In Myanmar/Burma a core preoccupation is the ongoing peace process that will end 65 years of armed conflict in the resource-rich borderlands populated by ethnic minorities. This takes place alongside a transition from totalitarian military rule towards democracy and a rapid influx of international aid agencies and investors. In this article we engage with the wider debate about ‘Demobilization, Disarmament and Reintegration’ (DDR) programmes and argue that in the present Myanmar peacebuilding context it is highly unlikely that conventional DDR programmes will suffice to support stability and sustainable peace (Jensen & Stepputat 2014; Munive & Jakobsen 2012; Munive 2013; Muggah 2005; McMullin 2013b). This is because of the predominant focus in DDR programmes on disarmament, as a first step in the process, and on economic incentives to successful integration. In Myanmar this DDR template overlooks key political motives behind both the causes of conflict and the demands of the peace negotiations. For six decades 20 ethnic armed organisations have fought for self-determination and have, to varying degrees, enjoyed considerable state-like control over ethnic territories and peoples. For this reason ethnic Armed Non-State Actors (ANSAs) are strongly against laying down arms before a comprehensive political settlement is reached. A core demand of ANSA leaders is a federal system that not only gives them political positions but also allows them to retain arms in the different ethnic nationalities areas.

Overall, the Myanmar situation raises the question of whether conventional DDR in some contexts should be substituted by what is now referred to as ‘third generation DDR’ or ‘RDD’ (reintegration, demobilization and disarmament) (Sedra 2003). Reversing the conventional sequence, RDD begins with incentives – economic and political – for reintegration and only ends with some form of disarmament or arms control (Munive & Jakobsen 2012: 362). It is increasingly realized that ‘sequencing flexibility’ may be needed to adapt DDR to particular contexts. A UNDPO report (2010: 28) highlights that, ‘[p]rioritizing reintegration before starting disarmament and demobilization may be advantageous in cases where political will is lacking for disarmament.’ Reintegration opportunities, including non-material incentives such as political recognition, can serve to move a stagnant peace building process along and may also provide incentives to financially motivated combatants of lower-ranks (ibid). There are also matters of security and trust at play, as Walter (1999: 154–5) argues: ‘because combatants are likely to become highly fearful and insecure as they demobilize, they can gain an added sense of safety if they are not forced to disarm fully, especially not before the political terms of an agreement have been fulfilled.’

In this article we similarly argue that the lack of political will to disarm in Myanmar calls for an exploration of potential (re)integration options for ANSAs as a starting point for discussing DDR. This, we suggest, will not only help obtain sustainable peace but also increase trust in what currently is a contested peace process. Given that a political agreement is still in the process of being reached, it is equally necessary to ground a discussion of reintegration within a wider framework of conflict resolution (Walter 1999; Zartman 2001; Ramsbotham et al 2012). This means framing integration options as an element of disaggregated power-sharing guarantees (economic, political and military), which can help create trust in the peace agreement (Walter 1999).

On 15 October 2015 eight ANSAs signed a Nationwide Ceasefire Agreement (NCA) with the government and the head of the national Burmese army after two years of intensive negotiations. However, the remaining seven ANSAs who were invited to sign have not done so, and another three ANSAs have been excluded from the NCA by the government. Fighting between government and ANSA forces also continues in some ethnic areas and even the groups that have signed are internally split on the NCA. The peace process therefore remains contested. At a higher political level this reflects insecurity regarding whether the government and the army will commit to a federal system. At the lower level armed actors are insecure about their future options, including their sources of income, recognition and security, which creates mistrust in the peace process. The peace negotiations have so far left out any open discussions of what will happen to the many thousand middle- and lower-ranked soldiers and officials once an agreement is reached. Instead the current draft agreement focuses on high-level political and military aspects, including a promise to hold a political dialogue regarding changing the political system. While these high-level settlements will clarify the overall framework for the future possibilities of ex-combatant integration, more concrete options and guarantees also need to be discussed at this current stage of conflict resolution. As Walter (1999) argues, the incentives to accept and implement a peace agreement depend not only on resolving the root causes of conflict and agreeing on the overall political system, but also on more disaggregated security and power-sharing guarantees. These can help mitigate the enormous uncertainties that rival parties face in the implementation of peace treaties: e.g. fear of attacks, of losing income and positions and of being excluded from future government arrangements. Power-sharing guarantees can include the distribution of specific posts to ex-combatants in government, administration, service delivery and security forces, including at the local level. It may also imply ‘interim arrangements’ whereby ANSAs are allowed to maintain administrative and security roles in the territories they have controlled until integrated institutions are established. Hartzell & Hoddie (2003) further add the importance of economic power-sharing guarantees, especially in contexts where certain identity groups have a history of being economically marginalised. Apart from providing an incentive to sign an agreement, these guaranties are also important for sustainable peace. This argument emanating from the debate on conflict resolution is closely related to the integration aspects of DDR programming.

In light of these insights, this article discusses integration options for the ANSAs: what ‘exit’ options do ANSA members have after decades of conflict and how do they envision their future – as armed actors, civil servants, politicians, businessmen or something else? In addressing this question we draw on semi-structured individual and group interviews held in Mon and Karen states as well as in Yangon in January 2014 and on prior research.1 We discuss five different integration options. These consist of a combination of different forms of political, economic, civil society and security sector integration. The options are of our own creation but are inspired by the DDR literature and our interviews in Myanmar. They are in no way exhaustive but should be read as an initial contribution to the debate about reintegration in Myanmar. In fact, we make a call for a more in-depth analysis of the armed groups, their incentives and motivations, not as a homogenous group but as a complex and dynamic set of actors. It is important to note that the integration options will overall depend on what kind of political settlement is reached, including important power-sharing arrangements and the possibilities for building trust and ensuring security in the implementation process. It is therefore necessary to firstly provide a short background to the conflicts and to discuss the main elements and challenges of the nationwide ceasefire negotiations in Myanmar. We conclude the paper by reflecting on the role of international aid agencies in the peace process, with Myanmar representing a rather exceptional case of very low international involvement.

From Armed Conflict to Contested Peace

Myanmar has been blighted by civil war, ethno-nationalist conflict and outbreaks of communal and religious violence since colonial times. Thirty-three per cent of its 51.4 million residents belongs to ethnic minorities and the remainder to the Burman majority. There are today an estimated 20 active ethnic ANSAs, including numerous splinter groups, which represent the different ethnic minority groups (e.g. the Shan, Karen, Kachin, Mon, Chin, Kayah, Rakhine, Wa, Pao-o, Naga, Lahu, Lisu and Palaung, each with various sub-groupings) (Smith 1999).2 These groups have fought the national army to obtain a federal constitution since 1949, two years after independence from the British. Most of them also have political wings, their own flags and uniforms. Over different periods they have administered their own ‘micro-states’. Some, like the Karen National Union (KNU), have their own departments of education, health, justice, forestry and local defence. Today the ANSAs can muster an estimated 100,000 soldiers. The size is difficult to access, but it ranges from the large United Wa State Army (UWSA), with an estimated 20–25,000 troops, to the Karen National Union (KNU), with approximately 4–6000 troops, and the New Mon State Party (NMSP), with 500–700 soldiers (Gravers & Ytzen 2014).

The political economy of the ANSAs has been tied to a shadow economy, linked to cross-border trade, especially with China and Thailand, and income from the borderlands’ natural resources such as minerals, gems, timber and opium (Woods 2011). ANSAs are also known for taxing ethnic populations and larger ANSAs have received substantial donations from the ethnic diaspora as well as from donors and religious organisations. Over the course of the armed conflict, access to and control over trade and resources have played a strong role, along with identity politics. Shifting military operations and trade alliances with China and Thailand have also influenced the strength of the ANSAs. For instance, in the 1970s–80s the KNU was regarded as the de facto authority by Thailand at the local level in the border region, but in the 1990s the Thai commander-in-chief struck a deal with the Burmese military government to gain access to agricultural and mineral businesses in Karen state and dams and ports in Mon state. This substantially undermined the ANSAs (Oh 2013).

The armed conflict in Myanmar has resulted in establishing the borderlands as a segmented society where military organisation and a shadow economy amalgamate in an ethno-nationalist semi-state polity. Weapons have remained crucial to protect civilian supporters as well as businesses against Burmese army attacks and economic incursions. In addition competition between and within ethnic organisations has been strong.

History of conflict and previous ceasefires

The seeds for the ethnic-based armed conflict were already sown during British Colonial rule. Ethnicity was reified and politicised and a new order of classification and administration divided subjects according to not only ethnicity, but also culture and religion (Furnivall 1956: 304–307). Apart from ethnicity, religious diversity has been central in the conflict, which had begun during the British conquests in the 1800s when ethnic minority Christian converts helped the British fight Burman rebels led by Buddhist monks.3

During World War II Christian Karen and Kachin states cooperated with the British forces against the Japanese army, alliances that resurrected tensions between the ethnic minorities and the Burman majority. During the negotiations leading to independence ethnic minorities expected their loyalty to the British to be rewarded with autonomy. A conference in Panglong was organised in 1947 with the main ethnic group leaders, and here a federation was discussed that would grant autonomous administration to the ethnic minorities (Gravers & Ytzen 2014). However, the federal principles of the 1947 constitution never materialised. This laid the roots for the long civil war. The Kayah rebelled in 1949, followed by KNU, which almost managed to take over the then-capital city, Rangoon (now Yangon).4 The KNU retreated to the hills of present Karen state (established in 1952) where it established the de facto government of the Kawthoolei (Old Country) state. The Mon took up arms with the KNU in 1950 and, after a ceasefire agreement, rebelled again in 1958 with the formation of the NMSP (South 2003). The Kachin and other groups followed in 1961.

In 1962 General Ne Win staged a coup, overthrowing Prime Minister U Nu, who had promised states to the Rakhine and Mon groups. Fearing that other ethnic groups would secede, Ne Win took power and launched a military offensive, demanding unconditional surrender from the ANSAs. His idea of order was a corporate state of one nationality and he created a one party socialist union. Major military offensives during his rule weakened many of the ANSAs’ territorial control. After pro-democracy protests in 1988, the State Law and Order Restoration Council took over power from Ne Win and initiated the first ceasefires with at least 17 ANSAs (between 1989 and 1997).5 These ceasefires focused on economic and military matters, excluding any political settlements towards federalism. For this reason some ANSAs, like the KNU, did not agree to a ceasefire. Those who agreed were allowed to keep their weapons and were given lucrative local trade deals, territorial control over specified ethnic areas and even industrial and international trade concessions, in exchange for giving up the armed struggle (Oh 2013: 10).

The 1990s ceasefires split up many of the ANSAs due to internal disagreements, fatigue from fighting and conflicting economic interests. For instance in 1994 many Buddhist soldiers in the mainly Christian-led KNU/KNLA mutinied to form the Democratic Karen Buddhist Army (DKBA), under the leadership of a Buddhist monk.6 The DKBA signed a ceasefire with the government and cooperated with the army to take over the KNU headquarters. In exchange, DKBA was given logistical, military and financial assistance as well as permission to conduct businesses (South 2011: 19). After the DKBA exit, more splinter groups from the KNU appeared, led by officers who were fed up with the struggle and looking after their own business interests and the interests of their supporters. This reflected the general emphasis on economic incentives in the 1990s ceasefires. While they did bring some development projects that improved the lives of villagers, they also ended up strengthening the illicit businesses of ceasefire groups and the national army. Rather than create sustainable peace, they allowed for the expansion of the army’s territorial control and counter-insurgency strategies (Oh 2013: 11).

In 2009 the ceasefire groups from the 1990s became subject to the Border Guard Force (BGF) initiative, which followed the 2008 constitution’s demand for a single army. It was an open strategy for military integration of the ANSAs as special units under the command of the National Defence Services (Keenan 2013). The deal involved stable salaries, social benefits and continued armament for the ethnic soldiers. Whereas this arrangement resembles reintegration elements from DDR programs elsewhere, it involved neither disarmament nor demobilisation, and came with no political settlement. Consequently, many ANSA leaders refused the deal, resulting in renewed cycles of fighting and tensions.7 Like the 1990s ceasefires, the BGF initiative has been criticised not only for undermining ethnic political demands, but also for exacerbating abuses of villagers, illicit business and land-grabbing by BGF forces (Keenan 2013: 3–4). Until 2011, joining the BGF was made a precondition for any peace talks with the government.

The BGF initiative was implemented after a longer political process beginning in 2003 with the military government declaring a seven step roadmap to ‘disciplined democracy’, which in 2008 led to a referendum for the new constitution, followed by general elections in 2010. Both events were allegedly marred with fraud. The military proxy Union Solidarity and Development Party (USDP) was declared the overall winner of the elections, which in 2011 brought President U Thein Sein into power. He introduced a surprising political and economic reform agenda based on fundamental rights of citizens. He also made a peace process with the ANSAs a top priority. By early 2012 ceasefires had been signed with the majority of the ANSAs, followed by negotiations towards a nationwide agreement.

Nationwide peace negotiations – obstacles and challenges

In December 2013 a National Ceasefire Coordination Team (NCCT), comprised of 16 ANSA members, began the process of drafting a National Ceasefire Agreement (NCA) together with the Union Peace-Making Work Committee (UPWC), representing the government and army. The main political demands of the ANSA leaders are a federal constitution, the protection of the rights of ethnic nationalities and a federal army into which the ethnic ANSAs are integrated. Legal reform that regulates land ownership, development projects and control of the drug trade is also on their agenda. This last demand is closely related to the fear of unequal economic power and should be understood against the background of the large-scale development projects initiated by the government during the ceasefires, which ethnic communities have associated with land-grabbing and counter-insurgency. Importantly, the ANSAs are against disarmament before a political settlement and most envision this settlement to include some form of continued armament. A major change for the government is that since 2013 it no longer demanded that ANSAs first surrender weapons before entering into political dialogue; it also agreed to discuss federalism.

In August 2015 the NCCT agreed on a final NCA text with the government, which on 15 October 2015 was signed by eight ANSAs, including, among others, the KNU and two other Karen armed groups. However, the remaining ANSA members of the NCCT – including the influential KIA in Kachin state and the NMSP, representing the Mon ethnic group – decided not to sign. The large UWSP, representing the Wa group, which is not member of the NCCT, also rejected an invitation by the government to sign. The official argument is that they do not want to sign unless the NCA is all-inclusive and therefore genuinely national. At the heart of this matter is the government’s refusal to include three ANSAs that are in open combat with the army in the Kokang area (the Ta’ang National Liberation Front [Palaung], the Arakan Army and the Kokang Army [Myanmar National Democratic Alliance Army]) and three others regarded as too small and insignificant. However, apart from dissatisfaction with lack of full inclusiveness, the unwillingness of some ANSAs to sign the NCA indicates continued insecurities and mistrust about the future.

The NCA that was signed on 15 October guarantees an end to all hostilities, provisions for a military code of conduct and the promise of establishing a union based on the principles of democracy and federalism, which can be seen as a major step towards power-sharing and decentralisation. However, the guarantee of a change of the political system will depend on the outcomes of an inclusive political dialogue also involving civil society and political parties, as well as on constitutional amendments. The political dialogue will only begin in early 2016. As such the NCA does not include any concrete and disaggregated power-sharing guarantees, such as positions to ANSA members in government, the administration and security forces, which, as argued by Walter (1999) can be a very significant incentive to sign a peace agreement. The text does mention that political dialogue should include discussing a union army that will represent all ethnic nationalities, but it makes no concrete guarantees for the continued armament of ANSAs. In terms of economic power-sharing such as the equal distribution of resources and land rights (Hartzell & Hoddie 2003), the NCA also does not provide any concrete guarantees. It only provides guidelines for ensuring that ANSAs and local communities are consulted before the government rolls out major projects in the ceasefire areas. Although this could help counter the widespread fear that the government and the army will use the NCA to penetrate ethnic areas to make economic and political gains, such guidelines provide no legal guarantees in themselves. This is likely an area of concern, especially for larger ANSAs like the KIO, which controls areas with many natural resources and lucrative trade with China.

Although the NCA text illustrates that the parties have come a long way in agreeing on basic principles and in increasing trust between ANSA leaders and USDP government representatives, there are still many uncertainties when it comes to implementation and power-sharing guarantees. A main concern is also mistrust in the national army to commit to political changes and to end attacks, even after the signing of the NCA.

Deep issues are at stake that make the conflicts in Myanmar extremely difficult to end, despite progress in the NCA process. If we look at the three dimensions – contradictions, behaviour and attitudes – outlined by Galtung and others in the peacebuilding and conflict resolution literature (Ramsbotham et al 2012), we can identify some of the most important stumbling blocks.

The ethnic contradictions are still emphasized by all parties. The army and the USDP government adhere to a singular Myanmar national identity while the ANSAs maintain that they are independent ethnic nationalities. There is also fundamental political disagreement about a federal constitution. The current constitution provides the military with 25 per cent of the seats in the two houses of parliament and grants the president power to appoint ministers from the ethnic states. Proposed amendments to these two parts of the constitution were rejected by the parliament in June 2015 due to opposition from the military. This sent a strong signal that the military is not committed to federalism. The army also insists upon upholding the Unlawful Association Act, which makes those ANSAs that have not yet signed the NCA illegal organisations and makes any contact with them illegal. Moreover, there are contradictory economic interests, as mentioned above, as it is widely believed that the army wants to gain control of natural resources in ethnic areas. The army now insists on a DDR plan with a focus on ANSA disarmament as part of implementing the NCA, whereas the ANSAs want to keep weapons. All these are fundamental structural contradictions impeding an all-inclusive settlement. Thus integrative measures and structural transformations during the peace process remain difficult (Ramsbotham et al 2012: 175–176).

Attitudes have changed very little. Nationalism and ethno-nationalism are still dominant ideologies.8 Related to these ideologies and the long history of violence is a general and profound mistrust among all parties. The presence of mistrust and fear means that the non-compromise factions in the ANSAs and the army remain influential. There is an internal elite struggle within most ethnic nationalities and their organisations (Naing 2015; Gravers 2015a), including within those ANSAs that have signed the NCA. In fact not everyone in the KNU supports the 15 October signing. Those against the NCA particularly worry about the future status and power of the ethnic groups. The army works as a corporate unit but also seems divided between ‘hardliners’ and ‘liberal’ officers. Factionalism is therefore a major problem (Ramsbotham et al 2012: 174). However, the idea of marginalising sceptics and spoilers will only lead to further conflicts (ibid: 186; Gravers 2015a). Importantly, the ANSA leaders as well as their middle- and lower ranks worry about their future positions, which is further complicated by the fact that there are no concrete guarantees for inclusive integration in the current NCA.

Fighting during the NCA negotiations has demonstrated that behaviour has changed very little. More than 40 clashes between ANSAs and the army occurred from January to August 2015 in Kachin and Shan states, and further fighting intensified in the last few days before the NCA signing. The armed actors on the ground stick to their old ways, which are the values of armed struggle, despite trust building at the leadership level. One positive step is the creation of liaison offices, which establishes direct contact between combatants. Nevertheless, substantial transformation of behaviours will take time and depend on political results, which remain unclear.

Future peacebuilding through the implementation of the NCA is, as elsewhere, even more complicated (see Ramsbotham 2012 et al) but if it is successful from the perspective of ethnic groups in the areas covered by those ANSAs that have signed already, other groups may follow in signing. During the NCA negotiations none of the parties involved wanted to include international and third party mediation, and it is still unclear if the international community will be invited to play a role in monitoring the NCA implementation. As argued by a number of scholars, third-party commitments to assist ceasefire implementations can be important to create trust in peace agreements (Walter 1999; Zartman 2001). They can help reduce fear among combatant groups that the most powerful party to a conflict will fully take hold of government power as agreements are being implemented. In Myanmar, international aid agencies have only so far been officially invited to support with development projects and reconciliation in NCA areas. As a former British colony, Myanmar guards its independence and sovereignty and the military has always been suspicious of human rights, considered Western ideas. Although the EU was invited to sign the NCA as a witness, along with the UN and neighbouring countries, the government did not agree to the ANSAs’ suggestion to also include specific European countries, like Norway and the UK. The government only agreed to include these countries as observers. As witnesses and observers, the international community does not as such stand as a guarantor of the NCA process.

A final and important area of concern is the current political changes in Myanmar, influenced especially by the 8 November 2015 elections. Rather than waiting for all ANSAs to be included, the president and the ruling party, USDP, have undeniably pushed for the NCA before the elections so as to secure more votes. Conversely, those ANSAs that did sign were likely fearful that the negotiation process would drag on too long if they waited until after the elections. It is unclear how the National League for Democracy (NLD), led by Aung San Suu Kyi, which has won an approximately 80 per cent majority in both Houses of Parliament, will stand in relation to the ANSAs once it takes over the government in March 2016. The party has declared its support for federalism, but how much power it will concede or decentralize to the ethnic states remains unclear. In addition, the military still retains veto power over constitutional changes, which would be necessary to get a federal system, and it also continues to have 25 per cent of the seats in both houses of parliament and the power over three important ministries (defense, home affairs, and border affairs). Aung San Suu Kyi did not herself participate in the NCA signing in October 2015, which may indicate reluctance to support the current agreement. In the minds of those ANSAs that have already signed, the negotiation process had come to be perceived as what is often defined as a ‘hurting stalemate’ in conflict resolution literature, which indicates ripeness to sign an agreement (Ramsbotham et al 2012). Others may hope that they can get a better deal with the NLD, but this remains to be seen.

Another current political challenge for the ANSAs is the growing number of new ethnic political parties, who also claim to represent ethnic minorities (Hiebert & Nguyen 2014). Although these lost most seats in the November 2015 elections to the NLD they still stand as alternative ethnic political forces that may question the role of the armed groups as legitimate stakeholders in the political dialogue that follows the NCA (South 2014). The ANSAs may become more marginalized in the political process, and there is a risk that this can lead to renewed cycles of armed conflict, especially if ANSA members are not able to strike a deal that also benefits them, politically and in terms of economic survival.

Against this background, the different ANSAs’ incentive to sign and not to sign the NCA stands between what could be an urgent need to strike a deal before they potentially lose political clout and the fear that the current NCA will not assure the ANSAs significant positions, due to a lack of any concrete power-sharing guarantees. In line with Walker (1999), we argue that the lack of such guaranties creates insecurity about what the NCA will mean in practice, including for the middle- and lower-ranks. This calls for the need to discuss concrete (re)integration options, even if at this point conventional DDR is not a realistic first step. We therefore approach the debate about reintegration options in the next section as not only important to sustainable peace, but also an urgent matter to increase trust in the conflict resolution process.

Future Options for the Armed Actors

Interviews in Mon and Karen states confirmed the strong conviction that the ANSAs are not ready to disarm before any comprehensive political settlement. This was also the view of CSOs such as Mon Women’s Organisation: ‘the people do not want the armed groups to disarm, because they need them to protect their identity and freedom,’ adding that this is also a military power issue because as long as ‘there are only Burmans in the top army positions, the ANSAs do not accept the army proposals’ (group interview, 16 January 2014). Keeping arms was not only seen as reflecting insecurities about the military position of ANSAs, but also as necessary to protect ethnic civilians due to mistrust in the army. However there was also concern that the ANSAs are losing their popularity among civilians as armed actors. This legitimacy threat puts pressure on the ANSAs to explore alternatives to reinvent themselves as serving roles other than just armed protectors. Other interviewees also argued that the ANSAs will not be satisfied with only economic incentives to disarm: ‘To have peace the government has tried to give the armed groups opportunities like land, cars and business, but the groups still do not trust them. The leaders need to be given high positions. They hold onto arms still because they want a federal state’ (pastor, Karen state, 17 January 2014). This reflects, as discussed earlier, ANSA members’ political ambitions and the need for power-sharing guarantees as incentives to engage in and commit to the peace process.

In this section we consider five integration options for the ANSAs. Integration here refers to the process through which fighters change their identity from ‘combatant’ to ‘civilian’, and change their behaviour by ending the use of violent means and increasing activities that are sanctioned by the mainstream community (Torjesen 2013).9 Already, there are some examples of ANSA members in Myanmar who have voluntarily disarmed or self-integrated, for instance as members of political parties or civil society organisations (CSOs). While we draw on these examples, the options given below are of our own creation. As argued in the introduction we do not see these integration options as only following on from a process of disarmament and demobilisation, but also as an input to discuss more concrete and disaggregated power-sharing options, including military, political and economic, as part of the political settlement and implementation of the NCA. This goes beyond national level agreements on political institutions, like federalism or proportional representation, to also include local level positions. As Walter (1999: 142) argues, ‘the more political, military and territorial power can be disaggregated, the more enforceable and credible promises to share power will be’. For instance, as we discuss below, federalism and democratic institutions are not in themselves a guarantee that warring factions will obtain positions (ibid: 140). In line with Hartzell & Hoddie (2003) we further add the economic dimension to this equation, also considering livelihood survival and distribution of resources. The five options are: 1) integration into the security sector, including community policing; 2) political parties; 3) civil service and local government positions; 4) economic integration through job creation and skills training as well as the formalisation of large-scale businesses run by ex-combatants and; 5) CSOs. We consider the obstacles to and the dilemmas of these options.

Security sector integration: military, police and village defence

In contrast to the dominant view of DDR programmes that disarmament is a precondition for political stability, experiences from elsewhere show that military integration can work to create stability and pave the way for integration (Spear 1999; Mutengesa 2013; Berdal & Zaum 2013). For fighters who had known little else other than rebel life, Mutengesa (2013) argues, military integration can be a way to ‘decompress’ and make the transition to civilian life. It can also build confidence and give ex-combatants a much needed sense of employment security (ibid: 343). Conversely, hasty disarmament can mean ‘reintegration into poverty’ or, at worst, engagement in illicit activities or re-mobilisation into militia units because employment opportunities are scarce (ibid: 342).

According to the current NCA a good number of the ANSA members will likely be integrated into some form of restructured Union Army, ensuring ethnic nationalities representation. It seems that stability will depend on the development of relatively independent ethnic armed factions where ANSA commanders are given equivalent ranks and/or where the army is based on ethnic state divisions. This would be an important power-sharing guarantee, reducing the fear that the Burman army would take over full military power (Walter 1999: 141). Trust-building will likely also depend on giving ANSA commanders the guarantee to control their wartime armed units in the ethnic territories. In Tajikistan, for instance, such an arrangement was combined with positions in government for ANSA political leaders, resulting in considerable stability and trust in the peacebuilding process (Torjesen & Macfarlane 2007). However, due to a lack of political regulation of illicit economies, such stability came at the cost of sustainable economic development because commanders used their positions for personal enrichment (ibid: 327). There are thus important political-economic issues to consider with such forms of military integration.

In general we suggest that military integration should be combined with wider Security Sector Reform (SSR), including the police and the judiciary, as these institutions are also extremely important for de facto power-sharing. This can involve the integration of ANSAs into national and regional police branches as well as into more local level village defence or community policing (Knight 2009). Shared control of the judiciary and consideration of local level justice and dispute resolution mechanisms in ANSA areas are also important steps in reconciliation. Although SSR has yet to be discussed in Myanmar as part of the post-NCA political dialogue, the ANSAs are proposing federal police forces for the ethnic nationality states as well as recognition of ethnic justice systems.

As Knight (2009) notes, integration into the police is less straightforward than military integration, as it requires radically different skills and education than does the military. There is thus a need for comprehensive training and careful recruitment among ex-combatants if human rights abuses by police and/or their political instrumentalisation by former leaders are to be avoided. Moreover police integration – and SSR more broadly – need to take into consideration that even if the national police and courts are not present, there is seldom a complete security vacuum in conflict or ceasefire areas; armed and non-armed local security forces, with varying levels of legitimacy and effectiveness, usually exist. This is the case in Myanmar, although knowledge is needed on how these forces operate, are structured and relate to or overlap with the ANSAs (UNDP 2012; McConnachie 2014). Such knowledge could inform potential efforts to align local-level security provision with the integration of ex-combatants into more formalised village defence or community policing schemes that work with civilians and create partnerships with the police over time, as is the case, for instance, in Liberia (Hill & Bowman 2006). This also gives ex-combatants an occupation and a sense of worth in the community. However, experiences from, for example, Afghanistan warn against allowing such groups to remain armed as this can run the risk of them becoming independent militias who are not adequately accountable to their communities (Kumar & Behlendorf 2010: 13).10

Political parties

Transformation of ANSAs into their own political parties as well as political integration of ex-combatants into existing political parties predominantly targets ANSA leaders. However, it can also give lower- and middle-ranks a conduit for political expression so as to realise personal, social and economic goals through non-violent means (Mitton 2008: 202). Most of our interviewees supported political integration as an option in Myanmar but there were also concerns. One Buddhist monk stated: ‘The leaders [of ANSAs] are not ready to be politicians in a democracy. They are not educated and civilised, but speak in a too rough manner like military way. They do not understand that democracy is to be representative of the people’ (Interview, January 12, 2014).

In general it cannot be assumed that ANSAs already have the required political and technical skills to operate party apparatuses and engage in parliamentary politics. Thus in other post-war contexts political integration has commonly been supported by international aid agencies who provide capacity building (ibid: 198). Conversely, Nilsen & Tønnesen (2013) argue that the problem of adequate skills also concerns already existing political parties in Myanmar, and therefore a transformation of ANSAs into parties should be seen as part of a wider democratisation process. Likely successful political integration will depend on a demilitarisation of the political culture of ANSAs and beyond. Otherwise political integration could risk reproducing patronage politics and the mobilisation of military networks within electoral politics. Yet these issues cannot be generalised across all the ANSAs in Myanmar; the larger ANSAs, like the KNU, already have entrenched political structures and some internal democratic procedures in place, whereas smaller splinter groups do not.

A core challenge to political integration in Myanmar is the great complexity and heterogeneity of already existing parties that represent the same ethnic minorities. It is not clear to what extent current parties represent the ANSAs, or if individual ANSA members support them or would be willing to lay down arms to join them. Potentially, political integration could involve motivating combatants and commanders to join these existing parties, especially those that represent their political goals (like self-determination for ethnic nationalities). Some of the existing party representatives suggested that alliances could be built between ANSA parties and existing ones. A minister for the Karen People’s Party (KPP) asserted: ‘the KNU leaders can become party officials in the KPP or they could make their own party […] and then we can make an alliance. This would mean a strong constituency because KNU has support in the villages and KPP is strong in towns’ (interview, 15 January 2014). A key challenge now is that ethnic political parties can only get real national influence and adequate representation in the current political system if they create broad alliances (Nilsen & Tønnesen 2013). This is not only due to the ethnic groups being a minority, but also because the current single member constituency voting system favours larger parties. Entering elections is therefore no guarantee of de facto power positions for the ANSAs.

A related concern is the political legitimacy of the ANSAs in the ethnic constituencies. According to South (2012) many Karen communities in KNU-controlled areas display strong support for the KNU, yet there is concern that this is not the case in other Karen constituencies. Some ANSA leaders fear losing popular support and control over client populations during the current peace process, especially as civilians resettle in government-controlled areas. Transformation into political parties as part of a peace settlement and disarmament process will arguably only be attractive to the ANSAs if they believe they are able to mobilise enough votes. Even if federalism and democratic decentralisation were agreed on in the political dialogue, these institutional arrangements would not be a de facto power-sharing guarantee to the ANSAs (Walter 1999). In other contexts, such insecurity for ANSAs has led to a combination of political integration in the form of electoral competition with the guarantee of specific positions to ANSA members within the government and state apparatus (Torjesen 2013; Mitton 2008). The question is whether the incoming NLD government and other ethnic representatives would agree to such privileged positions for the ANSAs in Myanmar.

Political integration also needs to consider the potential power games for positions among top and mid-level ANSA members, which may also affect lower ranks (Torjesen 2013: 6). If lower ranks do not feel that they benefit and are represented through the political integration there can be a risk of violent remobilisation (Christensen & Utas 2008). According to Spear (2007) one of the problems in other post-war contexts is that many ANSA members do not regard being in the political opposition as providing for them economically, at least not sufficiently. This calls for a consideration of the heterogeneity of incentives to give up fighting and thus for different integration options.

Civil service and local government positions

Another possible option for ANSA members is positions within local government service provision and administration, based on already existing experiences and structures. As studies have shown the ANSAs, along with a range of ANSA-linked Community-Based Organisations (CBOs), have to varying degrees had quite extensive administrations as well as social service delivery in the areas of health, education, agriculture, land tenure and so forth (Jolliffe 2014). Instead of viewing these as oppositional to state-building, they should be seen as an asset in consolidating and improving service delivery during the peace agreement implementation. According to Jolliffe (2014: 10), cooperation between ANSA-linked service providers and the government can also contribute to reconciliation in the long term. This is already seen with the NMSP education sector, which has been successful in introducing Mon language into government schools through corporation with the Ministry of Education (ibid). Such cooperation can be contrasted with the fact that the rolling out of government schools and clinics, staffed only by Burmans, in ethnic areas during the ceasefires has created antagonisms and fears that the government is taking over control before a political settlement is reached. It also sends a signal that ethnic representatives are excluded from government, underpinning mistrust in the peace process and in governance reform in general.

Apart from job creation for ex-combatants, local government integration could already be part of a disaggregated power-sharing guarantee in the implementation of the NCA, allowing the ANSAs to continue to administer the ethnic areas they control. This would mitigate fears that the government and army will take full control during the NCA implementation. Ceasefire negotiations have already discussed ‘interim arrangements’ that could include ANSA structures at the local government level until official government institutions and services are rolled out, but the details regarding how this will play out are not yet clear and there are no direct guarantees in the NCA. Naturally, the longer-term institutional developments within this field will also depend on the extent to which federalism is accepted.

Local government integration needs to carefully consider already existing power arrangements at the local level, so as not to lay the ground for future tensions. Not only do local governance setups vary across the ethnic minority states due to the shifting contours of the conflict, but there are also areas with mixed local government, for instance areas where government-appointed village leaders and ANSA leadership structures co-exist, sometimes alongside village leaders accountable to other armed factions, like smaller splinter groups (Interview, KPF leader, January 2014). Against this background the post-NCA political dialogue should early on include a dialogue about what will happen with existing personnel within the various local government setups and the power positions they hold. This will be very important to the successful implementation of the NCA. South (2012) questions the governance capacities and technical expertise of existing personnel from the ANSAs, but adds that this also applies to government officials. As experiences from elsewhere (e.g. Aceh, Indonesia) show, it is important that not only ANSA leaders, but also rank-and-file combatants and civilians from other ethnic groups, are included in local government initiatives, so as not to produce new forms of dominance (Ansori 2012).

Economic integration: job creation, education and training

In DDR programmes economic integration is understood as a process whereby combatants are moved from livelihood support mechanisms associated with military networks to employment in formal and informal sectors (UN 2008). This covers a range of mechanisms like vocational and agricultural training, job placement, education for ex-combatants, income generation with microcredit schemes, and public works schemes (McMullin 2013a). Apart from providing an income that moves them away from combat or criminality, a job can also give ex-combatants a sense of pride in supporting their families and thus aid their psychological and social reintegration (Specht 2003). In previous years DDR programmes have also begun to involve whole communities in joint community development and reconstruction work, where civilians and ex-combatants participate and get on-the-job training (such as the rebuilding of schools, clinics, roads and wells) (Munive & Jakobsen 2012: 362; UNDPO 2010). This also potentially lessens distrust and increases tolerance between different conflict-affected groups, thereby also supporting reconciliation (Specht 2003: 96).

Myanmar is already experiencing large investments and new businesses (the country is opening up to foreign investors and the economy is being liberalised). This could also benefit the economic integration of ex-combatants. However, as experiences from elsewhere show, this will likely be more realisable if ex-combatants are given heightened skills and education as part of, for instance, internationally-funded skills training. Specht (2003) also suggests that there may be a need for the government to lobby potential employers and give them concrete incentives (like tax reductions) to recruit ex-combatants, as it cannot be assumed that private businesses will necessarily be willing to hire them.

Economic integration should not only be seen as a technical exercise of post-war employment creation, however, but also as an integrated element of the conflict resolution process. This implies framing integration within wider economic power-sharing guarantees, including share of resources and access to business concessions, as part of the peace agreement (Hartzell & Hoddie 2003). It also implies establishing incentives to enter an agreement that will outweigh the benefits of war economies for combatants (Zartman 2001). According to interviewees, a real worry in Karen and Mon states is that the new businesses, which are predominantly owned by the Burman majority or by foreigners, will not hire local Karen and Mon, but import Burman labourers. This tendency will not only make it difficult for ex-combatants and returning IDPs to get jobs, but also challenge the consolidation of economic power by the ethnic minorities in their own areas. This is also why the ANSAs demanded in the NCA negotiations a guarantee that all larger development and business projects planned for the ethnic areas only be approved on the basis of consultations with the ANSAs and local communities. However, to what extent this will be cherished in the implementation of the NCA in Karen state remains to be seen and may set an example for the Mon armed group, the NMSP, which has still not signed.

ANSAs also fear losing economic power themselves as a consequence of the implementation of the NCA. This concern is likely more acute among those groups that control large natural resources and trade, like the KIO, and is deeply embedded in what Woods (2011) has defined as ceasefire economies. These include the many economic activities that ANSAs have developed in the territories they were granted control over through bilateral ceasefires, ranging from cross border trade, mineral extraction and plantations to illegal drugs and gambling. They also include the Burmese military’s economic incursions into ANSA territories and the borders around them. Economic reintegration should consider such economies, which are deeply embedded in networks of power and contestations over control of territories and resources. Key combatants have often enjoyed the benefits of the war or ceasefire economy and this position can be hard to break. Some of our interviewees stated that a key challenge is that the armed conflict has created a kind of ‘lost generation’ of people who have known little but military conduct and who see few opportunities to join the licit economy. Specht (2003) speaks about creating a ‘political economy of peace’, which involves closing off illicit routes to economic gain. Another option is to formalise ex-combatants’ existing agricultural or mineral businesses and other forms of trade and grant them land and business concessions. This option already has historical roots in Myanmar. For instance the Karen Peace Force General used his ceasefire deal with the government in 1995 to commence large-scale agricultural and infrastructural projects, which also benefitted the population in his area. These experiences point towards a potential entry point to economic (re)integration via economic power-sharing between government and ANSAs. However, it is important to mitigate the risk that such a deal involves the personal enrichment of ex-combatants at the expense of other members of the local populations. This calls for a more concrete dialogue in the near future on ways to regulate and formalise the informal economic activities of ANSAs so that they serve the economic rights of ethnic communities at large.

Civil Society Organisations

Civil society organisations (CSOs) enjoy an expanding space for operation in Myanmar and with the increase in international donor flows inside the country there is a growing demand for local NGOs as partners in development. In one respect such new CSOs could be seen as challenging the ANSAs’ local legitimacy because many of them claim to represent those civilians who did not participate in the armed struggle. Conversely, many of those operating in the ethnic states still need the protection of the ANSAs and therefore have deep alliances. The possibility of CSOs becoming spaces for ex-combatant integration into civilian life is not something one reads about in the DDR literature, but in Myanmar could be relevant. This became clear in Mon state when we met the Ramanya Peace Foundation (RPF), established after the 2012 NMSP ceasefire. Two of its founders were former NMSP members and essentially had ‘self-integrated’ by setting up the RPF, which now receives considerable international donor funding to support the peace process by doing projects in the areas of water and sanitation, women’s empowerment and leadership training in NMSP ceasefire areas. Their work is ground-breaking because, as one of the first CSOs, it was registered with the government and allowed to carry out development projects in NMSP areas. The two founders already had some skills to enable them to re-invent themselves as a development CSO because they had been part of the NMSP’s education department. Similar examples exist in other ethnic states.


In this article we have argued that there is a need in Myanmar to begin a discussion on future (re)integration options for ethnic ANSAs, rather than highlight disarmament and demobilisation as the first steps in a DDR process. This alternative sequencing of DDR is already being discussed in the international DDR debate, where it has been argued that in contexts where there is a lack of political will to disarm it may be more sustainable to begin with (material and non-material) reintegration initiatives (UNDPO 2010; Thruelsen 2006: 36). This RDD approach could be a valuable option for Myanmar. However, we have also argued that there is a need to frame reintegration options within the wider debate on conflict resolution and peace negotiations (Walter 1999; Zartman 2001), rather than confine it to the technical aspects of DDR or RDD programming. This means framing integration options as part of more disaggregated power-sharing (political, military and economic) guarantees already during peace negotiations. Such guarantees can help create incentives to reach agreements as well as build trust in the implementation of agreements.

As discussed in this article, the ANSAs in Myanmar will not lay down arms before the political system changes towards a federal one, and even talks about disarmament at the moment can be detrimental to the peace process. Conversely, economic incentives to reintegration that focus on employment opportunities will likely only satisfy some ANSA members unless combined with political positions and guarantees that ensure the distribution of resources to ethnic groups. The strong emphasis on power-sharing guarantees should be seen in light of the exceptionally long history of ethnic-based armed conflicts in Myanmar’s borderlands under repressive military regimes which have, despite the growth of war economies, always been embedded in strong ethno-political agendas. Moreover there is currently a fear among the ANSAs, including members of those that have signed the NCA, that a peace agreement could be used by the government and military to take full control of the ethnic areas, through development projects and the expansion of state institutions that do not include ANSA members and the ethnic populations, but the Burman majority. This is a fear also reflected in other peace negotiation processes, as shown by Walter (1999), who argues that such fear can be a strong disincentive to commit to peace agreements. One solution to this impasse is to include in agreements more disaggregated power-sharing guarantees, such as positions in local government and security forces. Hartzell & Hoddie (2003) further add economic power-sharing guarantees, such as the distribution of resource control and access to state funds. Although eight ANSAs have now signed the nationwide agreement, there is still much concern among the remainder of the ANSA leaders, as well as among middle- and lower-ranks about their future options and positions. This also regards members of those groups that have already signed the NCA, such as the KNU, which is internally split on the agreement. As reflected in a media statement by a general of the KNU’s armed wing, many combatants feel that there is a need for a concrete ‘political roadmap’ in the NCA, including specific guarantees, rather than the promise alone of a political dialogue (Karen News August 26, 2015).

In this article we have discussed five different integration options that combine concerns for military/security, political and economic power-sharing, and which also consider middle- and lower-ranks. Thus we have also focused on local-level positions, such as in service delivery and sub-national administration, along with political parties and civil society organisations. It is clear that there are no quick solutions or blueprints. The modalities need to be based on particular contextual understandings and a consideration that armed actors are not homogenous groups. So far in Myanmar the incentives of lower- and middle-ranked armed actors to transform themselves have been silenced in the peace talks. Experiences from elsewhere show that such kind of exclusion can run the risk of creating autonomous spheres of violence and predation (Derksen 2014; Spear 2007). In Myanmar it has also meant continued mistrust in the peace process by over half of the ANSAs. Mid-level commanders are particularly important to consider here because they often enjoy considerable local power and access to informal businesses (Spear 2007: 181; Derksen 2014). In Myanmar such commanders have for years run de facto local micro-states. These positions they hold raise questions not only about future economic reintegration options, but also about politics and power-sharing. As Derksen (2014: 2) argues, it is important that a political settlement also involves ‘translating national power-sharing into local arrangements that give the main local actors access to power and resources.’ In this light, we suggest that the post-NCA high-level political dialogue about federalism should immediately include considerations of concrete integration options at the local level. This is much more important than rushing into DDR programming with its ultimate focus on disarmament and demobilisation. Equally important at the moment is a military code of conduct and inclusive monitoring mechanisms that will ensure effective implementation of a nationwide ceasefire and reduce the fear that fighting will continue. Right now a key concern is exactly that open combat between ANSAs and the army has not ended in all areas. Before this happens political talks may be futile. Civil society involvement is crucial in monitoring and in future reconciliation measures, including in dealing with traumas and repatriation of IDPs and refugees.

The remaining question is what role the international community, including development agencies, can play in the future peace process and the implementation of the NCA. According to conflict resolution scholars like Walter (1999) and Zartman (2001), third party mediators and external commitment to assist peace agreement implementation – e.g. through peace keepers – are key ingredients to successful peace processes. Mediators can provide credibility to ceasefire incentives (Zartman 2001: 300) and third party actors can help enforce commitments to demobilisation and power-sharing arrangements, thereby increasing trust and reducing fears that either of the parties will cheat (Walter 1999: 137). In Myanmar international agencies have not been invited to become third party mediators and at the signing of the NCA foreign representatives, including the EU, the UN, Japan and neighbouring countries, only acted as witnesses and observers. It is still unclear to what extent international actors will take part in any implementation and monitoring of the NCA, but a peace keeping mission is highly unthinkable. This meagre international involvement comes despite the massive influx of development agencies since the country opened up in 2011. The advantage is that the peace process can be seen as more home-grown and locally or nationally owned rather than internationally driven. However, there is also cause for concern that current donor modalities can undermine, rather than support, the peace process. This is because currently the vast majority of donors are principally involved in supporting the government-led reform process, along with providing humanitarian aid. They are already engaging in state-building, before and on the side-line, of the peace process.

In Mon and Karen states there was a strong view that until a political settlement is reached between the government and ANSAs, international donors should avoid supporting government development initiatives (including schools, clinics, etc.) in ethnic areas. Such initiatives have until now been seen as boosting the legitimacy and control of the USDP government as well as undermining the ethnic political agenda. Although it is likely that an NLD government will be more trusted by the ethnic minorities, it is still important that donors are considerate of being inclusive of ethnic nationalities’ concerns when they are operating through government agreements. There must at least be a strong awareness among internationals about their potential damaging effects on creating trust in the peace process.

Having said this, there are clear openings for support, especially after the signing of the NCA, have also in October 2015 launched a ‘Joint Peace Fund’ earmarked for development projects and reconciliation in ceasefire areas. Moreover, ethnic CSOs are welcoming donor funds for development assistance to areas still marked by conflict. International agencies could also support ex-combatant reintegration and provide assistance to institution and capacity building that supports agreed-upon power-sharing arrangements (Walter 1999). This also means being sensitive to ethnic inclusion more broadly in government institutions, including administration, police and so forth. However, as suggested in the critical DDR literature, such support should not take the form of export models but be based on careful contextual analysis that is sensitive to the power dynamics and heterogeneity of the ANSAs (Munive & Jakobsen 2012; Spear 2007; Torjesen 2013; Muggah 2005; McMullin 2013b).

Competing Interests

The authors declare that they have no competing interests.