RCCS
HomeIntroducing CybercultureBook ReviewsCourses in CybercultureEvents and ConferencesFeatured LinksAbout RCCS

View All Books

Twisty Little Passages: An Approach to Interactive Fiction

Author: Nick Montfort
Publisher: Cambridge, MA: MIT Press, 2003
Review Published: December 2006

 REVIEW 1: Russell Mills
 REVIEW 2: T. Michael Roberts
 AUTHOR RESPONSE: Nick Montfort

Nick Montfort's Twisty Little Passages is subtitled "an approach to interactive fiction," but it actually provides three interwoven approaches. First, the book is a genre analysis of interactive fiction (IF). Second, it is a manifesto proclaiming the value of this new genre. Finally, it is a survey of some of the most important works in this genre. Taken together, these three elements make this an essential resource for anyone interested in interactive fiction.

Montfort is uniquely qualified to analyze this literary genre, as both a scholar (Ph.D. student at Penn) and the author of several works of interactive fiction. He begins by distinguishing interactive fiction from other related literary forms. In his schema, a true work of interactive fiction must accept text input, analyze it using a parser, and finally modify the world of the interactive fiction based on that input. This definition of the genre excludes hypertext literature for two reasons: in hypertext, the input of the "reader" is limited to clicking on predefined links rather than entering text; and this input alters only the sequence in which the text is displayed and read, not the characteristics of the underlying fictional world.

A core element of IF is what Monfort calls the IF World. The IF World includes not only the place where things happen, but also the rules by which the events unfold. As Monfort notes, it "is no less than the content plane of interactive fiction, just as story is the content plane of a narrative" (26). Further, the IF World is malleable, not static. Its characteristics (location of objects, behavior of characters, even laws of physics) can change in response to input from the interactor.

The protean, user-definable nature of the IF World means that the narrative itself exists in a potentially unlimited number of varieties; indeed, the work itself does not incorporate an actual narrative but it makes possible a number of potential narratives. The relationship between the IF World and the potential narratives could be compared to the relationship between deep structures and surface structures in linguistics, where a set of rather simple and limited rules can give rise to a rich and varied set of surface structures (cf. Chomsky).

Montfort's idea of an IF world gives rise to important questions about the nature of causation and verisimilitude in IF worlds. These issues have been explored in the existing literature on representation and "fictional worlds" (for example, see Auerbach, Bachelard, Gelley). In one seminal treatment, literary critic Robert Champigny suggests that "to the extent that fictional events are stated explicitly, causality and probability apply to fiction as well as to history ... but causality and probability should not be applied to make explicit what is implicit in fiction" (989). Champigny particularly questions whether the laws of physics and biology apply to fictional worlds. He concludes that they do not unless the author specifically tells us that they do.

Champigny is probably overstating the case. In reading a work of fiction, most readers assume that the fictional world resembles their own. This is what gives violations of verisimilitude their affective power. In the case of IF, the case is more complex because the work is classifiable not only as a work of interactive fiction, but also as sci-fi, romance, detective fiction, spy thriller, etc. For example, Montfort provides brief descriptions of the 35 IF works developed and published by Infocom. Many of the works are classified according to traditional fictional genres: Border Zone, is a spy thriller, Deadline is a detective story, and A Mind Forever Wandering is a dystopia (122-124). And each fictional genre has its own set of expectations about the underlying fictional world. In a romance world, we can expect magical spells and sorcerers; a sci-fi world will have alternative concepts of space, time, and physics; a detective world will eschew miracles and time travel; etc. In adopting a particular literary genre for a work of IF, an implementer is also adopting a set of assumptions about the sort of IF world that can appropriately be constructed for such a work.

Montfort sidesteps the question of whether all of the various potential narratives have equal status. I would argue that in most IF works, there is a "target" potential narrative, which the interactor strives to discover. In many cases, the interactor is graded on how closely his or her particular traversal of the IF work corresponds to the target narrative. In the case of a "perfect" traversal, the interactor will receive the maximum score. In other cases, the interactor may blunder so completely that he or she can never accomplish a full traversal; often, this is signaled by the death of the adventuring character. Clearly, some potential readings are better than others.

This notion of a "privileged" narrative is consistent with Montfort's claim that the literary riddle is the most important precursor of interactive fiction. Montfort dismisses the influence of detective fiction, perhaps because a detective story explicitly presents the solution at the end, while no solution is explicitly presented in a literary riddle. But an IF work does explicitly congratulate the interactor if he or she is successful in finding the solution -- or chide him or her for failing to find the solution. No such "answer key" is present in a literary riddle. A work of detective fiction does contain a solution, which is announced at the end of the work, and the reader need not successfully solve the puzzle in order to learn the solution; however, a fully satisfactory reading of a work of detective fiction does require that the reader make the attempt. A reader of detective fiction can cheat by peeking at the back, but so can an IF interactor cheat by looking up hints on a bulletin board. So perhaps IF owes as much to detective fiction as it does to the literary riddle.

From Montfort's defense of interactive fiction as a literary genre, it is clear that many scholars hold IF to be a mere curiosity, in contrast to the more "serious" form of hypertext-based fiction. Thus, interactive fiction is not only a marginal form, but it is marginal even within the marginal world of computer-based fiction. Given the potential of the interactor to affect not only the sequence of the text, but even to alter the most fundamental qualities of the fiction, Montfort's commitment to the IF form seems warranted. Unfortunately, as Montfort acknowledges, few actual works of IF have fully exploited the literary potential of the form.

In his survey of interactive fiction, Montfort provides solid readings of the seminal works of the genre (Adventure and Zork), as well as brief synopses of the Infocom canon and its successors. Along the way, he gives readers glimpses of the IF culture. We learn how early works of IF contain references to the architecture of the MIT campus and to MIT personalities as well (98-107). We learn of the virtual "Z machine," which allowed Infocom games to be developed for a single programming environment that would make them playable on a wide variety of platforms (126-127). We are given a sense of the satirical, allusive style of the games themselves; for example, in Zork, uttering a magic word from the earlier Adventure game produces the reply, "Cretin" (99). But Montfort does not, perhaps because he cannot, make the case that works of IF have sufficient depth to merit careful literary analysis.

Throughout his discussion of the various works, Montfort seems apologetic about the lack of literary sophistication represented by most works of interactive fiction. The main exception appears to be A Mind Forever Voyaging, which Montfort finds to be a "work of unparalleled power" (152-153). In his discussion of the most recent literature, Montfort finds works that offer subtlety and sophistication of execution, but there does not appear to be a coherent tradition of IF practice -- or perhaps the tradition is too fragmented, with the territory divided between gamers, MUD and MOO aficionados, proponents of adult interactive fiction (i.e., depiction of sexual activity), and other fringe types.

Despite the lack of a well developed canon, IF appears to offer, at least in theory, the possibility of a much more sophisticated literature than what has been produced so far. One potential that does not seem fully realized is the possibility of true "co-authorship," where the interactor generates the actual text of the narrative rather than simply typing commands. Curiously, this potential was realized to a certain degree in the IF precursor, ELIZA. That computer program stores and repeats phrases that have been provided by the interactor. In a very crude fashion, the interactor is partially responsible for creating the text that constitutes the "story." Given the increasing competence of parsers, chatterbots, and other linguistic software, this idea of "interactor-generated text" appears to be an area that offers great potential. Moreover, this potential goes far beyond the theoretical limits of hypertext fiction.

Perhaps the most unexpected legacy of Interactive Fiction has been the development of the World Wide Web, which was strongly influenced by the seminal interactive fiction work, Adventure. According to Montfort, the idea of exploring "twisty little passages" in Adventure's virtual cave led naturally to the idea of following "twisty" hyperlink connections to traverse a web of interconnected computers (225). As much as anything in Montfort's book, this illustrates the conceptual importance of interactive fiction. If the creative play of a game like Adventure contains the seeds of the Web, what literary possibilities may be lurking in the form?

I would recommend this book not only for those who are interested in interactive fiction as such, but by scholars of literary theory. The characteristics of this genre (or "potential genre"?) call into question many of the fundamental categories of fictional theory. Concepts such as "character," "setting," "plot," and "narrative" all take on altered qualities in interactive fiction: qualities that offer hints of important new possibilities that are ready to be explored by future practitioners -- perhaps by Monfort, himself a practitioner of the form.

Auerbach, Erich. Mimesis: The Representation of Reality in Western Literature, tr. Willard Trask. Princeton: Princeton University Press, 1953.

Bachelard, Gaston. La Poétique de l'Espace. Paris: Presses Universitaires de France, 1957.

Champigny, Robert. "Implicitness in Narrative Fiction," PMLA, 85 (1970), 988-991.

Chomsky, Noam. Aspects of the Theory of Syntax. Cambridge, MA: MIT Press, 1965.

Gelley, Alexander. "Setting and a Sense of Reality in the Novel." Yale Review, 42 (1972), 186-201.

Russell Mills:
Russell Mills is professor of English, humanities, and social sciences at Vermont Technical College. He teaches courses in the relationships between technology and the broader culture. His research focuses on the role of technology in constructing and expressing cultural meaning.  <rmills@vtc.edu>

RCCS
 HOME   INTRO   REVIEWS   COURSES   EVENTS   LINKS   ABOUT
©1996-2007 RCCS         ONLINE SINCE: 1996         SITE LAST UPDATED: 12.10.2009