There are currently over 20 million people ‘of concern’ to the office of the United Nations High Commissioner for Refugees. Just over half of those are internally displaced or stateless, with 8 million having fled across an international border. Established in 1950, unhcr was charged by the 1951 Convention relating to the Status of Refugees with the protection of their interests: full political and economic rights in the country of asylum, with the hope of eventual voluntary repatriation. As a brutal testament to its contemporary failure, at least 3.5 million of those refugees currently struggle for survival in sprawling camps in Africa and Asia. Fleeing from genocide, imperial aggression and civil war, only to be herded into camps or sent back to the country they were escaping, these asylum-seekers and returnees are part of a seemingly endless human tragedy. If it was originally a guarantor of refugee rights, unhcr has since mutated into a patron of these prisons of the stateless: a network of huge camps that can never meet any plausible ‘humanitarian’ standard, and yet somehow justify international funding for the agency.

Like many of the un’s specialized agencies—the World Food Programme, the un Development Programme and others—unhcr functions independently of the General Assembly. Most of these bodies have their own assemblies and compete with each other for their portfolio, prestige and funds.footnote¹Responsibility for the 4 million Palestinian refugees remains with the un Relief and Works Agency but, partly through its support for both refugee camps and repatriation, unhcr has successfully encroached on the territory of the development organizations. Financed by donations and periodic appeals, rather than as a structural part of the United Nations, it has always been constrained by the interests of the rich ‘donor nations’, and its level of funding largely depends on how it sells emergency relief operations to the West. During the 1980s the United States criticized unhcr for being too ‘legalistic’, and concerned with protecting refugees in America and Europe: it wanted a focus on relief operations in the South. Jean-Pierre Hocké—a Swiss car-salesman turned Red Cross official—was appointed in 1986 to reform the agency. He began to focus the organization on the mass return of refugees, at one point provoking a staff revolt by cutting food rations to Ethiopians who declined ‘voluntary repatriation’, and failing to condemn the forcible repatriation of the Vietnamese boat people. When Hocké was pushed out of office in 1989 by financial crises and allegations of corruption, the Norwegian Thorvald Stoltenberg briefly held the fort before the appointment of Sadako Ogata as High Commissioner at the end of 1990. The agency that Hocké and Stoltenberg left behind was demoralized and unsure of its post-Cold War purpose. By the time Ogata left in 2000, its mandate would have been transformed. The publication of her memoir provides an opportunity to track unhcr’s evolution during that ‘turbulent decade’, and assess its changing responsibility for the camps and their inhabitants.footnote²

Ogata took up her post during the whirlwind of Operation Desert Storm. Within months, the reassuringly-named Operation Provide Comfort was launched, with un and unhcr support, in order to corral hundreds of thousands of desperate Kurds back into Iraq. Presented to the media as an emergency relief campaign, it signalled the beginning of a successful reform of unhcr’s mandate and methods, in the direction originally charted by Hocké. During the 1990s, Ogata—soon with the help of Kofi Annan—was to retool this apparently outdated agency as an instrument for the new age of humanitarian warfare. The free world could no longer score political points by opening its doors to the oppressed, and the nationalist and ethnic rivalries stoked by Western intervention in the Middle East, Africa and the Balkans led to refugee movements that the donor countries were no longer willing to accommodate. The often bloody compromise that Ogata brokered began with a shifting network of ‘safe havens’ and refugee camps, in war zones or near contested borders, guarded either by international troops or local militia. As soon as possible, the survivors would then be sent back to the territory they had tried to escape. The goal of repatriation—still the preserve of the far right in Europe and the us—had become the stated aim of the unhcr: a ‘humanitarian’ medicine well suited to the new disease of humanitarian war.

Reforming mission

Ogata is a scion of Japan’s political elite. Her maternal great-grandfather was Inukai Tsuyoshi, Prime Minister until his assassination in 1932, shortly after Japan’s invasion of China. Her academic career focused on this period, with a doctorate from Berkeley on Japanese foreign policy and the League of Nations. Ogata’s schooling in the language and bureaucracy of the un began in 1968 on the Japanese delegation to the General Assembly, and later continued at the Permanent Mission of Japan in New York. In 1979 Ogata led the Japanese plan for providing assistance to Cambodian refugees, and then during the early 80s she was Japan’s representative on the un Commission on Human Rights. Presumably it was her un background in the increasingly ubiquitous, and promisingly flexible, language of human rights that recommended her to Boutros-Ghali’s office in New York.

The Turbulent Decade is billed as an opportunity to reflect on the large changes in both the unhcr and its war-torn environment, whilst free from the pressures of office. None of the power-brokers of the period need fear any indiscretions, however: Ogata remains the consummate bureaucrat, firmly adhering to the accepted international narrative of un and nato heroism in tackling the post-Cold War world of ethnic rivalry, failed states and tinpot dictators. War is humanitarian throughout, havens remain safe even under fire, new and old nations are liberated by nato, and at one point Ogata even says that the war on Afghanistan was carried out on ‘reconstructional’ grounds. No direct terminology is risked when a euphemism lies easily to hand. The book tracks the development of four major refugee crises, selected for their scale and unhcr involvement, and as episodes that helped to reshape the way that the agency works: the first Gulf War and the Kurdish crisis, the Balkan wars from Bosnia to Kosovo, the Great Lakes region in Africa, and Afghanistan. Much of it seems to be compiled from Ogata’s diaries: uncomfortably interspersed with the unfolding human tragedies are name-checks of every airport landed in and functionary met, and a record of gifts from grateful recipients of unhcr largesse. Ogata’s conclusion, followed by her farewell speech and final briefing to the unsc, veers from lavishly praising every bureaucrat involved in this string of invasions, massacres and disasters to meekly requesting further unhcr powers and more resolute un action. If there were failings during this whole period, the remedy appears to be more troops, more of the time, with further powers to deploy police and other authorities regardless of sovereign boundaries.

Militarization of unhcr

For Ogata, the 1991 Gulf War was a resolute un action that unfortunately left Saddam Hussein still able to crush the Shia and Kurdish revolts that followed: no mention is made of the us role in encouraging these, or the empty promises of assistance. Tehran opened its borders and provided funds to assist the refugees, but even as their numbers swelled to 1.3 million, ‘coalition countries were loath to assist Iran’. The half a million Kurds fleeing towards Turkey faced a worse fate, as Ankara closed the border. They remained trapped on high mountain passes, without food and water in freezing weather. Ogata notes understandingly that Turkey faced a Kurdish insurrection within its own borders; therefore ‘the coalition member states, committed to maintaining the use of Turkish air bases for nato, were mute in response to Turkey’s refusal to grant asylum’. President Ozal pressed instead for ‘safe havens’ to be set up in the lowlands of Northern Iraq, backed by the us, uk and France. Resolution 688 was passed by the unsc to authorize Coalition access into Iraq to set up the camps, before handing them over to the un. The un appealed for funds for the operation, and Ogata was now thoroughly boxed in: ‘a bit perplexed by the complex setup of the un operational structure’, and further confused to find herself agreeing to look after ‘refugees’ still within their country of origin. This ‘severely tested unhcr’s protection mandate’:

Should we follow the legal dictate of not exercising our mandate inside the border and thereby refrain from helping those prevented from crossing, or should we stand on more realistic humanitarian grounds and extend whatever support we could?footnote³

Indeed: who would want to be unrealistic and inhumane, by demanding that refugees be offered passage to a safe country against Coalition wishes? Once the precedent had been established, the logistics of the transfer were largely handed over to the us Army, inaugurating an era of ever-closer cooperation with the military. Boutros-Ghali, concerned to preserve a veneer of independence, requested that un coordination with Coalition forces should be ‘discreet and not formal’, with particular attention paid to ensuring that us and un flags were not raised at the same time in the refugee camps. At this point the unhcr, ostensibly doubting the capacity of the camps, began to run in the direction in which it had initially been pushed, and argued for full repatriation of the Kurds back to their homes. Washington and Baghdad were agreed on one thing: the transfer of refugees from the mountains and the handover to the un was to be fast, to save face on both sides—Ogata quietly notes that ‘the pressure brought on the Kurds to return was immense’. The unhcr attempted to maintain the façade of a ‘voluntary repatriation’ by assuring the refugees that conditions were untenable in the mountains and that coalition forces would remain in northern Iraq to protect them. This last was little more than a lie: Washington offered Ogata nothing more than continuing to patrol the no-fly zone.

unhcr involvement in the Balkans was on a different scale: millions of refugees throughout the 1990s, with relief operations that began by providing the logistics for ethnic segregation and ended by mopping up after a nato invasion carried out without un sanction. unhcr involvement with displaced civilians in Croatia preceded independence, again bending the mandate; but as the break-up was ‘in prospect’ (not to say encouraged), Ogata ‘judged that the wisest course might be to become engaged ahead of time and take whatever preventive measures possible’. After the ec’s swift recognition of Slovenian and Croatian independence, us support for holding a referendum in Bosnia–Herzegovina—formerly the most ethnically mixed Yugoslav republic—increased nationalist tensions. As the operation there expanded over the following years, the agency gradually lost the will or means to distinguish between aid work and military action. Its operations were, according to Ogata, ‘political in objective and military in context’. By now unhcr had to consider its involvement in a process that could fairly be described as ethnic cleansing, whether through aiding separation by helping people to move, or advising civilians to stay in apparently unsafe areas. Ogata’s solution was a massive deployment of personnel throughout the region, with thousands of ‘field protection officers’ and dozens of sub-offices; it was clear that the relief effort was slowly becoming an occupation. The unsc-authorized Sarajevo airlift, which ran from 1992–96, involved military air and support crews operating directly out of unhcr offices.

Containment and repatriation

At an international conference on the Balkans in July 1992, Ogata had argued for two changes in the unhcr approach: ‘preventive protection’ and ‘temporary protection’. The former derived from both Iraq and the Balkans: ‘the right to be allowed to remain in one’s home’—or encouraged to stay in a war-torn area—as opposed to being given asylum abroad. The latter was a temporary form of asylum, intended to encourage reluctant European states to keep at least some borders open; the new status would provide limited protection without rights, least of all the right to remain. When the un Protection Force was expanded in order to support unhcr’s operations, Ogata was asked to address the unsc directly: ‘to me, it was like crossing the humanitarian Rubicon . . . no head of a humanitarian agency had ever addressed the Security Council.’ Resolutions 819 and 824 followed in 1993, declaring Srebrenica and other areas as safe havens, with Resolution 836 mandating unprofor to deter attacks. A unhcr working group acquiesced: ‘the overriding principle in Bosnia and Herzegovina should be to bring safety to the people, rather than to bring the people to safety’. The agency’s bureaucracy had now internalized the change in policy that had started in Iraq.footnote⁴

In a joint tv interview, Madeleine Albright clarified the purpose of the ‘safe havens’ for the us: ‘You cannot have a state without any territory, and what we’re trying to do through these safe havens is to establish these as areas that represent what might at some point be a Bosnian state’. Ogata claimed not to have ‘looked at the safe areas in such terms’—pawns for the us to redraw the map in the region—but conceded that they would affect the outcome of any later territorial division.footnote⁵ These un-run settlements soon became areas where troops could ‘rest, train and equip themselves’. As Serb attacks intensified the skeleton un force turned to nato, which had been champing at the bit: air strikes began in 1994, then nato’s Operation Deliberate Force in 1995, after which the Dayton Agreement split the state along ethnic lines. From this point, nato generals felt able to march into unhcr headquarters and give orders. A ‘pessimistic’ Boutros-Ghali, hamstrung by the un’s financial crisis—no mention of who might not have been paying their dues—could do little more than concede the field to nato, while Brussels and the World Bank were charged with reconstruction. unhcr staff on the ground struggled a little to help civilians stay in minority areas, but nato troops were escorting people across the new borders, with the effect of further entrenching ethnic division. Out of nearly three million refugees and internally displaced civilians, only a quarter of a million went back to their homes during 1996. All but a handful of those returned to areas where they were now the majority ethnic group.

Turning to Kosovo, Ogata does note that the American draft for the Rambouillet Agreement proposed that nato should have sole authority over freedom of movement throughout Yugoslavia; but she does not connect this provocative expropriation of sovereignty with the failure of the negotiations, merely observing that ‘the Federal Republic and the Serbs’ refused to sign. Nor is nato’s decision to act without un authorization discussed. Once the air strikes started, large numbers of Albanians began to flee to safety, apparently catching unhcr unawares; one hundred and fifty thousand in the first week, reaching three quarters of a million after two months. Macedonia objected to the numbers arriving, and—as with Turkey during the Gulf War—the alliance would not tolerate unhcr criticism of a country that hosted vital nato bases. Ogata meekly complied, even asking kfor to help build refugee camps on the border (‘nato was pleased to engage in a humanitarian mission’). unhcr handed over to nato almost all of its logistical work and, in addition, the job of leaning on Greece and other allied countries to get them to accept some of the refugees that Macedonia did not want.

Having presumably concluded that the un was powerless, Albania went straight to nato to get refugee camps built and provide assistance: nato swiftly deployed a force of eight thousand (afor). The un and unhcr were marginalized throughout. nato demanded a quick return of refugees after the Russian-brokered peace deal, because ‘the fast return of refugees was seen to crown the success of its mission’; and despite claiming to worry about remaining landmines, unhcr apparently ‘had no choice’ but to go along with this. The speedy repatriation led in turn to an exodus of two hundred thousand Serbs and Roma from Kosovo. Ogata notes some ‘confusion’ and ‘blurring of the lines’ between military and civilian relief operations, but concludes that the situation was exceptional, and co-operation necessary.footnote⁶

Great Lakes

The new framework for dealing with large refugee crises—containment in camps and ‘safe havens’, quickly followed by repatriation—was also applied in the Great Lakes region of Africa, following the Burundian coup and the massacre of up to 800,000 predominantly Tutsi civilians in Rwanda. One million mainly Hutu refugees moved into what was then Zaire in 1994, and unhcr immediately appealed for military help. French troops were there from the start, shortly followed by us Army logistical support—Operation Support Hope—then Swedish and German troops, and the Japanese Self-Defence Force (a minor breakthrough in Japan’s remilitarization, which Ogata celebrates). Setting up and running the camps required the support of Mobutu’s regime, which he gave with a view to strengthening his hand in the region and in international negotiations.footnote⁷ Ogata argued that ‘From the outset it seemed obvious that return was the main solution. The refugee outflow was too massive for the people to be absorbed by neighbouring countries or to be resettled in third countries.’footnote⁸ This policy was now becoming so entrenched that it no longer seemed necessary for unhcr even to ask whether repatriation was in the best interests of the refugees. Sporadic massacres of displaced civilians within Rwanda seem only to have slightly delayed the apparently inevitable return.

As in the Balkans, unhcr tried and failed to cope with the military elements mixed in with a largely civilian refugee population. Fifty thousand Rwandan soldiers and militia had crossed into Zaire with the refugees, and unhcr and the rest of the un were at a loss as to how to deal with them. The refugee camps, which at this scale had an economy of their own, provided funds and a power base for part of the Hutu leadership that had carried out the genocide. Boutros-Ghali was in favour of a un-headed force of ten to twelve thousand to separate armed groups from civilians—Annan, perhaps quicker to adapt to international trends, proposed using an ex-sas private militia: Defence Systems Limited. But the Security Council, divided between French support for the Hutus and us support for the new Rwandan government, could not agree to allocate either troops or money. Relations between Rwanda and Zaire deteriorated further, with the refugee camps being used as political pawns. There were some attacks on the largely Hutu camps within Rwanda’s borders, after the departure of the French troops that had protected them, and Zaire was arming militia in the camps on its side, facilitating attacks across the Rwandan border. While elements of the Zairean government were in favour of repatriation, President Mobutu—more closely aligned to the Hutus—found it useful to keep them in place. Communities near the border that had Rwandan Tutsi origins—the Banyarwanda and Banyamulenge, many of whom were denied citizenship by Zaire—were now vulnerable to attack, and had turned to elements of the Rwandan army for military training and support. In October 1996, Rwandan and Banyamulenge forces attacked many of the Hutu camps, forcing hundreds of thousands to flee. Staggeringly, given that many of them were escaping from Rwandan troops, Ogata broadcast radio messages telling the refugees ‘to consider returning to Rwanda’.footnote⁹

By November 1996, Rwandan forces and the Banyamulenge had joined with Congolese rebel groups to form the Alliance of Democratic Forces for the Liberation of Congo. Aiming ultimately to topple Mobutu’s regime, and so viewing the refugee camps as a local threat, the adfl attacked more of the camps. Half a million refugees were forced to converge in one small volcanic area in Goma, with no more than a few hundred unhcr staff trying to arrange sanitation, food and water. At the worst point, three thousand people were dying every day.footnote¹⁰ Pressure mounted on the un to act, and Canada offered to lead a multinational force. Rwanda and the Congolese rebel forces moved quickly, with likely us connivance, to make sure that an international military presence would not impede their war aims. Rwandan Vice-President Kagame ordered attacks on camps in Mugunga with rockets, mortar and heavy artillery, forcing half a million refugees in the direction of Rwanda. The unsc resolution ordering a deployment was passed the next day, but Kagame had ensured that it would not take effect: the camps it was to protect had been dispersed. He also negotiated with Tanzania to dissolve the camps there, leading to half a million refugees being forced to return, with unhcr providing logistical support in direct contravention of the 1951 Convention. Up to seven hundred thousand refugees had, in addition, fled in the ‘wrong’ direction from Rwanda’s offensive—west into Zaire, ahead of the frontline of the civil war—upsetting the agreement between Albright, Ogata and Kagame that they should all return to Rwanda. unhcr led a ‘search and rescue operation’—paying local volunteers and militia—to find them in the rainforest and ‘help them to return’ (Ogata later describes it as ‘the mop-up of Rwandan refugees’).footnote¹¹ Any façade of voluntary repatriation had been abandoned; unhcr was actively forcing refugees back into a volatile Rwanda. Ogata’s reasons for helping the Tanzanian repatriation presumably apply to the whole Great Lakes operation:

We might have stood aside and condemned the rough handling by the military. We might have disassociated ourselves from the operation. Instead, what we did is compromise, to save what little there was to save.footnote¹²

It is a familiar line of argument, which echoes throughout the twentieth century. In the new millennium, as unhcr becomes yet more integrated with American and British foreign policy, Ogata’s views further lose touch with reality. In Afghanistan, it was apparently ‘only in the aftermath of the terrorist attacks on the United States that the international community turned to help ensure its peace and stability’.footnote¹³ Ogata had left office in 2000. The peace and stability ushered in by Coalition aerial bombardment was after her watch, but she was asked by Koizumi to help with reconstruction. After the us invasion, the flurry of un and ngo activity around Karzai’s regime raised Iranian and Pakistani hopes of offloading their respective shares of the six million Afghan refugees. But as the un struggled to impose even rudimentary security beyond Kabul, it once again turned to nato, who took over the International Security Assistance Force and ensured its remit was expanded by Resolution 1510. As in Africa and the Balkans, a hasty repatriation saw three million refugees return by the end of 2003. Ogata lauds the ‘ceremonial and emotional’ loya jirga and the 2004 constitution as vital steps on the road to normalization. For those who have followed the intensifying conflicts in Afghanistan, her optimism will not prove too infectious.

Modernizing the Convention

The shifts over the past two decades in the way unhcr works are momentous, and have affected the lives of millions of victims of war and persecution. In part, the changes reflect the continuing debasement of the un. If, as Peter Gowan has argued, there was no Golden Age free of us influence, the Communist bloc and the non-aligned movement did provide a check on American control of global institutions.footnote¹⁴ As the Soviet Union collapsed, the first Gulf War showed the new extent of us power over the un. Ogata’s memoirs throw some light on Boutros-Ghali’s attempts to preserve the last shreds of un independence, for example in the nominal separation of us military and un humanitarian operations in Iraq in the early 1990s. In Bosnia, Boutros-Ghali argued for dual-key control of nato bombings, much to their irritation; while he was on holiday, Kofi Annan waived this requirement, allowing the us to step up the offensive without un oversight. America’s un Ambassador Richard Holbrooke noted that Annan ‘won the job’ of Secretary-General on that day; it certainly set the tone for his time in office.footnote¹⁵ In both the Gulf and the Balkans, unhcr felt unable to protest the closure of borders by countries that hosted key us bases. After the first Gulf War, the us and Britain strong-armed the un into accepting and setting up safe havens; Ogata pleads that she had to go along with this. During the Balkan crisis, nato was effectively giving orders to the un and unhcr, and the safe haven policy became a tool for Albright’s experiment in nation building. In Africa, unhcr was caught between American and French designs on the region, crippled by a lack of resources, and by this point fully complicit in a policy of containing refugee movements at any cost. By the time of the invasion of Afghanistan, unhcr was working closely with the military from the outset—the agency had seemingly lost the capacity for independent thought.

The justifications offered for these shifts in policy are far from unique to unhcr; they are billed as attempts to adapt to a new global environment, marked by ethnic and religious conflict and massive flows of refugees and migrants, requiring innovative tools to solve new problems. According to the modernizers, ‘there is no alternative’—this is an adaptation, rather than a derogation of historic rights and duties. Yet there are no ‘humanitarian’ grounds for updating the 1951 Convention, or the 1967 Protocol that turned it into a global instrument. These documents were designed to address situations of mass exodus and misery after ethnic cleansing and war. They are laudably clear on the rights and duties of refugees and states, and they contain explicit provisions for situations like those in Africa, where war criminals are mixed in with civilians. Refoulement, forced return from the country of asylum, is explicitly ruled out, with eventual voluntary repatriation as an aim, and settlement in a safe third country as a last resort (with a recommendation to find methods for sharing that burden equitably). The bulk of the Convention is concerned with specifying the political and economic rights of refugees in the country of asylum. The only real problem is that what began as a self-interested European approach to dealing with the mass displacements of the Second World War now has to operate in an increasingly unequal world, in which Western powers do not want to take responsibility for the results of their military interventions and destructive economic policies.

The doctrine of ‘preventive protection’ is at best a violation of sovereignty that exceeds unhcr’s mandate. It authorizes, with the swagger of a benevolent imperial power, extensive intervention in a country on the ground that it might generate refugees. At worst, and quite typically, it is a tool for a general policy of containment that tries to keep civilians in areas that are becoming increasingly dangerous. ‘Temporary protection’, the other side of this counterfeit coin, is nothing other than a straightforward abandonment of Convention provisions. The need to avoid ‘red tape’ and to reach speedy decisions are given as reasons to deny refugees the full set of rights to which they are entitled. Needless to say, these have both become central planks of the agency’s development since Ogata left office. Former Dutch Prime Minister Ruud Lubbers launched the ‘Convention Plus’ initiative in 2002—just reverse the sign—which further entrenched the strategy of containment. After he had to resign on sexual harassment charges, the reins were handed to former Portuguese Prime Minister António Guterres, an Opus Dei initiate, who continued these reforms.

The official doctrine is now the ‘4Rs’: Repatriation, Reintegration, Rehabilitation and Reconstruction—note the order in which they appear. unhcr continues to say that repatriation should be to a stable country that can uphold human rights, but its practice shows how hollow that proclamation is. Wide-ranging civil war—the Balkans, Great Lakes, Afghanistan, Iraq—has at no stage been regarded as a barrier to speedy repatriation, especially when admitting that there is a civil war might embarrass the us. Reintegration and Rehabilitation consist largely of community-based projects (like Imagine Coexistence in the Balkans) that would be better given back to the development agencies, and Reconstruction is usually handed over to the World Bank, with predictable results for the economy and employment prospects. unhcr not only cleans up after nato’s wars, but also helps to prepare the devastated economy for the invasion of international capital.

Imprisonment or integration?

That the 1951 Convention cited voluntary repatriation as one of its goals was understandable. Those fleeing war and persecution would usually prefer to return to their family home, if order could be restored, houses rebuilt, adequate protection provided and the economy rejuvenated. But now that the right to permanent settlement in a safe country has largely been dismantled, the horrendous conditions engineered by host countries, the military and the un in ‘safe areas’ and refugee camps have become a lever which the West uses to lower the standards for that repatriation. Given a choice of starving in a mountain pass, the Kurds ‘voluntarily’ returned to Iraq; interned in safe areas that were under constant bombardment, former Yugoslavian civilians ‘voluntarily’ repatriated into majority ethnic areas; from the camps and slums in Iran and Pakistan, Afghans also ‘voluntarily’ returned. Misinformation, the lack of alternatives, increasing coercion: the ‘soft power’ of the containment project is buttressed by the inhuman conditions of the camps. In the Balkans and the Middle East, where the un now allows nato to project its power at will, unhcr has helped to redirect millions of human casualties back into the brave new democracies, in the midst of raging civil conflicts. But Africa, as ever, is the real graveyard of the un’s charters. Just as the un had turned a blind eye to the genocide in Rwanda—which it had explicit warning of, and facilitated by withdrawing its troops—unhcr cooperated in the forced return of Rwandan refugees, mainly because Albright ordered it, thereby undermining the central provision of the Convention with apparent impunity. The un itself has estimated that, during 1996–97, two hundred thousand Rwandan refugees were killed in Zaire.footnote¹⁶

Refugee camps and repatriation are the twin pillars of the overall strategy of containment, generously funded by rich countries that wish to keep asylum-seekers from their borders. Conditions in one can be used to argue for funds for the other, but the debate rarely moves on to question both. During a lifetime of sociological research and legal advocacy, mainly in Africa, Barbara Harrell-Bond has documented the full range of rights violations and abuses in refugee camps: she argues forcefully that unhcr’s whole strategy should be recast.footnote¹⁷ The camps are, in the first instance, a legal anomaly. Although they are on the territory of the host country, that country attempts to absolve itself of responsibility by handing control of the camps over to ‘humanitarian’ organizations. That initial abdication of duty creates a legal void that makes it hard to prevent violence, corruption and neglect; however good the intentions of the majority of unhcr workers may be, there is no structural solution to the abuses of a minority. Once they are beyond the rule of law, refugee camps become ‘total institutions’ like prisons and mental hospitals, where the agency workers can exercise their powers unchecked:

The very organization set up to monitor the extent to which refugees enjoyed their human rights had assumed de facto sovereignty over them. Who could monitor the monitor?footnote¹⁸

The process of attracting donor money relies upon the visibility of the refugees, and the possibility of counting them; both are easier when they are held in a confined space.footnote¹⁹ It also encourages a portrayal of them as helpless and dependent: denied freedom of movement, and with little chance of economic self-sufficiency within the camps, this becomes self-fulfilling. Although the country of asylum is usually desperately poor, mortality rates remain higher in the refugee camps. For the host country, the camps become a parallel economy that draws money and skills away from the locality, places strains on the environment and facilitates the spread of disease; and after repatriation, hastily erected but much-needed schools and hospitals are just as swiftly dismantled. Yet if these refugees were afforded freedom of movement and political and economic rights in the country that granted asylum—as mandated by the Convention—they would be likely to become a net benefit to the society and economy. unhcr’s focus should not be on building camps, but on enforcing the Convention provisions upon its signatories: the freedom to move, work and remain in a country of asylum, with the initial costs of integrating refugees shared by those countries that can more easily afford it.

Not camps but integration into the host country is the best way to begin to address the plight of refugees. As the authors of the Convention seemed to appreciate, repatriation cannot properly be called voluntary unless the person making that choice is currently in possession of his or her civil rights; anything less should be viewed as a forced return. unhcr’s turn away from the political rights enshrined in the Convention—towards ‘emergency relief’, encampment and repatriation—has condemned an uncountable number of refugees to death and misery in the camps and ‘safe havens’. Many are there as a result of wars the West has prosecuted or armed, and others as a result of economic collapse and civil war; in both cases, Europe and the United States have unequivocally supported the strategy of containment.

Sadako Ogata’s initial challenge was: should the unhcr dissociate itself from military operations, refusing to provide cover in situations where refugees’ rights are being undermined? The answer has to be yes. If the agency were to accompany or transport vulnerable refugees to a genuinely safe country in accordance with the Convention—in defiance of the wishes of border guards and donor countries, whilst commanding media attention—would they be fired upon? We will of course never find out; its unstable funding structure keeps the agency on too short a leash. Dependent on its donors, unhcr also lacks the political determination it would need to enforce the Convention’s provisions upon its signatories. As with the Declaration on Human Rights, the lofty sentiments of an international treaty—one that has huge popular support—remain crippled by the lack of an independent and effective agency capable of enforcing them. The us and Europe are doing everything possible to keep it that way.