Civilian Harm Mitigation and Response Action Plan: Observations from a Combat Training Center
The August 25, 2022 publication of the Department of Defense Civilian Harm Mitigation and Response Action Plan (CHMR-AP) generated debate about whether it is capable of being applied by the DoD across the full continuum of conflict. That continuum ranges from counter insurgency (COIN) and counter-terrorist (CT) operations to large scale combat operations (LSCO). LSCO, which consist “of fighting a near-peer threat on a division and corps level under rapidly changing conditions ” are more intense and destructive than COIN. In LSCO, avoiding population centers, while ideal, is difficult due to the strategic and tactical value of urban terrain. The Army’s focus in LSCO is to defeat the enemy in ground combat.
A recent article suggested that four objectives in the CHMR-AP do not offer adequate legal maneuver space to commanders, particularly in LSCO. This view draws heavily on the idea that military and policy makers suffer from a COIN “hangover,” a concept elaborated in “The Eighteenth Gap,” by Lt. Gen. Charles N. Pede and Col. Pete Hayden. The “COIN hangover” suggests that commanders and staffs hesitate to fully employ military capabilities to achieve military objectives, based on flawed legal perceptions and their experience in Afghanistan and Iraq, because of concern with causing civilian harm. These articles advocate that the Army better train to achieve the military objective and not overstate or misapply law of armed conflict (LOAC) during LSCO.
That view is informed more fundamentally by the Army’s focus on building readiness to conduct LSCO – not COIN – as directed through the 2022 National Security Strategy, the 2022 National Defense Strategy, and in Army doctrine for conducting operations. Additionally, concerns with maintaining “legal maneuver space” stress the importance of acknowledging what the LOAC does not prohibit, as opposed to the situation-specific policy restrictions U.S. forces had the luxury to implement in slower-paced COIN and CT operations against numerically and technically inferior adversaries.
Other articles champion the CHMR-AP and dispute the concerns raised above by highlighting the legal requirement to implement all feasible precautions in attacks. They emphasize that the law, and by implication the CHMR-AP, adequately accounts for the strategic and operational contexts of LSCO through the feasibility standard. Commanders need only take precautions that are practically possible. This view also notes that feasible precautions must be conducted across the spectrum of conflict, to include LSCO against near-peer or peer adversaries.
This post provides concrete examples, observed at an Army combat training center (CTC), that illustrate the difficulty the CHMR-AP may encounter at the LSCO end of the spectrum of conflict. At the CTC, we routinely encounter commanders and staffs that excessively restrict their operations with respect to civilian harm protection to the detriment of their force and the mission. In short, the “COIN hangover” is a real phenomenon. This is not to say that the CHMR-AP is not a worthy endeavor—to be clear, the vast majority of the CHMR-AP is likely to yield extremely beneficial results to the operations process and strategic objectives of the United States. However, two CHMR-AP objectives – namely Objective 3, Incorporate Guidance for Addressing Civilian Harm Across the Full Spectrum of Operations and Objective 5, Incorporate Measures to Mitigate Risk of Target Misidentification – have strong potential to reinforce the COIN conflation of civilian harm prevention and civilian harm mitigation.
Practical Implementation in Training Exercises
The Joint Multinational Readiness Center (JMRC), based at Hohenfels, Germany, is the Army’s Europe-based CTC. JMRC trains leaders, staffs, and units at the brigade combat team (BCT) level to fight and win through multi-domain operations in LSCO. A typical BCT consists of approximately 4,400 soldiers, with three infantry or combined arms battalions, one cavalry squadron, one field artillery battalion, one engineer battalion, and one support battalion. Each BCT also contains a Brigade Legal Section (BLS), led by the Brigade Judge Advocate (BJA). Among numerous other duties, the BJA advises the commander and staff on the application of the LOAC and rules of engagement (ROE).
Currently, the greatest challenge BJAs face at CTCs is countering commanders’ and staffs’ (strongly held) beliefs, based on twenty years of COIN operations, as to what the law requires during combat operations. These beliefs persist even after coaching on the strategic and contextual differences between COIN operations and LSCO. The context does indeed matter when determining what precautions are feasible in the attack, but the education and past practice of leaders responsible for the execution of the attack distorts the operational assessment of what precautions are feasible in LSCO. Frequently, during training for LSCO, commanders and staffs believe the legal requirement for feasible precautions in the attack requires foregoing a military advantage when the operation cannot avoid civilian harm.
As Lt. Gen. Pede pointed out, “ROE can also create habits on one battlefield that may get you killed on another.” This comment is often substantiated during exercises at CTCs where staffs and commanders confuse policy-driven ROE restrictions from COIN operations with the baseline legal requirements applicable in any conflict. For example, the training scenario at JMRC routinely presents the dilemma of an enemy conducting indirect fire from a positional artillery area, then displacing from that area into a civilian population. The enemy tactic is intended to exemplify and test responses to lawfare (using your adversaries’ compliance with the law as a weapon against them). It often disrupts the brigade’s efforts to effectively target the artillery formation responsible for destroying friendly forces as the brigade staff spends precious minutes debating whether or not they can lawfully attack the enemy artillery. That debate often focuses on the mistaken belief from experience in COIN that causing civilian harm is per se unlawful – an understanding that is as wrong in COIN as it is in LSCO. As the Articles of War audience understands, the law does not prohibit firing into populated areas; it requires consideration and implementation of feasible precautions to prevent or mitigate civilian harm. The time commanders and their staffs deliberate on this fact, however, can make the difference between life and death for the friendly unit.
CHMR-AP Objective 3, Incorporate Guidance for Addressing Civilian Harm Across the Full Spectrum of Operations
The stated goal of CHMR-AP Objective 3 is to “describe the importance of the civilian environment and address its significance as a component of the operational environment.” A recent article described this objective as “prioritization of civilian protection during the planning and conduct of military operations.” The author explains, “DoD intends to achieve this by elevating the consideration of the civilian environment – including the civilian population, and the personnel, organizations, resources, infrastructure, essential services, and systems on which civilian life depends – as an integral component of the operational environment.” While the goal of Objective 3 is commendable, misinterpretation threatens to reinforce the mistaken belief that prioritizing – that is, giving preeminent consideration to civilian protection – is required, even over mission accomplishment or unit defense. Based on CTC experience, many commanders already prioritize civilian protection to the detriment of their unit’s successful, and lawful, accomplishment of the military mission. What commanders need from a civilian protection policy is clarity on how to prioritize defeating the enemy while minimizing collateral harm to civilians. Commanders need confidence that U.S. policies and civilian harm prevention bureaucracy will underwrite mission accomplishment.
For example, in the typical CTC exercise scenario, the enemy often establishes a defensive strongpoint in a heavily populated civilian area. The strongpoint consists of one or two companies of a mixed infantry and armor formation – a force of a couple hundred enemy soldiers and dozens of armored vehicles. The friendly BCT receives an order to seize a bridge near this city. Inevitably, as the BCT advances, the enemy engages the BCT with direct and indirect fires. In response, the BCT usually decides, whether through prior planning or dynamic planning, to forego the use of indirect fires (artillery, rockets, and mortars) to prepare or shape the objective as the BCT attacks the strongpoint. The BCT declines to use indirect fires out of a concern that the large caliber explosives will cause civilian harm, and that such harm is unlawful. Rather than employ these lawful capabilities to attrit the enemy before entering the city, the BCT decides to engage in the most resource intensive and highest risk operation – a building-by-building fight with dismounted forces using small arms weapons. By doctrine, defeating this strongpoint in a populated area requires the commitment of two to four battalions of combat power – one to two thousand soldiers and hundreds of vehicles or more. Instead, the BCT chooses to fight from a position of weakness to avoid even the potential for civilian casualties – and feared criticism from higher headquarters or the media.
As another example, during an exercise, friendly units acquired enemy artillery fire using radar, but are usually unable to return fire before the enemy unit changes position. The friendly units will follow the enemy using low-fidelity sensors. By common practice, the enemy artillery moves into a populated area, and the friendly unit will not even consider attacking the enemy for fear of causing civilian harm.
In both examples, because of hypersensitivity to harming civilians, the staffs do not develop intelligence and information on the population centers to inform the commander’s judgment as to whether the incidental civilian harm is excessive in relation to the concrete military advantage to be gained – the legal requirement for proportionality. There is no analysis on civilian structures or how many civilians may be in the area. The ingrained fear that any civilian harm is unlawful prevents the staff from analyzing what precautions may be feasible. Even when facing mission failure, the staff avoids using lawful force because of the expectation of civilian casualties.
After coaching one commander on how he could lawfully employ indirect fire against the enemy in a populated area, the commander remarked “you want me to shoot into the area so you can investigate me?” This statement perfectly captures the essence of our observations at the JMRC. Many leaders instinctively consider civil harm prevention as the highest priority when determining whether target engagement is authorized. Today’s U.S. commanders and their staffs believe the civilian harm outcome – the number of civilian casualties as a result of the strike – is determinative when analyzing the lawfulness of a proposed strike, regardless of the military advantage to be gained. Many staffs do not even present a target to a commander for a proportionality assessment when the staff believes any civilian harm will result.
Commanders and staff instinctively prioritize minimizing civilian harm when conducting military operations, and misinterpretation of CHMR-AP Objective 3 risks further ingraining the false notion that civilian harm outcomes are the determinative factor when conducting a proportionality analysis.
CHMR-AP Objective 5, Incorporate Measures to Mitigate Risk of Target Misidentification
At first blush, CHMR-AP Objective 5 captures lessons learned from the past twenty years of COIN. That is, “misidentification, including misinterpretation and mischaracterization, can be a frequent cause of civilian harm.” This objective tasks Combatant Commands with further restricting positive identification (PID) policies by mitigating cognitive bias with an appropriate level of certainty. The policy further discusses incorporating “red teaming” techniques into the targeting procedures as a method to mitigate cognitive bias. The U.S. Army has defined “Red Teaming” as
a flexible cognitive approach to thinking and planning that is specifically tailored to each organization and each situation . . . It uses structured tools and techniques to help us ask better questions, challenge explicit and implicit assumptions, expose information we might otherwise have missed, and develop alternatives we might not have realized exist.
While CHMR-AP Objective 5 points out that red teaming below the combatant command level would be ad-hoc and as needed, policymakers must resist the urge to mandate procedures feasible in COIN and apply those procedures across the spectrum of military operations.
No doubt, red teaming, PID, and other efforts to avoid target misidentification are worthwhile, but policy makers should be mindful of realities on the ground they contest. Commanders at the tactical level – “the level of warfare where battles and engagements are planned and executed” – do not own or control many, if any, high-fidelity sensors. Furthermore, most tactical commanders do not have the personnel or time to conduct extensive red teaming processes. And, if engaged in LSCO, the commander will almost certainly be overwhelmed by simultaneous operational requirements. Policymakers must understand that, in perhaps a majority of situations, commanders cannot be reasonably expected to conduct extensive red-teaming.
Returning to the earlier example of indirect fire in a populated area, the BCT did not have access to drones with full-motion video to allow the commander and his staff to “positively identify” the target because weather conditions prevented the BCT’s drone from flying. However, the BCT did acquire the target location when they detected the reverse trajectory of enemy rounds using counterfire radar, then followed the enemy artillery battery with moving target indicators. While useful and in fact reliable, this type of intelligence has not, however, met the threshold required by Combatant Command PID policies in the COIN operations of the past twenty years.
“Positive Identification,” or “PID” is a heightened requirement enshrined in the policy on collateral damage methodology. While PID requirements for pre-planned targets are commendable as feasible precautions, they do not reflect what is required under the law to engage the enemy. The law requires commanders to determine in good faith that a target is a military objective based on the information reasonably available at the time. The distinction is important, and takes on greater importance when CTC observations suggest that many staffs view PID, defined as “a reasonable certainty that a target is a military object,” as requiring full-motion video from a drone feed or “eyes-on” the target. While the PID requirement, as interpreted and applied in the past, is sound in the context of COIN and CT operations where the focus is on eliminating civilian casualties, it is a policy constraint that may result in operational failures where other factors such as poor weather, limited resources, and adversarial electronic and cyber effects limit or prevent a unit from employing highly technical capabilities to gain “positive identification” of a target.
The risk for operational failure through red teaming and PID is elevated in LSCO when the cost of waiting to gain “positive identification” permits the enemy artillery unit to maneuver to a new position and fire even more artillery rounds on a friendly target, all while U.S. forces are paralyzed because of futile attempts to gain “positive identification” of the enemy forces. BCTs often wait to gain “PID” even when the staff has sufficient information to provide a good faith assessment the enemy is in a certain location. Instead, the BCT attempts to improve certainty on target identity and location to prevent civilian harm. Put another way, the BCT’s effort to perfect the picture of collateral damage overcomes the analysis on whether the military advantage is not excessively outweighed by the anticipated collateral damage. By trying to perfect this picture, the friendly unit loses the advantage, and the enemy artillery unit maneuvers to a new position, free to fire even more artillery rounds on a friendly target.
In the example where the friendly unit used counterfire radar acquisition and moving target indicators, the staff did not ask the brigade commander to determine if the enemy artillery position was a military objective. The staff believed there was insufficient intelligence to be certain under PID requirements to calculate an exact location to deliver precision munitions on each individual artillery piece. In this situation, the staff frequently assesses that positive identification requirements and applying feasible precautions to mitigate civilian harm, requires waiting for drones to provide a precise location, or alternatively, waiting until the enemy artillery units move out of the populated area. During this time, friendly combat power is degraded and significant resources are expended in an effort to prevent civilian harm, even at the risk of mission failure. This example highlights how staffs hesitate to take lawful action to secure the military advantage while significant effort is undertaken to provide the near-perfect picture of information to prevent civilian harm.
The freethinking enemy opposing forces (OPFOR) assigned to JMRC routinely use populated areas to their advantage. The OPFOR recognize that “lawfare” and populated areas represent a significant tactical advantagebecause U.S. units will often refuse to fire into populated areas. The OPFOR is often able to execute multiple artillery missions, stage offensive operations, and control subordinate units from locations close to populated areas, allowing the OPFOR to inflict heavy losses on U.S. forces with little to no risk to their own combat power.
Conclusion.
By their very design, the training scenarios at JMRC present incredibly difficult dilemmas for commanders and their staffs. They demand thoughtful approaches and a nuanced understanding of LOAC. They also highlight the struggle of implementing civilian harm mitigation across the continuum of conflict. The context within which operational decisions are made does indeed matter, and the loss of U.S. combat power to a roadside bomb in a deliberate COIN operation is different than a constant barrage of artillery in a fast-moving LSCO. In other words, losing one or two vehicles and several killed or wounded soldiers from a roadside bomb, while devastating and destructive, is nowhere near as catastrophic as losing dozens of vehicles and hundreds of soldiers in a matter of minutes. While training at JMRC for LSCO, BCTs routinely lose most of their combat power because they are getting the context wrong – confusing the civilian harm mitigation policies of past conflicts for what the law requires – all in a well-intentioned, if mistaken, effort to avoid collateral damage when attempting to seize urban terrain.
The advocates informing the CHMR-AP debate generally derive their data from the last twenty years of military operations. The measures recommended are informed by lessons-learned in COIN operations, and are primarily a result of the investigation into the August 29, 2021, drone strike in Kabul, Afghanistan that resulted in the death of 10 Afghan civilians. Caution should be exercised in creating a policy that relies primarily on lessons from COIN and CT operations, and then making that policy broadly applicable, to include war with a peer threat. In COIN and CT operations, the United States demonstrated a great ability to mitigate civilian harm through feasible precautions such as high-fidelity sensors that surveilled a target location continuously over long time periods to develop a pattern of life analysis, and then deliver precision munitions to destroy the target with minimal civilian harm. In LSCO, policymakers should strive to provide clarity on prioritizing achieving the military objective over civilian harm mitigation while assuming that tactical units will not have the time and resources to red team or establish a heightened identification requirement when engaged with a peer competitor in close combat.
The debate described above is useful among lawyers, who presumably understand the nuance between law and policy, but the overarching concern is the CHMR-AP will reinforce erroneous beliefs held by commanders and staffs as to what is required by the LOAC. The Army needs our commanders and staffs to follow the LOAC, not revert to their experiences with restrictive COIN and CT civilian harm policies. As discussed above, mistaking old policies for law has had disastrous consequences for many units training for LSCO. Those policymakers charged with implementing the CHMR-AP must keep in mind that the continuing failure of commanders and their staffs to understand the difference between law and policy may be the difference between winning and losing the next conflict.
***
Major Jason Young is the senior observer, coach, and trainer at the Joint Multinational Readiness Center near Hohenfels, Germany.
Photo credit: U.S. Army via Wikimedia Commons
