‘These are primitive people. At the end of the day they will go where they are told to go.’ These were the words of a senior UN High Commissioner for Refugees (UNHCR) staff member in 1993, during a meeting convened to discuss potential solutions for the 250,000 Rohingya refugees who had recently fled from Myanmar to Bangladesh. The dismissive tone of the statement was emblematic of the organisation’s engagement with the Rohingya, which for many years showed limited respect for their rights and a readiness to abandon UNHCR’s own protection principles.+For a discussion of those principles and their implementation, see Jeff Crisp and Katy Long, ‘Safe and Voluntary Refugee Repatriation: From Principle to Practice’, Journal on Migration and Human Security, 4(3), 2016 (http://jmhs.cmsny.org/index.php/jmhs/article/view/65). Drawing on previously unpublished material from the UNHCR archives, this article shows how, in both the 1970s and 1990s, large numbers of Rohingya refugees were returned to Myanmar in a manner that was premature, involuntary and unsafe. The article concludes by asking whether a similar scenario could develop in relation to the 700,000 Rohingya refugees who fled to Bangladesh in the second half of 2017.
An unwanted community
The Rohingya are a Muslim minority group whose presence in Rakhine State in Myanmar dates back to at least the twelfth century. Their number expanded significantly as a result of labour migration from Bengal, both in the pre-colonial period and under British rule. Violence broke out between the Rohingya and Rakhine Buddhists following the Japanese invasion of Myanmar in 1942, and persisted once the country gained independence in 1948. While the new Constitution and Citizenship Act did not explicitly render them stateless, the Rohingya were excluded from a list of officially recognised ethnic groups. National Registration Cards issued by the military government after 1962 stated that ‘holding this certificate shall not be considered as conclusive proof of citizenship’. New restrictions imposed in 1974 reinforced their economic, social and political marginalisation. A military assault on the Rohingya began in February 1978, prompting some 200,000 to flee to Bangladesh. Between August 1978 and December 1979 almost all of these refugees returned to Myanmar under a bilateral agreement between the two countries.
The 1978 repatriation
According to a later assessment prepared for UNHCR’s Evaluation Service, it was ‘highly questionable’ whether the 1978 repatriation was voluntary. ‘Early repatriation,’ that paper explains, ‘was the government of Bangladesh’s priority from the outset of the crisis.’+The paper was not published due to its controversial nature. To achieve that objective, a variety of tactics were employed. From late 1978, Bangladeshi security personnel and government officials were involved in a number of attacks on the Rohingya. Commenting on these events, one UNHCR official observed that the refugees ‘were at best very reluctant’ to repatriate, but ‘to this end the Bangladesh government was prepared to go to considerable lengths, and ultimately, it seemed, to use force if necessary’. When asked by UNHCR headquarters to explain why a growing number of Rohingya were returning to Myanmar, the official replied that ‘one reason could be that the refugees were disturbed about the serious incidents in the camps which had led to a number of deaths … They realize that they really had no future in Bangladesh and preferred to return home’.
Another method used to induce repatriation was to withhold food and other essential assistance from the refugees. One experienced UN official described the Rohingya camps as ‘death traps – the worst I have ever seen’. Malnutrition was soaring, and in July 1978 the mortality rate in the camps was four times higher than in the rest of Bangladesh. Even so, a senior official in the Ministry of Relief observed in one meeting: ‘Well gentlemen. It is all very well to have fat, well-fed refugees. But I must be a politician, and we are not going to make the refugees so comfortable that they won’t go back to Burma’. One UNHCR staff member observed that ‘by this time, Bangladesh government policy had become one in fact of starving the refugees into leaving’. ‘Lack of food,’ he concluded, ‘resulted in widespread malnutrition and death rates that were avoidable, but their greatest significance was in increasing the momentum of the return operation.’+Alan Lindquist, ‘Report on the 1978–79 Bangladesh Refugee Relief Operation’, Institute for Development Studies, University of Sussex, June 1979 (http://www.ibiblio.org/obl/docs/LINDQUIST_REPORT.htm).
The role of UNHCR
In January 1980, once the repatriation had concluded, UNHCR acknowledged in a press briefing that up to 10,000 refugees had died in the camps, citing epidemics as the principal cause. Privately, however, staff in the field agreed that the high mortality rate was a result of the government’s failure to release and distribute food supplies that were readily available in the country. ‘Can there be an excuse,’ one asked, ‘for an international organization like UNHCR, whose brief is refugee welfare, to acquiesce in a policy which results in more than 9,000 unnecessary deaths.’
As this quotation suggests, UNHCR played a highly questionable role in the 1978 repatriation, especially at the more senior levels of the organisation. Early in the emergency, a fact-finding mission to Bangladesh quickly concluded that the Rohingya had been subjected to human rights violations such as rape, shootings, beatings and forced labour, and that they could consequently be considered refugees. But the organisation’s approach to the emergency was not consistent with that conclusion. In September 1978, when the refugees were being encouraged to return, the High Commissioner stated that ‘UNHCR would like to see that the repatriation agreement between Bangladesh and Burma is implemented successfully’. It was a position adopted on pragmatic grounds, based on an understanding that the refugees were unwanted in Bangladesh, that a long term relief operation would be financially unsustainable and that a prolonged refugee situation could lead to a border conflict between the two countries. In the words of another senior staff member, it was ‘collective wisdom that the refugees had to go back as soon as possible’.
This position ran contrary to UNHCR’s protection responsibilities. The refugees were not consulted about the repatriation operation or informed of the conditions awaiting them in Myanmar. Little was done to halt the abuses in the refugee camps, and two field staff who tried to raise the alarm were removed from the operation. In an attempt to deflect criticism, UNHCR stated publicly that it had little involvement in the repatriation and claimed (implausibly, given its mandate) that it ‘was not entrusted with the responsibility of ascertaining the voluntary character of the decision taken by refugees to return’. Privately, senior staff agreed that UNHCR should ‘persuade the refugees to repatriate’. ‘The procedure,’ one observed, ‘is not to ask them if they wish to repatriate, but passively to repatriate them if they do not object to the action.’ In the words of the assessment undertaken by UNHCR’s Evaluation Service, this approach demonstrated a ‘reckless regard for the voluntary nature of the repatriation exercise’.
A new cycle of displacement
A further flaw of the 1978 repatriation was that it was not accompanied by any serious effort to ameliorate conditions for Myanmar’s Rohingya population. Indeed, four years later the country introduced new legislation that definitively excluded the Rohingya from citizenship. A decade later, Myanmar’s armed forces launched a new assault on them, involving killings, sexual violence and the destruction of settlements and mosques. In 1991 and 1992, around 250,000 Rohingya fled again to Bangladesh.
Subsequent events bore a striking resemblance to those of 1978. Bangladesh began negotiating a repatriation agreement with Myanmar in November 1991, when Rohingya refugees were still fleeing in significant numbers. While stipulating that returns would be ‘safe and voluntary’, the agreement also said that repatriation should begin by May 1992 and be completed within six months. Once again, pressure was placed on the refugees in an attempt to meet that target, with restrictions on food and other assistance. In February 1992, Bangladesh stated its intention to limit the aid provided to the Rohingya, as ‘it would not wish to create a pole of attraction for more refugees’. The negative consequences of this approach quickly became apparent. Just four months later, a UNHCR staff member observed that ‘in 1978, over 10,000 refugees died from problems relating to inadequate assistance. The present line of the government is coming dangerously close to creating a repetition of this tragedy’.
Engagement or withdrawal
With the support of a number of NGOs and diplomats, the UNHCR office in Bangladesh managed to persuade the authorities to limit restrictions on the provision of assistance. The organisation also stood firm on the principle that repatriation should be voluntary, with one senior staff member observing in May 1992 that ‘we should be ready to pass on our present relief responsibilities to some other agency and leave the scene if we come across clear evidence that we cannot stop an unacceptable repatriation’.
With growing evidence that the refugees were being subjected to physical and psychological harassment in an attempt to force their departure from Bangladesh, UNHCR announced in December 1992 that it had no alternative but to disassociate itself from the repatriation operation. However, in subsequent discussions with the government the agency agreed to continue its involvement, based on its conclusion that ‘UNHCR will experience problems with the authorities in attempting to stop them from applying pressure mechanisms in the camps to force people to repatriate’. ‘It will be equally difficult,’ the same document observed, ‘to convince them that some refugees will stay longer in the country while waiting for a durable solution. Most likely, the attitude of the authorities towards the refugees and UNHCR may not improve.’
Wishing to avoid a confrontation with the government, and recognising that many refugees had resigned themselves to repatriation, in May 1993 UNHCR signed a Memorandum of Understanding (MoU) with the government of Bangladesh. While the agreement stipulated that the agency would ‘assist in the smooth repatriation of refugees who opt to return on the basis of their own judgement’, UNHCR also agreed to ‘undertake promotional activities to motivate refugees to return home’. A separate MoU was subsequently signed between UNHCR and the Myanmar government setting out conditions for the return and reception of the refugees.
With these agreements in place, UNHCR felt able to engage fully with the repatriation. In July 1994, a senior official from UNHCR headquarters (themselves a Bangladeshi) announced that the organisation would actively encourage the return of the Rohingya, based on the belief that it was now safe for them to go back. Rather than verifying the voluntary nature of return on an individual basis, the organisation would conduct a mass repatriation registration exercise, with the onus being on refugees to decline to register if they did not wish to return. Few felt able to choose that option, and by mid-1995 200,000 had returned to Myanmar.
While lauded as a success in official UNHCR statements, this second repatriation was in fact both controversial and contradictory to refugee protection principles. As the Evaluation Service’s assessment stated, some UNHCR staff ‘were not convinced that the refugees really wanted to return or that the conditions for voluntary repatriation in safety and dignity could be met’. Staff ‘could not understand the change of policy and wanted to leave the operation’, while UNHCR managers in Bangladesh ‘felt trapped between pressure from headquarters to repatriate refugees and field staff who contested the validity of UNHCR’s involvement in such a repatriation’. Noting these concerns, as well as the many criticisms levelled at the operation by NGOs and human rights organisations, the review concluded that ‘refugees believed they had no choice and accepted a repatriation that they feared and did not wish to undertake because they were told to do so by UNHCR and the authorities and believed they had no other choice’. In other words, and as had been predicted before the operation commenced, they went where they were told to go.
The obvious question raised by this account is whether the 700,000 Rohingya refugees who fled Rakhine State for Bangladesh in 2017 (again as a result of a murderous assault by the Myanmar military) will again be induced to participate in a repatriation that is premature, involuntary and unsafe, and in which UNHCR is fully engaged.+For a more detailed discussion of the issues raised here, see Jeff Crisp, ‘We Must Not Repeat the Shameful History of Returning Rohingya Refugees’, Refugees Deeply, 17 January 2018 (https://www.newsdeeply.com/refugees/community/2018/01/17/we-must-not-repeat-the-shameful-history-of-returning-rohingya-refugees). There are certainly reasons to fear such an outcome. Donor states made it clear from a very early stage of the emergency that they consider a speedy return to Myanmar to be the optimal solution to the crisis. And while Bangladesh has responded in a very generous manner to the Rohingya influx, the government has also made it clear that the refugees are placing excessive strain on the country’s economy, environment and infrastructure, and that their future consequently belongs in Rakhine State.
In Myanmar itself, the signals are mixed. Under pressure from the international community, the government has endorsed the notion that the refugees should return by means of a safe and voluntary repatriation. However, it remains unwilling to address the issue of Rohingya citizenship, and there are serious doubts about the willingness of the military to countenance the return of the Rohingya. Even so, under pressure from the international community the government has endorsed safe and voluntary refugee returns, and has signed an agreement with UNHCR and the UN Development Programme (UNDP) (which has not been published) with respect to the repatriation and reintegration process. As for UNHCR, the agency is under serious pressure from governments – not only in the Rohingya context, but also in many other parts of the world – to ensure the speedy return of refugee populations, and in doing so to compromise voluntariness and safety.+With respect to Syrian refugees, for example, see Jeff Crisp, ‘It’s Far Too Early to Talk of Return for Syrian Refugees’, Chatham House, 10 August 2017 (https://www.chathamhouse.org/expert/comment/it-s-far-too-early-talk-return-syrian-refugees).
Two conditions may militate against another repatriation operation that fails to meet UNHCR’s protection standards. The first is that the plight of the Rohingya is under unprecedented international and media scrutiny, and it seems unlikely that an induced return of the type that took place in the 1970s and 1990s could proceed without a major public outcry. Second, on the basis of past experience the Rohingya themselves have developed a much stronger understanding of the conditions under which they are – and are not – prepared to return to Rakhine State. According to one recent survey undertaken in Bangladesh, ‘99 percent said they would go back only if certain conditions were met, the majority mentioning citizenship of Myanmar with acknowledgement that they are Rohingya; freedom of movement and religion; and their rights and dignity restored’.+Xchange Foundation, ‘Rohingya Refugee Survey’, 23 May 2018 (http://xchange.org/rohingya-repatriation-survey/). In that respect, not to mention the incredible resilience that they have demonstrated since arriving in Bangladesh’s overcrowded refugee settlements, the Rohingya are anything but ‘primitive people’.
Jeff Crisp is a Research Associate at the Refugee Studies Centre, University of Oxford, and an Associate Fellow at Chatham House in London.