TOC | Summary | One | Two | Three | Four | Five | Six | Notes | About the Author

Preventing Genocide in Burundi Lessons from International Diplomacy

Burundi: The Politics of Genocide

What distinguishes the violent conflict in Burundi from so many others is the extent to which elite-led, politico-ethnic rivalry for power and economic and social advancement has become entwined with mass killing and fears of group extinction. Would-be peacemakers not only confront the challenge of fostering nonviolent processes of decision making and building trust among former enemies, they must also address the fears of the 15 to 20 percent Tutsi minority and the 80 to 85 percent Hutu majority that the other group will eventually commit genocide against it. As political scientists and human rights monitors have testified, much of the current violence is in fact genocidal, even if it is not always possible to document the “intent” to partially or wholly destroy an ethnic group required by the international Convention on Genocide.⁵ Rene Lemarchand, the foremost scholar of Burundi politics, explains,

Behind the murders of political opponents, the systematic ethnic cleansing of urban and rural districts, the armed attacks on refugees and internally displaced persons, and the ambushes of civilians lies the conviction held by both Tutsi and Hutu that unless the other’s crimes are retaliated against by retribution, planned annihilation will inevitably follow.⁶

      In Burundi, even relative “moderates” are enveloped in the atmosphere of fear and recrimination that ultimately justifies preventive violence. Thus, many Hutu political leaders who have sought nonviolent, compromise political solutions nevertheless strongly suspect that their “moderate” Tutsi counterparts were complicit in the murder of President Ndadaye. Some of these leaders, in reacting to the assassination, made remarks that may have unintentionally fueled ethnic massacres of Tutsi. On the other hand, many on the Tutsi side who supported the democratic transition believe that the main Hutu-led party planned for massacres of Tutsi even before Ndadaye’s assassination. They speak darkly about the “necessity of eradicating genocidal ideologies.”⁷ Neither side’s allegations have been proven, but both are “credible” in the context of Burundi’s political experience.

The Roots of Conflict

A look at history sheds light on the dynamics of Burundian political development and its implications for peacemaking.⁸ As in Bosnia, ethnic violence and genocide are by no means the inevitable results of ancient tribal hatreds. In fact, there were few violent confrontations between Tutsi and Hutu until a little more than 30 years ago. When German colonists arrived in the capital of Bujumbura near the end of the 19th century, Burundi was a long-established, decentralized kingdom. The principal political rivals were not Hutu or Tutsi but members of a small royal group, the ganwas or “princes of the blood.” Although the predominantly pastoral Tutsi benefited more from the extensive patron-client system than the mainly agricultural Hutu, the two groups inhabited the same lands, spoke the same language, shared a largely common culture, and often intermarried. Despite an overall distinction in social status, their economic circumstances were often quite similar. Even status differences were subject to qualification. Some Hutu clans were very influential, furnishing the ganwas with many of their advisers, managers, religious authorities, and local deputies. And certain Tutsi clans were forbidden to enter the king’s court. Generally speaking, regional and family identities appear to have been more central to traditional Burundi politics than ethnic ones.

      It was under the influence of colonialism that ethnic ties became more salient. Centralization and modernization eroded old identities based on locality, kinship, and dynastic rule. Moreover, Belgium’s colonial takeover after World War I produced authoritarian manipulations of the ganwa system that made it less able to satisfy its Hutu constituents. At the same time, Belgian educational, cultural, and administrative policies promoted the social and political advance of the “noble” Tutsi, whose “fine bearing alone guarantee them considerble prestige over . . . the worthy Hutu, less clever, more simple, and more trusting.”⁹

      As elsewhere in Africa, the transition to independence spawned a struggle for power among politico-ethnic elites who gravitated toward authoritarianism. But nowhere was the conflict as stark as in overpopulated Rwanda and Burundi where just two groups, with radically different populations and social statuses, confronted one other. There was hope that Burundi might avoid the worst because its ethnic dichotomies were less extreme than those of its “false twin,” Rwanda. But Burundi’s politics was heavily influenced by the sudden, Belgian-abetted rise of Hutu power in independent Rwanda, a power that literally drove much of the Tutsi population out of Rwanda. In 1963, Burundi prime minister-designate Prince Louis Rwagasore—a ganwa who had brought Hutu and Tutsi together in the dominant UPRONA (Union for National Progress) party—was assassinated by political rivals. After that, the political process unfolded with the fatalism of a Greek tragedy.

      Politico-ethnic conflict, centered in the urban elite, led to the assassination and wounding of two more prime ministers. In 1965, when it appeared that the Tutsi side had seized the upper hand, some Hutu officers and politicians attempted a violent coup. This led to the first ethnic pogroms, including the liquidation of almost the entire Hutu political elite. It also ushered in a long series of military-based governments. After another, more serious Hutu armed challenge in 1972, the regime murdered 100,000 to 200,000 people in three months, a “selective genocide” that targeted educated Hutu. Concerning this entire period, a former Burundian leader wrote, “A minority of tribalists . . . lit the spark that destroyed a whole building.”¹⁰

      The recurrent pattern of Burundi politics was now established: Political exclusion and repression generate extremist movements of resistance that propound ideologies of ethnic superiority and are willing to use indiscriminate violence against other ethnic groups. Many power-holders react in like fashion. In the process, relative moderates are either eliminated or move toward adopting more extreme agendas. “All Burundi,” former UN Special Representative Ould-Abdallah has observed, “at a given moment are extremists. For a very simple reason—politics in their country has a vital stake . . . life and death for each person.”¹¹ After a period of restored stability in which regional, clan, and other divisions within the ruling elite come to the fore, the cycle of interethnic violence begins again.

      These defining years also marked the rise of the Southern-based Tutsi military-political-business faction known as the “Bururi Lobby.” Its ascendancy continues to the present day. The corrupt regime of Captain Michel Micombero was overturned in 1976 by a coup led by his cousin, Lieutenant Colonel Jean-Baptiste Bagaza. Although Bagaza initially promised ethnic reconciliation and took some modernizing economic initiatives, he also pioneered new forms of repression, including a reinforcement of educational discrimination and restrictions on the dominant Catholic Church. The latter, along with moves to downsize the army and charges of government corruption, helped provoke a 1987 coup led by Major Pierre Buyoya, Bagaza’s cousin and Micombero’s nephew.

Recent Political Developments and International Reactions

Although Buyoya initially showed only a faint intention to change the political structure, he was soon transported by events. In 1988, rising ethnic expectations and fears erupted into genocidal killings of up to 20,000 people in two northern communes. Most of the dead were Hutu, victims of blind army retaliation. Unlike 1972, these well-publicized massacres led the U.S. Congress, Belgium, the European Union (but not France), and the World Bank to threaten to withhold aid unless the government moved toward reconciliation with its Hutu citizens. Dependent upon foreign assistance for a quarter of its gross national product, the Buyoya regime promptly adopted a top-down program of liberalization. With a push from the West, an international effort to prevent further genocide seemed to be having an impact.

      Under additional pressure from the spreading African democratization movement and Western donors, the government accepted multiparty democracy. Buyoya and the governing UPRONA party hoped to co-opt enough Hutu support to maintain substantial Tutsi domination. But the Hutu-led FRODEBU (Front for Democracy in Burundi) party swept the June 1993 elections. Within four months, the Tutsi military’s attempted coup fatally wounded democracy and reawakened the Furies of genocide.

      What lessons can be extracted from the international community’s first, abortive attempt to alter the generation-old pattern of Burundi politics? First, democratic change was profoundly threatening to Tutsi interests. But, in contrast with South Africa’s contemporaneous transition from minority rule, Burundi evaded negotiation of compromise solutions to virtually all of the flammable issues of national politics. These included the return and absorption of hundreds of thousands of Hutu refugees; the integration of Hutu into the army, police, judiciary, and civil service; and revisions of state policies on loans, contracts, and privatization of industry to incorporate Hutu interests. Disputes over these issues helped incite the attempted coup and subsequent violent conflict. In the absence of progress toward a new political consensus, relatively moderate leaders like Ndadaye, who incorporated substantial opposition forces into his government, and Buyoya, who relinquished power after his electoral defeat, could not contain the extremist forces of Burundi politics.¹²

      Second, although ethnic massacres had been central to political life, there was no effort to defuse intergroup fear and craving for vengeance by acknowledging the worst crimes of the past and providing some measure of accountability for them. As events in the former Yugoslavia were then demonstrating, leaving the “culture of impunity” intact only encouraged further genocidal violence.¹³

      Third, the transition, like Rwanda’s, contained no effective counterweight, domestic or foreign, to the overwhelmingly monoethnic military. This left the decisive card in the hands of a group that both mirrored continuing Tutsi civilian concerns and felt specifically threatened by the new government’s announced reforms of the security forces.¹⁴

      Following the assassination of Ndadaye and the ensuing interethnic massacres, Burundian politics became more extreme and fragmented. Tutsi political hardliners within UPRONA and an increasing number of small parties, together with elements of the military and civilian militia, embarked upon a “long series of provocations and killings . . . to weaken FRODEBU, intimidate moderate politicians (including those within UPRONA) and paralyze public life.”¹⁵ In this they drew support from the Tutsi-dominated Constitutional Court, which delayed the selection of a new president after Ndadaye’s successor was killed in a plane crash along with the president of Rwanda. On the other side, certain elements of FRODEBU began to arm and attack the Burundian military.

      During 1994, and especially in the wake of the Rwandan genocide, the international community, acting through UN Special Representative Ould-Abdallah, helped the Burundian political parties reach a series of transitional power-sharing agremeents. These accords were designed to ensure “a minimum of political stability” pending the 1998 elections.¹⁶ They failed for two principal reasons. First, the major agreement, the Convention of Government, virtually annihilated FRODEBU’s election victory by essentially superseding the 1992 constitution, guaranteeing the Tutsi-led opposition a 45 percent share in the government, and establishing a National Security Council in which the opposition could block key moves by the FRODEBU Hutu president. More fundamentally, as Filip Reyntjens later commented, “Rather than attempting to tackle the real problems of the country, these negotiations dealt with the distribution of offices and functions.”¹⁷ The UN’s emergency diplomacy was perfectly understandable, but in the special context of Burundi politics it proved inadequate to control the extremist dynamic unleashed by Ndadaye’s assassination and even seemed to fuel it. The Convention of Government and related power-sharing accords were undermined by opposition, intimidation, and violence from the Tutsi-led parties, the army, and associated militia. And disillusionment with the Convention and its consequences spurred the growth of violent Hutu-led resistance.

      By 1995, beneath a veneer of multiparty governance featuring FRODEBU President Sylvestre Ntibantunganya and a UPRONA prime minister, Burundi was largely controlled by an uneasy coalition of the army, various Tutsi militia, UPRONA hardliners, and small Tutsi parties. But it faced a growing challenge from Hutu-led insurgents, especially those of the National Council for the Defense of Democracy (CNDD), which was headed by former FRODEBU Minister of the Interior Leonard Nyangoma and backed by perhaps half of FRODEBU’s increasingly powerless parliamentary majority. Arms flowed freely to both sides, including notable shipments to the government from China, Russia, and Eastern Europe (via Belgium, Rwanda, and Tanzania) and to the CNDD from the former Soviet Union, Eastern Europe, Angola, and South Africa (via Belgium, Sudan, and Zaire, the latter channel involving the family of Zairian President Mobutu Sese Seko). As dramatic massacres and displacements proved, the main victims of the war were civilians.¹⁸

      Buyoya’s July 1996 coup brought greater coherence to the government while removing the last remnants of Hutu political power. Many Western officials were grateful that the army did not install a more “extreme” leader, such as former President Bagaza. Still, the coup generally signaled a more uncompromising political stance and a greater emphasis on military repression. Even as Buyoya somewhat acceded to international pressure for secret talks with the CNDD and cooperation with regional African mediation, his government legally harassed the powerless FRODEBU parliamentary and party leadership, doubled the size of the army, and undertook a massive forced temporary “regroupment” of hundreds of thousands of Hutu peasants. In late 1996, the regime became modestly involved in the conflict in eastern Zaire and benefited from one of its major results: the dismantling of the CNDD’s main external bases.¹⁹ By September 1997, security in some parts of Burundi had improved and the monthly death toll had been cut in half, to 400 to 600. But the insurgency still appeared well rooted and continued to have some access to the outside through Tanzanian and other porous borders.²⁰ And sputtering tensions between Burundi and Tanzania raised the possibility of a wider war.

Possible Political Solutions and an Appropriate International Role

Even if the government were able to temporarily contain the CNDD and two smaller rebel groups—PALIPEHUTU (Party for the Liberation of the Hutu People) and FROLINA (National Liberation Front)—the basis for future confict and genocide would still exist. The last 30 years of Burundi history suggest that the only way out of the labyrinth of violence is a compromise political settlement followed by a long period of peacebuilding. But what kind of settlement is feasible given this bleak past?

      Partition, a frequent recourse in desperate ethnic conflicts, seems unlikely here, although it cannot be excluded as a last resort. There are no traditional “Hutulands” and “Tutsilands” in Burundi. And while the war has recently brought about a substantial amount of ethnic segregation, this has been at the local rather than regional level; therefore, any attempt at territorial division would entail truly massive population transfers. Most important, the task of calculating a mutually acceptable partition between the politically predominant 15 to 20 percent Tutsi minority and the subjugated Hutu majority would be an extremely daunting one. Another major obstacle would be anticipated resistance from the Tutsi-led regime in adjoining Rwanda, which could not be happy about a precedent for territorial division and possible new security threats along the border.²¹

      Nor do Burundian circumstances seem propitious for the reconciliationist system of “consociational democracy.” The Tutsi comprise too small a part of the population to benefit much from the principle of proportional representation. A “grand coalition” of political elites would be impeded by the fact that these elites have been heavily implicated in past ethnic violence and have few cross-cutting relationships that transcend ethnicity. And any comprehensive minority veto on national political issues would undoubtedly stymie Hutu advances.²² Consociationalism is no recipe for long-term political stability in Burundi.

      Suggestions for bypassing ethnic conflict by incorporating Burundi into a larger regional system of economic and ultimately political cooperation have the potential for a “win-win” payoff for both Hutu and Tutsi. But this is a long-range remedy, one that will also have to overcome strong patterns of discrimination against both groups in neighboring countries.

      A political settlement that took adequate account of Burundian history and circumstances would probably have three basic characteristics. First, the essence of any enduring compromise would be a form of democratic power sharing that was more majoritarian than consociational but guaranteed minority security—perhaps in part through local autonomy—and provided some significant protection for minority socioeconomic and political interests. What is needed is a negotiated accommodation between the aspirations of the majority and the acquisitions of the minority, a specifically Burundian version of what has recently occurred in El Salvador and South Africa. Of course, in the Burundi case, the achievement of a negotiated compromise is greatly hampered by the history of genocide.

      Second, there would be measures to address collective fears and memories of genocide by acknowledging past crimes and fixing individual rather than ethnic group responsibility for them. There is an emerging international menu of ways to deal with this problem, ranging from truth commissions and amnesty programs in El Salvador and South Africa to national criminal prosecutions and international criminal tribunals for Yugoslavia and Rwanda. Also, according to the head of a leading Burundian human rights organization, Burundians have relevant local traditions of justice, including that of pronouncing sentences for certain crimes but not carrying them out.

      Third, there would be impartial outside military forces sufficient to control the Burundian military and other armed groups until an ethnically integrated and politically subordinate military institution emerged (probably three to five years). This is particularly necessary in Burundi, which has absolutely no tradition of constitutional rule and where the military has long governed as an instrument of Tutsi privilege and security. In this respect, the Burundian situation differs greatly from those in South Africa, Zimbabwe, and El Salvador.

      It is almost unthinkable that such a settlement could develop without substantial international support, including a deft mixture of carrots and sticks. “One thing I know,” observed an African leader who had worked to reconcile the Burundian groups for almost two years, “They can’t do it on their own.” Because fear and extremism rule the “land of a thousand hills,” moderate initiatives only develop with the help of outside pressure and assistance. Prime examples are the liberalization of the late 1980s and early 1990s and the power-sharing accords of 1994.

      The demise of these initiatives holds some important lessons regarding the way in which the international community needs to direct its pressure. As we have seen, these failures largely stemmed from the lack of genuine engagement of all important parties, especially the “extremists”, in successful political negotiations. Without such engagement, the “moderates” who helped spark constructive change were ultimately powerless. Ironically, a moderate political system in Burundi can only be established with strong participation from extremists, who constitute the dynamic forces in Burundian politics. The challenge to the international community is to use its carrots and sticks to engage the extremists in peacemaking.

      There is no more powerful argument for this proposed international role than that provided by neighboring Rwanda’s descent into civil war and genocide. Although the central issue there was the alteration of a system dominated by a narrow Hutu political elite, it took concerted international pressure to produce political negotiations. Yet these did not succeed, because key Hutu extremists did not participate, a mutually acceptable balance of majority rule and minority rights was not achieved, past crimes were ignored, and a UN peacekeeping force was too weak to control the Rwandan military and associated militia.²³

      As the Burundian civil war intensified in late 1995, members of the international community undertook new approaches to prevent genocide and foster political reconciliation. We now turn toward the analysis of these approaches.

TOC | Summary | One | Two | Three | Four | Five | Six | Notes | About the Author

Home  |  Jobs  |  FAQs   |  Contact Us  |  Directions  |  Privacy Policy  |  Site Map

United States Institute of Peace  --  1200 17th Street NW  -- Washington, DC 20036
(202) 457-1700 (phone)  --  (202) 429-6063 (fax)
Send Feedback