Working with Janet Murray's "Agency" chapter
Chapter 5: Agency
The more realized the immersive environment, the more active we want to be within it. When the things we do bring tangible results, we experience the second characteristic delight of electronic environments—the sense of agency. Agency is the satisfying power to take meaningful action and see the results of our decisions and choices. We expect to feel agency on the computer when we double-click on a file and see it open before us or when we enter numbers in a spreadsheet and see the totals readjust. However, we do not usually expect to experience agency within a narrative environment.Even in the rare circumstances when we are invited to participate in a traditional narrative form, our participation is circumscribed in a way that generally limits our sense of agency. For instance, if the audience at a performance of Peter Pan decided that Tinkerbell is a pest and refused to clap her back to life, the play would come to a halt. The participatory dinner theater plays that simulate an Italian wedding or an Irish wake or a Jewish funeral encourage audience participation by keeping the plot to a minimal level and the dialogue with the audience to social formulas appropriate to distant friends of the family. When the groom in such a play leans down to kiss me as a guest at his wedding, I can congratulate him and warn him about staying away from his ex-girlfriend now that he is married, but I cannot really influence his behavior. When audience members are included in the story, they serve only as the butt of a joke. They may be accused of adultery by the priest or shot by a mafioso relative. The slender story is designed to unfold in the same way no matter what individual audience members may do to join the fun.
In fact, participatory theater performances become participatory by incorporating folk art forms and festival behavior such as singing, dancing, and sharing a feast. Striking up a familiar song or dance tune—“Que Sera Sera” or a tarantella—is a reliable way to get the audience involved. Musical participatory forms are successful because they rely on careful cueing and formulaic behavior: We sing along with the chorus and remain silent for the verse; we answer the singer’s “call” with the appropriate response. And we do these things in unison as a single voice. In a square dance we perform whatever steps the caller announces because the repertoire of possible movements and the rules of combination are known to both parties before the music starts, and though everyone does not have to do exactly the same thing at the same time, all the square dancers do have to be part of a single overall pattern. Folk dancing in ballroom style offers a model of freer participation. In the Cajun two-step or the Brazilian samba, for instance, all the dancers share a repertoire of movements, and each set of partners has license to invent its own combinations and interpretations of these movements. Like jazz musicians, the dancers can improvise their own satisfying creations from these given elements. But the greater individual freedom in ballroom-style folk dancing means that the group as a whole has less coherence than at a square dance.
Electronic environments have similar formulas and rules for structuring participation. For instance, when users are merely asked to respond to a menu with a predictable begin/quit choice, they are performing a kind of response to the “call” of the machine. When we learn a complicated program, like a word processor, and run through its familiar steps in order to do a difficult job, we are like participants in a square dance, repeating formulaic sequences, with the relevant manual page acting as caller of the dance. When we are placed within a simulation environment and allowed to experiment with changing a set of parameters as we see fit (more nitrogen, less algae), we are acting more like the leading partner in a Cajun dance. The crucial difference, however, between folk art rituals and computer-based interactions is that on the computer we encounter a world that is dynamically altered by our participation. On the ballroom dance floor, we can at most influence our partner, but the musicians and the rest of the dancers remain relatively unaffected. Within the world of the computer, however, when the right file opens, when our spread-sheet formulas function correctly, or when the simulated frogs flourish in the model pond, it can feel as if the entire dance hall is at our command. When things are going right on the computer, we can be both the dancer and the caller of the dance. This is the feeling of agency.
Because of the vague and pervasive use of the term interactivity, the pleasure of agency in electronic environments is often confused with the mere ability to move a joystick or click on a mouse. But activity alone is not agency. For instance, in a tabletop game of chance, players may be kept very busy spinning dials, moving game pieces, and exchanging money, but they may not have any true agency. The players’ actions have effect, but the actions are not chosen and the effects are not related to the players’ intentions. Although gamemakers sometimes mistakenly focus on the number of interactions per minute, this number is a poor indicator of the pleasure of agency afforded by a game. Some games, like chess, can have relatively few or infrequent actions but a high degree of agency, since the actions are highly autonomous, selected from a large range of possible choices, and wholly determine the course of the game.
Agency, then, goes beyond both participation and activity. As an aesthetic pleasure, as an experience to be savored for its own sake, it is offered to a limited degree in traditional art forms but is more commonly available in the structured activities we call games. Therefore, when we move narrative to the computer, we move it to a realm already shaped by the structures of games. Can we imagine a compelling narrative literature that builds on these game structures without being diminished by them? Or are we merely talking about an expensive way to rewrite Hamlet for the pinball machine?
The Pleasures of Navigation
One form of agency not dependent on game structure yet characteristic of digital environments is spatial navigation. The ability to move through virtual landscapes can be pleasurable in itself, independent of the content of the spaces. A friend of mine whose son is an avid Nintendo player complains that when he tries out the games he is annoyed at having to be fighting all the time, since the combat is an unwelcome distraction from the pleasure of moving around the unfolding spaces of the maze. For my friend, videogames are about exploring an infinitely expandable space. Similarly, new explorers of the World Wide Web find themselves entranced with the ability to leap around the world, following links from one home page or Web site to the next mostly for the pleasure of the repeated arrivals. The navigational pleasures are richly exploited by the many forms of labyrinths, from Zork-like dungeons to informational webs, that fill cyberspace. All of them allow us to experience pleasures specific to intentional navigation: orienting ourselves by landmarks, mapping a space mentally to match our experience, and admiring the juxtapositions and changes in perspective that derive from moving through an intricate environment.This participatory pleasure is not unlike the enjoyment people find in the organized sport of “orienteering,” where players follow a series of geographical clues across a large and complex terrain, such as a portion of the Maine woods or downtown Boston. Construing space and moving through it in an exploratory way (when done for its own sake and not in order to find the dentist’s office or the right airport gate) is a satisfying activity regardless of whether the space is real or virtual. Electronic environments offer the pleasure of orienteering in two very different configurations, each of which carries its own narrative power: the solvable maze and the tangled rhizome.
The Story in the Maze
Zork-like puzzle dungeons and maze-based combat videogames derive from a heroic narrative of adventure whose roots are in antiquity. It was Daedalus who built King Minos of Crete a labyrinth to contain the deadly Minotaur. The horrible beast required the yearly sacrifice of Athenian youths and maidens, whom it devoured, until the hero Theseus arrived to slay it. In the story, Ariadne, the daughter of the king, fell in love with Theseus and gave him a sword to kill the beast and a thread to find his way out again. Minos’s maze was therefore a frightening place, full of danger and bafflement, but successful navigation of it led to great rewards.The adventure maze embodies a classic fairy-tale narrative of danger and salvation. Its lasting appeal as both a story and a game pattern derives from the melding of a cognitive problem (finding the path) with an emotionally symbolic pattern (facing what is frightening and unknown). The maze story celebrates the combination of intelligence and courage, and it depicts romantic love as the element that provides the hope that brings the hero into the confrontation and back out again to safety. Like all fairy tales, the maze adventure is a story about survival. The maze is a road map for telling this story.
As a format for electronic narrative, the maze is a more active version of the immersive visit (as described in chapter 4). Maze-based stories take away the moving platform and turn the passively observant visitor into a protagonist who must find his or her own way through the fun house. A typical maze-based puzzle game sends you, the player, through a multitiered space vaguely resembling an Arabian Nights palace. You operate an avatar who walks through the palace rooms, whose tiled floors and ornately decorated corners often hide treasures that are tricky to perceive. The palace is full of informants, who speak in text bubbles and whom you reply to from menus, and you must negotiate with them carefully, offering them icons representing money or other valuables. A mysterious peddler on one of the lower levels holds a talisman needed to get into the highest chamber. You must have it with you while you stand on a special spot that is hidden in the patterning of the floor. If you forget to get it, you must retrace your steps through many perils. The game is like a treasure hunt in which a chain of discoveries acts as a kind of Ariadne’s thread to lead you through the maze to the treasure at the center.
This kind of narrative structure need not be limited to such simplistic content or to an explicitly mazelike interface. In the right hands a maze story could be a melodramatic adventure with complex social subtexts. For instance, instead of a fairy tale palace it could be set in a Kafkaesque city where the secret police are rounding up and deporting citizens with the wrong kind of papers. The protagonist’s role would be to save them, a task that would require navigation through the corridors of power and through underground hiding places, elaborately conducted negotiations, clever manipulation of bureaucrats, and split-second timing. The characters in the menacing world could be subtly portrayed, in either graphics with text bubbles or in video segments. Saving people might involve horrifying choices, perhaps implicating the protagonist in the corruption of the violent world. The maze could be composed not only of spatial twists but of moral and psychological choices. Just as it is hard to see where a tangle of virtual corridors is leading, so too would it be hard to foresee the consequences of your actions and to determine what to value and whom to trust. Just as Kafka used the conventions of the fable to convey the profound depersonalization of modern life and Art Spiegelman used the format of the comic book to tell the story of his father’s Holocaust experiences, a digital artist might use the structure of the adventure maze to embody a moral individual’s confrontation with state-sanctioned violence.
Whether an adventure maze is simple or complex, it is particularly suited to the digital environment because the story is tied to the navigation of space. As I move forward, I feel a sense of powerfulness, of significant action, that is tied to my pleasure in the unfolding story. In an adventure game this pleasure also feels like winning. But in a narrative experience not structured as a win-lose contest the movement forward has the feeling of enacting a meaningful experience both consciously chosen and surprising. However, there is a drawback to the maze orientation: it moves the interactor toward a single solution, toward finding the one way out. The desire for agency in digital environments makes us impatient when our options are so limited. We want an open road with wide latitude to explore and more than one way to get somewhere. We want the “pullulating” web that Borges described, constantly bifurcating, with every branch deeply explorable.
Rapture of the Rhizome
The second kind of digital labyrinth, which has arisen from the academic literary community, is the postmodern hypertext narrative described in chapter 2. Full of wordplay and indeterminate events, these labyrinths derive not from Greek rationalism but from poststructuralist literary theory and are unheroic and solutionless. Like a set of index cards that have been scattered on the floor and then connected with multiple segments of tangled twine, they offer no end point and no way out. Their aesthetic vision is often identified with philosopher Gilles Deleuze’s “rhizome,” a tuber root system in which any point may be connected to any other point.1 Deleuze used the rhizome root system as a model of connectivity in systems of ideas; critics have applied this notion to allusive text systems that are not linear like a book but boundaryless and without closure. Stuart Moulthrop, a theorist and electronic fiction writer, states it this way:Seen from the viewpoint of textual theory, hypertext systems appear as the practical implementation of a conceptual movement that … rejects authoritarian, “logocentric” [i.e., truth-affirming] hierarchies of language, whose modes of operation are linear and deductive, and seeks instead systems of discourse that admit a plurality of meanings where the operative modes are hypothesis and interpretive play.2
The postmodern hypertext tradition celebrates the indeterminate text as a liberation from the tyranny of the author and an affirmation of the reader’s freedom of interpretation. But the navigational software designed specifically for this purpose and celebrated by many proponents of literary hypertext is anything but empowering to the reader, even in comparison to the earliest Web browsers.3 For instance, it offers the navigating reader no way to mark links as having been already taken, and no way to mark a lexia so it can be easily jumped back to. Many of the stories written in this framework do not even mark which words are hot links within the lexia text. Instead, the reader has to click on a pop-up display of cryptic link names. Moulthrop’s own Victory Garden, which is perhaps the most coherently structured literary hypertext, contains a clever overview map of the major story clusters, which are arranged like a Borgesian garden labyrinth. But readers cannot easily return to the overview in order to get a sense of where they are or how much is left to read. In trying to create texts that do not “privilege” any one order of reading or interpretive framework, the postmodernists are privileging confusion itself. The indeterminate structure of these hypertexts frustrates our desire for narrational agency, for using the act of navigation to unfold a story that flows from our own meaningful choices.
But the unsolvable maze does hold promise as an expressive structure. Walking through a rhizome one enacts a story of wandering, of being enticed in conflicting directions, of remaining always open to surprise, of feeling helpless to orient oneself or to find an exit, but the story is also oddly reassuring. In the rhizome, one is constantly threatened but also continuously enclosed. The fact that the plot will not resolve means that no irreparable loss will be suffered. The narrator of Afternoon (discussed in chapter 2) will not have to confront the fact of the morning’s fatal accident so long as the afternoon’s evasive wanderings continue, and the reader of Victory Garden does not have to accept the death of an appealing character so long as there are multiple paths to explore, including some that lead to alternate realities in which she comes back home from the war. In both stories the reader is protected from feeling the irreversibility of death by the fact that the stories do not have to end there.
The boundlessness of the rhizome experience is crucial to its comforting side. In this it is as much of a game as the adventure maze. In fact, it reminds me of a particular game my son William invented at about age five. At his own initiative he one day drew a large game board, assembled dice and playing pieces, and invited his father to join him in an inventively improvised game with ever-changing and ever more elaborate rules. After two hours of this surreal activity, my husband became restless and began asking every five minutes or so if the game was almost over. William responded by calmly walking into the kitchen, where I was sitting, and asking me to write his father the following note:
DEAR DAD—THIS GAME WILL NEVER END. WILLIAM
The rhizome has the same message. As we navigate its tangled, anxiety-laden paths, enclosed within its shape-fitting borders, we are both the exasperated parent longing for closure and separation and the enthralled child, lingering forever in an unfolding process that is deeply comforting because it can never end.
Giving Shape to Anxiety
Both the overdetermined form of the single-path maze adventure and the underdetermined form of rhizome fiction work against the interactor’s pleasure in navigation. The potential of the labyrinth as a participatory narrative form would seem to lie somewhere between the two, in stories that are goal driven enough to guide navigation but open-ended enough to allow free exploration and that display a satisfying dramatic structure no matter how the interactor chooses to traverse the space.The key to creating an expressive fictional labyrinth is arousing and regulating the anxiety intrinsic to the form by harnessing it to the act of navigation. Suspense, fear of abandonment, fear of lurking attackers, and fear of loss of self in the undifferentiated mass are part of the emotional landscape of the shimmering web. Moving through the space can therefore feel like an enactment of courage and perseverance, like Gary Cooper’s striding through the town in High Noon. Computer gamers often experience shivers of physical fear as they approach an unopened door in a text-based or graphics-based labyrinth. The drama of suspenseful approach does not have to be tied to combat or to jack-in-the-box effects. It can also have the feeling of a determination to face the truth, to stare directly at the threatening beast. It can be experienced by the navigating reader/viewer as well as by the player/protagonist.
One such use of the labyrinth as a means of evoking and controlling terror is a story format increasingly used by my fiction students, a story I have come to call the “violence hub.” Writers will place an account of a violent incident, often a real or invented newspaper article, at the center of a web of narratives that explore it from multiple points of view. A helicopter accident near MIT, a robbery in a convenience store, a canoeing fatality—these are all examples of an act of sudden violence that have served as the center of such a story web. The incident account itself is usually bare but evocative. People have died in violent and surprising circumstances, we are told. The incident happened at a particular place and time involving a particular group of people. The names in the account or in a diagram of the site of the accident lead outward with hot links to the satellite files that tell us how the incident appeared to the various people involved—the perpetrator, the witnesses, the rescuers, the victims, the survivors.
The proliferation of interconnected files is an attempt to answer the perennial and ultimately unanswerable question of why this incident happened. For instance, one convenience store robbery labyrinth follows the robber, the clerk, the owner, and the cop (who shoots the robber) back through the events leading up to the shootout and forward again into the moment of violence. Reading it we feel sympathy with all of them, and we see how they appeared to one another. A web story of a teenager who drowned on a white-water canoeing trip describes the traumatic experience from the points of view of the friends he was traveling with, the operators of the canoeing company, the emergency medics, and the family members receiving the dreadful phone call. These violence-hub stories do not have a single solution like the adventure maze or a refusal of resolution like the postmodern stories; instead, they combine a clear sense of story structure with a multiplicity of meaningful plots. The navigation of the labyrinth is like pacing the floor; a physical manifestation of the effort to come to terms with the trauma, it represents the mind’s repeated efforts to keep returning to a shocking event in an effort to absorb it and, finally, get past it. The retracing of the situation from different perspectives leads to a continual deepening in the reader’s understanding of what has happened, a deepening that can bring a sense of resolution but one that allows for the complexity of the situation and that leaves the moment of shock unchanged and still central.
A linear story, no matter how complex, moves toward a single encompassing version of a complex human event. Even those multiform stories that offer multiple retellings of the same event often resolve into a single “true” version—the viewpoint of the uninvolved eyewitness or the actual reality the protagonists wind up in after the alternate realities have collapsed. A linear story has to end in some one place: the last shot of a movie is never a split screen. But a multithreaded story can offer many voices at once without giving any one of them the last word. This is a reassuring format for encountering a traumatic event because it allows plenty of room for conflicting emotions. It lets us disperse complex, intense reactions into many derivative streams so that we do not have to feel the full flood of sorrow all at once. The multithreaded web story achieves coherent dramatic form by shaping our terror into a pattern of exploration and discovery.
The Journey Story and the Pleasure of Problem Solving
The navigational space of the computer also makes it particularly suitable for journey stories, which are related to mazes but offer additional opportunities for exercising agency. Journey stories date back to oral storytelling, from the fairy-tale convention of setting out from home to find one’s fortune to the voyages of Odysseus and Sinbad. It is a universal archetype recognizable across all the variations of culture, author, and medium. After the invention of the printing press, the journey story was reinvented as the picaresque novel, exemplified by Don Quixote, Moll Flanders, Tom Jones, Huckleberry Finn, and Catcher in the Rye. With the invention of the movie camera, the journey story was again revived, and its variations include everything from the comic (e.g., Buster Keaton’s The General) to the cowboy melodrama (e.g., Stagecoach, The Searchers) and the feminist buddy film (Thelma and Louise). When television came along, journey stories (Wagon Train, Route 66, The Fugitive, and, of course, Star Trek) were among the most successful series.Moving the journey story from the fairy tale to the novel meant moving it from a symbolic realm of universal actors (a king, a wicked stepmother) to a particularized social world and a particular time and place. In the novel the cruel things that happen to the hero are often treated as instances of a specific social injustice, like the English Poor Laws, rather than as the work of a generic antagonist like a big bad wolf. Moving the journey to the movies opened up the visual dimension of the archetype. Journey films often emphasize exotic landscapes, foreign cultures, and the lure of open spaces. Since television is best at portraying interior dramas and family-size social units, journey stories on TV generally focus on a succession of small communities or even replace the hero and sidekick with an entire traveling community, as on Star Trek. On the computer the journey story emphasizes navigation—the transitions between different places, the arrivals and departures—and the how-to’s of the hero’s repeated escapes from danger.
One of the consistent pleasures of the journey story in every time and every medium is the unfolding of solutions to seemingly impossible situations. We watch each new situation along the road and wonder how the hero will escape a beating or a hanging or a forced marriage or jailing. When Odysseus foolishly allows himself to be captured by the Cyclops, a huge, one-eyed man-eater, he is presented with a life-and-death riddle. The situation is carefully described so that it seems that he has no chance of getting out. The Cyclops is a brutal and heartless creature who brags that he is unafraid of Zeus and therefore free to do what he pleases. Every night two more of Odysseus’s men are eaten; the survivors know they must get out soon or die. They could kill the monster as he sleeps, but the cave is barred with a stone too heavy for them to move; if they kill him, they will never get out again. The Cyclops’s routine is unvaried: he goes off with the sheep in the morning, closing the cave behind him, and comes back at night with the sheep, ready for a dinner of Greek sailors. Then Odysseus (who is narrating the story) tells us how he solves the problem. He prepares some wine. He prepares a battering ram and gathers a group of strong helpers. He tells the giant his name is “Nobody” and gets him very drunk. When the Cyclops falls asleep, Odysseus and his men heat the battering ram in the fire and thrust it into his one terrible eye. Now the giant is blinded, but how will the Greeks get out? While the Cyclops is raging, Odysseus separates the sheep into groups of three and places each of his men under the middle sheep and himself beneath the strongest ram in the flock. Finally, the Cyclops lets out his sheep, as Odysseus has seen him do every morning, and out go all the Greeks with them. And when the wounded Cyclops cries out for help from his neighbors, whom does he say is tormenting him? “Nobody.” Odysseus’s description is constructed so that we can enjoy each individual step and gain increasing pleasure as the overall plan becomes clear. The story is as much a riddle as Oedipus’s, but the answer to the riddle is not in a single word; it is in a series of beautifully orchestrated steps, an elegant algorithm for defeating giants.
Computer-based journey stories offer a new way of savoring exactly this pleasure, a pleasure that is intensified by uniting the problem solving with the active process of navigation. On the computer the dramatic situation of capture and escape can be simulated by keeping the player within a confined space until the solution to a puzzle is found. These puzzles are most satisfying when the actions have a dramatic appropriateness, when they serve as a way of increasing our belief in the solidity and consistency of the illusory world. For instance, in Myst the wizard’s island includes an elevator hidden in a giant tree and operated by a nearby control panel. In addition to solving the puzzle of the panel, the interactor must move efficiently through the space to get to the elevator at just the right moment in its descent. The concreteness with which the space is detailed makes the sequence feel not like a test of coordination but like a dramatic moment. By contrast, in the computer game The Seventh Guest, the player is asked to cut up a cake into enough segments to match the number of murder victims. The puzzle is a satisfying one, but since there is no one there to eat the cake, the action takes us outside the immersive world instead of reinforcing our belief in it. The most dramatically satisfying puzzles are those that encourage the interactor to apply real-world thinking to the virtual world. For instance, a computationally sophisticated MIT student who is also an expert gamer instanced a particular dramatic moment from the text-based Zork II as among his lifetime favorites: The story involves a dragon that is slow to rouse but always lethal if you persist in fighting him. Elsewhere in the dungeon is a wall of ice that is impossible to pass. What you must do is attack the dragon enough to get his attention—but not so much that he “toasts” you—and then run and head for the wall of ice. The dragon follows, sees his reflection in the ice, and thinks it is another dragon. He rears up and breathes fire at his enemy; as he does so, the ice melts, drowning the dragon and eliminating the obstructing wall.6 Like Odysseus in the Cyclops’s cave, the player escapes by outsmarting a ferocious monster using only the materials at hand.
Games into Stories
Games seem on the face of it to be very different from stories and to offer opposing satisfactions. Stories do not require us to do anything except to pay attention as they are told.7 Games always involve some kind of activity and are often focused on the mastery of skills, whether the skill involves chess strategy or joystick twitching. Games generally use language only instrumentally (“checkmate,” “ball four”) rather than to convey subtleties of description or to communicate complex emotions. They offer a schematized and purposely reductive vision of the world. Most of all, games are goal directed and structured around turn taking and keeping score. All of this would seem to have nothing to do with stories.
In fact, narrative satisfaction can be directly opposed to game satisfaction, as the endings of Myst, widely hailed as the most artistically successful story puzzle of the early 1990s, make clear. The premise of the Myst story is the confinement of two brothers, Sirrus and Achenar, in magical books that serve as a dungeon. Through a video window we can see them in their imprisonment and hear them talking to us in short, staticky segments. Each one warns us about the wickedness of the other and asks us to rescue him. The brothers can only be freed by heroic labors of problem solving by the player, who must journey to four magical lands or ages and bring back a single page from each of them for either Sirrus or Achenar. Each time the player gives one of the brothers a magic page, he responds with a slightly clearer video segment. At the end of the game, when most of the puzzles have been solved, the player has most likely gone to each land twice in order to gather both sets of pages and to hear all of the messages from both brothers. At this point we are faced with a dramatic choice. The last magic page will release one or the other of them from the book. Which is it to be?
The game is well designed in that all the evidence on which to base a decision is, as in any good detective story, available to the player. Exploring the various lands reveals—through accusatory notes, hidden corpses, imperial furnishings, desolated landscapes, and multiple instruments of torture and destruction—the villainy of both brothers. The secret of the game is that although both brothers are evil, their father, Atrus, is alive and—with some more puzzle solving—can be found and rescued. The “winning” ending involves locating the good wizard Atrus and remembering to bring with you the magical item that will free him from captivity. This is a satisfyingly fair yet challenging mystery plot.
Yet surprisingly, the “losing” endings of the game are much more satisfying than the winning ending. In the winning ending one finds a beautifully rendered but dramatically inert video cutout of Atrus superimposed on a backdrop of a very shallow fantasyland. Unlike all the other lands visited during the game, this one is not really explorable and offers no pleasures of manipulation. It is a dead end. The ending in which you get to the wizard but forget to bring him the means of escape is more dramatic, because he gets quite angry at you. But the most dramatically satisfying endings are the near-identical losing branches, which are the result of choosing to rescue either of the evil brothers. The moment you release either Achenar or Sirrus from imprisonment, he will mockingly turn on you and lock you in the very same dungeon from which he has escaped! The visual effect is simple but brilliantly effective because it reverses your perspective. Throughout the game you have peered into each brother’s dungeon through a static-ridden, credit-card-size window embedded with the parchment page of an enchanted book. The brothers’ immobility has been marked by the fact that you could see little more than their faces. Now you are looking out through a similarly staticky window set into a totally black screen. Through the window you can see the evil brother now exultantly standing and moving around while looking down at you, just as you had looked down at him. A game that marked a breakthrough in ease of navigation appropriately ends by immobilizing the player.
The superiority of the losing endings of Myst suggest a basic opposition between game form and narrative form. How can we tell significant stories in a form that always has to end happily? How can we impose endings that yield complex story satisfactions on a form that is based on win/lose simplicity? Many would argue that computer-based narrative will always be gamelike and that such dissatisfactions are therefore inevitable. But when looked at more closely, games and stories are not necessarily opposed.
From: Murray, Janet H.. Hamlet on the Holodeck (pp. 126-142). Free Press. Kindle Edition.
