Editor’s Note: Just Security is holding a “mini forum” on the new Defense Department Law of War Manual. This series includes posts from Sean Watts, Eric Jensen, and others.
The Defense Department seems to think that it may kill civilians if there is any risk that the precautions necessary to avoid killing them might prove militarily disadvantageous. The DOD’s new Law of War Manual recognizes that “[c]ombatants must take feasible precautions in conducting attacks to reduce the risk of harm to civilians.” Such precautions may include giving effective advance warning before an attack, adjusting the timing of attacks, or selecting weapons that lower the risk of incidental harm. However, the manual declares that “if a commander determines that taking a precaution would result in operational risk (i.e., a risk of failing to accomplish the mission) or an increased risk of harm to their own forces, then the precaution would not be feasible and would not be required.” Apparently, the DOD thinks that “feasible” means “risk-free.”
In my view, the DOD’s position on precautions in an attack — like its position on target selection and killing human shields — is both legally incorrect and morally unsound. The stakes are high. While the proportionality rule sets an upper limit on the harm that attackers may inflict on civilians in pursuit of some military advantage, the precautions rule requires attackers to inflict even less harm on civilians whenever feasible. Indeed, since proportionality is a notoriously imprecise standard, the precautions rule may provide civilians with more effective protection.
Under customary international law, the precautions rule governs both the verification of targets for attack and the selection of weapons and tactics used to conduct attacks. Attacking forces “must do everything feasible to verify that targets are military objectives” in order to avoid mistakenly targeting civilians and civilian property. In addition, attacking forces “must take all feasible precautions in the choice of means and methods of warfare with a view to avoiding, and in any event to minimizing,” harm to civilians. Remarkably, the manual never mentions the precautionary obligation to do everything feasible to verify targets — not even to reject it — and endorses the precautionary obligation to choose weapons and tactics that minimize risk to civilians in a footnote.
Under international law, feasible precautions are “those precautions which are practicable or practically possible, taking into account all circumstances ruling at the time, including humanitarian and military considerations.” On any fair interpretation, this rule “must be interpreted by balancing” the humanitarian considerations in favor of taking a precaution against the military considerations against taking that precaution. If the former outweigh the latter, then the precaution is required. Conversely, if the latter outweigh the former, then the precaution is not required.
In contrast, according to the manual, humanitarian considerations require a precaution only if there are no countervailing military considerations. Conversely, military considerations always permit foregoing a precaution irrespective of countervailing humanitarian considerations. No balancing is necessary because military considerations necessarily prevail over humanitarian considerations should a conflict arise — as one almost always does.
Strikingly, the Joint Chiefs of Staff took a different — and better — view when interpreting the precautionary obligation to “give effective advance warning of attacks which may affect the civilian population, unless circumstances do not permit.” According to the Joint Chiefs, “circumstances permit” advance warning when “any degradation in attack effectiveness is outweighed by the reduction in collateral damage [e.g.,] because advanced warning allowed the adversary to get civilians out of the target area.” On this sensible view, attackers must weigh the military reasons to attack without warning against the humanitarian reasons to give advance warning. Sometimes the former will outweigh the latter, but sometimes the latter will outweigh the former. Unfortunately, the manual departs from this sensible view.
According to the manual, a precaution is not required even if the risks to civilians that the precaution will avoid far outweigh the risks to attackers that the precaution will create. The manual’s only support for its position is a 1991 Army statement that “‘[f]easible precautions’ are reasonable precautions, consistent with mission accomplishment and allowable risk to attacking forces.”
However, as Michael Schmitt observes, “[i]t is reasonable to require military forces to assume some degree of risk to avoid collateral damage and incidental injury. They do so regularly.” Conversely, it is unreasonable to forego precautions that are much more likely to avoid harm to civilians than to either endanger attacking forces or compromise their mission. Similarly, small risks to attacking forces are not only “allowable” but required when they avoid large risks to civilians.
As A.P.V. Rogers writes, “[m]ilitary necessity cannot always override humanity. In taking care to protect civilians, soldiers must accept some element of risk to themselves.” Rogers quotes British Defense Doctrine, which stated that
If there is a choice of weapons or methods of attack available, a commander should select those which are most likely to avoid, or at least minimize, incidental civilian casualties or damage. However, he is entitled to take into account factors such as his stocks of different weapons and likely future demands, the timeliness of attack and risks to his own forces. Nevertheless, there may be occasions when a commander will have to accept a higher level of risk to his own forces in order to avoid or reduce collateral damage to the enemy’s civil population.
Similarly, the Group of Experts responsible for the HPCR Manual conclude that “whereas a particular course of action may be considered non-feasible due to military considerations (such as excessive risks to aircraft and their crews), some risks have to be accepted in light of humanitarian considerations.” Unfortunately, the DOD Law of War Manual contains no such qualifications.
To see the implications of the DOD’s position, consider the following scenarios involving the choice of means and methods of warfare (for discussion of target verification, see Rogers):
Resist: Counterinsurgents use thermal imaging to determine that an insurgent leader is asleep in his home with his five young children. The commander determines that a missile strike on the house will certainly kill the leader, as well as all five children, at no risk to attacking forces. The commander also determines that a night raid by special forces will certainly kill the leader and kill none of the children, but that there is a small chance that the leader will immediately wake up, spring to his feet, and try to attack one of the special forces operators before the other operators kill the leader.
Escape: Same as Resist, except that the commander determines that a night raid by special forces will certainly kill the leader and kill none of the children, but that there is a small chance that the leader will immediately wake up, jump out the window, evade the special forces operators stationed outside, and escape.
Let us stipulate, for the sake of argument, that it would not be unlawfully disproportionate to collaterally kill the five children in order to eliminate the insurgent leader. Of course, the counterinsurgents can avoid killing any children by choosing weapons and tactics that involve either a small risk of harm to their own forces or a small risk of mission failure. However, according to the manual, the commander may order the missile strike, certainly killing five children, rather than accept such small risks. Indeed, according to the manual, the commander need not even weigh the certain deaths of the children against the small chance of resistance or escape. This rationale is almost as bad as the result. Almost.
Unfortunately, states sometimes appear to give the safety of their armed forces absolute priority over the safety of defenseless civilians. I fear that the DOD’s position will encourage this lawless and immoral trend. In ordinary morality, I may not take a large risk of harming you in order to avoid a small risk of being harmed myself. In my view, the morality of war is no different.
As Rogers notes, “No-risk warfare is unheard of.” In the words of one general, “[y]ou know, to be brutally frank about it, if your overriding objective is to protect your own force, then you probably should not have deployed in the first place, because the only way to avoid risk to your forces is not to get involved.”
I encourage readers who reject the manual’s position on precautions, target selection, or killing involuntary human shields, to respond to the DOD’s invitation and contact them at osd.pentagon.ogc.mbx.ia-law-of-war-manual-comments@mail.mil.