Limiting Civilian Casualties as Part of a Winning Strategy: The Case of Courageous Restraint

How much risk combat troops must accept in order to avoid harming civilians has long been central to moral and legal arguments about just conduct during war, or jus in bello. In his seminal book Just and Unjust Wars, Michael Walzer argues that it is a state's duty to accept greater risks for its own military forces as a means to limit harm to noncombatants in the course of armed conflict. He provides a vignette from a World War I British soldier's memoir for context in supporting this assertion. In this particular incident, Walzer describes a dilemma faced by British troops as they attempt to clear a French town of German soldiers hiding among some of its dwellings. When entering a home, the British soldiers had the choice of whether or not to shout a warning before throwing a grenade down the cellar stairs. This warning would alert civilian noncombatants that may be hiding there and give them the opportunity to make the British soldiers poised to engage with lethal force aware of their presence. Alternatively, however, this effort to safeguard civilians would also place the entering British troops at greater risk by giving any German soldiers that might also be hiding there the opportunity to attack first. The soldier who wrote the memoir admitted that attacking first would have felt like murder to him if it resulted in the death of an innocent French family member. According to Walzer's subsequent analysis, soldiers in such cases are in fact obliged to assume increased risk and – in an effort to limit the expected costs in terms of civilian casualties – issue a verbal warning prior to engaging with a grenade.¹

This World War I example rests on a moral argument. From a utilitarian perspective, however, if the British troops opted to make themselves safer by throwing the grenade without warning, it would matter little for the ultimate outcome of the conflict. While the resulting French civilian casualties would be tragic, might weigh heavily on the consciences of those responsible, and could potentially encourage in-kind retaliation from the Germans, they would be of little military consequence. In conventional interstate conflict, civilian casualties do little to inhibit the ability of military forces to mass firepower on enemy objectives, seize terrain, and ultimately achieve victory at the strategic level.

Asymmetric intrastate conflicts are different. In conflicts like those in Afghanistan, Colombia, Iraq, Northern Nigeria, Pakistan, and the Philippines, to name just a few, information flows, collaboration, and ultimately support of the local population are key to achieving strategic objectives. Limiting casualties to noncombatants and other actions that alienate the population have clear military value in such conflicts. But while military commanders in all types of war have moral obligations to abide by international norms and humanitarian laws governing their treatment of noncombatants, just how much risk to their own forces they must take on in the process is never completely clear. Indeed, the principle of proportionality in the law of armed conflict all but necessitates that they make a utilitarian calculation: potential harm to civilians must always be balanced against military value when considering actions that could hurt innocents. And if minimizing civilian casualties helps advance strategic goals in certain conflicts, then the standards for protection might be much higher.

These were the challenges that the International Security Assistance Force (isaf) was grappling with in Afghanistan in 2009. Protecting civilian lives had clear military value at a time when isaf and the government of Afghanistan were competing with the Taliban for the allegiance and support of the population. Standards of action that entailed protections for civilians, which were appropriate for interstate wars, and met requirements under international law were not necessarily protective enough. That observation prompted senior leaders within the organization to call for greater restraint when engaging an enemy that operated in close proximity to the civilian population. This increased emphasis on limiting civilian casualties, what became known as courageous restraint, was deemed critical to achieving strategic success.

In this essay, we first describe the genesis of courageous restraint in Afghanistan and discuss the arguments made for it on moral and legal grounds, as well as in terms of the expected impact on the success of isaf's campaign. We then highlight the challenges it faced in execution at the tactical level. We conclude with a discussion of the enduring lessons that can be learned from ISAF's experience implementing courageous restraint and its implications for the preparation and execution of future conflict.

In late spring 2009, nearly a decade after the U.S.-led invasion of Afghanistan toppled the Taliban and drove Al Qaeda from its former safe havens, the United States remained at war and, by most measurable standards, the war was not going well. According to nato/isaf statistics, there was a 156 percent increase in attacks on Afghan government infrastructure for the period of January to May 2009 compared with January to May 2008; a 152 percent increase in complex attacks (those involving more than one means of attack, such as small arms plus ied, or more than twenty insurgents); and an increase of between 21 and 78 percent in total attacks across the five Regional Commands within Afghanistan.² Newly elected President Barrack Obama considered Afghanistan a war of necessity, not of choice like Iraq, but his administration, like much of the U.S. public, was not willing to expend American blood and resources indefinitely in pursuit of their campaign's objectives.

In a very visible manifestation of the dissatisfaction with the status quo, Secretary of Defense Robert Gates called for the resignation of General David McKiernan, commander of U.S. and Coalition Forces in Afghanistan, in early May 2009, citing the need for “fresh thinking” and “fresh eyes” on Afghanistan.³ Lieutenant General Stanley McChrystal, the storied Army Ranger and Special Operations Forces commander who led the Joint Special Operations Command (jsoc) from 2004 to 2008, was tapped as McKiernan's replacement and leader of the new direction in Afghanistan. Shortly after taking command, he called for a comprehensive assessment of isaf's mission, objectives, and strategy.

Based on the findings of the June 2009 assessment, General McChrystal requested an additional forty thousand troops to “surge” to Afghanistan later that year and help provide much needed physical security to facilitate the broader aspects of a comprehensive counterinsurgency campaign plan. Perhaps even more significant than calling for a troop increase, General McChrystal determined that ISAF needed to fundamentally change how it operated in Afghanistan down to the level of how soldiers and small units interacted with the populations living where they were deployed. Specifically, he was concerned with the impact of the mounting civilian casualties that isaf was responsible for and his command's relationship with the population it was ostensibly deployed to protect. Reflecting on this assessment, General McChrystal recalls, “I quickly came to the conclusion – and had been talking about this for years – if we didn't change the Afghan people's perceptions about our use of power, then we were going to lose them.”⁴

On July 2, 2009, General McChrystal issued a revised tactical directive for isaf. The directive outlined policies for the employment of air delivered munitions, indirect fires (such as artillery and mortars), and other weapon systems, intending to reduce isaf-caused civilian casualties and other collateral damage.⁵ The principles and command intent laid out in this document would make up the foundation of the Commander of International Security Assistance Force's (comisaf) calls for restraint and tactical patience when determining how much force to employ in certain battlefield conditions.

The tactical directive remains a classified document, but portions of it have been released in an effort to educate a wider audience. The carefully worded and personally authored passages provide both insight and clarity on why General McChyrstal, as comisaf, was determined to limit the civilian casualties caused by isaf and his intent for how isaf troops were expected to exercise the restraint required to achieve these ends.

This was not a case in which the commander was inherently conservative about using force. General McChrystal commanded elite counter-terrorist operatives in jsoc at the height of the conflict in Iraq. In this role, he maintained a near singular focus on killing and capturing high-value targets and weakening terrorist networks through attrition of key leaders and individuals. But conditions in Afghanistan in 2009 and his role as the theater commander obliged him to expand, if not shift, his emphasis. To ensure that isaf's tactical and operational actions supported the overarching strategy he was responsible for pursuing, he wrote:

In addition to modifications of the tactical directive, isaf also issued new counterinsurgency guidance,⁷ a revision of its standard operating procedures (sop) for the escalation of force,⁸ and a tactical driving directive, all of which shared a common theme of directing isaf members to operate in a way that protects the population and limits civilian casualties and collateral damage. The revised tactical directive and these additional documents provided the basis for the concept of courageous restraint.

comisaf guidance and intent were emphatic. isaf soldiers are expected to operate in ways consistent with protecting the population and limiting civilian casualties. None of these directives explicitly denied isaf soldiers the ability to defend themselves, but they set explicit and implicit expectations that isaf troops would exercise restraint on the battlefield when civilian lives were potentially in danger. They acknowledged that exercising this restraint might require commanders and individual soldiers to accept an increased degree of risk as part of their effort to reduce casualties to the civilian population.

There was consensus between the comisaf and a number of senior leaders that soldiers exhibiting courageous restraint should be recognized for their actions. The isaf Counterinsurgency Advisory and Assistance Team (caat) responsible for helping communicate comisaf intent to isaf troops in the field described this interest: “We routinely and systematically recognize valor, courage, and effectiveness during kinetic combat operations. … In a coin [counterinsurgency] campaign, however, it is critical to also recognize that sometimes the most effective bullet is the bullet not fired.”⁹ Nick Carter, a British major general and commander of Regional Command South, which included the volatile provinces of both Kandahar and Helmand at the time, went as far as advocating for the creation of a medal recognizing isaf soldiers and marines for exercising restraint when appropriate on the battlefield. According to Carter, restraint and tactical patience should be viewed as an “act of discipline and courage not much different than those seen in combat actions.”¹⁰ Broader support for establishing a special award for courageous restraint never gained traction within isaf, but the increased emphasis by the comisaf and senior leaders within the command on reducing civilian casualties was palpable and could be felt down to the lowest echelons in the field.

Why did the comisaf demand that isaf troops exercise courageous restraint? For one, protecting the population by exercising restraint in combat and assuming risks to avoid civilian casualties is consistent with international law. Additional Protocol i of the Geneva Conventions, adopted in 1977, lays out the signatories’ obligations with respect to protecting victims of international armed conflicts.¹¹ Article 51, “Protection of the Civilian Population,” describes types of indiscriminate attacks prohibited by the treaty, including, for example:

1. An attack by bombardment by any methods or means which treats as a single military objective a number of clearly separated and distinct military objectives located in a city, town, village or other area containing a similar concentration of civilians or civilian objects.

2. An attack which may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated.¹²

Article 57, “Precautions in Attack,” further requires that “in the conduct of military operations, constant care shall be taken to spare the civilian population, civilians and civilian objects.” This article explicitly mandates that combatants:

1. Do everything feasible to verify that the objectives to be attacked are neither civilians nor civilian objects.

2. Take all feasible precautions in the choice of means and methods of attack with a view to avoiding, and in any event to minimizing, incidental loss of civilian life, injury to civilians and damage to civilian objects.

3. Refrain from deciding to launch any attack which may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated.¹³

Succinctly, failing to distinguish between civilian and military targets is a war crime as defined under the provisions of Protocol i. For most of the war, isaf units took great pains to follow the standards of distinction and proportionality enshrined in the 1977 Conventions. By mid-2009, however, a bias toward greater caution and a high threshold of military advantage to justify actions that risked civilian lives was seen as ensuring compliance with the law of armed conflict.

Beyond the moral and legal incentives outlined above, however, exercising restraint in asymmetric conflicts can be strategic; it has become a key component of successful counterinsurgency, such as in Afghanistan. Looking back on this time, General McChrystal explained: “So what we were trying to do was tell people – and I was trying to communicate it in a way that emphasized that the only thing that matters here is winning. Now, the only way we win is not by killing more Taliban, but by convincing people of the efficacy of our strategy, and of our commitment to their protection. I still passionately believe that this is absolutely the right answer.”¹⁴ General McChrystal's perspective is grounded both in concerns about how the Afghan population perceived isaf and in a long tradition of scholarship and practice that identified gaining cooperation from non-combatants as a critical part of winning a counterinsurgency campaign. Insurgent leaders – from Mao Tse-tung (1937) to Che Guevara (1960) to Marighella (1969) – emphasize the criticality of earning popular support so as to ensure insurgents can operate undetected by government forces. This view was echoed by a large group of Western counterinsurgency theorists who fought against communist insurgents in the postcolonial period, including Roger Trinquier, Robert Taber, David Galula, David Clutterbuck, E. P. Thompson, and Frank Kitson. Recent work by American counterinsurgency practitioners drawing lessons from Vietnam and Iraq, including Kalev Sepp, David Petraeus, Robert Cassidy, and H. R. McMaster, emphasize the importance of earning good will and avoiding actions that discourage cooperative non-combatant behavior because civilians can provide valuable intelligence.

A compelling example of the tangible returns that can be gained from displaying restraint and tactical patience can be seen in the experience of a U.S. Marine unit operating in the Garmsir District of the volatile Helmand Province in January 2010. In this case, the 2nd Battalion, 2nd Marines were confronted by a large and overtly angry crowd of local Afghans enraged by the rumor that members of the U.S. military had defaced a Koran. It escalated and the visibly frustrated locals began to throw rocks and bricks, many of which struck the young Marines that had formed a perimeter surrounded by the rioting Afghans. A number of Marines suffered concussions and other serious injuries; however, while justified by their rules of engagement, no Marines responded through an escalation of force. Eventually, word that the Koran burning was in fact a rumor planted by the Taliban subdued the crowd and they dispersed without further incident. The courageous restraint exercised by these disciplined Marines and their small unit leadership prevented a dangerous situation from escalating to something much worse and avoided inflicting casualties on the civilian population. Of note, this Marine unit was among the most successful in the theater at locating ieds and detonating them (so they no longer presented a risk) in the months following the crowd incident, largely owing to the battalion's ability to get tips from local Afghans regarding the locations of these bombs. Thus, as the unit's commanders acknowledged, building a strong relationship with the local Afghans provided their best protection.¹⁵

There are many anecdotal accounts of isaf members attributing the importance of their relationship with the local population with facilitating information sharing and other forms of collaboration to tactical and operational level success. Measuring attacks that did not occur (dogs that do not bark) is difficult, but quantitative tools provide some options. By using the fine-grained administrative data collected in the wars in Afghanistan and Iraq, we can assess the impact of inadvertent killings of civilians by both sides, the tragic side-effects of conflict, on subsequent violence and other outcomes. In prior work, we and our colleagues have found substantial econometric evidence that harming civilians can hurt counterinsurgent efforts.

In Iraq, insurgent attacks increased modestly for a one- to two-week period following Coalition-caused civilian casualties; the median Coalition-caused incident led to approximately two additional attacks over the next two weeks in the average district.¹⁶ Moving up to the province level, the next higher geographic unit, the flow of information to Coalition forces on tip lines also dropped following such casualties for a one-week period. In 2007 and early 2008, the median Coalition-caused civilian incident led to approximately 1.6 fewer tips in the following week.¹⁷ Afghan public opinion in 2012 was significantly more favorable to the Taliban relative to isaf for people who reported suffering harm from isaf operations.¹⁸ And in all three cases, the effects were asymmetric: government forces and their allies paid a greater penalty for causing the same level of harm as the Taliban did, though both sides paid a cost for harming civilians in terms of attitudes, information flow, and subsequent attacks.

Similar effects on insurgent attacks were seen in Afghanistan. In April 2010, members of the isaf caat conducted an empirical study of the impact of civilian casualties on future insurgent-initiated violence. The comprehensive study – later briefed to the comisaf and isaf senior leaders – found evidence that civilian deaths caused by isaf led to increased attacks against Coalition Forces that persisted for fourteen weeks. Interestingly, civilian casualties increased violence directed against isaf whether the Taliban or Coalition was responsible for the casualties, though the impact was much larger for isaf-caused incidents.¹⁹

It is not surprising that civilian casualties alienated the population, shifted support away from isaf, and contributed to an increase in violent attacks directed at coalition forces. And compelling empirical evidence of this causal relationship further validated isaf's emphasis on restraint. General McChrystal reflected: “That really affected me. Because I remember the takeaway from that [caat Civcas Brief] was, no matter who causes violence in an area – you do it or the enemy – it makes the area less secure and less stable over time. Get down the violence, period, and then you can start other things.”²⁰

Despite the moral, legal, and strategic justification for courageous restraint, it met significant resistance from many of the individual soldiers and marines in the field who were asked to use it, at least as interpreted by their immediate chain of command. Small unit combat in the restive areas of Helmand, Kandahar, the Korengal Valley, and other hot spots remained, as it has throughout history, a kill-or-be-killed exercise in survival from the perspectives of those closest to the fighting. It was difficult to convince these forces that accepting risk in a combat situation – deliberately jeopardizing the lives and safety of one's own forces – may be the optimal response in strategic terms. Neutralizing imminent lethal threats to yourself and fellow comrades using the most effective weapons systems and firepower available is a near reflexive action for combatants struggling to survive and triumph in the heat of battle with all the fear and visceral emotions that accompany it. Exercising restraint may very well be a morally correct and strategic response, but is exceptionally challenging to implement for those expected to pay the devastating human costs that can result from showing restraint.

A U.S. Army Ranger company commander described an incident involving another company in his battalion that underscored the reality troops faced when operating in compliance with aspects of the tactical directive:

Another experienced U.S. Army Ranger commander, deployed to Afghanistan in 2010, believed a major aspect driving the Rangers to comply with the directives was fear that their actions would be responsible for getting their unit in trouble with the senior isaf leadership: “We did not want to be responsible for shutting down the Task Force [special operations unit]. My Rangers understood the importance [of the tactical directive], but having been on a few hundred raids, I saw that no matter how well we did we were horrible at the io [information operations] fight and that once we left the target anything could be said and a Presidential Inquiry from Karzai typically just reinforced the negative story.”²³

Civilian casualties were at times difficult to avoid in the course of operations, even when the restraint and tactical patience called for from the updated isaf directives were followed. A special operations mission in Ghazni province in 2010 represents one particularly extreme example. The Rangers maneuvering during this raid came under fire from a compound on the target and they responded with well-aimed fire at the combatants engaging them. In the course of this engagement, some of the small arms rounds fired by the Rangers passed through the torso of the enemy combatant and struck a woman behind him not visible to the Rangers. An official inquiry conducted by the Ranger unit supported the information and images collected on target. A subsequent Afghan presidential inquiry, however, concluded instead that the U.S. military members were not provoked or in any danger when they killed multiple women and children and even claimed that there were not any Taliban at the residence where the civilians were killed. This type of misinformation from the Afghan government at that time was a most frustrating aspect and challenge to the implementation of the tactical directives for isaf forces.²⁴

A significant challenge in garnering support for courageous restraint was overcoming the inertia and default behavior within the ranks when U.S. forces made contact with the enemy. Traditionally, soldiers were rewarded for aggressive actions on the battlefield that inflicted casualties or damage on the enemy. A U.S. infantry officer serving in Afghanistan at the time recounted: “The first tactical directives were the hardest to embrace, because we had gone from total freedom of maneuver in Iraq and in the early years in Afghanistan to a more constrained mo in the later years in Afghanistan. Eventually, we figured out that Afghanistan in the 2010s was more politically sensitive, and we adjusted our attitudes and tactics accordingly.”²⁵ Additionally, the notion that restraint on the battlefield should be recognized and rewarded was not consistent with how soldiers were trained and largely not how they had operated in previous tours to date. Choosing to avoid kinetic engagements with the enemy under some conditions to avoid civilian casualties and support strategic objectives is a tough sell for troops at the tactical level.

General McChrystal described how ISAF troops that were able to work closely with the local population were much more inclined to appreciate the critical importance of protecting civilians than were those that had little real contact with Afghans: “When these people were in an area for a long time, and they got enough interaction with the local population so that they could see the complexity of that situation, they're the ones who get it.”²⁶

The command emphasis on reducing civilian casualties had a significant impact on the number of civilian casualties attributed to progovernment forces. In the year following General McChrystal's command directives, there was a 28 percent reduction in casualties attributed to American, nato, and Afghan forces; deaths from aerial attacks fell by more than one-third.²⁷Figure 1 shows the time-series of combat incidents and two measures of civilian casualties, as recorded by the isaf Civilian Casualty Tracking Cell, from January 2009 through January 2010. The top panel shows the four-week moving average of combat incidents per week: combat violence rises throughout summer 2009 and then begins to dip in the fall and into the winter. The middle panel shows the four-week moving average of insurgent-caused civilian casualties (killed plus wounded) per combat incident. This is a measure of how much risk civilians faced from insurgents given the intensity of combat. Throughout the period, there were approximately 0.15 civilian casualties caused by insurgents per combat incident. The bottom panel shows the four-week moving average of isaf-caused civilian casualties (killed plus wounded) per combat incident. This is a measure of how much risk civilians faced from isaf given the intensity of combat. Prior to General McChrystal's revised tactical directive, isaf forces caused approximately 0.04 civilian casualties per combat incident. Afterward, that number dropped in half, to 0.02.

These data highlight two important patterns. First and foremost, civilians were at much greater risk from insurgents in 2009. Second, the risk to civilians from isaf-given levels of combat dropped substantially starting in late May 2009, when the comisaf began emphasizing civilian casualties as a threat to accomplishing the mission. The courageous restraint concept was clearly being adopted by isaf personnel.

But these measures not only protected civilians; the restrictions on the application of firepower protected the Taliban as well. This was likely a contributing factor in the dramatic overall increase in civilian deaths during the year that courageous restraint was implemented given that the large majority of civilian deaths recorded were attributed to actions initiated by the Taliban. Additionally, in some instances, the increased restrictions on the use of firepower disappointed isaf partners in the Afghan National Security Force who depended on U.S. firepower as a key combat multiplier and thus were not always supportive of isaf units’ decisions to restrict their employment of these resources even when available.²⁸

General David Petraeus took command of isaf in a subdued ceremony on July 4, 2010, following the abrupt departure of General McChrystal in the wake of the publication of an article in Rolling Stone magazine in which his subordinates were quoted making disparaging remarks about senior U.S. political leaders. Critics of courageous restraint were hopeful that the new comisaf would revise or even retract some of his predecessor's policies and address a directive that some perceived as overly restrictive of their right to defend themselves. A senior British noncommissioned officer in Sangin, Helmand Province, lamented at the time: “Our hands are tied the way we are asked to do courageous restraint. I agree with it to the extent that previously too many civilians were killed but we have got people shooting us and we are not allowed to shoot back. Courageous restraint is a lot easier to say than to implement.”²⁹

General Petraeus literally “wrote the book” on population centric counterinsurgency, however, and the emphasis he placed in principle on limiting civilian casualties reflected more continuities than differences with that of his predecessor.³⁰ But the new comisaf appreciated the misgivings voiced from soldiers in the field and amplified up the chain of command and, in some cases, all the way to their representatives in Congress. He was concerned that his predecessor's policies on tactical level restraint and restrictions on employment of force had gone too far. He implemented key revisions to the explicit content as well as interpretation of comisaf guidance to ISAF troops operating in the field.³¹

Comparing the updated tactical directive that General Petraeus issued in July 2010 with McChrystal's 2009 version, it is clear that General Petraeus's directive strived to alter the risk relationship/balance between Afghan civilians and the U.S. military.³² The 2009 directive acknowledged that “the carefully controlled and disciplined use of force entails risk to our troops”³³ and that the imperative to protect forces may at some level, in some conditions, be subordinate to protecting civilian populations. General Petraeus's revision of the tactical directive one year later explicitly put protection of Afghan civilians and protection of service members as equal moral imperatives. Importantly, he adds additional emphasis that no members of isaf would be denied the right to defend themselves, nor could any subordinate commander make further restrictions to his guidance.

General Petraeus emphasized, however, that he expected isaf troops to display what he termed “tactical patience” in their operations, which was largely consistent with the intent of courageous restraint. General Petraeus admonished coalition forces in his revised directive: “We must continue – indeed, redouble – our efforts to reduce the loss of innocent civilian life to an absolute minimum. Every Afghan civilian death diminishes our cause.”³⁴ Thus, the war effort at isaf continued with a comisaf committed to limiting civilian casualties from isaf operations. While courageous restraint du jour left with General McChrystal, the de facto emphasis on being prepared to assume risk to avoid civilian casualties endured and was largely consistent with the ongoing comprehensive population-centric counterinsurgency strategy pursued.

Exercising restraint and limiting non-combatant casualties is nearly always justified on moral grounds and according to the applicable international law and conventions. The aggregate returns on accepting risks at the tactical level, however, vary based on the characteristics of the conflict. Enforcing policies that call for discriminate use of firepower and exercising restraint in its application can be a net gain for states combating insurgency and other internal threats. The anticipated gains from such restraint, however, will vary, and developing an appreciation of where and under what conditions these gains are most significant is important to understand. General McChrystal, in his 2009 tactical directive, acknowledges that the document outlined his intent but would have to be interpreted by junior leaders and individual soldiers in the context of the situation and local conditions at hand. “I cannot prescribe the appropriate use of force for every condition that a complex battlefield will produce, so I expect our forces to internalize and operate in accordance with my intent. Following this intent requires a cultural shift within our forces – and complete understanding at every level – down to the most junior soldiers. I expect leaders to ensure this is clearly communicated and continually reinforced.”³⁵ When soldiers, marines, and other combatants are asked why they performed in a certain way in the heat of combat, they are likely to respond: “I did what we were trained to do.” In the stress, uncertainty, and ambiguity of combat, individuals’ behavior defaults to how they were trained. It's critical to continue to invest in the quality of junior leaders and training of all combatants and ensure that their preparation and training provides a base to draw on when making these split-second life and death decisions both for themselves, the enemy they are attempting to engage, and the noncombatants potentially caught in the crossfire.

Right now we're losing the tactical-level fight in the chase for a strategic victory. How long can that be sustained?”³⁶ The exasperated U.S. military officer making these remarks in the spring of 2010 cast doubt on the tactical restraint and emphasis on reducing civilian casualties that isaf was promoting at this time. To him, no strategic goal was worth, or could survive, continual tactical failure. But asking isaf troops to embrace and display courageous restraint was made with clear strategic objectives in mind. Sun Tzu allegedly warned, “Strategy without tactics is the slowest route to victory. Tactics without strategy is the noise before defeat.”³⁷ This ancient dictum still applies today in that tactical gains are irrelevant unless they are accompanied by an overarching strategy that links the outcome of individual engagements to achieving a larger political goal.

We recognize that striking the balance between fostering conditions necessary to make gains at the strategic level and achieving tactical objectives, including force protection, is difficult, and ultimately believe the answer is conditional. In cases of asymmetric counterinsurgency, popular support and willingness to share information can be significantly impacted by perceptions and judgments shaped by tactical level actions and activities. Thus, tactics that are effective in the moment of an engagement, such as the employment of artillery or large volumes of heavy weapons fire, can undermine overarching strategic ends in ways not experienced in symmetric conflict, where support and information from the population are less consequential. If victory is defined in comprehensive terms, the route to achieve it must reflect the concerns that courageous restraint was designed and intended to address.

Protecting the population as a means to garner greater popular support and accepting increased risk to forces in order to limit casualties to noncombatants pays off in some cases under some conditions and less so in others. Courageous restraint was always intended to be interpreted in case-and situation-specific contexts. As General McChrystal acknowledged, “I wrote [the tactical directive] not to prescribe tactical decisions for sergeants and junior officers closest to the fight, but to help them understand the underlying logic of the approach I was asking them to employ.”³⁸ The potential returns on exercising restraint and tactical patience on the battlefield must be recognized and anticipated by military leaders at the small unit level.

Measures intended to minimize civilian casualties such as courageous restraint can be a strategic net gain for forces combating insurgencies and in other conflicts where information and support from the civilian population are critical enablers for success. Voluntarily displaying such restraint is a challenging concept to internalize, however, especially for troops who expect to make contact with a deadly enemy and are trained and conditioned to decisively bring to bear the superior combat power they possess. The near-term risks and costs of exercising this restraint are very clear to soldiers exercising it, whereas its anticipated strategic benefits in the longer term are far less compelling at the tactical level – especially in the heat of the moment in combat. For commanders and soldiers in the field, the optimal level of restraint – if any – in a given situation will vary based on a multitude of dynamic factors and conditions.

Investments in education and training, as well as in quality leadership down to the small unit level, can increase combatants’ capacity to make decisions tailored to the prevailing tactical and strategic conditions. In the heat of combat, however, the decisions that impact the lives of soldiers and noncombatants alike and that can influence the strategic direction of a military campaign are complex dilemmas often only made clear in hindsight, if ever.

Ultimately, third-party counterinsurgency campaigns such as the U.S.-led effort in Afghanistan can only be as effective and legitimate as the governments they support. Limiting harm to civilians in areas where government authority is contested is not only a moral imperative but also an important component of any comprehensive strategy to achieve victory in these conflicts. It can provide near-term tactical advantages and buttress efforts to convince civilians to support the government. When the incumbent government is viewed as corrupt, unrepresentative, or otherwise illegitimate, however, even the most discriminate military forces of the state and its allies will be constrained in their ability to gain the support from the population needed to roll back insurgent threats, much less to sustain that support and prevent a return to violence. For many military forces engaged in the complex struggle to combat insurgent threats, this is the essence of the challenge.