Professor Barry Stein
Department of Political Science
Michigan State University

United States Agency for International Development
Promoting Democracy, Human Rights, and Reintegration in Post-conflict Societies
30-31 October 1997
Washington, D.C.

second draft
14 October 1997


This paper discusses the challenge of contemporary repatriation which has been transformed from relatively small voluntary repatriations which are safe and follow fundamental changes to large returns to troubled countries where repatriation is often violent, compelled and premature. Because the nature of return has a direct effect on the nature of the reintegration program, the paper discuss issues such as: protracted complex emergencies; reintegration of returnees and restoration of civil society; coordination and closure of assistance; and, rehabilitation, development, and funding shortfalls.

This decade has seen a great variety of refugee repatriation and return. Although virtually all of the individual numbers and totals are suspect, they do reveal a broad pattern, of the fourteen million refugees who have returned home in the 1990s, almost ninety per cent are spontaneous returns, refugees making their own decision to go home without waiting for significant international assistance.(1) In addition, they are not returning to post-conflict societies, they are returning during conflict to societies where there is no peace to keep or it is a fragile peace at best.

Contemporary voluntary repatriations are unlike almost any that have occurred before. In most cases, the peace is fragile, security is tenuous, and the economy and infrastructure of the homeland are devastated. Most returns involve hundreds of thousands, even millions, of refugees returning swiftly and irregularly to ravaged homelands. Almost all of the returns are outside, or only marginally aided by, the system of international repatriation assistance. Mixed in with these movements are equal or greater numbers of internally displaced persons in great need.

Most voluntary repatriations occur amidst conflict, without a decisive political event such as national independence, without change in the regime or conditions that originally caused flight, with only a lull in the fighting or a shifting around of the contending parties. Countless refugees return home in the face of continued risk, frequently without any amnesty, without a repatriation agreement or program, without 'permission' from the authorities in either the country of asylum or of origin, without international knowledge or assistance, and without an end to the conflict that caused the exodus. Many refugees return to regions controlled by parties to the conflict other than their national government.

Refugees are the main decision-makers in voluntary repatriation, they determine how they will move and the conditions of reception. Refugees apply their own criteria to their situation in exile and to conditions in their homeland, and will return home if it is safe and better by their standards. Refugees strive for an outcome that achieves relative security and some small degree of control over their lives.

Other forces, however, particularly in the country of asylum, increasingly are trying to influence refugee decision-making and limit its voluntary character through pressure, harassment and direct violence. Although refugee decision-making is under unprecedented pressure, refugees retain a modicum of choice. In 1996, although civil war in Zaire permitted or forced a half-million Rwandan refugees to return home in the largest and swiftest repatriation in memory, several hundred thousand other refugees made the decision to flee further away from their homeland.

It is a paradox of the 1990s that we speak a perverse truth when we refer to a number of international, governmental, and nongovernmental humanitarian and relief agencies as "refugee agencies." These refugee agencies, such as the United Nations High Commissioner for Refugees (UNHCR), the United States Agency for International Development (USAID), and the International Rescue Committee (IRC), are being forced by circumstances to act as if they themselves were refugees. The violence of thugs, extreme nationalists, armed refugees, governments, and other parties to the conflict, the inaction of the international community, the spreading global threat to asylum, the weakness and collapse of numerous states, and fragile incompletely implemented peace accords, increasingly mean that those who provide aid are under attack, inadequately protected, and forced to choose between unsatisfactory and even tragic and terrible choices. A number of current refugee returns are forcing refugee agencies to choose the "lesser evil" and the "least worse" course of action. Refugee agencies do not have the luxury of an exit strategy. Even if the return is involuntary, denying assistance to the victims is hardly an option.

Definition of Terms

Some of the terms in my paper's title--repatriation, return and refoulement--need some clarification.(2)

Voluntary repatriation refers to two things, the refugee goes home voluntarily and there is a restoration of the bond between citizen and fatherland. Return is a better term for most problematic situations as it simply notes the fact of going home.

Refoulement refers to the expulsion or return of a refugee to a place where his life or freedom would be threatened. Refoulement, which is prohibited by Article 33 of the Refugee Convention, is increasingly being resorted to by asylum countries either directly, as in the US interdiction and return of Haitians, or more frequently indirectly as the global threat to asylum leaves refugees with little choice other than going home. During summer 1997, "UNHCR has had to protest three major, blatant acts of refoulement" (Vieira de Mello, 1997).

Asylum is not part of my title but it is essential for voluntary repatriation. Asylum is sanctuary, safety and security. Asylum protects refugees while they heal, while their homeland changes or at least conditions moderate, while they make decisions about voluntarily going home. The push-pull model of migration(3) is very old, however, it is useful in describing a major change in refugee decision-making in the 1990s. While the pull from the country of origin often remains unchanged, the push by the country of asylum has greatly increased. Thus, the role of the country of asylum in the process of voluntary repatriation has greatly increased. The global threat to asylum, and therefore to voluntary repatriation, is diverse sometimes caused by the country of asylum's security or economic concerns, other times by fragmentation or danger in the country of asylum.

Premature repatriation is a result of the lack of asylum. It is a repeat of the refugee experience. Premature repatriation comes when both the country of origin and the refugees are not ready for it. Conditions at home have not changed sufficiently to pull the refugees home. Neither the refugees nor their homeland are reconciled or ready for the return. Premature returnees are pushed out by threats, attack, and expulsion rather than pulled by peace and safety at home. In a flagrant abuse of basic humanitarian standards, a country of asylum refoules the refugees or civil conflict, danger, and fragmentation in the asylum forces the unreconciled refugees to flee homewards.

It is essential that we maintain an image of truly voluntary repatriation. Voluntary repatriation requires asylum, it respects the refugees and allows them to make unpressured decisions. Today voluntary repatriation is becoming rare. UNHCR, in a series of Conclusions of its Executive Committee, has outlined the proper conditions for an ideal repatriation.

fundamental change of circumstances, removal of the causes of refugee movements.

voluntary nature of the decision to return, freely expressed wish.

tripartite agreements between origin, host, and UNHCR to provide formal guarantees for the safety of returning refugees.

return in dignity under conditions of absolute safety.

These basic principles of protection and voluntary repatriation need to be recalled precisely because they are under great threat and are often violated or ignored.

Today we talk of premature repatriation, rescue repatriation(4) and violent repatriation as well as expulsion, deportation and refoulement. High Commissioner for Refugees Ogata (1997a) recently noted that: "we cannot ignore the fact that the voluntary nature of repatriation is increasingly being undermined by a mounting number of forcible returns in situations which are far from safe."

Many repatriation programs are intertwined with peace accords and multi-dimensional peace-building efforts. Compelling refugees to "repatriate too early" may give relief to the asylum countries through "a dangerous shifting of the burden back to the country of origin. Premature repatriation puts refugees at risk and may jeopardize a successful transition from war to peace" (Ogata, 1997b).

The Transformation of Repatriation

Repatriation has been transformed in the 1990s in both positive and disturbing ways. The most important positive change has been in the scale of repatriation; 14 million returns between 1990 and 1996 in contrast with less than 4 million returns from 1975 through 1989. However, fully half of the 14 million returns have been to very troubled countries such as Afghanistan, Angola, Bosnia-Herzegovina, Burundi, Iraq, Myanmar, Rwanda, and Somalia. And many of the other returns were to states that are now stable and peaceful but were on the knife edge of conflict and peace at the time of repatriation.

Repatriation today often is violent, compelled, and premature. Its hallmarks are the global threat to asylum and the militarization of repatriation. Repatriation has become a battleground:

Rwandan refugees who were been denied a durable solution in exile formed an army in order to fight their way home.(5)

Rwandan refugees in Zaire and Tanzania militarized their camps and violently prevented voluntary repatriation of those they held hostage.

Salvadoran and Guatemalan refugees forced their way home with political repatriations accompanied by international witnesses and the media.

In Bosnia-Herzegovina and Croatia, groups have marshalled militias and thugs to prevent the repatriation of minorities to their areas.

In Abkhazia, Nagorno-Karabakh, Bhutan, and elsewhere repatriation is violently blocked.

In this decade the search for durable solutions to refugee problems has undergone a transformation from a "reactive, exile-oriented and refugee specific approach" to a "proactive, homeland-oriented and holistic approach" (UNHCR, 1995). Initially, throughout most of the cold war, the pace of voluntary repatriation was slow and reactive with the timing largely determined by the refugees themselves. The ideal repatriation was voluntary, safe, and followed a fundamental change in the country of origin. A long exile was expected and refugees received protection and assistance "until the day they chose to return" (UNHCR, 1995).

A "homeland-orientation" is the equivalent of a repatriation-orientation. The shift towards repatriation is due to several factors including the end of the cold war and the lack of durable solutions other than repatriation. The end of the cold war created opportunities for peace and repatriation in Afghanistan, Angola, Cambodia, Central America, Ethiopia and Mozambique, although some of these disputes proved to have an independent life of their own. Humanitarian intervention in the country of origin, which was not feasible during the cold war, has attempted to improve conditions in Cambodia, Bosnia, Haiti, Iraq, Liberia, Rwanda, and Somalia, Rwanda.

Repatriation not only has become the preferred durable solution, it is the only available durable solution. Less than one per cent of the world's refugees are resettled in third countries and almost none of the countries of asylum are prepared to offer permanent status to their refugees. By default, if the number of refugees is to be reduced it will be by means of repatriation.

Proactive policies focus on the country of origin rather than the conditions of exile. Instead of passively waiting for conditions to change, refugee agencies "must work actively to create conditions conducive to their safe return" (Ogata 1995). UNHCR's Executive Committee (UNHCR, 1995b), while retaining its prior commitment to voluntary safe return, now emphasizes:

"the right of all persons to return to their country;"

"the prime responsibility of countries of origin" to establish conditions for safe and dignified return;

"the obligation of all States to accept the return of their nationals;" and,

"calls on all States to promote conditions conducive to the return of refugees and to support their sustainable reintegration."

Repatriation has been connected to international peacekeeping efforts. In northern Iraq, Liberia, Cambodia, Somalia, Haiti, Rwanda, Bosnia, Zaire, and Albania, United Nations or other multilateral peacekeeping forces have responded to threats to international peace and security, often with the objective of facilitating the voluntary, orderly repatriation of refugees or of protecting returnees.

Refugee returns are linked to much larger processes of peace with repatriation and reintegration viewed as a measure of successful peace-building. In Cambodia and Mozambique full refugee repatriation prior to the elections was crucial for the legitimacy of the new governments. In Central America refugee repatriation and reintegration was a principal part of the regional peace process. In Bosnia many view return of refugees to communities where their group is a minority as a litmus test for the entire peace process.

Premature repatriation often leads to militarized repatriation. Imposed returns mean unreconciled refugees return to an unsettled and conflicted homeland. There is an increased need for military protection of refugees from home elements not ready for their return, for military protection against refugees not ready to accept their place in their homeland, and for military protection of assistance programs and aid workers.

Donor countries and host countries share a concern with cost and speed. Both sets of countries want rapid repatriation in order to be rid of the problem and to lower their costs. Return is keyed not by a homeland-orientation, in terms of creation of conducive conditions in the homeland, but by asylum concerns of a protracted massive refugee burden.

The global threat to asylum has been increasing for years. The threat includes denial of access to territory, rejections at the border, and legislative restrictions. Safety during asylum is jeopardized by attacks on refugee camps, the militarization of camps, violence against vulnerable refugees, forced recruitment of refugees, abusive detention, intimidation both for and against return, and forcible returns (Ogata, 1997a). Confronted with massive influxes asylum states fear security problems as well as the economic and social burden.

Major Issues and Problems

The nature of the return has a direct effect on the nature of the reintegration program. Voluntary repatriation of 1.7 million refugees to Mozambique(6)

in the context of a peace accord was a prelude to fair elections and a remarkably peaceful and successful process of reconciliation, reconstruction and development. The mixed return, amidst rebellion and violence, of 1.3 million refugees from Zaire and Tanzania to Rwanda has been a prelude to renewed guerrilla activity and insecurity in northern Rwanda. The forced return of Rohingya refugees from Bangladesh to Myanmar has been followed by renewed persecutions and another exodus.

Repatriation and reintegration assistance has to be sensitive to the types and conditions of return. There are several clusters of repatriation issues and problems which present major challenges to the refugee assistance and policy communities and which could mar hopes for the future. These issues are: protracted complex emergencies; reintegration of returnees and restoration of civil society; coordination and closure of assistance; and, rehabilitation, development, and funding shortfalls.

Protracted Complex Emergencies

Not all refugee situations are part of complex emergencies. However, virtually all complex emergencies involve refugees or internally displaced persons. Sometimes the refugee/displaced person component is not the dominant portion of the overall complex emergency, other times, the refugee and displaced persons problem is the most important and visible aspect of the emergency.

Complex emergencies involve protracted internal conflicts of unusual violence and cruelty; guerrilla conflicts without a clear front line; structural deterioration of the framework of societies including the inability of a nation to feed itself and the collapse of the institutions of conflict resolution; failed development policies; and, even failed states. In the resulting emergencies, the role of international assistance is huge and decisive.

Refugee flows and the plight of displaced persons are often of great importance as a factor justifying United Nations involvement in a complex emergency. However, whether the refugee problems are a ancillary or a major issue, they are rarely treated as the central issue. The primary concern is to end the conflict and engage in post-conflict peace-building.

The failed states, weak governments, guerrilla factions, warlords, and combatant authorities lack the capacity to cope with the disasters they have caused, however, they are able to deliberately deny humanitarian access to vulnerable populations and to make humanitarian and relief workers direct targets of violence. As a result, the international community has increasingly resorted to multifunctional peacekeeping operations to get assistance to civilian victims of complex emergencies. International forces are deployed to facilitate and protect humanitarian operations while attempting to remain neutral and impartial between the warring parties.

Complex emergencies lack clear stages or endings. What looks like the end of a crisis or the start of progress may prove to be false. In many cases, multiple cease fires and peace agreements have been signed but the signatories lacked the institutional and governmental capacity to follow through on their commitments and deliver peace.

Comprehensive peace settlements that attempt to resolve complex emergencies often include refugee repatriation and a role for refugee agencies as part of a much larger package. Annex 7 of the Dayton Peace Accords gives UNHCR responsibility for the repatriation of refugees and the displaced but its programs are heavily dependent on the policies and actions of governments and other international agencies. As the international pullout from Somalia indicated, other parties may suffer a failure of political will, may limit mission goals so that humanitarian and nation-building activities receive insufficient protection or time to be accomplished, may fail to fund rehabilitation or development activities, or engage in other half-measures that contribute to the failure of the comprehensive package. Humanitarian activities may be endangered or curtailed, leaving returnees and other casualties to face renewed threats.

Reintegration of Returnees and Restoration of Civil Society

Voluntary repatriation of refugees to their homeland is a sign that safety and control over one's own life has the possibility of being restored, but repatriation is only a beginning, it does not necessarily mean that the bond of trust and loyalty has been restored between citizen and state. People who are physically home but are not participating in the economic and political life of their country are still uprooted persons. In many cases they may be back in their homeland but far from their own communities. The danger exists that repatriation alone is a relocation that converts refugees into internally displaced persons.

Reintegration of returning refugees is a complex political, economic, social and cultural process that goes beyond a simple physical reinsertion of refugees in their home communities. Violence and conflict and the passage of time have an effect on individuals and societies. Refugees may undergo major cultural and social transformations, while conflict and politics transform the homeland. Many of those who stayed behind may have been on the other side of the conflict. Others who stayed, have sufferings and experiences not shared by the repatriates.

Internal conflict, in particular, can polarize and politicize even the most friendly and benign activities thus contributing to social disintegration and the violent resolution of disputes. When civil institutions are weakened and attacked and security interests are paramount and pervasive, participation in civil society can be dangerous. Populations are uprooted when violence replaces consensus-building, when normal, legitimate concerns and needs are labeled subversive.

Most of the money spent on international assistance for refugee reintegration is concerned with improving the economic status of the returnees and their communities. This focus is understandable given the devastation caused by civil conflicts and the significant groups of needy and vulnerable people. However, the long-term obstacles to reintegration are only partially rooted in economics. The roots of the original refugee flows are deep and entwined around resentments of power and privilege, mutual suspicions, and fundamental economic and social injustices. Reintegration will be precarious without the re-creation of political or humanitarian space and a functioning civil society at the local, regional and national levels.

However, we need to be cautious about the impact of resources whether for civil society or development. Many ethnic or political leaders have extreme agendas which they will not abandon in order to receive conditional aid. Taliban, Bosnian Serbs, Rwandan Interhamwe, Abkhazia, and other groups consider their policies more important than international assistance.

Coordination and Phase Out of International Assistance

Although holistic reintegration goals are sometimes given rhetorical voice in repatriation planning, it is rare to find solid attempts to achieve them. The emerging conceptualization of reintegration as a part of a larger, more complex process is not accompanied by any internationally recognized institutional framework for reintegration or reconstruction assistance. However, recent comprehensive peace settlements in Central America, Mozambique, Cambodia have provided for international assistance programs that were accompanied by a significantly higher level of national political will and international commitment and funding to the tasks of reintegrating uprooted populations and rebuilding societies.

Holistic reintegration assistance is not a one agency task, indeed, rather than being a single task, reintegration is intertwined with conflict resolution and post-conflict development. No international agency can plan and direct, initiate and complete, the full range of activities needed to reintegrate returnees. UNHCR can begin the process of return and reintegration, however, it is not a development agency and thus cannot complete the job. The United Nations Development Program (UNDP) is a development agency, but it is not likely to be present at the creation of a reintegration program that it may be called upon to complete. Further complicating the picture is the role of the refugees themselves. Most repatriation is refugee-initiated rather than agency-initiated. Most repatriation is to societies amidst conflict, even to regions controlled by non-governmental forces. Programs and plans made for such returns cannot know what end and closure conditions they will have to confront.

If reintegration assistance is part of a comprehensive peace settlement or a complex emergency there may be dozens of UN actors from its political, military and humanitarian divisions both at headquarters and in the field, plus ad hoc task forces, working groups, emergency teams, and other special relationships will form within and between many agencies and offices (UN 1994c). The complexity of the UN response is mirrored by non-unitary structures in the primary bilateral donors. And recent programs have involved numerous regional international organizations and more than 100 international NGOs.

Of all the organizations and agencies involved, none necessarily stands out from the others as the leader in an emergency or a repatriation. Each of the separate entities has its own mandate, governing boards independent fundraising and resources. The existing system does not view a complex emergency as a whole problem, it is unable to offer a coherent and comprehensive approach and solution.

UNHCR can participate in a refugee situation without asking for permission. Refugee camps may be assisted for decades. However, if refugees voluntarily repatriate to their country of origin, UNHCR will run into mandate limitations. Although UNHCR has an abiding concern that returnees be firmly reintegrated into their societies, UNHCR assistance and involvement is limited to their return and an initial, albeit lengthening, arrival and settling in period.

In providing segmented assistance during a repatriation many agencies have to deal with a situation in which they will launch a program or process even though they are unlikely to be present at the finish. Pressed by other demands on their resources, many agencies need to bring some of their assistance programs to a conclusion. The decision to phase out an operation is normally made independently, based on an agency's mandate and on achieving certain criteria--such as repatriation or attaining food self-sufficiency. A problem for agencies seeking to phase out their involvement is the need to find other agencies to take over their unfinished work. Lacking coordination and a comprehensive approach at the beginning of a repatriation, agencies find their necessary departure may be dependent on and entangled with the operations and decisions of other parties. Experience has shown that waiting to make a handover to another party can be open-ended.--"As we try to phase down in Mozambique, the challenge is to ensure that others have a stake in the reintegration process and will continue when we leave" (Ogata 1995).

Inter-agency collaboration within the UN system tends to be weak. Childers and Urquhart (1994) concluded that for each fresh emergency: "The 1991 reform did not overcome the separatism and built-in competition that is so pervasive in the UN system even in the face of the human desperation of emergencies." Now we have the 1997 reforms, which again promise a major restructuring to deal more effectively with complex humanitarian emergencies.

Rehabilitation, Development and Severe Funding Shortfalls

There is a need to arrange better funding for repatriation activities. UNHCR and the UN system rely on voluntary contributions to fund their repatriation and rehabilitation activities. This means that the availability of resources is unpredictable and often inadequate, and repatriation opportunities may be lost. Persistent funding difficulties, in terms of funding appeals which are undersubscribed, have been greatest for reintegration programmes in countries affected by armed conflict, such as Somalia, Tajikistan, Liberia, Rwanda, Georgia, Azerbaijan and Afghanistan. However, even the peaceful return of more than 100,000 ethnic Tuaregs to Mali has gone unsupported receiving a mere $1.3 million towards an appeal for a $17.6 million reintegration program (UNHCR, 1997x). Overall, "repatriation programmes in Africa . . . are dramatically underfunded" (Vieira de Mello, 1997).

At a time when humanitarian needs are expanding, a third window for financing emergency rehabilitation activities is needed. Funding windows already exist for emergencies and development, but resources are lacking for post-conflict rehabilitation. The distinction between rehabilitation and development is important one because a great deal contemporary repatriation is not to post-conflict societies ready to engage in development efforts.

Rehabilitation is assistance to restore victims to self-sufficiency and viability, and to reduce their vulnerability. While rehabilitation activities should be consistent with development objectives, they are implemented only to the extent that they satisfy specific needs imposed by the crisis.

Refugee return often occurs amidst conflict, very early in the process of reconciliation and peacemaking and often before any resumption of development is possible. Attempting to tie reintegration assistance to national development may be a laudable goal but it is most likely to be an unrealistic goal, at least at first. At the time of refugee return the most pressing needs are for rehabilitation and reconstruction. Development is a very distant goal.

Permanent emergencies and protracted complex political emergencies are much more than a transitory delay in the development process. Two, three and four decades of internal conflict and disaster in Afghanistan, Angola, Sudan, and elsewhere reflect the development of normalized, institutionalized, systematic, long-term disasters that have shattered and reversed the development process.

Countries in the throes of complex emergencies are terrible candidates for development or relief-linked development. Past development failures have contributed to the onset of a complex emergency. At one time these devastated countries had "a functioning civil and social administration, a transport system and a modest, but real, economy" (Duffield, 1994). Now, those advantages have been lost and their absorptive capacity for receiving aid is very low, and without peace and political stability their likelihood of development is meager.

A development orientation does not serve the real needs of the victims of a complex emergency. Their needs are more direct and immediate than development, more modest and attainable. A certain modesty in the face of violence and intractable problems is advisable. It is very rare for lives torn apart by violence, displacement and disaster to be rebuilt with little cost. Rehabilitation and reconstruction with a focus on food security would set more proximate and realistic goals for the international community when coping with complex emergencies.

When refugee assistance moves from humanitarian aid toward development aid there is a shift in leverage in favor of the donors. Humanitarian aid has a compelling dramatic immediacy about it that makes it difficult for donors to stand on the sidelines. Returnee assistance, on the other hand, is after the emergency, conditions have stabilized, the danger is past, and whatever the compelling arguments in favor of development aid, the drama and urgency are missing. Donors asked to fund unsatisfactory development projects--developed from government or agency "wish lists" without a suitable institutional framework for implementation in societies with disrupted development processes--can sit on their purses. Increasingly, donors are resorting to "conditionality" regarding development aid to homelands, demanding progress on human rights, judicial and prison reform, demilitarization, democratization, and other issues, before funding projects.

Acceptance of rehabilitation as the strategic goal of humanitarian and reintegration assistance, particularly during complex emergencies, is based on the potential of rehabilitation assistance to improve conditions for victims of conflict. In designing a rehabilitation strategy a number of constraints and weaknesses present during a complex emergency need to be confronted. These include: limited access to and weak knowledge of the field; limited capacity and experience to design, plan and implement assistance at the local and national levels; fragmented assistance efforts and authority; substantial population dislocations; and, widely varying conditions throughout the society. This leads to a rehabilitation strategy that responds to local conditions and opportunities to provide assistance; is simple and focused on timely, quick-impact assistance; that is flexible and avoids plans and designs that cannot keep up with fluid conditions; and that monitors and reacts to, supports, enables, and facilitates the initiatives, decisions and coping strategies of the beneficiaries.


Within the wide range of conditions that surround refugee returns, there are several worlds of repatriation. Much repatriation is voluntary to changed societies, takes place in the context of a peace agreement that is being negotiated, signed or implemented, and has substantial international support, such as in Cambodia, Mozambique, and perhaps Liberia, ranging from money and assistance to international peacekeepers to committed diplomatic support. Other repatriations are premature and compelled to conflicted homelands, lack a peace agreement or have an agreement that the parties are unable or unwilling to implement, and lack sufficient international backing. The lessons from the first world of successful repatriation are difficult to apply in the second world of involuntary repatriation because of the presence of violence and the international communities reaction to violence.

Never again. After the Holocaust, after the Rwandan genocide, after the militarization of refugee camps by mass murderers; the international community signs conventions, passes resolutions, learns the lessons of the past, and pledges "never again." Lessons are easy to learn but difficult to apply. Especially if the parties knew they were ducking a right but tough action the first time.

Although there are repeated references to compassion fatigue, the reality is that there is great kindness, benevolence and charity towards refugees and other victims. However, humanitarian action cannot substitute for political action, in particular a willingness to confront violence. Many political leaders are focussed on keeping their military interventions casualty-free, in Bismarck's words they "will not sacrifice a single Pomeranian grenadier," when they support humanitarian rather than national interest goals. Indeed, recent bruises to American soldiers were enough to lead some to call for withdrawal from Bosnia.

There may be times when the international community should be grateful that a 'durable solution' of repatriation has not been achieved. Many premature returns represent a failure by the international community to provide for and protect refugees. One needs to question why adequately protected and nourished refugees would return home during conflict conditions to a country ruled by the government that originally caused the flight. A lack of a 'solution', non-return, may be a positive reflection on the attitudes and efforts of host countries, on the support of donors, on the protection by international agencies, and on the voluntary nature of return.

The international community has endorsed repatriation as "the most preferred solution, where feasible." In practice this has meant it is virtually the only solution available and repatriation's feasibility is rarely examined. It may be time to question whether some refugees are ever going to be able to go home, time to abandon the idea of repatriation in all cases. Recently the idea of partition has returned as an option in international discussions about Bosnia, Nagorno-Karabakh, Abkhazia.

Lastly, there is voluntary repatriation during conflict. Confronted with the harsh reality of temporary asylum and no durable solution, many refugees explore the possibility of going home. In the absence of coercion, refugee-induced repatriation is a self-regulating process. Refugees will voluntarily repatriate if and when they believe they will receive sufficient protection. Protection, security, and more control over one's fate are the key variables in repatriation during conflict. Protection is a perceived political "space" that provides refugees not only relative physical security, but also material and moral support. The space may be so narrow that only single refugees can return, or it may be understood so broadly as to permit a collective return.

There is a need to actively assist voluntary repatriation during conflict. These returns reflect the refugee's own needs, standards, and judgement of their situation, that they may have a better chance of survival amidst conflict at home than amidst the dangers--sanitation, food, disease, attack--of camps and asylum.

"Lessons" for Discussion

Peace agreements are a poor indicator that the post-conflict stage has begun. Peace first is not a requirement to promote reintegration and rehabilitation. In fragile and partial peace there are opporunities to provide assistance.

All aid is local. In a complex emergency marked by an absence of cental government control over large areas of the country it is important to think in terms of pockets of return and safety and to promote the development of civil society in those areas. These more peaceful pockets represent a localized post-conflict area within a country at war. It is useful not to be bound by sovereignty and to assist whichever party controls the area.

Assistance needs to be simple, flexible, and reactive. Refugees are returning spontaneously, perhaps prematurely. There is no national development plan and a lack of capacity to implement one if it existed. Simple assistance reflects a belief that you cannot design and plan for the many choices open to the people amidst conflict and that complexity would overwhelm the capacity of available systems. Reactive reflects a beleif that it is better to assist returnees in what they choose to do rather than plan the wrong thing and then try to cajole them to fit your design.


Childers, Erskine with Urquhart, Brian

1994 Renewing the United Nations System. Uppsala: Dag Hammarskjold Foundation.

Duffield, Mark

1994 "Complex Emergencies and the Crisis of Developmentalism," IDS Bulletin, v. 24 (4), October.

Ogata, Sadako

1995 "Opening statement by the United Nations High Commissioner for Refugees," in UNHCR, "Report of the Forty-sixth Session of the Executive Committee of the High Commissioner's Programme," A/AC.96/860, 23 Oct.


1997a "Statement to the 53rd Session of the United Nations Commission on Human Rights," Geneva, 1 April.


1997b "Opening Statement at the Eight Meeting of the Standing Committee of the Executive Committee of the High Commissioner's Programme," Geneva, 24 June. []

UNHCR [Crisp, Jeff]

1995 The State of the World's Refugees: In Search of Solutions. United Nations High Commissioner for Refugees. Oxford: Oxford University Press.

U.S. Committee for Refugees

1997 World Refugee Survey 1997. Washington, D.C.: Immigration and Refugee Services of America.

United Nations High Commissioner for Refugees

1997x "UNHCR seeks $17.6 million for its Mali and Niger Repatriation Programmes," Press Release, 30 May.

Vieira de Mello, Sergio

1997 "Opening Remarks" by UN Assistant High Commissioner for Refugees, EXCOM Standing Committee, Geneva, 9 September. []

1. Waiting is the key word. There is a time lag between improved conditions, a peace accord, and peace. Formal organized repatriation assistance programs wait for a firm peace. The refugees do not wait, they apply their own criteria and go home on their own. In 1993, some 625,000 Mozambican refugees returned spontaneously out of a total return of 650,000. They returned in the period between the 1992 peace accord and the 1994 election. In 1997, at least 40,000 Liberian refugees returned in the lead up to the elections and before the launching of a major voluntary repatriation program.

2. Some additional terms from UNHCR's Refworld:

Involuntary Repatriation: repatriation to the country of origin induced by the receiving country by creating circumstances which do not leave any alternative but the return.

Voluntary Repatriation: return to the country of origin on the basis of the freely expressed willingness of the refugees.

Spontaneous Return: voluntary repatriation to the country of origin of the refugees on their own without assistance by UNHCR or the international community.

Principle of Non-Refoulement: Principle of international law which states that no state shall return a refugee in any manner to a country where his life or freedom may be endangered on account of his race, religion, nationality, membership of a particular social group or political opinion.

3. Push factors encourage departure from a place of residence; pull factors draw one toward a destination. The decision to migrate is determined by push factors at home--bad weather, unemployment, family problems--which encourage migration and by pull factors in the destination--sunshine, good jobs, streets paved with gold--which attract the migrant. Refugees flee home because of an overwhelming push, persecution, with almost no pull factors, often they do not even have a destination in mind.

4. Referring to UNHCR's repatriation of Rwandan refugees from Zaire High Commissioner Ogata (1997b) stated: "this was a life-saving exercise under extreme circumstances, to save people from near-certain death, whether by starvation, disease, or attack." However, these were the same refugees who had fled further away from Rwanda rather than return in late 1996. The "attack" was by rebel forces created, aided, and directed by the Rwandan government.

5. Concern that the peace accords would lead to the repatriation of hundreds of thousands of Tutsi refugees was a contributing factor in the Rwandan government's decision to launch a campaign of genocide.

6. On December 31, 1996 UNHCR invoked the "cessation clause" for Mozambican refugees because the "circumstances that originally forced refugees to flee Mozambique had ceased to exist, due to fundamental changes in Mozambique" (U. S. Committee for Refugees, 1997:82).