On the Record

New Challenges for Refugee Policy: Internally Displaced Persons

Roberta Cohen

Roberta Cohen is Co-Director of the Brookings Institution Project on Internal Displacement. This presentation was based on Masses in Flight: The Global Challenge of Internal Displacement, by Roberta Cohen and Francis M. Deng, Brookings Institution, 1998.

When we look at uprooted populations in Africa, we find that those who are forcibly displaced internally outnumber refugees two to one. There are approximately 9 to 10 million Africans (the OAU gives much higher figures) who have fled or been forcibly displaced from their homes for the same reasons as refugees: armed conflict, ethnic strife, human rights violations, but who remain in their own countries.

Because they have not crossed a border, they have no assured source of protection and assistance. The international refugee regime, a complex network of institutions, laws and agreements set up after the second world war, had as its aim the protection of persons forced to seek asylum on the territory of a foreign state. The UN High Commissioner for Refugees (UNHCR) and the UN convention on refugees of 1951 sought to protect persons outside their country of origin. Excluded from this arrangement, in keeping with traditional notions of sovereignty, were those forced from their homes who remained under the jurisdiction of their own governments. It was expected that their governments would provide for their well-being and security. In fact, when governments failed to do so, or deliberately subjected their displaced populations to starvation and other abuses, the international community basically stood by. Even as late as 1988, international organizations and NGOs watched while a quarter of a million people died in the Sudan for lack of food and emergency supplies.

This gap in the international protection system has become more evident in recent years. For one, the increase in internal conflicts produced growing numbers of internally displaced persons. The numbers rose from 1.2 million in 11 countries in 1982 to more than 20 million by 1995. In Africa alone in 1996, 14 out of its 53 countries were embroiled in civil wars. Second, the CNN factor brought the issue to the fore. Watching starving Sudanese or Ethiopians on TV screens led to outpourings of international aid to persons displaced in their own countries. Most importantly, the end of the cold war transformed the issue into a humanitarian one. When the superpowers were engaged in proxy wars, as in Angola and Mozambique, it was pure geopolitics; no attention was paid to IDPs. It was only when these struggles began to wane that the plight of the displaced came into view and was recognized as requiring international humanitarian action. The end of the cold war also facilitated access to many countries reinforced by the idea that events taking place within a country are a legitimate subject of international concern. The human rights movement had long championed this view. But now humanitarian organizations began to insist that when governments deny access to populations at risk, the international community should find ways to become involved. Thus, in 1989 and 1990, in the Sudan, the UN used hard diplomatic bargaining to persuade the government and the rebel forces to accept Operation Lifeline Sudan for displaced populations. In Somalia and Rwanda, the Security Council went to far as to authorize the use of force to facilitate the delivery of relief to IDPs and other affected populations.

The issue of internal displacement became more prominent too because of the realization that peace and reconstruction in war-torn societies depended on the effective reintegration of displaced persons. Many of the countries devastated by civil war—such as Mozambique, Angola and Liberia—had anywhere from one-third to three-quarters of their population forcibly uprooted. It thus became impossible to talk about reconstruction and development without taking into account the return and reintegration of IDPs.

Increased attention to the problem, however, has not made it more tractable. Unlike wars between nations, civil wars, particularly when they divide countries along racial, ethnic, linguistic, or religious lines, do not resolve themselves easily or neatly. When states are monopolized by or identified with one ethnic group to the exclusion or marginalization of others, displaced persons easily fall into a vacuum of responsibility in the state. In Burundi, for example, the Tutsi-dominated army does not provide protection for internally displaced Hutu. In the Sudan, the Arab, Muslim government of the north is directly engaged in persecuting the non-Arab, non-Muslim population of the south, 4 million of whom have become displaced. Consider that within Africa, there are more than a thousand ethnic groups divided among artificially created states. With democratic and pluralistic governments few and far between, most internally displaced persons can not rely upon their government. It should be noted that even when the governments allow in international aid to their displaced populations, they often remain suspicious of such efforts. They fear that the assistance will undermine their own authority. For this reason, governments are often unwilling to allow humanitarian organizations to mount cross-border operations or to negotiate with the rebel forces. This has been the case in Angola, the Democratic Republic of the Congo and Sierra Leone.

Internal conflicts caused by warlordism are equally problematic. In failed states, belligerents or nonstate actors accept almost no ground rules of battle. The main target of the rebel force in Sierra Leone are civilians. Humanitarian assistance is plundered and perceived as a weapon, access is closed off, and no respect shown for the personnel and equipment of humanitarian agencies.

Access may be further complicated by the different manifestations of internal displacement. In some countries, IDPs do not congregate in accessible camps or settlements but disperse so as to avoid identification. Many displaced Hutus in Burundi, for example, hide in forests, making it difficult to reach them. Or in other countries, the displaced may merge into local communities, where gaining access will require programs that extend to the entire community. In Liberia, for example, as many as three-quarters of a million displaced fled to the capital, Monrovia, blending in with the rest of the city, whose population nearly tripled in size during the civil war.

It needs to be underscored that the impact of internal displacement extends beyond those displaced to disrupt whole communities and societies. The areas left behind and the areas to which the displaced flee often suffer damage. In Rwanda, the World Bank estimates that the damage done to national parks and forests will have long-term economic effects. In Angola and Liberia, the overloading of urban infrastructure has hastened its deterioration. Few African countries can afford such destruction. Ten of the African countries with significant internally displaced populations are among the thirty poorest countries in the world.

Conflict and displacement also spill over borders into neighboring countries. Internal displacement is not only a human rights and humanitarian issue but a political and security one. The Great Lakes region of Africa is a good example of how conflict and displacement in one country inflame the situation in others and lead to large refugee flows and also military invasions of other countries. Similarly in the Horn of Africa and West Africa, conflict and internal displacement can quickly spill over borders and help destabilize neighboring countries.

In the absence of an international system for dealing with internal displacement, it has become necessary to develop a framework of normative standards and institutional arrangements to guide the actions both of governments and of international humanitarian and development agencies. In 1992, the UN Secretary-General appointed a Representative on Internally Displaced Persons, Francis M. Deng of the Brookings Institution, to grapple with these issues. Deng found the doctrine of sovereignty as responsibility the most suitable conceptual framework for dealing with internal displacement. Sometimes called the Brookings doctrine, it basically says that states have primary responsibility for the security and well-being of their populations. If they are unable to provide protection and life-supporting assistance, they are expected to request and accept outside offers of aid. Should they refuse or deliberately obstruct access, they can no longer claim the prerogatives of sovereignty; the international community must become involved.

To provide the international community with a basis for legitimate action, a legal framework for the internally displaced is now being put in place. A set of Guiding Principles on Internal Displacement was introduced into the United Nations in 1998 by the Representative of the Secretary-General. The Principles set forth the rights of IDPs and the obligations of governments and insurgent groups to these populations. Although the Principles are not a binding legal document, they have already gained, in a short time, standing and authority. They have been acknowledged by the UN and endorsed by the major international humanitarian and development organizations. They have begun to be used in the field as an advocacy tool by international organizations, regional bodies and NGOs. In Africa, the OAU Commission on Refugees has taken note of them with interest and appreciation.

Related Books

At the institutional level, an array of international humanitarian, human rights and development organizations have come forward to offer protection, assistance, and development aid to IDPs. Nonetheless, no organization has a global mandate to protect and assist the internally displaced. As a result, the system is ad hoc: organizations basically pick and choose the situations in which they will become involved on the basis of mandates, resources or other considerations. Thus, UNHCR deems about 5 million internally displaced persons out of a total of 20 to 25 million to be of concern to the organization. And only 1 million of these are in Africa. Nor does UNICEF or the ICRC deal with all situations of internal displacement. The selectivity of the international system could be modified if there were an effective central point within the system to routinely and rapidly assign responsibilities to different agencies. But there is no such point, although for the first time there is a vigorous UN Emergency Relief Coordinator who is taking the plight of IDPs seriously and is trying to strengthen coordination among UN agencies.

OAU Secretary-General Salim expressed concern about the discrepancy in treatment between refugees and internally displaced persons in Africa in a public statement in 1995. Indeed, in Rwanda during the great emergency, most donor funding went to the refugee camps outside the country, whereas only a fraction went to assistance and development needs within Rwanda. This led to proposals to approach such situations from a regional or subregional perspective so that the distribution of international resources and attention would be more equitable.

Discrepancy in treatment can spawn conflict, which is just the opposite of what UN assistance is intended to accomplish. In Burundi, for example, in 1993, the conditions in UNHCR camps for refugees were so much better than the conditions in the camps for IDPs—and they were just up the road from one another—that resentments and violent incidents resulted. During returns too, IDPs may suffer neglect. Unlike returning refugees, they are often not provided with seeds, tools and other service packages to facilitate reintegration.

A definite locus of responsibility is needed in the field for the internally displaced. Masses in Flight recommends that in each situation, one operational agency on the ground be tasked with monitoring the conditions of IDPs, identifying their assistance and protection needs, and working with a coordinating mechanism to promote a division of labor to address those needs. The study also recommends that attention be focused on the physical security and human rights of the population as well as on its need for food, medicine and shelter. In 1995 in Rwanda, one of the largest massacres of IDPs in a camp took place in Kibeho, where UN agencies effectively provided relief, but where there was little or no attention to protection. In Angola too, in 1996, NGOs reported that the UN agencies engaged in reconstruction and development were paying scant attention to the protection problems facing IDPs.

It should be pointed out that protection for internally displaced persons is often different than for refugees. Access may be problematic in internal conflicts and have to be secured. Safe havens may need to be created. Evacuations may be needed. Strong advocacy may be essential. Not many international organizations have this kind of experience or expertise. Only ICRC does, although increasingly UNHCR and other organizations have begun to focus more attention on providing protection to internally displaced populations through increased field presence, reporting of protection problems and joint advocacy.

Of course, in some situations the only way to provide protection is through military and police action. Whereas refugee situations rarely require this kind of intervention, IDP situations more often do. In Africa, regional military intervention is notable. In 1990 in Liberia, the Economic Community of West African States (ECOWAS) through its military arm ECOMOG set up a safe haven for thousands of displaced persons in Monrovia and made possible the return of humanitarian agencies to the city. More recently in Sierra Leone, ECOMOG restored order in the capital and protected displaced and other populations. While it is true that in both cases, ECOMOG forces also looted and raped and committed human rights violations, the record had a positive side from the point of IDP protection. What is clearly needed is better military discipline, more stringent UN oversight and regular payment and support of the troops.


Regional arrangements are critical, especially since international intervention is often not forthcoming or problematic when it does take place. In the case of Africa, the intervention in Somalia in 1992 and the intervention in the Rwanda after the genocide to protect displaced populations have very mixed records. The 1992 military intervention in Somalia succeeded in preventing mass starvation, but did not disarm the local factions, became embroiled in the conflict, and in the end failed to provide protection for itself or Somali civilians. In Rwanda, in 1995, the UN military force, UNAMIR, was specifically tasked with protecting internally displaced persons in camps. But when the Rwandan army tried to force the IDPs to go home and attacked and killed thousands of them, UNAMIR didn’t intervene. Basically, it interpreted its mandate to not include defending internally displaced persons from actions by their own government. The lesson here, however, is not that international intervention should be avoided but rather that it should be made better through clearer mandates, sufficient resources, equipment and troops to do the job.

Of course, neither military action nor humanitarian assistance should be anything but temporary. To be effective, both have to be accompanied by measures to address the underlying causes of the conflict and displacement. Political settlements are needed to resolve disputes, mediate among groups and deal with the inequities at the heart of many conflicts. Internal displacement, after all, is but a symptom of a far deeper problem within a society. Yet far too little energy and resources seems to be spent trying to prevent or contain crises in Africa. There is need for regional and international organizations to bring pressure on warring parties, insist on the mediation of disputes, limit the supply of arms, and be ready to offer development aid, investment and debt relief to those who will work to bring conflicts under control. International development and financial institutions also need to become involved earlier on to try to help stabilize situations, prevent displacement and contribute to return and reintegration.

To conclude, we need to take a more comprehensive look at humanitarian and human rights crises. They produce displacement across borders but also within countries. It is time for national, regional and international response to go beyond the parameters set up after the second world war and address the totality of the problem.