transatlantic dialogue

Transatlantic Workshop on International Law and Armed Conflict: Wounded and Sick, Proportionality, and Armaments

By Geoffrey S. Corn
Tuesday, October 10, 2017, 5:28 PM

[Editor's note: This piece is the latest installment in a mutli-blog series building on the Fifth Annual Transatlantic Workshop on International Law and Armed Conflict, as explained in detail here.]

Transatlantic Workshop on International Law and Armed Conflict

Wounded and Sick, Proportionality, and Armaments

Geoffrey S. Corn

Protection of wounded and sick members of the armed forces, those exclusively engaged in their collection and care, facilities and equipment also exclusively engaged in these activities (what I will refer to herein as “military protected persons and equipment”) is at the very core of International humanitarian law, or the law of armed conflict. Ameliorating the suffering of these victims of war was, after all, the seed from which the entire Geneva tradition, to include the four Geneva Conventions and their three Additional Protocols blossomed. It is simply beyond question that these people, places, and things must be respected and protected during all armed conflicts.

But the 2016 International Committee of the Red Cross’ revised Commentary to the 1949 Wounded and Sick Convention has exposed an area of substantial uncertainty: does the law protect military protected persons and equipment from the collateral or incidental consequences of attacks on proximate lawful military objectives? And, if the answer is yes, what is the source of that protection and how does it apply in practice?

The Commentary provides an affirmative answer to this question, explaining that this protection derives from an a fortiori extension of the proportionality rule applicable to civilians and civilian property. While the Commentary acknowledges that the treaty enumeration of this proportionality rule in Additional Protocol I applies by its terms exclusively to civilians and civilian property, this is ultimately not considered a sufficient justification for excluding military protected personnel and equipment from the scope of the proportionality obligation. In contrast, the U.S. Department of Defense Law of War Manual expressly addresses this issue, and asserts that these protected military personnel and assets, while unquestionably protected against direct or deliberate attack, are not protected from incidental injury and/or collateral damage resulting from lawful attacks.

These two approaches to the question of whether incidental injury or collateral damage to military protected personnel and equipment highlight two almost undeniable conclusions: first, complete indifference to such attack effects is inconsistent with the humanitarian objectives of the law; second, the source of obligation to consider and, where appropriate, avoid such collateral effects is extremely uncertain. For the ICRC, the “greater” obligation to apply a proportionality rule for the purpose of protecting civilians and civilian property from such effects logically indicates an analogous obligation applicable to military personnel and equipment qualified for protection pursuant to the 1949 Geneva Convention I on the Wounded and Sick (GCI). But, as I argued at our workshop and in a forthcoming article, this extension of the proportionality obligation does not seem as logical as the Commentary asserts. First, as the Commentary notes, military protected persons and equipment were simply not included within the scope of the 1977 First Additional Protocol’s (AP I) provisions adopted to protect civilians from the effects of combat. Considering AP I also supplemented the protections provided by GCI, it is difficult to ignore the omission of military protected persons and equipment within the Protocol’s provisions. Second, and more problematically, the extension is operationally and tactically unworkable.

To illustrate the incompatibility of a wholesale extension of the civilian proportionality rule to military protected persons and equipment, I offered several hypotheticals. The first is most compelling. Imagine you are commanding forces that have just repulsed a combined arms enemy ground attack. The enemy is now withdrawing, and you observe what are obviously wounded enemy soldiers being loaded onto enemy combat vehicles. You fully anticipate the enemy to regroup in order to continue the offensive. These vehicles are not protected because they are not properly marked nor exclusively engaged in the collection and evacuation of the wounded and sick. Instead, the enemy is employing the common practice of evacuating wounded with any available combat vehicle. While this is occurring, other enemy forces are providing covering fires in support of the withdrawal. You have on-call close air support assets, and your air support coordination liaison asks if the enemy vehicles should be attacked? The enemy vehicles are lawful objects of the attack, but you know that the military wounded and sick must be respected and protected. It is therefore clear that an attack may not be directed against the wounded enemy soldiers. But the ICRC’s updated Commentary asserts that before launching the attack on the withdrawing enemy forces who are not hors de combat you must assess whether the risk created to the wounded enemy personnel is excessive in relation to the concrete and direct military advantage anticipated.

As has been noted by others, the Commentary approach would require combatants to engage in proportionality assessments in combat after the first salvo is fired against an enemy, because once that salvo is fired it is reasonable to assume that some of the enemy are hors de combat due to wounds. While it may be tempting to respond that in such a context, the military advantage will almost automatically outweigh the risk to the enemy wounded, this does not nullify the inconsistency with the Commentary position and tactical and operational practice. As the DOD Manual notes, the perhaps unfortunate but simple truth is that, “Combatants who are wounded, sick, or shipwrecked on the battlefield are deemed to have accepted the risk of death or further injury due to their proximity to military operations.” That risk almost inevitably includes the risk of incidental injury (or collateral damage to protected military medical equipment and facilities). Suggesting that such an obligation is logically inferred from the civilian proportionality rule is fundamentally flawed, because unlike military personnel, civilians (who do not take a direct part in hostilities) do not accept the risks of combat. Instead, civilians are the beneficiaries of a comprehensive legal architecture developed to protect them not only from deliberate attack, but also from the risks associated with proximity to hostilities.

Most of the experts assembled in Florence acknowledged this flaw in the Commentary’s logic. However, almost all the experts were of the view that the obligation to respect and protect military protected persons and equipment requires some effort to mitigate the risk of incidental injury and/or collateral damage when conducting an attack. On this point, I strongly agree: complete indifference to such risks seems incompatible with the underlying humanitarian foundation of the LOAC. In the article referenced above, I along with my co-author, Andrew Culliver (who will soon begin his career as a U.S. Army JAG officer), proposed that the Martens Clause may provide a credible basis for requiring consideration of feasible precautions to reduce this risk. I recognize that tethering the validity of such an obligation to the Martens Clause is not without its own weaknesses, most notably the longstanding uncertainty as to whether the Clause imposes substantive obligations beyond those established in treaty and customary international law. Other scholars are also offering an alternative approach to addressing this issue, most notably Aurel Sari and Kieran Tinkler, whose excellent article entitled “Collateral Damage and the Enemy” is forthcoming and will be available in draft form soon.

However, I continue to believe this approach has merit. Furthermore, since making this proposal, I have also become convinced that if protection for military protected persons and equipment from collateral risk is to be logically extrapolated from an existing treaty rule, a number of the feasible precautions (reflected in Article 57)  providea more compelling tether than the proportionality rule. Specifically, this extension would include the obligation to do all that is feasible to verify the nature of intended targets in order to implement the distinction obligation, which includes a prohibition against any deliberate attack directed against military protected persons and equipment.  And, where feasible, this obligation also requires commanders to issue warnings to enable the enemy to mitigate that risk; and, where feasible, employ methods or means of attack that mitigate that risk.

This appears to be the approach adopted by the DOD Manual, which continues the provision cited above with the following: “Although the presence of the wounded, sick, or shipwrecked on the battlefield does not serve to exempt military objectives from attack due to the risk that such personnel would be incidentally harmed, feasible precautions must be taken to reduce the risk of harm to the wounded, sick, or shipwrecked.” In support of this asserted precautions obligation, the Manual cites back to the section addressing the obligation to take feasible precautions to mitigate risk to civilians and civilian property, which in turn relies on Article 57 of AP I. This may seem somewhat inconsistent with the rejection of extending the proportionality rule of AP I to military protected persons and equipment, as a proportionality assessment is one of the precautions enumerated in Article 57. The Manual does not explicitly explain why it rejects the extension of one rule applicable by its terms only to civilians (i.e. proportionality), but then indicates that during attacks, feasible precautionary measures must be implemented to mitigate risk to military protected persons and equipment. There does, however, appear to be a plausible textual basis for this partial application of Article 57 precautions to military protected persons and equipment: the target verification precaution in Article 57 includes within its terms protection not only of civilians and civilian objects, but also people and places subject to, “special protection.” This “special protection” basis for requiring the target verification precaution, but not a proportionality precaution, is bolstered by the text of Article 57 (b), which requires an attack to be cancelled or suspended when it becomes apparent that an intended target is subject to “special protection”. The attack has to be cancelled or suspended based on a proportionality violation only when the attack will endanger civilians and/or civilian property, not military personnel or equipment subject to special protection

In my view, the DOD approach is not only based on a more logical treaty foundation than the Commentary approach, it is an especially important and positive development. This extension of the obligation to take feasible precautions to mitigate risk to military protected persons and equipment was notably absent in the original 2015 version of the same provision of the Manual (“Incidental Harm Not Prohibited).  The respect and protection due to the wounded, sick, and shipwrecked do not cover incidental damage or casualties due to proximity to military objectives or to a justifiable mistake.  Because combatants who are wounded, sick, or shipwrecked on the battlefield are deemed to have accepted the risk of death or further injury due to proximity to military operations, they need not be considered as incidental harm in assessing proportionality in conducting attacks.”). Why DOD added this obligation in the 2016 revision of the Manual is unclear, but from my perspective this strikes a logical balance between the risks these individuals and assets must accept as the result of their association with an armed force, and the interests of humanity that make complete indifference to this risk intolerable.

More importantly, the feasibility element of the precautionary measures approach provides a more rational answer to the dilemma highlighted in the example above than does extending the proportionality rule. In the midst of close combat, or after firing that first salvo, pausing or halting an attack on an enemy in order to mitigate risk to enemy wounded personnel and those tending to them would rarely be considered feasible. However, in many other tactical situations, risk mitigation precautions could be feasible. For example, it is quite common to co-locate medical aid stations and other forward collection and treatment facilities in military logistics clusters. This facilitates rapid reconstitution of resources with the backhaul of casualties. But this means that an attack on the logistics cluster jeopardizes inflicting collateral damage on military protected persons and equipment. In this context, precautions in relation to the means and methods of such attack could not only be feasibly implemented, but could mitigate this risk. Or, consider the equally common situation of non-protected enemy vehicles transporting wounded and sick enemy personnel. If a commander sought to attack the vehicles, it might often be feasible to wait for the wounded and sick to be offloaded at a medical treatment facility before conducting that attack.

In my view, there is another reason why the precautions approach adopted by the revised DOD Manual is more credible than the proportionality approach in the Commentary: the inclusion of a precautions type rule in GCI to protect military medical facilities. Article 21 of GCI provides that,

The protection to which fixed establishments and mobile medical units of the Medical Service are entitled shall not cease unless they are used to commit, outside their humanitarian duties, acts harmful to the enemy. Protection may, however, cease only after a due warning has been given, naming, in all appropriate cases, a reasonable time limit and after such warning has remained unheeded.

While not phrased in the express terms of a precautionary measure, this warning obligation expressly applicable to military medical facilities indicates that the Convention contemplated situations where subjecting the occupants of such facilities to  the risks of attack may be unavoidable. However, because it will almost always be the case that an attack in response to misuse of such a facility will subject the protected persons in the facility to incidental death and injury, the treaty imposed a precautionary warning obligation. This is notable for two reasons: first, it is a clear indication that the logical method to mitigate this risk was through the issuance of an appropriate warning; second, there is no indication that an attacking force must apply a proportionality assessment once it determines that attack is necessary because the enemy fails to cease the misuse after the warning.  

Ultimately, I don’t believe extension of the civilian proportionality rule is the ideal answer to the question of how to mitigate incidental and collateral risks to military protected persons and facilities, although I am not sure there is an ideal answer. But I do believe there is a better answer than what the Commentary asserts. I believe that the precautionary measures approach strikes a rational and necessary balance of interests that protects military protected persons and equipment from unnecessary risk of collateral effects of attacks without imposing an unrealistic constraint on tactical action. This approach is better aligned with the GCI rule related to the misuse of military medical facilities than the proportionality approach. I also believe it is better aligned with the instincts of military commanders. After all, why would such a commander forego the opportunity to reduce this risk when doing so in no way compromises or degrades his or her efforts to achieve military objectives and defeat enemy forces? In other words, the precautions approach makes sense, because as the tactically and operationally feasible opportunity to mitigate risk to enemy protected persons and equipment increases, the expectation a commander will integrate collateral risk mitigation measures into his or her attack decision should also be expected to increase.


Schedule of posts:

The joint blog series arising from the workshop follows on from our collaboration in hosting a similar series last year (see herehere and here). The Transatlantic Workshop is organized and sponsored by the Oxford Institute for Ethics, Law and Armed Conflict (directed by Dapo Akande), the Individualisation of War project, European University Institute, Florence (directed by Jennifer Welsh), the Washington DC & London delegations of International Committee of the Red Cross, the Houston College of Law (through the good offices of Geoff Corn), and the Robert S. Strauss Center for International Security and Law at the University of Texas (directed by Bobby Chesney).

  • Click here for the 2016 Series
  • Click here for the 2015 Series.
  • Click here for the 2014 Series.