Judging from the comments thread at TCS, it seems that I failed to adequately convey that political libertarianism is by no means an amoral theory. Political libertarianism assumes that a peaceful, stable, fair, extended social network of mutually advantageous cooperation — liberal order — is, if not morally good in itself, at least good as a means to other moral ends. The point, however, is that people with different commitments can support a liberal order, and can account for the moral value of the order in different ways. When you live in a large, incredibly pluralistic society like ours, the problem of how we all can live together, despite our differences, is a serious problem no matter what you happen to believe. A minimal set of social principles that accomodates the broadest array of commitments and worldviews can be seen by all sorts of people as the best solution to that problem.
This also does not imply that comprehensive justificatory strategies are false. Suppose, say, Ayn Rand is right. Then Ayn Rand is right. But the probability that everyone comes to agree with Ayn Rand is, well, zero, give or take. (The probability that the people who claim to agree with Ayn Rand will come to agree with each other is probably no better.) Whatever the correct comprehensive theory is, it's probably never going to be the case that everyone believes it. An authoritarian order can probably coerce agreement, to an extent, by restricting freedom of thought, speech, and inquiry. But that's not the kind of society we want. And a small, homogenous community, a group of Hutterhites, for example, might share a common conception of the good. But we're talking about a huge, diverse society.
So, one might arrive at the one true theory of the good, and even do a bang up job of spreading the word, but still be swamped by Babelian pluralism. The problem simply isn't how to get everyone to agree on fundamentals, because it's a problem that won't get solved in a big, free society. What we're left with is a sort of engineering problem. What terms of association, what social principles, can accomodate all these people, and all these diverse commitments, in a manner (almost) everyone has reason to affirm. The hypothesis is that political libertarianism is the best solution to the engineering problem.
Now, I'm by no means sure that this hypothesis is correct, or even exactly what political libertarianism entails (and thus what the hypothesis really is). I think I'd just want to call my own view liberal minimalism. I'm receptive to the idea that some small-scale redistribution might be a condition for stable liberal order, putting me in the company of Hayek, Milton Friedman, and Loren Lomasky. While people tend to identify these thinkers as libertarian, people also tend to think libertarianism by nature rules out redistribution. So I'm not quite sure what to call myself, not that it matters much.