I set out to begin to untangle the strings of argument in the DOJ white paper, and find myself, after three pages of writing notes myself, having untangled only as far as the paper’s page 5 (of 16). There is something wrong with a memo like this. Let me see if I can explain why.
The white paper says that it intends to set forth “a legal framework for considering” when the U.S. government can “use lethal force in a foreign country outside the area of active hostilities against a U.S. citizen who is a senior operational leader of al-Qa’ida or an associated force if al-Qa’ida – that is, an al-Qa’ida leader actively engaged in planning operations to kill Americans.” While the paper disclaims any intention of establishing the “minimum requirements necessary” to make such a killing lawful, I take it to be an explanation for how such an operation is both authorized by domestic law, and in compliance with the various provisions of domestic and international law that regulate the use of force. It is the closest thing we have seen – and perhaps the closest thing that exists – to a detailed argument for why such a practice is legal.
So let’s start where the paper starts, more or less, with the domestic source of authority on which the administration relies for the use of lethal force it describes. The paper points to two sources of domestic law and suggests the operation is authorized either by the Constitution itself or the statutory Authorization for Use of Military Force (AUMF). So which one is it in the highly specific context here – statute or Constitution? The paper never commits. But which one it is matters enormously. It is, in a sense, what matters most. Pull out the argument threads. What law does the paper mean to reference by invoking as a source of authority the “President’s constitutional responsibility to protect the nation” (pp. 1, 2)? I infer from this description the authors mean to reference the President’s power under Article II, though I could find no citation to that provision. In particular I imagine this is some reference to the President’s power to, as the framers put it, “repel sudden attacks.” In such limited form – i.e. as a defensive use of force – it is a principle that has been recognized in various ways back to the Civil War era Prize Cases (though I find no citation to those authorities either). Does the paper contemplate that this Article II power alone – without further congressional authorization – would be sufficient as a matter of domestic law to authorize the use of lethal force against any senior operational leader of al-Qa’ida or an associated force actively engaged in planning operations to kill Americans? If the Article II power so conceived includes lethal force in those circumstances, mustn’t it also include the authority to employ lesser forms of force, like detention, against those figures, without further congressional authorization? Given the Administration’s insistent reliance on statutory, and not purely constitutional, authority to support the legality of ongoing detention of folks like KSM, it is hard to imagine the paper here is really intended as a full throated embrace of that claim.
So maybe the paper means to limit the scope of the “President’s constitutional responsibility to protect the nation” in some other way. Indeed, having invoked this constitutional power, in whatever passing way, the paper immediately also invokes UN Charter, art. 51 – the key international law provision governing the use of force in self-defense. (“[T]he President’s use of force against al Qaeda and associated forces is lawful under other principles of U.S. and international law, including the President’s constitutional responsibility to protect the nation and the inherent right to national self defense recognized in international law.”) (p.2) So perhaps the idea is that whatever Article II power the President has under the Constitution, it extends only so far as (or is co-extensive with) a state’s ability to use force in self-defense under international law. Ok, so Article 51 (which the paper does not quote) says: “Nothing in the present Charter shall impair the inherent right of individual or collective self-defence if an armed attack occurs against a Member of the United Nations, until the Security Council has taken measures necessary to maintain international peace and security.” Is the argument, then, that the relevant armed attack against the United States occurred on September 11, 2001, and that a response now would still be timely? Or is the argument that there is some subsequent action by Al Qaeda against the United States that rises to the level of an “armed attack” that the Administration is counting? Or is the argument, rather, that it doesn’t matter whether an armed attack has occurred recently and that the targeting in this context is an exercise in anticipatory self-defense – that is, whatever “inherent right” of anticipatory self-defense survived the UN Charter as a part of customary international law? I could imagine an argument here, too – one that might begin, as international law casebooks do, with the Caroline’s statement requiring that “the necessity of self-defence, instant, overwhelming, leaving no choice of means, and no moment for deliberation.” One could then argue that customary international law has evolved from the Caroline standard, and that there is evidence of an emerging CIL rule requiring a more refined test of imminence. But the paper doesn’t cite the Caroline. It makes no argument about customary international law. And it is only as part of the paper’s constitutional due process analysis that it engages “concepts of ‘imminence,’ feasibility of capture, and compliance with applicable law of war principles” (citing a law review article from 1992, one from 2000, and a single statement of the former UK Attorney General contending that “it must be right” that states can act in self-defense against terrorist groups without knowing exactly where or how the group will attack next). In short, it is difficult to extract a discernible argument in the paper about the current meaning of self-defense under international law.
Maybe then the mention of the “President’s constitutional responsibility to protect the nation” (on p. 1 and again on p. 2) is really meant more for rhetorical (or protective) effect. After all, the first source of domestic authority the paper really engages is the statutory AUMF – and on its face this seems like a far better course. The statute gives the President the authority to use all “necessary and appropriate force” against Al Qaeda. If the U.S. citizen in the paper’s scenario is indeed a member of Al Qaeda, then isn’t that simply the end of the analysis in terms of domestic authority? The problem, of course, is that both the Supreme Court, and the Administration, and now Congress have recognized that the scope and meaning of the AUMF must be understood as informed by the law of war (or call it IHL or call it LOAC). So the paper would have to argue that the targeting of the U.S. citizen in this scenario is consistent with that law.
How would this argument go? First, the paper would have to take the position that the United States and Al Qaeda are in a non-international armed conflict (NIAC) – that is, that the United States and Al Qaeda are parties to a conflict recognized as triggering the law of armed conflict under international law. The paper says as much (p. 3), but that of course is a challenging argument, one not embraced as far as I can tell by most non-U.S. states. Hamdan, which the paper cites, need be read as going only so far as to recognize that there was a NIAC in Afghanistan at the time Hamdan was captured. Historically (i.e. in the 1970’s), NIACs were contemplated to be mostly internal (i.e. civil) wars. But even more modern understandings of what a NIAC might be require findings of a level of organization in the non-state party and an ongoing intensity of conflict between the state and the non-state party that it is far from clear the conflict between the United States and Al Qaeda now meets (though of course the white paper leaked to the press is undated). The paper does not engage this debate.
Second, the paper would also have to evaluate whether the targeting of the U.S. citizen in this scenario was consistent with LOAC targeting rules, rules that say civilians are not targetable “unless and for such time as they take a direct part in hostilities.” (E.g. AP II, art. 13.) The paper could argue that this rule is now outmoded, that even the ICRC contemplates targeting individuals who play a “continuous combat function.” Is it the ICRC’s CCF standard the paper means to incorporate, which standard would allow, for example, targeting those individuals “whose continuous function involves the preparation, execution or command of acts or operations amounting to direct participation in hostilities,” but not the targeting of “recruiters, trainers, financiers and propagandists” whose efforts fall outside the DPH zone? Maybe. But I don’t really know. The paper doesn’t mention either DPH or CCF or the ICRC’s interpretive guidance. The closest it comes to engaging the who-can-be-targeted-in-LOAC discussion is its citation to a set of sources all but one of which are from the law of international armed conflict – not at issue here – which stand for the proposition that all members of armed forces are targetable at any time (in which case what difference does it make whether the target is a senior operational leader or not?).
And this is all before one gets to the question of, for example, whether or not the process the United States follows in selecting U.S. citizen targets satisfies the requirements of due process under the Fifth Amendment of the U.S. Constitution (see paper, p. 6 requiring an “informed” (how informed?), “high-level” (how high?) official to “determine” (how? on what kind of information?) that there is an imminent threat of violent attack against the United States, that capture is “infeasible” (?), and that the operation would be conducted “consistent with applicable law of war principles” (but see supra)). There are scores of Supreme Court due process cases beyond the two relevant cases cited here that elaborate on the meaning of what process is due in various contexts. Cases that engage questions like standards of proof, burdens of persuasion, neutrality of decision-makers, and the possibility of, if not pre-deprivation, some form of post-deprivation adjudicative review. If we are going to weigh burdens and benefits, surely it is worth at least considering those, too.
As a “framework for considering” some of the applicable laws here, the paper is a step ahead of where we were before it was leaked. As a legal argument for why this kind of operation is lawful, there is substantially more work to be done.