Context

The Fourth Eelam War (2006–2009) was the last iteration of full-scale hostilities between the LTTE and the Sri Lankan government. In this ethnonationalist civil war, the LTTE strove to carve out a separate Tamil state on the Sinhalese-majority island.Footnote ¹⁸ Prior to the outbreak of the 2006 conflict, the LTTE controlled a de facto state in the Northern and Eastern Provinces and was party to a Norwegian-brokered cease-fire agreement.Footnote ¹⁹ This agreement was technically still in place until January 2008, although the Sri Lanka Monitoring Mission acknowledged that by 2007 an all-out war was occurring.Footnote ²⁰

While previous iterations of the conflict had ground to a stalemate, this time the government possessed a stronger hand. The LTTE's reliance on terrorist tactics had created a host of international enemies. India listed it as a terrorist organization in 1992 and the leadership remained a wanted entity in India over the murder of Rajiv Gandhi.Footnote ²¹ The LTTE was later proscribed in the United States (1997), the U.K. (2001), Canada (2006), and the EU (2006).Footnote ²² These proscriptions were significant in that they instigated law enforcement action such as the freezing of financial assets.Footnote ²³ It is also likely that because the international community viewed the LTTE through the terror lens, the Sri Lankan government was more confident that it could characterize the final onslaught as a hostage rescue.

The LTTE was further weakened by the 2004 defection of the Karuna faction, a group of former LTTE fighters from the Eastern Province.Footnote ²⁴ Meanwhile, the government received a steady supply of military equipment from China and Pakistan and vital maritime intelligence from India and the United States.Footnote ²⁵ In July 2007, Sri Lanka claimed it had recaptured the Eastern Province and it declared an intention to defeat the LTTE in the north.Footnote ²⁶ In late 2008, government forces advanced on the LTTE capital of Kilinochchi and the conflict entered its most deadly phase.Footnote ²⁷ The fighting culminated in the death of LTTE leader Velupillai Prabhakaran and the decimation of his remaining cadres.Footnote ²⁸ The following section of the article provides a brief summary of these final months of the conflict. Existing accounts express a common theme—that Sri Lanka's actions and the absence of consequences have been detrimental to IHL.Footnote ²⁹

As the Sri Lankan armed forces approached Kilinochchi in September 2008, there was a marked decline in international oversight of the conflict. The government announced it could no longer ensure the safety of humanitarian workers in the region and instructed UN and international nongovernmental organization staff to leave their offices in Kilinochchi. Later that month, the UN relocated its offices to the city of Vavuniya in government-held territory. Many other international organizations also withdrew their staff.Footnote ³⁰ From that juncture onward, there were virtually no international observers able to report on what was happening in the conflict zone.Footnote ³¹ As I show in the next section, the decrease in transparency corresponds to a series of denials about factors such as the casualty rate. By early January 2009, government soldiers had marched into Kilinochchi.Footnote ³² By January 25, they had captured Mullaitivu just south of where the conflict would end.

One of the distinguishing features of this phase of the conflict is the number of civilians who were at risk. As the international observers left and the government forces advanced, hundreds of thousands of civilians were caught up in the LTTE's retreat.Footnote ³³ There are a number of explanations for why civilians felt compelled to accompany the LTTE. Tamil civilians in the Northern Province feared what would happen if they went into government-controlled areas, and the LTTE's forced recruitment policy meant many of these civilians would have had relatives among the combatants.Footnote ³⁴ Weiss suggests that many civilians believed the LTTE would mount a successful counterattack, enabling them to return to the normalcy of life in LTTE-held territory.Footnote ³⁵ DeVotta maintains that the LTTE usually forced civilians to accompany the group whenever it retreated.Footnote ³⁶ Finally, to reach government-held territory, the civilians often needed to move in the direction of artillery fire, combat, and minefields.Footnote ³⁷

The retreat brought LTTE and the civilians to a narrow strip of land between a lagoon, the Indian Ocean, and the government-held town of Mullaitivu.Footnote ³⁸ According to the UN Secretary-General's Panel of Experts, by mid-April 2009, 170,000 civilians had left the coastal strip in the midst of heavy shelling.Footnote ³⁹ When the LTTE leadership was killed in May 2009, another 290,000 civilians emerged.Footnote ⁴⁰ The sheer number of internally displaced people (IDP) created a humanitarian crisis of its own. The army struggled to provide adequate shelter, food, clothes, and medical care for the escaping civilians, but it reportedly resisted UN requests for full access to the IDP camps.Footnote ⁴¹ Between January and May 2009, as many as 40,000 civilians died.Footnote ⁴² The scale of the crisis is one of the reasons that observers expressed concern about the ramifications of Sri Lanka's actions.Footnote ⁴³

While there was limited access to the conflict zone, both during and soon after the fighting, a range of actors moved to detail the alleged abuses. In April 2009, the U.S. National Geospatial-Intelligence Agency released images of the conflict zone. These images buttressed claims that Sri Lanka was deliberately underestimating the number of trapped civilians.Footnote ⁴⁴ In response, the U.S. Congress requested a report from the U.S. Secretary of State.Footnote ⁴⁵ In June 2010, UN secretary-general Ban Ki-Moon appointed a panel of experts to examine the nature and scope of alleged violations in the “final stages of the war.”Footnote ⁴⁶ Nongovernmental accounts did much to raise the profile of the allegations,Footnote ⁴⁷ while more recent United Nations Human Rights Council proceedings have centered on Sri Lanka's domestic accountability processes.Footnote ⁴⁸ In sum, there is a considerable volume of work establishing the nature of the offenses in the final stages of the conflict.

While both belligerents committed an array of offenses, the government's win-at-all-costs approach was the primary cause of civilian suffering.Footnote ⁴⁹ The UN Secretary-General's Panel of Experts gave this account of the government's shelling practices:

Moreover, the panel alleged that this shelling had caused the majority of civilian casualties.Footnote ⁵¹ It also provided this account of the LTTE's conduct:

As I will show in the final section of this article, the government of Sri Lanka cited credible evidence of LTTE crimes in a bid to absolve itself of any responsibility for the scale of civilian suffering. Nevertheless, it violated the rules of proportionality and discrimination significantly.

The relative absence of any immediate consequences is a defining feature of this historical episode. China and Russia supported Sri Lanka at the UN Security Council, where the crisis did not become the subject of a formal agenda.Footnote ⁵³ The Responsibility to Protect remained effectively unused as the permanent members continued to dispute its meaning and application.Footnote ⁵⁴

At the UN Human Rights Council, Sri Lanka and its supporters blocked the passage of a resolution calling for an independent investigation.Footnote ⁵⁵ Instead, the council passed Sri Lanka's draft resolution that welcomed “the liberation by the Government of Sri Lanka of tens of thousands of its citizens that were kept by the Liberation Tigers of Tamil Eelam against their will as hostages.”Footnote ⁵⁶ In this context, the EU's decision to revoke Sri Lanka's trading privileges is an exception to a general pattern of inaction.Footnote ⁵⁷ In the short term, the Sri Lankan government did not encounter the degree of reputational or material costs it could well have borne.

Nor has there been significant progress on accountability in the years since the end of the conflict. The Sri Lankan political landscape remains dominated by the same actors responsible for prosecuting the war. Most visibly, the former defense secretary Gotabaya Rajapaksa is the current president, while his brother Mahinda Rajapaksa (the former president) is the current prime minister. A 2021 Office of the United Nations High Commissioner for Human Rights report instead details six trends that characterize Sri Lanka's domestic political order since the war. These are militarization of civilian government functions, a reversal of constitutional safeguards, political obstruction of accountability, majoritarian and exclusionary rhetoric, a shrinking of democratic space, and new and exacerbated human rights concerns.Footnote ⁵⁸ This report preceded the most recent UN Human Rights Council resolution providing the human rights commissioner with additional funding to collect information and evidence with a view to facilitating future accountability processes.Footnote ⁵⁹ While the latest UN Human Rights Council resolution is evidence that Sri Lanka has not been able to escape international scrutiny, it is also symptomatic of the fact that these international processes have struggled to meaningfully change the course of domestic politics in Sri Lanka itself.

Many commentators have viewed Sri Lanka's actions as an attack on IHL and the broader human rights regime. Former Finnish president and international mediator Martti Ahtisaari suggested that “countries operating outside international norms” would be “taking courage from Sri Lanka's apparent success at avoiding international reproach.”Footnote ⁶⁰ Steven Ratner placed an emphasis on the scale of the crisis and the absence of an international response:

For Ratner, the size of the protection failure and the muted response compromised the principle of accountability. The International Crisis Group came to a similar conclusion:

These accounts convey a sense that Sri Lanka flouted IHL on an enormous scale and that by escaping any serious or immediate international pushback, the episode weakened IHL and the human rights regime. In the wake of the conflict, the international community could expect more noncompliance with the law. This article accepts the proposition that Sri Lanka challenged the legitimacy of IHL. Undoubtedly, the muted international response also enhanced the effect. However, I suggest that by adopting a conceptual account of what it means for the law to lose standing, we can show that Sri Lanka's approach was both more subtle and more detrimental to the law than many of the preceding accounts have intimated.

My account can also be differentiated from Gordon and Perugini's work on Sri Lanka's use of legal experts, which suggests that after the UN Secretary-General's Panel of Experts published its report on the conflict in 2011, Sri Lanka sought out legal opinions to vindicate its actions.Footnote ⁶³ As this article will show, Sri Lanka's denialist challenge to the law starts earlier than the period under examination in these authors’ work and it is functionally different from the performative challenge they observed later. The Sri Lankan government denied that it was killing civilians during and immediately after the conflict. However, after the publication of the Panel of Experts’ report in 2011, the government attempted to fit the highest civilian casualty estimates under the umbrella of proportionality.Footnote ⁶⁴ In this sense, Gordon and Perugini help us to identify where Sri Lanka's denialist challenge is supplanted by another form of contestation.

Conceptualizing Norm Contestation

To understand how Sri Lanka launched an indirect, acute challenge to IHL, we need to address a conceptual question: How does IHL or any legal institution come to lose standing? This section answers this question by engaging two constructivist debates. The first concerns the role of compliance in relation to norm strength. In contrast to approaches that view norm strength in terms of the rate of compliance, this article treats norm strength as a function of legitimacy or standing. The second concerns the question of whether challenges to the scope of a norm can erode its legitimacy. This article accepts that attacks on a norm's validity or righteousness can corrode its standing. However, when the attacks diminish the collective expectation that actors must conform to a norm, attacks on a norm's scope can also undermine its standing.

Assessing compliance is one obvious way to examine norm health. In his work on legal regimes, Glennon advances just this formulation:

For Glennon, a low rate of compliance is a good indication that a norm is ailing. According to this view, to the extent that Sri Lanka's bombing campaign is part of a broader pattern of noncompliance, we can say the IHL regime is unhealthy.

Yet while compliance is an intuitively appealing way to measure norm health, it is also problematic. There are at least two reasons this is the case. First, the concept of compliance is descriptive. It tells us nothing about why actors comply with the norm in the first place.Footnote ⁶⁶ If widespread compliance is a product of convergent interests, for example, it is a leap to say the norm is healthy. Second, the compliance approach cannot make sense of circumstances where a norm is both widely accepted and frequently violated. The norm against murder is an example. By the terms of the compliance approach, we would be led to conclude that the norm is weak. After all, murder happens every day. However, this seems like an unsatisfactory finding.Footnote ⁶⁷ If compliance is a poor indicator of norm health, what should analysts instead look for?

Rather than focusing on compliance, other scholars have looked to legitimacy. We can find variations on this interest in legitimacy in an array of constructivist accounts.Footnote ⁶⁸ For instance, Hirsch and Dixon examine the extent of collective expectations that the norm matters.Footnote ⁶⁹ Similarly, Clarke et al. call for a focus on whether the norm enjoys international standing.Footnote ⁷⁰ Deitelhoff and Zimmerman take compliance seriously, but they also introduce the concept of “facticity” to establish whether a given norm is guiding action. Facticity is complemented by “validity,” which refers to the degree of verbal acceptance of the norm.Footnote ⁷¹ The advantage of these approaches is that they do not conflate behavior with the question of whether actors collectively expect compliance. If we return to the norm against murder, this opens up the possibility of achieving a more satisfactory account of the norm's health. Although murder occurs every day, because the norm against murder enjoys legitimacy, it continues to motivate compliance by other actors.

If a norm's health is contingent upon its standing, then reason giving matters. Consider Sandholtz's discussion of justificatory strategies:

In an appeal to exceptional circumstances, the actor implicitly acknowledges the legitimacy of the norm in question. Whatever the nature of an actor's behavior, it has left the standing of the norm intact. Consequently, if a norm retains its standing, it will continue to guide the behavior and claim making of other actors.

Constructivists do not agree on which claims are likely to harm norms. One key debate concerns the effects of what Deitelhoff and Zimmerman call “applicatory and validity contestation.”Footnote ⁷³ Applicatory contestation occurs when actors debate whether or how a norm should apply to a given circumstance.Footnote ⁷⁴ By contrast, validity contestation occurs when actors challenge the righteousness of a particular norm.Footnote ⁷⁵ To provide examples, we can see applicatory contestation in debates about when it is appropriate to protect intellectual property. Within these public debates, the validity of intellectual property is not in question. By contrast, in the 1980s activists contested the validity of granting political amnesty to war criminals after violent conflict. In effect, they argued that the norm of granting amnesty was unrighteous. Deitelhoff and Zimmerman argue that validity contestation can precipitate norm weakness but applicatory contestation is likely to strengthen norms.Footnote ⁷⁶ Indeed, the argument makes a good deal of sense. An attack on a norm's validity is almost certainly an attack on its legitimacy. Meanwhile, other accounts have empirically shown how applicatory contestation has clarified a norm's scope and specificity.Footnote ⁷⁷ But is it correct to say that there are no circumstances in which applicatory contestation presents a challenge to norms?

Elsewhere, theorists have shown how certain forms of applicatory contestation are indirect pathways to norm weakness. For instance, Panke and Petersohn argue that if actors make claims that limit the scope of a norm, then over time they can render it obsolete.Footnote ⁷⁸ By this account of contestation, Búzás's concept of “norm evasion” is also bad for norm health. In norm evasion, actors legally institutionalize norms in a way that allows them to sidestep the very purpose of the norm itself.Footnote ⁷⁹ Using Panke and Petersohn's interpretation, if norm evasion continues to occur, actors will come to expect that they have no obligation to the norm. In other words, the norm will lose standing because actors will cease to collectively believe that it must be followed.

At this point, it is useful to introduce Clark et al.'s “rejectionist,” “revisionist,” and “denialist” challenges. While the rejectionist challenge can be understood as validity contestation, the revisionist and denialist challenges are forms of applicatory contestation. Importantly, and contrary to the suggestion of Deitelhoff and Zimmerman, all three challenges are acute in the sense that they undermine the standing of IHL's core principles or bargain. A “bargain” is an agreement about the foundations of a legal regime. It concerns “the nature of the governance domain, who the legitimate actors are, and how rights and responsibilities are distributed.”Footnote ⁸⁰ Consider this account of IHL's bargain:

Here, we can see the legitimate actors (combatants), the domain (war as opposed to other forms of violence), the rights (a right to pursue a military victory and to engage in a measure of force protection), and the obligations (an obligation to enact humanitarian restraints).Footnote ⁸² If a bargain enjoys legitimacy, it motivates compliance. However, rejectionist, revisionist, and denialist challenges can all undermine a bargain's standing and the chance that states will adhere to it.

In a rejectionist challenge, actors directly attack the foundations or bargain of a legal regime. Here, Clark et al. cite the beheadings of prisoners committed by the militant organization ISIL (Islamic State in Iraq and the Levant). By executing prisoners, ISIL is breaking the laws of war. In disseminating graphic videos of the process, the organization is publicly celebrating its act of noncompliance. The organization presents audiences with a war where IHL's central principles have no standing.Footnote ⁸³ This is a validity challenge insofar as ISIL presents its own ideology as the only legitimate source of norms.

In a revisionist challenge, actors set out to radically redefine the meaning of the law's foundational categories. For instance, in the early days of the War on Terror, U.S. lawyers coined the term “unlawful combatants” to be used instead of “former combatants.”Footnote ⁸⁴ The lawyers were trying to revise the meaning of an entire category of IHL to avoid the legal obligations of the United States under the Geneva Conventions.Footnote ⁸⁵ Another potential example (though not identified by Clark et al.) is Israel's use of the term “administer” in place of “occupy” following the 1967 Six-Day War. Israel granted itself the leeway to militarily control Gaza and the West Bank while avoiding its obligations as an occupying power under the fourth Geneva Convention of 1949.Footnote ⁸⁶ The examples are applicatory in that they alter the norm's scope through acts of revision.

Last, and most relevant for our purposes, there is the denialist challenge. Here, Clark et al. provide Russia's approach to IHL as an example: Russian foreign policy officials have portrayed their state as the chief defender of international law. By contrast, Human Rights Watch alleges that Russian-backed rebels in Ukraine are behind a series of IHL violations.Footnote ⁸⁷ Russia counters this criticism by simply denying that it is involved in the conflict. In the story that Russia invents, there are no opportunities for the international community to assess its conduct by the terms of IHL. This brings us to the essential quality of a denialist move: Actors invent a new set of facts to limit the reach of the legal regime's central principles.

This section has positioned the article's approach to norm contestation with respect to two key debates in the literature. First, working against the temptation to conflate norm strength with compliance, this article treats norm strength as a function of legitimacy. Second, against the view that applicatory contestation cannot undermine a norm, this article sides with those accounts that argue there are instances in which applicatory contestation can render a norm obsolete. Denialism is an example of such contestation. Applying these arguments, it is now possible to explore how Sri Lanka mounted such an attack on the IHL regime.

Sri Lanka's Denialist Move

Sri Lanka's denialist challenge to IHL encompasses more than just its conduct. It also includes its efforts to invent a new set of facts about the conflict's close. During the final months of the Fourth Eelam War, the Sri Lankan government maintained that the civilians caught up in the LTTE retreat were rebel hostages, with the implication being that the government was conducting a hostage rescue.Footnote ⁸⁸ Over the course of this military operation, it disputed criticism and pressured potential critics into withholding information.Footnote ⁸⁹ Sri Lanka's approach to explaining the civilian casualty rate tied these strategies into a denialist challenge in that it curtailed the reach of IHL's central bargain. Sri Lanka maintained that while the LTTE was using civilians as “human shields,”Footnote ⁹⁰ Sri Lankan armed forces were pursuing a zero civilian casualty approach.Footnote ⁹¹ With this move, Sri Lanka shifted the responsibility for the deaths of the civilians onto the LTTE. The government absolved itself of its legal obligations to enact restraints in the form of proportionality and discrimination. The remainder of this section provides an account of Sri Lanka's hostage rescue narrative and its attempts to deny any responsibility for the deaths of civilians. I show how this representational strategy constitutes a denialist challenge to IHL.

When a state of full-scale hostility was resumed between the Sri Lankan armed forces and the LTTE in May 2006, the government framed its military operation as an effort to ensure civilian access to water, and then as an attempt to pressure the LTTE back to the negotiating table.Footnote ⁹² The government's justificatory strategy began to change when it declared a military victory over the Eastern Province in July 2007.Footnote ⁹³ At this juncture in the conflict, Sri Lankan officials felt ready to assert that their military operation could “liberate” the remaining LTTE-held territory.Footnote ⁹⁴ In January 2008, Sri Lanka abandoned the cease-fire process.Footnote ⁹⁵ The government continued to develop its narrative of liberation for the remainder of the conflict.

In his speech to the UN General Assembly in September 2008, Sri Lankan president Mahinda Rajapaksa supplied his audience with the metaphor of a hostage scenario.Footnote ⁹⁶ At that time, government forces were advancing toward the LTTE's de facto capital of Kilinochchi,Footnote ⁹⁷ and they had already instructed the UN to leave the city.Footnote ⁹⁸ Hundreds of thousands of civilians were on the move.Footnote ⁹⁹ Rajapaksa drew on the LTTE's reputation as a terrorist organization to present his government's action as a matter of duty:

Other Sri Lankan administration officials treated the president's statement as the default account of the government's rationale. Indeed, in front of their respective foreign audiences in Europe and Australia, the ambassador to the EU and the foreign minister directly quoted this passage from the president.Footnote ¹⁰¹ As Gordon Weiss remarked, Sri Lankan officials were doing their best to “sell” the concept of the hostage rescue to the international community.Footnote ¹⁰² Of course, many civilians were accompanying the LTTE because they feared what would happen to them in government-held territory.Footnote ¹⁰³ Nonetheless, the government would ultimately celebrate what it described as a successful rescue mission.

After the deaths of the LTTE leadership, Sri Lankan officials assumed a triumphant note. While giving an address to the diplomatic community in Colombo in May 2009, Rajapaksa offered this appraisal of the military operation:

The president was depicting a near perfect hostage rescue. Speaking at the Shangri-La Dialogue in late May, Foreign Minister Rohitha Bogollagama also called on his audience to celebrate the Sri Lankan military operation:

In this narrative of a triumphant hostage rescue, we can already see part of the logic of the Sri Lankan government's denialist move: the government liberated the trapped civilians, while the LTTE had endangered them. By implication, the hostage rescue could not be evaluated by the criteria of proportionality and discrimination because the LTTE was responsible for the casualties. As discussed in the first section of the article, the UN Human Rights Council then effectively endorsed Sri Lanka's account. In the government's narrative, the final months of the conflict entailed a delicate and successful hostage rescue. This was the icing on the cake in “a war of liberation.”

Sri Lankan officials fervently contested any evidence that could unsettle their hostage rescue narrative.Footnote ¹⁰⁶ There were four particularly pertinent issues at stake: the number of trapped civilians, the nature of the bombing tactics, the civilian casualty rate, and, lastly, the question of how civilians were being endangered. I argue that Sri Lanka's answer to this fourth question makes its approach into a denialist challenge to IHL.

While the president ultimately took credit for rescuing 200,000 people, Sri Lanka's original approach involved underestimating the number of civilians caught up in the LTTE's retreat.Footnote ¹⁰⁷ At the beginning of April 2009, the government maintained that there were only 10,000 trapped civilians. By contrast, drawing on satellite imagery, the UN believed that at the end of April the figure was actually 127,177.Footnote ¹⁰⁸ The UN Secretary-General's Panel of Experts alleged the disparity between UN and government figures was a deliberate part of Sri Lanka's strategy. It allowed the government to restrict the volume of aid and enhance the severity of the siege.Footnote ¹⁰⁹

Additionally, the approach protected Sri Lanka's narrative of a hostage rescue. The more civilians were trapped, the more civilians the government was endangering by seeking to eliminate the LTTE.

Sri Lankan officials repeatedly challenged assertions that they were shelling the conflict zone in an indiscriminate fashion. On February 25, 2009, the government announced it was no longer using heavy weaponry in its designated no-fire zones.Footnote ¹¹⁰ During his interview with the BBC's HARDtalk in March of that year, the minister of disaster management and human rights, Mahinda Samarasinghe, had also given this account of how the government forces were fighting:

According to Samarasinghe's account, Sri Lankan military personnel were accepting a high level of personal risk in order to root out the LTTE from the civilian population. In fact, successive reports have concluded that throughout the hostilities of 2009, the Sri Lankan army was shelling in all three of the no-fire zones.Footnote ¹¹² As noted, of everything that occurred in the final months of the conflict, government shelling killed the largest number of civilians.Footnote ¹¹³ Moreover, by encouraging civilians to move to the no-fires zones and then shelling them anyway, government forces were not only failing to take precautionary measures, they were manipulating the requirements for precaution.Footnote ¹¹⁴ However, Sri Lankan officials were prepared to issue a series of statements defending a “clean” account of their conduct.

The civilian casualty rate was also a sensitive issue. The UN secretary-general's Internal Review Panel gives a telling insight into the relationship between UN agencies and the Sri Lankan government on this topic, noting that the “Government's harsh reaction to even the suggestion that there were civilian casualties led the UN in Colombo to limit the sharing of information on the casualties.”Footnote ¹¹⁵ Indeed, by early 2009, UN officials were withholding what they regarded as conservative casualty estimates.Footnote ¹¹⁶ UN High Commissioner for Human Rights Navanethem Pillay broke the pattern in March and said that “more than 2,800 civilians” had been killed.Footnote ¹¹⁷ In response, Sri Lanka simply told the European Parliament Subcommittee on Human Rights that the figure could not be substantiated.Footnote ¹¹⁸ Much like it had done in its efforts to underestimate the number of trapped civilians, the government was protecting its hostage rescue narrative by challenging any conflicting civilian casualty estimates.

So far, this account of Sri Lanka's actions has not necessarily demonstrated how it produced an acute attack on the legitimacy of IHL. Based on the preceding evidence, Sri Lanka's actions and rhetoric look more like a troubling example of noncompliance than an acute challenge to the laws of war. However, this article submits that the preceding actions and representational strategies were tied together by Sri Lanka's account of how civilians were being killed. The result is a denialist challenge to IHL.

By the government's account, to the extent that civilians were in danger, it was because the LTTE was exploiting the government's diligence. Officials repeatedly admonished the LTTE for using civilians as human shields.Footnote ¹¹⁹ As Bogollagama told his audience at the Shangri-La Dialogue in Singapore:

By this account, the government played the role of the valiant rescuer, while the LTTE was deliberately targeting civilians in a bid to avoid military defeat. As discussed, although there was credible evidence that the LTTE had put civilians in danger or directly targeted them,Footnote ¹²¹ Sri Lankan officials were not prepared to countenance the suggestion that civilians could have died because of government action.

The fate of Convoy 11 provides the clearest example of this denialist strategy. Between October 2008 and January 2009, the UN sent eleven convoys of humanitarian aid by road to the conflict zone.Footnote ¹²² However, this practice ceased in late January when the eleventh convoy was shelled from government positions.Footnote ¹²³ Unlike much of the shelling during the final stages of the conflict, two international UN officers witnessed the incidentFootnote ¹²⁴ and reported it directly to the UN headquarters in Colombo.Footnote ¹²⁵ The UN staff had transmitted their positions to the Sri Lankan government before and during the shelling; they were inside the first no-fire zone and they were surrounded by civilians.Footnote ¹²⁶ In response to this testimony, the government maintained that if the convoy and the safe zone had been attacked, it was from LTTE fire.Footnote ¹²⁷ In the Convoy 11 incident, the government dismissed the eyewitness testimony of UN officials and replaced it with a version of events in which the LTTE was responsible.

Alongside this discussion of the LTTE's tactics, Sri Lankan officials went as far as to suggest that their own military operation was designed to avoid civilian casualties altogether. While briefing the diplomatic community in May 2009, Rajapaksa introduced the zero civilian casualty claim:

Like the concept of the hostage rescue, this account became a standard government and military position.Footnote ¹²⁹ The ubiquity of the claim led Gordon Weiss to make this observation:

The zero civilian casualty approach proposed by Sri Lanka is arguably the starkest point of contrast from what actually happened in Sri Lanka's indiscriminate bombing of civilians. Importantly, the human shield argument and the zero civilian casualty claim are two sides of the same coin: Sri Lanka was attempting to shift all the blame for civilian casualties onto the LTTE. If Russia's denialist strategy in Ukraine was to say, “It's not us, we're not there,” then Sri Lanka's denialist strategy was to say, “It's not us, it's them.”

It was only much later that Sri Lanka acknowledged that its shelling strategy could have resulted in a high number of civilian casualties. After the UN Secretary-General's Panel of Experts published its 2011 report, the Sri Lankan government commissioned a string of lawyers to provide opinions on the case. At this point, Sri Lanka's strategy shifted to what Gordon and Perugini have described as an act of “computational acrobatics.”Footnote ¹³¹ By the terms of its newly sought-out legal opinions, if Sri Lanka had killed as many as 40,000 civilians, this was proportionate to the 295,000 civilians it had “saved.” Because the LTTE had used civilians as shields, Sri Lanka was entitled to adjust the ratio of acceptable civilian casualties. The legal opinions made no mention of the conditions that awaited the displaced civilians or the fact that many of them hoped to never encounter government forces at all. The shift in strategy is remarkable given that at the start of the onslaught, Sri Lankan officials had deliberately underestimated the number of trapped civilians and maintained their strategy was one of zero civilian casualties.Footnote ¹³² The change in justificatory strategy also marks the end point for Sri Lanka's denialist challenge to the laws of war. The government went from denying they could have killed civilians to stretching the criteria of proportionality to account for the civilian casualty rate.

Having outlined the features of Sri Lanka's denialist move, we should contend with some alternative ways of conceptualizing its approach. I group these alternatives into two categories: the “rejectionist” challenge and the “war as hell” doctrine. First, as noted, the rejectionist challenge is an overt attack on the law. If Sri Lanka had engaged in such a challenge, its approach would be one of boastful noncompliance. Yet Sri Lanka did not celebrate violating the law. It celebrated the so-called success of its hostage rescue and then it was able to publish this narrative in its very own UN Human Rights Council resolution.Footnote ¹³³ Sri Lanka did not make its actions appear consistent with the law, but it did not provide an outright rejection either.

The war is hell doctrine is the second alternative conceptualization of Sri Lanka's approach. Michael Walzer associates this kind of logic with General Sherman's campaign in the U.S. Civil War.Footnote ¹³⁴ While fighting Confederate forces in the South, Sherman forcibly evacuated the city of Atlanta and then destroyed it. In response to protests from another Union general, he noted that the leaders of the Confederacy had created the war. On this basis, he reasoned that the Confederate leaders were responsible for bringing “hell” to Atlanta.Footnote ¹³⁵ On the face of it, this conceptualization is a promising alternate avenue for thinking about how Sri Lanka framed its human shield problem. By using civilians as human shields, the LTTE alone had created hell. However, again, when we compare this conceptualization to Sri Lanka's representational strategy, we can see a crucial point of divergence. Unlike Sherman, Sri Lanka never accepted anything other than its hostage rescue account. The war is hell analogy does not offer a convincing alternative account of how Sri Lanka engaged with the laws of war because Sri Lanka invented a narrative where the civilians had been saved.

In describing the Sri Lankan justificatory strategy as denialist, this article sees Sri Lanka's action as one of multiple denialist challenges to the law. In outlining the concept of denialism, this article has already touched on Russia's efforts to prevent the laws of war from applying to the Ukrainian conflict. Russia's support for the Syrian regime is also characterized by invention. For instance, when the UN Security Council met to discuss the 2016 bombardment of Aleppo, Russian ambassador Vitaly Churkin dismissed accusations of indiscriminate bombing as “fake news.”Footnote ¹³⁶ Similarly, following the emergence of allegations of a chemical weapons attack on the town of Douma, Russian foreign minister Sergey Lavrov claimed his government had “irrefutable” evidence the events were staged.Footnote ¹³⁷ Like the Sri Lankan case, these statements seal military practices off from the remit of IHL. Collectively, these practices indicate a decline in the health of the laws of war.

In Sri Lanka's narrative, the military operation was a hostage rescue that brought democracy to an area long under the control of a terrorist organization.Footnote ¹³⁸ As part of this portrayal, the government underestimated the number of trapped civilians, refuted allegations that it was using heavy weaponry, and contested casualty estimates.Footnote ¹³⁹ Sri Lanka also engaged in a concerted effort to prevent independent access to the region, and it used the imperative of aid delivery to coerce the UN into withholding casualty estimates.Footnote ¹⁴⁰ Crucially, Sri Lanka's efforts to control the narrative were tied together by one important claim—that to the extent that civilians were being harmed, the LTTE was to blame.Footnote ¹⁴¹ With this move, Sri Lanka attempted to prevent its actions from being assessed by the criteria of proportionality and discrimination. As a result, it sealed its military operations off from the central bargain of IHL.