Introduction
Since Myanmar began a process of political liberalization in 2010, refugees outside the country have found themselves under increasing pressure to return. This has been the case across diverse contexts of asylum, including refugees living in camps in Thailand, Rohingya in Bangladesh, and ethnic Chin in India and Malaysia. Return of refugees from these countries has been supported (and in some cases instigated) by international donors and agencies under the auspices of peacebuilding, even though the political reality inside Myanmar is of a stalled peace process, continuing conflict, and state violence. The promotion of return in these circumstances has considerable personal costs for refugees, in terms of risks to security as well as the emotional toll of uncertainty and disempowerment in decision-making. I argue that it also has more far-reaching political consequences as efforts to repatriate refugees have the effect of reinforcing the Myanmar's government's narrative of transition and transformation over the political reality of continuing conflict and oppression. This has implications for peacebuilding (by favouring the government and weakening other political actors) and state building (by normalizing the political repression and military dominance that has driven displacement from Myanmar for so many years).
The analysis presented here is drawn from more than ten years of research with refugees from Myanmar, including substantial periods of fieldwork with refugee communities and extensive qualitative interviews with a range of stakeholders in Thailand between 2008 and 2014 and in Malaysia and India between 2012 and 2016, and continuing relationships with displaced communities and civil society in these countries. My most recent field trip was to Thailand in 2018 and included visits to refugee activists and community-based organizations in Chiang Mai, Mae Sariang, and Mae La Oon refugee camp. Some data from those interviews are included in this article.
The central contention of this article is that, while refugee policy is rarely foregrounded in analyses of international peacebuilding and state building, it nevertheless has the potential to influence political governance at multiple scales. This argument is presented in three stages. First, an overview is given of recent politics in Myanmar and the ‘triple transition’ of economic liberalization, democratic transition, and the peace process between the government and ethnic armed organizations. Second, an explanation of refugee repatriation in international policy and practice is provided, along with case studies of proposed refugee return to Myanmar: ‘facilitated voluntary return’ of refugees from Thailand, repatriation of Rohingya refugees from Bangladesh, and the cessation of refugee status for Chin refugees. Third, the effects of these policies are analysed in relation to their personal consequences for refugees and their national and international political influence. It is argued that this represents ‘border governance’ (that is, governance in, of, and through borders) at multiple scales, with implications for global border control, national border control, and the governance of political transition.Footnote 1 Recognizing these roles and effects has a number of implications, including showing that the relationship between refugee policy and national politics is not one way (that is, that refugees are affected by events in their country of origin) but that repatriation planning can itself be an intervention in national and international affairs. This is not only about the physical return of refugees across borders but a wider climate of acceptance and insistence that refugee return is viable. This support for repatriation has consequences for refugees in terms of the withdrawal of aid and assistance, the escalation of personal fear and uncertainty, and political bolstering of a Bamar-dominated government and military vis-a-vis ethnic minority groups.
Myanmar in transition
Myanmar was under military rule from 1962 and, after the violent suppression of pro-democracy protests in 1988, was largely closed off to the wider international community. Throughout those years, and indeed until the present, Myanmar has consistently generated internal and cross-border forced migration. An important driver of forced migration has been Myanmar's politics of ethnicity and identity. Myanmar is ethnically diverse, with an estimated two-thirds of the population of Bamar [Burman] ethnicity and one-third of non-Bamar ethnicity. The importance of ethnicity to Myanmar's national politics is reflected in its administrative division into eight regions (where the majority population is of Bamar ethnicity) and seven states where the majority population is a non-Bamar ethnic group (Chin, Kachin, Kayah [Karenni], Kayin [Karen], Mon, Rakhine, and Shan). Ethnicity is at the centre of Myanmar's constitutional law, with citizenship predicated on inclusion in a list of 135 ‘national races’. The exclusion of Rohingya from this list is a central contributing factor to their exclusion and persecution in recent decades.Footnote 2
Myanmar's post-independence politics (particularly post-1962) have been dominated by Bamar nationalism and an agenda of ‘Burmanization’ which emphasizes the superiority of Bamar ethnicity, Burmese language, and Buddhist faith. For non-Bamar ethnic groups this has led to discrimination in access to services and development, religious discrimination against non-Buddhists (including Christians and Muslims), and, ultimately, to conflict between ethnic armed organizations and the Myanmar military. It is this climate of discrimination, persecution, and political violence that has caused so much internal and cross-border displacement for so many years. There are large populations of refugees from Myanmar in Bangladesh, India, Malaysia, and Thailand, and at least 600,000 people are internally displaced within Myanmar.Footnote 3
Under military rule, Myanmar was a pariah on the international stage. This was rapidly reversed when the country embarked on political reforms in 2010. Aid and investment flooded in, even while the military remained in power. More than US$10 billion in international aid was disbursed in Myanmar between 2011 and 2015, and aid income accelerated after the installation of the NLD-led government in 2015.Footnote 4 The European Union (EU) alone provided almost one billion euro to Myanmar from 2012–2020.Footnote 5 In 2012, the World Bank characterized Myanmar as undergoing a ‘triple transition’: from military rule to democracy, from conflict in ethnic territories to peace, and from a centrally managed economy to a market economy.Footnote 6 By 2020, the depth of these transitions was being questioned. While economic reforms had driven a massive rise in GDP, which almost doubled between 2009 and 2017,Footnote 7 the political landscape was less encouraging. Despite highly visible milestones such as the 2015 elections and subsequent installation of an NLD-led government, the military continues to hold a constitutionally protected role in governance. Furthermore, the new government has retained or escalated many of the military's repressive policies, including restricting international aid to internally displaced persons and policing assembly and expression. Some commentators have questioned whether the real political transition is not to democracy but to autocracy, that is, the normalization and institutionalization of the military's role in Myanmar's politics.Footnote 8 In January 2019, the UN Special Rapporteur on the Situation of Human Rights in Myanmar concluded that, ‘instead of bringing about the democratic reforms that were promised, the civilian government is consolidating what military governments worked towards for many years’.Footnote 9
The risk of autocracy or ‘military-state capture’Footnote 10 is particularly acute in relation to the peace process with ethnic armed organizations. A lasting peace in Myanmar will only be achieved if the root causes of conflict in ethnic states are addressed: the discrimination, underdevelopment, military violence, and impunity that underlie ethnic demands for federalism and political autonomy. This would also require amending the 2008 constitution and beginning the work of rebalancing power between the military and the government as well as between Bamar and non-Bamar ethnic groups. To date the peace process has made little progress towards these goals, despite a Nationwide Ceasefire Agreement (NCA, currently with ten signatory ethnic armed organizations) and the 21st Century Panglong Peace Conference (convened in 2016, 2017, and 2018).Footnote 11
While the peace process has failed to deliver concrete outcomes, the Myanmar Army has continued to target ethnic territories. Ceasefires have been repeatedly violated in Chin and Karen states. In Kachin and Shan states, where no ceasefire has been agreed, the army has used air strikes against civilian populations.Footnote 12 Extreme violence against Rohingya and other Muslim communities culminated in a ‘human rights catastrophe’ in Rakhine state in 2017. An independent UN fact-finding mission found that human rights violations since 2012 ‘undoubtedly amount to the gravest crimes under international law’ and that ‘these abusive patterns are reflective of the situation in Myanmar as a whole’.Footnote 13 Exceptionally, there are cases pending against Myanmar before both the International Court of Justice (which, at the request of The Gambia, has already ordered provisional measures to prevent genocide against Rohingya in Myanmar) and the International Criminal Court (which is investigating alleged crimes of deportation and persecution committed in Rakhine state, where there is a jurisdictional nexus to Bangladesh, a state party to the Rome Statute of the International Criminal Court).Footnote 14
Two distinct narratives have emerged to interpret Myanmar's politics post-liberalization. One has accepted the ‘triple transition’ at face value and recognizes the government as an honest broker in pursuing political and economic reform, and peace with ethnic armed organizations. This version of events perceives continuing problems in the country as developmental rather than political, and that the best course of action for the international community is to support the government in its pursuit of political transition. This has been the approach taken by governments and international agencies such as the United Nations Development Programme (UNDP) and the United States Agency for International Development (USAID) (among many others) who have invested significantly in Myanmar, channelling all aid through the government in order to ‘strengthen Burman's transition to democracy’.Footnote 15 The second narrative considers that there is no transition or, at best, a very weak transition: that the government is either unwilling or unable to produce real change in its relations with ethnic minorities, that the military continues to dictate national politics, and that Myanmar remains a country at war. From this perspective, approaching the government as the only legitimate political authority in the country provides it with unearned political legitimacy (as well as an aid bonanza) and undermines other actors in the peace process. This view has been represented by some researchers and academics—particularly those working on issues of ethnic politics—and is very much the prevailing opinion among ethnic armed organizations and refugee communities.Footnote 16
International donors, investors, and humanitarian agencies have continued to work through and with the Myanmar government, despite evidence of repression, persecution, and pervasive military impunity.Footnote 17 There has been a tendency for international actors to flatten political understanding of Myanmar's ethnic conflicts, focusing on developmental questions and giving insufficient weight to the historical context of discrimination and oppression. This is a classic illustration of Ferguson's description of international development as an ‘anti-politics machine’.Footnote 18 A lack of contextual political understanding of Myanmar's historic and current conflicts bolsters an ‘over-reliance on developmental engagement strategies’, described by Stokke et al. as a key risk of international aid programmes, which may actually undermine peace and democracy in Myanmar.Footnote 19 Civil society activists have claimed that international donors have affected the balance of power in peace negotiations and ‘significantly weakened the negotiating power of the ethnic groups’ by uncritically accepting the official narrative of transition, and particularly by distributing all international aid through the government.Footnote 20 Paul Sein Twa explains that:
They [international actors] approach the situation as a normal case of statebuilding, which works in favour of the government's unitary system. The narrative is that poverty and the weaknesses of the government administration are behind the problems in Myanmar, while the root causes are actually political. The international community wants the NCA [Nationwide Ceasefire Agreement] to be signed no matter if the deal is good or bad for ethnic groups.Footnote 21
This article argues that refugee policy has occupied a similar, and thus far unrecognized, role as a political intervention into Myanmar's peace process which favours the government and undermines other political actors. Since the earliest stages of Myanmar's political reforms, refugees from camps in Thailand were encouraged to return. More recently, this has been accompanied by international support for the repatriation of Rohingya from Bangladesh, and a declaration of cessation of international protection for Chin refugees from Myanmar (issued in June 2018 and revoked in March 2019). With the exception of the Rohingya, these cases have attracted little international attention. These proposals have important implications in terms of their compliance with international refugee law (such as the requirement for repatriation to be voluntary or the threshold for declaring ceased circumstances that gives rise to refugee protection). They also have significant political implications. What are the consequences of supporting the government and military narrative of reform over refugees’ and citizens’ lived experiences of violence? Whether the country is safe for refugee return is a question with obvious and immediate consequences for refugees, but also for Myanmar's national politics, for the role of ethnic organizations in the peace process, and, ultimately, the position of ethnic minorities in the future of Myanmar.
Repatriation in principle and practice
Voluntary repatriation (that is, the return of refugees to their country of origin) is one of three recognized durable solutions to refugee situations, along with local integration in the host country and resettlement from the host country to a third country. Repatriation is frequently described by the United Nations High Commissioner for Refugees (UNHCR) as ‘the preferred solution’ for refugees, although it may be more accurate to say that it is the preferred solution for states. As a result, it is also the durable solution used for the largest number of refugees: by 2010, UNHCR had assisted in the repatriation of nearly 25 million refugees.Footnote 22 To be a durable solution, repatriation should offer a genuine end to displacement and a restoration of state protection. International standards to ensure the legitimacy of repatriation require that it is voluntary (that is, an individual choice) and that it is carried out ‘in safety and with dignity’.Footnote 23 As repatriation has become the ‘preferred solution’, these thresholds are often applied weakly, if at all, and the decision to repatriate is not determined by individual refugees but by international agencies and donors.Footnote 24
Repatriation is the act of returning refugees from one state to another across a defined national border. More than this, repatriation is understood as returning refugees to their place of belonging, their country of citizenship, where they will receive state protection. Drawing on Mary Douglas's analysis of social ordering in Purity and Danger, Malkki famously described refugees as ‘matter out of place’ in a world of sovereign states.Footnote 25 Repatriation ostensibly corrects this category error by, quite literally, putting refugees back in their place. By participating in repatriation, refugees are moving from conflict to peace in a spatial process of return which maps onto the political process of transition. A link between repatriation and peacebuilding has been explicitly asserted, at times as a direct and linear relationship, that is, ‘peacebuilding with people’ or ‘peace by repatriation’.Footnote 26 UNHCR policies have reinforced that approach by emphasizing repatriation as one stage in a larger process of personal and political transformation. For example, in the early 2000s, UNHCR's so-called ‘4Rs’ approach—repatriation, reintegration, rehabilitation, reconstruction—directly positioned repatriation as a tool of peacebuilding. This approach is no longer part of the agency's vocabulary, although the notion of repatriation extending beyond the act of return and into longer-term processes of reintegration remains very much alive.Footnote 27
A relationship between repatriation and peacebuilding has been repeatedly asserted but without much consideration of what such a relationship might say about the type of peace that is being built.Footnote 28 Analysing refugee returns in the 1990s, Chimni observed that a language of peacebuilding was being used to legitimize repatriation as a tool for the containment, rather than protection, of refugees.Footnote 29 He argued that this was not incidental but deliberate, that peacebuilding as a conceptual category was invented to ‘blur the distinction between war and peace’ and to normalize ongoing violence as ‘transition’ rather than conflict. This creates a number of political and economic opportunities for states: allowing global investment and structural adjustment programmes to be launched, and circumscribing the meaning of ‘peacebuilding’ by equating it with a rudimentarily accountable state, thereby condoning donor withdrawal from protracted refugee situations and the return of refugees. Chimni's position that international peacebuilding has not only failed to prevent, but actively legitimated the premature return of refugees is firmly relevant to current circumstances in Myanmar, where (as I explain below) international donors have disengaged from protracted refugee situations, invested heavily in Myanmar's domestic economy, and shown a marked lack of interest in demanding accountability for massive, ongoing human rights violations. Encouraging repatriation in these circumstances enables the containment of refugees and prioritizes national and international political interests above refugee protection. It is also an intervention into Myanmar's national politics, endorsing the state narrative of transition over the direct experience of refugees and other victims of state violence. This serves to strengthen the state politically and thereby weaken opposition actors, including ethnic armed organizations.
The discussion below examines three policies aimed at the return of refugees to Myanmar: facilitated voluntary returns of refugees from Thailand, proposed mass repatriation of Rohingya refugees from Bangladesh, and UNHCR's declaration of cessation of international refugee protection for Chin refugees (issued in June 2018 and revoked in March 2019). Refugees in these countries have very different living conditions and levels of acceptance and recognition in their countries of asylum. They also come from ethnic groups that have very different experiences of and potential for peacebuilding in Myanmar. Yet they have all been encouraged to return to Myanmar, despite the political realities of continuing conflict, massive human rights violations, and ‘military-state capture’.Footnote 30
‘Facilitated Voluntary Repatriation’ of refugees in camps in Thailand
Around 100,000 refugees from Myanmar are currently living in camps along the Thai-Myanmar border. These camps have existed for more than 35 years and have provided refuge primarily for people fleeing from conflict and persecution in southeast Myanmar, particularly Karen and Karenni states.Footnote 31 Refugees in Thailand received substantial international support during the years of military rule. Indeed, when external aid to Myanmar was restricted, refugees in Thailand became one of the primary recipients of international support, essentially becoming a symbol of the Myanmar state's brutality and of courageous democratic and ethnic opposition. These were years when the camps received considerable assistance from international donors as well as informal assistance from international volunteers.
Despite this history of international support for refugees, repatriation planning began at the onset of political reforms. By 2012 UNHCR had established institutional mechanisms for coordinating voluntary repatriation. In October 2014, UNHCR published its Strategic Roadmap for Voluntary Repatriation, which distinguished between three types of repatriation: spontaneous (organized by refugees without assistance), facilitated (UNHCR involved through the provision of information and possible assistance during voluntary returns), and promoted (UNHCR providing support for transport and a package of reintegration assistance). The Strategic Roadmap viewed repatriation as part of Myanmar's political transformation and predicted that ‘in the most likely scenario’, a Nationwide Ceasefire Agreement would be concluded in the latter half of 2014, promoted returns would begin in early 2015, and repatriation would be complete by 2017. Promoted returns would occur ‘once the conditions are appropriate for the promotion of return throughout South-East Myanmar’. When it became apparent that these conditions were far from being realized, UNHCR revised its approach from organized repatriation to ‘facilitated voluntary repatriation’. Voluntary repatriation centres were established in every camp and regular information sessions were held to educate refugees about return options and the support that UNHCR would provide (transport back to Myanmar, three to six months of food assistance, a mosquito net, and a hygiene pack).Footnote 32 The first ‘facilitated’ repatriation was carried out in 2017 between UNHCR, the government of Thailand, and the government of Myanmar. Notably, refugee representatives and camp committees were not included in the design and implementation of these returns. Only 65 people participated. Their reports back to the camps claimed that no arrangements had been made for them to access services, including education and health, and that 17 people returned to Yangon had no housing on arrival. UNHCR publicity material presented a very different story—one of successful returns and happy returnees—and posted laminated printouts of these stories around the refugee camps to encourage others to follow.Footnote 33 Three further repatriations were carried out in 2018 and 2019.Footnote 34 By July 2019, a total of 1,039 people (273 households) had participated in the voluntary repatriation programme.Footnote 35
Studies and surveys among refugees from Thailand have consistently found that they do not believe conditions are suitable for sustainable returns, giving a variety of reasons for this assessment: lack of confidence in the peace process; fears for their safety on returning, with risks perceived from the Myanmar Army and non-state armed groups; landmines in return areas; lack of houses, land, or jobs to return to; lack of documentation of Myanmar citizenship or of property ownership.Footnote 36 UNHCR estimates that 15,000 people have returned ‘spontaneously’ without notifying UNHCR.Footnote 37 There is evidence to suggest that this might be best understood as a coping strategy of ‘cyclical return’, in which families send one or more members back to Myanmar to ‘test the waters’, while keeping open the possibility of returning to Thailand should conditions change.Footnote 38 Furthermore, these returnees have foregone cash and transport assistance so as to avoid registering with the Myanmar government—in itself, a sign of lack of confidence in the new regime.
Refugees do not want to go back to Myanmar but life in camps in Thailand is increasingly difficult. Several donors have stopped or reduced support to refugees, often while opening new programmes inside Myanmar. In several cases this has been directly linked to the peace process. Both the UK and Norwegian governments—historically major donors to these refugee camps—opted to invest heavily in peacebuilding and reduce refugee funding accordingly.Footnote 39 The primary provider of food and shelter in these refugee camps, The Border Consortium, saw its funding drop by 50 per cent between 2012 and 2016. Funding cuts have affected programmes and services throughout the camps. The Border Consortium has repeatedly reduced food rations since 2012, stopped funding internally displaced persons in 2017, and suggested that in 2020 it will stop funding all camp staff.Footnote 40 Refugee-led systems of governance and service provision that have been built up over decades are crumbling due to a lack of resources. Camp-based schools have started charging high fees or closing entirely. Some donors have pressured schools to relocate to Myanmar and made future funding conditional upon doing so.Footnote 41 Services for refugees and migrants outside camps, such as the internationally renowned Mae Tao Clinic, have also been affected. UNHCR in Thailand claims it has not encouraged people to return but merely provided ‘necessary assistance’ to those who want to return.Footnote 42 Refugees perceive it differently, viewing the creation of voluntary repatriation centres, cuts to services, and repeated ‘preferences surveys’ as pressure to return that is eroding the voluntary nature of repatriation. These views came through very clearly when I visited Mae La Oon in May 2018, when every person I spoke with emphasized their fear of return and anxiety for the future:
In camp because of the noise of ‘return, return’, we worry. We talk a lot what will we do. I am still traumatized from running away from fighting. The noise of shooting is still in my ears. When will we have to go back? When? I cannot sleep. I worry. I cannot imagine if I have to go back.Footnote 43
I am already tired from running in the jungle. I don't want to see Burma Army any more. I am already afraid. My family have told me that women are raped by the Burma Army. If we are lucky we will live. If we not lucky we will die. It would be better to die [i.e. commit suicide] than return.Footnote 44
UNHCR was widely perceived as pressuring refugees to return but also as an unwitting stooge of the Myanmar government, tricked by its narrative of transformation:
UNHCR share news that everything is very good but actually it is not. Myanmar Government is also saying to Rohingya people, you can come back too. This is propaganda. It is not true. We cannot go back. It is very obvious that we cannot go back. But we know the Myanmar Government does not tell the truth. They will show international groups the good things and after the people will suffer. We know the Myanmar Government will not do the right thing. We can never win against them.Footnote 45
These donors don't know the real situation in Burma. They think everything is ok but they don't know the real situation.Footnote 46
There is a lot of pressure from UNHCR now about returns. Have they got a place where we can be safe?Footnote 47
Refugees saw the emphasis on return accompanied by the removal of other options, particularly the possibility of remaining in camps in Thailand for much longer:
Before, we never heard about return. Now many people come and say ‘You have to think about your homeland. It is time to go back’. We hear this all the time. It is a big problem. We cannot go back. We have nothing to go back to, no land. Also we cannot go back. So if we cannot go back and we cannot stay here it is a big problem.Footnote 48
Of course we know that we cannot go back. It seems like we can't stay here either. They say that we can't stay a long time but don't give us a date either.Footnote 49
It is very challenging. Different people come to the camp and a lot of information is given about returns, by UNHCR and even by the Thai authorities. The community have a lot of questions. ‘When will we go back?’ They are asked, ‘Where do you want to go back to?’ But they can't answer. They are not from cities and their area is still under fighting. So they come to the Camp Committee and ask ‘Where should we go? When do we have to go?’Footnote 50
Throughout all of these interviews there was a sense that the gulf between UNHCR's emphasis on return and refugees’ own understanding of Myanmar's politics was unbridgeable. The resulting fear and uncertainty had become all-consuming for many people.
Repatriation of Rohingya refugees from Bangladesh
A generous interpretation might say that UNHCR's role in encouraging repatriation from Thailand has been simply political optimism, a miscalculation of the degree of transformation in Myanmar rather than anything more cynical. This explanation is less convincing in relation to the proposed repatriation of Rohingya refugees from Bangladesh. Rohingya have been persecuted in Myanmar for decades, with policies restricting travel, marriage, birth, voting, and other rights that have established ‘a continuing situation of severe, systemic and institutionalized oppression from birth to death’.Footnote 51 Underpinning all of these policies has been the exclusion of Rohingya from recognition as a ‘national race’ and related exclusion from citizenship and widespread statelessness. The cumulative effect of these policies was described by Zarni and Cowley in 2014 as a ‘slow-burning genocide’.Footnote 52 In August 2017, this became a highly visible genocide when devastating military attacks were unleashed on Rohingya villages in Rakhine state, killing an estimated 10,000 people and forcing more than 700,000 people to flee to Bangladesh. A UN independent fact-finding mission recommended the prosecution of several high-ranking Myanmar military officials for acts of genocide against Rohingya, and for crimes against humanity and war crimes committed in Rakhine, Kachin, and Shan states.Footnote 53
Despite this context of extreme violence, endemic persecution, and entrenched impunity for the perpetrators, UNHCR and other international stakeholders cooperated with proposals to repatriate Rohingya refugees from Bangladesh to Myanmar. In November 2017, Myanmar and Bangladesh signed the Arrangement on the Return of Displaced Persons from Myanmar in Bangladesh. This arranged for return to commence in January 2018 for ‘residents of Rakhine State who have crossed over to Bangladesh after 9 October 2016 and 25 August 2017’, excluding around 300,000 Rohingya who were present in Bangladesh prior to that date. The agreement uses the language of international refugee law, recognizing ‘the need for sustainable and durable solutions’ and agreeing that Myanmar and Bangladesh will ‘start the process of voluntary return in safety, security and dignity’. There are no concrete provisions to ensure the safety, security, and dignity of the Rohingya, with the text of the agreement merely stating that returnees will return ‘voluntarily and safely to their own households and original places of residence’ where ‘their freedom of movement will be allowed in conformity with the existing laws and regulations’.Footnote 54
The repatriation agreement provoked an international outcry.Footnote 55 Refugees protested against the repatriation and outlined preconditions for safe return, including restoration of both citizenship and land. Despite this opposition, in June 2018 UNHCR and UNDP signed a memorandum of understanding to cooperate in repatriating Rohingya to Myanmar. The EU high representative Federica Mogherini described the agreement as ‘an important and welcomed step’ which the EU wanted ‘to be implemented without delay’ and pledged five million euro towards its implementation.Footnote 56 The foreign minister of Bangladesh responded to domestic and international criticism by saying, ‘Our only goal is to send the Rohingya back to their country, and there is no point in criticizing this agreement.’Footnote 57
This process must be understood in light of a history of forced returns of refugees from Bangladesh to Myanmar, including in 1978 and 1992–1994.Footnote 58 These processes are explored in an unpublished report written for UNHCR's policy and evaluation unit, which was based on research in UNHCR's archive and offers unique insight into the agency's internal communications surrounding these repatriation processes.Footnote 59 In 1978, UNHCR was not a party to the repatriation agreement but had a role in monitoring the returns process and the voluntariness of return. Refugees experienced physical violence and other forms of coercion to return, including withdrawal of food aid. By the end of November 1978, the death rate in the camps was eight-and-a-half times the Bangladeshi national average.Footnote 60 UNHCR was repeatedly notified of this by a staff nutritionist and through camp reports, yet took no action.Footnote 61 Over the course of the repatriation, as many as 12,000 people in the camps died unnecessarily.Footnote 62 Return conditions in Myanmar were also problematic. Violence occurred in return sites during 1978, and returnees faced serious difficulty in finding food and livelihoods. By the end of 1979, nearly 190,000 Rohingya refugees had been repatriated under this programme.
Conditions in Myanmar worsened for Rohingya during the 1980s and escalated to another phase of extreme persecution between 1991 and 1994, when more than 250,000 people crossed to Bangladesh. Again, a bilateral repatriation deal was quickly agreed and returns started in September 1992. UNHCR was initially critical but subsequently signed memoranda of understanding (MOU) with Bangladesh in May 1993 and Myanmar in November 1993, and began to actively promote large-scale repatriation on the basis that repatriations were voluntary and conducted in conditions of safety.Footnote 63 The MOUs were never made public but a statement issued by UNHCR at the time describes the agreement with Myanmar as guaranteeing that returnees would be issued with appropriate identification papers and would enjoy the same freedom of movement as other Myanmar nationals and that UNHCR would be granted full access to all return sites.Footnote 64 These guarantees were not fulfilled. An estimated 200,000 people were returned between 1992 and 1994, into the same conditions of insecurity and persecution.
These past processes established a template that has also been followed in recent events: a bilateral repatriation agreement between Bangladesh and Myanmar which was negotiated even as Rohingya villages were being destroyed in Myanmar, and a process which secured UNHCR cooperation despite failing to provide any guarantees of safety and security. Above all, these past processes show complete disregard for the safety of returnees.Footnote 65 The prevailing mentality within UNHCR in 1993 is reflected in a staff member's comment that, ‘These are primitive people. At the end of the day they will go where they are told to go.’Footnote 66
In 2020, it is not clear that much has changed. UN independent experts have insisted that repatriation cannot be countenanced: ‘it is clear that Rohingya refugees in Bangladesh cannot return to Myanmar in the near future’; ‘it is evident that Myanmar is not working to create conditions for return for the Rohingya but is engaging in a sustained campaign of violence, intimidation and harassment’;Footnote 67 and ‘in the current circumstances, returns are not possible’.Footnote 68 Yet, the UN secretary-general's special envoy to Myanmar (a post created by the General Assembly in 2017) has enthusiastically praised the smallest of gestures (‘lauded the drafting of an information sheet for the refugees in Cox's Bazar’) and accepted political statements at face value (‘welcomed the stated willingness of the Rakhine leaders’ to accept the refugees’ return and ‘was satisfied to hear the Rakhine State Minister explain that the National Verification Cards are not for foreigners, but for people residing in Myanmar. It is a first step towards citizenship.’)Footnote 69
Repatriation of Rohingya had not started by January 2021. In 2019, Bangladesh's prime minister located the delay solely with Myanmar rather than with UNHCR or the international community: ‘The problem lies with Myanmar [...] Everybody wants the Rohingya people to return to Myanmar […] Myanmar is not interested in taking them back.' In December 2020, the Bangladesh government began relocating refugees from camps in Cox's Bazaar to an off-shore island, Bhasan Char. Low-lying, made of silt and sediment, and located miles from the Bangladesh coastline, this island is at extreme risk of flooding, cyclones, and other natural disasters. Relocating Rohingya refugees to this location had been vehemently opposed by refugees and human rights organisations. The UN issued a statement that it had not been involved in preparing the movement of or identifying refugees to be relocated and reiterated ‘its longstanding position that Rohingya refugees must be able to make a free and informed decision about relocating to Bhasan Char’.Footnote 70
In contrast to refugees in Thailand, Rohingya refugees in Bangladesh are continuing to receive international support as a high visibility and high priority refugee population. However, this has not impeded the repatriation agreement, which was signed despite continuing violence in Myanmar and, indeed, despite international recognition of that violence constituting the most serious crimes in international law. The United Nations has continued to support repatriation to Myanmar, including by extending its repatriation MOU with Myanmar until June 2021. A new round of repatriation talks between Myanmar and Bangladesh were scheduled for January 2021, with China mediating. There is no question that Rohingya refugees must be able to return to Myanmar, but this can only take place when their status in the country has been secured. At present it is evident that the Myanmar government has no intention of granting citizenship or equivalent rights protection to Rohingya returnees. The result for Rohingya refugees is an impossible dilemma: unwanted in Bangladesh and unprotected in Myanmar. There is no easy solution to this situation, but there is also a direct historical comparison for the likely result of refugee return without meaningful political change. In 1978, and again in the 1990s, thousands, or even tens of thousands, of people died as a direct result of the premature returns process, while an unknowable number of returnees were subjected to violence and persecution in the years that followed. This is a stark lesson that premature return costs lives and that facilitating return in the absence of real legal and political reform will merely empower the Myanmar government and military to commit further violence.
Cessation for Chin refugees
A third example of international actors encouraging the premature return of refugees to Myanmar relates to Chin refugees. Chin state is in western Myanmar, bordering Rakhine state, and people from this area have predominantly sought refuge in India and Malaysia. In June 2018, against the backdrop of the extreme military onslaught in Rakhine state, UNHCR issued a declaration of cessation of refugee protection for the Chin refugee community, announcing that Chin refugees were no longer in need of international protection and their refugee status would be ended. The statement said (in part):
An analysis of the political, social and security development in Chin State over the course of several years has determined that there are positive developments which are durable and sustainable in nature. While the situation in Myanmar is still fluid and unstable in parts of the country, the situation in Chin State is now stable and secure from a refugee protection perspective. Based on this UNHCR presumes that Chin refugees are safely able to avail themselves to the protection of the Myanmar Government and hence are no longer in need of international protection from UNHCR.Footnote 71
Refugees were given the choice of accepting the declaration of cessation (in which case their refugee status would be removed at the end of 2019) or challenging the decision and asserting a continued need for international protection. Those who opted for the latter would have their refugee status immediately revoked while a case review was conducted. In the meantime, their UNHCR card was removed and a piece of paper declaring that the holder's case was under review given in its place. No time frame was given for the review process other than an ‘intention’ to complete it by December 2019. This was particularly problematic because, although UNHCR refugee registration is not officially recognized by the Malaysian government, UNHCR cards have strong informal protection value and card-holders are largely insulated from the widespread arrest, detention, caning, and deportation that is inflicted upon undocumented migrants.Footnote 72 Implementation of cessation began in Malaysia in August 2018 (affecting 16,000 registered Chin refugees and 35,000 asylum seekers) and in Delhi in December 2018 (affecting 5,000 registered refugees). The decision to end international refugee protection for Chin was justified in terms of Myanmar's transition and transformation. In 2018, UNHCR's representative in Malaysia Richard Towle commented, ‘We should be welcoming, not condemning the changes in a country that allow people to think about going home. The objective is to get people back to their communities, and the cultures that they know and love.’Footnote 73
The 1951 Refugee Convention recognizes the possibility of ending refugee status when there is no longer a need for international protection but sets a high threshold for doing so: ‘when the circumstances in connection with which he has been recognized as a refugee have ceased to exist’.Footnote 74 This is a deliberately high threshold, reflecting the life-or-death consequences of removing refugee status. UNHCR policy guidance has reinforced the importance of a strict interpretation of this standard,Footnote 75 with the 2003 Guidelines on Cessation indicating that the threshold might be satisfied when there is substantial and fundamental change that is enduring and where effective protection has been restored. UNHCR's statement of cessation for Chin refugees used this terminology (‘positive developments which are durable and sustainable in nature’) but was announced only months after the Myanmar Army had committed widespread international crimes in Rakhine state, which is directly adjacent to Chin state. The cessation declaration also contravened UNHCR guidance by applying to a single geographical region and leaving those affected by the decision with uncertain status.Footnote 76 Finally, it is significant that the cessation of international protection for Chin refugees was apparently a unilateral decision by UNHCR, without pressure from Malaysia or India (the states hosting Chin refugees). A new government was elected in Malaysia in 2018, the first new government in 61 years. Its election campaign had pledged human rights reforms, including better treatment of refugees. Successive Indian governments have historically adopted a position of benign neglect towards Chin refugees (who are overwhelmingly Christians), offering little direct assistance but also not insisting that they return to Myanmar. In declaring cessation for Chin refugees in June 2018, UNHCR was taking a harder line than the incumbent Malaysian and Indian governments, making this a truly ‘unnecessary cessation’.Footnote 77
The cessation decision was lifted on 14 March 2019, after nine months of intense stress, anxiety, and hardship for Chin refugees. The decision to lift cessation was as sudden as the decision to impose it. A UNHCR press release described the reversal as based on ‘new reports and assessments which did not support its original conclusion of fundamental and durable changes’ and ‘noted specific concerns raised by the Chin community and civil society organizations’.Footnote 78 The press release stated that ‘No Chin refugee had lost their refugee status’ as a result of the cessation declaration. This implies that the declaration was harmless, which was certainly not the case. For almost a year, Chin refugees lived in a sustained state of fear and panic at having their refugee status removed. People made irreversible choices based on this decision, including, tragically, several cases of suicide.Footnote 79
Commonalities across contexts
At the time of writing (April 2020) mass repatriation had not yet been implemented for refugees from Myanmar but the promotion of return has inflicted lasting consequences. In Thailand, Malaysia, and India, international acceptance that repatriation is feasible and desirable has resulted in the withdrawal of funding and services to refugees and a weakening of refugee-led organizations. There have also been consequences for refugees’ security, as their status in hosting countries was made less stable. This was particularly dangerous in Malaysia, where refugees who challenged the declaration of cessation had their refugee status removed until their case was resolved and were accordingly much more vulnerable to arrest, detention, caning, and even deportation. In Thailand, food rations and services have been reduced for several years, which is likely to have lasting consequences for children's education and health.
Refugee status exists to provide surrogate protection in the event of a failure of protection by the state of nationality. For refugees from Myanmar, the focus on repatriation as their only option has been accompanied by a removal of other durable solutions, either explicitly (as in the case of refugees in camps in Thailand or Chin refugees in Malaysia and India) or implicitly (in the case of the Rohingya). Refugees are unable to return safely but are afraid that they will no longer be permitted to live as refugees either. This raises serious questions about compliance with UNHCR's international standards for repatriation to be voluntary and conducted in safety and with dignity.Footnote 80 It has also been immensely damaging for refugees. In Bangladesh, the repatriation agreement ‘caused high levels of fear and anxiety amongst the already traumatised refugee population in Cox's Bazar, and led to refugees going into hiding or even attempt to take their own lives to avoid being forced to return to Myanmar’.Footnote 81 Similarly, in Thailand and Malaysia, community organizations report rising mental health concerns and an increase in suicides related to anxiety, uncertainty, and fear of return.Footnote 82
Another consistent thread across the contexts is that decisions for refugees’ futures have been taken externally and driven in particular by international agencies and donors.Footnote 83 Refugee committees in Thailand were not consulted by UNHCR in designing and implementing ‘facilitated voluntary repatriation’. Chin refugees and community organizations were not consulted before UNHCR declared the cessation of their refugee status. Rohingya refugees and their representatives in Bangladesh have not been able to participate in decision-making about their future. Despite their formal exclusion, refugees in all three cases discussed here have persistently asserted a political voice, using any and every mechanism available to them, including community-based organizations, diaspora organizations, and international media. Refugee organizations have argued for ‘nothing about refugees without refugees’ and clearly identified their needs and priorities regarding return, including legal status and citizenship (particularly for Rohingya refugees) as well as provision for housing and sustainable livelihoods, gendered assistance, and recognition of qualifications gained in exile.Footnote 84 Refugees have advocated for legal change inside Myanmar (particularly in relation to land policy), have organized protests outside UNHCR offices, and explained their position repeatedly to national and international media. These efforts have had little purchase. Failure to involve refugees in policy decisions is the norm in refugee situations but it is also significant, reflecting the reality that these processes are not really ‘about’ refugees at all but are driven by other political interests and priorities. It remains important to ask whose voices are being heard and whose are being silenced.
Refugee policy as border governance
As shown above, refugees in camps in Thailand, Chin refugees in India and Malaysia, and Rohingya refugees in Bangladesh have experienced consistent pressure to return, despite continuing conflict and state violence in Myanmar. It is the contention of this article that this policy approach has not only had serious consequences for refugees but also for Myanmar's domestic politics. In this section I outline those consequences and argue that the premature promotion of return reproduces boundaries and hierarchies of power, constituting ‘border governance’ (that is, governance in, of, and through borders) at multiple scales.
Withdrawing donor support from refugees and promoting return to Myanmar have been justified in terms of peacebuilding and transition. However, Myanmar's political transformation remains fragile at best and forced migration has actually increased since political liberalization began.Footnote 85 Critical peacebuilding scholars have long recognized that international interventions in conflict-affected countries often fail to deliver on their ambitions, for reasons including a focus on economic growth rather than poverty reduction, an emphasis on state-centric reforms at the expense of political pluralism and hybridity, and programmes designed around the perceptions of political officials rather than the lived experiences of a wider population.Footnote 86 Richmond and Pogodda have argued that:
As has become obvious over time, statebuilding and liberal peacebuilding strategies fail to connect with their target populations. Instead, they end up buttressing problematic elites and their often chauvinistic, nationalistic, or personal interests.Footnote 87
Similar outcomes have been observed in diverse settings, including Rwanda, Somaliland, Sri Lanka, and Afghanistan.Footnote 88 In Myanmar, researchers have investigated the impact of international aid and trade in a variety of policy areas, including development and peacebuilding assistance, energy, transport, rural development, peace and security, and humanitarian aid.Footnote 89 Refugee policy has not been analysed in this way, though the return of refugees can be read as a powerful political statement by international stakeholders, and one that risks buttressing (in Richmond and Pogodda's words) ‘chauvinistic, nationalistic or personal interests’ against the needs of refugees and Myanmar's broader population.
The policies for the return and repatriation of refugees outlined in this article challenge the central assumptions of repatriation as a restoration of citizenship, re-entry into a community of belonging, and as an act of peacebuilding. Humanitarian standards for voluntary return in safety and dignity have not been met, yet many international agencies and donors have apparently accepted return as the best and only option for refugees. Chimni has observed that ‘humanitarian factors do not shape the refugee policies of the dominant states in the international system […] non-humanitarian objectives are pursued by these actors from time to time behind the façade of humanitarianism’. I argue that three ‘non-humanitarian objectives’ are particularly evident in the pursuit of the premature return of refugees to Myanmar. One such objective is almost certainly global border control, that is, keeping refugees in their region of origin or returning them to their country of origin to ensure that they do not travel onwards to seek asylum elsewhere. There is a great deal of analysis of refugee outflows as a challenge to state borders, and of the evolving strategies of border control that seek to deter and prevent asylum-seeking.Footnote 90 Much less attention has been paid to the ways in which refugee return can also serve purposes of border control and migration management, particularly when it is carried out as an ostensibly ‘legitimate’ repatriation rather than state push-backs or deportations. This argument is relevant to Myanmar but also to several other contexts where repatriation has recently been encouraged or carried out under the auspices of peacebuilding, including Somalis in Kenya, Afghans in Pakistan, and Syrians in Lebanon.Footnote 91
A second objective promoted by refugee return is to reinforce national borders between Myanmar and its surrounding states, reducing the porosity of these borders by reducing the likelihood of refugee protection. This dynamic seems to be more important for India and Bangladesh than Thailand and Malaysia (where the pressure for repatriation appears to have originated with UNHCR). Bangladesh has tolerated the presence of Rohingya from Myanmar for many years, though in most cases it has refused to formally recognize them as refugees. At times of escalation and mass influx, however, the response has always been to push for repatriation at the first opportunity and as the only option.Footnote 92 Similarly, India continues to host large populations of Tibetan and Tamil refugees as well as other ethnic groups from Myanmar, but has taken strong action to prevent any onward migration of Rohingya from Bangladesh, stationing thousands of soldiers along the India-Bangladesh border, pushing back or returning Rohingya who attempt to cross into Mizoram, and attempting to deport around 40,000 Rohingya in New Delhi who have been recognized as refugees by UNHCR. The return of refugees is part of this wider portfolio of mechanisms to restrict asylum, and is another means for Bangladesh and India to assert their borders and insist that the Rohingya are Myanmar's responsibility. This aspect of border governance is also apparent from the text of the Myanmar-Bangladesh repatriation agreements, which not only outline the agreement to repatriate but reiterate general border control roles and responsibilities of the two countries.Footnote 93
A third ‘non-humanitarian’ objective is to define the boundaries of political transition by reinforcing an official narrative of peace and transition over lived realities of conflict, insecurity, and continued persecution. As in Chimni's observation of the 1990s repatriations, the promotion of refugee return has been accompanied by donor withdrawal from long-standing refugee communities and increased international investment in a rights-violating regime. However, conflict is not erased by renaming it ‘peace’ or ‘transition’, and premature repatriation can be understood both as an act of violence in itself and as a precipitating factor for other forms of violence. Furthermore, redefining peace has consequences for the balance of power in political transition. International acceptance of repatriation supports a particular type of peace—a ‘victor's peace’ that is aligned with the interests of Myanmar's government and military rather than other political actors. This risks reinforcing a unitary Myanmar state at the expense of ethnic and political pluralism and, ultimately, against the long-term interests of returning refugees.
Thus, policies for the return of refugees to Myanmar can be understood as border governance in multiple dimensions: governing global borders by containing refugees in their region or country of origin; governing national borders between Myanmar and neighbouring countries; and governing political transition by bolstering a narrative of peacebuilding and development, regardless of the authenticity of those claims. Recognizing these functions and consequences of repatriation highlights the importance of shifting the analytical frame of political transition to incorporate displacement, with a view to ensuring that policies for displaced people can be politically informed, contextualized, and rooted in the preferences of refugees themselves. This analysis also highlights the importance of broadening understanding and analysis of repatriation, and recognizing that the effects of promoting refugee return begin even before mass returns have been carried out and can reverberate far into the future.
