Libertarian Theories of War:
In my post below I describe the libertarian take on war to be undertheorized. Two VC readers independently pointed me to two different articles by libertarians in a symposium on "War and Liberty," published in The Reason Papers. Though quite different from each other, both offer the type of effort I was calling for. I have only had a chance to read each paper very quickly, but I found them both to be very thoughtful and I wanted to get links to them up here promptly.

The first, War and Liberty, by Aeon Skoble makes many of the same arguments about sovereignty that I made in my post but goes beyond that to present a modified theory of soveriegnty. Here is a bit:
The problem is that the notion of state sovereignty in the modern era leads to a view of the moral equivalence of all states—Communist China is then no different from Republican Switzerland—and this is detrimental to human rights, because it means that a tyrannical state is immune from outside pressures to liberalize. Michael Walzer goes some of the way in this direction, but not to the ultimate conclusion. The argument is that sovereignty needs to be based in service to people, that is, protecting their rights, so illegitimate regimes don't have sovereignty at all. There's a Lockean component here also: If rights are conceptually prior to the state, then state sovereignty must derive from a theory of legitimacy which is based on protection of rights rather than from a theory of moral equality of all states.

The rights component gets lost when we adopt a "realist" model of legitimacy, such as actually holding power or being "recognized" by the UN. Now, what are the causes which might count as "just cause"? Least controversial is defense against aggression. The right to respond to force with force seems fairly straightforward, although in a moment I will indicate why it might not be for some. A bit less obvious is defense of another. If B is invaded by A, B might have the right to repel the invasion, but utterly lack the power to do so. C's assistance would be justified on the grounds that B was unjustified in aggressing against A in the first place. C's right to use force against A follows from B's right. More controversial still are interventions; for example, taking sides in a civil war or preventing a genocide or removing a tyrant. It might seem as though only in this last case does it even matter what model of legitimacy we adopt. If A is attacked, isn't A's right of selfdefense absolute regardless of whether it is attacked by a republic or a tyranny? Traditional just war theory would answer yes, but I think it actually does matter. Since tyrannical states have no legitimacy, if they are attacked by free states, they cannot claim that their sovereignty is being violated. In other words, intervening to protect rights against a tyrant is not a violation of sovereignty—at least not any kind of sovereignty worth defending. (Nevertheless, the attack would have to satisfy other justice conditions, e.g., it would have to be intended to liberate oppressed people or prevent a genocide rather than to seize raw materials or to acquire territory.)

Some will argue that a free society has no business interfering in other societies' internal politics. But this is, ironically, or paradoxically, a holdover from the old monarchist mindset. The old order on which traditional just war theory is based, and on which sovereignty is the paramount value in international relations, depends on a moral equivalence between states which is derived from a statist view, not an individualist view. On a non-statist, individualist view, individuals, not states, have rights. States may have powers, but the just powers derive from the consent of the governed. The putative right of any state to sovereignty thus is a function of its protection of the rights of the people in its domain. So a free society may very well have some business "interfering" in tyrannical or genocidal states—namely, the business of protecting life and liberty. The very language—that this is "interference" in a state's own affairs--implies that the state has some right of action which is presumptively respected, and again, this can only be justified by old-order thinking, not by liberal thinking. (I am not here arguing that they are obliged to do so, only that they are permitted to do so, or that they do no wrong by doing so.)
The second The Justice and Prudence of War: Toward A Libertarian Analysis is by Rod Long. Here is how it starts:
The morality of warfare is an issue that has long divided libertarians. The spectrum of libertarian opinion on the subject ranges all the way from Leonard Peikoff, who defends the use of nuclear weapons against civilian targets, to Robert LeFevre, who denied the legitimacy of all violence, even in self-defense.

Needless to say, most libertarians fall at various points between these two extremes — though the divisions have become sharper since the 9/11 attacks. (One of the more ironic manifestations of these divisions is that French libertarians are far more likely to support current US foreign policy than American libertarians are; perhaps anti-government thinkers tend to be more attracted to whatever position their own government opposes.)

What view of warfare is most consistent with libertarian principles? Here I shall distinguish between libertarianism as a normative ethical theory — a theory of justice — and libertarianism as a descriptive social theory. Libertarians disagree with one another as to the extent of the former's dependence on the latter; utilitarian libertarians profess to believe the dependence total, while natural-rights libertarians profess to believe it nonexistent, but in practice both groups tend to treat the dependence as partial, and so will I.

Deontological Considerations

The non-consequentialist core of libertarian ethical theory is an egalitarian commitment; specifically, a commitment not to socioeconomic equality but to equality in authority. Indeed, libertarians' lack of enthusiasm for enforced socioeconomic equality stems precisely from their concern that it can be achieved only at the cost of this for libertarians more fundamental form of equality.

The libertarian "non-aggression principle" expresses the conviction that forcibly to subordinate the person or property of another to one's own aims is to assume an unjustifiable inequality in authority between oneself and the other. And it is because this equality in authority likewise holds between private citizens and public officials that governments are forbidden to exercise any powers not available to people generally; libertarianism requires not just equality before the law but equality with the law.

It follows that a consistent libertarian theory of warfare must apply the same prohibitions and permissions to governments and private individuals alike. In this respect it will be radically different from nonlibertarian theories, which typically grant government actors more latitude in the use of violence than private actors; a libertarian theory must be equally permissive — or equally restrictive — with both. A consistent libertarian cannot, for example, accept a mere apology as sufficient recompense when the US military accidentally bombs the wrong target and kills fifteen children in Afghanistan[5]unless she is prepared to be equally tolerant when Uncle Zeke's backyard bazooka target practice accidentally takes out a passing school bus. It can make no difference whether the perpetrator is or is not an agent of the government; nor can it make any difference whether the victims are or are not citizens of that government.
I really need to subscribe to The Reason Papers, which are edited by Aeon Skoble.