? It is true that documents on the Internet are often full of errors. But so, too, is the newspaper. And what does he mean by “value,” anyway? Deep waters, indeed.

O’Donnell and Thrush, meanwhile, have created a nice little edition of “Lapis Lazuli,” which they use to make some useful observations about the issues surrounding the creation of electronic editions. But they go astray towards the end of the article. Stating that the editor of a hypertext document should consider (among other things) “[t]he reasons readers have for reading the document,” “[t]he reading strategies that will be used,” and “[t]he amount of previous experience the intended users have with similar documents” (208), they are treading on treacherous ground. If you are creating a web site for use in elementary schools or to instruct museum-goers about the paintings on exhibit, these are certainly valid concerns. They are not, however, really germane to scholarly publication, and I don’t see that they apply to the sort of electronic edition the authors have been discussing throughout most of the article. It is one thing to say a piece of scholarship should be accessible to as many different types of readers as possible, but at some point an editor has to draw the line and say, “This is what I am going to write.” It is not a scholarly editor’s job to present the exact information that he imagines a reader or group of readers expect or want to see. It is not his job to please them—though if he does a good job he probably will—or to cater to their agendas; they must take his work as they find it. It is not, in short, his job to do their reading for them. His job is to present, in a comprehensive manner, that information which, in his expert opinion, is most important, informative, representative, or new.

In this view, Gould is a man after my own heart. Though digital technology certainly has the potential to provide unprecedented access to all kinds of information, presented in myriad new ways, he doesn’t feel we should rush into it just because it’s there. Digitizing the text is not enough; linking text together is not enough; there must be editorial vision behind the project. His digital utopia, moreover, does not involve databases laid out along the lines of some new-fangled linguistic-theory technological/metaphoric hand-your-scholarship-to-you-on-a-silver-platter super-hypertext. Rather, it is essentially a gigantic electronic bibliography that is also a library, which “you could browse at will, roaming the shelves, following your nose” (25). The words he uses here are not his, but they might as well be. This seems, to me, a sensible approach. An electronic scholarly edition has the potential to be much, much more than simply someone’s monograph (though those are important, too.) If we are talking about making the raw text of literature available—along with relevant background, interpretation, and criticism—it must be freely accessible. I am not talking about cost here; that would naturally vary by project and circumstance. What I mean is that it must be both easy to use and allow the reader to explore at will, using whatever method (idle browsing or specific searches) that reader is most comfortable with. Let me bring whatever ideas, assumptions, and reading strategies I want to the material; give me the information I request without hassling me with bothersome attempts to make the job of interpreting it easier. A case in point: in the electronic edition of “Lapis Lazuli,” there is a picture of the lapis lazuli carving on which the poem was based, with some details enlarged. But what if the details which most interested me were different from the ones which have been handily enlarged for me? I couldn’t get at them. It probably wasn’t possible with the technology of the time, but I would find it far more useful to be able to look at a 360-degree view of the object, zooming in wherever I chose to zoom in, discovering details for myself.

Works Cited

Gould, Warwick. “A Diet of Worms: Descriptive Bibliography and the Structure of Digital Editions: The Example of W. B. Yeats.” Literature and Digital Technology: W. B. Yeats, Virginia Woolf, Mary Shelley, and William Gass. Ed. Karen Schiff. Clemson: Clemson University Digital Press, 2003.

Hockey, Susan. “Creating and Using Electronic Editions.” The Literary Text in the Digital Age. Ed. Richard J. Finneran. Ann Arbor: University of Michigan Press, 1996.

O’Donnell, William H., and Emily A. Thrush. “Designing a Hypertext Edition of a Modern Poem.” The Literary Text in the Digital Age. Ed. Richard J. Finneran. Ann Arbor: University of Michigan Press, 1996.

Shillingsburg, Peter. “Principles for Electronic Editions, Scholarly Archives, and Tutorials.” The Literary Text in the Digital Age. Ed. Richard J. Finneran. Ann Arbor: University of Michigan Press, 1996.