Ben's most recent post on the al-Aulaki suit has precipitated an interesting discussion at Opinio Juris involving, among others, John Dehn, Kevin Heller, and Mary Ellen O'Connell (Mary Ellen's thoughts appear in Kevin's post and in his comment thread). The full discussion is well worth a read, in part because it teases out an important disagreement among those who take the view that IHL does not authorize the use of force against al-Aulaki in Yemen. Specifically, it illustrates the difference between the geographically-strict view advanced by the ACLU and CCR in al-Aulaki (reflected in Mary Ellen's views) and the more flexible but still-limited view favored by Kevin. Interestingly, moreover, the exchange also draws attention to a particular reason why the suit might ultimately founder on state-secrets grounds.
Mary Ellen's position (quoted in Kevin's post, and drawn from her declaration in al-Aulaki) is narrowest. On her model, the authority of the United States to use lethal force lawfully is confined to locations such as Afghanistan in which violence involving an organized armed group has reached an adequate level of intensity, at which point IHL norms become relevant:
13. Armed conflict has a territorial aspect. It has territorial limits. It exists where (but only where) fighting by organized armed groups is intense and lasts for a significant period.
14. It is my understanding that the government has argued that the armed conflict against al Qaeda is a global conflict, and that the law of armed conflict governs the detention, prosecution, and killing of suspected al Qaeda associates wherever they are found. This conception of armed conflict is inconsistent with the one recognized by international law. That the United States is engaged in armed conflict against al Qaeda in Afghanistan does not mean that the United States can rely on the law of armed contlict to engage suspected associates of al Qaeda in other countries. The application of the law of armed conflict depends on the existence of an armed conflict. Armed conflict exists in the territorially limited zone of intense armed fighting by organized armed groups.
Kevin's position is considerably more flexible:
I have no problem with the idea that an individual who is member of an “organized armed group” participating in a territorially-bounded armed conflict is targetable even when he is outside of that armed conflict. Nor do I have a problem with the idea that a civilian who directly participates in a territorially-bounded armed conflict is targetable for the duration of his participation even if it he is outside of that territory. The problem with the government’s argument in Al-Aulaqi is not that it takes a non-territorial approach to armed conflict, but that it cannot identify an actual armed conflict to which Al-Aulaqi is connected, either as a member of an organized armed group or as a civilian directly participating in hostilities.
If I'm reading this right, Kevin's objection to the proposed use of force against al-Aulaqi turns in significant part on the notion of a distinction between al Qaeda "central" ("AQ") and al Qaeda in the Arabian Peninsula ("AQAP")--the entity with which al-Aulaki usually is associated. On this view, membership in AQAP is not tantamount to membership in AQ itself, meaning that AQAP members are not subject to targeting consistent with IHL unless and until they engage in specific conduct aside from their membership implicating IHL (in Kevin's view, DPH provides the standard for this conduct inquiry).
This raises the question whether AQAP should be understood as distinct from AQ central for purposes of the targeting question. The issue would be simple were we dealing with states, of course; one can readily distinguish between Afghanistan and Yemen, for example. And it's not necessarily a hard inquiry just because the entities involved are non-state actors--no one should have trouble distinguishing al Qaeda from Hezbollah, after all. But determining the nature and salience of the organizational distinctions between AQ central and entities such as AQAP (or AQIM, AQI, JI, Abu Sayyaf and so forth) is a considerably more difficult task, one for which we need both clear legal metrics and good intelligence.
As to legal metrics: What law, if any, specifies the conditions for categorizing AQAP as distinct from or instead a subpart of AQ? Note that this is a distinct question from co-belligerency. That is, if AQAP is a distinct entity for IHL purposes, we might then have a discussion of whether it nonetheless has become party to a relevant armed conflict. But the first question is whether AQAP is really different in a meaningful way from AQ itself. And, related to that, we might also ask whether federal courts, when confronted with this question in domestic litigation, must defer to the executive branch's determination that AQ and AQAP in fact are one and the same.
That leads us to an aspect of the state secrets issue in this litigation. Let's assume that a federal court concludes that it is not obliged to entirely defer to the executive branch on the question of the AQ-AQAP relationship. Won't the resulting need for the executive branch to prove the nature of the AQ-AQAP relationship require the executive branch to make use of intelligence that would ordinarily warrant protection under the state secrets privilege? It certainly seems possible, if not probable.