08 Oct Eric Posner Rejects the “Unwilling or Unable” Test!
Julian beat me to Eric Posner’s new Slate article on the legality of drone strikes. I don’t agree with everything in it, but I think it’s notable that Posner — echoing his sometime co-author Jack Goldsmith — rejects the idea that international law permits self-defense against a non-state actor whenever a state is “unable or unwilling” to prevent the NSA from using its territory as a base for attacks. That rejection emerges clearly in the following passages:
The U.N. Charter permits countries to use military force abroad only with the approval of the U.N. Security Council, in self-defense, or with the permission of the country in which military force is to be used. The U.N. Security Council never authorized the drone war in Pakistan. Self-defense, traditionally defined to mean the use of force against an “imminent” armed attack by a nation-state, does not apply either, because no one thinks that Pakistan plans to invade the United States. That leaves consent as the only possible legal theory.
In other cases, including current drone operations in Pakistan, the United States has invoked a new idea of the “unable or unwilling” country, one that outside powers can invade because that country cannot prevent terrorists located on its territory from launching attacks across its borders.
The “coerced consent” doctrine, the “unable and unwilling” doctrine, and the exception for humanitarian intervention all whittle away at whatever part of the law on United Nations use of force blocks U.S. goals. If the United States ever decides to invade Iran in order to prevent it from acquiring nuclear weapons, expect a new doctrine to take shape, perhaps one that emphasizes the unique dangers of nuclear weapons and Iran’s declared hostility toward a nearby country.
I couldn’t agree more with Posner’s rejection of the “unwilling or unable” test. I’ve been watching with equal parts bemusement and concern as that standard spreads in the United States — with little or no effort on the part of those who defend it, of course, to identify the (non-US) opinio juris and state practice that ostensibly support it. The “unwilling or unable” test has even found its way into the Stanford/NYU report on drone strikes in Pakistan, which is otherwise so critical of US policy. Here is one of its statements about the jus ad bellum:
Further, it must be shown that the host state is “unwilling or unable to take [the appropriate steps against the non-state group].” Pakistan has at times failed to act decisively against non-state groups, raising questions about its ability and willingness to take necessary steps.
The quoted language, not surprisingly, is from Ashley Deeks’ article on the “unwilling or unable” test — an article that, as I have pointed out before, not only fails to establish that the test has achieved customary status, but actually admits (in a footnote) that it has not done so:
I have found no cases in which states clearly assert that they follow the test out of a sense of legal obligation (i.e., the opinio juris aspect of custom), nor have I located cases in which states have rejected the test. Even if one concludes that the rule does not rise to the level of custom, however, the rule makes frequent appearances in state practice and therefore is the appropriate starting point from which to determine how the norm should develop.
That footnote, of course, is never mentioned in articles and reports that uncritically adopt the “unwilling or unable” test — thereby further facilitating its spread. I’d like to think that Posner’s rejection of the test may help stem the tide. Unfortunately, as Posner himself explains, nothing — especially not international law — gets in the way of legal theories that enhance the US’s ability to use force abroad.
That is odd. I always understood “unable or unwilling” as a species of self defence, not consent.
Allowing for time zones, Jordan is probably asleep, so I will get in first!
‘Unable or unwilling’ is not a separate test — it is a subset of the ‘necessary’ part of national self-defence. Or put it this way. A NSA with access to heavy calibre weapons is firing those weapons from the territory of State B into the territory of State A. State A contacts State B and State B responds that while it is trying, unfortunately it appears that it is unable to effectively prevent the actions of the NSA (could be any reason, but let’s assume limited military capability). Importantly though, State B does not consent to State A firing into State B’s territory. In such circumstances, surely international law permits State A to use necessary and proportional force to neutralise the threat (pending UNSC action).
I’m not sure the “unable/unwilling” tenet of self defense is unavailing. For example, if non-state actors in Syria (pro or anti government militias) kill civilians in Turkey, and the Syrian government is unwilling or unable stop this, wouldn’t Turkey have the right to use force against belligerents in Syria to protect its citizens? If non-state narco criminals in Mexico conducted lethal cross border raids into Texas, California or Guatemala, and the Mexican government could not stop the raids, wouldn’t the governments of the US and Guatemala have a right to self defense that would include the use of force in Mexico against the narco criminals? Though self defense has to be proportional, and requires consideration of civilian casualties and the necessity of international use of force, there are instances where the inability or unwillingness of a host state to stop attacks by non-state actors triggers a right to self defense.
Martin,
I think Posner’s argument is that self-defence does not include “unwilling or unable” situations, thereby leaving consent as the only possible basis for the legality of the strikes under Article 2(4).
Ashley’s footnote is unsurprising. Kevin’s analysis of it is, at least to me. States do not generally assert a right to self-defense in certain circumstances out of “a sense of legal obligation.” Now, I suppose a state might assert an obligation, perhaps a “responsibility to protect” its citizens from unlawful and unjustified harm emanating from an extraterritorial location…but that hasn’t clearly happened, at least not yet. Ashley’s article, and others, have noted a historically common state practice that is consistent with first principles of international law — first principles that underlie the neutrality law of armed conflict and even the infamous Alien Tort Statute.
Charlie Martel: “there are instances where the inability or unwillingness of a host state to stop attacks by non-state actors triggers a right to self defense.” That’s the thing: inability and unwillingness do not “trigger” the right to SD. If that were the case, the conduct of inability/unwillingness would by itself constitute an armed attack by article 51 Charter, this surely cannot be the case. John Dehn, as far as I understood (this is me being humble to a far more senior ranked academic) the rule on SD is a permissive one, not an obligatory one. That would mean that the opinio juris element is the legal conviction that a state has the right to behave in a given way, but it is not, however, require to act accordingly (cf. Lotus and the permissive rule of Turkish criminal jurisdiction over the incident on the high seas). In that sense the traditional definition of opinio juris (“sense of legal obligation”) does not take into account permissive rules. Now, if consent is the only justification according to Posner, where is that consent then? While the law on state responsibility requires that consent is express (not implied) Pakistan has repeatedly protested against the… Read more »
RJ1983 is absolutely correct: in order to establish a customary rule that expands the concept of self-defense to include the “unwilling or unable” standard, there must be state practice (the use of force in such situations) and opinio juris (here, the claim that such uses are consistent with Art. 51 — not that states are somehow “obligated” to defend themselves). As I have explained before, and as Tom Ruys has documented, not only are there very few post-Charter uses of force against an NSA whose actions were not attributable to the attacked state, there is an even smaller number in which the attacking state claimed the legal right to do so under the Charter. And there are far more uses of force in such situations that have been widely condemned by the international community as being inconsistent with the Charter, such as Israel’s attack on the Osirak nuclear reactor in 1981. Dehn’s reference to “the neutrality law of armed conflict” is also problematic. He omits a critical word: “international.” Neutrality law applied only to international armed conflict and to belligerencies in which the rebel group was treated for purposes of the law as if it were a state. So the… Read more »
Kevin, why would the Osirak bombing be relevant to the question of inability/unwillingness in case of SD against non-state actors? The installation was IIRC government owned and operated as such. The casualties were soldiers (and one French civilian), not non-state actors. In my view this is not an example of “such situation” of use of force against NSA whose actions could not be attributed to the State, and the test of inability/unwillingness (even if it existed) would not be relevant here.
Well put, RJ1983. Regarding your comment to me: if states engage in a practice that appears to violate the Charter then I take it as given that they assert the “right” to engage in it by the mere act. To search for affirmative statements is often to hunt the snark. Heller has misrepresented my argument. I do not state that the international law of neutrality directly supports the unable and unwilling test. I said that first principles of international law do. These principles (I have called them neutrality principles, not neutrality law, elsewhere) underlie neutrality law. Prominent international legal theorists, such as Vattel, clearly indicate that states have an obligation to prevent its citizens and others resident in its territory from causing harm to foreigners. Some have argued that this basic principle led the U.S. to adopt the ATS, which is why I referenced it. Vattel is thoroughly discussed in important ATS scholarship; see Tom Lee’s article in the Columbia Law Review and particularly Anthony Bellia & Brad Clark’s article in Michigan Law Review. There is no doubt in my mind that natural law provided opinio juris for the “civilized nations” that developed international law as we know it today.… Read more »
I will respond as soon as I can figure out why the response I have drafted will not post after several attempts. The short version is that Heller misrepresents the gist my earlier comment, and that we should infer opinio juris from actions in this context.
RJ1983,
Oops, my bad! I meant to cite Israel’s 1985 raid of a PLO headquarters in Tunis. One could also mention Iran’s cross-border attacks throughout the 1980s on Kurdish fighters in Iraq, Rwanda’s attacks in the late 1990s on Hutu rebels in the DRC, or Colombia’s attacks on FARC in Ecuador. All of those attacks were condemned by large segments of the international community — including by the US in the case of Iran’s attacks.
As for Dehn’s comment, I would simply point out that his view collapses the distinction between opinio juris and state practice entirely, because it asserts that a state’s willingness to engage in a practice necessarily indicates that the state believes it is legally permitted to do so. I will leave it to others to decide whether that view of custom is sound.
I support Kevin’s arguments and conclusions entirely. It’s important to stand up to these attempts by the US to unilaterally reinterpret use of force law. With that being said, I wanted to mention that I considered the issue of the current status and normative future direction of use of force law, particularly in the context of and in light of the proliferation of WMD, in a 2010 article in the George Washington ILR. You can find it at this link: http://docs.law.gwu.edu/stdg/gwilr/PDFs/40-1/40-1-Joyner.pdf
I wouldn’t dare to hold myself out as an expert, but I do have two questions to KJH:
– Why single out post-Charter practice? I have to say that I think John Dehn’s argument for including pre-Charter practice is a plausible one. That certainly goes for recent and “Westphalian” practice.
– I agree that John Dehn unpermissibly collapses practice and customary law, but his reason for doing so sounds reasonable. Especially when it comes to the right to self-defence, which is at the same time an inherent principle of all law and a rule that is vulnerable to abuse. Observing that large numbers of states condemned, Israel, Iran, etc. seems much less pertinent than the fact that states from all over the globe, democracies and dictatorships alike, have attempted to rely on this kind of theory of self-defence.
The important premise of my argument above is that opinio juris should be inferred in this context — meaning — when states use force. The reason? States generally understand the moral and legal implications of using force, especially given the 2(4) prohibition of the Charter. Now one can disagree with that argument or not. And I am not as wedded to it as I was before Stuxnet. But Under Heller’s preferred approach (in this context, perhaps not others), there would never be adequate opinio juris unless a state laid out its legal arguments to support each and every use of force…a highly unlikely prospect.
Opinio juris can certainly be inferred, but you still need to have enough data points of state practice – which i dont think there are on this matter – AND those data points still also have to manifest in a supermajority of cases, even if not in explicit statements, some other compelling evidence of opinio juris in order for this inference to take place. I am not persuaded that this level of inferred opinio juris has even been remotely approached on this matter.
And what do we think of the precedent of Turkey’s 2008 incursion in Iraq?
http://www.defense.gov/transcripts/transcript.aspx?transcriptid=4162
Turkey’s incursion came to my mind as well. But even closer to the point, what about numerous and ongoing incursions into Afghanistan by Pakistan?
Three points, and then I’ll leave it to others to continue the debate. First, Dehn’s view simply presumes that no state ever uses force in a way that it knows, but doesn’t care, is inconsistent with international law. I don’t find that compelling; perhaps others do. But it is a necessary assumption, because if such uses of force are possible, then state practice can contribute to custom only when accompanied by specific statements of opinio juris. Second, Dehn’s view presumes we can know precisely what opinio juris silently accompanies the use of force against an NSA. That, too, is problematic. How do we know that a particular use of force implies that the state in question accepts the “unable and unwilling” test? Perhaps it believes that the attacked state is controlling the NSA, a situation in which the ICJ has concluded that the use of force is permissible. Perhaps it believes that force is permissible when a state aids an NSA but doesn’t control it — a weaker version of the ICJ test endorsed by scholars such as Tams. Dehn simply ignores those possibilities and assumes that any use of force against an NSA in which control is not obvious… Read more »
Briefly, and in order: (1) Heller glosses over my reference to Stuxnet, which I included to show that I recognize there are occassions when states (allegedly) use force in questionable ways. Of course, it would be moronic to suggest that one should infer anything from uses of force in which states do not openly engage, or at least indirectly or vaguely acknowledge (see “official” comments of US officials about drone attacks). If anything, an opposite inference might properly be drawn about the legality of uses of force in which states covertly engage and never acknowledge. (2) Heller utterly (and consistenly, going back to our 2010 PENNumbra debate and other discussions here at OJ) ignores the natural-law based first principles of international law that I and others have repeatedly raised in this and other contexts. The ICJ decisions he prefers are, by the Charter, applicable only to the states involved in a particular ICJ proceeding. In spite of this, Heller and others curiously refer to ICJ statements of international law as canonical and complete. He is right that I do not address ICJ opinions that do not directly address the points I raise, and I think this is properly so. Attribution… Read more »
[…] I see that John Dehn raises this point in comments to a post at Opinio Juris by Kevin Jon Heller; the post and comment thread are worth reading. On reflection reading that thread, I think this is the best way to understand the opinio juris […]
I don’t want to interrupt the flow of the symposium, so I won’t put this into a post now at OJ, but over at Lawfare, I have some reaction to Eric’s article. http://www.lawfareblog.com/2012/10/stop-presses-even-eric-posner-says-drone-strikes-in-pakistan-are-illegal/
It’s a bit odd, isn’t it, to be almost two dozen comments into this debate without anyone mentioning the Caroline incident? As I recall the facts there, the UK and the US agreed that the UK attack would have been a legitimate act of self-defense if the U.S. were unable or unwilling to stop the rebels from attacking — their only disagreement was about whether the U.S. was in fact unable or unwilling. (Eric’s bald assertion that the doctrine of self-defense *only* deals with the threat of attacks from a nation state (it “does not apply” here “because no one thinks that Pakistan plans to invade the United States”), simply ignores the Caroline incident, even though it’s widely understood to be a seminal case.)
Of course, this question would be relevant only if there is no Pakistani consent, a question about which neither Eric nor any of us has sufficient information to assess.
I actually work with Eric Posner and have written a piece on the “unwilling or unable” doctrine, which can be found here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2152648 On the points of discussion: P.S: What about ICJ opinions in the DRC and Wall Opinion cases, where the ICJ seemed to suggest that armed attacks cannot be carried out by non state actors? Further, concerning the “unwilling or unable doctrine” John C. Dehn: You state that “prominent international legal theorists, such as Vattel, clearly indicate that states have an obligation to prevent its citizens and others resident in its territory from causing harm to foreigners.” True, but any analysis of state responsibility principles also demonstrates that states that are UNABLE to do so because of lack of means, resources, remoteness of territory, knowledge of the impending attack etc. are not liable under principles of state responsibility. A requirement to exercise due diligence to prevent harm to other states emanating from ones territory is **not** an absolute burden. See e.g. Richard Lillich &John M. Paxman, State Responsibilityfor Injuries to Aliens Occasioned by TerroristActivities, 26 Am. U. L. Rev. 217, 279 (1976–1977). There are also other cases, for e.g. the Iran Hostages case where it was relevant that Iran actually had knowledge and the means to prevent the hostilities… Read more »
Dawood: That’s changing the subject. I didn’t say that the Caroline incident and any hypothetical strikes in Pakistan are analogous in every respect, and I didn’t write at all on the question of imminence, whether it’s always required, what constitutes imminence, what the Caroline incident has to say about imminence, etc.
What I did write is that the Caroline incident demonstrates that nations have long been of the view — contrary to Eric’s column — that legitimate actions in self-defense (i) can be taken against nonstate actors who threaten to attack; and (ii) can be consistent with sovereignty norms (as later reflected in the Charter) when the state hosting the nonstate actor is unwilling or unable to adequately address the threat.
One can certainly argue about the contours of the unable/unwilling doctrine; about its modern acceptance; about whether and how the Charter might have affected or rejected or incorporated it; etc.
But I don’t think it’s fair to argue, as Eric does, that this is an idea that the U.S. just hatched yesterday–its origins go back at least to the Caroline incident, if not farther.
Dawood,
I do not share your admiration for the ICJ’s opinions in the cases you mention. They are merely important data points to be considered. Thet are not the law.
Others have essentially asserted an “unable or unwilling” test to exercise extraterritorial law enforcement powers against terrorists. On this theory, the infringement of a state’s sovereignty is apparently a justified coutermeasure in response to a state’s failure to prevent its territory from being used to harm other nations.
At any rate, does such a situation come within Art. 9 of the Draft Articles on State Responsibility for Internationally Wrongful Acts? Are terrorist groups seeking in part to defend the territory of Muslims exercising elements of government authority of their own initiative, thereby calling for the exercise of government authority to control or suppress the behavior? For an analysis along this line of thought, see Major Michael D. Banks, Addressing State (Ir-) Responsibility: The Use of Military Force as Self-Defense in Internatinoal Counter-Terrorism Operations, 200 Mil. L. Rev. 54 (2009).
Marty: I did not specifically mention Caroline because that merely feeds into Heller’s disdain for reliance on U.S./U.K. practice in this area. The articles I cited survey both U.S. and non-U.S. practice.
Dawood:
(1) ICJ decisions and advisory opinions are very important data points to be considered, but are not law.
(2) For an article pursuasively addressing state responsibility for terrorism, see Major Michael D. Banks’ article at 200 Mil. L. Rev. (2009). I think his analysis in this context is consistent with Article 9 of the Draft Articles on the Responsibility of States for Internationally Wrongful Acts. See also an AJIL article referenced in Ken Anderson’s post at Lawfare (trackback link below). Note that “responsibility” in this context does not necessarily entail pecuniary liability.
(3) Others less favorably disposed to drone attacks have asserted that extraterritorial law enforcement measures are a justified countermeasure in these circumstances. One does not get to countermeasures without state responsibility for an internationally wrongful act. If you agree that states have an obligation to control their territory, then what is your take on the net effect of a failure in this area.
Marty: I did not mention Caroline specifically because it merely feeds Heller’s disdain for U.S./U.K. practice. For that reason, I cited articles that discuss both U.S. and non-U.S. practice.
Dawood, my response appears to be lost in the ether.
Dawood Ahmed: If I understand the argument correctly, Pakistan (the state) can claim no culpability for attacks made from its territory if it “lacks the means” to stop them, while simultaneously asserting sovereign immunity over the (same) area from which those attacks are launched? That just seems to violate the reasonability test (probably because I am layperson).
Look at the area in question, the FATA. National Pakistani law does not even apply and the government of Pakistan openly admits to only indirect control over the occupants. The boundary is in dispute. Pakistan acknowledges this by its own actions…its forces have moved to the west and taken over parts of strategic high ground (west of the Durand line) that technically belong to Afghanistan.
Marty – I agree with you that a handful of nations have, in the past, (sometimes facing many ex post objections) resorted to some kind of military response when another state has allegedly been “unwilling or unable”; and I also accept that the list of states attempting to do so is growing: e.g. Kenya, Pakistan, India claimed a right to undertake surgical strikes against Pakistan but of course, fears that Pakistan is well able to retaliate so has not yet (but there are also a number of instances of when states e.g. Pakistan, India, Afghanistan, could have asserted a “unwilling or unable” doctrine but did not). John: We could start a debate (on facts) about whether Article 9 would even apply to the situation in Pakistan, but I do not think whoever is being targeted through drone strikes there have disciplaced the national government there. In fact, the jirga and tribal system (much of which is anti-militant and have formed anti-militant militias) remains very strong. Yes, its not a central government in the Western sense but such alternative forms of government have had reasonable success in maintaing law and order in that part of the world for centuries. However, I do not know for how much longer… Read more »
Putting Pakistan aside, as far as I recall, unwilling/unable is not an ‘additional’ exception to the prohibition on the use of force, in that it falls within the scope of self-defense (i.e. UNC Art. 51). If I recall, Schmitt argues for it in one of his articles as: unable or unwilling means that a ‘sanctuary/host’ State has a “duty to ensure its territory is not used to the detriment of others”; if the ‘sanctuary’ State can & will act effectively to foil future attacks, defensive actions against hostile parties (e.g. non-state actors and ‘terrorists’) present in that State’s territory are impermissible absent sanctuary State consent.The case for unable/unwilling includes reference to …• Lotus Case (France v. Turkey, 1927): “It is well settled that a State is bound to use due diligence to prevent the commission within its dominion of criminal acts against another nation or its people.”• Corfu Channel Case (UK v. Albania, 1949): underlined both the duty to police one’s ownterritory & the **conditional nature** of the right of territorial integrity.• UNSCR 1373: requires states to “take the necessary steps to prevent the commission of terrorist acts, including by provision of early warning to other States by exchange of… Read more »