Just How Many Civilians Are We Comfortable Killing in War?
Armies that accept human rights rarely backslide
On March 17, 2017, soldiers from Iraq’s special forces fighting in the Mosul Jidideh neighborhood called in U.S. air strikes on three buildings being used by Islamic State snipers. One of the buildings collapsed, and between 61 and 150 civilians died in the strikes.
U.S. officials now admit that the strike likely came from U.S. aircraft, but a Pentagon investigation on the exact role of American forces is still pending. According to an Iraqi special forces commander, the soldiers calling in the strike did not know that civilians were hiding among the buildings.
However, a New York Times article found this assertion to be unlikely, considering leaflets were dropped encouraging civilians to leave.
The strike comes just days after two similar incidents allegedly involving coalition strikes reportedly killed dozens in Syria, and nearly two months after a special operations raid in Yemen cost the lives of numerous civilians.
Some news outlets have insinuated a connection between the rise in civilian casualties and the Trump administration’s signaling of a greater willingness to delegate strike authority to military commanders than Pres. Barack Obama’s administration.
Indeed, Iraqi ground commanders have confirmed that strike authority has been given much quicker since Trump took office. The commanders say they are often able to circumvent some time-consuming collateral damage checks, although U.S. officials say the rules of engagement have not changed.
The change has been welcomed by both Iraqi and U.S. officers who “chafed at what they viewed as long and onerous White House procedures for approving strikes under the Obama administration.”
Although media outlets admit that it is too early to identify a trend or make any causal claims between the Trump administration’s policies and an increase in civilian deaths, the insinuation does bring up a critical question for civil-military relations:
Does an increase in authority to military officers over battlefield operations also lead to an increase in civilian deaths?
Political scientist Barry Posen finds in The Sources of Military Doctrine that when military leaders are delegated a great deal of battlefield authority they typically choose offensive strategies — i.e., destroying an enemy within its own territory.
This is, in part, because offensives reduce uncertainty by allowing commanders to carry out preplanned attacks rather than reacting to the enemy while in a defensive position, according to Posen. In other words, offensives inherently align more with the institutional interests of militaries than do defensive operations.
Building on this work, historian Isabel Hull argues in Absolute Destruction, that a military with autonomy over battlefield decisions develops an organizational-culture with an extreme view of what constitutes a “successful” offensive.
For instance, officers may come to define successful operations as destroying all sources of enemy power, even when civilians stand in the way. Further, in Hull’s view, when civilians are seen as aiding the enemy they are likely to become the target during offensive operations.
To see whether Posen and Hull’s theory is supported by the historical record, it is helpful to look at how different regime types treated civilians during war.
By combining one dataset from University of Pennsylvania Professor Jessica Stanton on civilian targeting in civil wars, and another from UCLA professor Barbra Geddes on regime types, I find that 15 out of 21 wars fought by military-dominated regimes since the 1980s featured systematic civilian targeting.
The argument here is not that autonomous militaries are the only or even the main reason for civilians targeting. For instance, although the chart below shows that military-dominated regimes account for the largest share of civilian targeting, even democracies victimize civilians.
Instead, my argument is that military autonomy leads to an adoption of an offensive-oriented organizational-culture that carries with it a higher acceptance of civilian targeting when deemed necessary for success on the battlefield.
Although military autonomy can occur under all regime types, a few historical examples from military-dominated regimes helps to clearly illustrate this argument in the most extreme cases of autonomy.
A protest against the Indonesian occupation of East Timor. Photo via Wikipedia
Indonesia’s ‘fence of legs’
Although the Indonesian military officially seized power in the 1960s, they maintained autonomy over military operations since Indonesians began their rebellion against the Dutch in 1945. Beginning in that war, the military developed a number of brutal offensive measures, mainly involving scorched-earth tactics.
Building on successes during their rebellion against the Dutch, the Indonesian military formulated what became known as the “fence of legs” strategy while fight against the Darul Islam insurgency shortly after independence.
This strategy involved forcing dozens of civilians to act as human shields for Indonesian troops as they cleared insurgent strongholds. Civilians involved in these operations often died of starvation, disease or being shot by Indonesian troops for breaking the cordon.
This tactic, and the propensity to crush the enemy through offensive actions, became ingrained in the military’s organizational-culture through military manuals, training, and promotions.
Not surprisingly, the Indonesian military’s invasion of East Timor in the late 1970s featured saturation bombing of villages, the “fence of legs” strategy, mass re-concentration of civilians into disease-ridden camps and numerous instances of hundreds of civilians being massacred for suspected collaboration with insurgents.
A Kachin insurgent in Burma. Photo via Wikipedia
Burma’s ‘four cuts’
The Burmese military’s propensity for massacres followed a similar path. Although the military in Burma did not officially come to power until 1962, it maintained autonomy from civilian control since independence in 1948. Ever since then, the Burmese military has fought a civil war against multiple insurgent organizations.
Throughout the mid-1950s and 1960s, the military began to develop an offensive strategy that eventually became known as “the four cuts.” This strategy required cutting insurgents off from any form of resources — i.e., “food, funds, intelligence and recruits.”
Tactics used to accomplish this strategy included moving entire civilian populations to government controlled areas, conducting devastating scorched earth campaigns, and establishing free-fire zones after “clearing” the area of civilians. The tactic was usually accompanied by dozens, if not hundreds of civilian deaths.
The Burmese military internalized a preference for the “four cuts” strategy into its organizational-culture after numerous insurgent groups were destroyed or crippled by the brutality early on in the civil war. This outcome was aided by the fact that, much like the Indonesian military, there was no serious effort to stop the practice by domestic or international forces.
Thus, the Burmese military continued to use the brutal “four cuts” strategy to destroy or force into surrender most of the insurgent groups over the six-plus decade span of the civil war.
Vietnamese refugees board a U.S. Navy ship during the fall of Saigon in 1975. Navy photo
Democracies and civilian targeting
Indonesia and Burma offer fairly good evidence of a connection between military autonomy and civilian targeting. But what about democracies? Since the main question in this article concerns the U.S. fighting against ISIS, are there any similar cases where a military in a democracy abused their delegated power?
Turns out there are.
At the turn of the 20th-century Britain fought a war in South Africa, known as the Second Anglo-Boer War. After the conventional phase ended, the war devolved into a guerrilla conflict. Much like the Indonesian and the Burmese militaries, the British decided to cut off the guerilla’s ability to gain food and shelter by burning down most of the farms in the guerillas’ territory in South Africa.
The South African civilians displaced by the massive scorched earth strategy, both black and white, were placed into concentration camps where over 45,000 died of disease and starvation due to foreseeable negligence by the military.
This occurred despite the fact that the British civilian governor in South Africa, Alfred Milner, formally asked the cabinet in England to end the practice. His reasoning was two-fold. First, Milner believed that destroying the territory would make rebuilding the territory extraordinarily expensive. Second, he believed that such atrocities would make the population more likely to rebel in the future—also an extraordinary expense.
The cabinet reasoned that they should not intervene in military matters because doing so would break historical precedent and may interfere in military effectiveness.
And despite the popular contemporary conception of colonial brutality, the cabinet was under a great deal of public pressure to end the destructive campaign after the atrocities became widely publicized in the press.
Still, an example closer to our time is warranted.
Following the Six Day War in 1967, the Israel Defense Forces slowly gained autonomy over military operations. This was mainly due to Israeli society’s well-founded belief that their national survival was under threat from surrounding Arab states. Thus, like many societies throughout history, Israel accepted that the military would maintain extraordinary autonomy over national security policy and the conduct of war.
By the time Israel invaded Lebanon in 1982, the IDF maintained a great deal of control over how it conducted the campaign.
Israel’s offensive against Beirut in the summer of 1982 was extraordinarily brutal. Over the course of just eight days Israel’s siege and indiscriminate bombardment of Beirut cost the lives of 9,583 Lebanese and Palestinian civilians and another 16,608 were wounded, according to the Lebanese government.
The Israeli government put the number at 332 civilians killed, and 1,083 wounded. But in either case, the IDF’s pattern of indiscriminate artillery and airstrikes evidence a blatant disregard for avoiding civilian casualties.
But does this evidence mean that a sudden delegation of authority to U.S. ground commander’s fighting ISIS will mean systematic targeting of civilians when deemed necessary for successful operations?
I believe that such a sudden shift is unlikely. As will be shown below, civilian government officials with a preference for restraint have the tools to slowly conform the organizational-culture of a subordinate military to their own—even when brutality has been used in the past. Once this culture shifts towards restraint, it is difficult to shift back to brutality.
Israeli artillery fires during the 1967 Six Day War. CIA photo
The military in Burma, Indonesia in the 1970s, Britain in the early 1900s, and Israel in the 1980s all have one thing in common. At the time of those wars each military had enjoyed decades with almost no significant pressure from domestic or international forces concerning the conduct of military operations.
But not long after Britain’s war in South Africa, and Israel’s in Lebanon, domestic political leaders began taking steps to reign in the power of the military.
In Israel’s case, both the fiasco in Beirut and Israeli society’s general realization that it no longer had to fear for its national survival from invading armies, resulted in democratic institutions acting to reassert civilian control over the military.
This process took time, and is owed to a number of factors, including. The newfound willingness of Israel’s Knesset and the cabinet to challenge generals on the defense budget. The Israeli media’s invasive questioning of the military on a range of matters from gender discrimination to conduct during war. And the willingness of the Israeli Supreme Court to challenge the military, including on military tactics in the West Bank and Gaza.
But it seems that the Second Intifada — a political and military rebellion against the Israeli occupation in the West Bank and Gaza in the early 2000s — had the greatest impact on the IDF’s conduct vis-à-vis noncombatants during war.
The institutions of Israeli democracy, the courts, the media and the political establishment, consistently pressured the IDF to practice restraint during the Second Intifada.
For instance, according to Stuart A. Cohen, a professor of International Relations at Bar-Ilan University and a Senior Researcher at the BESA Centre for Strategic Studies, the uprising caused Israel’s cabinet to implement heavy control over military operations that had the potential to result in civilian deaths because of a “growing importance being attached world-wide to respect for international human rights law.” In addition, according to Cohen, the Israeli public pressured their elected leaders to practice restraint.
Both the Israeli government’s ability to monitor the military, and the Israeli public’s opinion was undoubtedly shaped by the media’s intense coverage of conduct on the battlefield.
However, the pressures from Israel’s democratic institutions were not always able to induce restraint. For instance, the IDF continued to practice what they called the “early warning” procedure during military operations.
According to the official IDF policy, this procedure allowed “Israeli soldiers wishing to arrest a Palestinian suspected of terrorist activity may be aided by a local Palestinian resident, who gives the suspect prior warning of possible injury to the suspect or to those with him during the arrest.”
In practice, this amounted to the IDF using Palestinians as “human shields,” according to the Israeli human rights organization B’tselem.
B’tselem and other human rights organizations have documented cases of IDF soldiers ordering Palestinians to “enter buildings to check if they are booby-trapped, or to remove the occupants; remove suspicious objects from roads; stand inside houses where soldiers set up military positions, so that Palestinians would not fire at them; and walk in front of soldiers to shield them from gunfire, while the soldiers point a gun to their backs and sometimes fire over their shoulders.”
The IDF used the early-warning procedure 1,200 times during the five years of the Second Intifada, according to the Israeli supreme court. In 2005, the court deemed the practice illegal and the IDF officially ended the tactic. Though some unauthorized uses of the procedure continue, soldiers are now prosecuted for such actions.
Other cases ruled on by the supreme court have helped engender a culture of restraint in the IDF as well, including restrictions on punitive house demolitions of terror suspects and their families and targeted killings.
According to Cohen, the court rulings have “generated a feeling amongst several officers that, to be on the safe side, they had best consult with a lawyer before undertaking specific operations.”
The two recent wars fought by Israel in Gaza, “Operation Cast Lead” in 2008/09 and “Operation Pillar of Defense” in 2014, generally exhibit the IDF’s current culture of restraint. This is a contentious assertion.
Chart via the author
Chart via the author
Civilians accounted for 69 percent of those killed during Cast Lead and 64 percent during Protective Edge. However, the IDF’s collateral damage risk-mitigation practices show a general pattern of restraint.
This assertion is actually borne out in the numbers. There is a 0.73 positive correlation between civilians and combatants killed per day during Cast Lead, and a 0.78 correlation during Protective Edge.
In fact, it is possible to visibly observe this correlation in the chart: the lines representing combatants and civilians rise and fall in a similar pattern. Again, although civilians usually died in greater numbers than combatants each day, the correlations, coupled with evidence of the IDF’s targeting practices, suggest the civilian deaths were acceptable under relevant international humanitarian law.
In Britain’s case, the process of inducing an organizational-culture of restraint among officers took an almost entirely different path. Britain’s reason for inducing restraint was motivated almost solely by economic factors.
By the early 20th century the colonial territories had been conquered and needed governance. The expense of funding the territories through the British taxpayer became too great, and thus it was decided that the colonies would have to be self-sustaining. To do so, the Britain’s colonial office began plans to ensure the political and economic development of the territories was feasible.
Part of this plan was to make subjects in British colonies essentially citizens of the crown. This meant that English common law applied during rebellions, and thus the rebels and civilian population were to be treated as British subjects. This forced the military to fight rebellions more as a restrained police force than an offensive-oriented conventional army.
The legal distinction also meant that British civilian officials that ran the colonies maintained significant control of the military unless martial law were established—which often meant handing complete power over the colonies to the military.
But British civilian authorities never allowed full control to the military again after Gen. Reginald Dyer misinterpreted his authority to justify a massacre of as many as 300 civilians during an uprising in Amritsar, India in 1919—an act for which he was legally censured.
From that point on, civilian authorities maintained control of the military during rebellion. Further, official military manuals such as 1923’s Duties in Aid of the Civil Power and Notes on Imperial Policing from 1934 helped officers understand how to effectively fight a counterinsurgency while also practicing restraint and remaining subject to the civil authority.
Interestingly, most civilian governors consistently held the same views towards brutality during rebellion as Alfred Milner did — i.e., brutality makes rebuilding infrastructure extraordinarily expensive, and increases the likelihood of rebellion in the future.
This should not be terribly surprising since civilian colonial authorities also built an organizational-culture over many years of standardized training and merit-based promotions.
By the time the British fought a similar rebellion in Palestine in the late 1930s, the military was heavily monitored by civilian authorities, and the fighting was extraordinarily restrained—even by modern standards.
For instance, civilian authorities went so far as to restrict the Royal Air Force from dropping bombs near villages.
However, Burma and Indonesia are different stories.
The Indonesian military was extremely brutal during its fight in East Timor in 1999, even though officers were aware that a democratic government would be elected in the midst of its operations.
Similarly, in 2015 Burma elected Nobel Peace Prize winner Aung San Suu Kyi to run the country. And yet the Burmese military has committed mass atrocities in its current fight against Rohingya insurgents in Rakhine State.
The lesson from these cases is quite clear. Brutality likely occurred in East Timor in 1999 and is ongoing in Rakhine State because the political establishments in Indonesia and Burma — both of which purport to believe in protecting human rights — took no real steps to reign in the military.
By contrast, the relatively restrained wars in Gaza in 2008 and ’09 and 2014 were preceded by decades of human rights-minded domestic politicians asserting their control over the military. And although the British were motivated by economic factors, civilian control of the military resulted in a restrained counterinsurgency strategy even before the advent of human rights law or international humanitarian law pertaining to internal wars.
Much like the Israelis and British, U.S. military officers have incorporated the norm of civilian oversight into their organizational-culture over many decades. Further, the decades of civilian control over the United States has also engendered a deep respect for international legal norms among military officers. Most importantly, unlike the Burmese military, U.S. officers generally believe that targeting civilians is strategically counterproductive, especially in counterinsurgencies.
And looking at the lessons of the CIA’s fiasco with “enhanced interrogation,” it is reasonable to assume that American military officers know that even if the Trump administration gives them free reign on the battlefield, the next administration might punish them for atrocities.
In short, it is just as difficult to suddenly change the preference for civilian targeting in the Burmese military’s organizational-culture, as it is to change the U.S. military’s penchant for restraint.