Home Armed Conflict The Bin Laden Killing: Clarifying the Normative Framework(s) Governing the ‘War on Terror’?

The Bin Laden Killing: Clarifying the Normative Framework(s) Governing the ‘War on Terror’?

Published on October 12, 2011        Author: 

Alon Margalit is Research Associate, Hotung Programme for Law, Human Rights and Peace Building in the Middle East, School of Oriental and African Studies (SOAS), University of London. The author wishes to thank the editors of EJIL:Talk! for their helpful comments on an earlier draft.

It has been almost six months since Osama Bin Laden was killed in Pakistan by a US commando team.  It is now worth reviewing some of the legal questions arising from the incident as the heat of the moment has passed.  The May 2011 killing of Bin Laden marked an operational apex in the US ‘War on Terror’ and was favourably received by the overwhelming majority of States.  Shortly after the raid on a residential compound in Abbottabad was concluded, and before its exact details were disclosed, a statement by the President of the Security Council welcomed “the news on 1 May 2011 that Osama Bin Laden will never again be able to perpetrate such acts of terrorism” and urged all States to intensify their fight against terrorism in compliance with international law.  UN Secretary-General Ban Ki-moon declared that “justice has been done to such a mastermind of international terrorism”.

Similar statements were made by the EU which described the American operation as “a major achievement”.  Afghan President Karzai said Bin Laden “had paid for his actions”, and Saudi Arabia, the national State of Bin Laden, expressed the hope that his killing “would be a step toward supporting international efforts aimed at fighting terrorism“.  In Pakistan, where the operation took place presumably without its consent, President Zardari chose to stress the “satisfaction that the source of the greatest evil of the new millennium has been silenced, and his victims given justice.”

If the question of where this operation stood in terms of international law were to be answered according to States’ responses, the killing of Bin Laden apparently did not raise any legal concerns.  States hailed the American operation, did not question its legality, and thus signalled that they saw no violation of international law.  Within this almost universal favourable discourse, two independent experts of the UN Human Rights Council, the Special Rapporteurs on summary executions and on human rights and counter-terrorism, issued an exceptional statement.  They urged the US to disclose the facts supporting the use of deadly force against Bin Laden in order “to allow an assessment in terms of international human rights law standards”.  They emphasised that “the norm should be that terrorists be dealt with as criminals, through legal processes of arrest, trial and judicially decided punishment”.

This statement reflected – contrary to what seemed to be the consensus shared by States – the ‘legal buzz’ among international lawyers, triggered by the American operation and concerned with its legality: was the US allowed to plan and execute a shoot-to-kill operation, or were its troops obliged to try and capture Bin Laden and give him an opportunity to surrender before turning to lethal force?  A significant discussion on this question emerged immediately after the incident, debating the applicable law and whether the operation had adhered to the required standards.  Different, at times opposite, views were expressed including on EJIL:Talk!, here and here.

Setting aside jus ad bellum questions, human rights law and the law of armed conflict (LOAC) are the two normative frameworks to consider in this context.  Human rights law does not bar State-agents from using deadly force against individuals, but this can never serve as a punishment without a trial; it must only be employed as a preventive measure of last resort.  According to this paradigm, US soldiers initially had to make an attempt to capture Bin Laden and allow him to surrender.  Only if an imminent danger to the life or physical integrity of the attacking team or other individuals arose from Bin Laden during the raid, could they be permitted to shoot him in self-defence.  As Bin Laden was believed to be highly dangerous, the use of deadly force was also legitimate if he had resisted an arrest or tried to escape.  Nonetheless, even then, force should have been proportionate: if injuring Bin Laden, rather than killing him, was sufficient to remove the danger he posed, force should have been used up to this level and no more.

Conversely, if LOAC applies, Bin Laden was a permissible target simply because he was an enemy fighter, and thus the attacking team was entitled to kill him even if he did not pose an imminent threat.  Active members of an organised armed group can be targeted at any time as they lose immunity from direct attack which is reserved only to innocent civilians, those who do not directly participate in the fighting.  Further, under LOAC, while it is strictly prohibited to attack an enemy who has become hors de combat by laying down his arms, there is no obligation to offer the enemy the chance to surrender before employing lethal force.  An intention to surrender must be clear and unequivocal.  US practice emphasises, correctly, that an offer to surrender must not be refused when communicated, but in certain circumstances of combat there is a risk that a soldier may not understand his opponent’s actions as an attempt to surrender “in the heat and confusion of battle, or may find it difficult (if not impossible) to halt an onrushing assault to accept a soldier’s last-minute effort at surrender”.

Hence what may be viewed as an illegal extrajudicial killing under human rights law may constitute a lawful targeting of an enemy belligerent under LOAC.  Nonetheless, the distinction between these two branches of international law is not rigid.  It is largely accepted today that both may be considered in assessing the legality of a military operation, however the exact relationship between the two legal regimes is not sufficiently clear.  For instance, the Israeli High Court of Justice has opined (in the 2006 Targeted Killings case) that while ‘terrorists’ are permissible targets under LOAC, they should not be targeted if they can be arrested, interrogated, and tried; if such less harmful means are available, the use of deadly force is unlawful.  This view is shared by the ICRC, and yet it is controversial as this is not required by LOAC.  Further, both the Israeli High Court and the ICRC admit that these additional requirements – based on human rights law – are not always practicable, especially when the attacking troops do not possess effective control in the area where enemy militants are present, or when an arrest attempt may increase the risk to the participating forces or to civilians.  It is important to note that this approach – seeking to restrict the use of lethal force when LOAC applies – is not identical with the human rights law standard; it permits a shoot-to-kill operation when capture is impracticable, despite the fact that the targeted object does not pose an imminent threat to life or physical safety.

This ambiguity as to the legal standards that govern the use of force in counter-terrorism operations also characterizes the US justification for the Bin Laden killing.  Its official position, pronounced by the Legal Adviser of the Department of State, maintains a clear LOAC paradigm as “the United States is in an armed conflict with al-Qaeda, as well as the Taliban and associated forces, in response to the horrific 9/11 attacks… a State that is engaged in an armed conflict or in legitimate self-defense is not required to provide targets with legal process before the State may use lethal force”.  More specifically to the killing of Bin Laden, “he was the leader of an enemy force and a legitimate target in our armed conflict with al-Qaeda… [C]onsistent with the laws of armed conflict and US military doctrine, the US forces were prepared to capture Bin Laden if he had surrendered in a way that they could safely accept”.

Nonetheless, other explanations provided by US officials contain elements from human rights law, meaning that capture was considered but due to the threat posed by Bin Laden during the raid, his death was the unavoidable outcome.  It was stated that “[t]he operation was planned so that the team was prepared and had the means to take Bin Laden into custody” and that the planning “was to prepare for all contingencies.  If we had the opportunity to take Bin Laden alive, if he didn’t present any threat, the individuals involved were able and prepared to do that”.  It was also stressed that the “sole focus of the operation was to kill or capture Osama Bin Laden.  But there were certainly capture contingencies, as there must be”.

An insightful analysis, posted on EJIL:Talk! shortly after the incident, addressed the US  position, denied the applicability of LOAC to the operation and doubted that it was lawful under human rights law.  As time passed and more details were uncovered, it became apparent that while the US team was engaged in a firefight in the first and second floors of the shelter house in Abbottabad (where three armed men were killed and a woman was injured), Bin Laden himself stayed in his bedroom on the third floor.  His wife rushed towards the first team member that entered the room and was shot in the leg as a result, but not killed.  Bin Laden was then shot in the chest and head.  Both Bin Laden and his wife were unarmed.  No US personnel were injured during the 40-minute operation.  Taking this into account, it is unclear whether Bin Laden, who was reported to be suffering from a kidney disease, posed an imminent threat to the life of the team or others; nor were details provided on resistance to arrest or on an attempt to escape that justified the use of deadly force in compliance with  human rights law.  There is also the issue of proportionality: whether any danger perceived by the attacking team could not have been eliminated by less harmful means, as one commentator has suggested, by using tasers or tear gas, or by directing the shots at Bin Laden’s lower body.  These speculations should be treated with caution as the factual circumstances remain uncertain.  For instance, according to The Associated Press, two guns were found on a shelf inside Bin Laden’s bedroom.  It was also reported that the team feared that individuals in the house were wearing suicide vests or that the house was booby trapped.

Assuming the report that no attempt to surrender was communicated to American soldiers is accurate, it is more likely that the operation was lawful under LOAC rather than under human rights law.  Moreover, there are some indications that the raid was planned from the outset to be a shoot-to-kill operation in accordance with LOAC targeting rules.  Bin Laden was immediately shot in the chest and head while the other person in the same room was only shot in the leg.  In addition, a special operations officer, described as “deeply familiar with the Bin Laden raid”, told The New Yorker in August 2011 that “[t]here was never any question of detaining or capturing him – it wasn’t a split–second decision. No one wanted detainees”.

The applicability of LOAC to the American operation can be challenged.  The most compelling argument is that al-Qaeda’s attacks on the US do not reach the necessary degree of intensity for an armed conflict to exist.  These attacks are claimed to be sporadic, meaning there are large time gaps and geographical distance between different attacks.  Some attacks are committed by individuals that may sympathize with al-Qaeda and be inspired by it, but their actions cannot be attributed to the latter in the absence of an organizational or operational link.  Accordingly, these discrete attacks cannot be aggregated when calculating intensity.  Even when an armed conflict does exist in a defined area, an isolated attack in another location cannot bring LOAC to apply there as well.  This approach also demands an accurate identification of the parties to each and every conflict involving the US; for instance, an armed conflict in Afghanistan/Pakistan with the Taliban does not necessarily cover al-Qaeda militants as their confrontation with the US should be assessed separately and might not reach the intensity threshold.

The counter-argument is that non-State actors such al-Qaeda present threats and capabilities that do not fall short of those dangers traditionally presented only by States, and their attacks may be as violent and destructive as those carried out during an inter-State armed conflict.  For instance, Hezbollah effectively controls the southern part of Lebanon – thus challenging its government’s authority – and senior US officials claim that it “has far more rockets and missiles than most governments in the world”.  Since al-Qaeda does not limit its attacks to a specific territory, the intensity of these attacks should be considered as a whole, and isolated attacks should be considered by their cumulative effect as long as they are part of an on-going violent campaign.  Given that the US is involved in an armed conflict with the Taliban/al-Qaeda in Afghanistan and in some parts of Pakistan, the application of LOAC can be extended to neighbouring Abbottabad because militants were pursuing their hostile activities from there; indeed a Pentagon official described the compound as “an active command-and-control center”.

Moreover, the US conflict with al-Qaeda may be extended to the latter’s associated forces as they belong to the same side of the conflict, directing their attacks against US targets at times in cooperation with al-Qaeda, or by fighters who affiliate themselves with both al-Qaeda and another armed group such as the Taliban.  The intensity test – required for an armed conflict to exist – should therefore not focus on the territorial dimension or on a formalistic differentiation between armed groups which is not followed in practice by those groups themselves, but rather on the material question of whether it is realistic to expect the US to handle specific threats with conventional policing powers.

The overwhelming majority of States answered this question by immediately supporting the US killing of Bin Laden.  As the details were then unknown – the exact dangers that the soldiers faced during the raid and whether Bin Laden resisted an arrest or tried to escape – it was impossible to make an informed judgement as to whether deadly force was used as a last resort; it was impossible to justify the killing according to human rights law.  After more information was revealed, States did not seem to change their initial stance even with the US announcement that the legal basis for the killing was LOAC.

When welcoming the American operation, States were already familiar with the violent situation between the US and al-Qaeda, and with the status of Bin Laden as its top commander.    Bearing in mind the ambiguity and controversy in this matter, it is arguable that by uniformly ‘certifying’ the Bin Laden killing, State practice accepted the American position and determined that an armed conflict exists between the US and al-Qaeda, and that members of the latter are permissible targets.  In this respect, the States’ reaction recalls the dynamics that followed the 9/11 al-Qaeda attacks when the Security Council and NATO States determined that an attack by a non-State actor triggered the right of States to use force in self-defence.

It is true that while the US articulated a legal argument, the favourable reception of States was not pronounced in legal terms.  However this uniform support might carry important legal  implications.  Indeed in their response to the Bin Laden killing, the independent experts of the Human Rights Council commented that “[a]ctions taken by States in combating terrorism, especially in high profile cases, set precedents for the way in which the right to life will be treated in future instances“.  Did the States implicitly say that the whole global ‘War on Terror’ is governed by LOAC?  Such a broad inference is possible since classifying violent situations as armed conflicts and bringing in LOAC, could be used for a State’s own advantage.  It would enable the targeting of those considered to be terrorists simply because they are enemy fighters without the need to provide a criminal process or to demonstrate an imminent danger to life.  As David Kretzmer comments in this context, the application of LOAC would free States of some of the constraints imposed on the use of deadly force during law-enforcement operations, therefore “it is in the very application of these laws and customs that the State might be interested”.

Nonetheless, a narrower reading of State practice in the Bin Laden case suggests that LOAC is applicable only to the conflict between the US and al-Qaeda and not necessarily to the US confrontations with other armed groups; neither does it necessarily apply to confrontations that involve other States and armed groups.  Similarly, it may only refer to the application of LOAC in a specific area, namely in Pakistan-Afghanistan and not in other places.

The final question concerns the States’ position towards additional obligations which are drawn from human rights law during an armed conflict.  Since States did not seek to clarify whether there was a prior attempt to capture Bin Laden or to allow him to surrender, at first glance it seems that targeting operations when LOAC applies are not subject to these requirements.  Another possibility is that the States did not reject these standards but rather were sufficiently convinced that in the unique circumstances of the raid, a capture operation was impracticable.  Given Bin Laden’s ‘vicious record’ and the fear from his flight, employing less harmful means than lethal force would have substantially increased the risk to the attacking team or to civilians (albeit not necessarily to the level of an imminent danger to life).  Further, the US explanation that notwithstanding that Bin Laden was a legitimate target in its armed conflict with al-Qaeda, “there were certainly capture contingencies, as there must be”, is indicative that the US sees itself bound by such a requirement.  Whether the implementation of this requirement was impractical or that a duty was breached, the US practice reinforces its binding status.

It has been six months since the incident and it is still difficult to make a conclusive finding on the legality of the Bin Laden killing due to an incomplete account of the facts and uncertainty surrounding the applicable law.  The uniform response of the international community indicates that the use of force in this case was governed by LOAC and lawful under this regime.  It does not, however, remove the doubts in relation to the application of LOAC and human rights law to future counter-terrorism operations carried out as part of the ‘War on Terror’.

Print Friendly

13 Responses

  1. Marty Lederman

    Thanks for that very helpful account, Alon. (FWIW, I posted an analysis based on the U.S. assumption of LOAC-application here, before several of the factual accounts were reported:

    One minor follow-up question: What if one were to conclude, contrary to the U.S. view, that this operation was not part of a NIAC between the U.S. and AQ, either because there is no such conflict at all, or because it doesn’t extend beyond Afghanistan . . . but that the operation was a legitimate act of self-defense for jus ad bellum purposes? Do the same LOAC apply as a matter of custom to such acts of self-defense, outside of armed conflict? U.S. military practice and Pentagon policy is to apply the LOAC in either situation (indeed, as to all uses of force); but is there any well-established IL on that question?

  2. Marko Milanovic Marko Milanovic

    Marty, I hope you wouldn’t mind me barging in with a comment (and obvs we”ll see what Alon thinks about this), but my understanding is that IL is very clear on this: not every use of force ipso facto constitutes an armed conflict, and if it doesn’t, as in your hypothetical scenario, then IHL/LOAC wouldn’t apply at all. The only pertinent body of law would be the jus ad bellum, and if you accept that human rights treaties apply extraterritorially, also IHRL. And if country X authorized the US to enter its territory and shoot a missile at a suspected terrorist who is not involved in any cognizable armed conflict, the jus ad bellum would also be out of the picture due to X’s consent, and the only relevant law would be IHRL.

    In short, that the operation is done in self-defense has no direct bearing on whether IHL applies, nor does it of itself render IHRL inapplicable or preclude any wrongfulness on that grounds. For a more detailed argument, see:

  3. Marty Lederman

    Thanks, Marko. I certainly have seen you and others take the view that IHL doesn’t apply to acts of self-defense outside armed confict (by which I assume you mean the inapplicability of both the combatant’s immunity and all of the jus in bello restrictions of IHL); but that seems deeply counterintuitive to me — in that I can’t see why nations would not insist upon both the privilege and the need to impose the basic IHL restrictions.

    Is there any actual authority, or state practice, to support your view?

  4. I agree that the U.S. cannot be at war or in an armed conflict with al Qaeda as such under traditional customary criteria with respect to an insuregency, Geneva Protocol II, art. 1(1), or even the non-customary preference in the Tadic decision within the ICTY. However, there is a real war in Afghanistan and parts of Pakistan — the latter because the de facto theatre of war has migrated into parts of Pakistan and was, at the time of his killing, right over the head of bin Laden, since he was still a DPH taking a direct part in the hostilities. Please see
    As such, bin Laden was targetable under the laws of war and under article 51 of the U.N. Charter — the latter since he was a DPAA taking a direct part in ongoing armed attacks against U.S. nationals in Afghanistan if not elsewhere as well. See the above article on Self-Defense Targetings of Non-State Actors…., at ssrn.
    Since the U.S. cannot be at war with al Qaeda as such, the targeting of al-Awlaki in Yemen (outside the theatre of the real Afghan war and where there is no known evidence of his being a DPH in that war) was permissible only under the law of self-defense, the self-defense paradigm and not the law of war paradigm. Please see the ssrn article above. But he was a DPAA at the time IF the U.S. Exec. is correct that he became part of an ops team and was directly participating in attempted armed attacks (Christmas underwear bombing, Fed-Ex bombing, etc.).
    Shortly, I will post the full short article of mine in the Denver J. Int’l L. on ssrn since it has a short addendum on the killing of bin Laden. An earluer draft, without the addendum, is at
    Marty: my main article on Self-Defense Targetints also addresses relevant principles under the law of self-defense, since they are basically the same re: distinction, reasonably necessity, and proportionality, and an implied immunity for those engaged in legitimate self-defense in view of practice and apparent expectations about such practice.
    Marko: and I address why human rights law applies globally and in time of war, but why persons who actually benefit must either be within the territorial jurisdiciton of the targeting state or under its “effective control” — which is not the case re:al-Awlaki.

  5. Marko Milanovic Marko Milanovic


    The authority for my argument is simply in the Common Articles 2 and 3 of the Geneva Conventions that define the thresholds of applicability of modern IHL. If these thresholds are not crossed, the treaties – as well as the custom the grew out from them – do not apply by their very design. No state argues differently; the laws of armed conflict apply only in armed conflict, other than the peace time provisions on dissemination etc. Even the Bush administration’s initial position was that there was an armed conflict with Al Qaeda which was an IAC, if some weird sort of IAC that did not fall within the Geneva definition. Some argue for a third, transnational category of armed conflict under customary law. But noone, to the extent that I am aware of, argues that LOAC will apply outside armed conflict – that’s the whole point of the thresholds of application. No major textbook, for example, has advanced the contrary view.

    Do bear in mind that the scenario you suggest is very, very rare. Self-defense will normally involve state against state action. It is generally regarded that any inter-state use of armed force will constitute an IAC, so IHL would indeed apply, or that the threshold of violence is minimal at best.

    If, on the other hand, non-state actor A operating from the territory of state B attacks state C, and C enters the territory of B to respond to A (which is the scenario you suggest), then we have two options: either B consents to the intervention or it doesn’t. If it doesn’t, there’ll be an IAC between B and C, while the jus ad bellum lawfulness of the intervention will depend on the strength of C’s self-defense claim. If, on the other hand, B consents, then C’s right to self-defense is perfectly irrelevant, as C doesn’t need to invoke it. And, as you know, generally speaking conflicts between states and non-state actors require a higher threshold, that of protracted armed violence.

    So, in short, I fail to see what is counterintuitive about this framework. If B does not consent, then you do have an IAC, and there will exist the combatant’s privilege and all the works. If B consents, then you don’t need the combatant’s privilege, nor can you have it, as you don’t have two states fighting each other. As for basic humanitarian restrictions, human rights law would continue to apply. There’s no regulatory gap here that I can see, nor are states somehow preventing from actually using armed violence so long as they have sufficient justification.

  6. There is a major flaw in this piece. Bin Laden’s relatives are prepared to sue the U.S. in an international court and would gladly challenge the U.S.’s legal argument in a court.

    Nothing unreasonable with that, since no decent legal system allows the suspects of murder to decide themselves that they are innocent.

    But there is no such international court where they can challenge this legal argument, meaning that whatever the U.S. government says, goes and the U.S. has the last word. This is a violation of one of the most fundamental principles of law, that nobody involved in a conflict should be at the same time the final arbiter of the conflict. No one can be judge in her own case.

    As long as the IL jurists are conditioned by their teachers to be blind to this major flaw, there will be no sound argument about such matters, but only fallacies that enable those with power to do whatever they like with impunity.

  7. Alon Margalit Alon Margalit

    Thank you for your comments.

    Marty – I agree with Marko’s response. LOAC does not apply outside of an armed conflict. Not every use of force (even lawful under the jus ad bellum) necessarily triggers an armed conflict.

    When force is used extra-territorially against a non-State actor with the consent of a foreign State, the operation is subject to IHRL as the State using force is bound by the international obligations of the authorizing State towards its nationals and/or since the individual object of the operation comes within its effective control.

    Depending on the surrounding circumstances, once a certain level of intensity is reached, a NIAC exists and LOAC will apply.

  8. Marko — and my article demonstrates why B and C would not be at war if C merely responds in self-defense against A in state B — and why when the U.S. used military force in a foreign state to target non-state actor attackers in the past the U.S. was not at war with that state — and why the U.S. and the U.K. (or Canada) were not at war because of the British attack on the vessel Caroline — no one thought that the U.S. and the U.K. were at war — there was simly disagreement whether the actual method or means used by the U.K. in response to ongoing non-state actor armed attacks fit within the principles of reasonable necessity and proportionality (applicalbe in case of self-defense) under the circumstances — and there was no claim of anticipatory or preemptive self-defense, only self-defenst against attacks that had already occurred.
    The U.S. had argued for a tough standard re: methods and means, since in those days the U.K. could have waited until the vessel Caroline went back into Canadian waters. One of the persons involved had later declared that no one would have denied the U.K.’s right to take out an artillery position on the U.S. shore that had started firing into Canadian territory. No one thought we were at war.
    Yes, human rights law applies globally and in all social contexts (esp. under U.N. Charter arts. 55(c) and 56 — “universal”), but who benefits from a particular right — only those on U.S. territory or where the U.S. has territorial-type jurisdiciton (e.g., in occupied territory) or those who are in the “effective control” of the U.S. outside such territory.
    Alon: we are not Europeans. The restrictions regarding lawful killing in the European Convention do not apply to U.S. targetings in Yemen or Pakistan. Bin Laden was not in the “effective control” of the U.S., but perhaps darn close — he did not surrender — and al Awlaki was not in the “effective control” of the U.S. Note also that the Israeli case involved special territorial circumstances and a mixture of the “law enforcement” paradigm with the law of war and self-defense paradigms — and the law enforcement paradigm was not applicable with respect to the killing of bin Laden or al-Awlaki. Self-defense and laws of war were applicable re: bin Laden and the expanded de facto war, only the self-defense paradigm was applicable with respect to al-Awlaki.
    Also, I doubt that states and others mentioned who thought that his passing was a good thing arguably agreed to such things as the U.S. can be in an armed conflict of any sort with al Qaeda as such. But, yes, the backgorund recognition after 9/11 has been that the U.S. could rightly use military force in self-defense against certain non-state actor attackers.

  9. p.s. Marko: if C engages in self-defense force against A in state B without B’s consent and B is not at war with C, it is still possible that there is an armed conflict between C and A if A has the status, as a non-state actor, of a “belligerent” or an “insurgent” (in a NIAC, which should be recognized as an armed conflict of an international character (internationalized by C’s actions) by states so that state C’s miltary will have combatant status and combatant immunity for lawful acts of war.
    Thus, if the U.S. uses military force in Mexico without Mexico’s consent to target merely local insurgents who fight with the Mexican army in Juarez, Mexico and elsewhere and who have engaged in armed attacks on U.S. nationals there and in Phoenix, Arizona and El Paso, Texas, the U.S. is engaging in permissible self-defense tagetings and is not at war with Mexico, but might be in an armed conflict with the insurgents.
    More generally, there have been many non-state actors in the international legal process other than the state for the last few hundred years, including states, nations, belligerents, corporations and companies (with rights and duties under international law). Please see
    Armed conflicts of an international character have involved those among states and nations, nations and nations, states and belligerents (as in the case of the U.S. Civil War — U.S. vs. C.S.A., when the C.S.A. met the traditional criteria for belligerents and all of the customary laws of war applied — and we adopted the 1863 Lieber Code in an effort to codify that customary law).

  10. Marty Lederman

    Thanks, everyone, for those helpful responses.

    Marko and Alon, we appear to agree, then, that IHL applies to virtually all uses of force between states. Taking a cue from the Tadic decision, you would simply conclude that all such uses of force, no matter how limited and discrete, constitute armed conflicts, even where one side does not respond in kind. The Tadic court doesn’t do very much to defend that categorical assumption: After all, it is not common to refer to the numerous discrete uses of force in self-defense over the past 200+ years — “one-off” responses from Greytown and the Caroline to the Osirik reactor strike and the U.S. strikes against Iraq and Libya, etc., etc. — as “armed conflicts” or wars. But I’m not sure the typology really matters, since what really turns on the label is whether IHL — its limitations and its immunities — applies. (And, I suppose, whether such a discrete use of force that does not result in sustained hostilities between states triggers the GCs.) And you appear to be of the view that it does, as a matter of custom, which was my assumption, too.

    Indeed, if I understand you, Marko, you argue that IHL also applies where State Actor A uses force against non-state actor B in Nation C, without State C’s consent, even if the intensity threshold for armed conflict between A and B is not met.

    If so, then the only real, practical question that we still need to resolve is whether IHL applies where State A uses force against non-State actor B *with* the consent of State C, and where the intensity threshold has not (yet) been met. You both argue that IHL doesn’t apply in such cases; but I don’t quite see why that should be so. To be sure, the GCs don’t apply in such cases, because there is no armed conflict. But common articles 2 and 3 do not say anything about whether customary IHL applies in such cases. And if it did apply in such cases before 1949, presumably it continues to do so.

    It appears that what motivates both of you is the idea that in such cases IHRL should apply. That’s an interesting and potentially important question. But of course the United States has traditionally denied that it (at least the ICCPR) applies in such cases and, in any event, presumably would resist the idea that the prohibition on “arbitrary” killing, even if it does apply, is not satisfied by compliance with IHL norms (proportionality, humanity, necessity, no perfidy or denial of quarter, etc.) in such a setting (as suggested in the Nuclear Weapons opinion).

    In light of that consideration, shouldn’t you be urging that IHL applies *at a minimum*? Not just for purposes of constraining the use of lethal force, but also as a better reading of the historical customary law? After all, wouldn’t that have been the better view of custom *before* the modern development of IHRL (wasn’t that the point of the Martens Clause?) — and, if so, is there any reason to think that post-WWII custom has now developed to *decrease* the obligations imposed on nations using force against non-state actors in self-defense? I wouldn’t think so.

    SO, by all means let’s continue to debate whether the U.S. and other nations should consider themselves bound by IHRL/ICCPR with respect to the legitimate use of force in self-defense against non-state armed groups, and about whether the norm against “arbitrary” killings does the material work in this setting that many appear to assume it does (I’m a bit skeptical, but I haven’t done the necessary research). But in the meantime, can’t we all agree that customary IHL establishes an irreducible baseline, or floor, that should constrain all uses of force, armed conflict or not?



  11. Marko Milanovic Marko Milanovic


    Many thanks for this, and a couple of points in response.

    First, while I would certainly be inclined to agree with you that some minimum standards should apply to all uses of force (how minimal is a different matter), states actually don’t. In particular, the CA3 threshold of internal armed conflict was specifically designed by states so that IHL would NOT apply to any use of force internally. This was the purpose, for example, of Art. 2(2) AP II, which explicitly provides that the Protocol ‘shall not apply to situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature, as not being armed conflicts.’

    Why? Because states (1) originally did not want to restrict the means at their disposal to suppress a nascent rebellion, and (2) did not want to legitimize the rebels by giving them protection and status under international law. This sovereignty-induced dynamic in central in understanding the concept of NIACs and how they evolved. In other words, as originally designed, the law of NIAC/CA3 simply did not apply if a state shot a couple of insurgents against a wall, or launched a missile or two. The only law that applied was the domestic law of that state, while from the international side IHRL only gradually filled that gap.

    Second, and in that regard, note that pre-WWII the threshold of application of the jus in bello was interstate war, and that the laws of war applied to internal conflicts only if belligerency was recognized. In other words, your reference to customary law pre-WWII is misplaced, as there was no such law regulating even widespread conflicts between states and non-state actors, let alone isolated uses of force. The pre-WWII position was that states could do whatever they wanted to their own people generally, and to rebels and insurgents specifically. To the extent that such activities crossed state borders, as e.g. in Caroline, the only concern of international law was whether the sovereignty of the territorial state was infringed or not; targeting rules or limitations were never in the picture.

    Third, it is certainly possible for the Tadic protracted armed violence threshold (which was btw clearly articulated in that decision but most certainly precedes it) to have been relaxed through developments in state practice etc. But where are these developments, other than in the immense uncertainties of the ‘war on terror’? Can you point them to me? As I see it, states certainly still do NOT want to apply IHL to all uses of force within their borders against individuals or non-state actors. Why should the extraterritorial context change anything?

    In other words, if you argue, say, that a US missile strike on Dr. Evil, a loathsome individual completely unrelated to Al-Qaeda, should be regulated by IHL if he were located in Canada and Canada consented to it, then you would have to argue that the same strike would be regulated by IHL even if he were located in Alaska. There is no reason of principle to distinguish the two. Even assuming that you actually could substantively apply IHL to this situation (i.e. that it is somehow possible to distinguish between military and civilian targets in the absence of an organized enemy or protracted hostilities), where is the evidence that this is something that states would be willing to accept?

    Finally, assuming that I am right, and that IHL would not apply to such a use of force; that I am also wrong, and that IHRL does not apply extraterritorially; and further that the territorial state consents, thereby obviating any jus ad bellum inquiry – from the standpoint of the state using force extraterritorially, this situation would be just perfect: there would be no international law at all regulating its use of force against that particular individual. It could do whatever it wanted, as in the good old days. This is a gap that both you and I are concerned about, I think – except that you would fill it with IHL, and I with IHRL. The reasons for my preference are simple: it is artificial to label as an ‘armed conflict’ an isolated use of violence by a state against a specific individual or group of individuals, while the sole reason why we fear this gap is that we feel that individuals have some intrinsic worth and should not be deprived of their life arbitrarily, which is precisely what human rights are all about.

  12. Incredible! nearly everyone knows that the laws of war apply to a short war, even the 30 minute war re: the U.S. and Libya in 1986! Incredible! Nearly everyone knows that the laws of war do NOT apply merely because self-defense targetings of non-state actors take place on another territory with consent of that state.