It was becoming obvious that some modicum of coexistence had to be found, and the man who managed its precarious negotiation was Prime Minister Anatole Kanyenkiko, a liberal UPRONA Tutsi. Kanyenkiko’s idea was not revolutionary or even very innovative; it was simply an effort at common sense. Because the struggle for power between the groups had everything to do with money, positions, and patronage, he would try to share those according to the rough proportionality of each group’s political weight (which of course did not mean demographic importance), taking into account as best he could the various private interests and nuisance capacities of the various “political parties” (i.e., coalitions of appetites). Haggling went on for about six weeks, and on September 12, 1994, a power-sharing agreement was signed.
The problem was that most of the signatories acted in absolute bad faith. As soon as the agreement was signed the Tutsi extremists started what Professor Filip Reyntjens has called a “creeping coup”:103 strikes, forced business closings, procedural guerrilla tactics to challenge the nomination of Hutu appointees, use of the still Tutsi-dominated justice system to obstruct what could be obstructed, and an effort at de-legitimizing the FRODEBU government in the press. On the other side, the “moderate” Hutu politicians did very little to discourage the extremist guerrillas, which kept operating even after the power-sharing agreement had been signed.
Under increasing pressure from extremists in his own camp, UPRONA Prime Minister Kanyenkiko was forced to resign in early 1995, declaring in his resignation speech, “Today, a handful of politicians from all sides are refusing to draw lessons from the Rwandese tragedy which happened at our doorstep. They want to lead the country towards a similar tragedy.”104 Although understandable, this moralistic interpretation was wrong. The Burundian politicians were not mad fanatics. They were simply cunning and greedy, cold-blooded poker players. Most of them (and not “a handful”) were engaged in a short-sighted game of brinkmanship in which they expected to come out ahead without a genocide. They thought they could control the situation, kill their enemies, and get away with it. The prize was undiluted power, or at least majority power. The power-sharing agreement could not work durably because there was always a group somewhere that felt somehow slighted, that thought its rightful demands had not been met. The Tutsi had formally lost power through an election in 1993, but they had retained enough of it through their control of the army, parts of the civil service, and the judiciary to effectively sabotage any power-sharing agreement.105 The result was ceaseless confrontation at all levels, with a rough balance of terror preventing genocide by either side.
This begs a question: Why did the general population accept its role in this ghastly scenario? The answer is tragically simple: because of poverty and the absence of an economy independent from state patronage. The articulation between an economic dead end and violence has been extremely well explained by a (Tutsi) Burundian university professor who is worth quoting here at some length:
A growing part of the peasantry gradually realized [during the 1970s and 1980s] that, through the system of export cash crops, it was caught in a situation that completely blocked the way of any social and economic promotion for its children. In turn these children realized that they could not escape from an agricultural economy whose remuneration steadily decreased… . The state remained the only hope… . For these poor educated youths, these low-ranking civil servants and their peasant families, Ndadaye was more than a president, he was a king, a god, the only hope.106
But after July 1993 the shoe was on the other foot. Even if President Ndadaye was the “only hope” for the Hutu, he was not God and the economy remained as structurally limited under his leadership as it had been before. It was now the Tutsi’s turn to be afraid, and the fight for patronage was on:
The nominations of Hutu in the administration after July 1993, followed by the replacement of both Tutsi and Hutu UPRONA civil servants, down to such level as communal secretaries and marketplace watchmen . . . frightened a lot of people into thinking that they were going to lose not only the symbols of their hegemony but their permanent sources of monetary income and familial patronage. The press magnified the feeling and the UPRONA party played on it… . Demonstrations organized by UPRONA to protest against losses of employment can be seen in the perspective of this organized panic This led some of the members of the armed forces to think that the whole of Burundi society was in a state of upheaval.107
This was a self-fulfilling evaluation. By killing Ndadaye these same members of the armed forces actually drove the whole of Burundi society into a state of upheaval. Regarding patronage, the fear of loss (Tutsi) or hope of gain (Hutu) became sharper than before because the economic situation went from bad to worse as the insecurity spread. Thus the downward spiral became extremely hard to escape: Tutsi and Hutu fought each other to retain access to patronage in a state-dominated economy that their very conflict made poorer by the day, and growing poverty in turn kept ensuring a steady supply of resolute and desperate supporters for the political class. The press played a dreadful role in this process. Most newspapers were without a shred of professional ethics, mixing facts with commentaries or even outright fabrications. Violent, vulgar, and outrageously partisan, they ceaselessly contributed to destabilization in the name of press freedom.108 The prize probably could go to the Tutsi extremist paper Le Carrefour des Idées, which did not flinch from asking in one of its headlines: “Do the Hutu have a soul?”109 or from offering a front-page bounty of one million Burundian francs for the head of Léonard Nyangoma.110 The worst is that no legal action was ever taken against these publications, which usually sold better than the serious papers.111
This created a climate of slow-motion civil war. But there were no compactly held territory and no battle lines in this war. The two sides were hopelessly intertwined because they had lived side by side before the conflict and because the insurgents never had the military means to cut off certain areas. During the October 1993 massacres, the Tutsi had run to the little towns and villages, even to schools, factories, and hospitals that were isolated in the rural areas but that had an enclosure and could be turned into a strong point for defense. They had remained there, under thin army protection, at times forming their own armed militias. Going back to farming was hazardous, and they often saw their Hutu neighbors pick their crops. Hutu guerrillas and Tutsi militias raided and counterraided. The army tried to control the situation, at times neutrally, at times siding openly with the Tutsi militias. Whatever the fight pattern, the combatants usually suffered fewer losses than the civilians.
In such a climate the power-sharing agreement of 1994 never had much of a chance. Bujumbura itself became a battlefield, with such areas as Kamenge, Cibitoke, Kinama, and Gasenyi turning into Hutu bastions while the Tutsi entrenched themselves in Musaga, Buyenzi, and Ngarara. The showdown ended when the army ethnically cleansed the whole city between March and June 1995, pushing thousands of Hutu into the surrounding hills or all the way to Zaire. Then the army went after the IDPs, whom it accused of supporting the guerrillas. By early 1996 the situation appeared totally out of control. The “small massacres” had finally bubbled over,112 and there was a growing fear of genocide among the international community, even if, as I showed earlier, the prerequisites of a genocide did not really exist.113 Former president Nyerere tried to organize negotiations between UPRONA and FRODEBU in Tanzania in the hope that they could talk more securely outside the country than in Burundi itself, but the conversations failed. By then more of a hostage than a real head of state, President Sylvestre Ntibantunganya went to the Arusha regional summit of June 25–26, 1996, with a radical request: the international community should intervene militarily to reestablish a minimum of law and order. This was a radical turnaround; when the idea had been first mooted by UN Secretary-General Boutros Ghali in January, it had been rejected out of hand by UPRONA. President Ntibantunganya had gone along with the UPRONA rejection at the Tunis mini-summit114 because he knew that th
e Burundian army and the Tutsi extremists would not accept it. And as soon as UPRONA Prime Minister Antoine Nduwayo, who had backed the president’s position in Arusha, came back to Burundi he was denounced by his own party as “guilty of high treason.” Paradoxically, the radical PALIPEHUTU guerrillas agreed with UPRONA in their rejection of foreign intervention.115
It was in that extremely tense climate that two successive massacres came to push the situation over the brink. First, on July 3, Hutu guerrillas hit the Teza tea factory, killing between sixty and eighty people. Then, on July 20, three hundred displaced Tutsi were killed at the Bugendana camp. Rumors were rife of an imminent intervention by the regional powers or by the UN, although plans were extremely confused. When President Ntibantunganya went to the Bugendana funeral, the crowd pelted him with stones. He fled directly from the funeral site to the U.S. Embassy in Bujumbura, expecting a coup at any time. In fact, the coup took four days in coming because the Bururi Tutsi community was divided about who should carry it out, former President Pierre Buyoya or his radical predecessor, Jean-Baptiste Bagaza. The army NCOs and the Tutsi extremist militiamen preferred Bagaza; the officers and the Bururi elders were for Buyoya, with the argument that “Bagaza is too radical for the Abazungu [Europeans].” Buyoya got the support and took power in a bloodless coup on July 26.
Buyoya, the army, and the semi-moderate elements of UPRONA had carried out the coup both to finish off the FRODEBU-led regime that had come out of the June 1993 election and to stop the threatening slide into anarchy. But the coup could not achieve any sudden turnaround of the situation. There were still rumors of a possible UN intervention, even if nobody in New York seemed too enthusiastic about it.116 But the regional summit that met in Arusha on July 31 was very hostile and clamped radical economic sanctions on Burundi. The prime movers of the hostility toward Buyoya were the Ugandan and Tanzanian governments, for two different reasons. Tanzania had long been semi-supportive of the Hutu guerrillas because former President Nyerere, who remained the real master of Tanzanian diplomacy despite his retirement, had never believed in the good sense of having an independent Rwanda and Burundi. He could be quite frank about it, such as when he declared at a private meeting in Washington, DC in September 1996:
We might succeed to bring a political solution to Burundi. But we must not leave Rwanda and Burundi hanging there, they are too small and unviable. The real problem is the pushing of space. Rwanda and Burundi were part of Tanganyika before World War I. If Tanganyika, Rwanda, and Burundi hadn’t been broken up after the Versailles Treaty, you wouldn’t be hearing about Tutsis and Hutus today. Even now . . . if we can give them space, if we can find a solution for five or six years, then we can make them part of East Africa which is growing up again.117
Unknown to most of his American hosts, “Mwalimu” had of course a rather radical solution in mind “for the next five or six years,” since he was at the time putting the finishing touches to his general plan for the attack on Zaire. The Buyoya coup had made his long-term strategic planning harder because it would tend to reinforce the Tutsi regime, which he wanted to crack open. If Nyerere’s position had nothing to do with Buyoya as an individual, this was not the case with Uganda’s Museveni. Shortly after Mwalimu’s visit Amama Mbabazi, Museveni’s roving diplomatic troubleshooter, also went to Washington and declared in defense of the Arusha Summit sanctions, “The Tutsi dominate the Army and the civil service in Burundi. We must open them to the Hutu, there is no other way. We must find an adequate system of power-sharing, as we have in Uganda.”118 If this was apparently more politically correct than Mwalimu’s reasoning it is probably because it was said in public. But the reality was that Museveni could not stand Buyoya, whom he reproached for two reasons: overthrowing his friend Bagaza in 1987 and organizing elections in 1993. The first reason was very affective: Museveni was grateful to Bagaza for his help during his own bush war119 and generally agreed with his authoritarian style. He thought that Bagaza was the right man to negotiate with Nyangoma and that peace could come from this rapprochement of opposite extremisms.120 For the same reason he disliked Buyoya, whom he saw as a wet and a fool for having organized the June 1993 elections. Museveni thought that Africa was not ripe for multiparty politics121 and that Burundi, of all African countries, was probably one of the least ripe. But Museveni’s and Nyerere’s views, although diversely motivated, converged on one key point: Buyoya was not the man they wanted in power in Burundi. Mwalimu would not have minded a continuation of Ntibantunganya’s weak presidency, and Museveni would have liked a Bagaza coup. But both were unhappy and Buyoya could expect hostility from them.
This was not the case with Rwanda, which was very concerned about the situation in Burundi. Rwanda knew that if the Burundi regime imploded, it would have to deal with at least half a million Tutsi refugees.122 In addition, it could expect a radical Hutu regime to provide the ex-FAR and Interahamwe with military bases. Rwandese troops had started to operate alongside the Forces Armées Burundaises (FAB) as early as 1995, collaborating with the Tutsi militias to attack the Rwandese Hutu refugee camps in Burundi.123 In early August 1996 RPA contingents crossed the border to support the FAB in large-scale antiguerrilla operations.124 The situation had grown worse on the ground as the rebel Hutu forces were now trying to cut off Bujumbura’s food supply and had managed to knock down electric power lines, plunging the capital in darkness. There was a sense of urgency in Kigali. In a tense and hurried effort, the two governments managed to deport the last 85,000 Rwandese Hutu refugees left in Burundi back to Rwanda.125 On August 17 the Rwandese delegation at the Kampala regional meeting could not disagree with Uganda and Tanzania and had to half-heartedly approve the confirmation of sanctions against Burundi, but it then immediately set about undermining them. On August 30 the United Nations gave the new president of Burundi a two-month ultimatum to negotiate with the rebels. Buyoya answered, “Peace will not be made at the Security Council,”126 a statement certainly not politically correct but definitely prophetic. Rwanda could not afford to let Burundi boil over. For Kigali the last element of the situation had now fallen into place and the time had come for action.
General Kagame goes to war
If it was the situation in Burundi that gave the final call for action in General Kagame’s judgment, it was far from being his only reason to move. The basic cause that led the Rwandese leadership to attack Zaire in September 1996 was the presence of the large, partially militarized refugee camps on its borders. But there was also a broader view, which was a systematic trans-African plan to overthrow the Mobutu regime in Zaire. Already in November 1994, in the wake of the Rwandese genocide, Museveni had called a meeting in Kampala of all the “serious” enemies of Mobutu to discuss the idea of overthrowing him.127 The conclusion had been that the time was not yet ripe. In early 1995 former president Nyerere had relaunched the idea, developing contacts with a number of African heads of state with the purpose of cleaning up what they looked upon as the shame of Africa. The heads of state involved were the presidents of Eritrea, Ethiopia, Rwanda, Uganda, Zimbabwe, and Angola. Mwalimu himself occupied the place of President Mkapa because the really important issues of Tanzanian diplomacy had always remained his prerogative.128 His basic reasons for launching this unconventional effort were coherent with his lifelong choices: socialism and pan-Africanism. Nyerere felt that a new generation of African leaders had recently come to the fore who were committed to the basic ideals of something that might not be called “socialism” but that was a basic and radical concern for the social and economic welfare of their populations. At the top of the list were Presidents Issayas Afeworki of Eritrea and Meles Zenawi of Ethiopia, whom he admired for having put an end to the thirty-year-long Eritrean conflict.129 President Museveni and his protégé, General Kagame, were next. To these young men Mwalimu was ready to add the older leftists whom he had supported during the years of the anti-apartheid struggle, Robert Mugabe of Zimbabwe and Agostinho Neto’s heir, José Eduardo dos Santos, of
Angola. As an enormous and unmentioned counterpoint to the project, there was the victory of the anti-apartheid forces in South Africa and the ascension to power of President Nelson Mandela. But Mwalimu had no intention of asking the new South African leadership to get into the act of overthrowing Mobutu, even if he knew that he could count on its sympathy. There were too many reasons against it. South Africa needed time. South Africa should use the moral authority of Nelson Mandela to act as a referee rather than get into the fight itself. And, last but not least, Mwalimu knew that all was not for the best between the African National Congress and its friends in Luanda and Harare and that getting them to collaborate on the ground might be difficult. Rwanda, because of the refugee question, was of course to be the entry point and the spearhead of the operation. Never mind that General Kagame probably had scant regard for the inclusion of Rwanda into either a resurrected version of Deutsche Ostafrika or a modernized version of the East African Community. In the short run he was satisfied to be able to count on a regional alliance to back him. A Rwandese journalist summed up the situation quite publicly when he wrote,
The present situation in Burundi is largely a result of Zairian support for PALIPEHUTU and CNDD. The final attack on Burundi would be a catastrophe for Rwanda because the plan is to allow Nyangoma to take power in Bujumbura and to bring the Interahamwe back in Rwanda. But Zaire should be careful. The RPA can fight back. In which case the Great Lakes Region might witness the end of the Mobutu dictatorship.130
Africa's World War: Congo, the Rwandan Genocide, and the Making of a Continental Catastrophe Page 13