The law of war allows, or at least tolerates, the killing of civilians not directly participating in an armed conflict, but the killing must not be “excessive” in relation to the “concrete and direct” military advantage that the belligerent expects to gain from the attack. This rule, known as the principle of proportionality, is designed to ensure that the ends of a military operation justify the means of the operation by weighing the anticipated military advantage against the civilian harm. If the weight of the expected military advantage is inflated, however, what would otherwise have been “excessive” civilian harm can appear justified, since the more a military effort can be presented as essential, the more civilian casualties the principle is willing to tolerate. Another misleading way to justify an excessive civilian toll is to downplay the harm civilians have been subjected to in order to render the military attack more acceptable. In legal debates about proportionality, computations determine the meaning, legitimacy, and ethics of violence. This can result in denigrating civilians and presenting them as killable subjects in order to justify the lethal violence deployed against them.
Since the war on terror was launched in 2001, proportionality has naturally assumed center stage in the debates on civilian deaths, as have human shields. In our research on the history of human shielding, we found that civilians trapped in conflict zones have often been framed as human shields both in order to shift responsibility for their deaths from those who killed them to those who “deployed” them, and to reduce the value attributed to the “civilian-shields” who died. In the aftermath of military campaigns in Afghanistan, Gaza, Iraq, Sri Lanka, and Yemen, legal experts—some of whom are stalwarts of human rights—have repeatedly defended State militaries that deployed lethal repertoires of violence against civilian populations by casting thousands and, at times, hundreds of thousands of civilians as shields. In other words, human shields have become a key element in the proportionality calculations accompanying debates about the ethical use of violence in contemporary wars.
Within this landscape, the Sri Lankan Civil War constitutes an unprecedented case in terms of the number of people who were framed as human shields and the mobilization of prominent human rights lawyers to defend war crimes. In our book Human Shields: A History of People in the Line of Fire, we examine the affidavits of six legal experts submitted at the request of the Sri Lankan government as part of its response to an incriminatory U.N. report (more on this below). An analysis of the affidavits reveals not only how the shielding accusation was used to justify thousands of civilian deaths during the final months of the war, but also how proportionality calculations can become a form of dehumanization—part of a legal exercise that can have the effect of legitimizing war crimes. Chronicling the computations carried out by two of these experts sheds light on this phenomenon.
The Sri Lankan Civil War
One of the confidential cables dispatched in mid-May 2009 from the United States embassy in Colombo to the State Department in Washington, D.C.—and published by WikiLeaks—describes the plight of civilians in the civil war’s final days. The cable recounts how the bishop of Mannar had called the embassy to ask it to intervene on behalf of seven Catholic priests caught in a so-called no-fire zone that had been set up by the Sri Lankan military as a space that was supposed to provide civilians protection from the fighting while providing them with humanitarian assistance. The bishop estimated that there were still 60,000 to 75,000 civilians confined within that particular zone located on a small sliver of coastal land about twice the size of Manhattan’s Central Park. Following the bishop’s phone call, the U.S. ambassador spoke with Sri Lanka’s foreign minister, asking him to alert the military that most of the people remaining in the no-fire zone were civilians stranded in what had become a death trap.
Thousands of Sri Lankan civilians were killed in the weeks before and after the cable was sent as the 26-year struggle waged by the Liberation Tigers of Tamil Eelam came to a horrific end. Having vied for a sovereign state in parts of Sri Lanka, the Tigers had spent years enhancing their social, economic, and military capacities. They demanded independence from Sri Lanka’s Sinhalese Buddhist government, which controlled the island’s linguistically and religiously diverse population of over 21 million people. From the early 1980s, the government had become more and more authoritarian, stifling criticism and deploying violence against the country’s religious and ethnic minority groups, not least its Tamil population. They experienced disappearances, unlawful killings, torture, rape, and sexual exploitation along with the ongoing clampdown on freedom of the press.
The Tamil Tigers, who controlled the majority of the territory in the northern and eastern provinces of Sri Lanka at the peak of their strength, responded to state violence with ruthless techniques of warfare, weaponizing human bodies through the systematic deployment of suicide bombers. Frequently targeting civilians, over the years they managed to kill several politicians, including the prime minister of India in 1991 and the president of Sri Lanka in 1993. The Tigers also massacred Sinhalese and Muslims living in villages bordering the front lines of the areas they controlled, while forcefully recruiting child soldiers from the Tamil population.
After four failed rounds of peace talks, an unsuccessful deployment of Indian peacekeeping forces, and a cease-fire agreement that lasted from 2002 until 2005, sporadic fighting resumed. By January 2008, the government declared a full military operation against the Tigers. Blasting its way through the northeast tip of the island, the Sri Lankan military slowly regained control of most of the regions previously held by the rebel group, while the civilian population that either sympathized with the Tigers or were afraid of the government forces moved deeper and deeper into the small swath of land still controlled by the Tamil guerrillas—an area characterized by the government as the “Tiger cage.”
At one point, the Sri Lankan military unilaterally declared the creation of three no-fire zones within the cage (including the one mentioned in the intercepted cable), urging the civilian population to gather in these zones by dropping leaflets from planes and notifying them through the radio and loudspeakers. As an estimated 330,000 internally displaced people assembled in these zones, the United Nations erected makeshift camps and, together with several humanitarian organizations, provided food and medical assistance to the desperate population—not unlike the safe areas the U.N. had created in Bosnia in 1993.
The Tamil Tigers also retreated to the no-fire zones on the coastal strip, where they had prepared a complex network of bunkers and fortifications and where they ultimately mounted their final battle. Like other instances of guerrilla insurgency in other conflicts, they positioned their artillery batteries among the civilians and prevented many of them from leaving the area once the shelling began, in some cases shooting those who tried to exit the zone.
While the Sri Lankan military claimed that it was engaged in “humanitarian operations” aimed at “liberating the civilians,” it was in fact reclaiming the northeast tip of the island that was still in Tamil hands. An analysis of satellite images as well as numerous testimonies revealed that the Sri Lankan military continuously pounded the enclosed land with mortar and artillery fire, transforming the designated no-fire zones into killing fields. Between 10,000 to 40,000 caged-in civilians perished in the so-called safe zones, while thousands more were severely injured, often lying for hours or days on the ground without receiving medical attention because virtually every hospital—whether permanent or makeshift—had been hit by artillery.
The Sri Lankan civil war was subjected to international legal scrutiny. Following the government’s defeat of the Tamil Tigers, a panel of experts appointed by the U.N. secretary-general published a damning report accusing both the Tamil Tigers and the Sri Lankan government of having carried out war crimes and crimes against humanity. In addition to charges of murder, mutilation, cruel treatment, and torture, the panel accused the Tamil Tigers of using civilians as a human buffer and killing civilians attempting to flee the no-fire zones. The report charged the Sri Lankan military with intentional, indiscriminate, and disproportionate attacks on civilians, starvation and denial of humanitarian relief, and attacks on medical and humanitarian objects.
Computational Acrobatics
In response, the Sri Lankan president established a governmental commission to examine the allegations and weigh in on the matter. To bolster its legal defense, the commission enlisted several leading experts in international humanitarian law from the United States, the United Kingdom, and Canada. These experts provided legal opinions exonerating the government and the military from the accusation that they had committed war crimes.
While the commission as well as the experts used international law’s human-shields clauses to analyze the actions of the Sri Lankan government, they interpreted the act of human shielding in a relatively novel way. One opinion supporting the Sri Lankan government’s stance was written by David Crane from the College of Law at Syracuse University and Sir Desmond de Silva from the United Kingdom, both of whom had served as chief prosecutors in the international war crimes tribunal for Sierra Leone. These two experts reveal how legal calculations can be used to justify excessive civilian deaths.
After claiming that the government forces never intended to strike civilian objects and were merely returning fire against enemy targets embedded amidst civilians and close to hospitals, Crane and de Silva accuse the Tamil Tigers of employing thousands of civilians as involuntary human shields. These civilians, they explain, cannot be considered as taking an active part in hostilities, and thus their presence would have to be weighed on the civilian side of the proportionality scale. However, they go on to claim, factoring involuntary human shields as civilians within the proportionality equation enables those who deploy the shields “to profit from a clear violation of the laws of war, and thus should not be allowed.” According to this rationale, belligerents would be incentivized to use civilians as human shields—an act that is prohibited by laws of war and considered an inhumane method of warfare—to deter their enemy from attacking.
Crane and de Silva add that there appears to be consensus among international legal experts supporting the notion that casualties resulting from the use of involuntary human shields “are at least somewhat diminished in the proportionality analysis,” by which they mean that in the proportionality equation a dead involuntary shield is worth less than a dead civilian. In this sense they reinforce Yoram Dinstein’s claim that “the actual test of excessive injury to civilians must be relaxed” in cases of human shielding, while also anticipating the 2015 U.S. law of war manual (and to a lesser extent in the updated version) which intimates that condoning the use of human shields would translate into a military disadvantage for the attacking forces and that human shielding alters the way proportionality is calculated.
Lethal Proportionality
Up to this point, Crane and de Silva appear to concur with the other legal experts who provided advice to the Sri Lankan government, but then Crane and de Silva do the math, revealing, as it were, how the calculations enable them to adopt an extreme stance. They begin by assessing the weight of the military advantage:
First, the humanitarian operation launched by the [government of Sri Lanka] was justified by a host of compelling military objectives, namely ending the nearly 30 year campaign of violence by the [Tamil Tigers] which included assassinations on duly elected officials and attacks on civilian objects. . . . It is clear the termination of such insidious and wholesale threats to civilian life represents a compelling military objective which already sets the bar fairly high relative to the acceptable level of civilian casualties in achieving that objective. This is a factor that could weigh heavily in favor of a finding of proportionality on behalf of [the government’s] operations overall as this is a factor which must be put into the balance of the proportionality equation.
Next, they factor civilian deaths:
Even taking the highest figures ascribed to the deaths of Vanni civilians [a primarily Tamil population living in the mainland area of the Northern Province of Sri Lanka], assuming that there were up to 330,000 civilians in the [no-fire zone] as the [U.N.] Report contends—7,000 of whom were killed—this presumes a loss of life of approximately 2% of that civilian population. . . . If there were as many as 40,000 killed, this would be a loss of approximately 12% of that population. Whatever the figure in terms of a hostage rescue operation where some 295,000 were saved—it is a successful operation.
In fact, this proportionality analysis is legal maneuvering, and not only because the authors leave out of their legal arithmetic important forms of civilian harm, such as injury, displacement, economic loss, and destruction of civilian objects. Crane and de Silva assume that incidents of involuntary human shielding are identical to hostage-rescue operations by stating that the 295,000 civilians who survived “were saved.” The computational ruse is clear: once one supposes that all civilians could have died and that only 12 percent were in fact killed, then the operation can be viewed as successful. Their argument follows the logic of the lesser evil, justifying the Sri Lankan attacks against civilians by claiming that a greater evil could have been carried out and was ultimately warded off. This line of thinking transforms the proportionality computations into a tool for legitimizing the wide-scale killing of civilians, because one can always find a greater evil with which the harm that has been carried out can be compared. Ultimately, in their eyes, it does not really matter what the military advantage was or how many civilians were killed, so long as some civilians were “saved” as a result of the operations.
Crane and de Silva’s legal gymnastics that render the civilians in the no-fire zones killable becomes even clearer when they argue that not all the civilians in these spaces were involuntary human shields. “As a matter of logic,” they maintain, “there is a powerful case for saying that it is extremely unlikely that some 20,000 cadres of [Tamil Tigers] could have taken up to 330,000 hostages against their will. The probability is that a large section of the civilians went voluntarily with the Tamil Tigers in order to play a part, albeit passive,” in the war effort. The two experts go on to suggest that these alleged voluntary human shields directly participated in hostilities and therefore should not be afforded the protections offered to civilians. Thus, when assessing the balance between military advantage and civilian harm, the deaths of these civilians should not be counted.
Crane and de Silva’s effort to compute civilian deaths in relation to military advantage is not entirely exceptional and reveals how proportionality can be operationalized by legal advisors in theaters of violence across the globe. Moreover, the way all of the legal experts hired by the Sri Lankan commission made use of human shielding in their arguments highlights the central role human shields have come to play in these calculations. The decrease of the legal value of the civilians who were killed as human shields is a strategy that has become increasingly common in recent years.
Some legal commentators maintain that proportionality is always an ambiguous concept because several significant variables cannot be quantified, yet when an incident is disproportionate, that imbalance cries out for all to see. However, violence is always open to interpretation, and, as past conflicts have taught us, widespread killings always generate political debates about the use of lethal force and whether it was driven by an ethics of humane violence. In Sri Lanka, prominent legal experts mobilized the legal figure of the human shield in an analytic exercise and their proportionality calculations ended up legitimizing the killing of many thousands of civilians.