I agree with most of the points Randy Barnett makes in his excellent op ed on libertarianism and war. Both the Iraq War specifically and defense policy more generally have historically divided libertarians, as I explained in two posts last year (here and here).

Randy makes a good case that libertarians who accept the idea of defensive war cannot categorically reject the possibility that sometimes the best defense is a good offense. They can, of course, ultimately conclude that limiting war to narrow defensive efforts against truly "imminent threats" is the best policy. But that conclusion requires additional evidence to support it, and cannot be deduced from the nature of libertarianism itself on a priori grounds.

I would go one step further than Randy, and suggest that even nondefensive humanitarian military intervention is potentially compatible with libertarianism. This is a paradoxical claim. After all, libertarianism is nothing if not an ideology of deep skepticism about government-controlled enterprises and wars of any kind clearly fall into that category. Certainly, libertarians oppose all or most large-scale government programs in the domestic realm; if they didn't, they would no longer be libertarians.

Why is warfare different? Because in foreign policy, unlike in domestic policy, the alternative to US government action is often not the free market or civil society, but instead the continued rule of other governments. If the US government abolishes Domestic Program X, that usually means that the field is left open to private sector actors. Libertarians generally assume that the private sector will handle the issue more effectively - and with less infringement on individual rights - than the government would.

By contrast, let us assume that President Reagan decided not to invade Grenada in 1983. The alternative to Reagan's action was not private sector control of Grenada or a libertarian minimal state in that country, but the continued rule of Grenada's communist dictatorship. That dictatorship was, of course, itself a government. Moreover, it was a much worse government - in libertarian terms - than the liberal democratic regime that the US installed after the invasion Reagan ordered. On balance, it is highly likely that the US invasion of Grenada was a net gain in terms of promoting libertarian values, even when one factors in the loss of life and property in the fighting. Grenada is perhaps an easy case for pro-intervention libertarians. Other cases, including Iraq, may be much harder. Nonetheless, it does illustrate one example where military intervention clearly advanced libertarian values far more than it undermined them.

The key insight here is that question of humanitarian military intervention is not a tradeoff between government and the private sector, but is usually a tradeoff between two different governments - the status quo and the one the invaders plan to install. Sometimes, the latter government will be much better from the standpoint of libertarian values than the former. Obviously, one also has to factor in the lives and resources lost in the fighting. Sometimes, these will be so great as to counsel against intervention even in a case where the status quo government is extremely oppressive. The current government of North Korea is perhaps the worst in the world, but its possession of a large army and nuclear weapons ensures that trying to forcibly remove it may be even worse - from a libertarian point of view - than leaving it alone.

Whether or not humanitarian military intervention can be justified on libertarian grounds will vary from case to case. It depends on how unlibertarian the current government is, how much better the new one is likely to be, and how much loss of life and property will occur as a result of the fighting. Libertarianism gives clear, determinate answers on most questions dealing with tradeoffs between the government and the private sector. It is not nearly so unequivocal on issues dealing with tradeoffs between two or more governments. The question of war and military intervention usually falls into the latter category.

None of this suggests that one cannot be a libertarian and still oppose virtually all military action other than narrowly defined self-defense. If you believe that offense is rarely an effective form of defense and that humanitarian intervention nearly always leads to the installation of governments as bad or worse than those they replace, libertarian isolationism becomes the right policy prescription. However, such a conclusion does not flow from the intrinsic nature of libertarianism itself. It requires extensive additional empirical and theoretical analysis to justify it on libertarian grounds. For that reason, libertarians will continue to disagree over war and military intervention.