EJournal Volume 6 Number 3 (August 1996)
Taylor and Saarinen ask, "If an electronic text can be published in printed form, is it really electronic?" This issue of EJournal explores that question by presenting a cluster of interrelated hypertexts: the feature essay "Hypertext Notes" by Richard Andersen and two shorter hypertext essays by John December and Charles Ess. These shorter pieces are in one sense comments on the Andersen piece but also develop independent discussions of the theory and practice of writing in hypertext.
Since these pieces are written in hypertext, they cannot be presented in the familiar downloadable-and-printable listserve version of EJournal except as what Stuart Moulthrop calls "a paper shadow of an electronic text" (in this case, a linear electronic shadow of a non-linear electronic text). In addition, following a suggestion from Nancy Kaplan (see below), I have created links from within Andersen's document back to the comments by Ess and December. The result is a web of interpolated and self-referential texts which only make sense in the hypertext medium in which they were born.
This issue is therefore something of a breakthrough for EJournal, as it represents our first real attempt to use the hypertextual capacities of the World Wide Web to do more than link together linear documents which are essentially print-like in nature. (In fact, having explored the Web in search of true hypertext documents, I can report that they are relatively rare in scholarly discourse, and that this issue, while far from unique, represents something of a departure for the entire discourse community.)
This format presents some problems for a journal which was not originally designed for WWWeb distribution. The linear-text version of this issue that is normally e-mailed to subscribers, unlike other issues of this journal, is not a mirror of the hypertext version. It contains only a linear version of the December and Ess pieces (and the editorial note you are reading now) without the rich layering of hypertextual references that makes them what they are. The Andersen piece is so tightly connected with the hypertext medium that it will be completely unavailable except for a first-node teaser. (Readers of the hypertext version, conversely, might be interested in going back to look at the listserve version to see what happens when a hypertext essay is un-hypertexted, an interesting reversal of the normal pattern. The effect is an odd sense of discontinuity that is invisible in hypertext but striking when the text is flattened into linear format.)
All three essays, in their content and by the example of their structure, comment in various ways on the creation of meaning in hypertext. Andersen attempts to exploit both the non-linear form of the medium and its ability to link to multiple documents, weaving a tapestry of quotations and links that challenge the reader to "make what you will of this essay." December, a veteran of hypertext publishing, uses Andersen's experiment as a jumping-off point for a discussion of the characteristics of this new medium and of the importance of good hypertext design in making it accessible to readers. Ess, equally well versed in the philosophical aspects of cyberspace, uses the essay as a jumping-off point for an exploration of how modernist perspectives can be applied to what is normally considered the most postmodern of media.
I will avoid the temptation to dive directly into this subject myself and refer readers who are interested in my comments on the matter to my hypertext essay "Rhetorics of the Web." Rather, I would like to use the rest of this introduction to address some of the many questions about the nature of text that the form of the issue itself opens up. Print has always been a highly intertextual medium, even if its physical form has tended to disguise the intertextuality behind the facade of the solitary Romantic author. But the exercise of editing this issue of EJournal has drawn my attention to a particular set of questions about textuality that are peculiar to the business of massaging other people's words into publishable form--a business common in the academic world but suddenly defamiliarized by the new medium of hypertext.
Question 1: The Status of the Editor
For me, perhaps the most pressing question raised by this experience is that of the role of the author versus the role of the editor. There has always been a collaborative relationship, usually unseen by the reader, between author and editor. The editor may exercise no more control than fixing up a few typos or putting references into a standard format. At other times the editor may make suggestions for extensive rewriting, often relayed from peer reviewers. Occasionally the editor of unusual stamina or unusually pressed for copy may undertake a thorough revision of an article. But always it is understood that the author has the last word, can always accept the "or else" if the push and pull comes down to "accept these revisions or else we won't publish your work." The unseen collaboration never rises to the status of complete co-authorship.
Similarly, the role of commentators is always clearly demarcated. Frequently an editor will solicit comments on a controversial article and publish them either in the same issue or in a succeeding one, usually giving the author an opportunity to respond. The text becomes a chain of texts, but always with original article, comments and responses arranged in linear sequence and clearly marked by authorship. The practice of shipping material by e-mail for a listserve journal such as this one has vastly improved the speed of the process but it is still a quicker version of the linear sequence.
But hypertext presents entirely new possibilities. When she reviewed this piece for publication, Nancy Kaplan made a very interesting suggestion:
I think inviting a few well-known hypertext theorists/critics to "comment" or respond to "hypertext notes" would be far too tame (and anyway this sort of thing has been done before). Perhaps commissioning some hypertextual essays and then providing extensive cross-linking among the whole set (perhaps even some visual blurring of boundaries, renaming whole nodes and links to bring all the texts you receive into an indistinguishable aggregate of nodes and links) would be more interesting, if only because no one "individual text" would be central, the others relegated to "comments on," yet the whole could also be read as an integrated, communal discourse not co-authored in the traditional way, but conjoined by the editor's activities.
When you think about it, the natural mode of hypertext is compilation rather than linear creation, especially as the WWWeb begins to be dominated by sprawling hypertext documents that are chiefly made up of links to other documents, or other lists of links. As Bolter points out, this aspect of hypertext in some ways takes us back to the medieval manuscript with its layers of marginalia that over time found their way into the heart of the text.
And yet I find myself deeply disturbed at the thought of submerging the author's text below a set of other texts as an "indistinguishable aggregate of nodes and links." This isn't the twelfth century. The new technologies have not yet gotten us back to a place where we can be comfortable with texts whose voices gradually become more and more blurred. And perhaps they weren't comfortable with this then either. Possibly writers in a manuscript society simply lacked the means to prevent it except by dire injunction. (See Revelation 22:18-19, in which a curse is levelled on anyone who adds or subtracts from the text of the book--God's copyright notice.)
And finally, I'm uncomfortable with the power imbalance this suggests. This arrangement would give the editor supreme authority to blend voices, an authority always in the past reserved to the author. I remember how thoroughly annoyed I was when I found that one of my works had, in being republished, been encrusted with other texts interpolated into the margins. I was not so much irritated at what had been done to "my" text: I was irritated at the fact that the new text looked as though I, not the editor, had done the interpolations. I am a little embarrassed by my reaction--I flamed the editor of the reprint seriously enough that he has never asked to reprint any of my work since. However, the incident points up the fact that, however much we want to share our work and are flattered when it gets cited, quoted or reproduced as a piece of the "intellectual commons," five hundred years of print has accustomed us to treat our words as extensions of our own identity, not to be messed with by others without our express consent nor to be inserted into others' works without acknowledgement.
To exercise the editor's power to do so just seems to me unwarranted, especially since the editor has that power, not because she necessarily has any claim to wisdom that the author does not, but more or less by accident: the editor, not the author, is the last person to handle the text before sending it off to the typesetter, and therefore has the last word de facto.
Therefore I have compromised. I have added links from Andersen's text to the comments on it, but I have not blended them into an indistinguishable aggregate of nodes and links.
This might be no more than a "papyrocentric" attitude (a particularly felicitous coinage by Stevan Harnad) that I have been unable to shake off. But in another sense, the medium of hypertext seems to invite exactly this sort of compromise. In print, most intertextuality is covert. Citations and acknowledgements pages cannot allow even the most diligent author to credit the myriad of influences on her work. But when hypertext allows many "influences" to be incorporated into a work as discrete chunks of text, literally stored as files on another host, it seems only natural to tag them by author. Even if all of these voices still have other voices embedded in them, if the authors are really no more than Foucaultean gaps through which others speak, at least the top level of intertextuality can be tagged.
Question 2: The Status of the "Publisher"
Another question, not a new one but always it seems being asked in new ways, concerns the status of "publication." The lead article in this issue, "Hypertext Notes," can be said to have been "published" by EJournal on this day on this time. Yet the act of publishing it in this case simply means setting up a link to a set of files. This issue carries many other links to many other sets of files, including some hypertext documents which have been "self-published" on the author's own web site and others which have been "published" by other electronic journals. In a very real sense the act of publishing, like the act of editing referred to above, is becoming more an act of compilation than of "making available to the public" in the usual sense. Every one of these texts could have been "made available to the public" by the author, with no intervention from EJournal. So what is EJournal in its role as publisher actually doing to these texts that adds value that their authors have not already added?
The question is made more perplexing that it needs to be by our traditional expectation that "publishing" has something to do with bridging the gap between the author and the distribution mechanism (from typesetter to bookstore). When there is no such gap, publishing only makes sense if thought of as the type of speech act Austin called a "performative" an utterance that accomplishes something which is not physical but which, like a promise, a marriage, a sentence in a court of law, is nonetheless real. Publishing in this medium collapses together with editing as an assumption of responsibility for the integrity of the work. A speech act does not take place unless certain "felicity conditions" are met: one of the felicity conditions of electronic publishing is that the journal has enough reputation to confer more status on the text that its author could by himself. This reputation is aided by a set of systems for ensuring quality )the peer review system etc.) but ultimately it too does not "really" exist except in the minds of the readers.
Print publication, too, is built on trust, but this trust has more physical correlatives. The expense of paper publication necessitated a set of mechanisms to ensure that the money was not expended on material that would not turn a profit (in commercial publication) or that would not advance human knowledge (in academic publication). With the onset of electronic publication, this filtering function, evolved more or less secondarily as insurance against wasted expense, now becomes the most important value-added service that a "publisher" performs. In short, one of the contributions of the electronic medium to the changing of relationships and functions is the disengagement of process of certifying value from the process of making material available.
Question 3: The Status of the "Text"
And lastly, there is the question of stability. Ted Jennings has articulated EJournal's editorial policy as follows: "If we are to be useful in the evolution of the network culture, EJournal has to be 'dependable' within the traditions of codex reliability." Therefore the journal is archived on a fileserver hedged about with restrictions to make sure that "there will be a place to find what every issue looked like on mailing day." This issue, too, is archived in such a fashion, so that all the internal text will be preserved in the state in which you are reading it today. But in the hypertext version, the "text" sprawls outside the boundaries of the fileserver. It is full of links to other texts that will gradually change as their authors update them, or point to nothing at all as the unstable web makes them obsolete. So the attempt to maintain a stable copy of this text for archival purposes begins to look more and more like an antique practice of Bolter's "late age of print." As a transitional policy to help ejournals earn the trust of people used to print, it has done an important job. However, it goes so against the grain of the WWWeb medium that the entire business of maintaining stable copy is being called seriously into question.
Moreover, Andersen maintains his own copy of "Hypertext Notes"--the "director's cut" if you like--free of the links that I have interpolated and potentially changing as he adds new links and modifies old ones. It is fortunate that the versatility of the medium allows a link from this archived version to the author's developing version and thereby renders meaningless the question of which version is the "right" one. It all depends on one's purpose in reading it.
The degree to which the new media complexify the relations between authorship, editorship, publication and librarianship is no longer surprising. But it always seems as though, just when you think that the major questions have all at least been asked if not answered, a new set of experiences brings you face to face with a whole new set of questions which are outpacing their answers. As Taylor and Saarinen remark, "Our dilemma is that we are living at the moment of transition from print to electronic culture. It is too late for printed books and too early for electronic texts. Along this boundary we must write our work."
Note: This reference list includes only the references cited in this "main node," since references in other nodes are internally cited. References to on-line material are also omitted since these materials are directly hyperlinked from the body of the essay.
This is another case in which the conventions of print must be re-invented as we go along.
Bolter, Jay David. Writing Space: The Computer, Hypertext, and the History of Writing. Fairlawn, N.J.: Erlbaum, 1991.
Taylor, M. and E. Saarinen. "Telewriting." Imagologies: Media Philosophy. New York: Routledge, 1993.