Coda: Ethical Considerations

I’ m writing this short chapter because, believe it or not, the designers of virtual worlds carry a responsibility for the consequences of their designs. Real people play these worlds, and the effects that a design has on them are real. Naturally, players themselves are in many ways accountable for their own actions: If you know that a virtual world has PD and you know that experiencing PD can be distressing, then when you play in that world and suffer PD and find it is distressing you can only blame yourself. Nevertheless, designers ought to be aware when some aspect of their design is likely to become an issue; its inclusion then becomes an artistic statement, rather than merely an element of gameplay.

This book is aimed at designers rather than players, so I’ll primarily be exploring the moral and ethical considerations that

arise concerning virtual world design ¹. All these are still open for debate: Where I express an opinion, that’s all it is, opinion; your opinion is worth just as much as mine in such matters.

I don’t claim to have any answers here, just questions. They’re questions that every designer of virtual worlds should ponder, though. With power comes responsibility; designers have power, therefore they must accept the responsibility that goes with it.

A word of warning; this section only concerns virtual worlds that do not have youngsters playing them. Virtual worlds that are written for (or that can be expected to attract large numbers of) children have a whole extra layer of ethical issues to their design, about which I do not feel qualified to comment.

Censorship

The first topic I shall look at is that of censorship: The deliberate exclusion of material that would warrant inclusion on purely informative, artistic, or gameplay grounds. The libertarian viewpoint is that all censorship is bad; everyone has a right to free speech, whatever they want to say and to whomever they want to say it. The common-sense interpretation of this is that there are certain conditions where what is said can be profoundly damaging (thus the libel laws) or dangerous (shouting “fire!” in a crowded theater ¹), and in these situations

free speech takes second place to some greater right. Otherwise, people can say what they want ².

There are still exceptions, though. Elizabeth Reid argues in a paper ³ about the ethics she employed when undertaking her research into virtual worlds that she should perhaps have self-censored some of the details that were given to her, even though they were provided willingly by people in the knowledge and expectation that they would be published. Her reasons for this are that her discussion of JennyMUSH provoked such interest among social scientists that they descended on it en masse to study the community; needless to say, this put it under such pressure that it almost fell apart. Legally—and ethically, as far as the players were concerned—Reid had every right to publish the details she did. Nevertheless, in retrospect she wished she had MADE every quotation anonymous (even if that meant being unable to use some of them) because then people wouldn’t have been able to find JennyMUSH so easily and their sheer number wouldn’t therefore have impacted it.

Currently, censorship in virtual worlds is performed almost entirely by the designers. The marketing group may insist on some changes, and licensed worlds have to remain true to the franchise, but beyond that there is only the law of the land (whatever “land” that may be). In the U.S., virtual worlds have been able to pitch themselves as examples of free speech in action, so they have been relatively free of legal interference. It may be just a matter of time before a major test case comes to

court over violations of anti-hatred, libel/slander, or property laws, but for the moment this has not occurred. This is in part due to two factors:

• It’s fairly clear that virtual world operators are no more responsible for what players say within them than telephone companies are for what people say on phones. They relay communication; they don’t publish it ⁴.

• Virtual world designers self-censor their virtual worlds so that contentious or provocative content does not appear.

So, let’s look at the kind of thing that might, at a glance, seem legitimate material for virtual worlds but that never seems to make it into them.

Unpleasantness

Some things are found unpleasant by enough of the population that even though they accurately reflect reality it would cause dissent were they to be included. Slavery, for example, was a fact of human existence for millennia; even now, some countries allow bonded labor that is slavery in practice if not in law.

So, you’re designing a virtual world set in the swashbuckling Age of Piracy. Players take on the role of a buccaneer, sailing the Caribbean in search of treasure, glory, and infamy. Within a week of beginning your research, you (the designer) have come to the conclusion that slavery was such an important part of Caribbean trade and industry that were you to omit, it your economic model would be missing a third of its goods. Do you put it in, or leave it out?

If your virtual world was for educational purposes, you’d probably leave it in. If it was primarily for entertainment, though…? The game Europa 1400: The Guild ¹ lets women become Catholic archbishops, even though this somewhat undermines its historical accuracy ². Virtual world designers tend to do this kind of thing too.

Unpleasant history is one example where self-censorship comes in. Another is unpleasant present. In the 19 th century, it was the general consensus in the western world that men were stronger and more clever than women. Now, it’s the general consensus that men and women are of equal intelligence. What about their comparative physical strength? Studies have shown that past academic examinations and intelligence tests were invariably biased in favor of men. Er, strength? Women and men may emphasize different aspects of their intelligence, but neither one gender nor the other is fundamentally the more intelligent. So will you answer the question about strength? Women excel at linguistic tasks, whereas men are good at spatial, and so on.

Most virtual worlds give men and women equal physical strength, even though in reality women are palpably not as strong ³. Some women are beefy enough to haul a truck, sure, but not ones that look like the female avatars in EverQuest. Those virtual worlds that don’t give female characters the same strength as male characters will usually try to compensate by making female characters shine at something else instead. It’s all for show, though. Real-life, non-cosmetic differences in gender and race are so charged with potential for problems that

virtual world designers simply edit them out of their worlds rather than suffer the political problems that come with including them.

A third example of unpleasantness is that which is fine now but which may not be in the future. Movies in the 1950s were full of people smoking cigarettes, but today’s certainly aren’t. Virtual worlds can have very long lives; what should designers omit that might cause future offense?

It may seem foolish to speculate on these possibilities: how can you know that something in your virtual world that is currently acceptable will cease to be so in a few years’ time? In general, of course, you can’t, but in some areas you can, particularly terminology. For example, in this book I have referred to people having same-sex orientation as “gay” and “homosexual.” It is practically guaranteed that in 50 years’ time even the phrase “same-sex orientation” will not meet with approval. The situation is yet worse for describing different races: Whatever currently prescribed term I use to describe a race or ethnic grouping, the historical evidence is that in 50 years’ time I’m going to appear a racist ⁴. It’s very tempting therefore for virtual world designers to counter this by omitting anything that has a track record of being controversial. This, of course, has other consequences (that we shall come to later).

As a final remark on the subject of unpleasantness in virtual worlds, it should be noted that the definition of “unpleasantness” varies between real-world cultures. For example, computer games that concern the hunting of wild animals are big-sellers in the United States but aren’t even released in the United Kingdom—there’d be a public outcry if they were. Many examples of such “localization issues” exist;

designers must choose either to ignore them (and lose sales), to make their presence switchable (which is expensive), or not to put in any offending content in the first place (hey, there’s an idea).

) The Real as the Virtual (and Vice Versa

Virtual worlds have their own communities and their own laws. Some things that are depicted in virtual worlds are legal in the context of that world, but would not be legal when practiced in the real world. For example, a virtual world set in the time of the British Raj may implement the Hindu custom of suttee, whereby widows threw themselves onto the funeral pyres of their recently deceased husbands. This would be allowed in the virtual world, but it’s been all but stamped out in the real world ¹.

Other examples abound. Virtual worlds routinely allow murder and the libeling of one character by another; some permit cruelty to animals, some promote the use of narcotics. It’s one thing to portray real-world crime as a virtual crime, as would be the case in a virtual world set in prohibition-era Chicago. It’s another thing, though, to make real-world crimes legal in their virtual-world implementation. Nevertheless, this is quite possible; in combat-oriented worlds, it’s even quite probable.

On the one hand, this distinctness from reality can be seen as a freedom from oppressive real-world laws: You may need to have your identity papers with you at all times in the real world, but you don’t in a virtual one. On the other hand, it could be seen as

the dangerous encouragement of antisocial behavior: Whom does it harm when an adult PC has virtual sex with a child NPC?

Although virtual worlds are generally more freedom-oriented than the real world, there are a few occasions where the situation is reversed. This is mainly because of restrictions that the real world places on the virtual world. For example, what two adults do in the privacy of their own bedroom is generally no one’s business but theirs, yet the graphical depiction of this behavior in a virtual world could well qualify as hard-core pornography. Virtual worlds that show things that would be illegal to broadcast to any television can rationally expect to be closed down. Nevertheless, people doing those things in real life are not breaking any laws ².

Another issue is the manner in which real-life objects are shown. Basically, you have more freedom to mess about with a name-branded object in the real world than you do in a virtual one. It used to be that if you wanted to show a brand in a computer game, you had to pay the brand-owner; nowadays, it’s more likely to be the other way around. These are mostly legal issues, of course, and therefore not of direct relevance here. Ethics become involved when things that can be recognized in real life make an appearance in a virtual world against the wishes of the people who own them.

For example, chase/combat game The Getaway incorporates a 3D model of central London built up from tens of thousands of photographs. People who own the properties shown have no right to object to this, under the same laws that apply to movie shoots in streets. A virtual world could use similar real-world data to produce an accurate visual model of some identifiable portion of the real world. Property owners can’t complain (well, they can, but the developers don’t have to listen). So what would happen if a player took over a virtual building that in the real world was a fast food franchise, then turned it into a brothel or a drug den? Could the owners object after that? To whom? What if shop-fronts displayed advertising logos for major corporations that players then defaced. Would that be right (because players are making a political statement in a virtual world) or wrong (because logos are free speech even in virtual worlds and defacing them is censorship)? Can designers deface these, as they might if they were creating a post-apocalyptic virtual world?

Supposing brands do appear with permission, players may still use them immorally. If you buy enough cups of virtual cola, you can lay them out in a pattern to look like any part of the human anatomy you like. You can write words with them. You can put them in demeaning contexts, such as public lavatories. Is it ethical to do so? It’s ethical enough in the real world, as long as you send the cans for recycling after you’ve made your point. In a virtual world, though, you may be paying a smaller subscription than you would be if the cola company hadn’t paid for product placement. Doesn’t accepting the reduction in price impose on you an obligation not to interfere with the message?

Virtual world designers handle these issues by self-censorship. Except in certain key areas where public expectation is on their side (for example, killing monsters), they avoid anything that could raise major ethical issues. It’s understandable why they would do this, of course, but it stifles debate. It’s left to the

designers of small-scale worlds to push the boundaries, but all too often they use off-the-shelf codebases that have already made the decisions for them.

If you’re a virtual world designer looking for an area in which to make your artistic mark, this could therefore be quite a fruitful area.

Passive Censorship

Censorship is generally active: You have a design that includes an element that you want, but you decide to take it out. If you did nothing, there would be no censorship. However, on occasion it can be passive: Information comes to light that is not in the public domain, and by doing nothing you are keeping it there. This amounts to censorship by inaction.

Here’s an example of what I mean.

Many virtual worlds have the concept of “guild ownership” of property. Characters pay money or give objects to a guild, and this can be used to obtain items collectively that individuals could not own themselves; the guild is like a company in real life—a legal entity that can be treated as a person in many respects. Real companies have directors, guilds have leaders. So what happens if the leaders decide to asset-strip the guild, then close it down?

Most virtual worlds don’t account for this kind of activity. On the face of it, why should they? If players sign up for a guild and are ripped off by the leadership, well, that’s just the way of things. Guilds are for groups of like-minded people; choose your friends more wisely next time.

If it were only that guild leaders gave guild property to their other characters, well, okay, a “dog-eat-dog” argument is

perhaps tenable. What happens when they sell the stuff on eBay, though? They could be a team working to a long-term plan, inveigling themselves into other players’ confidences to maneuver themselves into positions of power. Think this sounds improbable? It’s happened; it’s probably happening right now.

So, what does the developer do if they spot players ripping off one another? If they do nothing, this would be a form of censorship. So do they stop it? If it doesn’t break their terms of service, on what grounds could they do that? They would be breaking a trust. One solution would be to make a change to the virtual world design so that there was always an audit trail available and players can see where their stuff came from. That’s somewhat unrealistic, though; it’s not going to please characters who play as thieves in a PvP world, for example.

This is just one example of many. Censorship is the deliberate exclusion of material, but that material does not have to be tangible, and the decision not to act can be as censorious as the decision to act.

Players as People

“It’s just a game.”

When you play a game and you know the rules and something happens within those rules, you just have to accept it. You may not like it that the other team scored the touchdown, but you knew it was a possibility. All you can do is try harder.

This would be a fine analogy if virtual worlds were games. They’re not, though: They’re places. Some activities may be game-like, but not everything is. When players treat non-game

behavior like it was a game, this can have a nasty impact on other people in real life.

In normal computer games, you can do anything that the code allows you to. This is not an attitude that extends to virtual worlds, though. Most of the time it’s a good rule of thumb, but sometimes the virtual world lets you do things that you shouldn’t. As a trivial example, they aren’t coded to prevent your saying things that in real life would constitute verbal abuse.

Persona Issues

When people sign up to virtual worlds, they will have a number of preconceptions as to what they can expect to happen to their characters and what they can expect not to happen to them. This may come from publicity, from what friends have told them, or from a general sense of what “this kind” of product is like. If they know the virtual world is “about” combat, they can expect their character to kill things in it. If they don’t know initially, but in the character creation system discover that there is a wide range of combat options available, then they should read this as a warning that nevertheless there is combat in this world. If they don’t like the idea, they don’t have to play ¹.

If something happens that could not be reasonably expected, that’s of course a different matter. In many cases, it will be a surprise—possibly even a pleasant one—but not an issue. You didn’t know this Victorian Science Fiction game incorporated a race of troglodytes, but hurray, you do now. In other cases, it will be less pleasant but not emotionally damaging (for example, forcing players to make their characters use the lavatory). In some instances, though, it could be hurtful. The classic example of this is rape.

As I explained in Chapter 6, there are occasions when players are immersed to the extent that an assault on their character feels like an assault on them personally. If they knew this was a possibility before it happened, then no matter how bad it feels when it does happen they can blame only themselves. If I sit in the front row at the sea lion pool, I shouldn’t be surprised if I get wet; if I play a virtual world with PD, I shouldn’t be surprised if my persona gets killed. In either case, I should be surprised—and possibly deeply injured—if I were raped. It doesn’t fit the context; it’s not something I signed up for. Even if it did fit the context (plenty of wars in the past involved rape and pillage, so a brutally realistic historical world could perhaps justify including it), an explicit warning would be a good idea: So few worlds implement the concept of rape that it would come as a shock to find that one did even if it made perfect sense in the milieu.

Other real-world effects can also be disturbing. If you were to chop off some character’s leg in combat, a reasonable result might be that they couldn’t walk. Similarly, if you threw boiling water in their eyes, you might blind them. Some people are disabled in real life, and they could well play in your world. Is it right to show their disability in a negative light like this? What if their character suffers the same disability as the player? It’s tiresome being deaf in real life ²; why should anyone have to put up with it in the virtual world, too? It’s not like virtual deafness gives any great insight into the condition, so education is no grounds for inclusion.

What about illnesses and diseases? If your virtual world includes cigarettes, it might seem fun to give characters that over-indulge lung cancer. Hey, it’s responsible, too, encouraging people not to smoke. That’s not going to cut much ice with a player whose grandmother just died of lung cancer, though. Poisons also have to be handled thoughtfully: The sensibilities of people whose relatives went to Nazi death camps must be respected. I should point out that it’s by no means impossible to incorporate these themes into virtual worlds; the difficult part is doing so responsibly.

There are further examples of seemingly innocuous things in virtual worlds that can have real-world effects on people. In MUD2, there is (among other things) a piece of ham that characters can eat to restore lost stamina. Should a real-life vegetarian allow their character to eat the ham? Am I, having put the ham there, hurting the feelings of vegetarians? Should I remove it? Was I wrong ever to create it, knowing that it might offend vegetarians ³?

The problem with looking at the effects on players of virtual events is, of course, that people can be affronted by practically anything. Plenty of people in the real world are frightened of spiders, so should there be no spiders in virtual worlds so as not to alarm them? What about those people in the real world unlucky enough to be afraid of oranges? They do exist: Should their concerns not be given equal weight?

Personally, I take the view that the only morally consistent approach is for designers to be true to their own beliefs. Do what you want to do: If players don’t like it, they can stop playing. General content warnings, like film classifications, can

be used to forewarn people of what range of material they can expect. Beyond that, though, it’s up to players to decide whether they want to visit your world. Developers should not have to take out product liability insurance for content any more than publishers should have to for books. It’s for players to weigh the odds of participation, not for designers to second-guess them.

Morality: Another candidate for inclusion in a critical aesthetic?

Privacy

Anything you do in a virtual world can be logged or snooped. In commercial textual worlds, it probably will be; in graphical worlds, cover is rarely 100% at the moment but will become so eventually.

So, is this an invasion of privacy? No more than is a closed-circuit television system (CCTV) monitoring a shopping mall. Indeed, it’s probably less so, given that log files are only usually looked at when there’s been a complaint whereas CCTV cameras are watched the whole time (in theory, at least).

Well, that’s one point of view.

Here’s another. If I (as a player) issue a string of profanities at someone, impugning their real-life person, will they sue the developer or me? They’ll sue me, because the developer will argue ¹ that they are merely a common carrier with no control over what their players say. That being the case, they should need a wiretap warrant before they can record any of my communications. If they log me otherwise, they’re accepting responsibility for controlling what I say. They can’t claim I consented to be monitored when I signed their terms of service

agreement: That would mean I was consenting to a form of self-incrimination, which is counter to human rights legislation.

Before this drifts too far away from morality toward legality, here’s something else to think about. Suppose in a virtual world you stand around in a reasonably private space chatting to a friend. Both of you are adults, and both of you are okay with using strong language. The moment anyone else appears in the room, you tone down your vocabulary, but when they go you start inserting profanities again. Is this ethical behavior or not? Would it (still) be ethical if you knew that you were being logged, and that an administrator might therefore come across your gynecological language and be profoundly shocked? What if it wasn’t the presence of rude words that was offensive but your unorthodox political views?

Can you ever consider yourself to be in private in a virtual world, or must you always act as if someone is watching?

Addiction

Virtual world designers are creating something that they know is addictive. Addiction is a bad thing, because of its personal and social costs. At a moral level, are designers of virtual worlds little better than drug pushers?

The first person to question me on the ethics of making an intensely addictive experience available to unsuspecting members of the public was computer journalist Guy Kewney in the mid-1980s. His point was that even if I didn’t know MUD1 would be addictive when I started on it, I knew now and should therefore give some thought as to whether I should continue running it. This I did, and I concluded that although players might at times exhibit behaviors that could be considered symptomatic of addiction, it wasn’t that simple. Virtual worlds are places: You can’t be addicted to a virtual world any more

than you can be addicted to Las Vegas. You can be addicted to the things you do in a virtual world, as you can be addicted to the things you do in Las Vegas. Unlike Las Vegas, however, in MUD1 it wasn’t me who made the experience addictive, it was the players themselves through their imaginations and interactions.

No, it didn’t convince me then, either.

I certainly didn’t set out to make MUD1 addictive. I wanted it to be a place where people could be and become themselves. It just happened that for many people this turned out to be a very intense experience that they wanted to repeat again and again. The question I should have been asking was not whether playing in virtual worlds was addictive (if “addictive” is the right word, which in this context it almost certainly isn’t), but whether this addictiveness was a bad thing given the potential rewards. In this respect, I see it morally defensible on three counts:

• People come out of the experience as better individuals than they were when they went in.

• Most players would not develop as much if they did something else instead of playing in the virtual world.

• The process has a definite end. When your hero’s journey finishes, all signs of addiction/dependency/habit are gone.

Players of virtual worlds don’t all have periods of what could be called addiction to the experience, but even those who do will on the whole come out the better for it ¹. Virtual worlds are a force for good.

The problem is, not all virtual worlds satisfy the preceding criteria. Some, for example, use gambling-psychology feedback systems to drive their addictiveness, rather than the sheer freedom to be. The player is addicted but isn’t developing; they play to feed their addiction, not to end it. It’s all the worse when the virtual world is especially designed this way; accidental incompetence is a lesser crime than deliberate incompetence. As I mentioned in Chapters 5 and 6, most game-oriented virtual worlds have a certain inherent amount of addiction-promoting activity as a part of their gameplay; it helps their players have fun. Plenty of regular computer games also have it, however these don’t attract anywhere near the degree of dedication that virtual worlds are able to rely on. It’s the player’s development that is the critical “addiction;” that being the case, virtual worlds that deliberately implement in tandem with this a variable feedback system of rewards have something of an ethical case to answer.

The other common violation is in closure. If players are on some version of a hero’s journey and the virtual world designer, knowing this, interferes with their progress, they will have to play for longer to reach the end (if indeed they ever reach it at all). The virtual world itself may never “end,” but individual journeys must; the never-ending journey is an eternity of frustration. Give people an out and they’ll come back; force them to stay and they’ll go. Virtual worlds that move the winning post as players approach it are dealing out an unnecessary cruelty.

Mental Illness

This is more of a game management thing, but I thought I should mention it here because it does impact on designers.

Some players really are crazy. If you accept that the word “crazy” can be applied to an individual, then statistically some crazy people will play virtual worlds. A user base the size of EverQuest’s will include several bona fide psychopaths. That’s assuming that virtual worlds mirror the general population: It could be, for example, that the nature of these worlds is such that fewer crazy people play them. Alternatively, they could act as magnets. I suspect the latter

I’m aware that the word “crazy” is emotive; I used it here to make a point. Most of the people who suffer from various forms of mental illness are not crazy by any definition of the word. Some, however, are—dangerously so. What does your virtual world say to them? Are you responsible if they take what it says to heart?

The “crazy people” quandary is not unique to virtual worlds. Film, television, and computer games all suffer from the same problem. If you dramatize a killing spree, there is a possibility that it could inspire a seriously disturbed person to go on an actual killing spree. Where virtual worlds are unique is that they blur the boundaries between fantasy and reality until the two become one and the same. For someone who may have difficulty separating reality from fantasy at the best of times, this could prove too much. In such circumstances, can virtual world designers still hide behind the “you can’t legislate for crazy people” argument? If not, how could virtual worlds as an industry ever be viable?

Most mentally ill people who play virtual worlds are not, fortunately, crazy. They may have serious issues, and could even be a danger to themselves ¹, but they will listen to reason.

System administrators accept a duty of care for players while they are in their virtual world, and if it is apparent that someone is behaving worryingly oddly then it could be best for them to be barred entry. Of course, this could well make the individual’s problems worse, so it’s always wise to seek external advice first if possible. I’ve only ever had to do this twice (once with MUD1, once with MUD2), but for large-scale games it’s going to happen more often.

Designers of small-scale virtual worlds who are panicking at having just read this can probably calm down. I can’t comment on the legal issues, but from an ethical point of view you’re on solid ground as long as you act in good faith. If you see that someone needs help, and you give what you (or any reasonable person in your position would) believe to be help, then you can’t be expected to do more. The alternative is to do nothing, which may be the legally sensible position, but it’ll feel worse if something awful does happen.

There is ample anecdotal evidence that virtual worlds can help with certain kinds of psychological problems. Sometimes, they merely make life easier—an agoraphobic may not be cured of their fear by playing a virtual world, but at least they “get out” a little. It’s the same as with many physical disorders: You can play a virtual world with both legs in plaster. More interestingly, though, are situations where people use virtual worlds to work through personality problems and come out “cured.” I know of no formal medical studies of this phenomenon, but I’m convinced it happens. Whether designers can claim any responsibility for effecting such changes is another matter, of course; I suspect it’s more to do with the fundamental nature of virtual worlds, rather than anything value-added by designers. That said, it’s perfectly feasible for designers to remove this therapeutic value by interfering with the way players progress; it’s obvious, for example, that any self-

realization that comes from completing a hero’s journey is not going to happen if the designer never lets anyone finish.

Tempting though it is to feel all warm inside about how virtual worlds can help people with mental illness, there is another side to the coin. If we accept that virtual worlds can have tangible positive effects on the minds of players, then shouldn’t we also accept that they can have tangible negative effects, too? Could they make someone’s condition worse? Or could they give someone a problem who wouldn’t have had it otherwise?

The answer to the first question is yes; we saw this in the Chapter 6 discussion concerning psychology and virtual worlds. People can relive the same issues the same way time and time again, hardening their views rather than changing them. In most cases this is unfortunate, and it means they’re not going to get anything out of the virtual world; for the mentally ill, it could be beyond unfortunate and make their situation worse. If this happens, then it’s time for the player to seek professional help. Normally it would be the player’s friends who raised the alarm, but it could be the live team who has to deal with making the real-life phone calls to find out what the best response is.

On the whole, I believe that many more people benefit from playing virtual worlds than suffer, but then I’m a virtual world designer so it’s not like I’m going to say otherwise.

The second question—whether virtual worlds can cause mental problems—is very disquieting. It is almost certain that the intense experiences that virtual worlds are capable of delivering will push some people over the edge. They may develop anxiety attacks, delusions, fears of persecution—pretty well anything. The virtual world acts as a trigger. The crucial point is whether this is something that would probably have happened anyway, or whether it was induced by the virtual world?

The standard defense, used in many industries, is that some people are predisposed to this kind of thing and you can’t blame the author, composer, artist, director, or news editor whose work tipped the balance. Virtual world designers could argue this too. However, it’s not quite that simple. Virtual worlds have a strong psychological element to their design. Could it be that with their huge capacity to promote introspection they might cause some people to become neurotic or worse when no other everyday experience would? I’m not suggesting that designers would attempt such a thing deliberately, but that’s not to say it couldn’t happen by accident.

If it did, this would be cause for concern. There would be a danger to playing virtual worlds that was not associated with any other leisure-time activity.

It’s impossible to gauge the possible extent of the problem without a formal study. However, given that anecdotally it’s difficult to say whether it happens at all, the incidence is likely to be low. It’s also quite likely that obsessive behavior will be picked up by friends or the live team before things come to a head, and that any potential catastrophe can therefore be averted. Designers of small-scale worlds where everyone knows everyone else can therefore probably rest easy. Designers of large-scale worlds can also probably do so if the mechanisms are in place for detection and prevention.

Virtual worlds with mental healthcare schemes? It could happen.

Religion

As I mentioned in Chapter 6, I’m an atheist. All religion looks like an aspect of anthropology to me. I see where religions come from, I see why they succeed and fail, I see why people follow them, and I see how they propagate. I also see why my making

any observations about religions is likely to be highly contentious no matter how impartial I aim to be, therefore I shall keep this short.

The necessity of my having to make the preceding disclaimer in this book transfers directly to virtual worlds. If you want to know what the phrase “treading on eggshells” means, this is a great way to learn. In real life, it’s almost impossible to mention religion without offending someone; the same is true of religion in virtual worlds. Yet for many people, their religion is the absolute focus of their life. So what does the designer do?

There are basically four options:

• Discreetly ignore religion. Simply never mention anything about it.

• Make up a religion (or use a dead one). Here, the virtual world conveys the sense that religion or spirituality is important, but it doesn’t use anything that anyone actually believes.

• Include references to real-world religions, but only passively. You might have a cathedral, but it has no tangible effects in the virtual world beyond its physical properties.

• Embrace one or more real-world religions. This can be explicit (rituals and ceremonies that have tangible effects) or implicit (the virtual world conveys a system of values and beliefs without directly referencing the source). When implicit, it can be combined with one of the other options mentioned previously.

Discreetly ignoring religion is the safest approach, especially if it makes sense in the context. This is the way most movies do it: If a character in a screenplay is a priest, he’s a priest for a reason. Science Fiction virtual worlds work this way.

When the genre demands a religion, the usual solution is to make up one ¹. Fantasy religions in Fantasy worlds are generally accepted; those who don’t accept them usually don’t accept the basic Fantasy premise either. Interestingly, the “need” for religion in Fantasy worlds stems mainly from the assumption that cleric-class characters must obtain their healing-type spells as gifts of their deity. Although this is a common fiction, it’s not one that is a necessity; despite its religion-influenced creation myths there are no priestly types in Lord of the Rings, for example (and few wizards).

The third option is to put in real-world religious paraphernalia, but not imbue it with any special functionality. Players may choose to conduct weddings in temples or churches or wherever, but to the virtual world engine none of this has any significance. The designer treats a religious building like anything else—a castle, a cave, a boatyard—and religious objects as no different from other expensive or coveted items. This is the safest approach for an historical genre where religion plays a part without being the central theme ². For example, clergymen and churches have a place in Wild West mythology, but they are stylized to the extent that they can be used sensibly without offense. In later settings this also can work, although you’d probably need a reason to use it rather than simply ignore religion altogether ³.

Finally, designers will often work threads of their own philosophy into a virtual world, and this could include their religious beliefs. A deliberately allegorical work would be more difficult to construct, but it’s by no means impossible. Overtly religious virtual worlds that give tangible virtual world effects to implementations of real-world religious practices are making themselves targets for all kinds of troublemakers, whether from their own faith or some other.

Even seemingly innocuous things can be hurtful: Making Santa Claus appear every Christmas and give out presents is “only a bit of fun,” but to a non-Christian it could appear evangelical. Not having Christmas in the virtual world when it’s in full swing in the real world is also saying something, though. Should characters have a “religion” flag that turns on the effects of a religion? Or is it then unfair when Christian characters are given presents (that is, tangible advantages) by Santa Claus but no one else is?

The religious sensibilities of players can cause design issues elsewhere. Earlier, I mentioned that I put ham into MUD2 to challenge vegetarians to look at their beliefs, but it also challenges Jews and Moslems too. Even if it were some other foodstuff, that could still give Moslems a problem during Ramadan. If your virtual self eats virtual food in a virtual world during daylight hours, are you breaking your fast or not? Is it the principle or the actuality that matters? What if the virtual world had its own clock or calendar that was out of step with the real one? Would that make a difference?

Virtual worlds can depict or allow activities that are legal in the real world but that are disapproved of by established religions. Homosexuality, alcohol use, usury, and non-pacifism are all frowned on by one religion or another. Some religions look on the depiction of their symbols, buildings, or teachers as blasphemy, whereas for others it’s good publicity. Some think

that showing “evils” such as black magic or devilry is corrupting, but others think it’s a case of “know thine enemy.”

Of particular difficulty is when there is a schism within a religion. Virtual worlds are very good at giving male and female characters equal opportunities, but some creeds are divided over whether such equality transfers to them. Some versions of Christianity allow female priests, but others don’t. Not only does the designer have to choose which religion to depict, but which sect of that religion.

Another problem with using real-life religions is omission. A cyberpunk world might have references to fundamentalist Christian churches, high-tech Buddhist and Hindu temples, but have nothing about Islam at all. This definitely insults either Islam or the other religions mentioned, but it’s not obvious which!

Then there are the non-standard religions. Some years ago I wrote a series of children’s stories that I put on my web site. One of those stories concerned vampires ⁴. People would look up vampires in search engines and, finding 18 references in my story, would read it. They would then send me emails criticizing me for mocking their religion and publishing blasphemous untruths; about half would include dark threats about what they were going to do to me ⁵. I now have a link on the page explaining the context, which has cut the number of complaints to practically zero; these people may have unconventional views, but that doesn’t mean they’re stupid. The point is that they had beliefs every bit as strong as those held by adherents of conformist religions. A designer hoping to avoid confrontation

with zealots can reduce the possibility of its happening, but can’t eliminate it.

There are really two sets of morals involved here. The first is the system of morality that comes with a religion; there is a practical aspect to this, in that if your virtual world promotes a morality that runs counter to a particular religion’s views, you shouldn’t expect to get many players from that congregation. If you go really overboard, there could be real-world protests or worse.

The second set of morals involved is your own. I’ve assumed throughout this chapter that you will do what you think is “right,” of course, but religion raises the tricky point of how to treat views you think are “wrong” (but that other people think are “right”). I used to know someone who refused to believe Gödel’s Incompleteness Theorem; he accepted the proof, just not the theorem. Whatever your own system of morality, someone else’s could well seem as absurd as that. If a player asked you to remove references to Gödel’s Incompleteness Theorem because it was wrong, would you do it? What if they asked you to remove references to Easter? Or Ramadan? Or the Festival of Lights?

Real-life laws make some boundaries clear, but the others you must set yourself.

Groups of Players as Groups of People

If someone makes a statement with which you disagree, you might want to take issue with them. There are social norms that dictate how far you can go with your riposte—verbal abuse

warrants in the first instance only a verbal defense, for example. Although this can get out of hand, on the whole it works. If an individual says something, then other people can challenge them about it.

What if it’s not the individual who makes the statement but an organization of which the individual is a member? If a political party issues a general pronouncement, then that announcement is on behalf of the party’s members. If you disagreed, you could reasonably expect to bring it up with any members of that party you knew, to take them to task. Sure, they might not agree either, but what to do about it then becomes their problem.

It gets fuzzier for groups that people are members of non-voluntarily. If two countries are at war, should they round up each other’s nationals and intern them? It happened in the 1940s. The innocent suffer, but the argument is that far fewer innocents suffer overall than would if enemy agents were free to wage a campaign of sabotage behind the lines.

Terrorism is another area where this logic can be applied. If you know that all terrorists are members of a particular religious group, can you intern members of that group that haven’t committed any crime but that are acting like they might were they given the chance? Membership of a religion is voluntary—plenty of people convert to other religions—so why not? This happened in Northern Ireland in the 1970s.

I’m sure you can see the slippery slope here. By taking a series of seemingly short but justifiable steps, pretty soon you reach the stage where individuals are frightened to get on a bus because of the color of their skin or the way they dress. In virtual worlds, the effects of stereotyping are not quite so worrying, but nevertheless there are a number of areas where people accept the status quo without perhaps considering what that means.

Icons

Many virtual worlds draw on iconic or mythical images to sustain their immersive depth. If a player comes across a “little old man” NPC, it’s a source of great disappointment if it turns out he really is just a little old man and isn’t veiling some immense power or wisdom. If they came across a unicorn, they would feel cheated if it wasn’t white (and if they came across a herd of them—say what?!). Although these iconic figures are stereotypical, therein lies their strength; they represent an idealized “knowledge” of a reality that when correctly applied to a virtual world can deliver immersion. So tokenized have these metaphors become that it can forcefully be argued they aren’t stereotypes at all, they’re shorthand for a parcel of meanings.

The same argument can be used to explain why so many Fantasy virtual worlds have elves and dwarfs (or rebrandings of the same) as character races. Dwarfs are solid, materialist, quick to anger but dependable; elves are spiritual, aloof, beautiful, yet insular. Players immediately have identifiable stereotypes that they can use to help decide what kind of character they wish to play. Real-life dwarfs might complain about this depiction, but “no, these are dwarves not dwarfs.” Real-life elves rarely comment.

An ethical point that seems to have been missed by most of the designers who employ character races ¹ is that it promotes the notion that races are tangibly different. If it’s fine to regard all elves as being arrogant, fair-haired aesthetes, that validates the idea of stereotypical races. Why couldn’t a dwarf be a dreamer, a poet, a fun-loving bard? Why couldn’t an elf be a greedy, bad-

tempered, never-satisfied lover of ale? Because they’re “not like that”? Or maybe “a few are like that, but most aren’t’? So what are “all Japanese” or “all Jewish” or “all Mexican” people like?

Having races as stereotypes in virtual worlds endorses having races as stereotypes in the real world. It’s up to designers themselves to decide whether or not this is fine by them, but they should at least make that decision consciously. At the moment, most of them simply follow the flow without thinking about it. They should.

As I mentioned in Chapter 6, most virtual worlds are fine with giving tangible properties to Fantasy races, but they avoid doing this for real races. Players may decide at character-creation time or through their personal descriptive text to reflect a real-life race other than the default (there’s always a default), but it’s purely cosmetic. There is an argument that this homogenizes characters, making them all basically the same; this is indeed the case, but that’s unsurprising given that it’s the goal. Sure, ethnic groups that have a culture tied to their physical appearance won’t thrive in a world where anyone can steal that appearance, but then you have to ask the question as to whether this might not be a good thing. Racial supremacists have lost the war when people can be whatever race they choose to be.

There is a separate issue related to people for whom their race is a fundamental part of their identity. In virtual worlds without their race, or even with their race but with Fantasy ones alongside it, this could prompt a serious crisis. Should designers try to accommodate such individuals? Or is it simply a fact of life that race doesn’t transfer to virtual worlds (even though attitudes to race can)?

Other stereotypes exist in virtual worlds, of course, which are less emotive but nevertheless raise similar issues. Character

classes, for example, impose conformity. If you’re a designer who believes in conformity, have character classes; if you’re one who doesn’t, don’t. If you haven’t given the issue the slightest thought, you’d better do some thinking—you may be promoting social values with which you disagree.

Finally, a word on anthropomorphic animal characters.

As I mentioned in Chapter 6, anthropomorphic characters are all to do with symbols. Different animal species have different characteristics that (because animals aren’t as free-thinking as humans) they follow strongly. Yes, some can have personalities that are different from those of others, but they’re ruled by their instincts far more than are humans. Because of this, people have come to associate certain behaviors with certain animals—the cunning fox, the timid rabbit, the inquisitive mouse—and they use the animals as symbols for the behaviors. In some cases, the link is only metaphorical: A man who prowls around looking for innocent women to prey on might be likened to a wolf that prowls around looking for innocent smaller creatures to prey on, but there’s nothing sexual about the wolf’s activity. Nevertheless, the image of the wolf as a predator is strong enough that if you referred to a man as acting like a wolf people wouldn’t think you meant he attacked sheep.

For some designers, this degree of stereotyping could also be too much. If it’s okay to stereotype animals, pretty soon it’s okay to stereotype humans, and then where are we? For others, it’s merely an example of the general way that languages work using metaphor, and the moment foxes and rabbits complain is the moment their plight deserves to be taken seriously. You can make up your own mind.

Virtual worlds where players take on the role of anthropomorphic characters implicitly accept that this stereotyping is fine; furthermore, they have a well-developed

system of codes that associate different animal forms with different kinds of behavior. This leads to a system of self-fulfilling stereotypes: If people think that bears betoken having a macho exterior but a sensitive caring inside, then people who want that image will play as bears. People who want to play as a shy character will not choose to be a bear, even though in a virtual world where all the animals are anthropomorphic anyway there’s nothing to stop someone from being a shy bear.

What about non-anthropomorphic animal worlds? What if your character is, say, a dog? If it looks like a dog and can only do things that dogs can do, is that okay? Well if people want to be a dog, then presumably it’s up to them, but there are some awkward problems about identity drift here. It could be psychologically damaging for someone who didn’t know exactly what they were doing to play for extended periods in virtual worlds in the role of a dog or any other non-sentient creature. Playing a character of a gender, race or age not your own is different because they can all think just as well as each other; playing as a dog, though? You can change, but can the dog?

Although this may seem rather a far-fetched idea, there are already virtual worlds moving toward it ². Knowing what you know about immersion, would it be ethical or unethical to go the whole way?

Changing yourself is easier than changing a stereotype. You decide on you; everyone else decides on a stereotype.

Social Engineering

Are virtual worlds a force for good or for evil?

As I have already said, I strongly believe in the former. However, mere faith in my own (clearly biased) judgment is hardly justification. Computer games (along with television, film, and music) are routinely implicated when some disturbed teenager gets hold of a weapon and goes on a killing spree; virtual worlds are also starting to suffer from their perceived negative effects ¹. Do virtual worlds incite, reinforce, or train?

I won’t make the case for virtual worlds here, because I hope to have done that already in this book. However, I will examine some of the case against. Without any formal studies to call on, all I can do is provide anecdotal evidence; nevertheless, it’s evidence based on years of observation, and therefore not quite as speculative as it may seem.

My argument that virtual worlds are a force for good is based largely on the fact that players can learn to be better people as a result of playing them. That means I intrinsically accept that virtual worlds can bring about change. Given this, why must that change necessarily be for the good?

Well, it isn’t always, of course; in a well-designed virtual world it will be most of the time, but some people will indeed replay old issues repeatedly, consolidating points of view that would be better disbanded. The reverse is also true: Some individuals in real life become better people after committing horrific acts that leave others traumatized, but on the whole anyone taking up this kind of activity is on the road to personal destruction.

So, why is it that individuals take “good” things from virtual worlds, rather than “bad” things? I believe that it is due to the similarities between virtual worlds and the real one. The

majority of people who learn anything at all from virtual worlds will pick up things such as:

• Information specific to the virtual world (stats, goals, geography)

• Skills specific to the interface (dodging, aiming, feature recognition)

• Skills specific to the virtual world (tactics, strategy)

• More general skills (social, behavioral, educational)

• Wider cultural values (as exhibited by the other players)

• The difference between the real and the virtual

These are all either innocuous or of benefit to the individual player (the further down the list, the more likely it is to be of use ²).

People who play virtual worlds will not pick up things such as these:

• Desensitization (to evil, to other people’s emotions)

• The belief that people are just like NPCs, and rules that apply to NPCs apply to people

• The idea that everything virtual transfers to the real

If any of these were true, that would indeed be worrying. In general, though, I don’t think they are. Let’s take a closer look….

The desensitization argument is that people come across something in the virtual world so often that they become used

to it, so when the same issue arises in the real world they do not respond appropriately. This is a valid point about virtual worlds (as it is about things like TV news, too); where it becomes invalid, however, is in the suggestion that impressionable players will be unable to distinguish between virtual-world concepts and the real-world concepts that they’re based on.

Take “evil,” for example. The notion that most players of the big commercial virtual worlds have about evil actually translates as “cool.” In practice, the people on the “evil” side exhibit the same kind of behavior that they would were they on the “good” side, except they think they’re cooler because they have all the nasty-sounding spells and skills.

What evil really means is rule by fear. You do what your boss tells you to do because if you didn’t you would be tortured, maimed, and (if you were lucky) killed as an example to others who might think of stepping out of line. An evil person is as likely to kill someone on their own side as they are the “enemy.” In real life, people fight for evil because:

• They’re evil.

• They think they’re evil.

• They’re afraid of what will happen to them or to people they care for if they don’t fight.

• They have no way of escaping.

• They don’t realize they’re fighting for evil.

• They’re threatened by what they perceive as an even greater evil.

Now translate this into a Fantasy world. Let’s give evil commanding officers the power to take 10% of a player’s levels off them for “torture.” Who would fight on their side?

Well evil people would, because they get the same powers over their subordinates and they agree with the overall aims of the

evil empire. Non-evil people would scream like crazy at customer service the moment such a sanction was imposed on them. Those who felt particularly annoyed would quit and join the good side. If a greater evil was on the way, they’d quit and play another game.

In short, “evil” as a concept doesn’t work as a major force in virtual worlds. What we have instead is some kind of sanitized version that gives people a cloak to hide behind if they want to do something they think is a little naughty. It’s no more evil than the “good” side is good (when self-sacrifice is cheap, even evil people will do it).

“Evil” in virtual worlds is just an artificial construct that misapplies labels to promote a meaningless conflict. Is it without consequences, though? If players come to see “evil” as just a label, does that indeed desensitize them to the very concept of it? Will they be more likely to side with something labeled as evil in the real world?

At a very superficial level—that of the label—perhaps. If they are told that some dictator is “evil,” some individuals may not fully appreciate what this means; this could be true irrespective of whether they have played virtual worlds or not, though. Whatever, contact with the reality of evil will swiftly disabuse them of this naivety.

Nevertheless, the wicked will always prey on the gullible; if desensitizing people to the label means that those to whom it applies can draw innocents closer to them, it can’t be a good thing. Whether it’s unacceptably bad or merely one poor chord in the great song of popular culture is another matter ³.

Besides, the traffic is not one-way: Real-world ideas can affect virtual world concepts. In the great scheme of things, permanent death for characters is not actually as bad as people make out. However, because death in the real world is bad, this has caused people to redefine what “death” means in virtual worlds. They say their character “died” even when all that happened was it was teleported out of a battle. Are they becoming less sensitized to the notion of death in the real world because of this? Or does their use of “death” in this way arise precisely because they aren’t insensitive to it?

The remaining two primary concerns about the way that virtual worlds can be a negative influence are related to each other. They say that people who spend long periods in virtual worlds will transfer inappropriate emotional and intellectual understandings to the real world because they cannot subconsciously separate the real from the virtual. If you spend all day beating the guts out of NPCs, and NPCs look just like people, then in the real world you’ll start to objectify people too. Similarly, if you find that being lying and deceitful in the virtual world works, you’ll change in the real world so that you adopt this strategy too.

The objectification argument is a non-starter. People are acutely aware of whether opponents are computer-controlled or player-controlled. It makes a big difference to them ⁴. If anything, they are more likely to subjectify NPCs than to objectify players.

Virtual-to-real transfer does occur. The core of the attack against it is that it happens without critical reflection; in other words, it becomes part of your behavior through rote, rather than because you thought about it. Although I agree that it is possible to learn information and simple skills this way, I do not believe that behavioral changes can also be acquired except under very artificial circumstances. If you find that being lying and deceitful works, you’ll notice it consciously; you won’t just fall into it with zombie-like mindlessness. This is why, when other players start shunning you because of your behavior change, you may explicitly consider what it is you are doing wrong (that is, lying) and resolve your ways.

The way I see this is like the difference between episodic and semantic memory. Episodic memory is your memory of events (“yesterday I watched a movie”); semantic memory is your memory of procedures and processes (“when you watch a movie, you have to buy a ticket”). People who suffer from amnesia will forget their name, let alone what movies they have seen, but they’ll still remember that they have to buy a ticket. Semantic memory is an abstraction of episodic memory ⁵; from individual events, you build up a general picture that describes them, such that you can forget the individual events but retain what you have learned. I don’t remember how I found out I needed to buy a ticket to see a movie, but then I don’t need to; I only need to remember that I do need to buy a ticket.

With virtual worlds, the actions that you undertake lead to individual episodes. Although you remember the episodes immediately when they’re over, with time they will fade. Yes, some will be particularly memorable and will remain with you always, but most won’t. What does remain are the lessons you learned from the episodes, and these are quite different from the mere facts of the episodes themselves.

As a designer, I therefore don’t care whether people get a rush from finding a secret passage, from being a leader, from designing their own house, or from laying waste to a nest of vampires. What I care about is whether, in doing so, they develop as people. So, when someone says that virtual worlds encourage sex or violence, no, they don’t. Even if a designer went all out to try to encourage it, they wouldn’t. It may be that they encourage play of a sexual or violent nature, but play is play, and it’s virtual. The lessons that people learn are the lessons that they can take with them into the real world. They find out about people; they find out about themselves.

I believe that designing virtual worlds is a morally defensible occupation. I believe that they are on the side of the virtuous.

Mind you, I also believe that one day we’ll see more people getting married because of virtual worlds than we’ll see getting divorced because of them. Don’t let my idealism do your thinking for you. Consider the issues; reflect on them yourself; draw your own conclusions.

Confounding Expectations

There’s a gray area to this talk of what people learn from virtual worlds.

Immersion is driven by a virtual world’s similarity to Reality. It is founded on players’ implicit expectations. Some of those expectations concern virtual worlds themselves. Should these expectations be confirmed or frustrated? Do you, as a designer, give the players what they expect (for example, races) or what they don’t (for example, everyone’s human)? The issue is not the effect on immersion (it’s not all that great); rather, it’s the effects that confounding these expectations has on individuals.

If you give someone something that they are not expecting, but which is consistent with the way the virtual world operates, then they will either accept it or reject it. Either way, they’ll have to think about it; this is why challenging expectations is good. Obviously, if too many are challenged at the same time it could impact on immersion, but in general this isn’t going to happen.

Challenging expectations causes people to think about those expectations. Through such analyses, players can change as people. As I have explained, this is how things should be; the lessons that people learn come from abstractions of events, not from the events themselves. You don’t learn that killing is good, you learn how to trust your friends; you don’t learn that Zeus is all-powerful, you learn that being all-powerful means being all-responsible. This is how it should be.

Yet immersion subverts this mechanism. You accept the virtual world because of its similarity to the real one. Your implicit view of how the real world works informs your view of how the virtual world should work. Is it not possible that the reverse could at times be true? What you implicitly accept without thinking in the virtual world could transfer to what you implicitly accept in the real world? As I said earlier, obviously anything complicated would require conscious thought—I’m not suggesting that players are likely to be brainwashed like

this—but nevertheless there are smaller things that could have real-world consequences.

Example: In a virtual world, acceleration is instantaneous. You go from a standing start to running full speed without an intermediate period of speeding up. If someone who has been playing in a graphical virtual world for 10 hours gets into their car and drives off, could that be dangerous? This kind of low-level skill acquisition is easy to pick up—people who access text-based virtual worlds from a command-line interface will routinely find themselves typing the short forms of “look” or “inventory” at the command prompt.

When designers challenge player expectations at this kind of level (rather than at a more cerebral level), the players have to accept it or reject it. Rejection means stopping playing, so challenges aren’t ever going to be major; they do exist, though. For example, in a graphical world if you want to walk between characters who are physically blocking your way, the world may let you do it—even though in the real world you couldn’t. The reason this is so is because otherwise players could barricade an area with their own bodies; nevertheless, it challenges players’ notions of their own physicality in the virtual world. If they don’t like it, they’re not going to become immersed and will eventually leave; if they do like it, it becomes part of their implicit knowledge of the way the virtual world works. With deepening immersion, there is the risk that this world view could be taken back to Reality. It may not last there for long, but it could still be enough to cause an accident.

Whose fault is it when such lapses occur? The player, for putting themselves into such a situation in the first place, or the designer for challenging their expectations?

Above the level of immersion, there is still cause for concern. Designers can cause players to re-examine their attitudes, but

what happens if the players conclude that they should behave in a way that is bad either for themselves or for society? I argued earlier that players don’t learn from the facts of what they do but from abstractions of those facts, but what if what they learn is inappropriate for their real-world culture?

For example, a virtual world may have lax attitudes to sexuality. A player may, after many experimental encounters of virtual sex, decide that they are not attracted to members of the same sex and are therefore heterosexual. This is fair enough, but if as a result of their not enjoying their homosexual encounters they drew all kinds of false conclusions about homosexuals in general, this would not necessarily be regarded as a universally positive view to transfer into the real world. Is the virtual world’s designer in any way responsible for this unfortunate result? After all, if they had not promoted lax attitudes to sexuality, the player would not have experimented and would therefore not have drawn any conclusions at all (false or otherwise).

Here’s another example. Virtual world designers create an environment in which people can explore their own identity. They can find out who they “really are.” It could transpire that they “really are” someone deeply unpleasant, who has previously been subduing their urges behind some façade that they now feel comfortable about dropping. Are virtual worlds best left alone if they regularly produce individuals like this?

Designers can influence the ways in which people change in virtual worlds. When they do so, does this give them a responsibility for the outcome? Or are they merely providing trials that will allow players to find what was already there deep inside them anyway? If I challenge a player’s expectations, am I equipping them with the means by which to determine their own solution? Or am I guiding them in a direction that may be inappropriate.

I’d like to believe the former. I can lead a horse to water, but I can’t make it drink. Yet I can make it consider drinking. Is it my responsibility if it does drink? Or is it the horse’s?

Yourself

When all’s said and done, the ethics of a virtual world reflect those of its designer. If you don’t think about how to behave, about what’s right and wrong, about responsibility, about rights, then why should your players? If you think ethics are other people’s responsibility, so will your players. Your beliefs, your attitudes, your personality—they’re all reflected in your virtual world. You have to take responsibility, because (at least initially) you are the world.

Here’s a final ethical problem for you.

Back in Chapter 3, I outlined A Story About a Tree. This concerned the death in real life of Karyn, a Norwegian beauty queen who played in LegendMUD. I used this anecdote to pose questions concerning what is “real” in a virtual world. At the time, I said I didn’t know whether Karyn was a real person. Actually, I do know whether or not Karyn was a real person: She wasn’t. She was a complete invention.

In February 2002, film producer Tracy Spaight got in touch with me about a documentary he was making, Real People, Virtual Worlds. He wanted to know if it would be possible to interview me, and sent along a preview tape of some interviews he’d done already. The centerpiece was a powerful monologue by Raph Koster in which he described and commented on A Story About a Tree. Although I had long harbored reservations about Karyn, I’d kept them pretty much to myself. When it came to a movie that

was likely to be seen on television by millions of people, though, well, it would have been somewhat embarrassing for Tracy and Raph if Karyn turned out to be fake. I suggested to Tracy that he check out the story, just to be sure. Someone at LegendMUD would probably have met or spoken to Karyn in real life when she was alive, so it shouldn’t take long; it was just to put my mind at rest. Except, no one at LegendMUD had actually met her.

I hadn’t reckoned with Tracy’s documentary-maker’s instincts. He methodically investigated every shred of evidence that remained concerning Karyn’s life. It took a great deal of time and patience, but the facts he unearthed were undeniable. Here are some of the highlights:

• There are only three people in the whole of Norway called Karyn. It’s a highly unusual name there.

• The total number of deaths in road traffic accidents in Norway in 1995 was 352. Road death is relatively uncommon in that country.

• Only one person died in the crash that supposedly killed Karyn. The report reproduced on Karyn’s web site said three people; it had been altered in translation.

• The University of Oslo, which Karyn said she attended, had no records of a former beauty queen (or anyone else) having died in a road accident in January 1998.

• The Frøken Norge (Miss Norway) beauty pageant had no entries named Karyn in 1995 (when Karyn was supposedly a finalist).

• Karyn’s picture on her web site was of a bona fide Frøken Norge contestant. She had never heard of LegendMUD until Tracy tracked her down.

There was more—lots more. Police accident reports, depositions from Geocities, newspaper archives, and so on.

So Tracy and I knew that Karyn didn’t exist, and could prove so categorically. We sensed that the news would elevate Tracy’s documentary to mainstream levels of interest. Real people can have real relationships with imaginary people? That’s a great hook! I could imagine reading a syndicated version of the story in UK Sunday newspapers a few years down the line, as happened with Julian Dibbell’s A Rape in Cyberspace.

But just a moment…

Real relationships? Involving real people, who have suffered real grief? Couldn’t it be potentially devastating for them to discover the depths of Karyn’s betrayal. It might not be right for us to go public with the story. The world might be a better place if we just sat on it.

That wasn’t all we had to consider. A Story About a Tree is very well known; it’s almost iconic. It has been cited many times as a counter to the argument that “it’s just a game.” Yet for Karyn, it was “just a game,” at least at some level. Would it be better for the long-term development of virtual worlds for A Story About a Tree to keep its integrity? Or could more useful lessons be learned from a reappraisal of it in this new light?

So we had a problem: What should we do?

I won’t tell you what we did do, although obviously it resulted in our revealing the truth about Karyn (or you wouldn’t be reading this ¹). All I’ll say is that both Tracy and I felt our actions were right.

Instead, I’d like to pose the question: What would you have done in our position?

Now, ask what your answer says about you. What does that say about your virtual world? What in turn does this say about your players?

It’s your design; it’s your morality. From Essex to ethics in one book….