The following guest post is the latest in a series comprising a debate as to whether LOAC requires an attempt to capture rather than a first-resort to lethal force in some circumstances. The debate involves Professor Ryan Goodman, on one hand, and an array of posts from Professors Kevin Heller, Jens Ohlin, Geoff Corn, Laurie Blank, Chris Jenks, and Eric Jensen (the last four writing collectively). Here are the earlier entries:
Professor Goodman writes:
Lawfare added Jens David Ohlin’s critique of my EJIL article to this series. I am grateful to Lawfare’s editors for this opportunity to respond, and to Professor Ohlin for his thoughtful analysis.
I have already addressed Ohlin’s criticisms, in part, in my response to Kevin Jon Heller’s critique. My reply here will be briefer than one might expect, because there is not much left that Ohlin and I actually disagree upon—with respect to the major claims in my article. Indeed, much of the remaining disagreement can be resolved simply by clarifying some misunderstandings.
In his critique, Ohlin makes three arguments–each is directed at one of the three sections of my article. This reply is organized accordingly.
I. Allied rules of the Law of Armed Conflict (LOAC)
In my article, I first try to counteract the notion that LOAC contains little or no constraints on the right to kill enemy fighters (p. 16). I map specific prohibitions in LOAC that impose constraints on the right to kill. These include prohibitions on perfidy, the denial of quarter, and killing treacherously. The point of this exercise is fairly modest. I do not attempt an ambitious interpretive exercise to derive—through induction or through a structural analysis of the legal regime—general principles from these specific prohibitions. The point is simply to show that LOAC includes prohibitions of this character—enclaves of specific constraints that safeguard enemy fighters—which makes it “more plausible” (p. 20) that another specific prohibition may exist: a constraint on the power to kill needlessly in some circumstances.
Ohlin’s critique is based on the notion that the purpose of my analysis is to derive general principles (the ambitious interpretive method). On the contrary, I share Ohlin’s general skepticism about the success of any such endeavor in this domain of LOAC. My argument is, moreover, consistent with Ohlin’s level of specification of legal prohibitions that he accepts. For example, Ohlin states that “the relevant treaties codified specific prohibitions against weapons and methods that produced unnecessary suffering and specific actions, like perfidy and treachery, that made the return to peace too difficult” (emphasis added). My argument is simply that another such specific prohibition involves methods of war that produce superfluous injury–which I interpret to include gratuitous or unnecessary killing when capture is an equally feasible alternative.
I recognize that the burden may fall on me, as author, to be clearer if a reader misunderstands my writing. I should note, however, that when I presented the article at a workshop in February, a colleague’s primary question was why the article steered clear of deriving a general principle from the specific prohibitions and structure of LOAC (the very type of argument that Ohlin mistakenly thinks I make). As indicate above, the answer was that, on my view, that line of analysis is not sustainable.
Finally, for the record, I have my doubts about Ohlin’s conception of military necessity in the contemporary law of armed conflict, and his assumption that the international legal regime is tied to the US Lieber Code’s particular vision of warfare. Indeed, the more extreme Ohlin paints Lieber’s view—e.g., as it moves closer to a “total war” theory of LOAC—the more anachronistic becomes Ohlin’s project of applying Lieber to the modern context. That said, I leave it to others, or another day, to debate that part of Ohlin’s fascinating law review article.
II. An Alternate Path: The scope of hors de combat
Ohlin’s analysis of the second part of my argument also stems from a misunderstanding. And I believe further clarity on my part will resolve this disagreement as well.
A separate argument in my article maintains that—whatever one might think about the lesser evil rule and the prohibition on superfluous injury and unnecessary suffering—Article 41 of the Protocol prohibits armed attacks against fighters who become hors de combat (no longer taking part in hostilities). Article 41(2) provides:
A person is hors de combat if:
(a) he is in the power of an adverse Party;
(b) he clearly expresses an intention to surrender; or
(c) he has been rendered unconscious or is otherwise incapacitated by wounds or sickness, and therefore is incapable of defending himself;
provided that in any of these cases he abstains from any hostile act and does not attempt to escape.
I contend that it is erroneous to claim that fighters must surrender or be incapacitated by wounds to be considered hors de combat. The text plainly shows that there is another independent category: section 2(a) defines hors de combat to include a person who “is in the power of an adverse Party.” The question is what that category means. After reviewing the text and travaux, I conclude, more tentatively, that this category includes combatants who are defenseless and at the complete mercy of an attacking party.
The implication of this part of the analysis is not that it offers direct evidence of a least-restrictive means analysis for the use of force. The implication is instead that the hors de combat rule, if interpreted broadly, can perform much of the same work as the least-restrictive means analysis in similar factual scenarios (e.g., defenseless combatants who can easily be apprehended). And the hors de combat rule may yield even more extensive protections—because it precludes any attack against the potential target regardless of the military costs to the attacking party.
Ohlin’s analysis suggests that he thinks I am discussing a different category of hors de combat—section 2(c). Quoting the Protocol, he states:
Goodman notes that article 41(2) of Additional Protocol I defines a person as hors de combat if “he has been rendered unconscious or is otherwise incapacitated by wounds or sickness, and therefore is incapable of defending himself ….” Presumably, the argument here is that a soldier who shoots and wounds an enemy combatant is prohibited from deploying a second kill shot if the wounded soldier is truly “incapable of defending himself.”
That is not my argument. Those examples would clearly come under section 2(c). My argument is, explicitly and throughout, focused on the class of hors de combat under section (2)(a). I presume that this confusion is why, after presenting his version of my argument, Ohlin concludes “it is hard to see the relevance of it.” If my argument were about section (2)(c), I would agree with Ohlin’s assessment on that score as well.
Additionally, Ohlin states that I do not understand the distinction between the protections afforded hors de combat (Article 41) and the protections afforded fighters still in battle (Article 35). Actually, my article is predicated on that very distinction—hence my suggestion that the former provides an even more extensive constraint on the use of force than a least-restrictive-means analysis in the latter condition. Indeed, here is one of the ways that I expressed the distinction in the article:
In the final analysis, the rules defining hors de combat share much in common with RUF [restraints on the use of force]. … RUF regulate the kind and degree of violence that can be employed against individuals who are legitimate military targets. That analysis is obviated, however, if the relevant individual should not be considered a legitimate military target in the first place. Thus a threshold question is whether the targeted individual is hors de combat.
Ohlin, however, omits my last two sentences when he suggests I do not appreciate the distinction. Ohlin states:
Goodman concludes that “in the final analysis, the rules defining hors de combat share much in common with RUF… RUF regulate the kind and degree of violence that can be employed against individuals who are legitimate military targets.” But this represents a fundamental misinterpretation of the rules regarding persons hors de combat. The hors de combat rules are specific provisions that remove specified targets from the category of legitimate combatants. They do not, however, count as restrictions against the use of force against combatants.
For close readers: note that Ohlin’s final two sentences are substantively similar in content to the sentences that he omitted from my article. In other words, the omitted sentences show the same understanding as Ohlin’s. (And, it is a fairly basic understanding in the law.) In essence, Ohlin and I agree again.
III. Direct Support for the Unnecessary Killing Rule
Ohlin argues that the LOAC rule prohibiting methods of warfare resulting in “superfluous injury or unnecessary suffering” does not include unnecessary killing. Some of the heavy lifting on this argument is conducted by Heller, which Ohlin signs onto. And I have addressed Heller and, inter alia, parts of Ohlin’s arguments in my earlier post. Let’s now turn to some independent points that Ohlin raises.
First, Ohlin states that “Goodman can only point to article 35(1) … and article 35(2)” of the Protocol as the foundation for the unnecessary killing rule. I am not sure what work “only” does here, but, yes, my claim is that Article 35 is the textual provision containing the prohibition on gratuitous and unnecessary killing. Second, Ohlin refers several times to “unnecessary suffering” and omits reference to “superfluous injury” in his claims that Article 35 excludes unnecessary killing. This omission is significant, because, as I explain in my response to Heller, it is the “superfluous injury” prong that directly applies to unnecessary death.
Third, in assessing whether the Commentaries support my interpretation, Ohlin turns to one small part of the Commentaries (the denial of quarter). Like Heller, Ohlin says nothing about the more direct references in the Commentaries concerning “the minimum loss of life” of “the party attacked” (under Article 35 itself), and references in the Commentaries to restrictions on the use of lethal force against individuals who suspend their hors de combat status by resuming combat (Article 41) or by escaping (Article 41). The one passage in the Commentaries where Ohlin makes an argument is unconvincing. The Commentaries’ discussion of the denial of quarter (Article 40) states:
The deliberate and pointless extermination of the defending enemy constitutes disproportionate damage as compared with the concrete and direct advantage that the attacker has the right to achieve. It is sufficient to render the adversary “hors de combat.” The prohibition of refusing quarter therefore complements the principle expressed in Article 35 “(Basic rules),” paragraph 2, which prohibits methods of warfare of a nature to cause superfluous injury or unnecessary suffering.
It is unclear on what basis Ohlin concludes that “[t]he unifying element of this passage is pointlessness, not death.” A more plausible reading is that “extermination” of the enemy fighter that exceeds the legitimate objective of rendering the adversary hors de combat is important to the complementary nature of the two sets of rules. And, indeed, in its discussion of Article 35, the Commentaries state that “‘several representatives [in the treaty negotiations] wished to have it recorded that they understood the injuries covered … to be limited to those which were more severe than would be necessary to render an adversary hors de combat’” (see also Commentaries to Article 35 stating: “The object of combat is to disarm the enemy. Therefore it is prohibited to use any means or methods which exceed what is necessary for rendering the enemy hors de combat.”). Killing an individual when other methods of rendering them hors de combat are available is the unifying principle.
The same unifying principle is present in the Commentaries on another provision: Article 41. In discussing limitations on the use of force against an individual who loses his hors de combat status by resuming combat, the Commentaries state:
Any hostile act gives the adversary the right to take countermeasures until the perpetrator of the hostile act is recognized, or in the circumstances, should be recognized, to be “hors de combat” once again. Obviously the remarks made above with regard to Article 35 (Basic Rules), paragraph 2, concerning the prohibition of superfluous injury or unnecessary suffering, continues to apply in full. The retort should be proportional to the measure of danger. It should not amount to a refusal to give quarter. Whatever the situation, the criterion of having been rendered “hors de combat” suffices.
Once again, the link between the denial of quarter and superfluous injury and unnecessary suffering involves the gratuitous use of lethal force—determined by whether the force exceeds what is needed to render the adversary hors de combat.
Fourth, Ohlin warns against any emphasis on commentators—including the ICRC Commentaries—as evidence of legal obligations. The question for our discussion is the relevance of these sources for treaty interpretation. As I mentioned in my reply to Corn et al., the position of particular commentators directly influenced states—for example, the final report of a 1973 expert group was endorsed by states at an important intergovernmental meeting in Lucerne. Furthermore, Ohlin’s own analysis would support assigning great importance to the ICRC’s reports from the 1970s. That is, Ohlin states “the experts and commentaries are relevant not as scholarly opinion but only as evidence of legislative intent when a given treaty or protocol was being drafted by experts.” Notably, the ICRC was responsible for writing the draft Protocol, submitted it to the diplomatic conferences, and participated in the discussions. Finally, there is an odd aspect to this debate. Ohlin casts doubt on resort to the ICRC Commentaries. Yet Heller’s argument, upon which Ohlin relies, draws heavily from the Commentaries. Indeed, Heller’s argument that Article 35 applies only to “disabled men” effectively depends upon the Commentaries.
Finally, I should address Ohlin’s suggestion that the interpretation of the treaty cannot turn on the fact that Pictet’s maxim was reflected in several parts of the travaux and explicitly supported by governmental delegations. Ohlin writes: “Goodman must demonstrate that Pictet’s view carried the day and that it won over in a contest against his adversaries when it came time to draft a final version of the Additional Protocols.” But who exactly were the adversaries and what was their position? The only evidence of opposition that Ohlin cites is a passage (1) that broadened the definition of military necessity (p. 36 of the article); (2) that is otherwise consistent with the lesser evil rule; and (3) that Ohlin himself suggests may be limited to the law on weaponry. In his critique, Ohlin also states: “Arguing that the majority of the experts who helped negotiate the instrument believed in this interpretation—even if this were true—does not demonstrate that the Additional Protocol codifies a least-restrictive-means test.” The burden falls on Ohlin once one considers the affirmative evidence. My article, for example, includes strong historical evidence such as the ICRC explaining to the diplomatic conferences that the rule would entail a kill or capture prohibition; governments discussing the principle had expressed “a wide measure of agreement that as few as possible should be killed, no more than necessary should be wounded and those lightly rather than gravely.” In the absence of significant evidence of opposition, the most plausible explanation is that Article 35’s prohibition should be interpreted in accordance with this part of the travaux, the ICRC Commentaries, and the Bothe et al treatise—all of which agree that the prohibition on superfluous injury and unnecessary suffering includes unnecessary killing.