Blip Magazine Archive


Home : Archive : Links



Over the last years, hypertext has been generating too much interest and hype as to still require a formal introduction. Questions and arguments over the future of the printed book in the light of this innovation and the possibilities of building a Xanadu (Ted Nelson) from hyperlinks and nodes have become frequent in print and abound on the World Wide Web, which of course is itself a hypermedia text of unprecedented size and accessibility. A good share of interest has been devoted to hypertext as a medium for fiction, sometimes called "Interactive Fiction."

In this paper, I will examine the perspectives of hyperfiction as an art form and its relation to fiction as we know it, trendily modified to "flatland" fiction these days. Considerations of the possible economic or cultural impact of hyperfiction lie outside the focus of this study, as does the perennial objection, "You can't read hyperfiction in the bathtub." Instead, my attention will be devoted to the aesthetics and possibilities of hyperfiction, and how they differ from fiction.

Browsing the World Wide Web can serve as a quick antidote to the foaming euphoria proponents of hyperfiction foster. Most hyperfiction encountered here seems clumsy, unsatisfactory, and of little artistic merit. Once the novelty of clicking on underlined words or outlined icons wears off, there is not much left to be euphoric over-- the stories seem to be lacking in everything but innovative structure, and the structures seem murky and pointless. There is a sense that the basic elements of the form have not been understood properly and are used in a haphazard way by most of its pioneers, happily experimenting on the fringes of cyberspace. It might be the proper time to ask questions about the essential problems and assets of the form.

One of the most prevalent problems plaguing current hyperfiction is what I call the "poles-in-your-face" effect: Many of the hyperstories found online are lacking in content and quality writing because the novelty of hypertext makes all other aesthetic concerns secondary. This seems to be an intrinsic problem with newly discovered forms or variants of forms-- the same obsession with the possible explains why early stereo recordings commonly feature oscillating sounds between the left and right speakers and actors in 3-D movies tend to wield poles toward the camera. The special feature in a medium, at least during an initial phase of experimentation, tends to be overused while eclipsing other, more traditional qualities.

If hypertext fiction is to be used in a satisfying, artful way, its specific features need to be understood-- those that cause problems due to the early stages of experimentation, and those that will remain at any stage of development. This second class, in turn, can be divided into assets that have to be used in a certain medium-specific way, and those that remain irrevocable problems which the medium will not outgrow.

I will limit my concerns to so-called "explorative" hyperfiction, which is authored by a single writer, as opposed to the more experimental "constructive" hyperfiction in which every reader is invited to add to or even delete parts from the body of the story. (For an ongoing experiment in constructive hyperfiction, see Robert Coover's "Hypertext Hotel" on the World Wide Web at

Probably the single most-often quoted difference between hypertext and "plain" text is that of dimension: a printed page, the argument goes, is linear, while a hypertext document is three-, if not multi-dimensional. Much has been made of the "multilinear," "multicentered," "networked," "fluid," "changing," "unfixed," and, finally, "antihierarchical" nature of hypertext. (Fowler 17, 18, 19, 21). By comparison, what we knew as fiction has become "flatland fiction" that suffers from "the frozen structure of the printed page," the "tired artifice of linear writing" that needs to be "liberated" by the three-dimensional "writing space" of hypertext (Bolter 21,146, 21).

As so often with hyped novelties, the differences between text and hypertext are less pronounced but more complex than the hyperhypers care to admit, or seem to know.

"Flatland" text works through the linear structure that the writer has laid out before the reader after making certain artistic and aesthetic choices. The writer chose to tell a particular story, in a particular mode and register, with characters, scenes and details which he devotes a certain amount of interest and narrative time to. The writer chose to tell the events in a certain order, with certain emphases, with, possibly, additional ways of reading the story built into its fabric. The structure might adhere to a classic, time-honored model, a convention the reader is likely to be familiar with, or structure might have been toyed and tinkered with to parody or renounce. The reader has to accept the choices the writer made, like a visitor in an art gallery has to accept the painters' choices of color, tone, perspective, motive, etc. The reader then engages in a relationship with the text, lingering over certain sentences, rushing over others. Some images might resonate in his mind, triggering further images and memories that are the reader's own. A complex relationship forms; the reader produces a reading, a version, a personal production of the text in her mind as she reads. All readings will be individual, and it is within the reader's power to rush through the text without so much as barely noticing its fine texture, or to engage with symbols and words, going back to earlier passages, and so on. Any text creates a three-dimensional space in which words echo each other and refer the reader back and forth, intratextually or intertextually.

The much-praised hypertext structure, however, in most cases remains a black box to the reader, who is asked to make choices that will help create a linear text. This linear text, of course, has the same potential for fine texture and networks of meaning as the flatland text. What seems hard to grasp, however, is that the result of a hypertext reading is not anything different from a linear story--the hyperstructure remains invisible to the reader, it is only in place to generate multiple linear readings. The implied structure of the hypertext might be "multilinear," "antihierarchical," "decentered," with "no obvious beginning or ending, no top or bottom" (Fowler 19), but the resulting linear texts that it produces are not, or not any more than the "frozen" flatland texts.

Probably the most striking feature of hypertext is the link-- the word, sentence, or icon that refers to the next node, or piece of text, which in turn offers more links to the reader's incessant mouse clicks. As Bolter rightfully notes (201, 204), the link is a sign that signifies the node it links to, which in turn signifies other nodes, and so on, ad infinitum. This endless chain of signification accounts for the feeling of vertigo (Johnson-Eilola 195) often reported by hypertext and WWW users--they are caught in a signifying chain very much like the one threatened by Jean Baudrillard (10). For hypertext fiction, however, this is less important than the nature of signification employed by the link. It seems that the way in which the link signifies is not properly understood. This is obvious to anyone who ever wondered why they read what they read after clicking on a link--the gap that yawns between the link and the new window that brings the next node is often big enough to swallow a lot of the Web's hyperfiction in its maw.

In this space, I can only suggest angles for the examination of this problem. Commonly, the hypertext link would offer the reader a choice of action for the protagonist of the story--the classic Choose Your Own Adventure hypertext, or what Gareth Rees has called "tree fiction," which offers the reader binary decisions and creates a "Garden of Forking Paths" (Borges) for him.

According to Stuart Moulthrop,

Michael Joyce's afternoon: a story (Eastgate Systems, 1990) introduced a major technical enhancement. Joyce rejected the pragmatic commands found in adventure games ("Go North"; "Take gold"; "Hit troll with ax") in favor of "words that yield:" cues to further development imbedded in the language of the story itself. In an encounter with afternoon, the reader may find the sentence: "I want to say I may have seen my son die this morning." If the reader selects the word "son," she follows one narrative direction; if she chooses "die," "I want," or some other set of words, she will go another way entirely.

The semantics, however, of "words that yield" are unclear, and the reader can never be sure if her click will influence plot, perspective, time, character, or any other element or combination of elements in the story. While Coover is "always astonished to discover how much of the reading and writing experience occurs in the interstices and trajectories between text fragments" (23), I would point to the more practical aspect of an often confused and befuddled reader who struggles to make a connection between two nodes.

Another, more promising way of signification through links, has been called "advanced footnoting." Its premise is, or might be, that a link will lead to a node that yields more information on a given subject without conflicting with the continuity of the story. Reeth considers advanced footnoting pointless since the reader will return to the main path of the story without the additional information yielding any impact. However, examples from different media suggest how powerful this strategy might be. Julian Barnes' Talking It Over can be considered a text that draws most of its power from the tension between the different perspectives; the book derives its impact exactly from the additional information that is available to the reader through changes in perspective. These changes could be simulated through advanced footnoting links. The 1946 movie The Killers can be considered one long visual footnote to Hemingway's short story of the same name; the movie covers the territory of the story in a few minutes and proceeds with a plot that infuses the events of the story with additional meaning. The most striking example of the power of advanced footnoting, however, is Ridley Scott's director's cut of Blade Runner, where the addition of 30 seconds of material resonates so thoroughly with other events in the movie as to change its meaning completely. (See Blade Runner Faq at usenet/news.answers/movies/bladerunnerfaq for an in-depth discussion.)

If the connection between nodes in a hypertext of any kind is not to become as arbitrary as that between randomly flipped-through channels of television, then the rules for signification of a link have to be made clear. These can and should vary between different hyperfictions, but the author has to understand the link as a grammatical and semantic element of his work that has to be controlled and fully realized to function properly.

The area of worst overestimation of hypertext is the liberating and democratizing effect it is supposed to have on both reader and writer:

Users of literary hypertexts can come to realize their power as reader-writers whose actions appear to determine fundamental characteristics of the story. In such a text, the common distinctions between "writer" and "reader" begin to collapse in a way that has long been theorized for print text but not realized in such visible form. (Johnson-Eilola 195)


The interactive reader of the electronic word incarnates the responsive reader of whom we make so much. Electronic readers can do all the things that are claimed for themor choose not to do them. They can genuflect before the text or spit on its altar, add to a text or subtract from it, rearrange it, revise it, suffuse it with commentary. The boundary between creator and critic (another current vexation) simply vanishes. (Lanham, quoted in Fowler)

Many even see the makings of the return to an oral culture in the possibilities of hyperfiction--the story becomes a kind of puppet show, the writer a storyteller who performs his piece in cyberspace, like a Homeric epic (Bolter 57ff., Fowler, Ong) where the audience claps, boos, and clicks on "No" to influence the course of the story.

However, to further use the discourse of power that permeates this discussion, the author does not only transfer some of his authority on the reader, but also the responsibility to select. Contrary to many a published opinion, the author does not sacrifice much in terms of control, since all the nodes are still written by him. However, both the author and the reader are burdened by the element of choice that has been assigned to the reader in the form of clickable links.

The reader's problem is that she is bound to have insufficient information to make a selection. She will be continuously wondering which link to choose, which link is "the right one." There will be an ongoing sense of missing out on interesting nodes somewhere else, aided by a confusion as to her location within the framework of the story. The much-praised decentered nature of the hypertext, which is due to its black-box makeup, results in confusion. As Davida Charney points out (about non-fictional hypertext),

First, since readers must choose what to look at, they may never see all the "right" information, either because they cannot find it or because, for some reason, they fail to select it. Second, even if they do see the "right" information, they may see it at the wrong time. ... Third, readers may see a great deal of intrusive, irrelevant information that may skew their representations. Even if they recognize that some information they have read is irrelevant, there may still be adverse consequences of having spent time reading it. Finally, readers may lose a sense of the integrity of any given text in the network, since they may be unaware of crossing from one text to another. ... In short, in addition to suffering the frustrations of disorientation or cognitive overload that hypertext designers already acknowledge, readers may come away with a false or incomplete representation of the texts in the network or even the information relevant to their topics.(Charney 249f)

The same holds true for fictional hypertext: there is a danger of failure to see important text, failure to read it at the right time, intrusion of irrelevant text, and loss of integrity.

For the author, the sacrifice of choice is not really a sacrifice at all-- he is still in control over all the text and the links. However, he is responsible for constructing the black box, the hyperstructure that will generate a multitude of stories at the reader's discretion. Therefore, the writer might feel the need to come up with more links and nodes than are necessary. As Robert Coover puts it,

The structuring of the space can be so compelling and confusing as to utterly absorb and neutralize the narrator and to exhaust the reader....If everything is middle, how do you know when you are done, either as reader or writer? If the author is free to take a story anywhere at any time and in as many directions as she or he wishes, does that not become the obligation to do so? (Coover 24)

But even if the author can resist this urge to overachieve because he can (and fall into the "poles-in-your-face" trap outlined above), there will be several nodes, and this means some will be superior to others. Some nodes might have been fitted in because the structure dictated them, not narrative need. Some nodes might be absolutely essential to any reading of the story that is to be meaningful. Ultimately, every node, every piece of text, will fall into one of two groups: Either it is good and interesting and important to the story, in which case it should be required and necessary that the reader read it, or it is a filler, inferior, meaningless, and pointless, in which case the author should make sure no one ever gets to see it. The choice between the two should be made by the author, who created the fiction in the first place and is qualified to make the selection.

Art, as an artifact, a thing in the world, is the result of a number of creative choices that were made to construct it. These choices are the willful, purposeful expressions of the creative mind. If the artist, in this case the writer, refrains from making some of these choices and leaves them up to the reader, his work is not empowering and democratic, but incomplete. To create also means to decide what the artefact is not, to rule out possibilities. By including all possible forms, the work is deprived of an actual form and remains a hybrid.

Anybody who has the resources to read a hyperfiction is not in need to be empowered to create, but surely has his own word processor to write for himself. Empowerment through another's refusal to create is unnecessary, futile, and unsatisfactory. A reader chooses his role purposefully, and if he feels inhibited as such, he can launch a different application or pull out pen and paper and become a writer.

The cooperation, then, between the writer and the reader of a hyperfiction is not that of two collaborating artists, it is that between a game designer and a player. Hyperfiction is ultimately a game, and its proponents know that:

Playfulness is a defining quality of this new medium. Electronic literature will remain a game, just as all computer programming is a game .... In video games, the kind depicting spacecraft and deadly robots, the player competes against the programmer, who has defined goals and put obstacles in the player's path .... No matter how competitive, the experience of reading in the electronic medium remains a game, rather than combat, in the sense that it has no finality. The reader may win one day and lose the next .... The impermanence of electronic literature cuts both ways: as there is no lasting success, there is also no failure that needs to last. By contrast, there is a solemnity at the center of printed literature--even comedy, romance, and satire--because of the immutability of the printed page. (Bolter 130)

Of course, there is and should be room for games and playfulness, but this irrevocable, intrinsic aspect of hyperfiction must be acknowledged if we are to understand its nature.

One feature of hypertext that seems inaccessible to "flat" text is its meta-structure. We saw earlier that it cannot be discerned from a single reading, which will only yield a linear text that is structured in some way. However, repeated reading of the hypertext will, or can, reveal a structure of a higher order: The meta-structure of the linear narrative, the structure of the hypertext. This meta-structure does not exist in linear fiction which only consists of one linear narrative.

Structure is meaningful because it is ordered, and thus information. A lack of structure results in an absence of meaning. Therefore, it is unclear to me what Fowler is cheering about when he argues that "Hypertext has no beginning or ending, no center or margin, no inside or outside" (Fowler 18). If the hyperstructure is to be used as an element of the makeup of hyperfiction then it has to be employed in a meaningful way--the sum of linear readings should point to a meta-structure that communicates. Bolter examines what "Afternoon"'s meta-structure signifies: "The need of the reader to struggle with the story mirrors the struggle that the character goes through. 'Afternoon' becomes an allegory of the act of reading." (126). It seems, then, that "Afternoon" is ultimately about the act of reading hypertext, it is about hypertext, about itself. If that is all the medium invites writers to talk about, then its promises are bound to be exhausted shortly.

However, linear narrative is capable of simulating a meta-structure by placing different narratives in sequential order, very much like the multiple narrative a hypertext structure produces are in a sequential order. Krzysztof Kieslowski's movie Blind Chance is an excellent example for the exploration of alternate narratives in a linear medium. The three versions of Witek's life produce a meaning beyond what any of them could have said by themselves, through reference to a superior structure. As a matter of fact, it is notable that the movie is not hypermedia, and rightfully so: to appreciate the meaning of the meta-structure, the audience has to see all three narratives, and it is Kieslowski, not the audience, who is best equipped to choose the order of their presentation. There is something to be gained, some meaning to be wrested, from the meta-structure of several linear texts, but there seem to be no advantages, only drawbacks, in using hypertext for this kind of work.

In summary, analysis seems to suggest that as far as traditional fiction goes, hypertext is burdened with too many problems and no advantages. According to John Gardner, "if the effect of the [narrative] is to be powerful, [it] must probably be vivid and continuous" (31). By forsaking control over order and selection of his text, by sacrificing its continuity, the author is in danger of giving up whatever narrative thrust his work had. Theoretically, the hyperstructure could be pieced together so elegantly and perfectly as to always produce a satisfying linear story by avoiding the pitfalls I tried to sketch. However, it is not clear what would be gained even by such a "perfect" hypertext version over its linear counterpart. There simply does not seem to be a good reason to tackle the problems that hypertext fiction is burdened with. While it might offer great opportunities for playful interactions (in the form of interactive games, constructive hypertexts, or MUDs), it seems to be an artistic dead end as far as narrative is concerned. Again, the analogy of the 3-D movie might serve well: once the director of such a movie has understood that no amount of wielded poles or things thrown at the camera and prospective audience will improve the narrative qualities of the movie, he will have to recur to the time-honored means of cinematography, use them, abuse them, enter a working relationship with them. But then, those red-and-green glasses won't be necessary anymore to enjoy the movie.


Barnes, Julian. Talking It Over. New York: Vintage, 1991.

Baudrillard, Jean. Simulations. 1976.

Blade Runner. Dir. Ridley Scott. Warner Brothers, 1982.

Blind Chance. Dir. Krzysztof Kieslowski. Poland 1981.

Bolter, J. David. Writing Space: The Computer, Hypertext, and the History of Writing. Hillsdale, NJ: Erlbaum, 1991.

Borges, Jorge Luis. Labyrinths: Selected Stories & Other Writings. New York: New Directions, 1964.

Charney, Davida. "The Effect of Hypertext on Processes of Reading and Writing." Literacy and Computers: The Complications of Teaching and Learning with Technology. Cynthia L. Selfe and Susan Hilligoss, eds. MLA: New York, 1994. 238-263.

Coover, Robert. "The End of Books." New York Times Book Review. June 21, 1992. 1, 2224.

Fowler, Robert M. "How the Secondary Orality of the Electronic Age Can Awaken Us To the Primary Orality of Antiquity Or What Hypertext Can Teach Us About the Bible." Interpersonal Computing and Technology: An Electronic Journal for the 21st Century. July 1994. 12-46.

Gardner, John. The Art of Fiction: Notes on Craft for Young Writers. New York: Knopf, 1984.

Johnson-Eilola, Johndan. "Reading and Writing in Hypertext: Vertigo and Euphoria." Literacy and Computers: The Complications of Teaching and Learning with Technology. Cynthia L. Selfe and Susan Hilligoss, eds. MLA: New York, 1994. 195-219.

Joyce, Michael. Afternoon, a story. Cambridge, MA: Eastgate, 1987. Computer program.

The Killers. Dir. Robert Siodmak. UI, 1946.

Lanham, Richard A. The Electronic Word: Democracy, Technology, and the Arts. Chicago: U of Chicago Press, 1993.

Moulthrop, Stuart. "Traveling in the Breakdown Lane: A Principle of Resistance for Hypertext." HTML-Document.

Nelson, Theodor H. Literary Machines. Mindful Press, 1990.

Ong, Walter J. Orality and Literacy: The Technologizing of the Word. London and New York: Methuen, 1982.

Reeth, Garret. "Tree Fiction on the World Wide Web." HTML-Document. treefiction.html.

Copyright © 1995 Blip Magazine Archive, The Center for Writers, The University of Southern Mississippi

Maintained by Blip Magazine Archive at

Copyright 1995-2011
Opinions are those of the authors.