In the startup of a virtual world, Garriott says, the players—like Iraqis—face anarchy, confusion, and unclear rules. They are poor, they are at the mercy of brutal spoilers (players who rob and kill other players for kicks), and they are subject to a whimsical, alien overlord (the programmers). Of course players don't actually risk their lives, but they are passionate about constructing a successful society, and there are hundreds of thousands of them.
Virtual worlds with thousands of players may not offer much useful economic insight for Iraq or help anyone understand Iraqi social structure. But, says Garriott, the games do clarify the essential rules for stabilizing a chaotic society. Virtual worlds teach that there are really only two of those rules, one obvious, one surprising.
The first is the urgent need to protect lives and property. Ultima was plagued by murder and theft from its earliest days, as players exploited software loopholes to wreak havoc and get rich. As a result, other players quit the game or simply become villains themselves. Garriott says they had to fix the code and evict the anti-social players who were ruining the civilization for everyone else. Ultima didn't take off until the caretakers established security and law. Neither can Iraq.
The second requirement is an idea that hasn't gotten much attention from the U.S. occupation. It is that the ruler must let the people know he has heard their complaints. In a virtual-world startup, thousands of players gripe about the same thing (there's not enough money, my character keeps getting robbed …). It's incredibly important, Garriott says, that the ruler acknowledge he has heard the complaints. Not acknowledging complaints makes people nervous: It destabilizes and enrages them. Even if you have a plan to deal with a problem, you still have to let participants know they have been heard. Otherwise they panic or turn to some rival power that does admit their complaint. Broadcasting the acknowledgement to the whole community—"yes, we know you don't have enough running water"—is as essential as actually fixing the problem. Only once you have publicly recognized the problem, Garriott says, do you present your plan to remedy it.
3. The Kingmaker. Edward Castronova, an economics professor at California State University, Fullerton, is perhaps the most creative thinker about the real-world applications of virtual worlds. (Castronova's article on the economy of the game EverQuest is by far the most popular article in the leading online economics research archive. You can download it here.) Castronova has lectured at defense industry seminars on the real-world utility of online games. Last winter, before the Iraq war started, Defense Department officials asked him if he had any suggestions for how to use games in rebuilding what they called "a Southwest Asian country." Castronova recommended that DoD consider updating the 1970s game Kingmaker.
Kingmaker is an English board game based on the War of the Roses. It is now available as a computer game. Players, representing factions of English nobility, try to accumulate enough power to get their candidate crowned king of England. They have to seek the support of the church and Parliament, conquer castles and towns, and form alliances with other factions. "It's an accurate dynamic for a situation like Iraq—a country balkanized and regionalized, with an emerging democratic process," Castronova says. "People have different kinds of power, some military, some religious, some economic. Whoever is going to win is going to have to cobble together different kinds of resources."
In Castronova's ideal world, Defense and State would put their heads together and redesign the game to reflect the social and political realities of Iraq. It would be much more complicated than the original: It would need to estimate the possible strength of various Shiite and Kurdish factions, the influence of the Turks, the potential oil wealth, the ill-defined role of exiles, the extent of this mullah's power versus that one's, etc. Once they refitted the game, Castronova says, they would have a computer run the game a million times and see what kind of outcomes they get. The result of any particular game wouldn't tell much about Iraq, but taken as a whole, the results might indicate the probability of certain outcomes—for example, one-third of the time, the Kurds secede despite promising not to. In addition to handicapping the policy odds, the repetition would turn up results that planners might not have expected, the scary outliers: Perhaps you'd find that 1 percent of the time, Shiite radicals commit genocide against Sunnis.
The game would remind policy-makers not to get locked into a favorite theory. Even if some outcome seems most likely, there is still a possibility that, say, a Baathist guerrilla movement takes hold or Iran invades. A modified Kingmaker would be a cheap, fast way to get American leaders thinking about a wide range of possible outcomes, not just the few that seem likely.
Castronova, however, says he never heard back from DoD.
4. NorthKorea.com. Policy analysts are usually wrong for the same reason stock analysts are usually wrong. One person sitting at a desk at the State Department, even one person with the best information the CIA can buy, is trapped by bias and ignorance.
Wall Street solves the problem of individual bias with mass. Individual stock-pickers rarely beat the market over time because the collective intelligence of the market—millions of individual decisions—is always better than any one person's repeated guesses. In a fantasy future, online games could harness the collective intelligence of thousands of people to develop better policy.