Psychology

Anyone with the remotest interest in the psychology of players of virtual worlds must—that’s must—look at John Suler’s hypertext book ¹, The Psychology of Cyberspace. Much of it is 100% relevant to virtual worlds (he talks about them directly), and some sections are classics in their own right. Others are merely short monographs, but almost no subfield of Internet-related psychology goes untouched. Suler has something to say about everything I mention in the remainder of this review, and more besides. There are other, more cohesive and directed introductions to the subject ², but none that I have encountered are as comprehensive as Suler’s.

Psychologists have found many things to interest them in virtual worlds, but two topics dominate: identity construction and addiction. I’ll take these in the order in which they were first written; perhaps surprisingly for many people, this means I’ll be starting with addiction.

In 1989, the year Jim Aspnes wrote TinyMUD, Margaret Shotton of Nottingham University in England published a book ³ about computer addiction. It reported a formal study she had undertaken in the mid-1980s, comprising data from a number of questionnaires and psychometric tests, coupled with interviews. Shotton explicitly included players of MUD1 in her survey, and referred to that world during the discussion of the results; indeed, one of the people with whom she conducted extensive interviews was a MUD1 player.

Shotton confirmed some people were indeed computer-dependent, and they could be divided into three types:

• Networkers used the computer as a means for communication.

• Workers used the computer for programming and other work activities.

• Explorers used the computer to program in an investigative manner for self-education.

Networkers was the category into which the majority of MUD1 players fitted. Shotton discovered that they were less likely to suffer (from their dependency) in their real-life relationships than workers or explorers; that said, neither of these other categories were particularly bad either. There was no evidence to suggest that people who spent a lot of free time on their computers pained their family any more (or any less) than people who spent a lot of time playing golf or star-gazing through telescopes. Computer addiction is like an addiction to stamp collecting or photography, rather than like drug or gambling addiction.

Shotton actually goes further than this. Not only did she demonstrate that the intense use of computers doesn’t turn outgoing extroverts into pale, reclusive hermits, but she also suggested that for some people the experience was very positive. Given that the print media at the time were publishing scare stories about “computer junkies” ⁴, this was great news for the developers of computer games ⁵.

Most of what Shotton wrote about computer dependency continues to apply, even though computers and the Internet have now become mainstream. This has not, however, prevented people from talking about “Internet addiction” as if it were a worse social evil than gun warfare. It is only to be expected that when an individual indulges for long periods in an activity that their close friends and family don’t understand, it will be cause for concern; that doesn’t mean that the people who are concerned are necessarily right, though. Concern is best addressed (one way or the other) through information: Either an activity is basically harmless, or it’s dangerous enough to warrant intervention.

Sadly, some psychologists and psychiatrists side with the friends and relatives without even listening to the other side of the argument. Kimberley Young, founder of the Center for On-Line Addiction ⁶, conducted a three-year study ⁷ of Internet addiction and pronounced it generally a bad thing. People sit around in chat rooms when they should be saying goodnight to their children! Well yes, that is a bad thing, but then people sit around in bars or watch TV when they should be saying goodnight to their children; the Internet provides the opportunity to become obsessive about more things, but it doesn’t actually make people obsessive—they’re like that already.

Although I don’t doubt there are some things on the Internet that can cause real social problems (pornography, for example), lumping together everything Internet-based as if it were the same thing is, in my opinion, a mistake. I’m particularly unimpressed by Young’s profile of an “on-lineaholic” called Steve who is addicted to LambdaMOO (she follows this up with helpful advice for a recovery strategy he could use).

LambdaMOO? Excuse me? LambdaMOO is a place—how can you be addicted to a place? You can be addicted to what you do in a place, which in Steve’s case seems to concern increasing his low self-esteem, but you can’t be addicted to the place itself ⁸.

Besides, who is to say that the addiction isn’t a positive thing? Virtual worlds are places where social interaction occurs, which is more than can be said for many of the alternatives (for example, watching TV). People can be addicted to things for good reasons, not just bad ones; personal development is, I believe, one of the former. In this kind of situation, the dependency will disappear on its own after the individual has achieved their goal (assuming they’re allowed to); any “addiction” here is psychological, not physical.

David Greenfield takes a more open view of Internet addiction ⁹, recognizing that in many cases all that people need to do is cut back, rather than give up; in this sense, the “addiction” is really “intense interest.” His attitude is that the Internet can offer unique and stimulating experiences, and only becomes a problem when it truly interferes with daily life ¹⁰. This is good sense: Just because you have a virtual life, you can’t ignore the maintenance of your real life that is required to support it.

Dinty Moore summarizes ¹¹ the different points of view about why people can’t leave virtual worlds alone as follows:

• The players are addicted. Computer games are addictive, and gamelike virtual worlds are even more so because they blur into reality and players don’t realize they’re playing a game ¹².

• The players are learning valuable lessons. They get to meet interesting people, and can socialize better than in the real world. Some learn to code, too.

• The players are college students who will do just about anything to avoid doing academic work.

The third of these has probably more truth in it than the other two put together…. So, what does all this mean for the designers of virtual worlds? Well, it means that although virtual worlds may be highly conducive to activities that cause addiction, playing them is not itself inherently addictive. It means that when people do become addicted, whether it’s a good or a bad thing depends on what exactly they’re addicted to. It means that people who satisfy every field test for addiction may actually be pursuing an important personal goal with great determination. It means that you could be doing people a favor if you allow them to complete their goal.

Oh, and it also means you shouldn’t go out of your way to make the activities in your virtual world addictive; that way lies lawsuits. Popular is one thing, but real addiction is something else.

The possibility of deliberately addicting players to activities in virtual worlds has been investigated in a series of articles ¹³ by experimental psychologist John Hopson. His work confirms that goal-oriented virtual worlds already have powerful addictive features of conditioned reinforcement, which explains why some players will run on a leveling treadmill ¹⁴ for months on end and still want more. A morally unscrupulous or unthinkingly irresponsible virtual world designer could tune what has evolved anyway to intensify the experience even more; if they did, however, they’d be advised to seek premium legal counsel about what to put on their product’s box about it.

Incidentally, Hopson’s work shows it’s possible to look at the reasons people play virtual worlds purely in terms of this casino-style addiction; in such situations, there’s no requirement that players are undergoing the kind of potentially life-changing personal journey I have described in this book.

Which segues neatly to the topic of identity construction….

Addiction and identity issues are actually related; for many people, the chance to escape from the physical body is greatly attractive, and therefore the search for identity could be the addictive quality that keeps many people playing in virtual worlds. This was a point suggested by the first paper ¹⁵ to recognize the importance of identity experiment in virtual worlds, written in 1992 by Amy Bruckman.

Bruckman began her research as an investigation into what the cyberspace envisaged by Cyberpunk writers might be like when it arrived; when she discovered virtual worlds, she realized that it already had arrived, and switched the focus of her inquiry. Over the course of a month, she studied a number of virtual worlds, but concluded that she’d have to concentrate on a single one if she was ever to understand the phenomenon fully. She selected TrekMUSE.

Following Pavel Curtis, she noted that male-as-female cross-gender play occurred fairly frequently, commonly as a psychological exploration, but occasionally as a practical joke; as usual with practical jokes, they’re only amusing for the perpetrator, not the victim ¹⁶. She did some investigation into female-as-male play, too, but discovered evidence only of the “psychological exploration” style of cross-gender play ¹⁷. She also found that although men treated female-presenting characters with excessive courtesy, they often did the same to male-presenting newbies, too.

In examining addiction, Bruckman came to the conclusion that although playing virtual worlds was habit-forming, most players were able to stop if they wanted to and few long-term players were truly “stuck” in them. This makes any decision as to the goodness or badness of spending lots of time in virtual worlds essentially a value judgment (that is, the point unhappily missed by many of the “Internet addiction” brigade).

Although Bruckman linked addiction and identity issues, they were soon to become separated. This is perhaps due in part to the tremendous influence of Donna Haraway’s A Cyborg Manifesto ¹⁸. This is a utopian thought-experiment that envisages human beings as cyborgs in order to make a number of stinging points about the direction that feminism should go. It starts by stating three areas in which age-old boundaries have become blurred:

• Human and animal

• Organism and machine

• Physical and non-physical

Haraway argues that the lowering of these boundaries admits the possibility of the future existence of cyborg life-forms, then tokenizes such an entity so as to explore the consequences of its existence on feminist thinking.

It’s due to the second and (because it’s related) third of Haraway’s blurred boundaries that researchers became interested in virtual worlds. Traditionally, psychologists have separated mind, fantasy, and information from body, reality, and materiality, but virtual worlds don’t fit into this scheme at all well. By examining virtual worlds, perhaps some new ways of looking at issues of identity could be raised?

Lynn Cherny, taking Haraway’s point that cyberspace makes the position of “the body” problematical, looked at ¹⁹ different ways by which the player is split between being a physical entity at a keyboard and a virtual entity encoded in a computer.

As Cherny is a sociolinguist, much of her paper concerns describing some idiosyncratic (from the point of view of other virtual worlds) communication constructs used by players of her pseudonymous ElseMOO. She noted that although most of these activities were for fun, sometimes they were used in ways that are most definitely not fun: She was distressingly spoofed ²⁰ at one point during her investigation, and recommended that for ethical reasons players should not be able to manipulate other players as if they were puppets.

As I myself have argued in this book, I concur that there is a split between the physical and the virtual self. Furthermore, I believe that it is a purpose—no, a duty—of virtual worlds to facilitate the reconciliation of these selves, for the benefit of the real-life ²¹ individual.

As for the manipulation of other characters, well that’s an expectation issue. If you know it’s going to happen, and the degree to which it’s going to happen, there’s not a problem. If there were, then stage hypnotists couldn’t manipulate willing subjects for a living. What Cherny objected to when she was spoofed was basically that she did not have an expectation that this would happen; even if she knew it was possible, it was still a betrayal of trust. When people go into virtual worlds that look and behave in ways consistent with a particular kind of environment (real or imaginary) such that they can expect to have a reasonable understanding of what it will be like, then if something happens within that understood context it’s (morally, if not emotionally) fine. For example, a player in a fantasy world who goes up to a snake-eyed necromancer NPC famed for his mesmeric powers can hardly complain if their character is made to run amok while clucking like a chicken; they may not like it, but they can’t say it was unfair. If, however, the same thing happened in the same virtual world while the player’s character was buying provisions at a mundane downtown store, there’d better be a very good explanation (or an easy defense against the assault), otherwise it would be unfair.

Unfairness occurs when it’s your understanding that nothing awful is, in the current circumstances, permitted to happen by the virtual world’s rules (physics or terms of service), then it happens anyway. This is whether the consequences of the unfairness are (in virtual world terms) tangible or intangible. Most problems with unfairness come when groups of people (especially newbies) have a false view of cultural norms and do one of the following:

• React to what they perceive as unfairness, but most people think is fine.

• Perpetrate something most people think is an unfairness that they don’t perceive as such.

Allowing players to spoof other players’ actions is an absolutely classic and obvious way to cause this kind of trouble. In MUD1, I didn’t even consider giving spoofing capabilities to anyone but the virtual world’s administrators; it beggars belief that regular players were ever allowed to use it in MOOs except perhaps in the most rigorous of role-playing situations ²².

Charles Stivale analyzed ²³ the different opportunities virtual worlds present for players to harass one another to various degrees; he found identity assault to be the strongest form of harassment. The most interesting part of his paper is his discussion of how players of LambdaMOO reacted following an incident of “virtual rape” (the specifics of which I discuss in the next section ²⁴). They spent a great deal of time debating ways and means to define what forms of action could be construed as serious harassment, and although the discussion eventually floundered on arguments about what constitutes free speech in a world where everything is typed (and therefore covered by the First Amendment), it’s interesting to note that one obvious solution—remove the ability to spoof—was never on the agenda. In adventure-oriented virtual worlds, these commands are usually far too immersion-busting even to contemplate for everyday use, yet in early 1990s LambdaMOO they were apparently regarded as indispensable despite their obvious potential for misuse.

The major work on identity discovery in virtual worlds is Sherry Turkle’s superb Life on the Screen ²⁵. This is a wonderful book by a very influential author. In it, she makes a number of central points in a clear, easy-to-follow style ²⁶, charting her own personal experiences, but including such a rich seam of anecdotal excerpts from interviews with subjects that other researchers will be quoting her for years. What I particularly like about this book is that I agree with pretty well all her conclusions, despite having come from inside virtual worlds, whereas she comes from outside.

As I already mentioned in my Chapter 3, her Chapter 7 is the big one insofar as virtual world theorists are concerned; it’s there she gets to the heart of the identity issue. Is identity unitary or multiple? Is it a structure of mind or an illusion? She observes that on the Internet people “build a self by cycling through many selves,” which previously in real life this was very hard to accomplish. Virtual worlds, because they provide reality-presenting environments, are where this happens most intensely ²⁷; she characterizes them as “laboratories for the construction of identity,” and looks at them in considerable detail.

Turkle poses a series of important questions, most of which are yet to be answered (although I make an attempt to address some of them in this book). Those of most interest to virtual world designers are:

• What kind of virtual selves do people make?

• Do real-world selves learn lessons from virtual selves?

• How do the different selves communicate with each other?

• Why do people do this?

Using a number of case studies to support her theory, Turkle argues that virtual worlds allow individuals to express unexplored facets of themselves, or facets that they are unable to manifest in real life. She concludes experience in virtual worlds can greatly encourage personal growth and self-awareness, but cautions they can also be places where people lose their way and become trapped. The mechanism for this is identity drift (she calls the changes slippages), where character and player merge, where multiple selves join to form what the individual believes is their true self.

As a clinical psychologist, Turkle is keen to identify explicit ways by which people can work through issues in virtual worlds in therapeutic ²⁸ fashion—for example, by role-playing their real-life situation (or an analogy of it) in different ways. However, she notes that some people can repeatedly replay the same mistakes, learning nothing from them; this only serves to harden their attitudes and they hate themselves all the more for it, like ghosts doomed forever to echo the reality of their former-life’s deeds. This is reenacting, not reworking; it’s much less common than reworking, but it’s a serious problem for those who fall into it. Virtual worlds allow people to find who they want to be, but it’s up to the individuals themselves to become those people. Maybe for some, the distance between existing self and ideal self is just too great to travel.

On the other hand, it may be the design of the virtual world itself gets in the way. As John Hopson noted, there are ways to keep people playing in virtual worlds that have nothing to do with personal development. This is a point also made by Jonathan Baron, who describes ²⁹ a dialectic between game worlds that emphasize development over achievement and those that emphasize achievement over development. His assessment is basically that the former are the goal, and that people who aim for the latter are asking for trouble in the long term. For hero’s journey reasons, I totally agree with him on that point.

Although, as Sherry Turkle describes, some people do deliberately create characters with different personalities to see how they work, I contend that most people don’t. When they experiment with an archetype (for example, a new character class ³⁰), it’s usually because they’re bored with their first one, curious about the new one, or think they’ll have some in-world advantage as a result; they don’t make a conscious decision about which direction their personality should go in next (or if they do, it’s because they’re psychology majors or have experienced psychotherapy). To be fair, Turkle doesn’t say that everyone thinks explicitly about what identity they want either, but it could be easy for people unfamiliar with virtual worlds to come away from her book believing that identity drift is something that happens to only a few players; in my experience, it happens to pretty well all of them.

I suppose this is as good a place as any to describe my own views on identity. Personally, I view an individual as having a core self (or agglomeration of selves, it doesn’t really matter), which projects itself through the filters of action and word to present images to other individuals. The filter can be changed according to circumstances, but in the real world it’s difficult to do this because of the anchors of family, community, and society. People can find themselves projecting through filters they don’t like, or with which they don’t feel comfortable. Virtual worlds allow the filters to be changed experimentally and (as a consequence) experientially. Through playing them you can find the filter closest to clarity: The one for the person you feel yourself truly to be.

Identity exploration is generally seen as a good thing. Sherry Turkle noted that it can be problematical for individuals who discover their ideal self but resist moving toward it; Mark Slouka warns ³¹ that there are further difficulties when an individual wants to change but their identity issues are irreconcilable. He describes a male friend who, while masquerading as female, fell in love with another female character; she reciprocated his feelings without knowing he was male in real life ³². In this example, the virtual person and the real person can be reconciled intellectually and emotionally, but never physically; the female character’s personality can influence the male player’s (and vice versa), but they can never become one in the physical world. They can in the virtual one, therefore the male player is committed to spending as much time as possible there, but this is a relationship that can never be consummated. The fate of Slouka’s friend reminds me of that of Rilian, Prince of Narnia, enchanted so that he was only ever himself for one hour every night while bound in a silver chair ³³.

Slouka is right, too: There can be problems when the virtual world acts as a bridge between different parts of the real world. In general, players treat virtual worlds as if they were separate from the real world; this is an incredibly useful approximation to the truth, but (as I keep pointing out) virtual worlds exist courtesy of reality. Newton’s laws of motion work for everyday situations, but they break down when objects are traveling at close to the speed of light ³⁴; similarly, rules of virtual behavior work for everyday situations, but they break down when they travel close to reality ³⁵.

Every occurrence in a virtual world has a real-world effect: If nothing else, it is presented to the minds of those who observe its occurrence or who observe its effects. In most cases, this effect can be regarded as negligible compared even with the mere noise of existence surrounding the player. However, over time the accretion of these mental effects can be life-changing. People will develop as people, of course, but even in so doing there can be very unpleasant real-life physical effects. It’s quite conceivable, for example, that for the wrong person at the wrong time the permanent death of a character could cause the permanent death of the player ³⁶. Less drastically, falling in love can and does occur regularly in virtual worlds.

Virtual worlds may offer respite from reality, but reality is always, always in charge. You always have to come back. Of course, this could be used as a reason for not leaving it in the first place. Fortunately, my reality at the moment is in describing the psychological research I have discovered that addresses issues of identity exploration, so I’ll spare you my further speculations and return to that.

As with any science where the collection of data is expensive and time-consuming, psychologists have a tendency to talk about other people’s experiments rather than conduct their own. This means that eloquent or forceful essayists can pile interpretation on interpretation until they reach a collective explanation of an idealized state that may not equate with the truth. In looking at identity issues, MOOs, and other TinyMUD derivatives are the standard; much of what psychologists know about virtual worlds comes from MOOs (LambdaMOO in particular), even though they are hardly representative of virtual worlds in general.

Rather than relying on second-hand observations and idealistic interpretations, Elaine Raybourn conducted her own study ³⁷ of MOOs in 1998. She found that several of the tropes that had become current were actually quite wrong.

Raybourn started from the standpoint that players construct identities in virtual worlds. Her aim was to discover how they did it. Through interviews and participant observation, she discovered three themes that reflected the processes at work:

• Developing the programming skills that lead to privilege and status and to being a contributor to the community.

• Using creative communication skills to become more popular.

• Treating newbies to displays of active engagement, indifference, or abuse, so as to present an image of powerfulness.

These are not the kind of touchy-feely counterculture ideals normally described as being embedded in MOOs.

There are two main assumption failures:

• Although social virtual worlds do not have an explicit gain-points-for-doing-things hierarchy in place, that doesn’t mean there is no goal-oriented behavior. It’s not tangible (that is, programmed in), sure, but it exists. Players have, through their interactions, created a goal-oriented pecking-order system themselves.

• Community building is no more democratic than in real life. The earlier consensus among researchers was that more egalitarian communities formed in virtual reality because, freed from their bodies, people could project aspects of themselves which were received more positively, thereby creating stronger and more meaningful relationships. Raybourn discovered that community formation is actually heavily influenced by people’s efforts to gain respect, privileges, and popularity.

However much that players of adventure virtual worlds may welcome any research demonstrating that players of social virtual worlds are not their ideological superiors, the argument about goal-oriented behavior doesn’t actually say much. The “games” that people play in MOOs are the same ones “played” in any other social environment that has a power elite, such as an office, school, or factory (or, indeed, a too-easy adventure world that ceases character development before personal development has taken place; Michael Lawrie’s discussion ³⁸ of how he made MIST work as a game is very enlightening in this regard). It’s politics, and game-type virtual worlds have politics, too.

So people in a MOO may use political success to validate their identity constructions, but it’s the identity constructions that are at the heart of it because they’re the only material (real-world) change that can emerge from playing them (educational MOOs are an honorable exception).

One of the points that Raybourn makes most forcefully is that in virtual worlds “communication is identity”—with disembodiment, it’s all there is by which to judge anyone. This may be basically true in MOOs, but in adventure-style virtual worlds there are other ways to get a message across, too. All actions have a communication component: If I go off and fetch you something you need, I’m communicating an aspect of my personality even if I don’t say a word. Similarly, all communications have an action component: If I want to get through a locked door, then threatening you to give me the key could be considered an action undertaken by me to cause the door to open ³⁹. Actions speak louder than words, the pen is mightier than the sword; they’re different ends of a continuum. Raybourn is right, communication is identity, but communication isn’t only speech.

Raybourn’s paper has its flaws, but it goes some way to debunking the more idealistic views of virtual worlds that were becoming orthodoxy. Perhaps had she invested more than the 50 hours she did as a participant observer, she could have provided some insight into the way her three themes relate to one another among a community of players.

As for disembodiment, I offer the following summary: Actions can be physical or communicative, their effects real or virtual, to various degrees ⁴⁰. The same applies to people.

Although MOOs have a special place in the hearts of social science researchers, some do look at other codebases. Mizuko Ito’s object of study ⁴¹ was the (now defunct) LPMUD Farside. While, in general, toeing the Haraway disembodiment line and therefore saying pretty much the same as everyone else about virtual worlds’ capacity to blur the boundaries of mind and machine (reporting that it happens, rather than how or why it happens or what the consequences are for individuals), Ito also touches on something most other researchers missed by virtue of concentrating on MOOs and MUSEs: PKing (player-killing). Here, she does look at consequences and (because of the contested nature of these) accountability. She identifies several different reasons that players PK:

• Vigilantism, defending the innocent from PKs ⁴².

• In response to extreme provocation.

• They consider it to be part of the game.

• They are thrill-seekers, who enjoy the hunt.

• They are sociopaths.

Ito operates under the not unreasonable implicit assumption that everyone in a virtual world knows what the consensus attitude to PKing is (which, in general, is that the first two reasons are acceptable, as is the third between consenting adults). She notes that there is a qualitative difference between killing PCs and killing NPCs and mobiles, in that people are real.

It’s interesting to consider whether this will change as NPCs get increasingly better artificial intelligence; Haraway’s analogy cuts both ways.

The best article ⁴³ I have found about the reasons people PK was written by Lexley Vaughan, a MUD2 player rather than a psychologist (and a case study in identity exploration herself). In addition to Ito’s list, she adds

• Wannabes. People who think PKs are cool, and hang around in gangs talking up themselves into occasional acts of violence.

• Achievers. The classic player type, testing the viability of PKing as a means of scoring points.

• Explorers. Another classic player type, exploring this aspect of the virtual world just like they would any other.

• Broken achievers. Achievers who have been attacked once too often and snapped. They PK as a form of protest.

• Broken explorers. Like broken achievers, but deadlier. They don’t even care if they lose.

• Provers. These form the main body of PKs, characterized by their identity issues. They are people with low self-esteem, who choose to beat up other players (usually ones completely unable to defend themselves) to prove their self-worth. Unfortunately, more often than not this only reinforces their (and everyone else’s) belief that they’re losers.

• Big whackers. Alpha males strutting their stuff.

Lexley also identifies different types of PK run by administration-level players; she doesn’t, however, address the issues of marauding bandit gangs that plague large-scale virtual worlds where PKing is allowed.

Another player who has eloquently expressed how the identity issues of some people can spoil the fun of others is Arios Truthseeker. His 2000 Essay on D00dism ⁴⁴ describes the phenomenon of the d00d, a class of individuals prevalent in game-oriented virtual worlds (especially graphical ones) characterized by their stylistic spelling, their uniformity of self-perceived coolness ⁴⁵, and their complete disregard for the sensibilities of everyone else. They exploit virtual worlds as places to gain the acceptance and gratification that they lack in the real world, little realizing that their almost total objectification of other players means they are achieving nothing of the sort except perhaps among fellow d00ds. Truthseeker ⁴⁶ offers a number of strategies for dealing with d00ds as a player and, more importantly from the point of view of this book, as a designer.

The only major solution Truthseeker doesn’t appear to consider is that of letting d00ds complete their hero’s journey. Over the years, I’ve encountered several players who, had they been around today, would have qualified as d00ds. Some of them turned into the best wizards (even arch-wizards) I ever had. The d00d of today could be the responsible leader of tomorrow, if someone hasn’t tracked them down in real life and used physical threats to stop their playing ever again.

This almost brings me to the end of my discussion of the psychology of identity with regards to virtual worlds. There’s a lot more out there on the subject, especially in the cyberculture literature, some of which I’ll be coming to shortly from a different angle.

Firstly, as an aside, I confess that I was both surprised and elated when I discovered that psychologists had so quickly latched onto the fact that virtual worlds are catalysts for identity exploration. My only disappointment was that they didn’t realize that this was in part because virtual worlds had been designed as such. The best functional designs are supposed to be those that work while remaining invisible to the user, so I guess virtual world designers should take this as a compliment. On the other hand, it reinforces the common prejudice that we’re unrounded, geeky, emotionally adolescent individuals who couldn’t tell a Renaissance from a Reformation if one came up and bit us. Oh well, such is our lot.

Now the remaining points about identity construction.

For some people, an important component of their identity is their race. This is particularly true of individuals who find themselves in an environment where their own race is not dominant. Although for some people race is a physical notation of (their or others’) difference, this is not always the case: Other people may regard it as a cultural marker (genetic difference correlating with, rather than defining, cultural difference); others ignore race entirely. This latter approach (“color blindness”) is the liberal point of view, but it diminishes people for whom their race is a part of who they are.

This is a point made by Lisa Nakamura, who examined ⁴⁷ issues of race in the context of (you guessed it) LambdaMOO. Nakamura asserts that by not giving players the option of defining their characters’ race, LambdaMOO is in fact defining it for them: All players are assumed to be white ⁴⁸. Unfortunately for her, even if they were able to define their characters’ race, there’s still a strong chance that all players would be assumed to be white, in the same way that female characters in a male-heavy virtual world are assumed by default to be real-life male. LambdaMOO characters are not LambdaMOO players.

Nakamura discovered that race does make an appearance in LambdaMOO, however, through the recreational appropriation of stereotypes. She encountered many Asian-presenting characters in the form of replications of TV, movie, and book characters. She portrays the (presumed white) players behind these characters as identity tourists, staying awhile in another race before moving on somewhere else; naturally, as with all stereotyping, they were only ever going to reinforce their stereotypes this way, rather than actually learning something about the race they were pretending to be.

Nakamura characterizes this racial tourism as “passing,” but I don’t believe it is. The stereotypes chosen are almost universally obvious as being stereotypes: No one is going to believe that someone claiming to be a samurai warrior really is a samurai warrior, therefore the whole point of passing—getting away with pretending to be something you’re not—is undone. Were a non-Asian to pretend to be a Japanese student of English, that would be passing; pretending to be Sulu from Star Trek is no more passing than pretending to be Spock.

Many virtual worlds do have “races,” of course, it’s just that these just don’t correspond to real ones: You can be an elf, a dwarf, a halfling, and so on. In virtual worlds for virtual characters, these are as meaningful as any other race—more so in tangible terms, because whereas with real-world races it would be improper to give members of one race superior characteristics to those of another, for imaginary races it is acceptable (although see Chapter 8, “Coda: Ethical Considerations,” for a discussion of why perhaps it shouldn’t be).

These “races” are actually more like subspecies than races, of course, which blurs the issue even further. If I can be an elf, can I be a Jewish elf ⁴⁹ or a Chinese elf? Can I be an Inuit leprechaun, or are all leprechauns Irish?

To experiment with issues of race in virtual worlds, Beth Kolko has established a virtual world (MOOScape) specifically for this purpose ⁵⁰. Sensing that an important element of real-world identity is missing from virtual identity, her aim is to ask how the global marking of a character’s race affects the social environment of virtual worlds. It’s a reasonable thing to do; after all, if LambdaMOO lets you define your character’s age, why shouldn’t you define its race ⁵¹?

In setting up MOOScape, Kolko had to decide how to put in the racial specifier. She had basically three choices:

• Very intrusively, like gender. Reflect it in every third-person message concerning the player.

• Moderately intrusively, like mood. Reflect it in every reference to the player’s actions—for example, their movement.

• Moderately unintrusively. Reflect it only in the descriptions of players that people get when they look at them.

She eventually decided on the latter; you know what race other characters are, but it doesn’t intervene overtly in interactions between players.

A trickier issue was what races people were allowed. The problem with a closed data set (that is, a list of races from which people should select one) is that not everyone finds a description with which they feel comfortable. On the basis that MOOScape was introducing race into virtual worlds, rather than defining it, Kolko went with an open data set: Despite reservations that races like elves and lizardfolk would make an appearance, players are allowed to describe their race in whatever terms they wished.

The results of the MOOScape experiment should be interesting. Race only makes sense in the real world; by asking about race, MOOScape is asking players to reveal something about themselves, rather than about their characters. If your character is you, there’s not an issue here, of course: In situations where the virtual world is being used as an adjunct to reality—hosting a business meeting, for example—your character is merely an avatar and people will always relate to you rather than to it.

However, in most virtual worlds your character is not you ⁵².

In an absolute sense, race is a purely physical property. It does not transfer to virtual worlds except when referencing a physical entity (that is, the player—something to be avoided in virtual worlds if they are to maintain their immersive integrity). Put bluntly, if you say you want your character to be a certain (real-world) race in a virtual world, you are arguing only for a cosmetic change to your character’s appearance. That cosmetic change may be associated (by you or other players) with certain real-world cultural baggage, but the nature of this baggage is dependent on the real-world culture of the players that participate in or witness it.

For example, let’s suppose that race is a choice. People cross-gender play out of curiosity, so we’ll assume that some will cross-race play, too. Let’s further assume that some non-Asians play as Asians and truly (rather than in the Nakamura sense) try to pass. Would they learn about what it is like to be an Asian that way? No: They would only learn what it is like to be viewed as an Asian character by the players of that particular virtual world. They get insights into mind-sets, but not into genetics.

You can’t learn about the physical aspects of a race by role-playing a member of it; all you can learn about is the culture with which that race is associated (by its representatives, by other cultures, and by you). You might learn about any of these, but they only tell you about states of mind. In Reality, race (like gender) is a badge that people are forced to wear wherever they go. In virtual worlds, like it or not, that badge is removed. You can put it back on your virtual self, but so can anyone. In real life, people who wear the same badge may feel disposed to interact differently with other people depending on the badges they wear, and cultural differences will arise because of this. In virtual worlds, badge-wearing is arbitrary; race is therefore expressed only through the culture of individuals of that race (if that race has a race-specific culture).

Players of virtual worlds for whom their race is an important part of their identity have a dilemma here. To be true to themselves, they want their characters to exhibit the same indelible mark that they use to define themselves; however, in virtual worlds these marks are not indelible. Do these people therefore campaign to bind the cosmetic features of characters to those of their players (knowing it would be unpoliceable)? Or do they discard that part of their identity that is coupled to the nature of their physical body and align with the portable aspects of their culture instead?

Another important component of identity for people in the real world is their sexuality. Unlike the situation for race and gender, people in the real world do not have to wear irremovable badges that proclaim their sexuality (although they may choose to present indicators, overt or otherwise). This means that there’s not much call to implement sexuality as a physical characteristic in a virtual world ⁵³; sexuality-defined subcultures are not tied to physical markers (although in real life obviously they are tied to physical activity), and therefore they transfer reasonably easily to virtual worlds.

Most virtual worlds take no line on sexuality, which (apropos Nakamura’s argument about race) perhaps means they actually do take a line—that is, heterosexuality ⁵⁴ is the norm. Actually, for a fair number of them sexuality really isn’t an issue, because sex itself isn’t; the subject simply wouldn’t arise in a virtual world written for children, for example, and the same is true of many other (particularly gamelike) virtual worlds.

That said, even in gamelike virtual worlds, people do meet and fall in love. Both heterosexuals and homosexuals can operate in this kind of environment; it’s pro-heterosexual, but usually more tolerant than real life (you can play characters of the opposite gender, after all) and therefore not a major barrier for homosexuals who (like it or not) are obliged to operate under these conditions in reality anyway.

The real world has its homosexual subcultures, however, and it’s natural that people may want to bring theirs with them into virtual worlds. There are essentially three ways to do it:

• Set up a virtual world for which this is the dominant culture.

• Join a virtual world that has similar or related subcultures.

• Incorporate the culture in a ghetto within a regular virtual world.

Virtual worlds of the first kind do exist, although they are not in general advertised outside their target player base (probably so as to avoid attracting homophobes ⁵⁵). They are split by gender, because the subcultures for men and women are different. The two most interesting experiments are both aimed at women, Shoujo-Ai MUSH ⁵⁶ and Isle of Lesbos ⁵⁷.

General sex-oriented virtual worlds, which are usually friendly to people whatever their sexual orientations and preferences, have been around since Void. Despite what people may think, players do not spend their whole time in these worlds engaging in imaginative acts of virtual sex. They are as likely to sit around chatting about nothing as players in any other virtual world (although their very nature makes their cliques and communities quite tightly knit). HavenMUSH ⁵⁸ is probably the best-known example of this kind of “responsibly adult” world.

Randal Woodland describes ⁵⁹ the third of these approaches (all of which, incidentally, he calls queer spaces). In examining how four different online systems (the other three are BBS/forum-oriented) cater to homosexual identity, he describes a part of LambdaMOO called Weaveworld. This subworld is ⁶⁰ both accessible and distinct from the rest of the virtual world, using a series of very subtle symbols to indicate to those passers-by attuned to such things as the nature of its contents. The reasons for this tactfulness are so as not to appear in-your-face to other players of LambdaMOO (as both a courtesy to the general population and a defense against potential hate-players), and to impart a sense of sophistication (that is, to tell visitors that it’s not a place built for devotees of narcissistic sex).

In terms of the effect of these different queer spaces on identity, Woodland is reticent. He describes, for example, a case where the creation of a psychologically safe haven for gay men fell through because of freedom of speech objections. He doesn’t say whether in a virtual environment where no one knows who you are, the issues of psychological safety are different from those of real life. Can cultural tropes that have evolved in response to real-life hostility be shed? Are people who follow the same rituals in virtual worlds that they do in the real one passing up on liberation, or merely being pragmatic? What are the implications of a virtual world where every character (if not player) is the same sex?

To speculate further, for many people gender, race, and sexuality are part of their real-life identity. They are not of equal status, however, because of their visibility. Gender is visible in the real world (through physical characteristics) and in the virtual world (through pronouns in textual worlds and avatars in graphical ones). Sexuality is invisible in the real world and invisible in the virtual world; people who want to display it have to use signs, which may be understood by everyone (you walk around with a notice hanging round your neck saying “I am gay”) or only by those in sync with such matters (Weaveworld’s references to paganism, sensuality, ancient Greek warriors, and so on).

Race, however, is visible in the real world but (in the textual case at least) invisible in virtual worlds. Is race only ultimately important because it is visible in the real world (in which case, wouldn’t a virtual world undermine real-life identity?), or is it legitimately important for cultural or biologically imperative reasons too (in which case, will people for whom race is important have to develop their own codified signals for recognizing one another?).

There are many questions here to which it would be interesting to know the answer; unfortunately, the research on virtual worlds concerned with race and/or sexuality is relatively thin on the ground ⁶¹.

This is in stark contrast to our next topic: gender studies.