Hypertext and Creative Writing
Jay David Bolter Michael Joyce
University of North Carolina,
CB# 3145 Murphey Hall
Chapel Hill, North Carolina 27599-3145
ABSTRACT
Among its many uses, hypertext can serve as a medium for a new kind of flexible, interactive fiction. Storyspace TMis a hypertext sys tern we have created for authoring and reading such fiction, Interactive fiction in the computer medium is a continuation of the modern “tradition” of experimental literature in print. However, the computer frees both author and reader from restrictions imposed by the printed medium and therefore allows new experiments in literary structure.
THE IDEA OF INTERACTIVE FICTION
The idea of hyptertext, which seemedd aring only a few years, is now emerging as a serious and sensible way to use the computer for reading and writing. Technical writing and pedagogy (interactive communication between teachers and students) are obvious and important applications for hypertext systems. But hypertext may in fact apply to the whole range of human literacy, including the writing and reading of fiction. Using hypertext as a vehicle for fiction is both more and less daring than using it for technical writing or education. It is more daring because fiction seems frivolous in the pragmatic world of data processing. It is less daring because fiction, at least modem fiction, is by nature open to experiment, and being open or open-ended is precisely the quality that hypertext fosters in writing. The point of a hypertext is that it can change for each reader and for each act of reading. This flexibility can be exploited to make fiction interactive.
Interactive fiction has already existed for some time in the form of computerized adventure games. In an adventure game the player has a mythical world to explore—a dungeon or an enchanted forest or valley. The computer describes the scene, and the player issues simple commands such as “go ahead”, enter the room”, “pick up the dagger”, “get gold,” and the like. The goal is to amass treasure and dispatch monsters, although sometimes the game is more sophisticated, casting the player in the role of a detective who must solve a murder or other mystery. Even the simplest of these games is a fictional hypertext. For the computer is presenting the player with a text, and the player’s job is to understand and respond to that text. Depending upon his response, the computer presents more text and awaits a further response. The player, then, is an unusually powerful reader, whose decisions determine what text he will next see. Admittedly the text of the current games is simple-minded, but the method of presentation is not.
This method of presentation can now be applied to serious fiction. A printed novel presents its episodes in one order, but the computer removes that restriction. Instead of a single string of paragraphs, the author lays out a textual space within which his fiction operates.The reader joins in actively constructing the text by selecting a particular order of episodes at the time of reading. Within each episode, the reader is still compelled to read what the author has written. But the movement between episodes is determined by the responses of the reader, his interactions with or intrusions into the text, and the reader’s experience of the fiction depends upon these interactions.
In its simplest form, interactive fiction requires only two elements: episodes and decision points (links) between episodes. The episodes may be paragraphs of prose or poetry, they may include graphic designs or pictures as well, and they may be of any length. Their length will establish the rhythm of the story—how long the reader remains a conventional reader before he is called on to participate in the selection of the next episode. At the end of each episode, the author inserts his decision points—a set of links to other episodes together with a procedure for choosing which link to follow. Each link may require a different response from the reader or a different condition in the computer system The reader may answer a question posed in the text, and there will be one link for each possible response. The computer can also keep track of the previous episodes the reader has visited, so that he may be barred from visiting one episode before he visits another. Artificial intelligence experts would not consider such a simple scheme for interactive fiction worth pursuing. They would argue that we have to store knowledge representations in the computer and write a program that can generate new sentences in response to the reader’s replies. In other words, the program itself would be the author, not simply the medium for delivering what the human author has written. While this AI strategy is interesting, it is not feasible at present or in the near future. No AI specialists can tell us how to store a world of knowledge in the computer: nor can their programs generate sophisticated English prose in response to queries by human users. For the foreseeable future, interactive fiction can only be a hypertext of prose written by human beings. There is in any case no need to wait for such breakthroughs in artificial intelligence. Even with the simple matching technique and the tracking of previously visited episodes, the author can create a fictional space of great flexibility.
Such electronic fiction is not automatic fiction. Since the computer does not create the verbal text, the locus of creativity remains with the author and the reader. Nor is electronic fiction necessarily random, for the author may put any number of restrictions on the reading order. The extent of the reader’s choices and therefore his freedom in examining the literary space depend upon the links that the author creates between episodes. The reader may have to choose among a few alternatives or may range widely through the work. The author can relinquish as much or as little control as he chooses; he has a new literary dimension with which to work
STORYSPACE
In collaboration with Professor John B. Smith, we have created a system for interactive fiction called Storyspace™, which implements the scheme of episodes and links mentioned above. (The system is by no means limited to fiction, although fiction will be the focus of the discussion here.) Storyspace has two modes: one for the author and one for the reader. The author creates his fiction as a series of textual episodes, using what we call a structural editor. This editor gives him a graphic or diagrammatic view of the hypertext he is creating. The reader has a different and more limited view: he sees the contents of each episode and may then reply by typing a string or pressing a button in order to branch to the next episode. Storyspace is implemented on the Apple Macintosh computer. This machine provides good graphics for the editing process. It is also popular and inexpensive, so that fictions created with Storyspace can be widely distributed,
The author constructs his fiction as a network of units. In the editor, the units appear as boxes on the screen. The author manipulates the structure by moving, adding, or deleting these boxes. He can also open each box to type in the text of his fiction. Each episode may be as long or as short as the author wishes: one word or several paragraphs. The end of each episode should, however, indicate how the reader is to respond: whether he should answer a question, make a choice, or simply hit return.
To indicate the relationships among these episodes, the author draws links, which appear as arrows on the screen. These links indicate possible orders of reading for the episodes, in that each arrow points to a possible next episode.
Each link carries with it a condition statement (specified by the author), which must be satisfied in order for that link to be followed. At present Storyspace recognizes two kinds of conditions. The link may require that the reader match a string (answer a question) before proceeding. The link may also require that the reader already have visited a particular episode before proceeding. The author can also specify Boolean combinations of these conditions.
While the author sees and manipulates a diagram of the evolving structure of his fiction, the reader sees only the text, not the structure of connecting links. The reader begins in one episode designated by the author as the starting place. He reads the text of that episode and may respond by typing text or pressing an appropriate button. The system processes the reader’s reply simply by checking all the links that lead out of the current episode. It checks these links in the order of creation and takes the first one whose condition is satisfied. It then displays the text of the destination episode and awaits a further reader response.
INTERACTIVE FICTION AND THE EXPERIMENTAL TRADITION
Storyspace, then, is a simple system for hypertextual fiction, but even this system permits significant structural experiments. Its use of episodic branching challenges many conventional ideas about literature. If the reader is allowed to choose his path through the narrative, then the stability and certainty inherent in a printed text disappear. There may no longer be one plot, but several, and characters may no longer develop in a consistent fashion. The structure and rhythm of the text will be different for each reading. Every element of fiction is subject to electronic fragmentation and recombination. At the same time, by disrupting the stability of the text, interactive fiction belongs in the tradition of experimental literature (if I may use this oxymoron) that has marked the twentieth century—the era of modernism, futurism, Dada, surrealism, letterism, the nouveau roman, concrete poetry, and other movements of greater or lesser influence. The experiments of Dada, for example, were aimed at breaking down all structures of established art and literature, and in that breakdown some of the Dadists worked in the same spirit as writers and readers may now work in the electronic medium. JeanArp wrote that in his poems: “I tore apart sentences,words, syllables. I tried to break down the language into atoms, in order to approach the creative.” [Gros71, p.136] Tristan Tzara proposed a poetics of destruction, when he gave this advice for creating a Dada poem: “To make a dadist poem. Take a newspaper. Take a pair of scissors. Choose an article as long as you are planning to make your poem. Cut out the article. Then cut out each of the words that make up this article and put them in a bag. Shake it gently. Then take out the scraps one after the other in the order in which they left the bag. Copy conscientiously. The poem will be like you...” [Gros71, p. 125]
Dada is an early and influential example of the modem will to experiment. The modern attack has often been aimed at the conventions of the realistic novel, the nineteenth-century novel that told a story with a clear and cogent rhythm of events, and in the course of their attack modern authors have often found themselves straining at the conventions and limitations of the printed page. Because the linear-hierarchical presentation of the printed book was so well suited to the conventions of plot and characters of the realistic novel, to attack the form of the novel was also to attack the technology of print that helped to shape that form. The French often led the way with the nouveau roman and Philip Sollers and the Tel Quel group. From France and elsewhere, we have had programmed novels and aleatory novels. All these efforts were instances of subversion: they worked from within, attempting to undercut the conventions of printed literature while themselves remaining printed books. Subversion is an effective mode of attack precisely because of this irony—because in this way the printed novel is made to contain the seeds of its own destruction, or perhaps deconstruction.
Indeed, much important twentieth century literature may be, and has been, accused of subversion. The avant-garde movements like Dada were never so radical as they claimed to be; they were instead extensions or perhaps caricatures of the mainstream. Joyce, Virginia Woolf, Pound, Eliot, and others all participated in the breakdown of traditions of narrative prose and poetry; breaking with such traditions was the definition of being modem. Pound and Eliot set about to replace the narrative element in poetry with fragmented anecdotes or mythical paradigms. Joyce and Woolf called into question the strategy of the novel as a linear and objective narrative. They devised new ways of structuring their works based upon stream of consciousness (Woolf) or upon multiple layers of topical and mythical organization (Joyce). All of these writers were trying to set up new relationships between the moment-by-moment experience of reading a text and our perception of the organizing and controlling structures of the text. In this sense, hypertextual fiction is a natural extension of their work, redefining the tradition of modernism for a new medium.
BORGES
One contemporary writer whose work is very suggestive of interactive fiction is the Argentinian Jorge Luis Borges, whose short stories called Ficciones are a series of meditations upon writing or more broadly upon the human capacity to create and comprehend symbols. Borges writes tiny pieces without much plot or characterization—pieces that are utterly insignificant by the standards of the nineteenth-century novel. Often these pieces concern the problem of time and writing: Borges is intrigued by the fact that a frozen text cannot change to reflect possibilities that unfold in time. “An Examination of the Work of Herbert Quain” is the literary obituary of a writer who tried to liberate his texts from linear reading and static interpretation. Quain’s work April March is nothing less than an interactive fiction. It consists of thirteen chapters or sections representing nine permutations of the events of three evenings.The novel is therefore nine novels in one, each with a different tone. Borges tells us the work is a game. He adds that “[w]hoever reads the sections in chronological order...will lose the peculiar savor of this strange book.” [Borg62, p. 76] He even gives us a tree diagram of the ternary structure of thework, and another diagram of the binary structure that Quain later says he should have written.
Borges’ longer and more elaborate “Garden of Forking Paths”is a detective story. At its center the story contains a description of a Chinese novel, a novel that seeks to explain and in its way to defy time. It was thought that the author Ts’ui Pên had retired from public life with two objects: to write a book and to build a labyrinthine garden. In fact Ts’ui Pên had only one goal, for the book was the labyrinth. The manuscript Ts’ui Pên left behind was not, as it seemed, “a shapeless mass of contradictory rough drafts,” [Borg62, p. 96] but instead a ramifying tree of all possible events. “The Garden of Forking Paths is an enormous game, or parable, in which the subject is time.” [Borg62, p,99] Ts’ui Pên “believed in an infinite series of times, in a dizzily growing, ever spreading network of diverging, converging and parallel times.” [Borg62, p. 100]. In the end “The Gardenof Folking Paths” is no more serious than any of Borges’ fantasies. Borges’ point is not to sketch a new philosophy of time, but a critique of writing itself—to imagine a book that calls into question the finite, fixed character of writing or print. Of course Borges’ fiction is also a book, as he realizes, and therefore subject to the same limitations that he is describing. His understanding of these limitations may explain why Borges himself only envisions experiments rather than undertaking them. The Ficciones are themselves conventional pieces of prose, essays or stories meant to be read page by page. Yet the works he describes, the novels of Quain or the “Garden of Forking Paths,” belong in another writing space altogether. Borges never had available to him an electronic writing space, in which the text can constitute a network of diverging, converging and parallel times. The literature of exhaustion in print by no means exhausts the electronic medium. In fact, a number of Borges’ pieces suggest themselves for translation into the computer’s writing space. The“Garden of Forking Paths” has been converted into an interactive fiction by Stuart Moulthrop of Yale University. This Storyspace module contains a web of over 100 units and 300 connections. Moulthrop has added his own meditions to those of the Borges’ story.
THE STRUCTURE OF INTERACTIVE FICTION
Like Borges, many experimental writers have concerned themselves with the problem of writing or printing: their concern is shown by the difficult relationship between the narrator and text and between the text and its reader. Hypertextual fiction has much in common with these experiments in print. It too will have to introduce new procedures of reading that violate the reader’s expectations of a linear narrative. (Perhaps the first indication that electronic fiction has matured beyond the adventure game will be the appearance of fictions that are about writing, about capturing experience in writing, and therefore about themselves.) Electronic writers therefore need conventions, genres, traditions by which their medium can be governed. They must find new ways to maintain a tension between the reader and the text. The source of that tension will surely be the participation of the reader in making the text. In electronic fiction, the struggle between author and reader to appropriate the writing space can become visible, as the reader admits or tries to avoid admitting particular elements into his particular reading of the text. Moreover, electronic authors will need a new concept of structure. The structure of an electronic fiction will change with each reading, because the order and number of episodes will change. Authors must therefore learn to conceive of their text as a structure of structures, and this is a concept that is new in the history of literature.
The temporal character of interactive fiction is also something new. In printed fiction the author is free to manipulate the time in which his story takes place, and every good author does so. However, the plot, the author’s manipulation of story time, is itself static. Printed fiction is one-dimensional in the sense that we need only one dimension in order to represent the experience of reading it. The episodes (chapters, sections, cantos, books, volumes) are realized through time as we read. The links between the episodes are fixed in the course of writing, and the reader has no obvious and effective way to alter the order of reading. In electronic fiction multiple links among episodes allow our temporal experience of the plot itself to vary. Time may be fluid in a printed novel, but the presentation of time is fixed, as the fixed pages of the book mark the progressive stages of the narrative. The author manipulates words to create a single narrative structure. The author writes with words, not with structures. The electronic medium permits writing of a second order, a writing with narrative units, in which the structure of the text becomes truly fluid and indeed geometric. The author becomes a geometrician or architect of computerized “space” (as computer memory is in fact called by programmers); he fills his space with a special pattern of episodes and links that define a kaleidoscope of possible structures. The success of his work will depend upon the poetic rightness of the way in which the pattern is realized in the act of reading.
Storyspace provides the author with the opportunity to see the structure of his tale and therefore to use the structural geometry as an aesthetic principle. Because the reader doe snot see the diagrammed structure of the text, he is left to gain an intuitive sense of the structure by reading the episodes themselves. He might have to read the tale many times to understand a structure that changes, in a controlled fashion, with each reading. Such a reader is like a mathematician who attempts to envision a four-dimensional object by looking at several projections in three dimensions: each projection is a snapshot, and all the snapshots must be synthesized to win a sense of the whole, if indeed such a sense is possible. The synthesis of many readings will be the ultimate experience that the electronic writing offers its readers.
Multiple reading does not necessarily mean multiple plots, although the most obvious way to construct a hypertextual fiction is by presenting the reader with choices that affect the plot. Hypertext could borrow from modern literature the techniques of stream of consciousness or multiple points of view as methods of organization. An electronic author could for example, create rings of episodes representing the same events as told by several characters. The technique of telling the same events from different points of view is familiar from Faulkner’s The Sound and the Fury and Darrell’s Alexandria Quartet. In exploring these multiple points of view, the reader of a hypertext can enjoy far more freedom than the reader of a printed text. He can move back and forth through the episodes, comparing one narrator’s version with another. Even visiting the same episode twice is not mere repetition, because his experience of the narrative will be affected by the other episodes he has recently read.
Here is an example of a narrative hierarchy composed from the ancient myth of Oedipus, as elaborated by the Greek playwright Sophocles. The three levels correspond to three points of view from which the story might be told. The lowest and most confused version takes the perspective of the shepherd who gave the baby Oedipus away rather than obey the order to kill him Years later, this kindly shepherd is brought before Oedipus the king, threatened with torture, and made to reveal what he did with the baby. Then he is released and left to wonder at the extent of the horror that he has revealed. Oedipus himself provides an heroic perspective; he is of course most directly affected by the horror of the events. The third and clearest perspective is the divine view of these view: the story as the god Apollo might tell it. This third perspective is one that Sophocles himself would never have permitted, for Sophocles’ gods, unlike those of Euripides, had to remain dignified and remote, their ways inexplicable and terrifying.
The reader begins at the shepherd’s level, and his task is to break into the higher levels representing greater understand of the events. The divine level is of course harder to reach than the heroic. It is also a detached view the reader himself may find repellent. Unlike the electronic version of Dungeons and Dragons, the electronic Oedipus is a game in which winning is not clearly defined. (One of us, Jay David Bolter, has discussed the Oedipus example, other possible geometries, and the idea of interactive fiction as a game in greater detail elsewhere [Bolt 85]).
All electronic literature takes the form of a game, a contest between author and reader. Unlike the static and monumental character of printed fiction, hypertextual fiction is characterized by impermanence and a lack of monumentality. A playful attitude prevails, as it should in any computing task. An interactive fiction is, after all, a program that the author creates and the reader executes, and any computer program can go unexpectedly and ridiculously off track. Fortunately, with most programs, it is possible to restart the system and try again. The reader of an interactive fiction can return to the starting episode and take another path. This flexibility of interactive fiction stands in sharp contrast to the solemn rigidity of the printed book, which cannot change in response to its readers’ previous experience. No matter how often the reader returns to the first chapter of a printed book, he still has only one path to choose. Printed fiction is proud of its rigidity, and hypertextual fiction, which cannot rival this quality of the printed book, must therefore draw its aesthetic strength from its capacity for change.
[Storyspace is available to creative writers on an experimental basis. One of us (novelist Michael Joyce) is currently working on interactive fiction using this system, and his first effort, “Afternoon, a Story” is available to interested readers. The same methods of editing and presentation can be used for nonfiction and particularly for teaching. A group working at the University of San Diego under the direction of Professor Bart Thurber is presently considering an interactive course using Storyspace as an interdisciplinary introduction to the humanities.]
REFERENCES
[Bolt85] Bolter, Jay David. “The Idea of Literature in the Electronic Age,” Topic: A Journal of the Liberal Arts. 39 (Fall, 1985), 23-34.
[Borg62] Borges, Jorge Luis. Ficciones, edited with an introduction by Anthony Kerrigan. New York: Grove Press, 1962.
[Gros71] Grossman, Manuel L. Dada: Paradox, Mystification. and Ambiguity in European Literature. New York: Bobbs-Merrill, 1971.
Permission to copy without fee all or part of this material is granted provided that the copies are not made or distributed for direct commercial advantage, the ACM copyright notice and the title of the publication and its date appear, and notice is given that copying is by permission of the Association for Computing Machinery. To copy otherwise, or to republish, requires a fee and/or specific permission.
Hypertext 87 Chapel Hill NC USA
Copyright 1989 ACM 089791-340-X/89/001 1/0050$1.50 50
[Source: Bolter, J. D. and Joyce, M. 1987. Hypertext and Creative Writing. HYPERTEXT ’87. Proceedings of the ACM Conference on Hypertext, 41–50. https://doi.org/10.1145/317426.317431. ]
Powered by Mintter + The Hypermedia Protocol