Violence between Buddhists and Muslims has been a worrying aspect of Myanmar’s politics over the last five years, and has once again taken centre stage with the October 2016 attacks in Rakhine State and the subsequent brutal crackdown by security forces. Yet, despite its apparent importance in the narrative of Myanmar’s broader political transition, large scale acts of violence, such as riots, have only taken place in a comparatively small number of places.

The Myanmar Media and Society Project (M.MAS, a collaboration between the Programme on Modern Burmese Studies at St Antony’s College, University of Oxford and MIDO, the Myanmar ICT for Development Organization) was initially set up to do research into the narratives that justified violence and discrimination against Muslims. While conducting that research, we regularly heard accounts of “almost” riots, situations in which tensions between religious communities had come to a head and residents were worried that violence would result. However, local individuals and groups were seemingly able to de-escalate the situation and prevent large-scale violence.

Investigating these cases was challenging since some had pending legal components, and people weren’t always interested in discussing them, or were concerned that too much attention could regenerate conflict and escalate into violence. Yet even within such constraints, there was much to be learned. On this we agree with others who have studied collective violence, and who argue that we can only understand things like riots if we take time to look at places where they seem to fail. “Until we study ethnic peace,” argues the political scientist Ashutosh Varshney, “we will not be able to have a good theory of ethnic conflict.”

We have published four case studies and our analysis and conclusions in the project’s second Working Paper, titled “Failed riots: Successful conflict prevention in four Myanmar cities.” Even though we can only provide basic details of the cases (and sometimes even these are contested or unclear), we believe that there are important insights to be drawn from these “failed riots.” As is often true in conflict prevention, there is no universal recipe for the de-escalation of conflict and stopping violence, but studying these negative cases (where violence did not occur, although it might have been expected) is critically important in expanding our understanding of the dynamics of peace and conflict. This blog post briefly summarizes some of the main conclusions, but we invite people to read the full Working Paper.

Each of these cases entailed a contest, between and among both individuals and groups operating with their own ideas about what was good and right. Some were engaged in activities that, while we can’t say whether they intended to cause collective violence, certainly risked such escalation. Insofar as such violence was a potential outcome, this was prevented by the active efforts of others, including civil society and religious leaders. Although those who intervened to stave off violence included a diversity of people from different faith groups, they generally were not associated with “inter-faith” groups. Interestingly, those that played key roles in calming, mediating, or bringing together opposed groups did not necessarily identify as “peace-builders” either. Instead, the common trait of the individuals involved was that they had built up trust and credibility in communities and with local authorities through previous activities that enabled them to counsel against violence and mediate disputes. They were land activists, advocates for disadvantaged populations, or community elders who could act as trusted interlocutors in situations where communities either were not directly communicating with one another, or when they would not have been inclined to believe statements from the other community. This highlights the critical importance of sustaining funding and support for a wide range of community-oriented and civil society activities, in addition to more focused efforts on inter-faith and peace-building activities.

The four cases also highlight the way in which framing conflicts in religious terms can serve to expand and escalate them. This re-framing both reinforces and is reinforced by already-existing stereotypes of religious others. M.MAS Working Paper 1:1 explored some of the narratives circulating about Muslims in non-Muslim communities. For example, that they are potential threats and that, as a group, they are intolerant, violent, and rapacious (or at least, potentially so). With this image of Muslims in circulation, religious framing of conflicts between individual followers of Buddhism and Islam can help to draw in many more people not involved in the original disputes and conflicts.

At the same time, religious framings should not be understood as always and only instrumental; some of these conflicts were over specific tenets or practices of different religious traditions, not as a matter of framing but as a matter of content, such as disputes over Muslims slaughtering cattle. In these cases, ignoring the way disputes are taken as religious would be to ignore the broader implications of state-sanctioned intolerance and discrimination against minority religious groups.

The efforts of those trying to de-escalate employed various arguments and ways of framing or re-framing the situation, that undermined or directly opposed both the individual conflict that appeared primed to escalate and, in some cases, the larger negative narrative regarding Muslims. Some highlighted the many devastating impacts of a potential riot, drawing on examples from other riots to argue that everyone would suffer—but poorer communities would suffer disproportionately—if a dispute turned into wider violence, regardless of religious identity. Others emphasized alternative pathways for resolving the disputes at hand, particularly—and interestingly—intervention via state authorities. Many people drew on past experiences of solidarity, inter-dependence, or peaceful coexistence with neighbors of other religions. These included times of crisis when members of different religious communities had supported one another as well as local, town and region-based, identities that could help people to rise above momentary tensions or misunderstandings. Finally, an effective argument seemed to be that the tense situation and apparent potential for conflict was the effect of political machinations, orchestrated by people in power. While it might be difficult to prove these claims, it is important to note the effect they seemed to have in convincing people that they should work against the escalation of the conflict.

Even though these cases were “successful” in that wider violence was prevented, our research also raised some concerns. First, the unstated premise present in some people’s explanation of the events was that the central precondition for peaceful coexistence was the appropriate (read: quiescent, deferential and submissive) actions of the minority religious group. Second, in assessing local accounts of riotous violence that attribute instigation and participation to “outsiders” or political machinations, we must bear in mind that this can obscure deeper problems, especially the need to honestly engage with discriminatory or exclusionary attitudes towards religious minorities, as well as to assign culpability for particular actions in moments of violence.

Finally, preventing riotous violence is not the same thing as promoting peace. We must recognize that the impact of systemic discrimination and disenfranchisement can be just as devastating as a riot, and the solutions to these require broader structural changes in politics and society. While the protagonists in our cases were successful in their efforts to prevent large-scale violence, the reasons they gave for intervening and the arguments that proved effective each come with their own risks. These arguments would not be as effective in situations where the escalation of violence poses no threat to the protagonists’ own community. In both these cases and our other research, for example, we have seen many people in Myanmar describe their work to prevent violence as motivated from the belief that this was necessary to support the National League for Democracy (NLD) in a political struggle against the military, political opponents, and others. People in Myanmar may be unwilling to go to such lengths if violence seems to align with the desires of the NLD, or if the NLD and its erstwhile opponents are seen as aligned on specific questions such as “counter terrorism.”

Even as we consider some of the limitations of the actions taken in these “success” stories, we see where some actions and dynamics could not only de-escalate violence but also serve as the basis for work to promote a more comprehensive and meaningful state of peace. These arguments about shared local identity, shared forms of vulnerability born from poverty, or common experiences of hardship and cooperation to surmount difficulties or respond to local crises, are born from lived experiences in Myanmar. They are in circulation and can help to promote peace, in the form of maintaining and encouraging forms of coexistence and solidarity that already exist. We should do our best to support them.