Analysing the US Invocation of Self-Defence Re: Abu Khattallah

Analysing the US Invocation of Self-Defence Re: Abu Khattallah

Most of the discussion about Abu Khattallah’s capture in Libya has focused on the operation’s basis — or lack thereof — in domestic US law. Less attention has been paid to whether international law permitted the US to use force on Libyan soil. As Marty Lederman recently noted at Just Security, Abu Khattallah’s capture can potentially be justified on two different grounds: (1) Libya consented to the capture operation; or (2) the capture operation represented a legitimate act of self-defence under the UN Charter. The first justification does not appear open to the US; the available evidence indicates that the operation was conducted without Libya’s consent. So it’s not surprising that the US has claimed — in a letter submitted to the UN by Samantha Power on June 17 — that Article 51 permitted the operation:

The investigation also determined that [Abu Khattallah] continued to plan further armed attacks against U.S. persons. The measures we have taken to capture Abu Khattallah in Libya were therefore necessary to prevent such armed attacks, and were taken in accordance with the United States’ inherent right of self-defense. We are therefore reporting these measures to the Security Council in accordance with Article 51 of the United Nations Charter.

Power’s letter obscures far more than it reveals. In fact, the US’s invocation of self-defence raises four very difficult questions:

  • Can a non-state actor launch an “armed attack” that triggers the right of self-defence?
  • If so, must that armed attack be attributable in some fashion to the state whose territory is the object of “self-defensive” force?
  • Do all uses of armed force qualify as an “armed attack” for purposes of Article 51?
  • Does the right of self-defence permit force to be used anticipatorily?

In this post, I want to put aside the first two questions. I have no doubt that a non-state actor can launch an armed attack within the meaning of Article 51, and my views on the “unwilling or unable” test are well-known. It’s worth spending some time, though, on the third and fourth questions.

The third question is interesting because it’s not clear that all uses of force qualify as “armed attacks” for purposes of Article 51. The UN Charter itself distinguishes between the “use of force” (Art. 2(4)) and “armed attack” (Art. 51), and the ICJ has suggested in both Nicaragua and Oil Platforms that at least some uses of force may be so de minimis that they do not entitle the victim state to use force in self-defence. (As opposed to taking other countermeasures.) On the other hand, customary international law seems to indicate that the threshold of force for an armed attack is extremely low. Here is Tom Ruys’ conclusion in his magisterial book “Armed Attack” and Article 51 of the UN Charter (p. 155):

In the end, customary practice suggests that, subject to the necessity and proportionality criteria, even small-scale bombings, artillery, naval or aerial attacks qualify as ‘armed attacks’ activating Article 51 UN Charter, as long as they result in, or are capable of resulting in destruction of property or loss of lives. By contrast, the firing of a single missile into some uninhabited wasteland as a mere display of force, in contravention of Article 2(4) UN Charter, would arguably not reach the gravity threshold.

In sum, the following general conclusions can be made: (1) the travaux of the Definition of Aggression suggest that a minimal gravity is indeed required and seem to rule out the aforementioned Option 3; (2) ‘concrete’ customary evidence nonetheless makes clear that the gravity threshold should not be set too high and that even small-scale attacks involving the use of (possibly) lethal force may trigger Article 51.

If Ruys is right — and he has examined state practice and opinio juris far more carefully than any other scholar writing on the use of force — the attack on the US embassy in Benghazi almost certainly was, in fact, an “armed attack” for purposes of Art. 51.

What, then, about the fourth question? Here is where the US claim of self-defence regarding the Abu Khattallah operation becomes problematic. The US clearly cannot use the original Benghazi armed attack to justify the operation — although a state’s response to an armed attack may not have to be immediate, the prohibition on armed force in Art. 2(4) of the UN Charter would be meaningless if a state could “pocket” an armed attack and respond to it with armed force much later — nearly two years later, in the case of Benghazi. Indeed, Power seems to acknowledge as much when she emphasises that Abu Khattallah was planning further armed attacks. Does that planning mean the capture operation was a legitimate act of self-defence by the US?

Answering that question, of course, requires us to address the temporal limits of self-defence under Art. 51. Three basic positions on that issue are possible:

  • Self-defence permits the use of force only in response to an armed attack; force cannot be used pre-emptively or preventively (“responsive self-defence”)
  • Self-defence permits the use of force to pre-empt an imminent armed attack but not to prevent a temporally more remote armed attack (“pre-emptive self-defence”)
  • Self-defence permits the use of force to prevent even a temporally remote armed attack (“preventive self-defence”)

Unfortunately, because of the US’s typical lack of transparency concerning its use of force, Power’s letter says nothing about the time-frame of the armed attacks Abu Khattallah was supposedly planning. (Nor does it provide any evidence of that planning, but that’s another question.) The time-frame doesn’t matter, however, if responsive self-defence is the correct position — as noted, the capture operation cannot be justified as a response to the original Benghazi attack.

Most readers — at least those in the West — will no doubt be inclined to reject responsive self-defence as too narrow, even though it is the only position consistent with the text of Article 51, which permits self-defence “if an armed attack occurs.” Surely customary international law does not require a state to wait until an armed attack has already taken place to defend itself, no matter what the UN Charter says.

This issue is much more difficult issue than it may appear. Those interested should read the relevant section of Ruys’ book; I’ll just quote his bottom line (pp. 341-42):

In light of the available evidence, it can be concluded that there has indeed been a shift in States’ opinio iuris insofar as support for pre-emptive self-defence, fairly rare and muted prior to 2001, has become more widespread and explicit in recent years. At the same time, it seems a bridge too far to claim that there exists today widespread acceptance of the legality of self-defence against so-called “imminent” threats. Such assertion tends to forego the opposition of a considerable group of mainly Latin-American, north-African and Asian States. In the present author’s view, it would therefore be more appropriate to argue that the crack in opinio iuris among States has widened, without, however, identifying one approach or the other as the majority view. The implication is that, taking account of the Charter “baseline” and the absence of a concrete precedent in State practice which convincingly demonstrates the international community’s support for some form of anticipatory self-defence, it is impossible to identify de lege lata a general right of pre- emptive – and a fortiori preventive – self-defence.

Ruys’ reference to the UN Charter’s “baseline” is important, because Art. 51’s adoption of responsive self-defence indicates that states who support a more relaxed concept of self-defence, such as the US, have the obligation to find sufficient state practice and opinio juris to establish a broader rule. And such state practice and opinio juris is simply lacking — unless, as is too often the case with custom, we simply ignore the views of the Global South.

Even if responsive self-defence is too narrow, however, that does not mean the Abu Khattallah operation was a legitimate act of self-defence. If the US had evidence that Abu Khattallah was about to launch another armed attack, it is reasonable to assume Powers would have said so in her letter. That she failed to do so thus seems to indicate — though is clearly not dispositive — that the US did not believe another armed attack was imminent when it launched the capture operation. Power’s letter may well indicate, therefore, that the US is promoting the broadest understanding of self-defence possible — preventive self-defence instead of pre-emptive self-defence. If so, as Ruys notes (pp. 336-38), the US is on shaky ground indeed:

[T]here can be no doubt that even among States adhering to the “counter-restrictionist” view, support for self-defence against non-imminent threats is virtually non-existent. Apart from the fact that the sponsors of Operation “Iraqi Freedom” avoided this justification, it may be observed that many States, such as Germany, Japan, Switzerland, Uganda, Singapore or Liechtenstein, which professed support for anticipatory self-defence after 2002, nonetheless placed great weight on the imminence requirement. Germany, for instance, expressly denounced an erosion of the Charter framework and State practice via the notion of “preventive self-defence.” Likewise, the French politique de defense unequivocally “rejects… the notion of preventive self-defence.”

What is more, even the “traditional” adherents of the counter-restrictionist interpretation of Article 51 generally appear to uphold the imminence requirement. Despite bold statements by its Prime Minister on the need to adapt the UN Charter, Australia’s response to “In Larger Freedom” was rather cautious: it simply “[supported] reaffirmation by the Secretary-General that Article 51 of the Charter adequately covers the inherent right to self-defence against actual and imminent attack.” Israel called for an explicit recognition in the World Summit Outcome that States may use force in self-defence “in the event of both actual and imminent attacks.” As far as the British position is concerned, Attorney- General Lord Goldsmith in 2004 declared before the House of Lords that: “It is… the Government’s view that international law permits the use of force in self-defence against an imminent attack but does not authorize the use of force to mount a pre-emptive strike against a threat that is more remote.”…

[W]e may therefore conclude that the trend in State practice has been broadly similar to that in legal doctrine: support for anticipatory self-defence has increased, but has by and large restricted this concept to imminent threats.

Again, in the absence of additional information, we cannot categorically reject the US’s insistence that the Abu Khattallah operation was a legitimate act of self-defence. But there is considerable reason to be skeptical. Indeed, the US’s lack of transparency concerning its understanding of Art. 51 of the UN Charter may well indicate it has adopted a position that even its closest allies formally disavow.

Print Friendly, PDF & Email
Africa, Courts & Tribunals, Featured, Foreign Relations Law, International Criminal Law, International Human Rights Law, National Security Law, North America, Organizations
Notify of
Benjamin Davis
Benjamin Davis

What if after the armed attack, the state starts a process of investigation before using the force to comply with necessity, etc. indiscriminate response being outlawed. Then once you find the right target you use the force.

Aaron F

And what if, hypothetically speaking, the use of force is limited to seizing an individual for the purpose of trying that individual for conspiring to engage in an attack? I have not analyzed state practice on this question, but I would imagine that states would be much more willing to accept this preventive use of force compared to, say, a full-scale invasion.

Ian Henderson

And further, as acts in self-defence cannot be merely for retribution, if X attacked Y but then lacked any capability or intent to launch further attacks, Y could not act against X in self-defence. But its seems logical that where X has attacked Y, and Y has evidence that X plans further attacks, it is not pre-emptive self-defence at all. The first attack has already occurred. Rather, Y is acting to prevent further attacks by X. Isn’t that the very essence of self-defence?

Itzchak Kornfeld
Itzchak Kornfeld

Kevin, as always you raise extremely insightful and thought provoking issues in your comments.‎ One area that has not been touched upon, however, is the issue of the U.S.’s invading Libya’s ‎sovereignty, without that country’s permission. That may be implied in your comments, ‎particularly the references to the UN Charter’s Art. 51. Nevertheless, it is worth noting that ‎newspaper accounts state unequivocally that U.S. Special Forces’ first kidnapped, Abu ‎Khattallah and then advised Libya of the undertaking. I suggest that such notice is in direct ‎violation of Article 2(1), which provides that “[t]he Organization is based on the principle of the ‎sovereign equality of all its Members.”‎ Similarly, Article 2 continues in sections 2, 3 and 4‎ ‎(2) All Members, in order to ensure to all of them the rights and benefits resulting from ‎membership, shall fulfill in good faith the obligations assumed by them in accordance with the ‎present Charter.‎ ‎(3) All Members shall settle their international disputes by peaceful means in such a manner that ‎international peace and security, and justice, are not endangered.‎ ‎(4) All Members shall refrain in their international relations from the threat or use of force against ‎the territorial integrity or political independence of… Read more »

Benjamin Davis
Benjamin Davis

Also, isn’t this another version of Eichmann in Argentina and the Israeli and Argentinian dispute? There it was Israeli agents, here it is military special forces. Assume it was illegal – does Libya acquiesce or contest like Argentina and have a state to state resolution. The UN filing of a notice is a cross the T’s and dot the i’s approach to support the self-defense argument.

Benjamin Davis
Benjamin Davis

Under US domestic law, the view would be it would be on Libya to protest to the US and the individual would have nothing to stand on – whether there was an extradition treaty or not. Public or private agents are treated like bounty hunters in the old West where the US court does not give a hoot how they got to the courthouse.

Alex B.
Alex B.

Thank you for this post. I agree that the attack on the US embassy in Benghazi is an “armed attack” for purposes of Art. 51. But I doubt that the self-defence can be invoked by the US in the present case for the following reasons. I believe that the Tom Ruys’ definition you refer to predominantly covers attacks performed by state actors (correct me if I am wrong – I am not able to access the book atm). While it is possible to apply the definition to the present case, it does not mean that an attack committed by a private actor from the territory of a third state can trigger the right to self-defence as easy as if this attack was performed by the state itself. It seems that the gravity threshold for an armed attack to trigger the right of self-defence can vary depending on the perpetrator. Single strike hardly can be enough to justify the violation of a state’s sovereignty even if one can pass the “unwilling or unable” test. The last one seems almost impossible to me as the “unwilling or unable” test is tailored to indicate the inability of a state to act against some… Read more »

Kevin Jon Heller


As I said in the post, I was bracketing the attribution issue, because I have discussed on numerous occasions why I don’t believe “unwilling or unable” reflects customary international law. So yes, the capture operation could still fail to qualify as self-defence even if the factors I discuss were satisfied.

Your point is also well taken that even if “unwilling or unable” was the standard, a single attack would probably not be enough to invoke it.

Kinga Tibori-Szabo
Kinga Tibori-Szabo

Good analysis, Kevin.
I would add to the self-defence options the “pinprick” (accumulation of events) theory, according to which a string of past attacks, not qualifying in themselves as armed attacks, but creating the conviction that future attacks would occur, could justify a response in self-defence.
In any case, I don’t think that any of these theories justify the recent use of force in Libya as self-defence. As I said in my book (Anticipatory Action in Self-Defence), the crux of self-defence is that it addresses a necessity, present and inevitable, to end the danger posed by an imminent, an unfolding or an already occurred armed attack or string of attacks. When the use of force exceeds this purpose, the measure ceases to be defensive and becomes a punitive action. What the US is really advocating through Power’s letter and other statements (especially in relation to their drone policy) is an enhanced Bush doctrine, i.e. an enhanced preventive self-defence approach, according to which it is not only the time and place of the future attack that are unknown, but its occurrence per se is not certain either.

Kevin Jon Heller


Needless to say, I couldn’t agree more.

Mary Ellen O'Connell


The Khattala detention cannot be justified by consent or Article 51 self-defense. But as it was undertaken following policing standards, it was arguably a lawful countermeasure as I explain in this comment for CNN:

Mary Ellen O’Connell
University of Notre Dame