Protesters take part in a demonstration against the military coup in Yangon on March 11, 2021.

Neither the military nor the protest movement can be certain about what the ultimate outcome of this present crisis will be. Here, protesters take part in a demonstration against the military coup in Yangon on 11 March, 2021. (Photo by STR / AFP)

Back to the Future? Possible Scenarios for Myanmar

Published

In Myanmar, neither the military nor the protest movement can be certain of the ultimate outcome following the former’s overthrow of the country’s elected government on 1 February. Myanmar appears stuck in a scenario where the military and the protest movement are each attempting to steer the situation towards their optimal outcomes. A tipping point may occur in favour of either side, depending on the resources or support that it obtains, either from domestic or international actors.

The generals who overthrew Myanmar’s elected government on 1 February 2021 envisioned a swift, smooth and bloodless action that would check the power of the National League for Democracy (NLD) and entrench the military’s role in the administration of the country through its model of “disciplined democracy”. They acted on the morning of the day on which the parliament elected last November in polls swept by the NLD under Aung San Suu Kyi’s leadership, was to take office. Now, surprised by overwhelming public resistance to its coup, the new junta has resorted to increasingly repressive measures to bring protests under control. 

After a brief period of restraint that saw the arrest and reshuffling of key decision makers at the Union, state and regional levels, the military prohibited peaceful protests and public gatherings of five or more people, imposed overnight curfews, cut off internet connections between 01.00 and 09.00, and released 23,000 prisoners into the community. It allegedly incentivized some of those prisoners to create disturbances and provoke violence. The junta also reintroduced mandatory reporting of overnight visitors to households, began using lethal force against demonstrations, and, on 8 March 2021, revoked the licenses of five independent media outlets. As of 11 March 2021, around 60 protesters have been killed, hundreds of others injured, and 2,008 arrested. Yet these efforts at suppression have only stiffened the resolve of protesters, who have used creative and diverse strategies to oppose the coup. 

Neither the military nor the protest movement can be certain about what the ultimate outcome of this present crisis will be. Observers have offered three scenarios so far. Anthony Davis argues in the Asia Times that the military has the “experience, skills, and resources” to ultimately succeed in bringing the civil disobedience campaign to heel. Others, including Su Min Naing writing in Frontier Myanmar, believe that the military cannot succeed against the united and widespread opposition to its rule. This view is shared by Tom Andrews, the United Nations Special Rapporteur for Myanmar, who said, “If I were a betting person, I will be betting for the protesters; I think they are going to prevail”. Thant Myint-U, on the other hand, has been more equivocal. He tweeted on 22 February 2021, “I have been a student of Myanmar history and politics my entire adult life; I’ve lived and worked in the country for over a dozen years; I know all the key actors in the present drama; and I can honestly say I don’t know what the coming months will bring.” 

A FRAMEWORK TO ANALYSE POTENTIAL OUTCOMES

Whether the military succeeds or fails in asserting its authority over the country is beyond our powers of prediction at the moment, but we can devise a framework to analyse potential outcomes and what they will look like on the ground. I have therefore mapped out nine different scenarios based on the interaction of strategies employed by both the military and the protest movement.

Chart: Post-Coup Scenarios for Myanmar

Post-Coup Scenarios for Myanmar

Let us begin by looking at the different alliances on either side of the coup and anti-coup divide. The military relies on a narrow base of support, including its business associates and members of the immediate families of its officers, a handful of civilian technocrats, and 23 small political parties. Many of the latter failed to secure representation in last year’s elections, overwhelmingly won by the NLD, or else indicated that they felt alienated by the NLD government of 2015-2020. The anti-coup movement on the other hand has a diverse base ranging from NLD members and supporters to ethnic minority youth, healthcare professionals and teachers, students, intellectuals, civil society organizations, left-wing groups, farmers, workers, and local businesses. 

On the sidelines are ethnic armed organizations (EAOs), which have in the past taken up arms against the government to fight for greater autonomy and have at various points been engaged in intense fighting with the military. There are over 20 ethnic armed groups with a combined estimated strength of 80,000-100,000 troops – significantly fewer than the Myanmar military, which has an estimated 2021 strength of 516,000 soldiers in addition to a police force numbering 80,000 as of 2018. Ethnic armed groups vary in size, legitimacy, and relationship with the military. Generally speaking, members of the Northern Alliance, based along the Chinese border, are less vocal than those who operate on the Thailand-Burma border. Among the latter, an alliance led by some of Myanmar’s oldest armed groups — including the Karen National Union (KNU) — has vocally denounced the coup and cooperated with prominent members of the anti-coup movement. The aim of this cooperation is elimination of Myanmar’s 2008 constitution and the establishment of a federal democracy. The Arakan Army, based in Myanmar’s west near the Bangladesh border, was at war with the military in 2015-2020; it now appears to be war weary, has not condemned the coup, and displays no sign of breaking the ceasefire that it signed with the military at the end of 2020.

The Military’s Choices

My chart of potential outcomes shows a range of choices or strategies on the part of the military, ranging from non-accommodation to partial accommodation and full accommodation. The military is unlikely to make any concessions (‘non-accommodation’ in the chart) as long as it receives cooperation and support from domestic and international forces, particularly China, Russia and ASEAN; if it can exercise control over civil servants; and as long as it is not opposed by the many ethnic armed groups that have so far remained on the sidelines. The military could make some concessions or full concessions if its financial or logistic resources were significantly affected; if the scale and degree of defection from its ranks or among civil servants vastly increased; and/or if there were an internal military putsch, though this in itself would not guarantee a change of strategy, Effective mediation by external actors might lead to one of those same outcomes, with some or full concessions. On 7 March, China publicly expressed its willingness to engage with all involved parties to improve the situation in Myanmar. On 26 February, to great surprise and to the satisfaction of the anti-coup movement, Myanmar’s United Nations Ambassador Kyaw Moe Tun denounced the military takeover and pleaded for the international community to help restore democracy in Myanmar in a speech on the floor of the General Assembly. Whether in emulation of his open defiance of the military or not, further defections by diplomats posted to Myanmar missions in Los Angeles, Washington, Geneva, Berlin, Tokyo, and Jerusalem followed. So far more than 100 police officers have also defected to the protest movement, including a police colonel in Yangon, while a captain in the military became the highest-ranking defector in the armed forces on 4 March.

The Protest Movement’s Choices

The chart above also outlines choices for the protest movement, from full protest to partial protest or outright compliance with the military regime. Full-scale protest occurs when protesters are able to mobilise comprehensive protest strategies that both threaten the foundations of military support and also offer alternative mechanisms to fulfill the basic needs of ordinary people and thus sustain the movement in the long run. These mechanisms include domestic and international support to help finance, plan, and coordinate the protest campaign and put pressure on the military, as well as an internationally recognized parallel government with ministers overseeing assorted responsibilities including self-defense (potentially provided by ethnic armed organizations). The degree and scale of the protest movement can gradually diminish until it reaches the point of compliance as a result of the arrest of key leaders, fatigue, economic insecurity and/or political repression.

Potential Scenarios

Scenario 1A reflects the endgame initially envisioned by the military, with full compliance from the protest movement. Upon seizing power, the military declared that it would reform the country’s election commission during the “emergency” period and host another “free and fair” election. Unlike Myanmar’s previous era of military rule between 1988 and 2010, when the State Law and Order Restoration Council established following a coup was made up only of military commanders, half of the members of the 16-member State Administration Council formed in the wake of the 1 February coup are civilians. The military would like to consider its present role to be similar to that of the “caretaker government” in 1958-60, when a civilian government asked the military to reestablish order and stability for elections. The arrests of and charges against people elected to parliament in November and prominent NLD leaders, as well as the interrogation of the administrator of Aung San Suu Kyi’s charity foundation, and the military’s call to consider reform of the electoral system, are signs of the military’s intention to eliminate the NLD as a political force. This plan has been stalled by nationwide resistance, and is therefore likely to result in the extension of military rule for an indefinite period of time (Scenario 1B in the chart). This outcome would be similar to the period between 1988 and 2004, when the military intensified its repression while exploring an exit strategy by drafting a new constitution. By the end of the first week of March, in fact, state media indicated that the military had extended its timeline for interim rule from one year to 12-24 months.

Riot police face off with protesters during a demonstration against the military coup in Naypyidaw
Riot police hold their firearms as they face off with protesters during a demonstration against the military coup in Naypyidaw on 8 March, 2021. (Photo by STR / AFP)

At present, the situation in Myanmar most closely resembles Scenario 1C, with neither side displaying any willingness to concede, the breakdown of law and order, and the cessation of basic operations of government. The protesters are predominantly members of younger generations, but they also feature a wide variety of people across professional backgrounds and different ethnicities, including those who were unhappy with the policies and practices of the NLD government. They have been able to deploy a diverse range of nonviolent strategies never seen during the opposition to military rule in the 1988 nationwide anti-coup movement. Both widespread internet use and the involvement of the vast Myanmar diaspora have made many of these strategies possible. They range from street protests to banging pots and pans every evening, naming and shaming perpetrators of violence and their families on social media, boycotts of military businesses, refusal by civil servants to show up at work, and protests outside the Chinese embassy. The Committee Representing the Pyidaungsu Hluttaw (CPRH) was formed to represent the ousted civilian government by 15 NLD members elected to parliament in November. The CRPH had expanded to 17 members by 17 February 2021, now including two elected members representing ethnic minority parties. It attempted to establish itself as a parallel governing body, with four acting ministers overseeing various responsibilities and two international representatives.

At present, the situation in Myanmar most closely resembles Scenario 1C, with neither side displaying any willingness to concede, the breakdown of law and order, and the cessation of basic operations of government.

The Civil Disobedience Movement (CDM) was started by medical doctors who built on courage and moral leadership that they had developed during the Covid-19 relief campaign. The CDM has spread to the education, transportation, banking and local government sectors. It has reached an unprecedented scale, with two in every three civil servants either on strike or unable to work because of the collapse of transport and government machinery. As a consequence, many basic operations of government and economic activities such as trade, banking and construction have come to a halt. Signs of economic stress have appeared everywhere, from a gradual rise in food prices to increased unemployment and shortages of cash and essential goods like gasoline. 

If the military decides to make concessions in order to gain public support or due to pressure or international mediation, it may allow the NLD to contest elections and to win a number of seats under a modified proportional representation system that prevents the party from capturing a majority of elected seats (Scenario 2A). This scenario would represent a slight improvement on Scenario 1A, which would see the NLD abolished or forced to re-establish itself under a different name. Partial protest could result in the military extending its rule indefinitely, but with some degree of political and economic relaxation (Scenario 2B). Scenario 2B would be similar to the situation between 2004 and 2010, when the military relaxed restrictions on foreign and domestic private investors and civil society organizations that refrained from political mobilization against the military. It is also a slight improvement on Scenario 1B, in which extended rule would be based on full-scale political repression. If resistance continued at its present level, however, one could see the emergence of localised self-governing mechanisms of the sort that have already appeared in some areas to fill a vacuum of political authority (Scenario 2C). In the Thai-Myanmar border town of Myawaddy, and in Kachin State in the country’s north, and in Kayah State, armed groups have protected and guarded protesters. In the Chin State town of Mindat, several villages jointly issued a statement announcing that they would administer their territory according to Chin customs and practices, while some armed groups, including several KNU brigades, declared that they would ally with neither the CRPH nor the military. Most areas in the country are currently being administered by local communities composed of religious leaders and respected elders and guarded by volunteer night watch groups. Peripheral areas home to minority ethnic groups were also already being governed by ethnic armed groups before the coup. Scenario 2c is a slight improvement over Scenario 1C, which is characterized by complete chaos. 

In the event that the military makes a full accommodation, there are three potential outcomes. Scenario 3A is the pre-coup reality, in which the military would recognise the November 2020 election results but retain its privileges under the 2008 constitution — such as controlling a quarter of reserved seats in parliament and thus retaining veto power in the legislature, along with control of the defense, border affairs and interior ministries. This is a scenario initially envisioned by the CRPH/NLD. More public protest could also result in the resignation of top military leaders responsible for the coup (Scenario 3B). Full military concession (Scenario 3C) would completely revolutionise Myanmar’s political landscape by abolishing the 2008 constitution, potentially transforming the country from a quasi-democracy to full democracy with the military controlled by civilian politicians, and from a unitary system to a genuine federal democracy. These are the objectives of protesters indifferent or hostile to the NLD, such as younger protesters and members of ethnic minority groups, including ethnic armed groups, who see the anti-coup movement as a once-in-a-lifetime opportunity to abolish the 2008 charter and achieve a genuine federal democracy and the protection of minority rights. 

In the short term, the more the anti-coup movement proves able to adopt comprehensive and diverse strategies, the more intense and even desperate the repression imposed by Myanmar’s military will become.

The nine scenarios developed here are based on a simplified model of interaction between the military and protesters. They could overlap or prove volatile over a very short period of time. For instance, three different scenarios could emerge in separate parts of the country at the same time. For example, an anarchic situation (1C and 2C) could prevail during military rule (1A and 1B). The chart nonetheless offers a framework to analyse potential historically grounded scenarios. It also provides opportunities for the main players to explore the most desirable outcomes, which would benefit the majority of people whose lives have been ravaged or destroyed by the coup.

The chart is also useful as a tool to examine the objectives of different players and their attempt to influence the nature and direction of this crisis. The best case for the military is Scenario 1A or 1B, with one-year or indefinite military rule, while the best case for protesters is any scenario that falls along the lines of Scenarios 3A, 3B, or 3C. Currently, Myanmar seems to be stuck in the chaotic Scenario 1C, while the military and protest movement are both attempting to steer the situation towards their optimal outcomes. 

CONCLUSION

In the short term, the more the anti-coup movement proves able to adopt comprehensive and diverse strategies, the more intense and even desperate the repression imposed by Myanmar’s military will become. The military has, for instance, increasingly relied on the use of brute force and the extrajudicial killing of unarmed civilians, along with the torture of detainees. These tactics have replaced its reported original plan to use a “war of attrition” to wear down and conquer the public. In the meantime, key figures in anti-coup movement have been able to expand the CDM and mobilise supporters toward pushing for a situation resembling Scenario 3C. The CRPH, for instance, has added the elimination of the 2008 constitution, along with the promulgation of a new constitution based on principles of federal democracy, as one of its objectives. This situation could be brief or last for a long time, and it could manifest differently in different geographical areas. For instance, border areas governed by ethnic armed groups are more likely to do better, with their extant self-governing structures and access to neighbouring countries, than core urban areas susceptible to the military’s strict control. 

A tipping point may occur in favour of either side, depending on whatever additional resources or support they can obtain from such domestic actors as ethnic armed groups and from international actors and defectors from the other side. Many groups and organisations will bandwagon with the stronger party. International mediation led by the UN or regional actors such as ASEAN, China, or Japan is a likely possibility if accepted by both the military and Aung San Suu Kyi. Mediation is, however, unlikely to result in a situation similar to the pre-coup political order, as that order will be unacceptable to both the military and the segment of the protest movement that wants a complete transformation in Myanmar politics in the form of genuine federal democracy and total civilian rule.



This is an adapted version of ISEAS Perspective 2021/30 published on 12 March 2021. The paper and its footnotes can be accessed at this link.