[Editor’s note: Ryan Goodman replies to Professor Heller in a subsequent post.]

In a recent post here at Just Security, Ryan Goodman offered a novel – and characteristically intelligent – defense of the US position that it is involved in a non-international armed conflict (NIAC) not only with al-Qaeda, but also with al-Qaeda’s “associated forces.” According to Ryan, the US is involved in a NIAC with al-Qaeda’s associated forces because they “belong to” al-Qaeda for purposes of the rules of IHL governing targeting and detention. Here is what he said, nominally in response to Christof Heyns’ assertion in his recent UN report on extrajudicial killings that an associated force must “form part” of al-Qaeda for its members to be targetable and detainable:

Nevertheless, the law of armed conflict stipulates that members of armed groups (e.g., AQAP) with a particular relationship to a party to a conflict (e.g., al-Qaeda) are legitimate targets. Specifically, Article 50(1) of Additional Protocol I states that a person cannot be considered a civilian (e.g., for the purpose of lethal targeting) if he is a member of an organized armed group with such an association. That association is defined in Article 4(a)(2) of the POW Convention: “members of other militias and members of other volunteer corps, including those of organized resistance movements, belonging to a Party to the conflict.”

In short, “belonging to” a party to the conflict is a form of an associated group, which renders its members subject to lethal force and detention.

I disagree with Ryan that the concept of “belonging to” can be applied in non-international armed conflict (NIAC). In this post I explain why.

“Belonging to” in NIAC

As Ryan explicitly acknowledges, conventional IHL addresses the concept of “belonging to” only in the context of international armed conflict (IAC) – as one way in which irregular armed forces (partisans) can be considered part of a state’s regular armed forces. The basic question Ryan must answer, therefore, is this: what justifies applying the concept of “belonging to” to NIAC? Ryan’s answer is two-fold – it applies either by analogy or by custom:

Of course these rules would also have to apply by analogy, or by custom, to non-international armed conflicts. But that is not as much of a stretch as rules that were developed to deal with particular relationships between states and neutrality in international wars. Indeed, the category of “belonging to” a party is derived from relationships that include nonstate actors. And the 2009 ICRC Interpretive Guidance on Direct Participation in Hostilities (which discusses the test for “belonging to” at length) is also replete with references to the same terminology — of organized armed groups “belonging to a party” — in the context of purely noninternational conflicts (e.g., “Members of organized armed groups belonging to a non-State party to the conflict cease to be civilians for as long as they remain members….”).

Let’s start with analogy. To begin with, it is not clear what theory of international law permits a state to take a concept that exists solely in conventional IAC and apply it by analogy to NIAC. This is an issue that Ryan does not address; he simply assumes that, if the “belonging to” analogy makes sense, the US would be entitled to rely on the concept – which would mean, of course, using it as a basis to kill individuals that would not otherwise be targetable and detainable under IHL. In my view, that’s a problematic assumption. Had the drafters of the Third Geneva Convention (GC III) wanted to make the concept of “belonging to” available in NIAC, they could have included it in Common Article 3. Or (perhaps more likely) the drafters of the Second Additional Protocol could have included the concept in Article 43. Neither did, so it seems to me that, at the very least, Ryan must be able to show that “belonging to” is available in NIAC as a matter of customary international law. (An issue I will return to below.)

Ryan’s argument also assumes, of course, that the analogy works. I disagree. Ryan invokes the analogy to establish that the members of an associated force that “belongs to” al-Qaeda are no less targetable and detainable than the members of al-Qaeda itself. Art. 4(a)(2), however, is not primarily concerned with justifying targeting and detention. On the contrary, its primary concern is to establish that members of a non-state actor that “belongs to” a state are entitled to POW status upon capture.  In the late 1940s, when the Geneva Conventions were drafted, there was no question that partisans could be targeted or detained. The issue was whether they should be given the same status upon detention as the armed forces of the state for which they fought. The inclusion of “belonging to” in Art. 4 thus represented a resounding – and unprecedented – victory for those states that wanted to ensure that partisans could participate in armed conflict on the same terms as regular soldiers.

So here is the problem: Ryan is not actually arguing that the concept of “belonging to” should apply to NIAC by analogy. Instead, he is arguing that the concept should apply to NIAC selectively – only with regard to those aspects of the analogy that would make it easier for the US to target and detain al-Qaeda’s associated forces. Nothing in Ryan’s post suggests that members of al-Qaeda’s associated forces would have to be treated as POWs upon capture, which is what the analogy between IAC and NIAC would imply. (And, of course, the US would never accept such a suggestion.) Ryan thus appears to believe that the US is entitled to all of the advantages of the analogy with none of its disadvantages. That is a problematic understanding of analogy.

But what about custom? Could customary international law make the concept of “belonging to” applicable in NIAC? Absolutely. Indeed, it could even make the concept applicable in the asymmetrical way discussed above: justifying targeting and detention but not requiring POW status. The problem is that there is absolutely no state practice or opinio juris in support of that customary rule. Not even the US has claimed that “belonging to” applies in NIAC by custom. And even if it did, the US is not entitled to unilaterally dictate the content of customary international law. We would still need to find that other states believe that “belonging to” applies in NIAC. And they do not.

Finally, it’s worth noting that the ICRC’s Interpretive Guidance on Direct Participation in Hostilities does not actually strengthen Ryan’s argument. Ryan doesn’t specify whether he believes the ICRC’s references to “belonging to” justify applying the concept in NIAC by analogy or by custom. But it doesn’t matter, because those references don’t support either argument. First, the ICRC only discusses the concept “at length” in its traditional context – IAC (see pp. 23-24). Second, and more importantly, although the ICRC mentions “belonging to” in its discussion of NIAC, it does not use the concept in the manner that Ryan does. Ryan wants to use the concept to connect non-state actors to other non-state actors – “associated forces” to al-Qaeda, in particular. The ICRC, by contrast, uses the concept to connect the military wing of a non-state actor to its political wing. That is what the ICRC means by “an organized armed force or group belonging to a party to the conflict,” as it makes clear on p. 32 of the Interpretive Guidance:

[I]t is crucial for the protection of the civilian population to distinguish a non-state party to a conflict (e.g., an insurgency, a rebellion, or a secessionist movement) from its armed forces (i.e., an organized armed group). As with state parties to armed conflicts, non-state parties comprise both fighting forces and supportive segments of the civilian population, such as political and humanitarian wings. The term organized armed group, however, refers exclusively to the armed or military wing of a non-state party: its armed forces in a functional sense.

The ICRC’s Interpretive Guidance, in short, does not support the idea that the concept of “belonging to” can be used to connect associated forces to al-Qaeda in a manner that would make members of those associated forces targetable and detainable.

“Unintended Consequences”

Ryan’s argument is not simply legal. He also claims that failing to accept the “belonging to” analogy would lead to two unintended – and undesirable – humanitarian consequences.  Here is the first:

[L]et’s consider the POW protections for members of associated forces in past and future conflicts. Recall that the definition for “belonging to” a party was set forth in the POW Convention. Indeed, its modest requirements were designed in part to ensure that members of resistance movements—such as the French resistance to the Nazis—would receive humane treatment upon capture. Raising the threshold for what it means to “belong to” a party to a conflict could negate such protections.

This claim only makes sense if Ryan believes that applying the concept of “belonging to” in NIAC requires members of al-Qaeda’s associated forces to be treated as POWs upon capture. If it does, then yes – raising the threshold for “belonging to” would indeed limit the protections for detained members of those associated forces. But if it doesn’t (which, again, would certainly be the US position, if not Ryan’s), then members of associated forces lose nothing at all by not “belonging to” al-Qaeda. Insofar as their detention is governed by IHL, the law governing their treatment would be the same regardless of whether they analogically satisfied the requirements of Art. 4(2)(a) – Common Article 3 and any provision of the Second Additional Protocol that reflects custom.

Ryan’s first argument assumes, of course, that the treatment of al-Qaeda’s associated forces is governed by IHL. Many scholars – myself included – would argue that at least some of those associated forces cannot be connected to any actually existing NIAC, which means that the targeting and detention of their members is governed by international human-rights law (IHRL), not by IHL. And that leads to Ryan’s second argument, which is that individuals who care about human rights should accept the idea that IHL governs such targeting and detention, because the US does not accept that IHRL applies extraterritorially:

Second, consider the implications for the current use of force in Yemen and elsewhere. Followed to its logical conclusion, the Heyns Report  (para. 63) (and perhaps Emmeron’s as well (para. 67)) would deem US drone strikes against AQAP in Yemen to fall outside of armed conflict. In my first post for Just Security, I warned of a future in which the US Government rids itself of the restrictions of the Geneva Conventions (jus in bello) by arguing that “there is no longer an armed conflict in different areas or with different groups” and instead relies on the law of self-defense. That day would be upon us if the Heyns Report were correct. US drone strikes in Yemen would, on the vision set forth by the Report, not be subject to jus in bello, and no matter how errant could never be considered war crimes. It may be far better for human rights groups to rely upon the Geneva Conventions than hope that the US military and CIA will ever bend to human rights principles applied extraterritorially in matters of self-defense in which US airborne forces arguably do not even have effective control.

There is more than a grain of truth to this argument. It may well be the case that members of al-Qaeda’s associated forces would be better off under IHL than under domestic US law and the law of “self-defense” (which cannot justify using lethal force against anyone, as Marko Milanovic has patiently explained). But that does not mean that we should give in to the US’s extortionate approach to international law – its demand that we either protect members of associated forces by embracing its flawed understanding of IHL or accept the undesirable humanitarian consequences of leaving those members exposed to its even more flawed understanding of both the jus ad bellum and the extraterritorial applicability of IHRL. That is not an acceptable choice, and we must refuse to buy into it. Instead, we must continue to insist that the US accept that not every drone strike is governed by IHL and that at least prohibitive rules of IHRL apply extraterritorially. In the long run, insisting on those basic principles will protect humanitarian values far better than accepting the US’s antediluvian approach to international law.