The central jus ad bellum debate emerging from the War on Terror is how to justify the use of force against a non-state actor (NSA) located within the sovereign territory of another state. The standard move (and the US position) is to view the use of force as permitted as an exercise of self-defense sparked by an attack committed by the NSA. The 9/11 attacks are the paradigmatic example.
The problem, of course, is what to do about the sovereignty of the state on whose territory the defensive force is applied. First, if the state consents, the problem disappears. Second, if the actions of the NSA can be attributed to the state on whose territory they reside, the problem also disappears. The real question is what happens if attribution is not possible.
One view is that the defensive use of force is permitted if the host state is unwilling or unable to stop the NSA from using its territory as a staging ground for attacks. Ashley Deeks has a new essay expounding on this standard. Under her view, the host state is liable to attack if it is unwilling or unable to prevent its territory from being used by the NSA -- a standard with historical roots in the law of neutrality which permits attacks against co-belligerents who refuse to declare their neutrality or who are unable or unwilling to prevent their territory from being used by belligerents. Kevin Heller at Opinio Juris objects to this standard (as he has before) as unsupportable under international law. According to Heller, there simply isn’t sufficient evidence of state practice backed up by opinio juris to demonstrate that the law of neutrality has been used within non-international armed conflicts with NSAs.
I want to step back for a second and address this question at a deeper level of abstraction. The question is how one understands the violation of sovereignty. To my mind, there are two possible views here. In the first, any finding of a violation of sovereignty is per se an indication that the defensive force is illegal. In the second, though, a finding of a violation of sovereignty is just the beginning of the analysis, but not necessarily determinative that the action was illegal.
How can a finding that a violation of sovereignty has occurred not trigger a finding that the action was illegal? Well, consider first the doctrine of necessity in international law (see this article on the topic by Robert Sloane). I don’t think the doctrine could apply in this case, because its preconditions are quire narrow, but the structure of its analysis is instructive. In a case where necessity were to apply, one would fine that a state’s sovereignty had been violated but also find that the other state’s actions were nonetheless lawful because they were necessary to avert a profound national interest, such as saving a large number of their citizens from a natural disaster. Incidentally, this is the same type of analysis one finds in domestic necessity cases, particularly in private law. The ship owner docks his ship to another person’s dock during a storm, damaging the dock in the process. The dock owner’s property interest is violated, but the action is lawful because it was necessary for the ship owner to do. (And, on an efficiency level, the ship-owner can compensate the dock owner for the smaller damage to the dock, as in the famous Ploof v. Putnam).
A similar structure exists in how civil law jurisdictions hand cases of individual rights. Whereas common law jurisdictions will simply ask whether an individual has a particular right, civil law jurisdictions will sometimes ask first whether there is a right, and then only second ask whether it was an abuse of rights (abus de droit) to assert or exercise that right or whether that right should give way under the circumstances. As George Fletcher famously described it in The Right and the Reasonable, this is a two-step or textured structure of analysis, as opposed to a flat or one-step structure that one finds in American-style rights analysis.
My point here is to insist that the so-called "extraterritorial" use of force against NSAs can be analyzed in this fashion. One can concede, for example, that the territory and sovereignty of the host state is violated during a defensive attack against the NSA, but this does not necessarily entail that the use of force against the NSA was, all-things-considered, illegal.
Of course, one needs some doctrine to explain why force is appropriate against the NSA. On this point, I think it is essential to understand the natural law origins of Article 51 of the UN Charter. As I have argued before, Article 51 carves out the inherent right of self-defense from the Article 2 prohibition on the use of force. The reference to inherent is, in my view, a recognition that the right stems from natural law, a point made even more clear in the equally authoritative French language version (droit naturel) of the Charter. The natural law origins of the right suggest that the right is not limited to the state-centric confines of positivist international law; attacks by any group, including NSAs, trigger the right of self-defense within natural law, and the positivist international law embodied by Article 51 preserves (or carves out) that pre-positivist right. As a final note, the fact that Article 51 simply says “if an armed attack occurs” rather than “if an armed attacks occurs by a state” further emphasizes that attacks by NSAs trigger the natural right of self-defense.
Ah, but what of the ICJ’s opinion in The Wall case, the most recent and authoritative decision by the World Court to address this issue? (Heller also criticizes the Deeks essay for failing to adequately consider ICJ jurisprudence). That’s a good point, but I would ask any reader to look at that Wall opinion and see what sources of law the ICJ opinion cites for the proposition that attacks by NSAs do not trigger the right of self-defense…