Chapter 6. Security Forces During the June Violence
Section 1. Police and Lon Htein: From Failure to Protect to Alleged Participation
The role of Myanmar's security institutions during the June 2012 violence cannot accurately be described as though the police, Lon Htein, NaSaKa and Tatmadaw formed a single force acting identically in every location. They had different commands, responsibilities and geographic roles, and the evidence concerning their conduct varies. Rohingya testimony nevertheless contains a persistent allegation running through the June crisis: institutions that were legally responsible for protecting civilians frequently failed to protect Rohingya and, in a number of documented cases, members of those institutions were accused of participating directly in violence against them. Later investigations by Human Rights Watch, Amnesty International and other organisations provided substantial independent evidence supporting important elements of those Rohingya accounts.
The Myanmar Police Force was the state institution most immediately present in many towns and urban neighbourhoods when violence occurred. Its first responsibility was ordinary law enforcement: separating hostile groups, protecting civilians and property, arresting individuals suspected of crimes, and preventing retaliatory attacks. Evidence from June, however, shows repeated failures to perform those functions effectively. Human Rights Watch concluded that during the early phase of the violence authorities failed to provide adequate protection to both Rakhine and Muslim civilians. Rohingya witnesses went further, repeatedly alleging that local police displayed hostility toward them and in some places treated Rakhine attackers with greater tolerance than Rohingya residents.
This perceived partiality had particular significance in Arakan because local police were not viewed by Rohingya as socially detached from the communal conflict. One Rohingya interviewed by Human Rights Watch contrasted the local police with Burman soldiers by observing that the police themselves were Arakanese and, in his view, hostile toward Rohingya. Such testimony cannot establish the conduct or attitude of every police officer. It does, however, help explain why Rohingya civilians frequently interpreted police inaction not merely as institutional weakness but as communal partiality.
The allegations became substantially more serious than simple failure to intervene. Human Rights Watch documented cases in which local police were accused of participating directly in violence against Rohingya. In Narzi quarter in Akyab/Sittwe, witnesses described police firing live ammunition at Rohingya while Rakhine mobs burned Muslim homes. One Rohingya witness told investigators that he personally saw police shoot six people, including a woman and two children, before the bodies were taken away. The destruction of Narzi itself was reconstructed in Chapter 5; its significance here is different. The episode provides evidence concerning the conduct attributed specifically to police officers during an attack on a Rohingya neighbourhood.
Similar allegations arose in northern Arakan. Human Rights Watch documented witnesses describing police participating in violent raids on predominantly Rohingya villages around Maungdaw, firing on residents, beating people and looting homes and businesses. Rohingya community reporting produced during the violence also alleged police participation in attacks on Rohingya property. On 11 June, Kaladan Press Network reported from Maungdaw that police and Hluntin personnel were acting alongside Rakhine attackers after army personnel had left parts of the town, including allegations concerning the burning of Rohingya property and religious buildings. These contemporary reports should remain attributed to their Rohingya sources, but their broader pattern is consistent with abuses subsequently documented independently by Human Rights Watch.
The second force requiring separate examination is Lon Htein, often rendered in contemporary Rohingya sources as Hluntin. Lon Htein was a paramilitary/riot-security force and therefore occupied a different role from ordinary local police. Its personnel were deployed for crowd control and internal security, precisely the circumstances created by the June unrest. For Rohingya witnesses, however, Lon Htein became associated not simply with controlling crowds but with shootings and coercive operations against their communities.
The evidence from Narzi is again important. Human Rights Watch concluded that while Rakhine mobs burned Rohingya homes on 12 June, police and paramilitary Lon Htein forces opened fire on Rohingya with live ammunition. The allegation was therefore not merely that riot police failed to stop an attacking crowd. Witness testimony indicated that members of the force themselves used lethal force against people from the community whose neighbourhood was under attack.
Rohingya reporting from Maungdaw made similar accusations. Kaladan Press described Hluntin personnel acting with police and Rakhine groups during attacks on Rohingya property on 11 June. By 16 June, the same Rohingya news organisation was reporting that police, Hluntin and NaSaKa were conducting arrests of Rohingya in Maungdaw and alleged that Rakhine civilians continued to accompany security personnel into Rohingya villages. It also published names of Rohingya whom local sources said had been detained. These reports are valuable because they were produced while the operations were still underway, although individual allegations require attribution because independent observers could not verify every incident in real time.
The use of firearms by police and Lon Htein raises a question distinct from whether some Rohingya had participated in violence. Security forces were entitled to intervene to protect lives and could lawfully arrest individuals reasonably suspected of serious crimes. The critical issue is whether force was directed at an immediate threat and whether it was necessary and proportionate. Testimony describing police or paramilitary personnel firing at civilians attempting to flee, defend their neighbourhoods or extinguish burning homes points instead toward potentially unlawful lethal force. Amnesty International subsequently emphasized that the declaration of emergency did not give security forces authority to violate human rights and documented unnecessary or excessive force resulting in deaths and injuries.
Police conduct after the main days of open violence further deepened Rohingya fears. Human Rights Watch documented large-scale operations in which police participated with other security agencies in rounding up Rohingya men and boys. Witnesses described people being dragged from homes, beaten and taken to unknown locations. The total number detained was not publicly established, and many arrests were carried out without transparent information concerning charges or detention locations. Amnesty International similarly reported that hundreds, predominantly Rohingya men and boys, were detained during large security sweeps and that nearly all were initially held incommunicado, with reports of ill-treatment.
These operations also revealed what Rohingya witnesses regarded as a fundamental disparity in law enforcement. Human Rights Watch reported that while authorities said they were searching for Rohingya suspected of crimes committed during the violence, the mass arrests in northern Arakan appeared discriminatory because comparable efforts to investigate and apprehend Rakhine suspects were not evident there. The issue was therefore not whether Rohingya individuals suspected of killings or arson should have been investigated; they should have been. The concern was whether criminal responsibility was being individualized for both communities or whether the Rohingya population was being subjected to broad punitive sweeps.
From the Rohingya perspective, this distinction was decisive. The police institution that should have distinguished an accused individual from his family, village or ethnic community was increasingly perceived as treating Rohingya identity itself as grounds for suspicion. Contemporary Rohingya reports of police and Hluntin entering villages alongside Rakhine civilians intensified this fear. Human Rights Watch later independently documented several cases in which Rakhine men carrying sticks and swords accompanied security forces during raids on Rohingya villages around Maungdaw.
At the same time, historical accuracy requires avoiding the conclusion that every police officer or every Lon Htein unit participated in anti-Rohingya violence. The surviving evidence does not support such a universal claim. Security personnel were operating amid genuine violence committed by civilians from both communities, and restoring public order was a legitimate state responsibility. The evidentiary problem is instead institutional: repeated and geographically separate testimonies describe police and Lon Htein personnel failing to protect Rohingya, using lethal force against them, participating in abusive arrests, or operating alongside Rakhine civilians. These accounts were sufficiently consistent for independent human-rights investigators to conclude that serious abuses had occurred.
For Rohingya communities, the consequence was a collapse of confidence in the most immediate institutions of public security. A civilian threatened by an armed crowd ordinarily seeks protection from the police. In parts of Arakan during June 2012, Rohingya witnesses described confronting a radically different reality: the people from whom they sought protection could themselves become participants in the danger they were trying to escape.
The role of NaSaKa and the Tatmadaw, however, requires separate treatment. NaSaKa already exercised extraordinary control over Rohingya life in northern Arakan before 2012, while the Tatmadaw entered the crisis with far greater military authority after the state of emergency. Their conduct included some of the most serious allegations of shootings and mass arrests—but, particularly in the case of the army, the record also contains documented instances in which soldiers protected Rohingya civilians.
Section 2. NaSaKa and the Tatmadaw: Sweeps, Arrests, Shootings and Contrasting Army Conduct
The role of NaSaKa, Myanmar's border security force, differed fundamentally from that of ordinary police. By 2012, NaSaKa had exercised extensive authority over Rohingya communities in northern Arakan for many years, particularly in Maungdaw and Buthidaung. Its checkpoints, local commands and administrative powers were already embedded in everyday Rohingya life before the June violence began. Rohingya required official permission for many forms of movement and family life, and NaSaKa personnel were repeatedly accused by human-rights organisations of extortion, arbitrary detention, forced labour and other abuses. When emergency violence erupted in June, therefore, NaSaKa was not entering an unfamiliar population as a neutral outside force. It was operating among communities over which it already exercised extraordinary coercive control.
This pre-existing relationship is important for understanding Rohingya testimony about the security operations that followed. Human Rights Watch documented mass arrests beginning around 12 June in Maungdaw town and Rohingya villages to the south. Witnesses described combined operations involving NaSaKa, police, Lon Htein and army personnel, sometimes moving from village to village and house to house. Some security personnel reportedly carried lists containing names of people suspected of involvement in the earlier disturbances. In other cases, Rohingya witnesses described arrests that appeared much broader, with men and boys taken away without any clear explanation of the evidence against them.
NaSaKa featured repeatedly in these accounts. Human Rights Watch reported testimony describing NaSaKa personnel participating in raids in which Rohingya were arrested, beaten or otherwise abused. Amnesty International likewise documented a wave of arrests after the state of emergency, predominantly involving Rohingya men and boys, and reported allegations of ill-treatment in detention. For Rohingya communities already accustomed to NaSaKa surveillance and coercion, the June sweeps appeared to transform an existing system of control into a mechanism for large-scale punitive operations.
Rohingya organisations documented the same pattern while the operations were taking place. Kaladan Press Network reported arrests across Maungdaw and surrounding villages involving NaSaKa together with police and Hluntin. Its contemporary reports named individual Rohingya said to have been detained and alleged that security personnel entered Rohingya settlements looking for men accused of participation in the earlier violence. Such reports could not always be independently verified at the time because access to northern Arakan was heavily restricted, but their broader description of mass arrests was subsequently corroborated by international human-rights investigations.
The question was not whether authorities had the right to investigate crimes committed during the violence. Killings, arson and attacks against civilians required investigation regardless of the ethnicity of the suspected perpetrator. The concern raised by Human Rights Watch was the apparently discriminatory character of the sweeps. Large numbers of Rohingya were being arrested in northern Arakan, while investigators found little evidence at that stage of a comparable effort to identify and arrest Rakhine suspects implicated in attacks against Rohingya. This disparity strengthened Rohingya perceptions that security operations were treating their community collectively as responsible for the crisis.
Some operations went beyond arrests. Human Rights Watch documented testimony concerning security personnel firing on Rohingya, beating villagers and looting property during raids. Witnesses also described Rakhine civilians accompanying security forces into Rohingya villages, in some cases carrying sticks or swords. The composition of individual security teams varied, and the evidence does not always allow responsibility to be assigned to one particular agency in every incident. This is precisely why NaSaKa, police, Lon Htein and the army should not be collapsed into a single category. Nevertheless, NaSaKa's participation in the broader sweep-and-arrest system is well established in subsequent documentation.
Rohingya testimony also described extortion surrounding detention. Families searching for arrested relatives sometimes reported demands for money in exchange for release or information. In an environment where detainees could be taken away without transparent records, the power held by individual security officers was enormous. Families could not easily determine whether a missing man had been formally arrested, transferred elsewhere, injured or killed. This uncertainty contributed to the large number of people whom Rohingya organisations subsequently classified as missing.
The Tatmadaw, Myanmar's armed forces, occupied a different position. Following President Thein Sein's declaration of a state of emergency on 10 June, the military assumed an expanded role in restoring order. Soldiers were deployed across affected areas and possessed far greater firepower and operational authority than ordinary police. Rohingya testimony concerning the Tatmadaw is therefore particularly important—but also more complex than testimony concerning some other security institutions.
In northern Arakan, army personnel were implicated in abusive security operations. Human Rights Watch documented military participation alongside police, NaSaKa and Lon Htein in raids and arrests affecting Rohingya communities. Rohingya witnesses described soldiers entering villages during searches, detaining men and participating in operations that generated widespread fear. Amnesty International also included the army among security forces accused of unnecessary or excessive use of force, arbitrary arrests and other abuses against Rohingya following the declaration of emergency.
Some Rohingya testimony alleged direct shootings by soldiers. These accounts require careful treatment because different security uniforms were not always easily distinguished by civilians during chaotic operations, and witnesses sometimes used broad terms for armed state personnel. Where an investigation specifically identified the Tatmadaw, that attribution should be preserved; where the responsible force cannot be established, the historical record should not assign responsibility more precisely than the evidence allows. This distinction strengthens rather than weakens documentation of abuses.
At the same time, the evidence concerning the Tatmadaw contains an important contrast. Some army units intervened to protect Rohingya civilians. Human Rights Watch researchers visiting Akyab/Sittwe in late June observed soldiers protecting displaced Muslims and recorded accounts indicating that army intervention had helped bring some of the open violence under control. In certain circumstances, Rohingya residents regarded the army as providing greater protection than local police or Rakhine-dominated local security personnel.
This distinction appears directly in Rohingya testimony. Some Rohingya witnesses told Human Rights Watch that the arrival or presence of Burman soldiers could provide a degree of protection from Rakhine attackers. One Rohingya contrasted soldiers with local police precisely because he perceived the army personnel as outsiders to the local communal conflict. Such testimony does not erase allegations of Tatmadaw abuses elsewhere. Instead, it demonstrates that the behaviour of military units varied according to location, command and circumstances.
The contrast is historically significant. It shows that violence against Rohingya was not simply an automatic consequence of every armed representative of the Myanmar state encountering Rohingya civilians. State forces possessed the practical capacity to protect threatened populations when they chose or were ordered to do so. Where soldiers separated hostile groups, guarded displaced people or suppressed attacks, civilians could survive. Where security personnel stood aside, joined attackers, fired on Rohingya or participated in abusive sweeps, the consequences were radically different.
This evidence also complicates any claim that every Tatmadaw unit operating during June 2012 followed one uniform anti-Rohingya policy at every moment. The documentary record does not support such a conclusion. What it does establish is that the Tatmadaw participated in a security response during which serious abuses against Rohingya occurred, while some army units simultaneously performed legitimate protective functions. Both realities belong in the Rohingya historical record.
For Rohingya survivors, however, occasional protection did not remove the wider fear produced by militarized operations. A civilian could not know in advance whether an approaching unit would provide protection, conduct arrests, search houses or use force. In villages already affected by rumours of raids and disappearances, the sight of armed personnel approaching could cause men to hide in fields and families to flee before determining which force had arrived.
The conduct of NaSaKa and the Tatmadaw therefore reveals two different dimensions of state power during June 2012. NaSaKa represented an established system of coercive control over Rohingya life in northern Arakan that became deeply involved in the post-violence arrests and security operations. The Tatmadaw represented the state's greater military power: implicated in abusive operations in some locations, yet demonstrably capable of protecting Rohingya in others.
That contrast makes the next question unavoidable. Security forces did not operate independently of civilian government structures. Township administrators, village officials, local political networks and Arakan State authorities possessed information, administrative authority and varying degrees of influence over what occurred. Determining how those institutions responded—and whether local officials merely failed to prevent violence or in some cases facilitated it is essential to understanding the wider question of state responsibility.
Section 3. Local Authorities, Collaboration and the Limits of Protection
Security forces did not operate in an administrative vacuum during the June 2012 violence. Township administrators, village and ward officials, Arakan State authorities and other local government structures formed the civilian framework within which police, NaSaKa and other forces operated. For Rohingya communities, these institutions were particularly important because local officials already exercised extensive authority over movement, residence, marriage, travel and other aspects of everyday life. The conduct of local authorities during the violence therefore requires examination separately from the conduct of uniformed security forces.
Rohingya testimony repeatedly alleged that some local officials failed to act impartially. Residents described seeking protection or permission to move while attacks were occurring and receiving little effective assistance. In other accounts, officials were accused of cooperating with Rakhine community leaders or failing to intervene despite possessing advance knowledge that violence was likely. Human Rights Watch subsequently documented broader evidence of state and local authorities failing to prevent attacks despite the presence of substantial security forces. These allegations do not establish that every local administrator participated in abuses, but they raise serious questions about how governmental authority was exercised when Rohingya civilians sought protection.
The problem was particularly acute because Rohingya were already dependent on local administrative permission for activities that other communities could undertake more freely. A Rohingya family could not simply respond to insecurity by moving wherever it considered safest. Longstanding travel restrictions, checkpoints and documentation requirements limited movement, particularly in northern Arakan. During a crisis, an administrative system originally used to control Rohingya movement could consequently become a barrier to escape. A population subjected to violence while simultaneously restricted in its ability to leave occupied a fundamentally vulnerable position.
Local officials also played a role in transmitting information between communities and higher authorities. Township administrations possessed knowledge of local geography, population distribution and communal tensions. Police and security forces depended partly on local structures to identify villages, neighbourhoods and individuals. When post-violence arrest operations began, witnesses described security personnel arriving with lists of Rohingya suspects. The precise origin of every list cannot be established from available evidence, but their existence demonstrates that security operations involved administrative information rather than consisting solely of spontaneous arrests by patrols encountering people in the streets.
Rohingya witnesses regarded this process with particular suspicion because they believed local Rakhine officials and community figures could influence who was identified as a suspect. Whether individual accusations were justified cannot now be established without the underlying arrest records and evidence. What can be documented is the broader pattern identified by Human Rights Watch: large numbers of Rohingya men and boys were subjected to arrests and raids while equivalent enforcement against Rakhine suspects appeared far more limited during the same period in northern Arakan. This disparity contributed to the Rohingya perception that local administration and security structures were operating together against their community rather than conducting an impartial criminal investigation.
Another serious allegation concerned collaboration between Rakhine civilians and state personnel. Human Rights Watch documented several instances in which armed Rakhine men accompanied security forces during raids on Rohingya villages around Maungdaw. Witnesses described civilians carrying sticks, swords or other weapons while moving alongside police or other state personnel. In Akyab/Sittwe, testimony similarly described security forces failing to prevent Rakhine mobs from attacking Muslim areas and, in some instances, firing upon Rohingya while those attacks occurred.
The presence of civilians alongside security forces is significant because the state's legal responsibilities were fundamentally different from those of communal groups. A government officer was obligated to protect civilians irrespective of ethnicity or religion. Allowing armed civilians to accompany an official operation could expose residents to retaliation, looting or collective punishment. Where security personnel themselves facilitated such participation, the distinction between an official law-enforcement operation and communal violence became dangerously blurred.
Contemporary Rohingya sources interpreted these patterns as evidence of coordinated persecution. Rohingya organisations reported that security personnel and Rakhine groups were acting together in particular locations and argued that authorities were permitting attacks while restricting Rohingya movement and self-defence. These allegations were made while events were still unfolding and sometimes contained claims that could not then be independently verified. Nevertheless, later investigations substantiated enough instances of cooperation, discriminatory arrests and security-force violence that the central Rohingya concern cannot be dismissed simply as wartime or communal rumour.
At the same time, evidence of intervention and protection must remain part of the record. State institutions were not uniformly inactive. Security forces sometimes separated opposing groups, imposed curfews, guarded displaced populations and intervened to halt violence. As examined in the preceding section, some Tatmadaw units protected Rohingya civilians in Akyab/Sittwe. There were also circumstances in which state intervention protected Rakhine civilians threatened by Muslim attackers. These cases demonstrate that governmental intervention could save lives when forces acted impartially.
This is important for assessing the failures elsewhere. The question is not whether Myanmar's authorities possessed the physical capacity to intervene. The presence of police, paramilitary forces and eventually substantial military units demonstrates that the state had coercive resources available. The more difficult question is why protection was effective in some places but absent, delayed or discriminatory in others. Human Rights Watch's 2012 investigation concluded that the government could have done considerably more to prevent violence and subsequent abuses.
The evidence also requires care when moving from individual misconduct to institutional responsibility. A police officer participating in an attack, a local administrator failing to respond, a NaSaKa unit conducting abusive arrests and an army unit protecting displaced civilians are not identical acts and should not be collapsed into a single accusation. Establishing whether abuses resulted from local prejudice, informal collaboration, command decisions, official policy or some combination of these factors requires evidence beyond the fact that state personnel were present.
For this reason, the June record should be preserved at several levels. At the individual level, witnesses identified officers and security personnel accused of shootings, beatings, arrests and other abuses. At the unit level, investigations documented police, Lon Htein, NaSaKa and Tatmadaw involvement in particular operations. At the administrative level, questions arise about local authorities' failure to provide equal protection and their role in the machinery of arrests and movement restrictions. Finally, at the state level, the government's response must be assessed against its responsibility to protect all populations under its authority.
From the Rohingya perspective, however, these distinctions did not lessen the immediate experience of vulnerability. Rohingya civilians encountered different branches of the same state at checkpoints, police stations, NaSaKa posts, township offices and during security operations. When several of those institutions simultaneously restricted movement, conducted arrests or failed to protect people from attack, the cumulative experience could reasonably appear not as isolated misconduct but as persecution by the state apparatus itself.
That perception was intensified by the Rohingya's pre-existing legal position. They entered June 2012 already deprived of effective citizenship and subjected to exceptional administrative restrictions. They therefore lacked many of the political and legal mechanisms through which other populations might challenge abuses. Complaining about a police officer, contesting an arrest, recovering property or demanding freedom of movement was particularly difficult for people whom the state itself had progressively excluded from citizenship.
The evidence reviewed in this chapter consequently reveals a security response containing both protection and serious abuse, but not in equal measure or with equal consequences for every population. Police and Lon Htein were implicated in shootings, discriminatory arrests and failures to protect Rohingya. NaSaKa participated in sweeping operations against a population it had already controlled through a restrictive border-security regime. Tatmadaw personnel participated in some abusive operations, while other army units demonstrably protected threatened Rohingya and helped suppress open violence. Local authorities operated within this system and faced serious allegations of partiality, inaction and collaboration.
Recognizing the protective actions of some security personnel does not cancel evidence of abuses by others. On the contrary, those interventions demonstrate that another response was possible. When state personnel acted to separate communities and protect civilians, violence could be restrained. Where authorities failed to intervene impartially—or where state personnel themselves participated in attacks—the consequences for Rohingya communities were severe.
Chapter 6 therefore establishes the institutional record without yet attempting a final judgment about the organization of the 2012 violence. Whether the documented patterns amounted to isolated misconduct, systematic discrimination, coordinated participation or broader state responsibility requires examination of political decisions, advance organization and evidence extending beyond the June security response. That question is reserved for Chapter 13 — Was the 2012 Violence Spontaneous? Evidence of Organization, State Involvement and Responsibility.
Before reaching that assessment, however, the physical and human scale of the June violence must first be established. Government statistics, Rohingya organisations, Rakhine sources, international agencies and human-rights investigators produced sharply different accounts of how many people were killed, how many homes and religious buildings were destroyed and how many people were displaced.
Continue reading → Part 13, Chapter 7 — Deaths, Destruction and Displacement: Establishing the June 2012 Record