Determining Literariness In Interactive Fiction Nell RandaU Department of English, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada
Abstract: The authors of interactive fiction are beginning to demonstrate a concern for the literariness of their product. Literariness, as defined by Shklovskij and the Russian Formalists, is the quality of "making strange" that which is linguistically familiar, a quality Shklovskij termed ostranenie. By applying the principle of ostranenie, as well as other wellknown literary principles, to the most serious interactive fictions, we can determine if this new genre exhibits the features of literariness. A study of Mindwheel, Brimstone, Breakers, A Mind Forever Voyaging, Portal, and Trinity suggest that the literariness of interactive fiction comes out of its concern both for "making strange" what is familiar and for "making familiar" what is strange.
In an article entitled "Indeterminacy and the Reader's Response", Wolfgang Iser asks: What is it that makes the reader want to share in the adventures of literature? This question is perhaps more for the anthropologist than for the literary critic, but the fact is clear that people have always tended to enjoy taking part in the fictitious dangers of the literary world; they like to leave their own security and enter into realms of thought and behavior which are by no means always elevating. Literature simulates life, not in order to portray it, but in order to allow the reader to share in it. He can step out of his own world and get into another, where he can experience extremes of pleasure and pain without being involved in any consequences whatsoever. It is this lack of consequence that enables him to experience things that would be otherwise inaccessible owing to the pressing demands of everyday reality. (1971, p. 45)
While Iser, as evidenced by his examination of Fielding's Joseph Andrews, Thackeray's Vanity
Neil Randall has published several role-playing games, and is working on a full-length critical study of interactive fiction. He is currently assistant professor of English.
Fair, and Joyce's Ulysses (pp. 23f.), is interested in the reader's participation in the traditional1 literary text, his discussion of reader involvement in serialized novels (pp. 14f.) displays a concern for other presentations of fiction. In the past decade, the entertainment software industry has developed "interactive fiction", a new presentation of fiction that uses the computer to act out the commands of the reader of the text (Barol, 1985). While the degree to which the reader controls the interactive text is debatable (Niesz, 1984, p. 123), the basis of this new genre is clearly, as Iser says, "that people have always tended to enjoy taking part in the fictitious dangers of the literary world". Interactive fiction indisputably fulfills the requirement of participation, as the cult following of Infocom's adventures suggests (Barol, 1985); what needs to be determined, though, is whether or not the world presented in these stories is a "literary" one. As interactive fiction grows more complex and more serious, its authors themselves have begun to ask that question. For the purposes of this article, I am considering only computerized, commercially-available, published interactive fiction, of the kind computer hobbyists still term "text adventures" (see Appendix). The reader of these works takes the role of the story's central character, responding to an onscreen textual description by typing an order for that character to follow. The most primitive kind of order consists of two words, such as "Go North", after which the computer processes the order and replies with a new description. Currently, the most complex commands allow the reader to address a secondary character, exclude specific objects, or co-ordinate two separate orders. The instruction manual for Steve Meretzky's A Mind Forever Voyaging (1985), for example,
lists "OLD MAN, TAKE THE SACK THEN FOLLOW ME" and "TAKE THE BOX THEN OPEN IT. PUT THE PELLET IN THE BOX. CLOSE IT" (pp. 16--17) as possible commands. Eventually, the reader completes the text by fulfilling her character's quest in the story. Finishing a work of interactive fiction can take as long as one hundred hours, and the challenge, or the narrative suspense, lies in trying to get past the next hurdle in the story. In early interactive fiction, the hurdles themselves provided the basis for the quest; to solve the quest the reader was forced to solve puzzle after frustrating puzzle. Recent works, though, such as those chosen for this article, do not usually allow a hurdle to stand between the reader and the completion of the text, because to do so would be to deny the traditional reading experience, the anticipation of the final page. As interactive fiction turns away from its origins in gaming and tries to attain some degree of literary stature, it has begun a process of considering the role of the new technology in the traditional relationships between the reader ("literant" in Norman Holland's terminology) and the literary text. Other kinds of interactive fiction exist, including the non-computerized Choose-Your-OwnAdventure novels originated by Bantam Books and the computerized, graphics-oriented adventures perhaps epitomized by Ultima IV (British, 1985), the most sophisticated installment of a series of fantasy role-playing games for the computer. Arguably, literariness can also be attributed to the non-computerized role-playing games of which Dungeons and Dragons (Gygax, 1978) is pre-eminent, especially when considered from a narratological point of view, but the group interactivity demanded by such games differs markedly from the exclusive computer-reader interactivity experienced in the text adventure. To limit interactive fiction to the computerized text adventure is to ignore several increasingly sophisticated attempts at interactive story-making, but the text adventure format, at least until now, has offered the greatest literary possibilities within this rather eclectic fictional genre. 2 I am also going to disqualify interactive adaptations of existing novels, but only because the number of original interactive works has grown to the point where we can safely ignore adaptations.
Interactive adaptations abound, re-creating such works as Roger Zelazny's Nine Princes in Amber. Michael Ende's The Neverending Story. J. R. R. Tolkien's The Hobbit, and even William Shakespeare's Macbeth. Adaptations have the advantage of instant literariness, at least if the original can be considered literary, but since they have so far adapted only by changing plotlines, and occasionally characters, they have not fully crossed the boundary from one art form to another. An adaptation of a novel into a motion picture (The Name of the Rose) or of a poem sequence into a Broadway musical (Cats) is a change of art form; interactive fiction has not yet defined itself into an art form, and thus has effected no significant adaptations. A new series of non-computerized interactive novels based on existing fictional worlds, but offering wholly new stories and characters -- a series which includes my own Storm of Dust (fromDavid Drake's The Dragon Lord) and Seven No-Trump (from Roger Zelazny's Chronicles of Amber) -- offers a further attempt at interactive literariness, but I have already decided against non-computerized interactive fiction. One especially interesting adaptation is Douglas Adams and Steve Meretzky's The Hitchhiker's Guide to the Galaxy (1984), a re-plotting of Adams' original novel. However, I exclude this work because its puzzle-solving orientation misses the sophistication of the most recent efforts in the genre. Canonization and Literariness When Terry Eagleton writes that the literary canon has to be recognized as "a construct, fashioned by particular people for particular reasons at particular times" (Eagleton, 1983, p. 11), he is saying, of course, that the composition of the canon is the result of "value-judgements" that are "historically variable" and which "have a close relation to social ideologies" (p. 16). But even if we object to Eagleton's ideological interpretation of the canon, we can scarcely object to his assertion that the canon is an historical construct, and that it is subject to reconstruction as times, tastes, and values change. We need only look, of course, at the fluctuations in the critical fortunes of Donne, Spenser, and Hemingway to see how quickly (and decisively) the canon alters itself. Furthermore, despite the trend in post-structuralist thinking
L I T E R A R I N E S S IN I N T E R A C T I V E
towards the decanonization of literature, a trend that Ihab Hassan (1986) has recently noted, for pedagogical and ideological reasons a literary canon still exerts authority. Whether snobbishly or philosophically, we still place King Lear above the Compuserve User's Manual on the hierarchy of works of literary value. For Viktor Shklovskij, canonization had its own law, the "canonization of the junior branch" (1923, p. 227). In this "law", which Shklovskij claimed to have discovered, literary works move into the canon when canonized art reaches some sort of impasse with itself; the impasse is broken through, or moved, only with the changes caused by the appearance of the new works. Through this procedure, suggests Victor Erlich, "[p]roducts of popular culture, leading a precarious existence on the periphery of literature, are thus admitted into the parlor, raised to the status of bona fide literary art" (1955, p. 227). Shklovskij's model misses, of course, the canonization of previously neglected but not necessarily "minor" works (as in the cases of the canonization of Donne by the New Critics and Blake by the archetypalists), but more seriously it says nothing of how new forms, those considered neither dignified nor popular, proceed onto the canon. It does not explain, in this respect, the canonization of the avant-garde. For the Russian Formalists, though, canonization was only one concern, and not per s e a particularly large one. More important by far was the notion of what Roman Jakobson termed literaturnost, usually translated as "literariness". To examine literariness is to find "the distinguishing element of literature itself" (Jameson, 1972, p. 43). For the Formalists, this meant determining how literary language differs from everyday language, a distinction expanded by Wellek and Warren (1956) to include scientific language as well. Fredric Jameson explains: The Formalists began by demonstrating that in many ways poetic speech stood to everyday language as a type of dialect, governed by its own peculiar laws, indeed often even pronounced differently. . . . The deeper implication is that poetry is not merely a specialized part of everyday language, but constitutes a total linguistic system in its own right (1972, p. 49).
Furthermore, Jameson goes on to say, literary language "is one which attracts attention to itself, and such attention results in renewed perception
of the very material quality of language itself" (p. 50). What this comes down to, for Shklovskij at least, is that the key feature of literary language is its "making strange" our perception not only of what the language is referring to, but also our perception of the language itself. This essential feature, called ostranenie, is the means by which literary art renders new that which has become familiar, mechanical, unable to impress us other than in a functional way. As Terence Hawkes puts it, for Shklovskij the aim of poetry is "to defamiliarize that with which we are overly familiar, to 'creatively deform' the usual, the normal, and so to inculcate a new, childlike, non-jaded vision in us, . . . so that we end by seeing the world instead of numbly recognizing it" (1977, p. 62). For Roman Jakobson, the "poetic function" of language "projects the principle of equivalence from the axis of selection into the axis of combination" (1960, p. 358). Jakobson sees two major dimensions in language, the metaphoric and the metonymic, where the metaphoric is also the selective and the synchronic, and the metonymic is the combinative and the diachronic. Poetic language is the superimposition of the first onto the second dimension. Jakobson realizes this function as part of his diagram of the communicative act, which shows the link between addresser and addressee consisting of four interrelated features: context, message, contact and code. When the transaction centres not on context, code, or contact, but rather on the message itself, then we have a poetic utterance. Robert Scholes goes a step further, suggesting that "we sense literariness in an utterance when one of the six features of communication [i.e. sender, receiver, context, message, contact, code] loses its simplicity and becomes multiple or duplicitous" (1982, p. 21), concluding that the doubling of context produces "the beginning of the kind of literariness characteristic of fiction" (p. 27). It is not far from here to R. A. York's insistence that "a pragmatic account of literature will tend to stress . . . the utterance rather than the content, the act of enonciation rather than the enonce" (1986, p. 10). Nor is it far to Shklovskij's ostranenie. Ostranenie, in fact, provides the basis for much current theorizing about literature. If we examine Hassan's "catena of post-modernist features" (1986, p. 504), we find, among others: Inde-
terminacy; Fragmentation; Decanonization; Selfless-ness, Depth-less-ness; The Unpresentable, Unrepresentable; Irony; Hybridization; and Carnivalization. What marks the catena, surely, is precisely its "making strange" the critical act itself, an ostranenie that many students, coming upon such concepts for the first time, find utterly objectionable. Traditional criticism grounds itself in the opposites of Hassan's catena -- determinacy, wholeness and unity, canonicity, representation, etc. -- and post-modernism almost effortlessly destroys them. If Shklovskij is correct in asserting that literary language hangs on the very notion of ostranenie, then the post-structuralist position that the criticism of a literary work is part of the work itself is provable. Making strange a literary work is a literary act all on its own. But surely ostranenie does not fully explain what we mean by "literary". Do not a text's literary features include such well-known, un-strange concepts as allusion, extratextual and intertextual referentiality, the expression of ideas, the portrayal of characters, the making of worlds, and any number of themes, symbols, and images? Perhaps each of these concepts is a type of making strange, at least insofar as each scarcely exists in nonliterary works, but the simple fact that critics continually recognize and study them suggests that they have become familiar, expected features within the literary text. In fact, we have come to employ the number of such features as a means by which we can measure the degree of literariness a work possesses, as is evidenced by such items as concordances to Joyce's Ulysses and Eliot's The Waste Land. Authors do not only make strange, they also make literary. Even in post-modernist literature, strangeness, playfulness, and disruptiveness have simply become the literary norm, to the extent that this Shklovskijan "junior branch" has now achieved canonization. Ultimately, it seems, literariness is a merging of the strange and the familiar. "Shakespeare," writes Dr. Johnson, "approximates the remote, and familiarizes the wonderful" (1986, p. 224), and this is the role of all literary art. The Literariness of Interactive Fiction
The most obvious attempts at a merger between traditional fiction and interactive fiction is the Electronic Novels series published by Broderbund
Software (formerly Synapse). Robert Pinsky's Mindwheel (1984), Bill Darrah's Essex (1985), James Paul's Brimstone (1985), and Rod Smith's Breakers (1986) all begin as traditional fiction, with roughly thirty printed pages serving as the first chapters of the story. Once through this introduction, the reader turns to the computer to finish the tale. Most interesting and encouraging about this series is its quality. Of the four, only Essex resembles earlier interactive fiction, which, as Niesz and Holland (1984) explain, was predominantly a series of puzzles that the reader grew frustrated trying to solve. But Breakers begins with an invitation to science-fiction mythmaking as allusive and thought-provoking as any such invitation: Starlight on naked rock: a phalanx of meteors charges through space -- fan mail from some dead planet, hurtling across the universe like blazing pinballs to flame out, rock by rock, in gravitational fields along the way. After eons of tumbling through darkness, the last meteors veer towards two stars and the golden planet looping around them in an endless figure-eight . . . . "When the Creator shall be masked," intoned one of the seven, and the others chanted, "then the world will live again..." "When the mask shall fall away," chanted the first, and the others responded, "then the world will live a g a i n . . . " The first one said, "The darkness quickens -- it is the time of renewal. When the blood star vanishes, the dark storm will scour the planet clean. All our people must be returned to Borg. And then we must perform the ritual of the elements, to recreate the Creator so that the Lau may live." (Smith, pp. 11--12).
The figure-eight image of infinity, the infinite ritual of re-creating the Creator itself: these are the mythological features on which science fiction is based. No reader of science fiction, or for that matter of The Golden Bough, will fail to recognize Breakers' genealogical line, and the book's (and the program's) insistence on myth is sufficient initial proof of its grounding in the literary. In its familiarity, it suggests one type of literariness. And yet, like all interactive fiction, and perhaps especially the works in the Electronic Novels series, it is also literary in its ostranenie. In fact, it makes doubly strange the reading of story. For, while a reader of traditional fiction will find the comfort of familiarity in the opening few chapters, only to be jarred into the non-passive, non-familiar experience of the fiction on the disk, a reader
familiar with interactive fiction will find strange (as well as time-consuming) the need for reading the introductory chapters in the first place. In Breakers, Essex, and Mindwheel, the introductory chapters are essential to the solving of the disk portion of the story, since they provide the reader with needed information, clues, and a sense of the invented world. Essex offers the background to the story, and Mindwheel outlines the reader's role in the solution. The introduction to Breakers is an immersion into the alien world's central mythos, thereby providing not only background but also a moral purpose to the reader's quest. Robert Pinsky's Mindwheel is self-consciously literary. Not only is the main character named Hay-Seuss Pederson, a rather clumsy and obvious rearrangement of "Jesus" and "Son of Peter", but the novel's mind machine, through which the reader travels into the minds of four deceased persons, is invented by a scientist named Virgil. The reader's quest is to journey through the dreams of a musician, a dictator, a poet, and a scientist, in order to retrieve an artifact with which civilization can be saved. Based on Pinsky's own poetry, especially "The Figured Wheel", the adventure is, from start to finish, self-referential and dream-like. Closest perhaps to Jakobson's formulation of poetic language is the interactive portion's combination of synchronicity and diachronicity; there are four minds, but three can be travelled in any sequence. Pinsky makes full use of the non-linearity possible in interactive fiction to set the associative, synchronic nature of poetry against the combinative, diachronic nature of fiction. Of all interactive works to date, the most literarily allusive is James Paul's Brimstone. In its self-referential opening chapter, "The Confession", narrator Jeremy Diddler recounts how he found a manuscript, The Dream of Gawain, and has subsequently confessed to its theft. In the "Explanatory Note", James Paul explains more about the manuscript, and about his friendship with Diddler. On the disk, Diddler appears again, when the reader, now playing the role of Gawain, visits him in jail. But Gawain (i.e. the reader) is by now acting out the dream-vision referred to in the stolen manuscript, thereby bringing the self-reference full-circle. The literary allusions begin early in the disk portion of the novel, as Gawain meets
IN I N T E R A C T I V E
Blake's Enitharmon and later Blake himself, and finally the Green Knight. Gawain's journey takes place in the ice-fields of Dante's Hell, described in great detail and doubly effective since the reader discovers Hell's features by journeying blindly into them. Inside that Hell is Blake's house, on the wall of which a painting changes continually, depicting in succession several of the Songs of Experience. If a lack of exterior referentiality characterizes literariness, Brimstone is in this regard, too, perhaps the most literary of all current interactive works. Unlike the introductory chapters of Essex, Breakers, and even Mindwheel, those of Brimstone are not functional. The reader gains no essential information from them, and can in fact skip over them completely. By doing so, however, she loses much of the novel's literary worth, because "The Confession" is an aesthetic opening to an aesthetic text. In other words, reading Brimstone without reading "The Confession" would be much like reading The Canterbury Tales without bothering with the General Prologue. It can be done, but the aesthetic loss is enormous. Like the Electronic Novels, Steve Meretzky's A Mind Forever Voyaging begins with a printed introductory chapter, and like Brimstone this chapter has aesthetic value only. In this work, the reader plays the role of a robot, whose early years (recounted in the introduction) were spent living as a human being, gaining the experiences necessary to function in human society. In the novel's interactive portion, the reader/robot operates a simulation of future societies, going further and further into the future to discover the effects of a restorative plan about to be put into effect. Two things make A Mind Forever Voyaging literary, at least in relation to other text adventures: first, it does not keep "score" of the reader's accomplishments (or "progress", as the Electronic Novels term it); second, its story is rhetorically powerful. Not keeping score eliminates the reader's perception of doing the fight thing versus the wrong thing, thereby making the story itself, as opposed to its game-playing characteristics, more immediate. The novel's power stems from the reader's gradual understanding of the incessant decay of the societies of the future. As she enters each new future period, the reader encounters a world growing increasingly violent, increasingly decadent, and increasingly alienating. As in Mindwheel
and Brimstone, the entire adventure takes place in the mind of the reader/character, but unlike those two books A Mind Forever Voyaging does not offer an obvious feeling of control. In fact, the future societies can be corrected only by changing the course of the present, not through future actions at all. The five works discussed so far differ from previous interactive fiction in several ways, but one substantial difference is in the length of textual passages. Besides the introductory chapters, which are of short-story length, the on-screen passages of these novels frequently occupy more than a full screen (or roughly 200 words). While length alone certainly does not imply literariness, it does suggest a greater concentration on the creation of text than does the more common, short, purely descriptive interactive text. Bucking this trend, though, and yet maintaining a literary stature, is Brian Moriarty's Trinity (1986), a fantasy that takes the reader/character back in history to the moment of the first atomic bomb detonation. Quotations from literary works are scattered throughout Trinity, each with the purpose of placing the reader inside a literary tradition. Like A Mind Forever Voyaging, Trinity depends in part on imagistic strength, but here the powerful images are those of the atomic explosion itself. Particularly forceful is a scene in a Hiroshima playground, in which the reader meets and befriends a young girl inside a bomb shelter. As it turns out, the reader has already encountered her, at the beginning of the novel, a facially-scarred Japanese woman in the park. Not until the explosion is the source of the scars apparent. Like the interactive novels discussed above, Trinity derives literariness from the combination of strangeness and familiarity. The familiarity is obvious, in its literary and its extraliterary references, its concentration on a crucial moment in the history of the twentieth century. Most recent, serious interactive fiction forces moral action upon the reader, thereby abandoning (in apparent disgust) the kill-and-steal mentality of the genre's early works. As the authors assume the role of literary artist, they are beginning to see themselves as part of the Western literary tradition, and they are also beginning to question the power and the value of their new genre. Much of this work, in fact, fits well in the Sidney-Shelley-John Gardner
school of moral fiction, a school that stresses literariness as a function of artistic responsibility. The novel's strangeness, its peculiar ostranenie, lies in its juxtaposition of bizarre fantasy and depressing history. In Trinity, Alice in Wonderland collocates with the nuclear holocaust. I have claimed that the Electronic Novels series represents the most obvious attempt to merge traditional and interactive fiction. A less obvious but perhaps more successful merger occurs in Rob Swigart's Portal (1986), certainly the most unusual and arguably the most interesting interactive fiction work yet written. The printed introduction sets for the reader a fascinating task: to discover why the population of the world ha s disappeared. There is no evidence of nuclear warfare, and no bodies to suggest biological warfare. Everyone is simply gone. To find them, the reader has only the resources of Homer, the interpreter of an enormous but malfunctioning database. The reader's job is to access :the database in the proper order, thereby re-creating the data inside Homer's own data banks and allowing him to narrate the story of Peter Devore, the person who ultimately arranged the disappearance. As the reader accesses the database, information becomes available about people, places, historical events, scientific and parapsychological discoveries, weapons research, and communications problems. Each interaction re-creates material for Homer's continuing narration and, indeed, for Homer's growing personality. Content at first with simply narrating the story, Homer slowly becomes uncertain of the value of the narration. He questions his abilities to interpret data, and he wonders how he can possibly write of human experience when he himself is not human. But what he narrates, in fact, is not only the story of humankind's disappearance, but also the growing to maturity of Peter Devote and the mythological love story of Peter Devore and Wanda Celeuf, whom Peter seeks among the stars. I say "mythological" because Homer tells us repeatedly, through Peter's memory, of an ancient myth of the confluence of two stars, and both Peter and Homer hint that Peter and Wanda are enacting that myth. This Homer, then, is set intentionally against the narrator of the Iliad and the Odyssey, himself also a mythmaker and a teller of heroic legend. Portal is, ultimately, a metaphor for the writing
L I T E R A R I N E S S IN I N T E R A C T I V E
process. We watch on the screen as the story's intertext is molded by the narrator into a psychologically valid myth of heroic scope, and we watch too as the narrator himself questions the act of narration. The end of the story is dissatisfying, not because it is abrupt but rather because we lose Homer, for whom we have grown to care, as we return whence we came. I am not sure exactly where in the story our concern for Peter Devore fades in favour of our deeper concern for Homer, but I do know that it happens. And it happens, ! think, because Homer, in the very literary activity of making familiar that which is strange, rather than the other way around, gives us a glimpse of the emotional, psychological, and mythical drawing power of narrative. In Homer's activity, we may have a further element of literariness. While ostranenie insists on making the familiar strange, something that interactive fiction accomplished originally simply by using the computer as medium, Homer takes the strange -- the uninterpreted, uncollated, unnarrated facts and studies -- and turns them into narrative, something with which we are familiar. This is, of course, the reverse of ostranenie, but it fits well with the idea that interactive fiction attains literariness by its combination of ostranenie and familiarization. Perhaps because interactive fiction is still a strange genre, because unlike other literary genres it does not rely on the ages-old Gutenberg technology, it establishes itself within the literary tradition precisely to provide us with familiarity. If we judge a traditional text's literariness partly by how it breaks from its own traditions, we judge the literariness of interactive fiction partly by how it adheres to those traditions. In this way, the works I have discussed are worthy in every sense of incorporation into the sub-canon of science fiction and fantasy literature, an incorporation that will demonstrate once again Shklovskij's "canonization of the lower branch". But interactive fiction's essential ostranenie, its reliance on computer technology, cannot be passed over so quickly. Even if we grant Niesz and Holland's claim that interactive fiction "simply pushes the role of the text back a stage" (1984, p. 124), forcing the reader to decipher the "underlying nexus of puzzles or conundrums [that] is the creation of the author", the fact remains that the reader of an interactive text must physically act if
she is to complete, decipher, discover the text she is reading. As each new narrative section appears on the screen, she must type something, or the story will simply not continue. Two interesting theoretical concerns arise from this situation. First, interactive fiction disrupts the concept of the novel's "last page"; whereas we often look to the last page of a printed novel, either to cheat on the ending or simply to see how many pages remain to be read, an interactive fiction work's only indication of ending (unless it keeps score, a rapidly disappearing feature) is the story's announcement that the quest has been solved. Second, the demand that the reader type something creates very real gaps in the narrative, and here we return once more to Wolfgang Iser. "The indeterminate sections or gaps of literary texts," writes Iser, "are in no way to be regarded as a defect; on the contrary, they are a basic element for the aesthetic response" (1972, p. 12). They are basic primarily because through these gaps the reader interacts with, responds to, even co-creates the text, and the fact that "every literary text invites some form of participation on the part of the reader" (p. 13) is itself, for Iser, a definition of literariness. In some fiction, he argues, particularly that published before the twentieth century, "the reader will not even be aware of such gaps" (p. 12), but the gaps affect the reading nonetheless. In interactive fiction, however, the reader is automatically aware of the gaps because until she fills them in the text will not continue. In other words, interactive fiction displays two types of narrative gap: the traditional type filled unconsciously by the reader, and a manifest type shown on the computer's screen. The effect of this additional gap on the interactive text's literariness is profound. Even in fiction as disjointed as Italo Calvino's If On a Winter's Night a Traveler or as long-to-read as James Joyce's Ulysses, the experience of reading is at least potentially continuous. Simply by turning the pages, the reader of a traditional text establishes reading continuity, even if she puts the book down and returns to it later. The reader of interactive fiction, on the other hand, establishes reading continuity only by typing in commands that will force the narrative one step further, and at a certain point may find herself unable to proceed. Here literariness seems denied. For all its reliance
on ostranenie, on strangeness, literariness has not foreseen a narrative that stops because the reader simply cannot figure out what to do next. The continuity implied by the consecutive numbering of pages seems, even though theorists do not mention it, one of the determining marks of the literary text. In fact, though, interactive fiction allows its own form of continuity. Even when the reader cannot formulate a solution to advance the interactive text, she can usually back-track to a previous screen or side-step to another. Continuity here is not necessarily linear, but it certainly exists. By back-tracking, the reader re-reads portions of the text, often many times over, in an effort to find a clue that will allow the barrier to be breached. By side-stepping, the reader hopes to return to the barrier with a new sense of how to surmount it. In either case, what is continuous is not the plot but rather the development of the reader's knowledge of the world in which her character is travelling. To return to the Iser quotation that opened this article, the reader here "can step out of his own world and get into another, where he can experience extremes of pleasure and pain without being involved in any consequences whatsoever." Just as it makes the familiar strange and the strange familiar, interactive fiction allows the reader to partake, first-hand, of a new literary world, and the unfolding of that world is continuous, even if the plot is not. NOTES i Throughout this article I will use the term "traditional" to refer to literature printed on paper, usually in book (rather than magazine) format. I considered such terminology as "book fiction" or "Gutenberg literature" to avoid collocating such diverse novels as Mitchell's Gone With the Wind and Barth's Sabbatical, which have in common only the fact that they are written by Americans and printed on paper, but such terminology proved more cumbersome than it was worth. "Traditional", therefore, says nothing of the degree to which a work is or is not avant-garde; it is simply a reference to medium. 2 Calling interactive fiction a "genre" creates its own set of problems. More properly, perhaps, interactive fiction should be considered a sub-genre of several existing genres: Fantasy, Science Fiction, Murder Mystery, Espionage, Horror. But to do so is to place interactive fiction under the constraints of those genres, and even though it certainly feeds off them it does not, by its interactive nature, conform to them. Because of interactive fiction's unique claim, that it is necessarily
rather than speculatively interactive (even if that interactivity is still at a primitive stage), I have elected to consider it a distinct genre.
REFERENCES Adams, Douglas and Steve Meretzky. The Hitchhiker's Guide to the Galaxy. Adapted from the novel by Douglas Adams. Cambridge, Mass: Infocom, 1984. Barol, Bill. "Zorked Again: Lost in Computer Fiction". Newsweek, December 23, 1985, p. 70. British, Lord. Ultima IV. Manchester, NH: Origin Systems, 1985. Darrah, Bill. Essex. San Rafael, California: Synapse-Broderbtmd, 1985. Eagleton, Terry. Literary Theory: An Introduction. Oxford: Basil Blackwell, 1983. Erlich, Victor. Russian Formalism: History-Doctrine. 'SGravenhage: Mouton & Co., 1955. Gygax, Gary. Dungeons and Dragons. Lake Geneva, Wisc.: TSR, 1978. Hassan, Ihab. "Pluralism in Modern Perspective". Critical Inquiry, 12, 3 (1986), 503--20. Hawkes, Terence. Structuralism and Semiotics. Berkeley and Los Angeles: University of California Press, 1977. Holland, Norman. "Literature as Transaction". In What Is Literature? Ed. Paul Hernadi. Bloomington: Indiana University Press, 1978, pp. 206--18. Iser, Wolfgang. "Indeterminacy and the Reader's Response". In Aspects of Narrative. Ed. J. Hillis Miller. New York: Columbia University Press, 1971, pp. 1--46. Jakobson, Roman. "Closing Statement: Linguistics and Poetics". In Style and Language. Ed. Thomas Sebeok. Cambridge, Mass: MIT Press, 1960, pp. 350--77. Jameson, Fredric. The Prison-House of Language. Princeton: University Press, 1972. Johnson, Samuel. "Preface to Shakespeare". In Criticism: The Major Statements. 2nd ed. Ed. Charles Kaplan. New York: St. Martin's, 1986. Macdonell, Diane. Theories of Discourse. Oxford: Basil Blackwell, 1986. Meretzky, Steve. A Mind Forever Voyaging. Cambridge, Mass: Infocom, 1985. Mitchel, Philip, et al. The Hobbit. Reading, MA: AddisonWesley, 1985. Moriarty, Brian. Trinity. Cambridge, Mass: Infocom, 1986. Niesz, Anthony J. and Norman N. Holland. "Interactive Fiction". Critical Inquiry, 11, 1 (1984), 110-29. Oxford Digital Enterprises. Macbeth. Baltimore: The Avalon Hill Game Company, 1986. Paul, James. Brimstone. San Rafael, California: SynapseBroderbund, 1985. Pinsky, Robert. Mindwheel. San Rafael, California: SynapseBroderbund, 1984. Pinsky, Robert. "The Figured Wheel". History of My Heart. New York: Ecco, 1984. Randall, Neil. Storm of Dust. New York: Tor, 1987. Randall, Neil. Seven No-Trump. New York: Tor, 1988.
D E T E R M I N I N G L I T E R A R I N E S S IN I N T E R A C T I V E F I C T I O N Scholes, Robert. Semiotics and Interpretation. New Haven: Yale University Press, 1982. Shklovsldj, Viktor. Literatura i Kinematograf. Berlin, 1923. Quoted in Erlich, p. 227. Smith, Rod. Breakers. San Rafael, California: Broderbund, 1986. Swigrat, Rob. Portal. Mountain View, California: Activision, 1986. Telarium Corp. Nine Princes in Amber. Cambridge, Mass: Telarium, 1985. Ward, D. C. and Ian Weatberburn. The Neverending Story. Chatsworth, CA: Datasoft, 1986. Wellek, Rene, and Austin Warren. Theory of Literature. 3rd ed. New York: Harcourt, Brace & World, 1956. York, R. A. The Poem as Utterance. London: Metbuen, 1986.
APPENDIX The principal publisher of this kind of interactive fiction is Infocom (Cambridge, MA). Broderbund-Synapse (San Rafael, CA) has halted its publication of Electronic Novels, but other publishers have entered to fill the void. Among these
are Electronic Arts (P.O Box 7530, San Mateo, CA 94403), Rainbird Software (P.O. Box 49, Ramsey, NJ 07446), and Aegis Development (2210 Wilshire Blvd., Suite 277, Santa Monica, CA 90403). These software packages are available for almost all popular microcomputer systems: MSDOS machines, Apple Macintosh, Apple II series, Amiga, Atari ST, Commodore 64, Atari 8-bit, and even (in Infocom's case) CP/M. In some cases, 128k of memory is required, but usually 64k is sufficient, if the user is prepared for a large amount of disk access. Because of the publishers' concentration on sales, no widely-available interactive fiction has appeared that is written exclusively for the machines with larger memories (IBM-AT, Amiga, Atari ST, Macintosh SE), or with advanced operating systems (Amiga, ST, Macintosh II, IBM's OS/2). Such fictions are, we are told, in the works.