Home EJIL Analysis The Power to Kill or Capture and the DOJ White Paper

The Power to Kill or Capture and the DOJ White Paper

Published on February 11, 2013        Author: 

Ryan Goodman is Anne and Joel Ehrenkranz Professor of Law at New York University School of Law

Thank you to the editors of EJIL: Talk! for inviting me to reflect a bit on the US Department of Justice’s White Paper on Targeted Killings in light of my forthcoming article: The Power to Kill or Capture Enemy Combatants, 24 European Journal of International Law (2013).  The article argues that international humanitarian law (IHL) prohibits killing instead of capturing an enemy fighter in important cases—where many other commentators assume that IHL is highly, if not completely, permissive (see meaty footnote 17, especially if you’re among the latter). My analysis does not require one to take a position on whether the current fight against Al Qaeda constitutes an armed conflict. The analysis is, instead, an exhaustive study of the pertinent jus in bello rules—for whenever an armed conflict triggers their application. Thus with respect to U.S. targeted killings, we can adopt a conditional position: If one accepts (even just for the sake of argument) that the law of armed conflict applies, then one needs to recognize that IHL will sometime impose a duty to capture instead of kill.

The DOJ White Paper is—unsurprisingly–premised on the assumption that the United States is in an armed conflict with Al Qaeda and associated forces, and that IHL applies. Refreshingly for students of international law, the document goes further and accepts that IHL is directly relevant to major questions of domestic law. First, in terms of procedural due process rights, the DOJ states that the constitutionality of the decision to kill a U.S. citizen hinges, in part, on whether “the operation would be conducted in a manner consistent with applicable law of war principles.” Second, it is a federal offense to kill or attempt to kill a US citizen in a foreign country (18 USC 1119(b)). The DOJ explains, however, that the federal law on foreign murder provides a safe harbor—an exception for killings that occur in armed conflict and comply with the laws of war. Finally, the question of Presidential power turns in part on international law. Shortly after September 11th, Congress authorized the President to use force to hunt and kill Al Qaeda members. The DOJ recognizes, however, that authority came with the following proviso: the President’s actions must abide by the laws of war. If they don’t, the President presumably would be acting without affirmative congressional authorization. In short, the White Paper identifies important areas of domestic law that are predicated on compliance with IHL.

The document, however, misunderstands the content of IHL. The White Paper suggests that, according to IHL, enemy fighters can be killed instead of captured anytime, anywhere unless and until they surrender. At one level, this is an understandable mistake and by no means idiosyncratic. Indeed, I researched and wrote my article because so many commentators expressed a similar assumption—a meme began to spread—and I wanted to correct it. That said, there may be a greater responsibility on the part of the DOJ not to simply repeat an idea like this one, but to research and justify it. Indeed, once you start exploring the relevant negotiating history and other important background texts, it becomes apparent that the DOJ’s position is unsustainable as a matter of contemporary law and the history of Geneva.

In The Power to Kill or Capture, I argue that, in certain well-specified and narrow circumstances, the use of force should be governed by a least-restrictive-means analysis. That is, I contend that the modern law of armed conflict supports a maxim that has often been associated with Jean Pictet: if enemy combatants can be put out of action by capturing them, they should not be injured; if they can be put out of action by injury, they should not be killed; and if they can be put out of action by light injury, grave injury should be avoided.

Through original research, the article recovers a history that has been lost in contemporary debates. In particular, I present and analyze voluminous support by international authorities that contradicts the meme. The full record that I foreground sheds significant light on the proper interpretation of key legal instruments such as Additional Protocol I to the Geneva Conventions—especially its definition of “superfluous injury or unnecessary suffering.” And my analysis shows how a parallel set of rules—on the definition of hors de combat—achieves many of the same effects as Pictet’s maxim. Admittedly, there are all manner of caveats and conditions that will qualify the application of the maxim. Nevertheless, the general formula—and its key components—should be understood to have a solid foundation in the structure, rules and practices of modern warfare. In sum, belligerents must comply with an important (albeit conditional) set of IHL constraints in planning and conducting kill or capture operations against enemy fighters. These implications obviously extend well beyond the conditions that allow for lawful targeting by the United States, and include the conduct of parties to armed conflicts in other corners of the globe.

Print Friendly
Filed under: EJIL Analysis

8 Responses

  1. Jordan

    A mistake? See Hague Convention No. IV, Annex, art. 23(c) — it is impermissible to kill an enemy “who, having laid down his arms, or having no longer means of defence, has surrendered” See also re: killing of bin Laden, and re: killing of al-Awlaki, [but noting why the U.S. cannot be at war with al Qaeda].
    Of course, the laws of war apply in the theatre of war in Afghanistan that has migrated de facto to parts of Pakistan and, in any even, self-defense permits targetings of DPAA see above and
    Human rights law applies globally, but persons protected outside the U.S. are those who are in the “effective control” of the U.S. (e.g., after they surrender).

  2. Jordan

    p.s. “if they can be” would have to be considered in context in any event. A soldier does not shoot at an enemy combatant in order merely to shoot the enemy’s arm, much less to shoot the rifle or pistol out of the hand(s) of the enemy combatant. Also a soldier in battle does not have to stop and ask “are you willing to surrender and be captured?”

  3. [...] selon Ryan Goodman, professeur à la New York University School of Law. Dans certaines circonstances bien [...]

  4. [...] Goodman énumère les fondements directs à l’emploi limité de la force. Il s’agit [...]

  5. Prof. Goodman,

    I enjoyed your article, even if I mostly disagree with the conclusion. A few reasons for my disagreement:

    1. As you acknowledge, there is no direct treaty support for the proposition you advance on concerning the restraints on the use of force (RUF). At most, there are principles that can be used to argue for such a conclusion.
    2. With a few exceptions, the support for your conclusion comes from commentators, not States.
    3. As RUF would be a limitation on what States can do under international law, then referring to a lack of known (public domain) State practice for a broader right to kill under what I might term a conservative view of hors de combat (ie, not including mere defencelessness) reverses the onus.
    4. I submit the fact that this was a live issue during the negotiations for API (as you point out, the 1973 Draft made specific mention of it) but the text of API does not clearly adopt the position you are advocating remains a telling point. If States had supported the view that a defenceless combatant should be considered hors de combat, then they could have very easily adopted that position (and even had draft wording to that effect).
    5. The ICRC Commentary on art.41 API starts by saying that the issue was about ensuring no gaps in protection prior to PW status. So far, so good. While the Commentary goes on to claim that a defenceless person is hors de combat, I suggest this is a position being advocated as a preferred interpretation by the authors and it is not a incontrovertible explanation of the text.

    Of course, the above points apply to situations where API is the law. Jordan has already mention HIVR. I just add that Rule 47 of the ICRC Study on Customary International Law seems to limited the concept of hors de combat due to defencelessness to those who are defenceless “because of unconsciousness, shipwreck, wounds or sickness”. This was a clear opportunity to define what type of defencelessness amounted to hors de combat. Further, if the authors of the Study intended for someone who is defenceless to be included in the category of “anyone who is in the power of an adverse party”, I imagine they would have said so seeing how the two sub-rules appear right next to each other.

    I should conclude with saying that you make some interesting arguments de lege ferenda, I just do not agree that your proposed maxim is lex lata.

  6. [...] leader of Al-Qa’ida or an associated force” at Opinio Juris (here and here), EJILTalk (here and here), Lawfare (here), and ASIL Insights [...]

  7. Ryan Goodman

    Dear Captain Henderson:
    Thank you for your very thoughtful comments on my article. I admire your work, and our exchange will benefit my own thoughts and research on these important legal questions.
    As you note (your point no. 1), I acknowledge in my article that there is generally a lack of explicit treaty text for restraints on the use of force (RUF) against combatants, and certainly no treaty provision akin to the express provision on protection of civilians (API, art. 57). That said, I also state: “The exception [to this lack of explicit text] includes the prohibition on superfluous injury and unnecessary suffering in Additional Protocol I”—and it is that provision (article 35) which includes RUF. My analysis relies for support (contra your point no. 2) on the ICRC Commentaries (which reflect the travaux préparatoires to article 35 and related articles); the leading treatise by Bothe, Partsh, and Solf; my own research on the travaux, and intergovernmental meetings prior to 1977. And, when I rely on states, it includes statements that reflect broad state support such as the Australian Ambassador’s conclusion that among governmental delegations, “There already seems a wide measure of agreement” supporting the rule. Finally, the interpretive presumption you suggest (your point no. 3) appears to be more appropriate for analysis of customary international law, not treaty obligations. States that have ratified the Protocol have accepted an obligation to constrain their—and their enemies’—behavior. The question is what that obligation specifically entails. In any case, my article then goes on to identify a significant body of state practice consistent with the RUF obligation.
    With respect to the definition of hors de combat, my article acknowledges the weaknesses you suggest, but I conclude that the better interpretation is nevertheless the one contained in the Commentaries to article 41. The drafters opted for a very broad construction. And, in the conclusion, I state that the negotiations, at the very least, “alerted the drafters to the substantial support for the idea that defenselessness” would be covered under this broad heading.
    Finally, you raise two other sources of law: the Hague Convention of 1907 (per Prof. Paust’s comment) and customary international law. On the former, a major point of my article is that a lot has changed since 1907, and the drafters of the 1977 Protocol obviously chose to reject the Hague formulation. As for custom, the ICRC Study on Customary International Law, which you raise, reaffirms that the category of “anyone in the power of an adverse party” provides an independent basis for hors de combat protection. Does the ICRC Study indicate whether that category includes a soldier who is defenseless? Yes, albeit indirectly. In the next couple sentences, the Study states that this legal rule “has been confirmed in numerous military manuals,” and the manuals that it quotes include the following: Australia (“Combatants who are unable to continue hostile action and refrain from attempting to do so must be treated in the same fashion as noncombatants. … The basic principle is that any person who is hors de combat, whether by choice or circumstance, is entitled to be treated as a noncombatant … A person is hors de combat if that person … is under the control of an enemy….”); Cameroon (“all combatants who are unable to fight must be spared”); Ecuador (“Combatants cease to be subject to attack when they have individually laid down their arms to surrender [or] when they are no longer capable of resistance”); Kenya (“the enemy combatant who is no longer in a position to fight is no longer to be attacked . . . It is forbidden to kill or wound someone who has surrendered having laid down his arms or who no longer has any means of defence …”); Peru (it is “prohibited to kill defenceless persons” [caveat: ICRC paraphrasing-not a direct quote]); Sweden (“combatants in situations where they have laid down their arms or are no longer capable of defending themselves”). The Study also includes similar provisions in the military manuals and national legislation of other states including Croatia, Egypt, Ethiopia, Georgia, India, Nicaragua, Rwanda, Slovenia, South Korea, Spain, and the United Kingdom. This is not to say whether manuals are sources/evidence of customary international law. In answer to your specific point about the ICRC Study, however, the study does provide affirmative support for a customary rule that defenseless soldiers in the power of an adverse party are considered hors de combat.

  8. Dapo Akande

    Thanks to all those who have commented on Ryan’s article.We are closing comments on this post at this time. However, we will resume discussion of Ryan’s article when it is published in EJIL. At that time, there will be a debate in the Journal on Ryan’s article, as well as further exchanges here on EJIL:Talk!