Image: Jefferson Memorial by kjd (CC BY-NC-ND).
While the discussion of liability rules and property rules (below) may be adequate for fungible interests, it may not be appropriate in cases of interests which more closely touch the personality of the participant. For these latter interests, inalienability, or partial inalienability may be the best method for protecting the personhood of the participant.
In a 1987 article (Radin, Market Inalienability (1987) 100 Harv. L. Rev. 1849), Margaret Jane Radin suggested that there are three main arguments to justify market-inalienability based on personality interests. Lets take the example of the term in the contract which states that a participant can be removed at any time for any or no reason, and consider the arguments for inalienability of the corresponding entitlement not to be removed from a virtual environment without due process. The analogous real-world right, as against the government, is (increasingly, somewhat) inalienable. Against private actors, it is a property entitlement held by the owner of the land. Absent strong arguments to the contrary, the presumption in a virtual environment will be that participants remain in the environment by the consent of the owners of the environment which, while subject to agreement, is revocable.
The first ground canvassed by Radin is a prophylactic argument – where the risk of harm to personhood of giving up the interest is so great that we are willing to constrain the choices available to those who would willingly give it up. In this case, we would be saying that we are willing to presume that all instances where a person gives up the right to due process to be coerced (see Radin at 1909 using slavery as an example). I am reminded here of Bartle's warning that there are any number of reasons that a person may wish to play a game with entirely arbitrary rules. The assumption that all such agreements are coerced simply cannot stand, and the question of consent must accordingly be reduced to a question of fact. However, if we remove the Bartle-world case, we begin to get an idea of the risk faced by participants – this is not simply an issue of losing access to a gaming platform, but of being cut off from one's social network, of having one's property forcibly removed, and of losing touch with the avatar – in the most extreme cases, of being forcibly alienated from a part of oneself. The danger posed can be evaluated quite strongly, and, particularly as the purported agreement is made before access is granted and before any attachment has formed, it may be fair to say that in all but the borderline Bartle-world cases, we are prepared to presume that the decision was coerced or otherwise not freely made.
A second justification given by Radin is that of prohibiting the commodified version of the 'good'. In this case, we may be able to say that there is a moral requirement that participation in the environment should not be commodified. The argument here is that allowing market forces to dictate whether we can associate with our friends and family or our avatars “creates and encourages an inferior conception of human flourishing”.1) In a world which encourages rich and diverse social relationships, to have those relationships subject to arbitrary severance by the platform owner may be damaging in itself. The counter argument is that we often allow rich and diverse human relationships to be governed by markets – although, in an idealised form, we may prefer that they were not. Radin's pragmatism deals with the non-ideal scenarios, and accepts that there can be a continuum of degrees of commodification, and that partial market-inalienability may “sometimes substitute for a complete noncommodification that might accord with our ideals but cause too much harm in our nonideal world”.2)
The third justification Radin gives is a domino theory; where commodification changes the nature of the 'good', such that non-commodified and commodified versions cannot co-exist, and there is a moral requirement that the non-commodified version is available, then a prohibition on the commodified version can be desirable. Radin explains that this “can be conceived of as the opposite of a prohibition: there is assumed to exist some moral requirement that a certain “good” be socially available”.3) In this instance, an argument may be that if we allow platform owners the ability to commodify and sever social relationships and avatar connections at will, we are unlikely to see the emergence of non-commodified systems. This suggestion is borne out, to a degree, by Andrew Jankowich's study showing that three quarters of virtual world agreements surveyed “allowed the proprietor to delete a player account at the proprietor's discretion.”4) If we believe that non-commodifiable versions of social relationships in virtual worlds *should* exist, and we also believe that while we allow commodifiable versions they will not emerge satisfactorily, then we may prefer a prohibition on the commodified version.
While there are many flaws in my under-developed reasoning, a model of partial market-inalienability may sometimes be suitable. While it would be folly to suggest that participants have an inalienable right not to be ejected from a private space, it may make sense to suggest that participants have an inalienable right not to be removed from a private space which very closely mimics public space without due process. This qualified market-inalienability may also be sufficient to allay the concerns raised by Bartle, in that participants and proprietors of virtual worlds which do not closely resemble public spaces will not be unduly burdened in their liberty to choose arbitrary rules. Similar to the way in which labour is partially commodified, where we allow the overall sale of one's productive force but impose limits in the form of minimum wages and unfair dismissal rules, we can envisage that participant rights in virtual worlds can be productively made partially market-inalienable. The difficulty, as always, will lie in establishing the boundaries.
Overall, I think that Radin's theory provides a fascinating way of approaching the topic of 'avatar rights' which could be very useful in determining which interests can be modified by Terms of Service, by code, and by internal norms in virtual worlds.
Comments, thoughts, or suggestions?