In "Signs, Symbols and Discourses: A New Direction for Computer-Aided Literary Studies", Mark Olsen argues that "three decades of literary computing have failed to have any substantial impact on the mainstream of literary criticism and scholarship" (Olsen 1993: 311). To be sure, projects involving computer-aided critical editing, authorship attribution, online databases, and electronic concordances, as well as the more conversational forms of computer mediated communication -- e-mail, electronic conferences, and newsgroups -- have been well received, but these advances have been regarded primarily as sophisticated tools that help scholars to continue doing what they had already been doing, only faster and, perhaps, with a greater sense of community. Olsen quite rightly comments on the lack of substantive challenge to the core values of literary studies as a discipline when he notes that humanities computing specialists have "failed to convince their colleagues that computer-aided analysis can help to explain, in any significantly meaningful way, how a text achieves its literary effect" (Olsen 1993: 17). However, writing in 1993, Olsen describes the field before the explosive growth of the World Wide Web. In light of this growth, the more critical object of discussion in literary studies now focusses less on how computing can help the critic to describe a text's literary effect than on the value assigned to scholarship presented in hypertext form.
Repeatedly, the claims of literary studies webs to serious, authoritative scholarship evoke skepticism. George Landow observes that resistance to these claims is often expressed as some variation of the argument that such projects privilege technology over substance: an argument that inverts the aesthete's battle cry and objects to "technology for technology's sake" (Landow 1994: 6). Literary studies webs still possess the status of primarily informal communications, while professional peer-reviewed electronic journals in English studies number only in the handful. The key issue here is that of quality or "value" of information. One computing services expert, speaking about how the exciting growth of the internet has quickened the pace of academic exchange, innocently comments that it is like having a television with a million channels -- "with that many channels, you could see all the television that has ever been done...; you could see anything you wanted, anytime, all the time". As critics often point out, there is something fast food-like about this increased access to and increase of information, something amounting to, to use Ian Lancashire's phrase, a "McDonaldization of knowledge" (Lancashire 1995: 4).
What these analogies intriguingly suggest is just how closely tied technology is to an aesthetics of pastiche. Critical work on hypertext has tended to focus largely on hypertext fiction or on the more general concept of hypertext as an infinitely expandable set of linked documents or lexia, rather than on the suitability of hypertext for publishing scholarly material. In this paper I want to explore the relationship between the discourse of pastiche as a genre and the discourse of critical work on hypertext in order to examine assumptions about hypertextual scholarship. Powerful value associations colour this issue of "suitability", associations that may be grounded as firmly in aesthetics as in a reaction to the technology itself. My argument consists of two sections, the first linking definitions of pastiche to hypertext webs and the second concentrating more specifically on pastiche and hypertext as value-laden communication processes.
The word "pastiche", derived from the Italian for "pie" or "pastry", is defined in the OED as "a jumble", "a medley of various ingredients", "a pot-pourri", or "a design made up of fragments pieced together or copied with modification from an original". The term applies well to electronic projects that combine a variety of media, such as text, graphics, audio clips, and video clips. The term would also seem to apply well to the range of materials -- from course listings and syllabi to pointers to other relevant sites -- found on the text-only versions of many literary studies webs, as well as to electronic articles and fiction whose use of hypertext's referential characteristics compartmentalizes the material into a non-linear collection of linked lexia. Significantly, the term "pastiche" has undergone a progressively more negative valuation, particularly in the ways it has been distinguished from parody, and in the so-called critiques of postmodernism delivered by Lyotard, Baudrillard, and Jameson, as well as in the works of Derrida and Hutcheon. The various related definitions of pastiche fall under three general categories: 1) intentionality; 2) superficiality; and 3) plenitude.
Under the category of intentionality fall those definitions that attribute to pastiche either a duplicitous or deceitful intention and those definitions that attribute to pastiche a lack of any intention whatsoever. More closely linked to forgery, hoax, and plagiarism than any other genre, pastiche is often associated with malicious and false claims to authority and ownership -- as Margaret Rose details, more a base theft than a constructive borrowing (Rose 1993: 75-80). At the same time, Jameson has discussed pastiche's defining characteristic as being a lack of underlying intention; pastiche is "empty", "blank" or "blind" -- "pastiche is, like parody, the imitation of a peculiar or unique style, the wearing of a stylistic mask, speech in a dead language: but it is a neutral practice of such mimicry, without parody's ulterior motive, without the satirical impulse, without laughter" (Jameson 1984: 65). In this way, pastiche might be said to resemble anarchy: it is both challenging to established authority, and/or lacking in directed intention or political agenda.
Indeed, stereotypes fostered by cyberpunk culture have contributed to the popular image of the Net as an anarchic space. The Hacker Hero -- or, perhaps more accurately, the Hacker Anti-Hero -- is a lawless renegade, laying claim to information which does not "belong" to him and playing malicious pranks. He is a paradoxical combination of both maliciousness and lack of directed intent, since his reason for creating computer chaos is simply that he can. Less dramatically, the Hacker embodies contradictions that have their corollary in the bog of legal issues concerning ownership and the Net together with the optimistic touting of the Net's potential for democratization. The complex question of copyright in the electronic age has been comprehensively dealt with in several recent articles, and it is not my wish to rehash it here, only to suggest that its existence is a component aspect of anarchy anxiety. In terms of democratization, perhaps the most oft-cited effect of Internet growth is a tendency towards a levelling of hierarchies that many computing and communications theorists see as profoundly positive. George Landow, noting that "this new technology has the power to reconfigure our assumptions about [authority]", counters such fears put forth by Lyotard and Derrida concerning greater state control through the new technology with the assertion that it offers "anarchy, liberation" (Landow 1994: 25). In a democratic environment of multiply-authored texts, he calls for a participatory criticism.
However, a general, underlying assumption concerning scholarship remains, one that a participatory criticism might not only fail to address but also to which it might also possibly contribute: that there is some type of basic incompatibility between the authority and critical distance of serious scholarship and the anarchic tendencies of the Net in general, and hypertext in particular. Writes Stuart Moulthrop, "some would say [hypertext] invalidates the informing 'master narratives' of modernity, leaving us with a proliferation of incompatible discourses and methods. Such unchecked variation, it has been objected, deprives social critique of a clear agenda" (Moulthrop 1991: 2). The invalidating subversiveness posited as an effect of hypertext is, notice, merely an effect of the proliferation of voices, not a directed intent to challenge hierarchies, or indeed, a directed anything. Again, hypertext comes to be seen as "blank" or "blind" -- in essence, uncritical. In addition, pastiche has been defined in terms of its "representational depthlessness"; theorists such as Baudrillard, Hutcheon, and Hal Foster posit pastiche as a mode of façadism, representative of superficiality and ornamentalism (Rose 1993: 221-6). Pastiche becomes associated with the idea of putting on a stylistic mask -- an empty form, a privileging of style over content or meaning. Web page designers recognize this problem better as the "cool or content" conundrum. The ideal, of course, would merge efficiency and the aesthetically-pleasing, but as professionals in charge of rating web sites point out, this sort of balance is a rare find. Michael van Wetering, Coordinator of Computer Services at the Rotterdam School of Management, notes that one difficulty in getting useful materials onto the Web has been that individuals who possess more advanced Web skills do not also possess an in-depth understanding of the material, except perhaps in the field of computer-mediated communication and some computer science fields (Wetering 1994: 6). The point is perhaps more simply demonstrated by the following anecdote: Sun Microsystems recently ran a contest at Queen's University, offering as first prize a SunSparc Station for the best student home page. The most frequently asked question from contest participants was, "Can I just show off all the neat stuff I know how to do -- like imagemaps, animation, frames, cool back-lit tables -- or does my page actually have to be about something?"
By the same token, the charge of façadism leads to criticism that information presented in the form of hypertext risks separation from its context, so that, like the false front of a building with the rest of the structure stripped away, it leaves us with too few clues as to meaning and modes of interpretation. Not surprisingly, advice in web page construction style guides is aimed, in part, at trying to reduce the problems caused by lack of context. This concern may seem ironic, since the ability of hypertext to refer to additional materials, to provide an unlimited context, which the reader can access immediately, has been regarded as one of the most powerful and positive features of the form. As anyone who has used their browser's "Save As" feature can attest to, however, lack of context can be exactly the problem. Pages may be missing crucial information, such as the author's name, date of initial composition, last revised date, and a complete URL. When downloading an entire series of linked pages, the use of complete paths in the link tags, rather than relative ones, can make the entire set of hypertext links unusable offline. Such pragmatic concerns may seem to suggest a need for more sophisticated browsing software and/or more thoughtful web authors. The need to revert to hierarchical models -- referencing a document in terms of where and to whom it "belongs" -- underscores the fact that with hypertext's promise of limitless context may come an encouragement of what educators and cognitive scientists are starting to call a one-dimensional cognitive process: a superficial type of learning involving an accumulation of small bits and pieces of information, rather than a thorough and clearly related understanding of a specific topic. Context expands infinitely, becoming simply more content. We are entering the age of "the infobyte generation" -- a million channels, but it may be even worse than we thought: it's all "Jeopardy", all the time.
Small wonder, then, that current humanities computing discourse is obsessed with questions of "authority", or "value" of information. While enthusiasts who cite the democratization of the Net and hypertextual reader empowerment comment that individuals will both be increasingly able to find information of value and increasingly able to forge their own sense of what is valuable, both pastiche and hypertext have been criticized as a democracy turned demonic. Baudrillard and Jameson implicitly link pastiche and electronic technologies when they prophecy a pastiche-like sea of details, a fragmented but seamless information soup in which the user will inevitably drown (Baudrillard 1984: 242; Jameson 1984: 64). Comments Don Byrd, "popular culture" -- and he is speaking specifically about the Web as an instance of popular culture -- "has led to the distinction between value and existence breaking down; history reaches its end not with the hoped-for apocalyptic fireworks but with a chilling hub-bub of universal but empty exchange" (Byrd n.d.: 3). This relationship between pastiche as a genre and the Web as hypertexual labyrinth underscores one of the most interesting linkages between the theoretical discourses surrounding the two, that of a simultaneous superficiality and abundance: in short, the classic fetish, offering a seemingly infinite substitute for a loss or lack, a substitution endlessly deferred along a chain of metonymic signifiers, illusory referential plenitude with a vengeance. One experimental web site plays on this relationship by presenting an introductory screen explaining that the site uses a new interface, Fetish, and providing a series of links that purport to point to sites organized into familiar subject headings, such as gender studies, postcolonialism, and so on. When the user follows these links, however, they branch out along a string of sites successively offering seemingly unrelated images. Clearly transgressing the boundary between criticism and art, sites such as these both exacerbate the problems that result in scholarly disdain and suggest that a transformative force is at work making over conceptual categories that the humanities hold dear.
Wolfgang Karrer's theory of pastiche as communication process has not been readily taken up by parody theorists, but it is one that has tremendous significance for the types of interactions that occur on the Web specifically, and for the concept of hypertext in general. Karrer posits parody, pastiche and travesty as communications processes that have "a retroactive effect upon social factors and conditions", by which he means that these forms are interactive processes affecting and being affected by their producers and receivers (Karrer 1986: 5). As critics like George Landow and Paul Delany have pointed out, hypertext is increasing our sense of the text as a process, rather than a product (Delany &Landow 1993: 12). Theoretically, each reader could experience a completely different version of a text -- in reality, completely different text -- than every other reader, and it is the process of making choices throughout the reading that determines the shape the text will assume. Further, if the reader leaves traces of her reading, in the form of annotations and additions, this will in turn affect the shape of the text for future readers. This emphasis on process rather than product tends to militate against the assignation of value or authority, since there is no stable, or final version, of the text on which to base a value judgement -- "value" comes to reside in the individual experience of each reader.
Margaret Rose, one of the only parody theorists to fully examine Karrer's conception of pastiche as a communication process that behaves retroactively, extends the meaning of the term "retroactive" for pastiche when she posits it as a form in which "all styles are dialectically available in the interplay between the Now and the Not Now"; pastiche is profoundly nostalgic, "preserving and presenting past forms for the pasticheur's own age" (Rose 1993: 73). This idea of nostalgia reveals a fascinatingly contradictory assumption in much humanities computing discourse, an assumption particularly apparent in discussions of the electronic publishing of scholarly journals.
Hypertext is currently fulfilling the role of a transitionary communication, preserving aspects of print culture in the digital age. In "Electronic Journals and Legitimate Media in the Systems of Scholarly Communication", Rob Kling and Lisa Covi provide a concrete example (Kling &Covi 1996). They begin by commenting on the very practical issues of peer-review of electronic journals, the reluctance of senior and well-known scholars to contribute, and the problems with integrating these journals into library acquisitions, but their example of one journal that has broken through the barrier of "electronic illegitimacy" is very revealing in that the journal's success depends largely on its "look". The Journal of Artificial Intelligence Research offers both a print version and an electronic version of each issue, and places stringent limitations on the electronic version -- authors may not include graphics beyond the charts and diagrams they would offer in the print version, citations must be in conventional format, pointers to additional material are not allowed, margins are set, and the work must appear as one complete (non-segmented) document. Note Kling and Covi, "[w]e think of JAIR as the Stealth E-journal of Artificial Intelligence Research. Its editors cleverly exploit the broad rapid international distribution afforded by Internet services such as WWW, while simultaneously calming authors' fears of publishing in a stigmatized electronic medium because it always looks like a p[rint]-journal" (Kling & Covi 1996: 4.6).
Notice that we have now moved from the "cool or content" conundrum -- style versus content -- to the somewhat paradoxical pragmatics of the "look" actually guaranteeing the authority of the "content"; why? "The ultimate goal of computer technology is to make the medium transparent, make the computer to all intents and purposes, invisible to the user", remarks Greg Ulmer, illustrating his point with the analogy of "the suture" in cinema. The "suture" -- the means by which the medium is made "invisible", transparent, or naturalized to the viewer -- can be interrupted with special effects approaching montage and "mise en scène", so that the film displays its own making and makes the viewer aware of the medium (Ulmer 1991: 1). Tellingly, it is those cinematic techniques approaching pastiche (mise en scène, montage, collage) which effect this interruption. A similar effect occurs with hypertext, as the often still-clumsy, pastiche-like integration of media in many hypertext systems makes us hyper-aware of the media and of their mixing. We are not yet at the stage where the computer is a transparent medium, and it is this problem of transparency that scholars most frequently cite. As one electronic discussion list participant stated during a recent thread concerning the electronic book, the "extras" such as graphics, audio, and video distract her from her goal, a text document in which she hopes to find information relevant to the topic she is researching (Osted 1996). The less aware she is of the medium, the more comfortable she is with the status or scholarly value of the information.
I began this paper by asking why, using Mark Olsen's words, "humanities computing specialists have [to date] failed to convince their colleagues that computer-aided analysis can help to explain ...how a text achieves its literary effect" (Olsen 1993: 309). In teasing out some of the implications of an aesthetics of pastiche as applied to hypertext, I have thus far emphasized this failure to convince in terms of the status of hypertextual criticism in the discipline. Let me now briefly address the question of hypertext and what we, as literary critics, actually do with texts when we analyze them.
In the hypertext environment, the blurring of distinctions renders value problematic, which has its impact on the practice of literary criticism. To summarize, look and information, active and passive, creative and critical merge into an indistinguishable mass. The dissolution of the opposition between style or look and content in hypertext is fascinating terrain, one which is just beginning to be mapped. Hypertext also blurs the distinction between passive and active where the notion of hypertext as an interactive form shades into the notion of hypertext as implying a certain passivity or laziness in encouraging a superficial acquisition of information. And we have a similar blurring where projects like the Fetishturgy site clearly transgress the boundary between criticism and art. Particularly in its function of preserving aspects of print culture, hypertext and the Web come to subvert even the distinction that most postmodern theorists make between parody and pastiche. Parody's intentional re-working of material is defined in opposition to pastiche's "blind" re-presentation of material, but on the Web, it is no longer easy to tell the difference between "dumb" imitation and intended critical recontextualization. In other words, what the inability to tell the difference between parody and pastiche signals hypertext's emphasis, to use Don Byrd's words, on existence rather than value. Thus, the current debate concerning improving the status of electronic critical materials not only arises most contentiously from the threatened dissolution of the idea of value as group standard, but also precisely suggests how the kind of activity we engage in as literary critics may change.
In "The Rationale of Hypertext", Jerome McGann puts forth a compelling argument for the use of hypertext in scholarly editing: "Hypermedia editions that incorporate audial and/or visual elements are preferable since literary works are themselves always more or less elaborate multimedia forms" (McGann 1995: 18). He offers such examples as Blake and Dickinson. He concludes by discussing the Rossetti Hypermedia Archive, but it is his insistence on the project as an archive, with his careful distinction between "archive" and "edition", that most intrigues me. Why this careful distinction, when a hypertext edition could, theoretically, take in the same materials the archive does, though probably presenting them in a slightly different way? Explains McGann,
When a book is produced it literally closes its covers on itself. If its work is continued, a new edition, or other related books, have to be (similarly) produced. A work like the Rossetti Hypermedia Archive has escaped that bibliographical limitation. It has been built so that its contents and its webwork of relations (both internal and external) can be indefinitely expanded and developed. (McGann 1995: 61)
Certainly, a blurring of editor and literary critic is implied both in the project McGann describes and in the document in which he describes it -- this particular terrain of debate is not new. Yet I think that his careful separation does not adequately address the problematics of a blurring between the concept of an edition and an archive that a hypertext edition unavoidably raises; or, if it does, it leaves unspoken the sense of a limit to the use of hypertext that an edition should -- but not, theoretically, could -- make. Hypertext may also come to blur the distinction not just between author/critic and reader, but also between the editor/literary critic and the archivist. The emphasis is on existence again, rather than value, or, by extension, effect. Will how a text achieves its "literary effect" still be a question it makes sense to ask when the computer has been naturalized as a medium?
Please remember to put your answer in the form of a question.
 See David Wilson's "Internet Quickens Pace of Academic Exchange and Collaboration", in which he quotes from a verbal interview with Michael Joyce, professor of English and humanities computing specialist at Vassar College (Wilson 1992: 7). The citation here is to the paragraph number; all citations of web pages throughout this paper will be documented in this manner.
 For a concise and cogent account of this issue, see Janet Fisher's "Copyright: The Glue of the System" (Fisher 1993).
 Contest details, the FAQ, and announcement of the winning entry and its URL, can be found at http://www.ccs.queensu. ca/ccs/contest.html.
 Education professor John Eklund points out that hypertext has been called by cognitive scientists a "non-pedagogical technology [because of its]...static, non-adaptive [nature, and allowance for]...minimal possession of structural knowledge of content" (Eklund 1995: 2).
 Fetishturgy. Some of the pages offer material easily related to the subject headings, but many do not. Initiated and maintained by Greg Ulmer, the site is an experimental collective authoring project. Author name(s) and posting date(s) often do not appear on the pages, providing an example of one of those cases where this lack of contextual information may be a deliberate component of the site's parodic strategy on the part of some authors.