Ken Anderson has an excellent, very interesting post regarding the US strategy of using drones for targeted killings of suspected terrorists in Pakistan and elsewhere (a topic we’ve addressed at the blog before). He argues that, as a matter of both law and policy, the current justification of the US government for its targeted killing practices is insufficient, because it relies far too much on IHL concepts like ‘combatant’ and IHL rules on targeting which are dependent on such concepts.
Such a justification is of course deeply problematic because IHL applies only in armed conflict, while the position taken by the Bush administration that the ‘war on terror’, or the US struggle again Al-Qaeda, is some sort of armed conflict unlimited in time and space to which IHL applies, is not very tenable. Unfortunately, the Obama administration has also used IHL in this way, if with some adjustments, while in Hamdan the US Supreme Court ruled that Common Article 3 was applicable and that the US is in some sort of global, amorphous non-international armed conflict with Al-Qaeda. (The Hamdan judgment is however quite unclear, and would bear other readings as well, such as that there is a non-international armed conflict between the US and Al-Qaeda in Afghanistan).
There has long been a consensus outside the US legal academia that IHL is inappropriate to regulate the fight against terrorism, outside specific situations where armed hostilities actually occur and have a certain level of intensity, as in Iraq or Afghanistan. For God’s sake, just how absurd would it be to look at the most recent Christmas/Underwear Bomber as some sort of ‘unlawful combatant’ in a global armed conflict?
Ken has thus valiantly argued on several occasions that the US government should desist from invoking IHL in this unconvincing manner, and that it should rely instead on its customary right of self-defense to justify the targeted killings of suspected terrorists:
That’s the legal authority that permits the US to strike at its enemies whether in a combat theater or not, in safe havens far away from any regular battlefield, and it is the traditional authority on which the US has always relied. And it is the authority on which the Obama administration is actually relying, if one looks to how it behaves. That is, if you asked US government lawyers twenty or twenty-five years ago on what basis the US would strike Al Qaeda targets in Somalia, the answer would likely be, customary law of self-defense — if a state is unable or unwilling to control its territory, the non-state actor safe havens are liable to attack.
Now, assuming that targeted killings of certain suspected terrorists are desirable in at least some cases as a matter of policy, is Ken’s self-defense argument for justifying them sufficiently persuasive? I respectfully submit that it’s not.
First, there are the obvious structural problems in invoking self-defense against non-state actors, ranging from whether such self-defense is even permissible (I think that it is), to what the actual conditions for it are (e.g. should there be a link of attribution to a state), whether a particular terrorist attack crossed the Article 51 of the Charter threshold of an armed attack, etc., etc. (For a discussion of these issues, see Christian Tams’ recent article in the EJIL.)
But then there is more fundamental problem: what is the actual wrongfulness that is being precluded with self-defense, and can self-defense even conceivably do it?
As I see it, when the US uses a predator drone in Pakistan to kill a terrorist, it thereby commits two distinct acts which can in principle be characterized as wrongful: it violates the sovereignty of Pakistan, and it violates the right to life of the person killed. It is the wrongfulness of the former only that can be precluded by an invocation of self-defense, just like Pakistan’s (or Yemen’s, or whoever’s) consent would preclude it. But, assuming the (extraterritorial) application of human rights treaties to a given situation, I don’t see how self-defense could be used to preclude the wrongfulness of an act contrary to the individual rights enshrined in such treaties. It is indisputable, for example, that self-defense as a matter of the jus ad bellum cannot preclude the wrongfulness of the violations of the jus in bello, i.e. IHL. How could it be any different when it comes to human rights? Indeed, the ILC explicitly says so in its commentaries on the articles on state responsibility, at 74.
Thus, I don’t see how Ken’s proposed solution can actually do all the work that it needs to do. It’s one thing to say that a state can’t complain about another state violating its sovereignty when it responds to an armed attack by a non-state actor which the former state did not prevent. It’s quite another to say that individuals somehow lose their equally inherent rights just because a state exercises its inherent right to self-defense.
What states wishing to use targeted killings can argue – and have argued – to avoid this problem is that human rights treaties don’t apply at all, for example because of extraterritoriality. These arguments are, in my view, entirely unprincipled and unpersuasive, but I won’t develop that here. Assuming that a human rights treaty does apply, self-defense just doesn’t cut it. On the other hand, if the human rights treaty doesn’t apply, then the invocation of self-defense becomes redundant.
However, targeted killings may, in some circumstances, be justifiable within the framework of the human rights treaties themselves. Article 6(1) ICCPR thus only prohibits arbitrary deprivations of life, while Article 2(1) ECHR allows for intentional taking of life when it is ‘absolutely necessary’ to do so to defend other persons from unlawful violence, to effect a lawful arrest, or to quell a riot or insurrection. It should not be forgotten that even democratic states frequently employ lethal force, even in situations with possible collateral damage – just take the examples of a hostage taking, or shooting fugitives to prevent their flight.
What human rights treaties invariably require, however, is a showing by a state that the killing was necessary; that other, non-lethal options were explored but were not viable. If a drone were to some day kill Osama bin Laden, whether he’s holed up in some cave in Pakistan or, to get rid of the extraterritoriality problem, is found hiking the Appalachian trail, this would not constitute a violation of his human rights if his actual arrest was difficult or impossible. It is only this proportionality analysis which can determine the lawfulness, vel non, of targeted killings. This is, indeed, how the Israeli Supreme Court looked on the matter in its Targeted Killings case. Accordingly, what is necessary is an individualized and factually supported determination – a blanket policy of killing is not permissible.
This is so, in my view, even in an indisputable armed conflict, since it is questionable to what extent human rights law can be displaced by IHL targeting rules: lex specialis in particular is not worth the Latin its written in. And since no Suday would be complete without some shameless self-promotion, the readers might be interested in a forthcoming article of mine on norm conflicts, IHL and human rights which deals with some of these issues, and is now available on SSRN. I’ve also posted an article on the Human Rights Committee Sayadi case on Security Council sanctions against suspected terrorists, which is derived from my initial take on the case here at EJIL Talk. Comments are, as always, most welcome.