In the spirit of Steve Jones's excellent keynote on his edition of Mary Shelley's The Last Man and Michael LaPlace-Sinatra's thoughtful response at the inaugural RC virtual conference, I want to address some of the issues raised in editing "The Devil's Walk" as a hypertext.
Don Reiman and I designed "The Devil's Walk" hypertext as a kind of experiment in which the work already done for our letterpress edition of The Complete Poetry of Percy Bysshe Shelley could be translated into electronic form. Most of its current features and content are not "native" to electronic form, unlike, say, the features of Steve's Last Man , which were all conceived for publication on the Web. As a kind of textual hybrid, "The Devil's Walk" hypertext invites one to think about textual editions as "information structures" and brings into relief the book itself as a kind of elegant machine for helping us organize and process this information. The latter is no small point, since until fairly recently, the information techniques of the book have been largely transparent to most of us because they have always been a "natural" part of our reading experience. Various postmodern and technologically-driven textual practices, however, are bringing back into focus the materiality of the book as a structure through which information is produced, structured, and distributed.
On-line editing presents an array of alternatives to the book for the production, structure, and distribution of knowledge. Books are undoubtedly still the best machines for traditional modes of reading and will remain so until our current screen technology is transformed: no one (or, at least, no one I've met) wants to read a triple decker novel from cover to cover on a computer screen. But I have turned to on-line editing for some crucial advantages it offers in terms of accessibility, searchability, extensibility, and interactivity. These advantages all stem from the different spatial and temporal properties between letterpress and on-line editions, and I want briefly to consider these two categories of difference before returning to "The Devil's Walk" proper.
Let's consider first what may be called the Temporality of Editions. Letterpress editions exist in what Keats called "slow time": they are printed once and distributed--and only sometimes updated and corrected in subsequent editions if they are profitable enough. The most ambitious of these editions exist not just in "slow," but "monumentalized time"; they are produced to last a generation or two of readers. The answer to the question, "When does an Edition end?" is normally, "When its final volume is published." But when does an on-line edition end? That question is not so readily answered. When the editor dies, perhaps? When she grows bored with it? When no one else wants to continue it? When users stop using it? On-line editions exist in "fast time": they can literally change over night, and not just once, but again and again. They not only can grow through time, but also through a kind of electronic mitosis, they can even split apart into separate forms that each begin an on-line life of its own (see the Mary Shelley short stories that began as contextual pieces for The Last Man and are now being prepared as separate editions by Sheila Minn Hwang.
But perhaps a better phrase than "fast time" to describe on-line editions is "fluid time." They need not always be built for the long haul; unlike letterpress editions, they can even be constructed specifically as ephemera meant to tackle a specific issue for a specific audience at a particular moment (e.g., the Medusa edition I mention below). Moreover, unlike letterpress editions, which begin distribution with the first finished volume, on-line editions can begin as works in progress; indeed, most exist ontologically as "works in progress." Among its other effects (and they are many), this ontology creates an advantage for editorial accuracy: although every editor aims at perfection, inevitably errors creep into complicated editions. Unlike the letterpress monumentalized edition, an on-line edition can be constantly corrected. We expect, for example, that our hypertext of "The Devil's Walk" will be continually improved from corrections and suggestions made by the many scholarly readers who visit Romantic Circles and use its electronic editions.
I'll now turn to the Space of Editions by reformulating a previous question: If the front and back covers of a volume present us with the material boundaries of a book, where does an electronic text end? As we all know, the potential for including contextual material in words, images, and sound in on-line editions is enormous. An on-line text lives in a macrospatial or better yet, hyperspatial environment that radically alters the letterpress boundaries of the textual edition, especially as the on-line edition connects with the larger hypertext that is the World Wide Web as a whole. In this way, an on-line edition can radically change the protocols of reading and complicate our very notion of "Editor," a term whose unitary and proprietary force is opened up to the multiple and collaborative forms of knowledge production of the Web. This is one of the greatest attractions of on-line editing to me, I might add.
In addition to the macrospatial environment of the on-line edition, we might also consider what may be called its microspatial potentials. That is, publishing an edition in letterpress form generally requires one to edit material of sufficient length to fill a good-sized volume. This quantitative threshold, of course, has to do with the economics of how books are produced and distributed. On-line editors now have the opportunity, however, to mount small editions, based on a single short poem or group of poems, for instance. Such editions can be geared toward classroom use, say, two different versions of Keats's "La Belle Dame Sans Merci," or a single fascicle of Emily Dickinson's poetry. Through the microspace of on-line editions, scholars can narrow their focus and, at the same time, expand the limits of what can be edited.
For me, the most distinctive features of on-line edition space are that it is "searchable," "extensible," and "shared": searchable in ways that make the most comprehensive letterpress index look impoverished; extensible in that it can be added to almost infinitely; and "shared" not just in its ability to link out to other resources, but also in its potential for interactivity with its users, creating the possibility for truly dialogic editions, co-produced, in effect, by editor and users. See, for instance, the dialogic edition of Percy Shelley's "Medusa" that Melissa Sites and I have tried to construct as an experiment in dialogical editing. The experimental forms of both the "Medusa" and "Devil's Walk" hypertexts highlight the fact that there is no one "right" way to build an electronic edition (any more than there is to build a letterpress one), that the shaping of edition space can be made responsive to the the purposes of the editors and the needs of the material. It is especially important now, however, before on-line conventions of editing get rigidified to experiment with a multitude of possible forms and formats.
We chose to edit "The Devil's Walk" on-line partly to make accessible a work that can rarely be found in any but complete editions of Shelley's poetry, and one that seemed ripe for further critical attention as well as classroom use. The self-conscious intertextuality of the poem also seemed best addressed in a format in which some of the most important intertexts could be made directly available to readers. Indeed, we were able to make the Southey-Coleridge satire "The Devil's Thoughts" available not just in one, but in several different versions, including a side-by-side comparison on the same screen of two of Coleridge's subsequent versions of the poem.
The use of frames to make such a comparison possible is one advantage of coding the present hypertext in HTML, even though SGML offers unquestionably richer possibilities for rendering complex orders of information available to users through searching. While the debate about the relative virtues of HTML and SGML rages on, HTML was our initial choice for making our hypertext available on-line because most Web browsers cannot presently read SGML at all, and even the most advanced software we could find for making SGML visible on Web browsers (created at the University of Virginia) made for a much less visually rich presentation of the material. Our compromise has been to mount the current hypertext in HTML, to take advantage of the increased functionality in the presentation of the material and to make it immediately available to users of Romantic Circles. Meanwhile, UVa has offered to make its software available to us so that when the SGML version of the "Devil" is ready (and that of The Last Man, incidentally), users will be able to view it in their browsers and take advantage of its more sophisticated search protocols. SGML format also promises to make for better stability across different platforms and over time: it creates, then, the best archival form of the text. For a cogent discussion of the relative merits and problems of using both SGML and HTML on-line, see Lou Burnard's "SGML on the Web: Too Little Too Soon, or Too Much Too Late," which offers the possibility that these problems might be overcome in a new markup format called XML, which like HTML is also a subset of SGML. For more on XML, see "Extensible Markup Language (XML)."
When we first designed the hypertext, there were few good models of what a critically edited poem might look like on-line, and none that did an excellent job of incorporating a full textual apparatus, with collations and annotations, into the text proper. Our first attempt at such an integrated text was serviceable, but clunky: users had to leave the screen containing the text proper to follow links for annotations and collations. Using frames, however, we have recently created a new version of the text and apparatus, in which the main screen is broken into three frames: the largest frame containing the critically edited poem, a vertical frame on its right containing the annotations, and a horizontal frame below both of these containing the collations. We think that this format is a far more efficient way to order and present the relations between text and apparatus, creating a "page" on the screen with more available information than the printed pages of any textual edition we have seen. To date, then, we have created a hypertext edition of "The Devil's Walk" that in terms of the above discussion might be described as an electronic translation of a monumentalized edition, which takes advantage of the contexualizing, searchable, extensible space of on-line editions while offering widespread access to a poem not usually included in collections of Shelley's poetry. But that is not where this particular story ends.
Given that this hypertext can grow over time, what might be added? Well, for one, the keynotes, responses, and discussion of this virtual conference will all find their place in the hypertext, along with other critical discussions of "The Devil's Walk." The contexts section will be expanded to include other related contemporary political satires, beginning with a work of which we have seen no prior critical mention: a 1793 prose satire by Thomas Hastings, The Regal Rambler; or Eccentrical Adventures of Devil in London: with manoeuvres of his ministers towards the close of the eighteenth century. Translated from the Syriack MS. Rabbi Solomon, recently found in the foundation of the Hebrew Synagogue. We also hope to add links to, if not create ourselves, other diabolical works by Shelley that speak in his "lower" style.
Then there are the possibilities that have yet eluded us. If the PRO will permit it, we would like to mount transcriptions of the fascinating official documents that accompanied the one surviving copy of the broadside ballad of "The Devil's Walk" on its way to Lord Sidmouth at the Home Office, who ordered that Shelley be watched closely and his mail interecepted. Similarly, we still hope to get permission from the British Library to mount a photo of the actual MS of the letter version. Permission problems have also so far prevented our mounting the 1830 version of the Southey/Coleridge ballad, with illustrations by Cruikshank (which we, of course, found in the Library of Congress but were not allowed to mount on the Web). The thought of what we might add to the hypertext in our perpetuity as its editors, then, makes its future mode of existence compelling to me. But even more compelling is the thought that hypertext links work in both directions, that "The Devil's Walk" hypertext might itself might be linked from--and thus participate in--on-line texts, editions, student papers, and critical essays yet to be.
--NF, OCTOBER 1997