Trans

Internet-Zeitschrift für Kulturwissenschaften

9. Nr.

Oktober 2000

Texts without boundaries^*

Gerhard van der Linde (Pretoria)

1. In the following, I present an overview of various types of electronic texts, divided into two broad classes. The division offers a convenient framework for discussion, but it is only a first attempt at systematic classification. The rate at which electronic texts and formats proliferate makes it very unlikely that an exhaustive and generally acceptable taxonomy will soon be achieved.

In the second part of the paper, I suggest how the electronic book, following the direction indicated by certain kinds of electronic text, could eventually realise the concept of the boundaryless text in a networked information environment.

2. In general terms, I define an electronic text as any text consisting of words only, or using a combination of words, graphics and sound, which is created by using specialised software, and which is read by means of a personal computer or other digital device. Within this broad definition, one can distinguish between texts which can be read both electronically and in paper format without affecting their integrity, even though certain functions may only be available electronically, and those which can only be read properly in electronic format.

3. The first class of texts includes purely verbal texts published on CD-ROM and via the Internet, which only offer the level of interactivity necessary for reading, printing and finding information in the text. Because of the large amounts of data which can be stored electronically, the format is especially suitable for providing access to collections of texts, such as the collected works of a certain author or texts from a specific period. Networking the texts via a local area network or the Internet provides remote access to simultaneous readers. One should distinguish between texts distributed in this manner by commercial publishers, which are accessible on a subscription basis, and non-commercial ventures. Examples of the latter include Project Gutenberg, which involve the digitisation and publication on the Internet of culturally significant texts in the public domain, and authors’ self-publication of their work. In principle, anybody with access to a computer and a knowledge of hypertext authoring can make his or her work available to a potentially worldwide audience. Sabrina Ortmann (1998) notes that this is an especially attractive option for aspiring young authors whose work would not easily be accepted by commercial publishers.

Literary texts which belong to this class and which merely use the Internet as a distribution channel can be termed "Literatur im Netz", "literature on the Internet", as opposed to "Netzliteratur", "Internet literature" (Simanowski 1999). These texts are not inherently different from traditional printed texts. Their style and structure are not determined by the nature of the Internet or the authoring language used for the World Wide Web. Roberto Simanowski (1999) describes this kind of text as "‘unechte’ digitale Literatur". They are read in basically the same way one reads traditional printed books.

4. The other categories of texts to be discussed all belong to the second class. All are fully accessible only in electronic format, though some of their components could be printed out and read in hard copy.

4.1 The first category of texts in this class includes CD-ROMs which simply combine words, graphics and sound. An example is the multimedia CD-ROM of Dante’s Divina Commedia. The reader can listen to a voice reading the poem while following the printed words and viewing illustrative graphic material. As in the previous class, all elements of the text are pre-determined, and there is only a minimal level of interactivity. Simultaneous use of an audio and a printed version of the work enables the reader immediately to correlate the poem with the sound world it projects. The co-existence of these parallel versions in a single digital environment does not produce a completely new kind of text. It simply offers a technology-enabled combination of versions which previously could only be used separately.

4.2 A second category includes literary texts, written in hypertext by a single author, which only make use of verbal signs, and in which the hyperlinks and their destinations are pre-determined. The reader can follow the links in any available sequence, but the underlying text is fixed. This kind of text originated in the mid-Eighties, before the advent of the Internet, as an attempt at exploring the literary potential of hypertext (Ortmann, 2000). Early examples were only available on disk, but examples can now also be found on the Internet. These include the hyperfictions by Adrienne Eisen. Greg Maier (cit in Ortmann, 1998) designates this category as "classical hyperfiction."

4.3 A third category includes texts, written in hypertext, which use the multimedia format to challenge the boundaries inherent to traditional print-based texts. Verbal signs, graphics and sound interact to generate a complex reading experience. Multimedia literary texts are related to the comic or graphic novel, to the extent that the story is told by means of graphic images. However, unlike in the graphic novel, the verbal text is not necessarily secondary and simplistic. The graphic material could be merely illustrative; it could add layers of meaning to the text; it could supplement gaps in the text by providing additional information; it could add narrative strands to the storyline; certain sections of the text could be narrated only through graphic images; and so on, through as many permutations as the author’s creativity and the constraints imposed by the available technology will allow. Similarly, different kinds of sounds can be integrated into the narrative.

The integration and interaction between text, images and sounds in the multimedia hypernovel enable a non-linear, multi-sensory reading experience. Multimedia hyperfiction published on the Internet sometimes makes considerable use of interactivity. As Sabrina Ortmann (1998) indicates, the reader can be invited to solve problems, to participate in discussion forums, to leave messages on a billboard, and to e-mail feedback to the author. Greg Maier (cit in Ortmann, 1998) designates this category as "modern hyperfiction."

Although this kind of text allows a significant level of interactivity, the author controls the ways in which interaction can take place, and the primary text is pre-determined. As with the traditional printed text, the author occupies a position of authority, dictating the choices available to the reader.

4.4 A fourth category, which offers a high level of interactivity is the so-called interactive novel, which came on the scene during the Eighties, preceding hyperfiction. Jerz (2000b) defines interactive fiction as "computer-mediated narrative, resembling a fine-grained ‘Choose Your Own Adventure’ story". Elsewhere, interactive fiction is described as presenting a "text-only interface which allows the reader to type in short phrases as navigation or action commands [...] All generally presume that the reader will assume the persona of the protagonist" (Keep and McLaughlin, 1995). According to Jerz (2000b),

"the classic IF interface is a command-based textual feedback loop: the computer displays a few lines or paragraphs of text: the interactor types a command; the computer describes what happens next, and then waits for additional input."

The so-called "interactor", who replaces the traditional reader, is required to "write part of the narrative, and the computer ‘writes’ back in response to real-time user interaction". While in tree fiction, a sub-category of classical hyperfiction. the reader is offered simple choices and taken to a pre-determined point in the narrative depending on the choice he prefers, interactive fiction "requires the user to intuit, deduce, or otherwise stumble upon alternative actions" (Jerz, 2000a).

There is a certain resemblance between interactive fiction and computer games in which the player as protagonist has to navigate a number of obstacles or solve certain problems, thereby generating a virtual adventure, playing out variable scenarios.

One of the most significant innovations of this form is that the distinction, theorised by Barthes and others, between a pre-determined "work" and an indeterminate "text" becomes redundant, as only an indeterminate, or at least a variable text remains (Ziegfeld, 1989:364). The notion of the reader as co-producer of the text is concretised in the interactive novel.

Interactive fictions tend to have a clearly defined threefold structure: "a prologue, a middle game and an end game..." (Nelson, cit in Traenkner, 2000). Despite its open-endedness and the active participation of the interactor-reader in generating the narrative, despite its variability, which does away with the traditionally inviolable status of the text, interactive fiction satisfies the need for a definite sense of closure. By way of comparison, in the traditional printed detective novel, the reader has the illusion that he participates in finding a solution; yet in reality, there is a single correct path to the ending, and it is carefully designed and often camouflaged by the author. In the interactive novel, the paths to the ending can be variable and indeterminate, and the reader actively participates in generating it. The ending itself can also be variable and indeterminate. By contrast to the traditional printed detective novel, the ending is at least to some extent determined by the reader. Yet both the traditional printed detective novel and the interactive novel invite the reader to navigate a route in order to arrive at an ending which provides a sense of closure. In both forms, most of the interest is generated by the journey towards the ending, but it is the ending which justifies what goes before.

4.5 A fifth category of pure digital literary forms is what Ortmann (1998) calls "adventure hyperfiction". It is related both to interactive fiction and to computer-based adventure games in that the reader acts as protagonist. The reader-protagonist enters a virtual scenario in which he has to solve problems, overcome obstacles and so on. The interest revolves around the reader’s participation in virtual scenarios.

4.6 The sixth category, collaborative hyperfiction, can be divided into at least two sub-categories.

4.6.1 The first sub-category includes projects in which a number of authors collaborate to develop a theme or set of themes or an initial narrative unit. In the Wörterwald project, contributions are invited on various themes specified by the developers. The Internet novel Beim Bäcker was initiated in July 1996 by Carola Heine. One of the participating authors wrote a sequel to her initial story, which was taken up by one of the others, and so on.

Each sequel is both a continuation of and a response to the preceding. The idea was to develop a linear narrative. According to Roberto Simanowski (2000), the group dynamics reflected by authors’ responses is particularly interesting, perhaps more so than the end product of the project, which was terminated after about four years.

4.6.2 The second sub-category includes the true Netzroman, a term which I borrow from the site of one of the examples of this genre, Die Säulen von Llacaan. The term refers both to the fact that the text resides in cyberspace, and to the concept of the Internet novel as an open network of multiple voices and stories.

Usually, an author or authors propose a framework or a text which acts as a starting point. Readers navigate the existing text by following hyperlinks and add further text according to the rules of the game specified by the author or authors. In some cases, contributions are filtered by the authors, while in others, contributions are directly added to the existing text.

    4.6.2.1 An example of this type of collaborative hyperfiction is the Cybernovel found at http://www.nodeadtrees.com. The developers state that

"for a novel to be interactive, it must have every possible beginning, middle and ending. Characters must be able to move in random directions with different outcomes and consequences as a result of given actions and reactions."

Their aim is to realise a novel which "must be able to move in any direction at any given time." The process used to accomplish this is quite different from the principles underlying classic interactive fiction. In classic interactive fiction, the reader reacts to prompts by the computer, which provides feedback to his reaction, setting off a further response, and so on. The resultant storyline can be quite indeterminate and non-repeatable. The process is linear and sequential, but the links in the sequence are not pre-determined. In the case of the Cybernovel, each of the authors involved

"will write scenarios, sketches or stories about the characters, and these will be linked together so that different links take characters in different directions. Readers will follow the links they choose, thus creating a different novel each time it’s read."

The reader can enter the story at any point and explore the available links in any sequence. Alternatively, the reader can click on one of the characters and follow the links leading from there in any sequence. In addition, the reader is invited to become a co-author: "The only way this will work is if you submit your work with these characters. The people are hanging ... directionless in Cyberspace. Take them, mold them, move them. They are yours."

The Cybernovel is doubly unstable and variable. Firstly, in that it does not exist as a novel outside the reading process. It is quite literally constituted through the reading process, and each reader can generate a different novel. All the potential readings are embedded in the surface presented on screen, thus creating a multidimensional network of possible novels. Secondly, the participation of numerous writers, adding further links, continuously increases the options available to the reader, and thus, the alternative novels which can be generated. This subverts the myth of the perfectionist author who endlessly rewrites page after page to arrive at a text in which not a word or a comma is out of place. There, the presumedly perfect text to be realised creates a movement towards finality: it is the sacrosanct object which drives and justifies the writing process. The author controls the process from beginning to end. In the case of the Cybernovel, authorial control is restricted to specifying the rules of the game and setting up links. There does not seem to be any quality control or editing. Instead of being driven by the desire to construct a perfect target object, the writing process is its own justification, an end in itself, with no finality.

    4.6.2.2 A different concept is represented by the collaborative hypernovel Amnesia. According to the authors, "The idea stems from the belief that the only problem with exciting novels as they are now is that the protagonist pulls you along kicking and screaming whether you like his decisions or not." They "hope to create an exciting novel where the readers control their own fate." Accordingly, readers are invited to "read Chapter 1 and vote what you’d like to do next". On the basis of certain clues in the first chapter, readers guess what will happen next. Since they are, in the authors’ words, "without the luxury of a world-class detective supporting character who manages to solve the mystery", and since there is no pre-determined solution, readers’ indications as to what they think will happen also suggest what they want to happen. The authors allow all further sequences to be determined by readers’ preferences. There is even a "discussion area", where readers can exchange ideas about the story and the characters. The authors assure readers that they also monitor the discussion area, "pulling out popular ideas and basing courses of action on these". Interestingly, and perhaps significantly, Amnesia is described as "an idea born in an Academy for the Advancement of Science and Technology classroom". It uses technology to generate new kinds of literary forms, but has no serious artistic pretensions. The authors set out to create an alternative literary form, but the initiative is primarily technology-driven.

Although it is written in hypertext and lives in cyberspace, Amnesia seems closer to classic interactive fiction than to contemporary hyperfiction. Instead of offering "a network of multidirectional links among the various pieces of a text" (Mihajlovie, 1998:214), which enables randomly variable readings, Amnesia makes use of the feedback loop principle: readers are invited to react to certain narrative material, producing a series of randomly variable sequences, with their reactions being used as basis for the next narrative unit, to which they are invited to react, and so on. In the case of Amnesia, the responses to the reader’s feedback are produced through human intervention, instead of by a computer programme, while feedback is provided by any number of readers. Amnesia combines collaborative writing with interactivity, but makes limited use of the possibilities offered by hypertextuality. Despite the democratisation of the production process, the authors remain responsible for the eventual text.

    4.6.2.3 Another example is Die Säulen von Llacaan, a project initiated by Roger Nelke. Three initial stories are provided. The reader can either add to the initial stories or to any of the continuations proposed by other readers. The reader can also link his or her contribution to any other story, which other readers can then continue in the same way. The result is an open-ended network of stories. The correlation to this is a open-ended network of possible reading experiences.

    4.6.2.4 The sister project to the Säulen is Die Perlen von Cala-Elen. The rules of the game specify that the reader can select any of the existing units of text, write a continuation and send it to the coordinators, indicating which unit it continues, or by which unit it should be continued. If the reader does not specify this, it is regarded as an open-ended text and any other reader can continue it as he or she chooses. The readers can also participate in the search for the pearls by interrogating the character Orca and reflecting the knowledge gained by this in his or her contribution. To help them, readers can click through to information on the fictive world Caala-Elen, the legend of the pearls, and the prehistory of Caala-Elen. Four parallel initial stories are offered as entry points.

    4.6.2.5 The third novel in the series, Magische Welt Íja Macár works on the same principles. According to the rules of the game, the reader should choose an existing chapter or beginning, write an alternative or parallel narrative, even if it already exists for that particular unit of text, and send it to the coordinator. Apart from a number of practical instructions, readers are instructed to avoid proposing actions or events which would have too large an impact on the fictive land Íja Macár and its inhabitants.

    4.6.2.6 On the surface, in the second sub-category of collaborative hypernovels, the author as locus of control is either abolished or substantially diminished. However, in all the cases I have discussed, what I would call the primary author or authors remain responsible for Web design, layout, and creating links; for determining the rules of the game; and sometimes also for deciding how readers’ contributions should be incorporated.

4.7 Reading the hypernovel can be a random and non-linear process, and many are designed to be read in this manner. That in itself is not radically new. A print-based novel such as Cortázar’s Rayuela is also designed to be read like that. Even traditional printed novels can be read in a random fashion, if the reader so wishes. However, on the whole, hypernovels present a number of conceptual innovations:

1. The blend of technical expertise and traditional writerly skills required to create a hypernovel.
2. The blurring of the boundaries between reader and writer. The reader can become a writer, turning the primary author into a reader. Communication between reader and writer can be instantaneous.
3. Especially in modern and collaborative hyperfiction, there is no definite, final text; each reading can literally generate a different text. Additions and revisions can be made continuously.
4. The boundaries between readers are abolished. Rather than a solitary dialogue with the text, reading becomes a communal exercise. Readers can interact in real time by using the discussion areas linked to some hypernovels.
5. Similarly, instead of a struggle between the author and the blank page, writing becomes a collaborative process, resembling a game played out according to a set of rules specified at the outset.

4.8 Although I am not aware that the concept has yet been used for literary purposes, I would like to outline the emedicine project, since it illustrates how the boundaries of the electronic text can be eroded so as to integrate it into a networked information environment.

In this project, thousands of authors collaborate in the production of medical textbooks in cyberspace. Anybody, anywhere can access the work in progress and comment on it. Updated information and changes derived from readers’ comments can be added continuously. Terms in the text are hotlinked, leading the reader to definitions and even to the most recent literature on a topic. From there, the system automatically searches relevant material in predesignated Web sources. Readers can get seamless access from the text to abstracts and journal articles, enabling them to explore topics in depth. They can be notified by e-mail of the latest revisions and of references relevant to the topics they specified.

4.9 From this, it can be seen that the hypernovels I have discussed above do not yet fully exploit the possibilities offered by hypertextuality on the Internet. For instance, any keyword in a hypernovel could be linked to other texts using the same, or similar topics, or to critical analyses of the topic or of hypernovels using the topic. This would enable parallel comparative readings of different hypernovels. It is even possible to generate new texts by combining links to elements from different hypernovels. For example, it would be possible to link characters and events in different hypernovels and to enable the reader to explore these in a random fashion. Each set of permutations would in effect produce a different text. It is technically possible to incorporate sound and graphic material as well. It would even be possible to write real life events and persons into the novels by linking them to Webcams and creating appropriate contexts. Alternatively, the reader could be invited to write units of text inspired by Webcam images; the units could be linked to create hypernovels. Already, the site maintained by the artist Ana Voog presents a continuous stream of random images of the artist’s private life, linked to her own comments, a discussion forum, comments on the site which appeared in the media, and images of her artworks. The whole could be read as a multi-layered text. Ana insists that the actions shown on screen are completely spontaneous, and that the images are not edited at all, yet points out that we only see a picture of her life, not the life itself. It seems almost impossible to distinguish between the "real" Ana and the media construct she has become. There are other examples of this kind of site, but this is really a field of study on its own.

5. Finally, I would like to discuss the so-called electronic book. In my opinion, it could open the way towards a truly boundaryless information environment.

5.1 As the format has not yet been standardised, there is no single exhaustive definition of the e-book. One publisher ("Definition", 1999) describes it as "a collection of pages (files) accessed by computer with no discernible sequence in which the pages are to be read." The information in the e-book is "presented...in files that are connected (linked) to one another. To read the book ... one merely selects various links using point and click actions. And to help readers decide which links to select, we (the publisher) offer a set of questions which we refer to as ‘motivators’." According to this model, the important point is that "electronic books are written in such a way that there is no particular sequence in which one has to read the book. No matter where one begins one will eventually cover all of the material. But, this is accomplished in an order and sequence that is determined by the reader, not by the writer." Another publisher states that there "is no standard way to read an e-book. Start at the middle, move backwards, forwards, or in random order. Bookmarks aren’t necessary because any section of an e-book can be found and re-visited" An electronic book can be "multimedia, interactive, web-connected, etc" ("New release", 1999).

These descriptions do not really help to define the electronic book as a distinctive phenomenon. Described in these terms, it seems quite close to the modern hypernovel. The descriptions also relate more to the structure of the contents than to the e-book as object and mechanism.

Since it is called an electronic book, one would expect this object to resemble or somehow to behave like a printed book. In fact, the term "electronic book" is commonly used to refer to a specialised device used for accessing and reading electronic texts. In this sense, electronic books are defined as "basically high-tech reading tablets that hold the equivalent of thousands of paper pages." The reader downloads text into the device, which displays it "on built-in screens" (Robb, 1998).

The electronic book was pioneered in 1998 by the Rocket eBook from Nuovo Media and the SoftBook from SoftBook Press. The SoftBook can "store up to 100,000 pages of text, pictures and graphics, which you can read, highlight, bookmark, and annotate." To download text, the SoftBook is plugged into a telephone jack, without using a personal computer, the reader buys the desired book online, and it is then delivered into the SoftBook. By contrast, for the Rocket eBook, text has to be downloaded to a PC, from which it is transferred into the reader. The Rocket eBook "holds up to 4000 pages of text and graphics. It also holds a dictionary and supports adjustable fonts, multiple bookmarks, and encryption." Books for the Rocket eBook can be downloaded from a Barnes and Noble Web site, where more or less one million titles are available for electronic distribution.

An alternative device is the EB Dedicated Reader from Everybook. The producers describe access and downloading as follows: "You attach your EB to any touch tone phone line anywhere in the world to connect to the Everybook Store. A three-dimensional virtual reality bookstore, personalised to your choices, appears on the full-color, high resolution screens. You use the full-page touch screens to wander through the bookcases, pull books off of shelves, and look at covers, tables of contents, and excerpts. You then touch the icon for purchase information, and put the publication in your shopping cart." After paying for the books online, "you can choose to download them now, or let the EB dial out in the middle of the night for a remote download to your EB." According to the publishers, "Only the EB is a true electronic book. All other reading devices on the market are electronic tablets." Thus, "Everybook uses existing print files from publishers, and displays them exactly as they would appear in print."

Another publisher, Librius.com focusses on offering content which can be read on a variety of devices. Texts are downloaded to a PC from the Librius World Bookstore, and from there, to the reading device.

5.2 In these terms, the electronic book seems little more than a distribution channel for delivering text alternative to traditional printed books. Client acceptance probably depends on the extent to which the e-book can equal the convenience and flexibility of printed books. Accordingly, publishers of e-books seem to strive for producing devices which mirror the positive attributes of printed books as closely as possible. However, it has been pointed out that "electronic books should have many more uses than simply storing electronic versions of paper tomes." The Boeing Company, for instance, has indicated the need for "a portable electronic viewing device for accessing shared reference libraries consisting of books, code libraries, programming tips, hyperlinked standards, magazine articles, Web pages, and newsgroup postings." Ideally, it should also be possible on the same device to read, compose and send e-mail, to write and edit documents, and to browse Web pages. Such a device "would combine the best of a laptop, a PDA, and a mobile phone." It should allow the reader to "access shared reference libraries from anywhere in the world via land-bound wireless technology or a low-earth satellite cluster" (Cline, 1998).

Instead of being merely a digitised version of a printed text, a literary text residing in an e-book could thus be integrated into an information network or informatised environment. From the downloaded text of a novel, seamless access could be offered to reference sources such as encyclopaedias and dictionaries residing in the e-book, to relevant articles downloaded from fulltext databases of electronic journals, and to downloaded Web pages. With the addition of connectivity, it would be possible to switch between reading an electronic novel and surfing the Web, in search of information supplementing or elucidating aspects of the text one is reading. Obviously, the various forms of hyperfiction could also be integrated in this manner. A critic would be able to have all the texts required for producing an article on the e-book, to switch between writing and editing the article and viewing relevant texts stored in the device itself or accessed on the Web, and even to have e-mail or online discussions with other readers on the topic he is studying, while working on the article. Writers could interact with online communities of readers, perhaps even submitting parts of their work in progress for comment. The possibilities are legion.

5.3 At this stage, one can only speculate on the impact such developments would have on literary production, reading and the literary marketplace. One possibility is that, in many cases, the "standalone" text will be replaced by carefully designed electronic packages. Instead of merely downloading the text of a novel, the reader will buy the package. In addition to the primary text, a novel for instance, this might contain (links to) reviews, possibly written before publication, the option to correspond with the author by e-mail, to access a discussion forum on the novel, to buy and download other books by the same author or on related topics from an Internet bookshop, and so on. In terms of the Everybook model, the reader would be able to browse available texts in a virtual bookshop, probably in three-dimensional virtual reality. The whole package might be available for downloading at the author or the publisher’s Web site, or at a site dedicated to the primary text. Such an electronic package would probably not be produced by a single author, but by the author of the primary text in collaboration with an editorial team, which would include a Web designer. The electronic package might be designed as a small, customisable virtual library. It would be updated continuously by adding reviews, hyperlinks, and revisions to the primary text; obviously, online discussions of the primary text will change all the time. Components of this virtual library would be discarded or replaced from time to time, according to the reader’s changing needs.

6. There is little evidence to suggest that either hyperfiction or the e-book and its successors will bring about the often heralded "death of the book" as we know it. However, it seems certain that technology will soon realise book-like media which offer capabilities far beyond those of traditional printed books, without sacrificing convenience and ease of use.

Researchers at the Massachusetts Institute of Technology Media Laboratory have envisaged that technology capable of storing massive amounts of data will be available in the near future. In the medium term, technology will become commercially available which can store up to 35 gigabytes, "already more books than the average individual reads in a lifetime" (Jacobson et al, 1997). A drive has already been demonstrated "with an information density capable of holding 10 terabytes ... Such a capacity is able to hold (using only modest compression) the entire United States Library of Congress," that is, more than 20-million volumes. The memory capacity of portable computers and of specialised reading devices will probably increase quite dramatically. The speed at which data can be delivered via the Internet will increase just as dramatically.

With this kind of technology on the horizon, the electronic package described above will become feasible sooner rather than later. It has been estimated that a device capable of delivering such a package will become commercially available to the general public within the next ten years (Makimoto and Manners, 1997:93). Once it has been introduced, it is likely that there will be a demand for customisation. Instead of merely accepting what is on offer, readers will probably demand that the package be structured around the primary text according to their needs and preferences, which may vary from time to time. The question of which texts to offer in this format will be determined by the demands of the marketplace. This could lead to the marginalisation in cyberspace of texts usually deemed worthy of academic literary study.

7. The more advanced categories of hyperfiction and the electronic book suggest modes of interaction between text and reader and between author and public which differ fundamentally from those dictated by traditional print-based texts. Information technology and advances in telecommunications, which seem to drive this field, rather than literature or literary studies, develop at a rate which would be unthinkable in academic literary studies. Developing a body of theory which would offer an informed response to electronic literary forms therefore is a considerable challenge. This is made even more problematic by the changeable and sometimes ephemeral nature of Internet literature, and by the traditionally adversarial relationship between technology and the humanities.

© Gerhard van der Linde (Pretoria)

References