previous chapter
Marie-Laure Ryan
Fictional Worlds in the Digital Age
next chapter


Fictional Worlds in the Digital Age

Marie-Laure Ryan

Of all the pleasures of literature, none is more fulfilling to the embodied mind than immersing itself in a fictional world. Some works, admittedly, discourage this pleasure, and there are other kinds of satisfaction: "high brow" literature loves to distance itself from popular culture, which thrives on immersion, by promoting the more cerebral experiences of self-reflexivity and critical distance from the fictional world. But immersion remains the most fundamental of literary pleasures (Schaeffer 1999; Ryan 2001). It would be pointless to demystify textual worlds as constructed by language or other types of signs, if the imagination were not spontaneously inclined to pretend that these worlds are real, or, as the romantic poet Wordsworth put it, to "suspend disbelief" in their autonomous existence. Digital media have made important contributions to both immersion and self-reflexivity: whereas computer games absorb players for hours at a time into richly designed imaginary worlds, hypertext fiction explodes these worlds into textual shards, code poetry promotes awareness of the machine language that brings text the screen, and what Noah Wardrip-Fruin (2006) calls "process-intensive" works direct attention away from the surface of the screen, toward the virtuoso programming that generates the text. In the present essay, I propose to concentrate on the immersive pleasures, by asking: how do digital media affect the experience of fictional worlds and the practice of fiction? My investigation will lead beyond the narrow domain of literature, if by literature one understands primarily language-based art, to investigate forms of digital entertainment that rely on multiple sensory channels and semiotic supports.

As a preliminary, let me outline my conception of fiction. For this notion to be valid of a medium, it must possess a cognitive value. This means that the question "is it a fiction?" must influence the use of a text or the interpretation of a behavior. In Cigars of the Pharaoh (1975: 16–17), the famous comic book hero Tintin gives an eloquent demonstration of the danger of mistaking fiction for reality: witnessing a woman being savagely beaten in the Sahara desert, he rushes to her rescue, only to discover that he has stumbled upon a movie set. Rather than being thanked by the victim, he must suffer the wrath of the entire film crew. Similarly, a reader who mistakes a novel for a representation of reality (as did Don Quixote) will be led to mistaken beliefs. A text of fiction invites its users to imagine a world, while a text of non-fiction conveys information, whether accurate or not, about reality.

Relying on the work of John Searle (1975), Kendall Walton (1990), Jean-Marie Schaeffer (1999), and David Lewis (1978), I regard fiction as the product of an act of make-believe whose prototype can be found in children's role-playing games, such as playing house, cops and robbers, or big bad wolf chasing little pigs. Through their act of make-believe, readers, spectators, or players transport themselves in imagination from the world they regard as actual toward an alternative possible world — a virtual reality — which they regard as actual for the duration of their involvement in the text, game, or spectacle. Once transported into this world, they either enter the body of a specific individual (dramatic acting; playing big bad wolf; controlling an avatar in a computer game) or they pretend to be an anonymous member of the fictional world who receives the narration or observes the unfolding of fictionally real events (reading a novel, watching a play). I call this projection into a virtual body an imaginative recentering (Ryan 1991: 21–3), and I regard it as the precondition of the experience of immersion.

Central to this definition is the idea that a fictional text must be able to conjure a world to the imagination. By world I mean a space serving as container for concrete objects and individuated characters, obeying specific laws, and extending in time, at least implicitly. A fictional text cannot be a philosophical treaty dealing with abstract ideas, nor a description concerned exclusively with universals. It must say "Ivan Ilych is dying," rather than "all men are mortal." When a text explicitly represents the evolution of a fictional world through time, this text presents a narrative dimension. Most fictions do indeed tell stories, but narrativity and fictionality are theoretically distinct problems. The narrativity of a text is a semantic issue, this is to say, a matter of content, and the user can decide whether or not the text tells a story by simply decoding its meaning. Fictionality, by contrast, is a pragmatic issue: not a matter of what the text is about, but a matter of how the text is supposed to be used. There may be some semantic restrictions on the content of a fiction (as I suggest above, a fiction must project a concrete world), but there are no positive conditions that specify a certain type of subject matter. It follows that one cannot always pass judgments of fictionality by simply inspecting the text.

In this chapter I will examine fictional practices that take advantage, to variable degrees and in variable combinations, of the most distinctive properties of digital media: interactivity, multimedia capabilities, volatility of inscription, and above all networking. Underlying all these features, and making them possible, is a more fundamental property of digital media that I will not discuss separately because it is involved in all digital texts: the property of being algorithmic, i.e., operated by computer code (see Winder, Chapter 27, "Writing Machines," this volume). I will follow the implications of these properties not only for fictional worlds that exist in the digital medium — classic computer games, MOOs (MUD Object Oriented), web-based micronations, and multi-user online games — but also, in Section 3, for worlds originally created in the "old media" of print literature, cinema or TV.

1. The Pleasures of World-building

As a network, the internet is a collection of interlinked nodes, or sites, serving as recipients for information, and distinguished from each other by unique addresses. From the very beginning of networked computing, users have treated some of these sites as places to meet and as territories to colonize. While some meeting places, such as chatrooms, remain empty spaces, others have been elaborately built up and decorated, so as to offer a more congenial forum, to provide topics of conversation, or simply for the sheer joy of building and decorating (pleasures often denied to us in the real world for lack of funding). Through this creative activity, the sites of the internet become "worlds" to the imagination.

The earliest manifestations of the pleasure of world-building were the text-based MOOs and MUDs (Multi-User Dungeons) that flourished in the 1980s and 90s. MOOs were public spaces where users met under the disguise of a fictional persona, or "avatar," which they created themselves. (My use of the past tense suggests that text-based MOOs have been largely supplanted by graphically rendered worlds, such as Active Worlds [] or the online game Second Life [<>]). To enliven this game of make-believe, the administrators of the system designed a permanent setting, typically a building with many "rooms" furnished with virtual objects. Avatars and objects were created in the same way a novelist brings a world to life: by posting on the system their textual description. Advanced players acquired the privilege of building their own room, and of decorating it with objects of their own invention. Upon entering a room, visitors were able to explore it by typing commands such as "look around." The system would display a list of the objects contained in the room, and the players could continue their inspection by typing instructions such as "look [name of object]," to which the system would respond by posting a more elaborate description, or by activating a behavior encoded in the object (for instance, exploding). The presence of objects created by other members of the fictional world gave players the pleasure of exploration and the thrill of discovery, while creating their own objects provided them with props in their games of make-believe: for instance, users (or rather, their avatars) could get married in the fictional world of the MOO by creating and exchanging a ring, and they could build their own love nest.

Whereas the MOOs and their graphic successors are ready-made spaces maintained by professional programmers and waiting to be filled with content created by users, another non-commercial form of online world, the micronation, is built from scratch by amateurs. An offshoot of the literary genre of the utopia, a micronation is an independent "state" created by individuals, who design its geography, define its form of government, write its laws, describe its customs, and invent its history. Micronations differ from those literary works whose main source of interest lies in the creation of exotic worlds — science fiction, the fantastic — through their lack of plot. It is not necessary to be a novelist to create an online world. Whereas literary works dramatize their imaginary worlds through a narrative line that follows the travels of a character and details his discoveries — Gulliver's Travels being the prototype — micronations are collections of mainly descriptive documents, and it is the user exploring the website who plays the role of traveler.

While some micronations exist in real space, by the will of a founder who declares the independence of a certain territory (a phenomenon documented by the Lonely Planet guide to micronations [2006]), most of these fictional countries reside on the internet, where they can be reached by cyber tourists and prospective citizens through a click of the mouse. They are named Bergonia and Talossa, Uteged and Reuniao, Lizbekistan and Aerica, Freedonia and Aristasia, the Feminine Empire. The reasons for building micronations are as varied as their landscapes, customs, and political systems: childhood dreams of inhabiting secret kingdoms, adolescent revolts against a world ruled by grown-ups, exercising royal power, articulating social visions, simulating processes of self-government, but above all the pure pleasure of writing the encyclopedia of an imaginary country.1

To illustrate the diverse motivations that drive the builders of micronations, let's take a look at two of them, Bergonia and Talossa. The founder of Bergonia2 has a clear political agenda: promote socialism and environmentalism by describing a viable alternative to our capitalist economy, which he regards as "certain to dehumanize humankind and likely to end up wrecking the planet, with catastrophic results of epochal proportions." But Bergonia is also a private game, the product of an enduring childhood's fantasy. Like Robert Louis Stevenson's novel Treasure Island, it was born out of a love of maps and of drawing maps. "In the beginning," writes the founder,

I had no purpose at all, only the joy of playing and inventing I'm sure that the discovery of my parents' grand atlas was the immediate hook. Drawing the first maps and dreaming up weird names was entirely a matter of play. I drew lots of maps when I was a kid, and Bergonia was then just one of many on-going projects. Over the years the work (play) of contriving a continent & nation has massively challenged me — a great expenditure and demonstration of imagination that wound up serving a multitude of incidental uses. The job has given me a vehicle and a focus for all my interests in geography & meteorology, anthropology & history, religion & mythology, and philosophy & psychology. (<>)

The texts posted on the Bergonia website represent an encyclopedic sum of knowledge: where else, but in an imaginary country, can one be at the same time an ethnographer, geographer, political scientist, linguist, cartographer, historian, and ecologist? But despite its aspirations to build a better society, Bergonia remains the creation of a single individual, and except for an invitation to send email to the author, it does not encourage active participation.

Like Bergonia, Talossa originated in a young person's fantasy: at age fourteen, its founder proclaimed his bedroom to be a foreign nation called Talossa, Finnish for "inside the house." Less thoroughly imagined than Bergonia, Talossa is defined by a made-up language (a blend of English and Romance language, with lots of strange diacritical marks), a history, a culture, a real-world territory (located around Milwaukee, but divided into provinces with invented names), a law, a constitution, a government, a system of three political parties, and a series of elective political offices such as prime minister, secretary of State, and minister of culture. As in most micronations, the rulers of Talossa revel in the insignia of statehood (a flag, a slogan, a national anthem, a system of nobility titles), as well as in a pompous legal language that stamps the decisions of the government with the seal of authority. Originally conceived as a constitutional monarchy with an elected King, Talossa underwent a revolution in 2004 and split into a Kingdom and a Republic because, according to the webpage of the Republic, the citizens spent most of their time arguing uselessly with the King. In contrast to the largely non-interactive site of Bergonia, Talossa is an active social forum. Visitors are invited to apply for citizenship by filling in an online form (there are currently 88 citizens in the kingdom), and they are expected to participate in the political life of the country through a blog that serves both the Kingdom and the Republic.

But the degree of participation in such worlds remains very low, compared to either the MOOs or online games of the EverQuest, World of Warcraft, and Second Life variety, because they do not allow real-time interaction between their members. All the communication takes place through a blog or through email, and the possibilities of taking an active part in the political life of the nation are limited to voting on topics such as changes in the constitution, who will hold public offices, and the acceptance of new citizens. Holding an elected office means little more than having one's picture posted on the website with a fancy title. Micronations are not dynamic environments but collections of static texts that express the creativity of their founder, and once you have taken a tour there is not much else to do.4 This explains why their population remains in the double digits, compared to the thousands, if not millions, who join the virtual worlds of online games.

2. Worlds as Playgrounds

It is everybody's secret that the interactivity of digital media allows their most important and most popular contribution to the experience of fictional worlds, namely the development of video games. In a classical print fiction, users play the role of a passive witness of the represented events, but in most video games, they impersonate and control a character who takes an active part in the evolution of a fictional world. Through this act of pretense, video games not only strengthen the connection between fiction and children's games of make-believe, they also reconcile two types of game which had previously remained largely incompatible with each other. These two types of game are what the French sociologist Roger Caillois calls paidia and ludus:

At one extreme an almost indivisible principle, common to diversion, turbulence, free improvisation, and careful gaiety is dominant. It manifests a kind of uncontrolled fantasy that can be designated by the term paidia. At the opposite extreme … there is a growing tendency to bind [this uncontrolled fantasy] with arbitrary, imperative, and purposely tedious conventions … This latter principle is completely impractical, even though it requires an ever greater amount of effort, patience, skill, or ingenuity. I call this second component ludus.

(2001: 13)

The best example of paidia games is building imaginary scenarios with toys, using them, in the words of Kendall Walton 1990: (21–24 and throughout), as "props in a game of make-believe." These games do not aim at a specific goal, and they do not lead to losing or winning. The pleasures of paidia reside in the free play of the imagination, in adopting foreign identities, in forming social relations, in building objects, in exploring an environment, and above all in creating a representation: paidia games are fundamentally mimetic activities. If there are rules, they are spontaneously created by the participants, as when a group of children decides that a certain tree will be the house of the wolf, and they can be renegotiated on the fly. Ludus games, by contrast, are strictly controlled by pre-existing rules accepted by the participants as part of a basic game contract, they lead to clearly defined states of winning or losing, and their pleasure resides in the thrill of competition and in the satisfaction of solving problems.

Of these two types of game, it is clearly paidia, with its mimetic dimension, that forms the ludic origin of fiction. Caillois goes on to say that the make-believe of fiction is incompatible with the rules of ludus: "Thus games are not ruled and make-believe. Rather they are ruled or make-believe" (2001: 9). (The original French text reads: "Ainsi, les jeux ne sont pas réglés et fictifs. Ils sont plutôt ou réglés ou fictifs" [1958: 41].) In the 1950s, when Caillois was writing, the vast majority of ruled games were abstract: chess, go, cross-word puzzles, and the various forms of competitive sport take place on a playfield divided into strategic zones (the squares on a chess board, the penalty box on a soccer field), but this playfield does not represent anything: it is not a world in an imaginative sense. The goals of the players are not the kind of events that matter to people in practical life, but actions only made desirable by the conventions of the game, such as capturing the opponent's pieces, or shooting a ball into a net. There had been, admittedly, some attempts to render board games more interesting from a thematic point of view: for instance Monopoly simulates real estate, many dice games enliven their board with pictures that tell a rudimentary story, and in Dungeons and Dragons, players impersonate characters and pursue quests whose outcome is decided by rules set up by the game master. But it is only with computer games that the conflict observed by Caillois is fully resolved. More and more, video games take place in a space that is not only a playfield but a richly designed world offering diverse landscapes to explore and identifiable objects to manipulate. Players do not move tokens but impersonate an avatar, and their goal is no longer the achievement of a conventional state of winning but the fulfillment of concrete tasks corresponding to what people might want to do in real life if they were placed in the proper world and in the proper situation: goals such as winning wars, saving the Earth from invaders from outer space, or establishing civilizations. On the other hand, these games are even more strictly controlled by rules than sports and board games, because the rules are established by code, and they are as imperative as the laws of physics: you can no more cheat with them than you can cheat with gravity. (Cheating in computer games is finding imaginative ways to get around the code, rather than transgressing it.)

In the fictional worlds of computer games players no longer have to choose between an activity of make-believe that speaks to the imagining imagination and an activity of problem-solving that relies on the strategic imagination. Video games are both "ruled and make-believe," and it is now possible to engage in both ludus and paidia within the same world. Some of them, like the shooter Doom, are admittedly more ludus than paidia, fiercer, more narrowly focused competition than make-believe and the pleasure taken in simply inhabiting the fictional world. These games are for the type of player that Richard Bartle (1996), in a classic typology, calls "killers" and "achievers," as opposed to his "explorers" and "socializers." Conversely, so-called simulation games, such as The Sims, SimCity, or Civilization, are exclusively paidia. But a single-user game like Grand Theft Auto (Bogost 2006), or a multi-user online game like EverQuest, combines both types of pleasures: competition and problem-solving through the "quests" given to the players; but also, depending on the particular game, chatting with other players, building avatars, acquiring or creating valuable objects for the avatar, exploring the diverse landscapes of the game world, and engaging in activities for the fun of it. The worlds of such games are both territories to traverse in search of missions to accomplish and spaces for flânerie; both combat zones full of challenges and playgrounds filled with a variety of toys.

3. Expandable Worlds and Worlds out of Worlds

The formation of computer networks, together with the volatility of inscription of digitized information, allows fictional worlds to grow, to be modified from the inside and the outside, and to give birth to other worlds. By volatility of inscription, I mean that digital texts are the projection on a screen of a binary information held in the computer's memory that can change value with every clock cycle of the machine. Digital texts thus differ from works inscribed on a static support, such as books or film reels, through the impermanence of their material inscription: it is not necessary to manufacture a brand-new copy to update them. This property is not specific to fiction, as we can see from the constant growth and rewriting of the online encyclopedia Wikipedia, or from the possibility to revise endlessly texts composed with a word processor. But the volatility of digital texts has particularly important implications for the fictional worlds of video games, especially when it is coupled with networking. It makes it possible to modify the operation of a video game by downloading from the web expansions, modifications ("mods" in the jargon), or error corrections known as "patches" (Salen and Zimmerman 2003). Various expansions to the world of The Sims allow, for instance, the characters of the game to leave their suburban homes for exotic vacation spots, to live the life of college students, to adopt pets, or to enjoy an exciting nightlife. Whereas the expansion packages of The Sims are commercial products developed by EA, the company that markets the game, many mods are the spontaneous creation of amateur players who make them available for free to the player community.

When game worlds reside on the internet, players can take advantage of updates without having to download a module. In online games, the combination of networking and volatile inscription enables the production team to expand the territory of the game world, to introduce new active objects, or to create new missions when some players have attained the highest level, so as to keep their interest and more importantly, continue collecting their monthly membership fees. In contrast to the worlds of CD-ROM-based games, online worlds can, consequently, be modified unbeknownst to the players.

The expansions described so far operate "from the inside," since they are seamlessly integrated into the fictional worlds of games. But by providing public forums that lead to the formation of communities, the internet also facilitates a kind of rewriting that takes place from the outside. This phenomenon, known as "transfictionality" (Saint-Gelais 2005; Ryan 2006), consists of producing and posting texts that complete, modify, or stretch in time the worlds of preexisting literary texts, or that transpose their plots and characters into new environments. Transfictionality expresses the reader's desire to free the fictional world from the control of the author and to see it live and evolve independently of the text that originally brought it to life. By offering familiar worlds into which readers can immerse themselves from the very beginning, rather than having to slowly construct these worlds on the basis of textual information, transfictionality represents a substantial saving in cognitive effort. This explains its popularity with texts of popular culture. The phenomenon is admittedly not specific to the digital age: in oral cultures, the performance of bards brought infinite variations to worlds familiar to the audience; in the age of print, popular novels such as Don Quixote and Robinson Crusoe inspired numerous imitations and apocryphal continuations; and in the early days of television and science-fiction, fans exchanged plot suggestions through photocopied publications known as "zines" (Hellekson and Busse 2006; Jenkins 2006). But with the internet, the production of transfictional texts flourished into a mass industry thanks to communicative advantages too obvious to merit discussion here. Popular TV series (Buffy the Vampire Slayer, Xena: Warrior Princess, Star Trek), as well as films (Star Wars) and cult novels (Harry Potter, Lord of The Rings), give birth to numerous websites where fans post stories based on the fictional world of the text that forms the object of the cult. Other fans provide comments, which may lead to revisions. There are, for instance, 38,000 texts of fan fiction inspired by the world of Harry Potter on the website www.harrypotter Energized by what Henry Jenkins (2006) calls the collective intelligence of the fan community, fictional worlds become interactive and participatory in a much more imaginative way than in hypertext or even video games, even though the individual texts do not contain interactive devices.

The computer provides not only a channel of transmission for the texts of fan fiction, it can also become a tool of production. It all began with the built-in game camera of The Sims. By taking snapshots of the screen, and by combining them with text of their own invention, players create comic strips (or graphic novels) that expand the rather limited narrative possibilities of the game. There is indeed no need for these works to reproduce actual game scenarios. Thanks to the representational power of language, it is possible to make the characters talk (while in The Sims they converse in "Sim-speak," an incomprehensible jargon); to represent their private thoughts (only shown as numeric coefficients of likes and dislikes in the game); and to have them perform other actions than those offered on the menus (for instance, violent and antisocial actions). By selecting individual frames out of the game's continuous flow of images, authors are also able to skip the repetitive events of daily life, such as taking showers or going to bed at regular intervals, that account for a major part of The Sims' gameplay, and to concentrate instead on events of much greater tellability. Thousands of these creations are posted on The Sims' official website.5 A related phenomenon is the manipulation of the engine that operates video games. (Some games, but not all, make this engine available to the players.) Through their ability to capture animated sequences, these engines allow players to take control of the characters and to produce short films known as machinima by adding a sound track of their choice (Jones 2006: 261). Artists have created machinima with original plots out of Quake, Half-Life, The Sims 2, EverQuest, Worlds of Warcraft, and Grand Theft Auto.6 With both still cameras and machinima, the user's creative freedom is only limited by the images provided by the game. The original game world becomes a quarry of visual materials, a matrix out of which players generate other worlds. Lost in the process, however, is the interactive character of the source world.

4. Living Worlds

When imaginary worlds exploit to the maximum the four properties of digital technology mentioned above, namely interactivity, multi-sensory dimensions, volatility of inscription, and networking, they become dynamic spaces where events are constantly happening and where fiction is continually being produced. These living environments are known in the jargon as MMORPGs ("massively multi-user online role-playing games"), but they have also been called synthetic worlds (Castronova 2005), virtual worlds (Klastrup n.d.), and gaming lifeworlds (Taylor 2006). Their names are EverQuest and Ultima Online, Second Life and World of Warcraft, Lineage and Entropia. Sacrificing precision to convenience, I will simply refer to them as "online worlds" or "online games." The remarks below are those of a lurker, rather than of an active player, and they are highly indebted to the vivid descriptions found in the growing scholarship devoted to these words.7

How does the world of an online game operate? Drawing inspiration from a list established by Castronova 2005, (chapter 4),8 I reduce their basic components to the following elements:

1.  The visual representation of a vast geography with diverse landscapes that offers specific possibilities of action: for instance, dangerous forests inhabited by monsters whose killing earns the players experience points, sanctuaries where players are immune to attacks (much as churches were in the Middle Ages), convalescent houses or magic springs where players heal their wounds, markets where they trade goods, and mountains where they can mines precious metals. The Atlas of Ever-Quest, a 500-page tome, provides a detailed guide to the symbolic geography of the game, listing for each area its back story, the dangers to be expected, the challenges being offered, and the benefits to be acquired by passing the tests.

2.  A repertory of roles to choose from, divided into races (for instance gnomes, elves, humans), genders, and classes (druids, knights, necromancers). These types differ from each other through their abilities: a druid can heal wounds, a knight excels at sword fighting, a gnome prefer to defeats enemies through ruse than through brute strength. These diversified talents encourage players to form social bonds (known as guilds) with other players, in order to complement their abilities. The higher one progresses in the game, and the more difficult the tasks to be accomplished, the more imperative it becomes to ally oneself with other characters.

3.  A system of progression, based on the acquisition of merit. All the players start, at least in principle (we will see an exception below), from the same lowly condition, but depending on their skills, and above all on the time invested in the game world, they reach different levels in the game hierarchy. The mechanism of leveling consists of the acquisition of experience points by performing various tasks, typically slaying monsters. Belonging to a higher level not only increases the power of a players to perform difficult tasks, it also gives them greater prestige in the game community.

Online worlds differ from those of standard narrative — film or novels — as well as from single-player, CD-ROM-supported games through a quality of persistence that takes the simulation of real life to unprecedented heights. An online world is like a TV show that runs every day, twenty-four hours a day, and if players play hooky in the real world, they will return to a changed world, like Rip van Winkle awakening from his hundred-year sleep. As T. L. Taylor observes: "A player who is away from the game for a couple of weeks may find that his level-25 character can no longer group with his friends' characters, which have now advanced to level 40" (2006: 49).

Another aspect of real life that multiplayer online games simulate very efficiently, while standard narrative does not, is our limited perception of the world we live in. In a novel representing a huge world with a large cast of characters, parallel lives, and intersecting destinies, the reader is informed of all the plot-lines through an omniscient narrator who can move freely in time and space. In an online world, by contrast, as is the case in real life, players are only aware of what happens in their immediate surroundings. Other players may occasionally inform them of events that took place in remote regions, but no player has a god's eye view of all the destinies that are simultaneously unfolding in the game world.

In one respect, however, online worlds differ significantly from both life and traditional narratives. In the real world, once you successfully perform an action to solve a problem, this action has durable consequences and does not need to be repeated, unless it is negated by another agent, or belongs to the daily maintenance type such as eating or sleeping. The same principle holds for narratives: when a knight slays a dragon in a fairy tale, the dragon is dead, and the feat does not need to be performed again by another knight. But in an online world, events have only very limited consequences. Not only is death reversible (players can retrieve the corpse of their avatar, collect its belongings, and bring it back to life, losing no more than a few experience points), but, as Fotis Jannidis (2007) has observed, the quests performed by players are endlessly repeatable by other players. In online worlds, history doesn't have to be definitive: if you kill a monster, he re-spawns in a few seconds; if you dig out gold from a mountain, the gold remains available for other players; otherwise, the earlier settlers would deplete the resources of the game world. Here the principle of causality, which forms the basic mode of operation of both nature and narrative, is sacrificed to the need to give equal opportunities to all players. But since players are ignorant of what happens elsewhere in the game world, the relative futility of their actions does not seriously damage their sense of being engaged in a meaningful pursuit.

Of all the distinctive properties of online worlds, none has been more highly praised than the feedback loop that takes place between developers and players. Exploiting the conditions imposed top-down by the developers of the game, players improvise bottom-up their own scenarios by communicating with each other. The game developers react to these behaviors by introducing new code that either facilitates them, when they enhance the interest of the game, or discourages them, when they diminish the pleasure of most players. As example of this emergent quality is the development of an economy in the world of EverQuest. By creating a monetary system and by placing NPC (non-playing character) merchants in the game world, the coding authorities put into place the foundations of a trade system. Through this commerce, players are able to sell the loot they acquire by slaying monsters and to buy more desirable commodities that either display their high status to other players, or let them challenge more ferocious monsters, in an endless loop that takes them through the levels of the game. But rather than trading with roaming NPCs, players soon developed the habit of meeting in a certain place, known as the Tunnel of Rô, to do commerce among themselves. To help them in this activity, the developers built an official bazaar, and the tunnel quickly lost its business (Klastrup n.d.: 314). The players took advantage of this new situation by creating robot-salesmen that represent them in the bazaar even when they are away from the computer. But the bazaar became so crowded with vendors and buyers that all the activity cluttered the display, overtaxed the system, and made response very slow. To remedy the problem, the developers redesigned the bazaar, so that "instead of showing all the vendors selling their goods, the screen shows only the character's avatar in a large hall, maybe with a few other shoppers, and the one vendor chosen to purchase from" (Taylor 2006: 64). Now the ball is in the players' court: who knows what they will do next?

There are times, however, when the feedback loop breaks down, leaving the developers unable to prevent activities that are detrimental to the enjoyment of other players. The most celebrated case of a loss of control over player behavior is the practice of buying objects in the game world with real-world money. As Castronova writes, "World builders always intended for goods and services to have prices and markets inside the world in term of gold pieces and credits and so on, but they seem not to have anticipated that these things would acquire robust markets in dollars outside the world" (2005: 149). Game developers and most players disapprove of this activity because it destroys the equality of opportunities that is so fundamental to the operation of games: a rich player who does not intend to devote much time and effort to the game world can rise instantly to the highest level by buying an avatar on eBay. It is the digital age equivalent of the practice that drew the ire of Martin Luther: gaining easy access to Heaven by buying indulgences rather than through faith and good works. As Julian Dibbell has chronicled in his entertaining book Play Money, Or, How I Quit my Day Job and Made Millions Trading Virtual Loot (2006), some people make a living manufacturing or collecting objects in online worlds, in order to sell them for real-world money. It has always been possible to work within fictional worlds, for instance as a professional actor or by parading as Mickey Mouse in Disneyland, but this possibility to create capital within a fictional world is unique to online games. Once the virtual objects produced in these games acquire real-world value, the inevitable next step is the creation of an exchange rate between game money (for instance, the platinum pieces of EverQuest) and real-world currencies. According to the New York Times,9 the developers of an online game named Entropia plan to introduce a credit card that will allow players to obtain US dollars from ATMs (automated teller machines) by deducting the desired amount from their account of Entropia dollars.

5. Online Worlds between Fiction and Reality

I mention the practice of selling virtual objects for real-world money not only to illustrate the autonomy of the inhabitants of online worlds vis-à-vis game developers, but also to introduce the issue that makes these worlds particularly interesting from a theoretical point of view: the validity of the distinction between fiction and reality. Western thought has a long tradition of questioning of the reality of the real: we find it in Plato, with the myth of the cave; in the Baroque age, with the idea that the sensible world is an illusory dream; and nowadays, in the philosophy of Jean Baudrillard, who claims that the real has been replaced by its simulation. While for Plato and the Baroque the real resides in a transcendental sphere situated beyond the realm of appearances, for Baudrillard the only reality is that of the image. But with online games, the ontological question takes a new twist: rather than exposing the real as a fiction, the trend is now to ask whether the fictional worlds of online games should be regarded as an authentic form of reality. It is not a matter of claiming that online worlds have replaced the lifeworld, as Baudrillard would have it, but rather of according the same reality status to both worlds. This can be done in two ways: by regarding online worlds and the lifeworld as separate but equal in ontological status, or by questioning the validity of the boundary that separates them.

The "separate but equal" theory prevails in psychological approaches to the online world phenomenon. Sherry Turkle's book Life on the Screen (1995) contains several testimonies of users who consider their avatar to form a part of their identity as authentic as their "RL" (real life) self. An informant tells her: "RL is just one more window, and it's usually not my best one" (1995: 13). Another asks: "Why grant such superior status to the self that has the body when the selves that don't have bodies are able to have different kinds of experience?" (1995: 14). For these players, as for the French poet Arthur Rimbaud, "true life is elsewhere." But these declarations may be more judgments of quality than ontological pronouncements: people would not be so eager to proclaim the reality of online life if they did not derive more enjoyment from it than from everyday life. The most fanatic players know that the game world is an image, and that this image is unable to nourish the body. Imagine a society where half of the people spend all of their time playing online games, while the other half takes care of their bodily needs by supplying food, drink, and other necessities. Even in such a society, the supplies that maintain the bodies of the players alive could be produced in only one of the two worlds. Moreover, the developers of online games could pull the plug on their worlds, and they would disappear without trace, but the members of online games could not pull the plug on the real world. This asymmetrical relation makes it amply clear that the two worlds do not belong to the same ontological level, even if many players find an essential mean of self-fulfillment in the creation of imaginary identities. But don't we find similar self-fulfillment in private fantasies, and in the virtual world of traditional fiction?

If the worlds of online games cannot achieve reality in an ontological sense, can they do so in an economic sense? Here the argument does not treat online worlds and the lifeworld as distinct yet equal, but rather, emphasizes the permeability of their borders. Espen Aarseth (2005) thinks that the currency of online games is real, in contrast to the money of a game like Monopoly, because it can be exchanged for dollars, yen, or euros. But what does it mean to call money real? In contrast to commodities that have both a use and an exchange value, money is only desirable because of its buying power. As coins and bills disappear in favor of plastic cards that represent numbers in an account, it becomes more and more obvious that money is a purely virtual entity. Moreover, if the possibility of trading game-money for real-world currency made this money real, the same could be said of all the pixel-made objects acquired or built by players within the game, since these objects can also be sold for dollars, yen, or euros. But we would never call a horse in an online world real, because it is a mere image. What may be regarded as real is not the objects themselves but rather their value for other players. This value can be either a genuine use value within the game (for instance that of a fancy horse), or the exchange value typical of game currency. As Castronova writes: "when most people agree that [a] thing has a real value to somebody, it really has that value" (2005: 148). This distinction between "real object" and "real value" marks the difference between ontological and economic reality.10

Does the existence of real values in online worlds mean that their economy could merge with the economy of the real world? Let's consider what would happen if the currency of online worlds disappeared entirely, and all transactions took place in real-world money. If the developers of the online world had the right to make money and to freely distribute some amounts to players by changing the numbers in their accounts, this would create serious disruptions of the global economy. Normally the production of money is strictly controlled by the Federal Reserve or any similar state-run institution. By putting money on the market the game masters would become a rival Federal Reserve; but then, on what ground could anybody else be prevented from doing the same thing? There would be more and more money on the market but not more commodities with a real-world use value, since game objects are useless outside the game. This imbalance would lead to uncontrollable inflation. On the other hand, if money could not be produced within the game, the buying power of the players would correspond to what they already own in the real world, and the game world would become nothing more than an online store for virtual products. This is the situation to which the current habit of buying avatars and other game objects for real money would lead, if it were generalized, rather than remaining a very restricted niche activity existing side by side with trade using the game's own currency. The membrane that surrounds online worlds cannot be entirely removed without ruining their ludic dimension; for, as Huizinga (1950) observed, what makes a game a game is the magic circle that separates it from reality.

It is not necessary to regard online worlds as an extension of reality to do profitable commerce with them. Since the dawn of civilization, trade has opened doors in political and natural boundaries while keeping them solidly in place. With online worlds, exchanges occasionally cross ontological boundaries. But there remains an important difference between trade across political boundaries and trade across ontological boundaries. In the first type of commerce, commodities can be transported to the other side of the border, while in the second type, traded goods cannot leave their world of origin. Objects manufactured in an online world are made of bits and pixels, and they only exist through the code that supports their native environment. Moreover, as I have already noted, their use value is entirely linked to a particular world: an avatar from EverQuest can no more be imported into the world of Ultima Online than a real-world body can immigrate into the world of EverQuest.11 In the commerce between the actual world and online worlds, there are consequently no direct exchanges, but rather, two distinct transactions. Within the game world, Orlando gives a magic sword to Furioso, while in the real world, John, who plays Furioso, transfers a certain sum of money into the account of Mary, who holds the strings of Orlando. To foster these parallel, mutually compensating transactions by which one world reacts to what happens in the other, the boundaries between game and reality need to be not so much porous as transparent.

What, in the end, is the reality status of online worlds? It all depends on what we understand by "real," but the meaning of the word varies with the nouns it qualifies. When it applies to an object or body, it stresses its materiality, its solidity, and it contrasts with its image. When it refers to emotions, it suggests their authenticity and sincerity, and it contrasts with pretending. When it qualifies the economic value of an object, it means that there are real people (in sense 1) who want this object, and it contrasts with lack of demand. When it describes a world as a whole, it refers to its autonomous existence and it contrasts with man-made, fictional, virtual, or simulated. The worlds of online games contain objects with real value; advancing in the game or falling in love with another player may trigger real emotions; stepping out of role, players may discuss the affairs of the real world; but what online worlds will forever lack is reality as autonomous, material existence. There cannot be a univocal answer to the question "are online worlds real?": they contain both real and unreal elements.


1  Jorge Luis Borges offers the literary model of this encyclopedic activity in his story Tlön, Uqbar, Orbis Tertius, where he describes the collective creation of Tlön, an imaginary county, by a "secret society of astronomers, biologists, engineers, metaphysicians, poets, chemists, algebrists, moralists, painters, geometers" (72): "I now held in my hands a vast and systematic fragment of the entire history of an unknown planet, with its architecture and its playing cards, the horror of its mythologies and the murmur of its tongues, its emperors and its seas, its minerals and its birds and fishes, its theological and metaphysical controversies — all joined, articulated, coherent, and with no visible doctrinal purpose or hint of parody" (71–2).

2   <>.

3  Kingdom of Talossa: <>; Republic of Talossa: <>.

4  The micronation of Aristasia, the Feminine Empires (<>), remedies this situation by inviting its members to meet in the online game world of Second Life. But the fact that live interaction has to take place in another world only underscores the inability of micronations to sustain active participation.

5  At: <>.

6  See <>.

7  I am particularly indebted to Fotis Jannidis for sharing his personal experiences in the world of EverQuest.

8  In the chapter in question, Castronova discusses the following dimensions of online worlds: Roles, Advancement, Status, Risk and Danger, Scarcity and Forced Cooperation, Messaging, Personality Content and AI.

9  < _r 1&oref slogin>.

10  The whole passage indicates, however, that Castronova is not aware of a difference between ontological and economic reality: "When a society in cyberspace hold that a certain glowing sword is really and truly magical, in the sense of having great and extraordinary powers, the judgment is not only impossible to deny within the membrane [of the online world], but it starts to affect judgments outside the membrane too. By this process, virtual things become real things; when most people agree that the thing has a real value to somebody, it really has that value" (2005: 148; my emphasis). The words in italics suggest, wrongly in my opinion, that real value implies real existence.

11  The only kind of object that can be brought into online worlds, or taken out of them, is digitized pictures or sound files. Some people have been known to sell their digital artworks in the world of Second Life. The buyer can either use the picture to decorate his virtual castle, or have it printed and framed for his real-world house. (See < page 2>.) What cannot be done, however, is to import a picture of a horse into a game world in order to provide a mount for an avatar. An imported image can only be used as an image.


Aarseth, Espen (2005). "Doors and Perception: Fiction vs. Simulation in Games." Proceedings of Digital Arts and Culture Conference, ITU, December 1–3, pp. 59–62.

Bartle, Richard (1996). "Hearts, Clubs, Diamonds, Spades: Players Who Suit Muds." <>.

Blumberg, Alex (n.d.). "It's Good to be King." <>.

Bogost, Ian (2006). Unit Operations: An Approach to Videogame Criticism. Cambridge, MA: MIT Press.

Borges, Jorge Luis (1998 [1941]). "Tlön, Uqbar, Orbis Tertius." In Collected Fictions (Andrew Hurley, Trans.). New York: Penguin, pp. 68–81.

Caillois, Roger (2001 [1958]). Man, Play and Games. Urbana: University of Illinois Press. French edition: Les Jeux et les hommes. Le masque et le vertige. Paris: Gallimard, 1958.

Castronova, Edward (2005). Synthetic Worlds: The Business and Culture of Online Games. Chicago: University of Chicago Press.

Dibbell, Julian (2006). Play Money, Or, How I Quit my Day Job and Made Millions Trading Virtual Loot. New York: Basic Books.

EverQuest Atlas (n.d.). Sony Entertainment Co.

Hellekson, Karen, and Kristina Busse (Eds.) (2006). Fan Fictions and Fan Communities in the Age of the Internet. Jefferson, NC: McFarland and Co.

Herge Á (1975 [1955]). Cigars of the Pharaoh. Boston: Little, Brown.

Huizinga, Johan (1950). Homo Ludens: A Study of the Play-element in Culture. Boston: Beacon Press.

Jannidis, Fotis (2007). "Event-Sequences, Plot and Narration in Computer Games." In Peter Gendolla and Jörgen Schäfer (Eds.). The Aesthetics of Net Literature: Writing, Reading and Playing in Programmable Media. Bielefeld: transcript Verlag, pp. 281–305.

Jenkins, Henry (2006). "Interactive Audiences? The 'Collective Intelligence' of Media Fans." In Fans, Bloggers, and Gamers. New York: New York University Press, pp. 134–51.

Jones, Robert (2006). "From Shooting Monsters to Shooting Movies: Machinima and the Transformative Play of Videogame Fan Culture." In Karen Hellekson and Kristina Busse (Eds.). Fan Fictions and Fan Communities in the Age of the Internet. Jefferson, NC: McFarland and Co., pp. 261–80.

Klastrup, Lisbeth (n.d.). Towards a Poetics of Virtual Worlds – Multi-User Texuality and the Emergence of Story. Unpublished PhD dissertation, IT University Copenhagen.

Lewis, David (1978). "Truth in Fiction." American Philosophical Quarterly 15: 37–46.

"Micronations" (n.d.). Wikipedia, The Free Encyclopedia. <>.

Ryan, John, George Dunford, and Simon Sellars (2006). Lonely Planet Micronations. Lonely Planet Publications.

Ryan, Marie-Laure (1991). Possible Worlds, Artificial Intelligence, and Narrative Theory. Bloomington: University of Indiana Press.

Ryan, Marie-Laure (2001). Narrative as Virtual Reality: Immersion and Interactivity in Literature and Digital Media. Baltimore: Johns Hopkins University Press.

Ryan, Marie-Laure (2007). "Transfictionality Across Media." In John Pier and José Angel García (Eds.). Theorizing Narrativity. Berlin: Walter de Gruyter.

Saint-Gelais, Richard (2005). "Transfictionality." In David Herman, Manfred Jahn, and Marie-Laure Ryan (Eds.). The Routledge Encyclopedia of Narrative Theory. London: Routledge, pp. 612–13.

Salen, Katie, and Eric Zimerman (2003). Rules of Play: Game Design Fundamentals. Cambridge, MA: MIT Press.

Schaeffer, Jean Marie (1999). Pourquoi la fiction? Paris: Seuil.

Schiesel, Seth (2006). "Entropia Universe Players Can Cash Their Online Earnings at the A. T. M." < 1&oref slogin>.

Searle, John (1975). "The Logical Status of Fictional Discourse." New Literary History 6: 319–32.

Taylor, T. L. (2006). Play Between Worlds: Exploring Online Game Culture. Cambridge, MA: MIT Press.

Turkle, Sherry (1995). Life on the Screen: Identity in the Age of the Internet. New York: Simon & Schuster.

Walton, Kendall (1990). Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, MA: Harvard University Press.

Wardrip-Fruin, Noah (2006). Expressive Processing: On Process-Intensive Literature and Digital Media. PhD dissertation, Brown University. <>.

previous chapter
Marie-Laure Ryan
Fictional Worlds in the Digital Age
next chapter