LOADING
Myanmar
Conflict Map
Myanmar conflict update

Threat of communal violence grows in western and central Myanmar

Anti-junta forces continue to make gains on the battlefield. But tense social dynamics and the consequences of protracted violence threaten to derail their momentum.

By Morgan Michaels
Graphics by Brody Smith
Published May 2024

Six months since the launch of Operation 1027, Myanmar’s regime continues to suffer punishing setbacks across the battlefield. Anti-junta groups are now looking to expand their territorial control and consolidate their own administrations. Yet despite their progress and ambition they face a number of daunting obstacles, including rising communal tensions, territorial disputes, and a stubborn enemy that remains well resourced and determined to maximise the costs of war.

Faced with mounting losses in Rakhine, the regime has resorted to arming members of the Rohingya ethnic minority to counter the Arakan Army’s (AA) advance. The AA has reacted with inflammatory rhetoric and violence directed at the Rohingya. A dangerous dynamic is emerging, reminiscent of the communal tension in Rakhine State between 2012 and 2018 which led to episodes of violence perpetrated by communities along ethnic lines. In the centre of the country, the regime has made a turn back towards indiscriminate violence and is upping support for local militias, a move that could supercharge inter-village conflict. But some attacks by local People’s Defence Forces (PDFs) may be inadvertently driving the growth of these regime-aligned outfits, referred to as Pyu Saw Htee.

Captured by AA as of March 2024
Building damage in Buthidaung Town (20.875056° N, 92.524194° E) following purported arson attacks in mid-April. Source: Maxar.
Building damage in a Rohingya village (20.86074° N, 92.55313° E) following purported arson attacks in late April and early May. Copyright Planet Labs PBC.
Violent events in the Dry Zone
01 May 2023 – 30 April 2024
0
50
100
150
200
May ’23
May ’24
01 May 23
01 Jun 23
Attack/armed clash
Remote explosives/IEDs
Air/drone strike
Infrastructure destruction
012345 , 01234567890 01234567890 01234567890
Number of Rohingya estimated to be fighting for the regime in Rakhine.
Tweet
Tweet
Keep scrolling ↓
Scroll to begin ↓
Following the regime’s expulsion from nine townships between November and March, much of the fighting shifted towards Buthidaung and Maungdaw. These two townships were the primary site of the military’s ‘clearance operations’ in 2017, which sent some 740,000 Rohingya fleeing across the border to Bangladesh.
To stem its losses, the regime has used both coercion and inducements to recruit Rohingya who remain in Myanmar. It is also cooperating with Rohingya armed groups, which are recruiting from the refugee camps in Bangladesh. As many as 5,000 Rohingya could now be fighting for the regime in Rakhine.
The introduction of Rohingya fighters has severely agitated the AA. The group has issued statements decrying Muslim ‘terrorists’ and ‘extremists’, despite little evidence that religious ideology is driving Rohingya behaviour. The group’s leader has described the Rohingya as ‘Bengali’, a pejorative term used to justify previous rounds of ethnic cleansing.
The AA’s frustration has already boiled over into violence against civilians. On 17 April, the AA reportedly abducted and executed five Rohingya villagers from Tha Yet Oke village in Maungdaw.
According to the AA, regime soldiers and Rohingya fighters razed hundreds of Rakhine homes in Buthidaung Town between 11 and 16 April. But according to Rohingya accounts, the AA torched nearby Rohingya villages between 27 April and 2 May. The AA also allegedly razed hundreds more homes after capturing downtown Buthidaung on 18 May. Satellite data and imagery confirmed fires and structural damage in these areas.
Heatmap shows fire hotspots recorded by NASA's Fire Information for Resource Management System, near Buthidaung Town from 11 April to 21 May 2024.
Allegations also emerged of a large-scale massacre of Rohingya civilians taking place on 2 May near one of two villages in Buthidaung: either Tha Yet Kin Ma Nu or Htan Shauk Khan. IISS researchers could not verify reports of the incident.
The overall incidence of armed violence in the Dry Zone slowed following the onset of Operation 1027, which drew the regime’s attention and resources towards the borderlands. But the nature and impact of violence in the Dry Zone remain severe.
The first two weeks of May were particularly bloody. On 9 May, a coalition of PDFs raided Sone, a village occupied by pro-junta Pyu Saw Htee forces. The raid killed at least 31 civilians, including women and children.
On 11 May, regime soldiers raided Let Htoke Taw, a village they knew maintained a PDF presence. The soldiers rounded up the villagers for interrogation before executing 30 men and one woman, all of whom were reportedly civilians.
Regime airstrikes also killed scores of civilians. On 9 May, aircraft targeted a monastery where resistance officials and residents had gathered in Akyi Pan Pa Lun village, killing 15 and wounding 34 more. Regime attacks continue to account for most of the combat-related civilian deaths in the Dry Zone.

Early warning signs

The regime’s decision to arm the Rohingya — a population that it attempted to violently expel just seven years ago — reflects a remarkable transformation of the conflict landscape in western Myanmar. That the regime has resorted to this behaviour underscores how badly weakened it has become in the last six months. But it is also a reminder that, even when losing, the junta retains a variety of options for countering its opponents and raising the costs of war. The regime’s exploitation of the Rohingya has not only slowed the AA’s battlefield momentum but has also forced the group to fight members of one of the world’s most persecuted minorities. Though it is expected that the AA would defend itself, the group has directed much of its aggravation towards the Rohingya through acts that resemble collective punishment. In doing so, the AA has undermined its own public-relations messaging regarding communal harmony and likely its ability to govern the Rohingya in the future.

Nonetheless, the AA leadership appears angered by the actions of some Rohingya. While the regime has clearly engaged in forced recruitment, there is evidence to suggest that some Rohingya have decided they are better off joining the fight against the AA. The reasons for this are likely varied and complex, and should be considered within the wider context of the Rohingya’s plight. For example, the regime has reportedly offered some Rohingya citizenship, a right that successive regimes have long denied them. And given the desperate situation in the camps in Bangladesh, others may simply be willing to take their chances back in Myanmar. But one motivator that cannot be discounted is that the AA’s own actions are pushing the Rohingya into the arms of the regime.

Other important variables include the roles of Bangladesh and India. Since 2019, Bangladesh has appeared relatively lenient towards the AA, turning a blind eye to the group’s activities on its soil. But this accommodation could be ending. Bangladeshi security forces may be indirectly facilitating the recruitment of Rohingya in the camps and their movement across the border, which would help the Myanmar regime. Possible explanations for this behaviour may be that some Bangladeshi officials seek to enhance their leverage over the conflict in Rakhine, or that others see an opportunity to repatriate the Rohingya by other means.

Unlike their counterparts in Bangladesh, Indian security forces had never tolerated the AA, while New Delhi has struggled to come to terms with the regime’s rapid deterioration. For one, the war between the AA and the regime has derailed the Kaladan Multi-Modal Transit Transport Project, which aims (by sea and by land) to connect Kolkata to northeast India via Rakhine. The conflict also has implications for the security situation in northeast India, where a number of groups maintain either rivalries or partnerships with Myanmar armed outfits. Some weapons captured from the Myanmar regime have shown up in India’s Manipur state, where ethnic violence flared last year. Pressure from New Delhi may also partially explain Bangladesh’s indirect support for the regime.

Society and militarisation in the Dry Zone

Throughout the first two years of the post-coup war, the regime dedicated most of its assets to countering threats from ethnic armed organisations (EAOs) in the borderlands. Rather than directly confront the new PDF movement in the Dry Zone, it adopted a containment strategy based primarily on acts of collective punishment against the PDFs’ civilian support base. Beginning in 2023, however, the regime shifted its approach by committing more resources to directly confront the PDFs. A campaign of mass arson and indiscriminate attacks was largely, though not entirely, replaced with more systematic search-and-destroy operations against PDF cells.

The recent pattern of violence in the Dry Zone, which includes a resurgence in arson attacks and mass killings, suggests that the beleaguered regime may be returning to a strategy of containment predicated on widespread punitive attacks against civilians. Stretched thin in the aftermath of Operation 1027, the regime could potentially be expected to redirect its efforts towards consolidating control over the country’s centre. But to survive, the military needs to first halt EAO expansion. It is therefore more likely to prioritise threats posed by EAOs in Kachin, Rakhine and Shan states, which are at present greater than those presented by the fragmented array of comparatively ill-equipped PDFs in the Dry Zone. Taking the fight back to the borderlands would also impede any EAO’s ability to project force into the centre by supporting PDFs.

Over the last two years, conflict in the Dry Zone has steadily evolved beyond a mere contest between the PDFs and the regime, and is increasingly marked by intense rivalry between villages with either a PDF or Pyu Saw Htee presence. One reason the regime can afford to deprioritise the Dry Zone is that it can tap into the hundreds of militia units there that are opposed to the PDFs. Similar to its arming of the Rohingya, the regime has reluctantly started to provide more and better arms to the Pyu Saw Htee, despite the possibility that doing so could allow these units to gain autonomy. This behaviour is likely to raise the cost of armed clashes and propel localised animosities across the Dry Zone.

But the recent attack on Sone village shows how PDFs also drive village rivalries and push segments of the population onto the side of the regime. Speaking to local media shortly after the raid, one resistance fighter appeared to justify the deaths of the women and children by insisting that everyone in the village was a member of the Pyu Saw Htee. This rationale closely mirrors the regime’s reasoning for targeting innocent civilians that live in villages where PDFs operate. But just as many civilians are not associated with PDFs, many living in villages with a Pyu Saw Htee presence support neither the regime nor its operations. When facing attacks from combatants, a security dilemma arises: civilians must either accept security from that particular combatant outfit and the label that comes with it, or flee their homes. The regime’s support for Pyu Saw Htee outfits in otherwise neutral villages, and the PDFs’ tendency to attack these villages, have helped drive the rise of pro-regime militias across the Dry Zone.

More than three years into the war, many PDFs in the Dry Zone continue to operate with no oversight from or accountability to the National Unity Government (NUG). In the absence of a clear command structure, autonomous anti-junta outfits have committed a mounting list of abuses. On 14 May, for example, a video emerged that appeared to show a PDF leader torturing and beheading four men he accused of stealing. Frustration with PDFs among their own supporters is so widespread that the NUG has formed a task force whose mandate includes countering PDF crimes. But growing concerns over weak command and control are not unique to PDFs: fear that it will be unable to control the Pyu Saw Htee is why the regime did not equip them sooner.

Should it continue along the current trajectory, the contest between already independent PDFs and increasingly autonomous Pyu Saw Htee outfits could become detached from the broader war between the NUG-led movement and the regime. The community-based, inter-village nature of these now bitter rivalries means that it will be difficult to halt the violence even if the regime is overthrown or a national political settlement is reached.

View more updates