Over at Opinio Juris, Gabor Rona of Human Rights First offers an extended critique of John Brennan’s speech on the use of lethal force. It is an interesting and provocative post, leading me to share a few thoughts in response.
The Scope of Domestic Authority to Engage in Drone Strikes
Rona’s critique opens by arguing that the Obama administration has exceeded the scope of its domestic law authority to use force. He writes that:
al-Qaeda today is not the al-Qaeda of 9/11, and the 2001 Authorization for Use of Military Force specifically limited military force to those who planned, authorized, committed, or aided the terrorist attacks that occurred on September 11, 2001.
If he means to suggest that the AUMF is limited to the specific persons within al Qaeda personally linked to the 9/11 attacks, I disagree. Such a reading would certainly be news to the broad range of judges who, in the context of the Guantanamo habeas cases, have repeatedly construed the AUMF to encompass al Qaeda as a whole rather than just the small number of al Qaeda members personally involved in the 9/11 plot. The AUMF certainly does require a link to the 9/11 attacks, but that link according to the plain language of the AUMF can be supplied at the organizational rather than the individual level—and that is precisely how the AUMF has been interpreted for more than a decade now.
It may be, though, that Rona accepts the organizational interpretation of the AUMF’s 9/11 nexus requirement, and is merely arguing either that the original al Qaeda network no longer exists and hence the AUMF has run its course, that AQAP and various other entities associated with al Qaeda are sufficiently distinct so as to lie beyond the AUMF’s reach, or both. Both are serious considerations requiring a careful assessment of the best available information. For my part, I don’t think that core al Qaeda is quite dead yet, nor do I think that AQAP is sufficiently distinct from core al Qaeda to treat it as being eyond the scope of the AUMF (whereas al Shabaab is, I think, a tougher question). Of course, neither Rona nor I have access to the intelligence that would be most salient in making definitive judgments about the connections among these groups.
Now, showing that a group is distinct from al Qaeda and the Afghan Taliban should not end the inquiry for domestic law purposes. For example, any construction of the AUMF that would lead to the conclusion that the government is acting unlawfully in using force against groups like the Haqqani Network (a group quite literally engaged in combat operations against the United States, but not properly described as part of al Qaeda or the Afghan Taliban) strikes me as unpersuasive. The Haqqani’s are the NIAC equivalent to a classic “co-belligerent” state, and it seems to me that either the AUMF must be read to implicitly include such scenarios or else that the same domestic authority must be located in Article II. This is one way to understand the much-discussed “associated forces” concept. Of course, that does not answer the question of how to define the set of entities that would come within this implied authority. AQAP, if not part-and-parcel of al Qaeda, would be an associated force in my view. Again, al Shabaab is trickier. A group like Boko Haram would be much trickier still, though I’m in no position to say anything useful as to its ties to al Qaeda or lack thereof.
In any event, once beyond the AUMF, there remains the question of whether the executive branch would be within it domestic law rights if it were to use force in self-defense, without legislative pre-approval, as for example Bill Clinton did in response to al Qaeda’s violence in 1998. It it were in fact necessary to make such an argument as to AQAP, there might then be a very interesting issue as to whether a proportionality requirement (not in the sense of collateral damage proportionality, but akin instead to the jus ad bellum idea of proportionality between the harm and the response) is inherent in Article II self-defense and, if so, whether the scope of our air campaign in Yemen has been and remains consistent with that requirement.
Targeting, IHL, and CCF
Turning to IHL, Rona argues that
The law of armed conflict does not limit targeting to enemy armed forces, but it does prohibit targeting others unless, and only so long as, they are directly participating in hostilities or otherwise perform a continuous combat function. In other words, while members of the armed forces are targetable due to their status, any other targeting powers derive from the individual’s conduct, not his or her status. Brennan lumps all these people into one targetable category – a clear misapplication of international humanitarian law that offends the most fundamental principle of that law: the principle of distinction between combatants and civilians.
FWIW, it may help to clarify the range of possible positions that are in play:
First, one might contend that no one is assimilated to combatant status for targeting purposes in a NIAC, and thus that the only legitimate uses of force involve targets who are DPHing at the time of attack. That is not the position I understand Rona to be advancing.
Second, one might contend that some members of an organized armed group in a NIAC can be assimilated to combatant status for targeting purposes–specifically, those members whose function in the group entails repeat DPH, thus enabling us to say that they have a “continuous combat function” (“CCF”). Those members may be targeted when not hors d’ combat, even if not DPHing at the time, but other members may be targeted only while DPHing. This position further subdivides into a spectrum of views, depending on how one calibrates the nature and frequency of activities required to categorize a person as having a CCF. Strictly construed this may pick up relatively few members of the group, and broadly construed it could encompass most if not all of them. I understand Gabor to be expressing this general position, while calibrating the test toward the strict end of the spectrum.
Third, one might contend that all members of some organized armed groups in NIAC are assimilated to combatant status, either directly or perhaps indirectly by virtue of a broad conception of CCF. On this view, a group might be the functional equivalent to a military force, as opposed to a group that has both military and significant non-military wings, and thus all its actual members (including the cooks, etc.) are assimilated to combatant status for targeting purposes without need for a person-by-person analysis. This may be the US government’s position; I’m not sure (though there are indications in the government’s brief in opposition to cert in al-Bihani that this is indeed their viewew).
Fourth, one might contend that all members of the military wing of an organized armed group in NIAC are assimilated to combatant status, even if there is also a non-military wing to the group, such that no individualized CCF inquiry is needed once the person is linked to the military wing. And fifth, one might simply go whole hog and say that all members, period, are so assimilated. Either one of these might also be the US government’s position, though I think the latter is less likely. The trick, in any event, is defining the boundaries of the “military wing” in connection with dual-use personnel such as financiers and propagandists.
Bearing all this in mind, I’m a bit worried about the way in which Rona talks about continuous combat function (CCF), as he might be read to collapse the concept back into mere DPH. That is, one might read Rona as suggesting that even with CCF, there is some need to determine the conduct of the target at the particular time of the strike. If that is indeed his intent, I think that is incorrect; the point of CCF is that the person is approximated to the status of a combatant for targeting purposes—i.e., the person can be targeted at all times so long as not hors d’combat—though without the usual benefit of combat immunity from domestic law criminal prosecution or entitlement to treatment as a POW upon capture. But it’s not clear to me this is what he means. In any event, he highlights an interesting issue more generally.
“Associated Forces” vs “War against Terrorism”
Rona contends that, by embracing the “associated forces” concept, the Obama administration has effectively embraced the notion that we are at war with any group that might employ terrorism, ala the broad claims of the early Bush administration:
This administration has taken great pains to distinguish itself from the Bush “war against terrorism,” asserting instead that the U.S. is not at war against terrorism per se, but only against al Qaeda and its “associated forces.” In practice, and as most recently articulated by Brennan, the overbreadth of war against “al Qaeda and its associate forces” leaves little daylight between Bush and Obama on the subject of who is the enemy.
If one thinks that there is no limit to the associated forces concept—i.e., if for some reason one thinks the concept does not actually require some meaningful degree of association with al Qaeda—then I suppose you could say this is a return to the notion of a war against all groups employing terrorism. But what evidence is there that this is what the administration understands “associated forces” to mean? None that I’m aware of. It would be far more fruitful to press in on the question of how precisely to define the degree of association, if any, that warrants treating a distinct group as an “associated force.”
Rona criticizes the use of the signature method of determining when strikes are appropriate. A few thoughts on that point:
Bear in mind that, in an ordinary armed conflict, very few if any uses of force are “personality strikes” in which the government actually knows, or thinks it knows, the precise identity of the person it is targeting. Yamamoto was the rare exception, not the rule. In that setting, it is vastly more common to attack based on what might well be described as the signature model. It’s just that no one thinks this odd, since the circumstantial evidence informing the signature model in a conventional conflict setting—i.e., the uniforms, the presence on the front lines or in a military vehicle/plane/vessel, etc.—provide relatively overwhelming evidence of the validity of the inference that the target is a combatant or proper military objective. That, of course, is not always or even often true in the context of Afghanistan, Pakistan, and Yemen (though it probably is true in some cases); the risk of false positives as a general proposition is much higher in this setting. Thus it is very appropriate to think very carefully regarding the calibration of the signature strike approach in these unconventional settings, while not rejecting the model as somehow inherently improper.
Unfortunately, but understandably, the public record does not reveal a great deal about how the signature-strike model is calibrated in these settings. This no doubt reflects at least in part the idea that if we were to state very clearly the precise array of circumstances that are deemed sufficient to infer that a strike is proper, this could be used by the enemy as a guide as to what not to do. In any event, it seems to me that oversight, both from within the executive branch and from the relevant Congressional oversight committees, plays an especially important role on this issue. Which leads to the next point….
JSOC vs. CIA Transparency and Accountability
In discussing and lamenting the use of the “signature strike” model, Rona observes that this model has “previously been conducted by the military’s secretive Joint Special Operations Forces and have just been opened up to the even less transparent and less accountable CIA.” It is certainly routine to depict the CIA as less transparent and accountable than the military. But is it less transparent and accountable than JSOC? I’m not saying it is or it isn’t, but I think we should be cautious with our assumptions on this question.