In a recent post, Marty Lederman echoed criticisms previously leveled by Oona Hathaway about the US DOD’s Law of War Manual. The thrust of their criticism is that several Manual provisions improperly state customary international law regarding the protection to be afforded to civilians. I will not address that aspect of their critique here. Instead, I want to focus on Marty’s inclusion of Manual provisions dealing with military medical and religious personnel, because I do not believe those particular objections have merit.
Marty identifies the Manual’s statement in section 220.127.116.11 that “medical and religious personnel” accept the risk of death or injury and therefore “need not be considered” in the proportionality analysis as somehow misstating the law. I disagree, at least with respect to military medical and religious personnel — that is, uniformed doctors, nurses, and chaplains who are a part of the armed forces and near a military objective or a military unit that may lawfully be targeted. It is true that a military doctor, in a known or marked military hospital or battlefield aid station where she is performing medical duties, may not be targeted as such. However, based on a plain reading of Additional Protocol I and the text of the vast majority of military manuals surveyed by the ICRC, a military doctor in a military headquarters, or in the field with a ground combat force, absolutely need not be accounted for as collateral damage in a proportionality analysis when that headquarters or unit is targeted. The proportionality rule as expressed in AP I Art. 57 is quite clearly limited to protecting civilians, and military personnel — even non-combatants like doctors and chaplains — are by definition not civilians. To be fair to Marty and Oona, as I explain below, the Manual could have perhaps been clearer about the differences between military and civilian personnel in this context. But any reader of the DOD Manual is by now familiar with its sometimes frustrating tendency to cross-reference itself, circling back (or jumping forward) to other provisions in order to paint a complete picture of any given position, and I think a close reading of the entirety of the Manual’s provisions on medical personnel would have made it clear that the points they object to largely apply to military personnel.
The precise status of military medical and religious personnel is often complicated, at least on the surface. Most lawyers are accustomed to making a binary status choice between “civilian” and “combatant.” But the truth is, military medical and religious personnel do not fit neatly into either category; they are certainly not combatants, but they are just as certainly not “civilians.” Both the Hague Regulations (here and here) and AP I implicitly recognize this “non-civilian but non-combatant” status. Art. 3 of the Hague Regulations acknowledges that “[t]he armed forces of the belligerent parties may consist of combatants and non-combatants.” Art. 43(2) of AP I begins to show us what this “non-combatant” membership in the armed forces might mean: “members of the armed forces of a Party to a conflict (other than medical personnel and chaplains covered by [Art. 33 of the Third Geneva Convention (GC III)]) are combatants, that is to say, they have the right to participate directly in hostilities.” More importantly, the very definition of “civilian” in Art. 50 of AP I refers back to Art. 43: “A civilian is any person who does not belong to one of the categories of persons referred to in Article 4(A)(1)(2)(3) and (6) of [GC III] and in Article 43 of this Protocol.” As noted, Art. 43 says military medical and religious personnel are not combatants, but also that they are members of the armed forces of a Party. They thus “belong to one of the categories of persons” that are not civilians.
Furthermore, even the ICRC Customary International Law Study recognizes that “non-combatant” and “civilian” have different meanings. Its interpretation of Rule 3 specifically notes that “[n]on-combatant members of the armed forces are not to be confused, however, with civilians accompanying armed forces who are not members of the armed forces by definition.”
The DOD Law of War Manual clearly adopts this position at 4.9, noting that military medical and religious personnel “are treated as a special category under the law of war,” and explaining at 4.10 that “military medical and religious personnel are members of … the armed forces,” but are “afforded special privileges so that they may fulfill their humanitarian duties.” 4.10.1 then goes on to state the proposition that Marty and Oona object to: while they “may not be made the object of attack,” such personnel also “must accept the risks of incidental harm from military operations.” It is worth noting here that this is not a new position for the United States — the 1956 Army Field Manual 27-10 took the exact same view: “The accidental killing or wounding of such personnel, due to their presence among or in proximity to combatant elements actually engaged … gives no just cause for complaint” (para. 225).
Medical and religious personnel who are members of the armed forces are decidedly not combatants, but nor are they civilians. The former fact gives rise to the prohibition on directly targeting them, but the latter means that their injury or death, proximate to an attack on a lawful military objective, does not constitute the “incidental loss of civilian life” that the rule of precautions as expressed in AP I requires the attacker to avoid, minimize, and consider when assessing proportionality.
The rule in Art. 25 of the First Geneva Convention is clear: Medical and religious personnel in the armed forces must be respected and protected; they may not be intentionally attacked. But that does not mean they become civilians, and therefore subject to the proportionality rule. There are practical consequences to holding otherwise that make it clear why this must be so. If one attacks an infantry formation, and military medical personnel are travelling with that formation, their presence poses no proportionality concern to the attacker, does it? They must clearly be respected and protected — not directly targeted, and not prevented from performing their humanitarian function, even in captivity. But must a commander conduct a proportionality analysis? How would this work? Is he entitled, or required, to consider enemy military doctrine to draw conclusions about the probable density of medical or religious personnel in a formation of any given size? From standoff distance, a column of uniformed dismounted troops bearing small arms — which could easily include medics or doctors — looks like … a column of uniformed dismounted troops bearing small arms. Absent civilians in that column, why must a commander conduct a proportionality calculation?
A similar approach can be applied to Marty’s objection to the same language in the DOD Manual regarding military medical units and facilities, at section 18.104.22.168. Again, these facilities clearly cannot be directly attacked if their status as such is known and they are not being misused. Fixed medical facilities are indeed afforded special protection; even if misused for prohibited military purposes, they may not be attacked without a warning and time to comply (AP I, Art. 13). But they are not “civilian objects” requiring a proportionality analysis under the strict terms of Art. 57.
No responsible military commander committed to adhering to the law of armed conflict would intentionally direct an attack at a known military hospital, and at least in the US, our collateral damage estimation process requires us to consider not only civilians, but also other specially protected persons or objects. But that collateral damage methodology is a process, which is not the same thing as a position on the law.
The question posed here is whether, as a matter of international law, a commander is required to conduct a proportionality analysis when targeting a military headquarters which may, hypothetically, be located right next to a military hospital tent. Clearly, the defender in that scenario has an affirmative obligation not to so locate his military hospital (GC I, Art. 19). But in order to effectively provide urgent trauma care to military forces, military medical facilities must be located reasonably near the forces they serve — a conundrum, perhaps, but one whose only resolution is that such medical personnel and facilities do in fact accept some risk of incidental damage, destruction, and death. The willingness of States to accept such risks is ostensibly why these personnel are part of the armed forces in the first place — so that they can be commanded into areas and situations of great danger.
Again, the points I make above are limited to military medical and religious personnel and military medical facilities. I note this because the Manual’s provisions on the respect and protection to be afforded to medical and religious personnel specifically includes authorized staff (presumably civilian) of voluntary aid societies as among those so protected. When it then caveats this protection by declining to extend the proportionality rule to medical and religious personnel, it fails to make clear that the caveat only applies to military personnel. In my opinion it certainly should have done so; the drafters saw fit to qualify the 22.214.171.124 provision on facilities in precisely that way (7.10 relates to “Military Medical Units and Facilities”), and it is not clear to me why they failed to do the same for personnel. It ought to have been made clear that there is a difference between the Médicins Sans Frontièries (MSF) facility in Kunduz and a military medical unit that is part of an enemy armed force. So in that sense, 126.96.36.199 is perhaps poorly worded.
Civilian members of voluntary aid societies remain civilians and remain protected as civilians, of course — provided that they are not subjected to military laws and regulations, in which case (as the Manual notes at 4.11) they are “placed on the same footing” as military medical personnel (citing GC I, Art. 26). And if that is what the drafters intended to convey by this section, then perhaps the Manual is not quite so poorly worded on this point after all. Just as there is a difference between military medical personnel and MSF, for example, there is a difference between MSF (completely neutral and not affiliated with any State) and a national aid society that has been mobilized in support of the armed forces and made subject to military discipline. When that line is crossed, voluntary aid societies are fairly exposed to the risks posed by combat and “have no just cause for complaint” if they suffer damage or death as a result of an attack on a lawful military objective.
Undoubtedly, medical and religious personnel, as well as medical units and facilities, are entitled to special protection under the law of armed conflict. But with respect to Marty, the customary law (and the treaty law) regarding military medical and religious non-combatants is different than that regarding civilians, and I do not think this particular criticism of the Manual is an entirely valid one. To be sure, as Sean Watts noted recently, the Manual is quite often engaged in a bit of “bootstrapping,” and on these points it would have benefited by citing to the plain language of the proportionality rule in order to make its point.
This is particularly the case as that language strongly supports the position taken by the Manual. At a minimum, the burden would seem to be on the critic to explain why a rule (like Art. 57 of AP I) that expressly applies to “civilians and civilian objects” must be read to include other (non-civilian) protected persons and property, particularly when Art. 57’s own terms indicate that the drafters were cognizant of the difference. Art. 57(2)(a)(i), which relates to target verification, clearly includes both civilians/civilian objects and those persons or objects “subject to special protection.” The remainder of the Art. 57 precautions do not. It seems to me that ordinary principles of construction warrant concluding that the drafters intended to exclude other protected persons and objects from the proportionality rule. In the context of distinction, the inclusion of the specially protected makes perfect sense — in that of proportionality, in my view, it does not.
This views expressed in this article are those of the author in his personal capacity and do not necessarily represent those of the reflect those of the US Department of Defense.