The Capture-or-Kill Debate #10: Goodman Responds to Heller

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 both Professor Kevin Heller and a group consisting of Professors Geoff Corn, Laurie Blank, Chris Jenks, and Eric Jensen writing collectively.  I have also linked to several relevant contributions from Professor Jens Ohlin.  Here are the earlier entries:

1. Goodman’s full-length EJIL Article

2. Goodman’s short essay on the same topic in Slate

3. The first critique from Corn, Blank, Jenks, and Jensen

4. Jens Ohlin’s recent paper relating to this topic (see also this post from Jens)

5. Goodman’s reply to CBJJ’s critique

6. CBJJ’s rejoinder

7. Goodman’s further response

8. Heller's critique of Goodman

9. A new and relevant paper from Ohlin

Kevin John Heller develops an interesting and vigorous challenge to my analysis of Article 35 of the 1977 Protocol to the Geneva Conventions (in which Jens David Ohlin essentially concurs in part III of his essay). Once again, Lawfare’s editors graciously invited me to respond. The relevant article of the 1977 Protocol states in pertinent part:

Article 35 -- Basic rules 1. In any armed conflict, the right of the Parties to the conflict to choose methods or means of warfare is not unlimited. 2. It is prohibited to employ weapons, projectiles and material and methods of warfare of a nature to cause superfluous injury or unnecessary suffering.

Here is my argument: the law of war does not necessarily impose a duty to try to capture enemy combatants—whatever such a duty might mean. The law forbids, in some circumstances, killing an enemy fighter when doing so is manifestly unnecessarily--for instance, when capture is equally effective and does not endanger the attacking party’s armed forces. That prohibition is the Law of the Protocol: unnecessary killing is a form of “superfluous injury or unnecessary suffering” under Article 35(2). On balance, the evidence favors this straightforward interpretation including: the plain text of the Protocol; the ICRC Commentaries to the Protocol; the leading treatise on the Protocol; and a considerable amount of historical sources, including intergovernmental meetings, in the 1970s leading up to the Protocol. Heller, on the other hand, contends that Article 35 could not possibly include unnecessary killing. He argues that if unnecessary killing were part of the prohibition, “the drafters of Art. 35(2) … would not have chosen two words (“injury” and “suffering”) that focus solely on non-fatal wounds” (emphasis added).  In that part of his analysis, the argument boils down to the question whether “superfluous injury” includes the loss of life or only non-lethal injuries. And Heller doubles down by asserting that my reading is suspect “because no modern scholar reads Art. 35(2) like he does.” As I discuss below, Heller’s analysis is riddled with errors: (1) Heller’s contention that no other scholar supports my reading elides some of the most important writings on this subject. (2) His analysis of the treaty text is inconsistent with a plain reading and would produce illogical results. (3) He does not realize that the specific treaty terms he identifies (“superfluous injury”) were actually inserted to take account of the very harm that he thinks the treaty omits (superfluous death). (4) His criticisms of my references to the ICRC Commentaries ignores  -- rather than grapples with -- the strongest and most direct pieces of evidence in my article showing that unnecessary killing is part of the prohibition. This omission is especially noteworthy because I previously (see my no. 3 here; and see here) faulted the first round of criticisms—by Corn, Blank, Jenks, and Jensen—for conspicuously ignoring these parts of the Commentaries. (5) Heller presents no affirmative evidence from the travaux or Commentaries stating, or even strongly implying, that fatal wounds or unnecessary killing is not a form of “superfluous injury.” I. The scholarly consensus? Let’s start where Heller begins his critique of my article by claiming that “no modern scholar reads Art. 35(2)” as I do. That statement is deeply flawed. Several of the most highly respected law of war experts and some of the most important works of scholarship ever written on this topic clearly support my view. Let’s examine one of the most influential and widely cited (though not by Heller or Ohlin) essays on Article 35 of the Protocol: Henri Meyrowitz’s article, The Principle of Superfluous Injury or Unnecessary Suffering: From the Declaration of St. Petersburg of 1868 to Additional Protocol 1 of 1977, 34 International Review of the Red Cross 98 (1994). Not only does Meyrowitz support my reading of Article 35(2): he directly contradicts Heller’s (and Ohlin’s). Recall that Heller notes, in a passing reference to the Commentaries, that Article 35 incorporated the French expression “maux superflus.” But Heller doesn’t say anything about the significance of that textual development. If he did, Heller would be hard-pressed to explain his notion that “superfluous injury” excludes death. According to Meyrowitz, the French term, which was first used at Saint Petersburg in 1868, specifically concerned unnecessary fatalities:

The French version … replaced, or rather corrected, the notion of unnecessary suffering by using the term "maux superflus", which conveys the further notion of superfluous deaths ….

As Meyrowitz further explains, the English translation of the official French text included a mistranslation, and that error persisted in similar provisions of the 1874 Brussels Declaration and the 1907 Hague Regulation. What was the mistranslation? The English version “fail[ed] to render the additional meanings of superfluous deaths” (Meyrowitz: 104). Article 35 of the 1977 Protocol, however, aligned the English and French texts: “In the English version of Protocol I, which is not a translation, this mistake was corrected as far as the language allowed by using the term ‘superfluous injury or unnecessary suffering’ to convey the meaning of ‘maux superflus’” (Meyrowitz: 104). So far so good, but one might object that (perhaps) the concept of maux superflus is limited to methods and means of warfare that render death inevitable (e.g., dum-dum bullets). Not according to the history or to Meyrowitz.  First, since Saint Petersburg, it is clear that states (and, yes, commentators) generally agree that the only legitimate military objective is to use force to render enemy soldiers hors de combat (taken off the battlefield) or, in a collective sense, to defeat enemy forces. If capture can immobilize as easily as killing, then a deliberate killing would constitute “superfluous death” (in Meyrowitz’s words). Secondly, as I note in my article (footnote 97), Meyrowitz presents examples that closely align with my reading of Article 35. Meyrowitz explains:

“To declare that it is unlawful, for example, to shower bombs and shells on troops that are completely defeated, encircled or retreating, and in any case practically defenceless, thereby not even affording them the opportunity to surrender, the principle of superfluous injury or unnecessary suffering expressed in [Article 35(2)] … may exceptionally be invoked.”

If I stood alone with Meyrowitz, I would be satisfied. But, let’s examine some of the other most important commentaries ever written on Article 35. Consider the leading treatise on the Additional Protocol by Michael Bothe, Karl Partsch, and Waldemar Solf. Notably, Heller excerpts a passage from the treatise that I quote in my article, and Heller states: “I think it’s likely that Bothe, Partsch, and Solf would have used the expression ‘unnecessary death’ instead of ‘unnecessary suffering’ had they been endorsing Ryan’s argument.” First, I quoted that passage in my article for a different purpose (Colonel Hays Parks and others suggested that one of Pictet’s statements at a famous conference was anomalous; yet Bothe et al effectively reproduced that part of Pictet’s statement in their articulation of the rules). Second, it is a different passage of Bothe, Partsch, and Solf that squarely supports my interpretation. Heller is correct that the treatise does not use the exact words “unnecessary death.” But just as good, in discussing Article 35, the treatise refers to minimizing the unnecessary “loss of life” of enemy fighters. Specifically, Bothe, Partsch, and Solf state that the prohibition applies to “measures of violence which are not necessary … loss of life and destruction inflicted must have some rational tendency to prompt achievement of a definite military advantage” (p. 195). That is flatly inconsistent with Heller’s (and Ohlin’s) view, and is perfectly consistent with mine. Notably, this is not an opaque part of the treatise that Heller may have overlooked for that reason. Indeed, other leading works of scholarship on this topic refer to the Bothe, Partsch, and Solf treatise as an exemplification of the view that the prohibition on superfluous injury and unnecessary suffering requires that “combatant casualties” not exceed the military advantage anticipated in an attack. See, e.g., Judith Gardam, Necessity, Proportionality and the Use of Force by States (Cambridge Univ. Press 2004), at p. 74. And, it bears emphasis for readers not familiar with these types of debates: Bothe, Partsch, and Solf are not simply a trio of important scholars writing on the subject. Their treatise is generally considered (including by Hays Parks on this very topic) one of the most authoritative interpretations of the law of the Protocol alongside the ICRC Commentaries. The list of leading experts supporting the view that superfluous injury includes superfluous death does not end there. Professor Christopher Greenwood (now the U.K. Judge on the International Court of Justice) wrote one of the most highly acclaimed and important works on the methods and means of warfare. Indeed, it is a “must-read” for any student of this area of the law—though Heller (and Ohlin) does not mention it. Here’s what Greenwood has to say about our topic: “the principle that belligerents may not employ weapons or methods of warfare of a nature to cause unnecessary suffering serves the objective of protecting even combatants from suffering and death which is not necessary for the achievement of legitimate military goals” (pp. 189-90 emphasis added; see also p. 197; 198; see also Christopher Greenwood in The Handbook of International Humanitarian Law, Dieter Fleck 2d ed. at p. 35). And, another leading law of expert, Leslie Green stated that the prohibition on superfluous injury and unnecessary suffering entails that “there should be no resort to measures which entail suffering beyond that necessary for achieving the purpose of the attack. This is merely the modern version of Sun Tzu’s statement that ‘to capture the enemy is better than to destroy it. … To subdue the enemy without fighting is the acme of skill.’” (Leslie C. Green, The Contemporary Law of Armed Conflict (3d ed. 2008), at p. 151). Add to the list Lawfare’s Jack Goldsmith with Curt Bradley in their 2005 Harvard Law Review article: “The laws of war may require a belligerent, even when targeting a legitimate military target, to avoid unnecessary violence and suffering. This principle might preclude killing a nonthreatening enemy combatant who can easily be arrested without the use of force.” Curtis A. Bradley &  Jack L. Goldsmith, Congressional Authorization and the War on Terrorism, 118 Harvard Law Review 2047, 2120-21 n. 325 (2005); see also Nils Melzer, Targeted Killing or Less Harmful Means? – Israel's High Court Judgment on Targeted Killing and the Restrictive Function of Military Necessity, 9 Yearbook of International Humanitarian Law 87, 97 & n. 54 (2006). Finally, several of the experts invited to advise in the process leading up to the ICRC’s Interpretive Guidance (2009) also took a position at least broadly consistent with, if not strongly supporting, my line of argument. Admittedly, some experts involved in the ICRC study (e.g., Colonel Hays Parks and Michael Schmitt) were vocally displeased with the final report’s conclusion favoring a capture instead of kill rule. However, the records of the ICRC proceedings show that a group of experts disagreed with Parks and Schmitt: “several other experts expressed their support for the view expressed in Section IX of the Interpretive Guidance and rejected any suggestion to delete it. According to these experts, the interpretation provided in Section IX accurately reflected contemporary IHL” (p. 11, emphasis added; see also summary record of fifth expert meeting, p. 11). And my EJIL article challenges the foundation of Parks’ and Schmitt’s published criticisms of the Guidance’s conclusions. II. Plain meaning of Article 35 Heller’s analysis of the plain meaning of “superfluous injury” under Article 35 is unconvincing. In the following discussion, for the sake of argument, I accept Heller’s idea that “unnecessary suffering” does not include death and instead relates to psychological and physical pain of disabled soldiers. Bracketing the “unnecessary suffering” prong of Article 35, let’s revisit Meyrowitz’s explanation of the meaning of “maux superflus.” While Heller (and Ohlin) asserts, without any direct evidence or even a dictionary definition, the strained notion that “superfluous injury” cannot encompass unnecessary death— Meyrowitz’s explains that the term “superfluous injury” was, in fact, added to ensure that superfluous death was covered! At least that is according to of one of the leading works on Article 35, and Heller’s (and Ohlin’s) analysis makes no mention of this account. Moreover, there are several analytic weaknesses in Heller’s plain-meaning argument. First, he presents the following proposition as a truism: “Dead combatants are not injured and do not suffer.” I appreciate Heller’s pithy rhetorical style (I do), but dead combatants are not injured? Well, something killed them (a mortal injury). Put another way, Heller has a considerable burden to show that “injury” does not include this subset (and most severe form) of injury. Second, Heller’s vision of the law would produce absurd results.  Let’s consider an example under his conception of Article 35: an attacking soldier in a wanton and cruel way shoots an enemy fighter 5 times, even though the latter could have easily been apprehended without any real risk to the attacking party. In one scenario, the enemy fighter somehow survives; and in another scenario, he dies. In Heller’s conception of the regime, only the former involves an unlawful act, and the latter--because the target succumbed to his injuries—does not. That is an absurd result. III. ICRC Commentaries to the 1977 Protocols Heller contends that I have “misread” the ICRC Commentaries to the 1977 Protocol, and that the Commentaries fail to support the conclusion that “superfluous injury” includes the unnecessary loss of life. First, Heller (and Ohlin) ignores the strongest and most direct evidence that I present from the Commentaries. And he does not attempt any explanation or reconcile his vision of the law with this contrary evidence. For example, as I explain in my article and in both of my responses on Lawfare to Corn et al., the ICRC Commentaries directly to Article 35 itself, states:

“the right to apply that amount and kind of force which is necessary to compel the submission of the enemy with the least possible expenditure of time, life and money. … [I]t should be quite clear that the requirement as to the minimum loss of life … refers not only to the assailant, but also to the party attacked. If this were not the case, the description would be completely inadequate.” (emphasis added)

In other words, the Commentaries on the very article that Heller claims has nothing to do with preventing unnecessary deaths, in fact, state that a component of the rule involves “the minimum loss of life” of “the party attacked.” Additionally, in my EJIL article (and in both of my responses to Corn et al.), I also explain that the Commentaries clearly discuss an obligation not to use lethal force in specific situations. The situations include use of force against a person who is no longer hors de combat because the individual tries to escape and a person who is no longer hors de combat because the individual resumes combat. Recall that in the former situation the Commentaries state: “It is only permissible to kill” the person “if there is no other way” of preventing his escape. And, in the latter case, the Commentaries state: “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.” These passages clearly and directly indicate that, at least according to the ICRC Commentaries, Article 35(2) applies to unnecessary killing. In the first two rounds of discussion on Lawfare, I highlighted the fact that Corn et al. also conspicuously failed to mention my analysis of these parts of the Commentaries—especially since these passages  so directly contradict their criticisms. Heller’s (and Ohlin’s) reply continues to avoid discussion of this evidence. Second, Heller’s claim that I “misread” the Commentaries is based, in part, on my quoting the Commentaries’ statement that the final text of Article 35 “corresponds to the position of the ICRC and to the intent of the original rule.” Heller claims that I overlook a footnote following the phrase “the position of the ICRC.” Two points about Heller’s argumentation here. First, Heller’s analysis borders on an explicit contradiction of itself. That is, Heller proceeds to quote the following words showing that the ICRC’s draft text prohibited “methods and means which uselessly aggravate the sufferings of disabled adversaries or render their death inevitable in all circumstances.” Remember Heller agrees that this ICRC draft language corresponds to the meaning of the final text. Yet, this very language includes a reference to “death.” How then does Article 35 itself does not include “death” as a form of “superfluous injury or unnecessary suffering”? Secondly, the issue is not just the position of the ICRC but also “the intent of the original rule.” The scope of Article should thus be interpreted with reference to both. And, here’s the important point: in the early 1970s, when the ICRC submitted its report to an intergovernmental meeting, the ICRC explained the intent of the original rule in these terms:

“[R]ecourse to force must never be an end in itself. It will consist in employing the constraint necessary to obtain that result. Any violence reaching beyond this aim would prove useless and cruel. The principle of humanity enjoins that capture is to be preferred to wounding, and wounding to killing ….”

This passage shows that the intent of the original rule includes the principle that “capture is to be preferred to wounding, and wounding to killing.” Yet Heller seems to agree with me that Article 35 “corresponds to the position of the ICRC and to the intent of the original rule” Heller’s own reliance on the Commentaries is flawed. First, he seizes on the following statement in the Commentaries: “It is clear that in the eyes of the victim all suffering is superfluous and any injury is unnecessary.” Why is that relevant? Heller thinks the implication is obvious. He states, in full: “That sentence provides further support for the idea that the ICRC views Art. 35(2) as limited to combatants who survive an attack; dead combatants have no feelings.” Okay, but in the eyes of the victim, being dead is an unnecessary injury. Who would think otherwise? Second, Heller seizes on a part of the Commentaries that discusses the difficulty of setting objective medical parameters, such as pain thresholds, for suffering. He then explains that death is not a condition that would give rise to such a measurement problem. But, neither is something like having acid poured on one’s face—it does not mean that such an injury is beyond the scope of the prohibition because its harm is so obvious and measurable. Heller also presents the following proof that my argument is “irreconcilable” with the Commentaries: “Being dead is regrettable for the combatant, but it is not a medical issue.” Being dead is not a medical issue? What is it then? Indeed, the government conference in which these medical issues arose is replete with discussions of total casualty rates (including deaths) with different types of weapons. Furthermore, it is well known that the ICRC’s SIrUS project in the 1990s involved a study of the medical effects of weaponry—indeed, Judith Gardam explains that the SIrUS project “mirrors th[e] approach” of the medical expert group that Heller quotes. And the SIrUS medical study notably included battlefield and hospital mortality rates as an indicator of methods and means of warfare that violate Article 35. The SIrUs project was aborted for other reasons—the mortality rates, for example, were unreliable because of their dependence on available medical care and technology. Nevertheless, the idea that unusually high mortality rates—if reliably measured—constitute an important indicator of superfluous injury or unnecessary suffering has long been accepted. And the notion that mortality is a medical issue was well accepted. Greenwood explained: “The identification of these criteria [e.g., “field mortality” and “hospital mortality” rates] and the medical study on which they are based is of considerable value in helping to show how the balancing act required by the unnecessary suffering principle can be made more precise and less anecdotal than at present.” Finally, there is an enormous part of the history that Heller leaves untouched. In my EJIL article, and in my second reply to Corn et al., I present a plethora of statements by governments during the 1970s conferences expressly supporting a “capture versus kill” approach and the prevention of unnecessary killing. These statements include key players in drafting the 1977 Protocol. Heller not only says nothing about the evidence of the negotiating history that I present. He presents no contrary evidence along that dimension. At the very least, our assessment of of the ICRC Commentaries should be evaluated in light of these parts of the negotiating history itself.

* *  *

Mine is not such a novel claim. It is consistent with a long line of some of the most highly respected law of war experts who reached the same conclusion on this issue. The Law of the Protocol prohibits manifestly unnecessary killing as a form of “superfluous injury or unnecessary suffering” in important circumstances. That conclusion, I submit, is based on a straightforward reading of the treaty text, and a close reading of the Commentaries and the travaux préparatoires. I am grateful to Lawfare and Heller for this valuable discussion. I will respond to Jens Ohlin’s other important criticisms separately.