The digitalisation of literature has proceeded on at least three fronts simultaneously, and these fronts have only lately merged – this merging being the main cause for the attention on digital literature we are presently witnessing. In what follows I will approach the history of the digitalisation of literature from the viewpoints of word processing and desktop publishing, hypertext, and text adventure games. Then I will discuss the new forms of writing and reading made possible by digital media and Internet. Special attention will be paid to two essential properties of digital texts, namely, interactivity and temporal manipulation.
One of the evolutionary lines behind digital literature is the development of digital text processing devices: word processors and layout programs. It is hard to estimate the impact word processing has had on literature, since there are no thorough studies on this topic; by this I mean studies which would scrutinise if the use of word processors has in any ways changed literature content wise or form wise. Some notions based on personal experience can be presented, though: one Finnish publishing editor, for example, told in a seminar talk how the amount of manuscripts submitted to them increased dramatically by the mid-eighties – apparently in close relation to the increase in the amount of personal computers sold. This assumption is backed by the further notion that most of the manuscripts were more or less autobiographical writings by middle aged, middle management, male writers. That is, exactly the same segment who first got computers on their office desks. From this we might draw the conclusion that word processors at least lowered the threshold to start writing. There is a similar implication in Günter Grass’ recent claim that he can tell from a manuscript, after reading just ten pages, if it is written on computer or not – because writing by computer is too easy, and text proceeds faster than thought… (Kettman 2000).
Another important feature of word processing are the manifold possibilities to manipulate the visual aspects of text, from font types to page layout. The newest word processors are so close to desk top publishing software that they are often actually used in making smaller publications. On the other hand, the word processors have their limitations, and hyperpoet Jim Rosenberg, unsatisfied with the clumsy working of wordprocessors, has called them 'word constrainers' (1996a, 110). As an opposite example we can take fiction author Raymond Federman – the author of extreme avant garde novels has been espeacially fond with typographic tricks: the 240 pages of his first novel Double or Nothing (1972) are all unique in design, none of them confirms to the normal ”evenly spaced lines from the top of the page to the bottom” format. Each page has its unique typographic design – Federman used two years to write this novel with a typewriter! Later he totally abandoned typographic experiments because, according to him, word processors made it ”too easy”.
In any case, it is much easier, compared to the typewriter, to play with visual aspects of text when writing by a word processor – this was stressed, for example, by Richard A. Lanham in his influential book Electronic Word (1993). But how much has this possibility been used, in practice? After my personal experiences as a reader and literary scholar – whose interest has been directed especially towards the so called experimental writing – literature still largely confirms to the traditional page format. We should not forget, however, that there have been works like Laurence Sterne’s Tristram Shandy, from the 18th century, which uses the page layout much more effectively than most of the present fiction.
It can be argued, then, that digital desktop publishing has had a more profound effect on literature at large than word processing. Together with postscript coding it enables the editing of text ready for printing more effectively, faster, and, cheaper than the traditional printing press. This way it has produced a new agenda for small publishers, who, in their turn, are the central, and often the only, channel available for experimental and marginal literature. Thus, desktop publishing has significantly reformed publishing industry and helped marginal, that is, more or less non-profitable, literature to survive.
Through layout processors and digital presses the digitality has sneaked almost unnoticed into literature, without too much fuss. How many, after all, of the poets still writing with a feather stylus in candlelight have realised that their manuscript, after they have left it to the publisher, for starters has been coded to digital format? Keeping that in mind, the worry many people have showed for the decay of literature because of digitalisation – I am thinking now of Sven Birkert’s Gutenberg Elegies, especially – is badly belated, which, in its turn, indicates that they are mistaken: if the problems they point out were truly caused by digitalisation, the problems should have been visible some ten to fifteen years ago already (Birkerts 1994).
Layout and word processors, then, have been one channel for the digitalisation of literature. Another channel has been the development of hypertext authoring tools and the networking of computers – which together have made possible Internet and World Wide Web.
The father of hypertext, as commonly seen, was Vannevar Bush with his microfilm based device for associative writing, Memex. Memex was intended to combine a large data base and a possibility to link different sections of that data base to each other. This was meant to make data retrieval and modeling easier, but also: it was Bush’s idea that when a person starts reading a Memex document written by another person, she could, through the link structure, access the associative reasoning chain which was behind that particular document. Memex was never constructed, and digital computers, developed at around the same time, would soon have made it obsolete in any case (Bush 1945).
Computer based hypertext systems were developed, however, only from the late fifties onward. Especially two of the projects need mentioning. First, Douglas Engelbart’s NLS System (also known as Augmentation) – the basic idea was close to Memex, and as the name suggests, Engelbart’s aim was to build an augmentation for the human intellect, a program which would make possible even more complex mental processes. Additionally, it was intended to strengthen mental cooperation, so that many persons together could develop a common idea. Several spin-offs of Engelbart’s work are still in use today, and nowadays he is mainly known for two innovations: the mouse pointer - a device without which no computer today seem to manage, and, the windowing of the screen, which was first successfully implemented in MacIntosh operating system, and later, with too well known results, appropriated also by Microsoft – the latest version is Windows 2000 and today it is hard to find a computer without a windows-based operating system (even though the Mac users joke: Better a room without windows, than a room with Windows 98). Even the Linux OS has gotten its windowed versions (X-windows).
If Engelbart has been succesful with, at least, some of his ideas, the same also applies to the other hypertext visionnaire, Theodore ”Ted” Nelson, who coined the terms hypertext and hypermedia. His ideas and visions have been highly influential in computer world. His own hypertext project, Xanadu, has not been implemented to general use so far, but techniques developed in its construction have been used in other forums succesfully. In some sense Xanadu was overrun by Internet, but Nelson is still working with his grand vision.
When speaking of Nelson, we should notice the strong emphasis he lays on hypertext being a literary thing. The book he published in 1980 - it has gone through several editions and modifications since - was titled Literary Machines. The subtitle says:"The report on, and of, Project Xanadu concerning word processing, electronic publishing, hypertext, thinkertoys, tomorrow's intellectual revolution, and certain other topics including knowledge, education and freedom." Also, there is in the cover the following statement:”This book describes the legendary and daring Project Xanadu, an initiative toward an instantaneous electronic literature; the most audacious and specific plan for knowledge, freedom and a better world yet to come out of computerdom; the original (perhaps the ultimate) hypertext system. Do not confuse it with any other computer book.”
These examples should already show how crucial the relation between hypertext and literature was, and still is, for Nelson. In quite a poststructuralist way he understands literature as a vast network of citations – it should be noted that Nelson started working with the Project Xanadu at the same time as Jacques Derrida and Roland Barthes laid the base for poststructuralism and deconstruction; that is, around the mid-sixties. Literature is, in its essence, citations between texts, and hypertext is specifically a device intended to enable, citing the cover, ”instantaneous literature”; it means that the text or passage cited can be instantaneously accessed – thus: hypertext system is a literary machine.
Nelson defines hypertext as follows:"Well, by 'hypertext' I mean non-sequential writing -- text that branches and allows choices to the reader, best read at an interactive screen. As popularly conceived, this is a series of text chunks connected by links which offer the reader different pathways."
Nelson uses in his definition the word ’non-sequential’ – usually, in hypertext theory today, the term ’non-linear’ is used. This has been a topic for lively discussion, and the common understanding seems to be that ’non-linear’ writing or reading is an impossibility – as a better alternative, the expression ’multi-linear’ has gotten strong support, to emphasise the fact that there are several possible reading orders in hypertext (see, for example, Landow 1992, 4).
The individual ’text chunks’ of hypertext are usually called, depending on the context, either nodes, pages, frames, workspaces, or, as is quite common in theoretical usage, lexia. This term was borrowed by George P. Landow from Roland Barthes’ essay S/Z, and incorporated to hypertext theory[1]. When dealing with the Macintosh based Hypercard environment, the term ’card’ is used, while in the WWW the highly misleading term of ’homepage’ is employed. Link is a link – a connection between any two lexias. Here a terminological clarification is in order. ’Anchor’ is a term closely related to link. An anchor is the exact place in a lexia to which a link is attached – that is, the starting or ending point of a link.[2] Stuart Moulthrop and Nancy Kaplan have also suggested the term ’cue’ to signify the way in which an anchor is indicated from the surrounding text. In WWW, for example, it is a standard way to indicate an anchor by blue font color and underlining – if an image serves as an anchor, it has a blue frame. The word underlined, or the image framed, in these examples, are anchors, even though they are often called links – to be exact, link is the connection between two lexias, and a cue indicates an anchor and shows the presence of a link, and usually gives some information about the destination of that link. In some cases, in some of the hypernovels, for example, the links / anchors are not indicated by any cues at all; by this, certain aesthetic effects are sought after.
The lexias may include not just text, but also images, sound, video clips etc. – technically speaking there is no difference between linking two text chunks, or text and image, image and sound etc., so we can understand the term hypertext as synonymous to the terms hypermedia and multimedia. It would seem as quite a natural practice to use ’hypertext’ as an abstract term to describe the structure of certain documents – ’hypermedia’ and ’multimedia’ would be more concrete expressions which would indicate that the hypertextual structure of that particular document includes several types of lexias – text, sound, images, and their combinations.
Of hypertext editors the Story Space is specifically designed to literary use. It has been the software environment for the best known hypernovels, Michael Joyce’s Afternoon. A Story (1987), Stuart Moulthrop’s Victory Garden (1991), and Shelley Jackson’s Patchwork Girl (1995). Each of them uses the hypertextual possibilities in its own way, thus offering the reader various ways to participate in the unfolding of the story – but what is common to them all is that they heavily rely on alphanumeric text, audiovisual augmentation is used scarcely or not at all. Also Hypercard editor has been used to write literature, even though the focus of its use has been on instructional materials and manuals. Of Hypercard based texts at least John McDaid’s Uncle Buddy’s Phantom Funhouse (1993), and Deena Larsen’s Marble Springs (1996) should be mentioned. There have been various other experiments, like the collaboration of Jaime Hernandez (best known for drawing Love and Rockets comics) and Monica Moran, titled Ambulance. An Electronic Novel (1993) – because of poor marketing these are all already nearly forgotten.
A Screen from Michael Joyce's Afternoon. A Story (1987)
A Screen from Shelley Jackson's Patchwork Girl (1995)
A Screen from Monica Moran's and Jaime Hernandez' Ambulance. An Electronic Novel (1993)
New mass storage means, especially CD ROM, have made it possible to distribute much bigger files conveniently. Because of this, hypernovels may now include more audio-visual materials. A good example of these ’new wave’ hypernovels is M. D. Coverley’s Califia (forthcoming), which, in the way of a digital Gesamtkunstwerk, by combining images, sounds, and text, produces a highly immersive fictional world.
Screens from M. D. Coverley's Califia (forthcoming)
At the same time elsewhere: computers were connected to each other using phone cables – the US Ministry of Defense started the ARPANET project. It didn’t take too much time for the first nerds in history to develop computerised versions of Dungeons & Dragons style role playing adventure games (f.ex. The Adventure). They were text based adventure games, in which the player directs the action through simple typed commands like ”go south”, ”look”, ”open door” etc. This is, clearly, the beginning of interactive fiction. Quite soon these games were adapted to the computer network, where several players could simultaneously play in the same dungeon, and fight not only the machine, but each other, too. That was the first MUD (multi user dungeon) ever. These multi user text based virtual realities proliferated, and in some of the new MUDs fighting and the ’dungeon’ were marginalised or totally discarded, and the communication with other users was taken as the main function. Thus, the nature of MUDs changed significantly – although the adventure game environments still form a big part of MUDs – and in order to hide the embarrassing origin the acronym MUD is now usually explained to stand more neutrally for ’multi user domain’, or, ’multi user dimension’.
The ARPANET rapidly expanded and crossed the boundaries to become an international network. Finally, in the eighties, it had reformed to the global Internet. When Internet, then, was developed to employ hypertextual techniques and graphical user interface (the browser), World Wide Web was born. WWW is the most visible part of Internet and usually when people talk of ’Internet’ they actually refer to the WWW – but Internet includes, in addition to WWW, such parts as e-mail, newsgroups and ftp (file transfer protocol). New browsers, however, have integrated these parts to such a seamless whole that the difference between Internet and WWW is a thin line indeed.
One of the first Internet based literary experiments was the Hypertext Hotel project initiated by Robert Coover at Brown University. It was a collaborative text built on a MUD – in the Internet it was available for reading to anyone interested, and also, all readers had the possibility to participate in the writing of the text. Most of the early literary pieces published in the Internet were stories written by amateur writers, and they only occasionally employed some minimally hypertexual techniques – mainly the forking paths model, where, at certain turning points, the reader could choose one from several possible directions for the story.
While Internet and the browsers have rapidly developed, Internet literature has increasingly shifted towards the dominance of the visual and the multimedial. The way how alphanumeric text has turned to just a one element among others in the multimedial whole, can be clearly detected, for example, in a German Web text titled Das Epos der Maschine – in it the text has a merely visual, or better yet, ornamental role.
Currently we are in a situation where we have, along the traditional print literature, digital literature both in the Internet, and, on floppydisks and cd-roms – but the tendency seems to be towards merging these together while the so-called hybrid technologies are getting more common. In hybrids, a cd-rom (or something similar) and Internet are combined so that the cd-rom may include an interface to a certain database (containing, for example, hypernovels) and additionally, buying the cd-rom gives the buyer access rights to that database, which then can be used via Internet to get those hypernovels (or whatever) to the user's terminal.
When micro computers started their invasion into people’s homes in the early eighties, their graphical capabilities were very limited from the viewpoint of computer games. The computers could much more effectively process alphanumeric text than images, so it was quite natural that along the shoot’em ups and other action games, also text based games were developed (and of course, there were already the precursors in the Dungeons & Dragons games written for the early mainframe machines). Text adventures offered the player several alternative ways to proceed through textual environment; in practice the proceeding was usually strictly limited to solving different kinds of riddles and problems, and there was usually only one succesful way to go through the game.
Screen from Michael Detlefsen's text adventure The Star Portal
A big portion of the adventure games were based on The Lord of the Rings style fantasy, to some extent on science fiction, and also, on mystery and detective stories. They maintained their devoted reader/playership for quite a long time despite the development of highly graphical action games, and some of them were rather ambitious both in their structure and in their text. Anthony Niesz and Norman N. Holland (best known for his studies in the reader response research) published the article ”Interactive Fiction” in Critical Inquiry (1984) – in the article they were discussing just these text adventure games. Even though the article makes it quite evident that Niesz and Holland did not really know the games all that well, they did, however, bring the term interactive fiction to public notice.
The early hypernovel authors seemingly had a strong desire to distinguish themselves from the world of the text adventure games – the slogan used by hypertext publisher Eastgate Systems, ”the source of serious hypertext”, seems to be echoing that desire. More recently the hyperfiction authors have accepted the game or playing aspect as an essential part of hyper and cyberliterature, and it is only appropriate to acknowledge the role of the text adventure games as the first, and even economically lucrative, form of digital literature.
There is still one more line of historical development which should not be forgotten – that of the avant garde tradition of literature itself, the tradition which has challenged and transgressed the limits of literary expression. Composition No. 1 by Marc Saporta is a deck of one hundred loose leafs, meant to be shuffled before each reading, thus getting a random order for the text in each reading. Pale Fire by Vladimir Nabokov consists of a Foreword, the poem ”Pale Fire”, and extensive Commentary and Index for the poem – the reader has several options to integrate and interpret these four sections, resulting in potentially very different outcomes as for the ”real story”. Rayuela by Julio Cortazar includes 155 numbered chapters which can be read in any one order (the author, however, suggests one particular order in the beginning – an order according to which the reader has to ”hopscotch” from one chapter to another through the text). Robert Coover’s short story ”Quenby and Ola, Swede and Carl” is a skillfully constructed text in which vague references and the hiding of the personality of the focalisers enables several interpretations of what is really going on in the story. Brian McHale has paraphrased Coover’s story as follows:”Carl, a businessman on a fishing holiday, either sleeps with one of his fishing guide's women or he does not; if he sleeps with one of them, it is either Swede's wife Quenby or his daughter Ola; whichever one he sleeps with (if he actually does sleep with one of them), Swede either finds out about it or he does not; if he does find out, he either plans to kill Carl in revenge or he does not. All of these possibilities are realized in Coover's text.” (McHale 1987, 107-108). In Raymond Federman’s Double or Nothing (1972) and Take It Or Leave It (1976) exceptional typography is used extensively, and as an additional site for signification; in the opening page of Double or Nothing, for example, the text circling round the edges of the page form a rectangle which represents the small room into which the narrator has locked himself in order to write his novel (see picture; to be exact, the narrator in the room is not the primary narrator in the novel). The OuLiPo group has experimented for decades with various kinds of procedural writing, producing texts according to formal rules – inventing the rules themselves has been the primary focus for the Oulipians, producing actual texts not nearly as important, thus the ”potential literature” in the name.
These and numerous other experiments have in various ways anticipated the directions which digitality has opened for literature. From today’s point of view we could call the texts mentioned above as ”protohypertexts”[3]. It could even be argued that these authors have, in some ways, gone further than the authors of hypertexts proper so far – knowing the work of these predecessors should prove most useful for novice hypertext authors.
A page from Raymond Federman’s Double or Nothing (1972)
As a topic, digital literature is very diffuse and very hard to define. As a crude working formula, the following classification could be used.
1. digitalisations of print literature – this group includes such great archival pursuits as The Project Gutenberg, and its Scandinavian counterpart, The Project Runeberg, which aim at digitalising as comprehensively as possible the old, mostly canonical, literature. These projects have several practical ends. They conserve old texts physically deteriorating, make rare works (which may exist in only one known copy) available for the larger public, and create useful corpuses for researchers and students. The possibility of word and phrase search in the texts alone greatly expands the usability of these archives for the researchers. In addition, the digital text archives enable all kinds of statistical style analyses – the term ”humanistic computing” has actually come to mean just this stylistical analysis of digital text corpuses – the study of whether a certain play is to be attributed to Shakespeare or not.
2. the digital publication of original literature – texts in this category do not employ any hypertextual techniques at all, or do so only modestly. Literature confirms to the established conventions, and digital form is primarily used in the distribution of the texts. This kind of publishing has been very small so far, and it has been, mainly, done by amateur writers. There are also established authors who have tried digital publishing – like the Finnish author Leena Krohn who published her story collection Sfinksi vai robotti? (1996; The Sphinx or the robot?) only in digital format – in this case the reason was that Krohn wanted to include digitally enhanced illustration for the stories, which would have been very expensive in print book. There has been a small amount of parallel publishing too, that is, publishing the same material both in print and digitally. This has been mainly done by different magazines (amongst which there are rather many literary magazines).
3. literature using the new techniques made possible by the digital format – this group includes everything from hypernovels to interactive poetry and multimedia encyclopedias, etc. Michael Joyce’s Afternoon. A Story (1987) is commonly seen as the first hypertext novel.[4] It is a hypertextual story consisting of 539 lexias and 951 links connecting them, and it tells about a man who sees a car accident site on his way to work in the morning, and who is afraid that the accident victims (probably dead now) may have been his ex-wife and son. It is possible to read Afternoon in several ways (different set of lexias, in different orders), resulting in potentially quite diverse stories, but it is still possible to construct a frame story above the readers’ stories, telling about a man who procrustinates in finding out the actual facts about the accident victims’ situation and identity – and traces the complex human relations and love affairs with his ex-wife and colleagues, and ponders about his own life. Traversing the text systematically through, one can find the ”real story”, which explains why the man is simultaneously so shocked about the accident, and, frantically, tries not to think about it[5]. There has been, so far, only one commercial publisher of hypernovels (and stories), but there are also diverse hypernovels produced in universities around the world – but usually these are not easily accessible.
There are also poetic works using interactivity and kinetic techniques. Jim Rosenberg’s Intergrams (1997) are especially interesting – they have two significant features: simultaneities, that is, several layers of text juxtaposed so that by moving the cursor, single layers can be read, and, syntax ”externalised” to graphic symbols which represent the syntactical interrelations between text fragments[6]. In Robert Kendall’s poem A Life Set for Two (1996) the reader may choose the ”atmosphere” of the text and other variables from a ”menu”. There are also individual words which may change into other ones in the screen according to a programming.
A Screen from Intergrams (1997) by Jim Rosenberg
4. networked literature – this is hypertextual literature using the special features made possible by, and only in, the Internet. For example, in Leporauha [Rest Peace] by Matti Niskanen (1998), the linking possibilities of the open Web environment are used subtly but effectively. Most of the links operate inside the textual field of Leporauha itself, but some of the links, without notice, lead outside, to sites of political parties, or, to the front page of a yellow paper etc. With Internet based texts it is also possible to update the texts continuously – all the changes made to the text are immediately present for the readers, without the need to purchase new versions on cd-roms, for example. In Internet the reader feedback is easily arranged and immediate, which makes it possible to let the readers participate in the writing and rewriting of the text. A very interesting example of this kind of interactivity is Markku Eskelinen’s Interface (1997), a work in three phases, which started as a print novel and now goes on living its own life in the Internet, steered by the author’s own input, and reader feedback.
In addition to these four categories there is a variety of digital text types which to a variable extent include literary aspects – narrative structures, fictionality – but which cannot be really called literature; especially computer games, simulations, all kinds of MUDs, IRC (Internet Relay Chat), virtual realities etc. These are all highly interesting phenomena in their own right, and they clearly have much to give to literature and literary studies, but it would violate their true nature to categorise them as ’literature’.
On the other hand, it should be noted that all pages in the Internet are hypertextual in nature. Thus, all the much read pages, like the sites of the popular youth tv-channels, or, news sites – they all familiarise their readers to hypertextual presentation. Sites like these usually make an effective use of direct reader feedback, all kinds of audience polls, and other means foregrounding reader interactivity. It does not seem too far fetched to expect new kinds of Internet literature to grow out of that soil in the near future. In any case, they familiarise their readers with the possibilities of Internet communication in ways which are accessible to much larger audiences than any of the texts discussed in this paper.
To put it briefly, the novelty of digital textuality can be located, mostly, to interactivity and temporal manipulation.
’Interactivity’is a highly problematic term in regard to literature, since all literature, after all, is interactive – a fact which was systematically formulated by Roman Ingarden already in thirties in his work Das Literarische Kunstwerk (1960). This idea was further developed, especially, in reception aesthetics and reader-response studies; these research practices take it as their starting point that the reader is an active participant in the literary signification process. There are several kinds of interactivity, and the difference between the interactivity of conventional literature and that of digital literature was, for the first time, clearly expressed by Espen Aarseth, who decribes four categories of reader(/user) functions: interpretation, navigation, configuration, and writing.
Interpretation is an inseparable part of all reading. When reading hypertext, one has to, in addition to interpretation, actively navigate her way through the net of hypertextual paths.[7] Furthermore, the reader/user may be allowed to configure the text, for example, add her own links to the hypertext. Configuration, thus, means reforming the text within certain limits. The last user function is writing, that is, the user is allowed to participate in the writing of the text – and writing may also be programming. It is a commonplace in hypertext theory to claim that because of interactivity, the ”reader becomes an author” – this is, however, not accurate but only in those texts which offer their readers the writing function (in other texts the claim can be accepted only in some metaphorical sense), and such texts are very rare so far.
In a convincing way Aarseth also shows how the distinction between print and digital texts is, actually, quite ineffective. In many cases a certain print text may be much closer to some digital texts, than to other print texts, and vice versa. Aarseth, then, rather speaks of cybertextuality, which he defines as a perspective on all texts, independently of their medium: if a text makes use of configurative and writing functions, then it clearly is a cybertext[8] – on the other hand, if a digital text does not use any other user functions than interpretation, then it does not, in any significant way, differ from traditional texts. This is a healthy notion, and mostly acceptable, but in practice the difference between print and digital texts may be more significant than Aarseth claims.
The manipulation of the temporal dimension is a topic which in Aarseth’s, otherwise exemplary, study is almost neglected, and it seems to be exactly that direction where the biggest differences between print and digital texts can be detected – in print texts it simply is not possible to control the temporal aspects of reading activity as is the case with digital texts. As examples of possibilities for temporal manipulation we can list the following:
1.limiting the reading time – the text will stay on the screen only for a limited amount of time. For example, in the Web text Hegirascope (1997) by Stuart Moulthrop the text on the screen changes every thirty seconds. Additionally, there are hyperlinks on the pages, clicking which the reader may ”direct the course of the text stream”. There are also texts which can only be read once – these can be seen as special cases of limited reading time. Surely the best known read only once type of text is Agrippa by the cyberpunk science fiction author William Gibson. In Agrippa the text scrolls by itself on the screen, and when a line has scrolled out of the screen, it vanishes – thus the reader can never return to the text already read[9]. Uruguayan multimedia artist Gonzalo Frasca has also written highly interesting ”one session narratives”, texts which change each time the reader starts reading them. Once the reading session has ended, the reader may never return to the exactly identical text.
2. delaying reading time – the reading cannot proceed but after a certain waiting period. Let us say there is a scene in a text, where the protagonist takes a fifteen minutes nap – now the traversing through the text may be halted for that fifteen minutes, or whatever. (This has been, so far, mainly a potential option which hasn’t been used too much intentionally – in Internet one is, of course, confronted with this effect way too often, unintentionally).
3. restricting the reading period – this is also mainly speculation, but Markku Eskelinen, in his collection of essays Digitaalinen avaruus (1997; The Digital Space), has suggested several ways to employ this effect: a novel which can only be read during office hours etc. On the other hand, the text may vary according to whether it is read on day time, or night time etc.
4. text ”living” in time – a digital text may be updated at various intervals. As a minimalistic case we can take Leporauha mentioned above with its links to the front page of a newspaper. The contents of the front page, naturally, change daily, and with this simple device the author has achieved a work which changes daily without having to do anything to it himself. Interactive texts (which employ at least configurative user function) do change continuously through the work of the active audience. Thus we can have ”living” (or evolving) texts, whose existence is processual in nature.
Now we can see that possibly the greatest challenge literature confronts because of digitality is the step from closed, clear cut, and static text (a novel, a book) to an everchanging and evolving amoeba: this concept undermines our notions of great works, canons, creative genius, aesthetic object etc. This kind of writing is always hybrid, in the process of becoming something else; but isn’t just this the essence of literature, and especially of novel – this is, at least, how Mikhail Bakhtin sees it. And one cannot but think Finnish author Matti Pulkkinen’s definition of novel: novel is like a pig, it eats anything; digitality has just once more taken that piggishness in to the fore.
What will the reaction to that be like? We are currently in a situation where digital texts are located on a no-man’s land – the traditional print publishers have not been too interested in digital publishing (and now I mean specifically hyper and cybertextual literature; naturally, digital techniques have been employed in print on demand experiments etc.), but there are neither too many established digital publishers. Virtual book shops (Amazon.com as the flagship) belong to the rare species of functional Internet trading (and even they are not turning any profit yet), but it is almost totally focussed on selling print literature, traditional books. It is quite possible that the traditional publishers will stay as such, and digital textuality produces its own trade – it is, after all, clear that ’book’ as a concept is less and less suitable for most of the digital texts integrating even more stronger multimediality and immersive virtual reality aspects to itself.
It may be, then, that cybertexts like the ones presented in this paper will be seen as a totally new media alongside literature, multimedia, computer games, and cinema. This wouldn’t mean that the digitalisation of texts would have no significant effects on the book trade. Already Internet is an important distribution channel for books. In future, more and more of literature will be distributed in digital format through Internet, so that the customer may read the texts in the form she prefers – from computer screen, with an electronic reading device, or, as a personal hard copy. Tailored books are also possible, and even now one can order children’s books with characters named after their own children etc. A print book is nowadays also a product of digital processing, and thus, more flexible a medium than ever before. Print book today is by no means an innocent object any more in this digital era. Another growing trade may very well be the total opposite to cybertexts: handmade, unique, definitely non-digital artist’s books.
[1] Actually, the use of the concept of lexia is quite misleading here - for Barthes the lexias are always interpretational constructs, not organizational principles inherent for the textual structure (like the nodes in hypertext):"The tutor signifier will be cut up into a series of brief, contiguous fragments, which we shall call lexias, since they are units of reading. This cutting up, admittedly, will be arbitrary in the extreme;…" (1993, 13)
[2] In some cases a whole lexia may be one anchor, but usually a word or sentence inside a lexia works as an anchor. One lexia may naturally include several anchors, and there can also be several links from one lexia.
[3] ’Proto-hypertext’ as a term is a bit problematic, though; Espen Aarseth sees the term as a form of colonisation by the hypertext community, on the other hand, many of the texts described here could be more accurately called ”proto-cybertexts.
[4] The hypertextual oeuvre of Michael Joyce includes, in addition to Afternoon, ”W.O.E” (1992), Twelve Blue (1994), and, Twilight. A Symphony (1996a). Joyce has also played a significant role in the designing of the hypertext editor Story Space.
[5] A detailed reading of Afternoon is in Douglas 1992; see also Klastrup 1997 and Walker 1999.
[6] Jim Rosenberg works as a programmer – he has written, for example, the code for John Cage’s first computer compositions. Most of Rosenberg’s print poetry is included in the Diagrams series. He has also done conceptual poetry and installations. From the mid-eighties onward he has mainly concentrated on the digital, interactive Intergrams series. Rosenberg is also known as an important hypertext scholar.
[7] With the so called proto-hypertexts, especially, it is common that there are only few alternatives between which to choose – in cases like these it seems a bit exaggerated to speak of ’navigation’. Instead, it would be more accurate to speak of ’selection’ (as a subset of navigation).
[8] In connection with cybertexts, Aarseth also writes about ergodic literature (’ergodic’ is coined from the Greek words ergos and hodos, which mean work and a path, respectively, and thus, ergodic literature requires from its reader active work in seeking/following the textual paths.
[9] To be exact, the text did not just vanish, but was coded to aminoacid chains. There is the possibility to decode the coded chains back to original text.