When does conduct by an online player in a virtual world game trigger liability for a real-world crime? In the future, will new criminal laws be needed to account for new social harms that occur in virtual worlds? This short essay considers both questions. Part I argues that existing laws regulate virtual worlds with little or no regard to the virtual reality they foster. Criminal law tends to follow the physical rather than the virtual: it looks to what a person does rather than what the victim virtually perceives. This dynamic greatly narrows the role of criminal law in virtual worlds. Existing law will not recognize virtual murder, virtual threats, or virtual theft. Virtual worlds will be regulated like any other game, but their virtualness normally will have no independent legal resonance from the standpoint of criminal law.
Part II turns to the normative question: Are new laws needed? It concludes that legislatures should not enact new criminal laws to account for the new social harms that may occur in virtual worlds. Virtual worlds at bottom are computer games, and games are artificial structures better regulated by game administrators than federal or state governments. The best punishment for a violation of a game comes from the game itself. Criminal law is a blunt instrument that should be used only as a last resort. The state's power to deny individuals their freedom is an extraordinary power, and it should be reserved for harms that other mechanisms cannot remedy. Online virtual worlds may seem real to some users, but unlike real life, they are mediated by game administrators who can take action with consequences internal to the game. Internal virtual harms should trigger internal virtual remedies. It is only when harms go outside the game that the criminal law should be potentially available to remedy wrongs not redressable elsewhere.
The article raises some very interesting points, but I was a little disappointed that Kerr gave such short consideration to online sex crimes. I wholeheartedly agree with Kerr that virtual rape is in no way like actual rape and should not be punished as such. However, I have argued that at least in some contexts, virtual rape could be cognizable as a property crime (since a person's avatar could be tainted, or diminished in monetary exchange value because of the sex crime). I probably agree with Kerr that a hands-off approach is preferable, but given that the Mr. Bungle incident is one of the more (in)famous online "crimes," I thought he could have done a bit more in that area.
However, I've been thinking a bit more about lately about a different sort of crime that can occur in virtual worlds. I wonder about treating enticement or corruption of a minor crimes differently in an online world. Certainly, if an adult beckons a minor to have sexual relations with that adult, and the minor does so in the real world, that is a "real" crime. However, I wonder if conversations about sexual matters between minors and adults represent more or less danger in the virtual world. On the one hand, there is a higher degree of safety with two persons separated by the wide expanses of the World Wide Web. On the other hand, there is greater opportunity for victim grooming where potential molesters can build trust in the virtual world. Ultimately, I think the government might want to develop different rules for these types of enticement cases for virtual communications. However, I haven't thought through how these new rules would work. Kerr's article certainly addresses some meta-questions in the area, but I wonder if a hands-off approach makes sense in all sex crime contexts.
Kerr's article is now the subject of an interesting online debate between Saul Levmore and Kerr. You can find the first few posts in the debate at these links (I'm sure there will be more, but I will be out of the country when this post actually appears):