If you mean along the lines of illegal restraint of trade, ICANN, even by its own admission, is a unique public/private hybrid (or lobrid), I don't think there is any legislation or caselaw that could easily be shoehorned to apply. That in itself would be a multi-year process. Look at the still unfolding attempts to rein in Micro$oft. That has taken years and $multi-millions and M$ is a known private entity presumably governed by existing legislation and caselaw.|
BTW, I don't see this as entirely bad, if M$ was more sure of where it stood it could singlehandedly take over web addressing functionality from ICANN using a RealNames like process. Their browser market share is at a stunning 95%. Purists (and even the somewhat impure like myself) might argue that that leaves out the other 5%, but if it came into common use there would be great impetus for that other 5% to come up with compatible solutions. And if it could make itself more useful and functional than the DNS then it could come into common use. This wouldn't be that difficult to do, much of the DNS is littered with forsale and dropjacked names and unreachability errors. Note well that one of Micro$oft's biggest beefs with RealNames was apparently that they didn't ensure the usefulness and functionality that they had promised. I suspect that someone could (and perhaps will) write an Internet Explorer plugin that enhanced the functionality and usefulness of the DNS (eg: through a database that removed unreachable, useless, and dropjacked addresses) and it could take off just through viral marketing because it would be an improvement on the DNS. With a greater than 90% captive audience, M$ doesn't even have to worry about viral, or any, marketing. They can just go ahead and do it, and something tells me once they feel they've dodged most of the current hail of bullets, they will. And as most of the pure and semi-impure are from the primarily English speaking world, the inclusion of IDN functionality would really rout, and route around, ICANN's root.
But I digress. Even if ICANN was clearly covered by existing legislation or caselaw regarding restraint of trade, they could, as Ray points out, play the 'stability card', they could also point to the 'proof of concept card' in which one of the proofs looked for was that there was a market for new gTLDs. If one factors out the speculators and defensive registrations and non-unique content of the new gTLDs and the existing gTLDs (and those ccTLDs that serve the same market), the new gTLDs have somewhere along the lines of 2% to a probably wildly over-optimistic 5% of the registrations of existing TLDs. Ben's research is particularily instructive in this regard. This doesn't imply that there is much of a market for new gTLDs, that there is any meaningful trade being restrained.
I know there have been some dark mutterings from some failed gTLD applicants regarding legal action, but they haven't initiated such yet and if/when they do, it could well take years to be resolved, and even then perhaps in ICANN's favour. By then I suspect that the DNS wars will be a historical footnote. We see this with M$ or the Auerbach lawsuit. Delay, and otherwise go about your business, until a ruling, even if it goes against you, becomes irrelevant. -g