According to my indirect utilitarianism, we should institute whatever legal rights would best promote human flourishing. Legal rights derive whatever normativity they have from this consequentialist foundation. We have no reason to believe in such entities as "natural rights" -- they are at once metaphysically mysterious and ethically inadequate. Moreover, it would be fetishistic to care about such entities fundamentally and for their own sakes. What really matters are people, and rights are only important insofar as they benefit real people.
Moreover, it is a contingent matter what rights would most help people. (To prove this point as strongly as possible, my next post describes a hypothetical situation wherein a society would be morally required to limit even the right to life itself.) So you cannot just start off with some set of rights as "given", a priori. What rights are worth having will depend on empirical facts about their likely consequences. So the rights themselves cannot be fundamental. They're instead derived from a utilitarian foundation, in conjunction with contingent empirical facts.
This means that rights may legitimately vary from society to society, in much the same way that the moral status of an act-type like "lying" might vary across situations. The difference is not due to different opinions, "beliefs", or other such relativistic nonsense. Rather, it's because the objective facts about consequent harms and benefits may differ between the situations, and the moral facts must be responsive to these non-moral facts. (That's why moral absolutism is so misguided -- it entails a gross insensitivity to morally relevant considerations. Just think of the standard "enquiring murderer" cases, where lying is entirely appropriate.)
Now, we must distinguish between two levels of moral criticism. At the 'practical' level, we assess particular actions "internally" against the standards of our everyday morality and legal/institutional processes. This is appropriate insofar as those standards are themselves justified. But to "externally" assess the standards themselves, we must ascend to the 'critical level' of utilitarian assessment. As previously explained, we should settle on whatever practical standards would do the most good.
Now, my previous post sought to illustrate the oft-neglected fact that property is a contingent "social construction", and hence we should "open our eyes to the possibility of various different systems of property rights" in order to "make an informed moral choice" between them. Along the way, I pointed out that the propertarian claim that "taxation is theft" involves a category error. Theft is an 'internal' violation of instituted property rights, whereas taxation is part of the institution and hence must be critically assessed against the appropriate 'external' standard (which I take to be utility).
It is important to recognize this external standard as part of my view. That's why I emphasized in my previous post that not all possible institutions would be equally legitimate. This is not some wishy-washy "anything goes" relativism. (Again, I took care to make this very explicit.) Governments are very much open to criticism if their institutions are detrimental to human wellbeing. They are morally required to establish the basic legal rights that would do the most good -- and hence morally culpable to the extent that they fail to do this.
(It's also worth noting that government actors might violate their own institutional requirements. Just look at the Bush Administration's disregard for the U.S. Constitution and limits on executive power. It would still be wrong to amend the Constitution to achieve the same effects. But it would clearly be wrong in a different way. It is a virtue of my theory that it can account for this.)
If one is particularly enamoured of "rights"-talk (and I'm not), one could go beyond legal rights by adopting a conception on which "Human rights are... moral claims on the organization of one's society." In this sense, talk of a "right" to X is really just shorthand for saying that we ought to establish institutions which grant us a legal right to X. That may be true enough, but we should recall that the ultimate basis for the "ought" claim is a utilitarian one, as argued above. On my view, moral significance fundamentally derives from real harms and benefits, not abstract quasi-magical "natural rights".