May 7th, 1998

Who Are These People Anyway?

People tend to think that [virtual worlds] alter how people perceive one another. That gender and race and handicaps cease to matter. It is a noble vision, sure... In truth [they] reveal the self in rather disturbing ways. We all construct 'faces' and masks to deal with others. Usually in [real life] interpersonal relationships, the masks can slip, they evolve and react, and they have body language and cues. [In a virtual world], on the Net, whatever--they cannot. And people see specifically this: what you choose to represent yourself as, and that is more revealing of your true nature than gender, race, age, or anything else... it's not a matter of how we hide, it's a matter of how we are revealing ourselves.

The above paragraph comes from an unpublished interview I gave many years ago now. It came in response to the question, "How do you think virtual worlds affect people's perceptions of each other?"

A tangled question. Many seized on the sentence, "Thank heavens for playerkillers" in the last essay, and used it as evidence that I, or UO, am "on the playerkillers' side." Unfortunately, that's not only incorrect, but a reductionist view of a tangled situation. A better question to ask is, what exactly is the population of an online world, and what social forces drive it?

In discussing the Other yesterday, one word seemed at the center of the issue: Power. The conflicts that arise are there precisely because competing agendas (and often, as in the case of the playerkillers versus the roleplayers , competing play styles) attempt to exercise power over one another. I got a letter from Kazola, proprietor of the Treetop Keg and Winery on Great Lakes, saying that the tavern is not famous for being a target, but for being a roleplay haven first. It became a a target because of that fame. Yet I would still argue that it had the roleplay fame within a narrower segment of the overall UO community than its fame as a "flickering light in the darkness." And it is worth examining why exactly this is so. Why did it become a target just for being what it is? And why was its struggle so compelling?

Richard Bartle, who along with Roy Trubshaw is generally credited with writing the first mud (multi-user dungeon, if you wish to call it that, but let's say "virtual online world" instead) wrote an essay which among designers of virtual worlds is often considered essential reading. In it he classifies players into four types:

Those who seek to interact with other people, or Socializers
Those who seek to dominate other people, or Killers
Those who seek to learn and master the mechanics of the world, or Explorers
Those who seek to advance within the context of the world, or Achievers

Now, these are simplistic definitions, of course, and there is plenty of debate over the exact mix, and whether these are reductions to stereotypes, etc. It is interesting to note that "roleplayers" aren't even on his list, though they are generally considered to be a major force in online gaming--under this system, they are merely a variant of socializers, and the line between in-fiction chatting and out-of-character chatting is blurred.

The fascinating part of the essay, however, is where Bartle discusses the interactions between these groups. Killers are like wolves, in his model. And therefore they eat sheep, not other wolves. And the sheep are the socializers, with some occasional Achievers for spice. Why? Because killers are about the exercise of power, and you do not get the satisfaction of exercising power unless the victim complains vocally about it. Which socializers will tend to do.

Further, Bartle pointed out that eliminating the killers from the mix of the population results in a stagnant society. The socializers become cliquish, and without adversity to bring communities together, they fragment and eventually go away. Similarly, achievers, who are always looking for the biggest and baddest monster to kill, will find a world without killers to be lacking in risk and danger, and will grow bored and move on.

Yet at the same time, too many killers will quite successfully chase away everyone else. And after feeding on themselves for a little while, they will move on too. Leaving an empty world. However, since killers tend to know the world really well, there are not many ways of keeping them in check. From the playerbase, the explorers are the only ones with a real chance, because they know the game better than anyone.

Among some virtual world designers, the dichotomy is simpler: you have what they term "GoP" players, or goal-oriented players, and you have everyone else: the roleplayers, the socializers, etc. And the conflict is always between these two types. My own preferred metaphor goes back to the work of child psychologist Bruno Bettelheim, who discussed the two ways children tend to amuse themselves. One form is the "game," where there is a winner and a loser. It is competitive, and may or may not involve team play. The other is "play," which is non-competitive. It can be as simple as chatting a lot, or it can be building blocks, or (as it often is with children) it can be make-believe-- which is after all, roleplaying.

You may have recognized yourself in one of these models, and may have recognized situations and events from UO as well. Now, many of the responses to the first essay on the web boards and the newsgroups discussed how idealistic a vision it expressed... and thus viewed the second essay as a reversal. Yet really, the dichotomy of game and play are two sides of the same coin. One does not tend to exist anywhere without the other. Whereas the most idealistic vision of a virtual world, the fully community-oriented one, would seem to be composed of only "play," in fact it would founder. It's in human nature to need both, in one way or another. It's the nature of reality--and therefore the nature of virtual reality as well.

From a strict in-context perspective, the actions of a killer within a virtual world can be seen as sociopathic: they do not recognize the mores of the society in which they operate. This is not to say that they are bad people-- it has been well-established that interactions in a virtual setting create a level of psychological disinhibition that encourages freer action, less inhibited speech, and perhaps a little less thoughtfulness; this is probably largely due to the intoxicating sense of anonymity that we feel online. One has to wonder what the proper method of controlling people is, when they are generally not bad people, but merely "drunk" on the sense of anonymity.

Ideally (yeah, back to those pesky ideals), we bring them to an awareness of the virtual community they are disrupting, whilst at the same time still permitting people to (in final analysis) exercise power over one another, because people tend to seek status and power, and it's an important mechanic we cannot do without.

To boil all the high-flown stuff above down into simple premises: we must have playerkillers in UO, because the world would suffer if we did not have them. But they also must be channeled, so that their effect is beneficial, and not detrimental. And they have to learn to act within the context of the "play" space, and not perceive UO as just a "game" space.

In other words, a lot of it is about education. And most roleplayers have a story to tell about the time they first introduced someone to that form of play, and the way in which the former killer or hack 'n' slasher tentatively started trying out new waters, and eventually discovered that "hey, this roleplay things isn't all bad!" They may not become true roleplayers, but they may also adapt their play style to conform more to the virtual context. The mud designer and theoretician J C Lawrence terms this "functional roleplay," where behavior patterns of those who do not roleplay are conditioned by the presence of things like social systems, reputation systems, and other "channeling" devices.

It's largely about perspectives. The issue for the killers is whether they will gain the wider perspective and cease to be "virtually sociopathic." And the issue for the socializers is whether they will recognize that the killers are a part of their society too, and not always a bad one.

The thorny issues that then remain are the nitty-gritty of virtual community building: how do we govern in a world of anonymity? How do we police, and who polices, the players or the game administrators? What sort of punishment is appropriate for virtual crime? What sort of punishment is even possible for virtual crime? The answers to these questions that the UO community seeks out will shape UO for years to come, because they are questions that we the designers must ask of the players--no tool we give to players will work unless players take it up. And it could be that not a large enough proportion of players are ready or willing to take them up. But with the formation of governments and militias, we already see that the UO community is "self-aware"--aware of itself as a community, and therefore implicitly asking for tools to define its own society.

In the end, being a "killer" or a "roleplayer" is just as much a mask as the character one chooses to play online. It reveals something about how the player perceives UO, but not necessarily about their actual nature. (As a classic example, not all playerkillers are 13-year old boys, as popular legend would have it. What makes a playerkiller is a perspective on on the world, not an age.) As designers, our role is to juggle the often conflicting perspectives.

The answer to "who are these people anyway" is better phrased as "who am I, in this virtual reality?" And until a player can answer that well enough to understand their motivations, they may not even be playing the way they really want to. The Greeks put it as gnothi seauton--Know Thyself. If you find these simplified classifications of player styles to be confining--make your own. UO is both a play space and a game space, and that is at the root of all the most wonderful things about it, and also at the heart of the most painful issues it faces with virtual community, playerkilling, and the like. "Giving up" and targeting only half of that equation is not a fruitful approach, in the long run. Only education, self-knowledge, and an awareness of others is.

And education, self-knowledge, and an awareness of others sounds a lot like the process of growing up. Hearkening back to yesterday--something is being born. But we do have to teach it how to walk. More on that tomorrow.

-Designer Dragon

For further reading, for the interested:

Richard Bartle's article Hearts, Clubs, Diamonds, Spades: Players Who Suit Muds
Julian Dibbell's classic article on a virtual world community discovering itself, A Rape in Cyberspace
A simplistic personality test can maybe help you judge which of Bartle's types you fall under

In addition, Kazola mentions that the Treetop Keg and Winery is thriving-- congratulations. Another victory against the forces of darkness--the darkness of social collapse, that is!

Many many people have asked to reprint the first essay, "A Story About a Tree," and just as many have asked if it is a true story. Yes, it is a true story. Please do feel free to pass it around to friends, if you feel it has touched you.

One last note: some people were troubled by the idea that the concepts about the necessity for playerkillers and the like translate back into real world society. The answer is, of course, no. The essays are about virtual worlds and not about the real one. Often concepts translate in one direction, not the other.