Africa's World War: Congo, the Rwandan Genocide, and the Making of a Continental Catastrophe

by Gerard Prunier

  It is crucial to keep in mind both the fragility and the interconnectedness of the Kivus if we want to understand the process that turned a genocide in the tiny country of Rwanda into what eventually became a continentwide war. The Kivus were (and have remained) a high-danger zone on the continent. Conflicts had started there before the global war of 1996, and they have proven extremely difficult to control even after the generalized conflict receded. They would remain so even after the successful Congolese elections of 2006.

  The slide from what had been a nationally focused genocide into a global war had one basic cause: there was no political treatment of the genocide in Rwanda by the international community. No efforts were made to prevent it, no efforts were made to stop it, and no efforts were made to remonstrate with those who spoke in the name of the victims when they started to abuse their role.44 Mature political treatment was replaced by humanitarian condescension and diplomatic bickering. As a result over two million refugees poured over the borders of Rwanda, complete with all the trappings of quasi-sovereignty, including an army, a treasury, and a complete set of criminal politicians. Because the treatment of the crisis, or should I say the nontreatment, was purely humanitarian, the situation was allowed to fester. In Tanzania, where the political and social makeup of the refugee asylum areas was basically sane in spite of extreme poverty, this did not create too grave a problem. But the Kivus were a highly complex and volatile environment in a country that, due to the corrupt leadership of his lifetime president, had sunk below the level of even the most deficient African polities. And the international community did not react to that any differently than it had reacted to the Rwandese genocide: it sent in the humanitarian battalions, ready with plastic tents, emergency food from our groaning farm surplus stocks, and devoted health care workers out to relieve human pain and suffering. The political side of the equation remained solidly blocked at zero. The following section examines how the results, which could be expected, eventually came to materialize.

  The area of Zaire known as the Kivus (North and South) is about four times the size of Rwanda. But even though these provinces are bigger and less densely populated than Rwanda, they are far from empty. The population patterns are closely linked to the ethnic and political situation. To understand the situation better, consider North and South not separately, but in sequence.

  North Kivu: ethnicity and the land conflict

  The two dimensions of ethnicity and land are intimately linked. For example, the table specifies the densities in North Kivu as humans per square kilometer.

  Zone

  1957

  1984

  Goma

  59

  286

  Rutshuru

  26.4

  91

  Masisi

  38.9

  101

  Walikale

  2.1

  6

  Source: J. C. Willame, Banyarwanda et Banyamulenge (Brussels: CEDAF, 1997),39.

  These figures show that the rate of growth is very high, whereas the densities are very unevenly distributed, depending on the various areas of the province. Densities are one dimension; the nature of the population is another. In North Kivu the local population is made up of the so-called autochtones45 and of Kinyarwanda speakers. There have been Kinyarwanda speakers in North Kivu since time immemorial and they were divided into many small communities (Banyabwisha, Bafumbira migrants from Uganda) that did not have any special relationship problems with their autochthon neighbors. Then came the Mission d’Immigration des Banyarwanda (MIB), created by the Belgians in 1937 to bring agricultural workers from an already overpopulated Rwanda into what was seen as an underpopulated Kivu. During its eighteen years of existence the MIB imported about eighty-five thousand Banyarwanda from Rwanda, mostly into North Kivu, although some were sent down to Katanga to work in the mines.46 They settled in various areas, but with a preference for the Masisi and Walikale zones, then thinly populated. The newcomers, both Tutsi and Hutu, tended to reinforce the community feelings of those Kinyarwanda speakers who were already present. Given their growing numbers (today they represent about 40 percent of the population province-wide, with peaks of over 70 percent in Masisi) this tended to create resentment among the autochthons, especially among the smaller tribes, such as the Bahunde and Banyanga, who already tended to be marginalized by the majority Banande.47 This was due to the fact that the Banyarwanda presence seemed at least partly sponsored by the Belgians and to the fact that their concepts of landholding did not mesh easily with those of the locals. With the support of the colonial administration they received “customary rights” to local tribal lands, which did not imply possession in the Western sense. But the social transformations brought about by colonization tended to weaken the powers of the local lineage chiefs over land attribution, replacing those by more individual and monetarized transactions.48 The Banyarwanda were seen as taking advantage of this new order of things to turn their “acquired customary land rights” into permanent ownership, and a very tense situation developed during 1960–1965, when the first civil war totally upset the forms of government inherited from the colonial administration. Contrary to the South, North Kivu was not directly touched by the war. But the war was often a pretext for local administrators to persecute the Banyarwanda and to illegally seize their land. This finally led to a limited outburst of violence in 1965, after the war had stopped in the rest of the country.49

  President Mobutu did not mind. For him the North Kivu Banyarwanda, especially the Tutsi, were an interesting political investment. The region was potentially rebellious and the Banyarwanda were in a difficult position, especially since many new arrivals had come after the massacres of Tutsi in Rwanda between December 1963 and January 1964. Therefore, using the Banyarwanda was good politics since they could not get a local power base independent from the central government. This led to the promotion of the Banyarwanda in the postwar years, especially during 1967–1977, when Barthélémy Bisengimana, a Tutsi refugee from Rwanda, was Mobutu’s righthand man as the chief of the presidential office. With the support of Bisengimana many Kivu Tutsi went into lucrative businesses and “acquired” land, especially land that had been abandoned by Belgian farmers in 1960 or confiscated from them during the 1973 exercise in “Zairianization.” The land grabbing reached such incredible proportions that in 1980 the Land Ministry in Kinshasa had to cancel the “attribution” of 230,000 hectares (575,000 acres) to the notorious Munyarwanda businessman Cyprien Rwakabuba. At a time when the average amount of land per person in Kivu was 0.81 hectare (2 acres),50 such insolent agrarian “success” won the Banyarwanda very few friends. Even if many of them did not partake of the riches corruptly acquired by their elite, the whole community became stigmatized.

  Parallel to the land struggle, there was another battle about citizenship. In January 1972, at the height of his power, Bisengimana had managed to get the Political Bureau of the Mouvement Populaire de la Révolution (MPR) to pass a citizenship decree.51 whereby all persons originating from “Ruanda-Urundi” and residing on then Belgian Congolese territory in or before January 1950 were automatically Zairian citizens (Article 15).

  When Bisengimana fell from power in 1977 there was intense pressure to change the law, and a new one was passed on June 29, 1981 (Law 81002), abrogating the famous Article 15. Although the new law had vague provisions for the eventual “acquisition” of Zairian nationality, it left this important point to political arbitrariness.52 The results were immediate. In 1987 elections could not be organized in North Kivu because nobody was capable of saying exactly who was or was not Zairian in order to draw polling lists. Then, using the 1981 law, local Banande worthies counterattacked and tried to cut down Banyarwanda landholdings and businesses.53 They managed to get the support of the Bahunde and Banyanga minority tribes (each one representing 5 percent of the North Kivu population), who had always been at the wrong end of the deals, whether it was the Banyarwanda or the majority Banande who had been
on top. To further their aims they used the democratization movement then taking place in Zaire. The autochthons sent delegates to the Conférence Nationale Souveraine (CNS) in Kinshasa and managed to bar Banyarwanda from taking their seats at the Assembly under the pretext that they were not Zairians. Then they used the CNS decisions to completely overhaul the local administration in North Kivu, putting new judges and police in place who were Banande, Bahunde, or Banyanga. The whole justice and repressive apparatus then became slanted against the Banyarwanda.

  In the meantime, war had broken out in neighboring Rwanda and the Tutsi-Hutu conflict there was carried over to Kivu. President Habyarimana and his clan had organized a pro-Kigali network under the guise of a peasant association, the Mutuelle des Agriculteurs des Virunga (MAGRIVI). While the RPF recruited young Zairian Tutsi into its ranks, the MAGRIVI did the same thing with Hutu, thus deeply dividing the Banyarwanda community at the very moment when it was under assault from the autochthon tribes. Tensions rose and finally broke into open violence in February 1992. To paraphrase the title of a contemporary report by Médecins Sans Frontières, the French medical NGO, the struggle for land had burst into interethnic conflict.54 After a period of low-intensity skirmishes during 1992, violence escalated when a “large number” of Hutu Banyarwanda were killed by armed Bahunde militiamen at the Ntoto market near Walikale on March 20, 1993. Within ten days, after about a thousand Banyarwanda had been slaughtered, they counterattacked, doing some killing of their own. By August the casualties numbered around 20,000 and there were about 250,000 IDPs of all tribal origins.55 President Mobutu himself came to North Kivu, set up residence in Goma, and deployed several thousand elite troops from the famed Division Spéciale Présidentielle (DSP). Violence abated by late 1993 and negotiations were undertaken by local leaders in November. By February 1994 an uneasy peace had been reestablished. It was to be shattered five months later by the arrival of 850,000 Hutu refugees from Rwanda.

  South Kivu: the Banyamulenge and the memories of 1965

  Although similar to North Kivu in climate and vegetation, South Kivu is different because it is less populated, which results in less pressure on land problems. While there are 102 humans per square kilometer in Uvira Province, the southern Fizi-Baraka area has only 13.56 The question of ethnicity and nationality is also posed in different terms. Here, the main “nonnatives” are the Barundi, who probably represent up to 15 percent of the population. Curiously, they have had only limited problems with their autochthon neighbors (mostly Bavira and Bafulero), unlike the much smaller Banyamulenge group.

  The Banyamulenge are a group of Banyarwanda migrants who have come from Rwanda at various points in history.57 The first arrivals might go back as early as the seventeenth century.58 But they seemed to have especially come at the close of the eighteenth century from the southern Kinyaga region, to escape the growing power of Mwami Rwabugiri in Rwanda.59 Some came to escape the repression unleashed after the Rucunshu coup d’état of 1896. But in any case they were few in number and they were mostly Tutsi. Their Hutu abagaragu (clients) had been icyihuture, turned into Tutsi, thus dissipating any intragroup social tension. They settled on the Itombwe plateau above the Ruzizi plain, where the altitude (up to 3,000 meters) prevented normal agriculture but where they could pasture their cows. They received further influxes of migrants in 1959, 1964, and 1973, as anti-Tutsi persecutions took place in Rwanda. Poor and somewhat aloof from their Bafulero and Babembe neighbors, they played an almost involuntary political role during the civil war, when the Simba rebels,60 on the run from Jean Schramm’s mercenaries and government soldiers, came up the plateau and started killing their cows to feed themselves. The Banyamulenge, who up to that point had had no group involvement in the rebellion,61 then turned solidly against it, later accepting weapons from the Mobutu forces and joining in the slaughter of the remaining rebels. Because many of those in the area were drawn from the neighboring Babembe tribe, this created a durable resentment between the two groups.62

  After the war, in a political context that was favorable to them, the Banyamulenge expanded, some moving southward to Moba and Kalemie in northern Katanga, others going down into the Ruzizi plain, where a few became chiefs among the local Barundi through gifts of cattle, and still others going to work in the provincial capital, Bukavu, or in the mushrooming gold boomtown of Uvira.63 They supplied meat and milk to the creuseurs, the illegal but not clandestine gold diggers, and made a fair living. Unlike the North Kivu Banyarwanda, they had only a very small educated elite and no political connections in Kinshasa.

  But the bloody memories of 1965 remained between them and the autochthon tribes, feeding an ever present low-key tension. Even their name, Banyamulenge, meaning “those from Mulenge” (a village on the Itombwe), was chosen in the early 1970s to avoid being called Banyarwanda and being seen as “foreigners.”64 Thus when the war broke out in Rwanda, the RPF recruited widely among the Banyamulenge, who went to fight there not so much for the liberation of a country that was not theirs but to gain military experience and to acquire weapons for a possible future showdown with their neighbors. But going to fight in another country’s civil war was an ambiguous course of action for people who wanted to be accepted as Zairians. One of them, Monsignor Gapangwa, bishop of Uvira, was not able to decide clearly which were the various motivations of the young RPF volunteers, mentioning “a search for security, opportunism, joining as mercenaries or perhaps finding a solution to their questions about nationality.”65 Like the Banyarwanda in North Kivu, they followed the ebb and flow of the legal status of Kinyarwanda speakers in Zaire. But their position was different. In North Kivu the Banyarwanda were 40 percent of a population of about 2.8 million, and getting rid of them belonged to the domain of wishful thinking. But in South Kivu the Banyamulenge were a much smaller proportion of the population and understandably felt more threatened. Their enemies tried to belittle them by saying that there were no more than 35,000 of them,66 while they themselves tried through dubious computation to push their numbers up to 400,000.67 The reality stands probably at around 60,000 to 80,000, a very small fraction (3 to 4 percent) of the approximately 2.4 million people who live in South Kivu. Thus, here as in North Kivu, the question of the relationship between Kinyarwanda speakers and their autochthon neighbors remained a barely healed sore that the arrival of the refugees was going to reopen.

  The impact of the Rwandese refugees on the Kivus

  To understand exactly what happened, we must steer clear of two diverging ideological interpretations.68 The first, which we could call pro-RPF, sees the Kivus as an extension of Rwanda, where “extremists” and “negative forces,” inspired by Rwandese Hutu génocidaires (almost a pleonasm, in this view) are systematically pursuing the annihilation of the Banyamulenge and the North Kivu Tutsi.69 This interpretation, which was supported internationally by pro-Tutsi NGOs and the U.S. government,70 tends to overlook RPA human rights abuses in the Congo, or at least excuse them as simply unfortunate collateral damage in a war of self-defense. The second interpretation, which we could call anti-RPF,71 sees the autochthon tribes as innocent lambs savagely attacked by evil Banyarwanda predators. This view, which is mostly supported by the Kivu civil society and its European NGO allies, used to dwell on the evil of the Hutu in 1994–1996 but has since shifted its ideological aim to the evil of the Tutsi while trying to hide or minimize violent racist outbursts on the part of the Kivu tribes as “legitimate anger.” The reality is of course much more complex, as I have tried to show. The struggle for land in North Kivu, the memories of the civil war in the South, the problems of citizenship and of dual loyalty on the part of the Kinyarwanda speakers, general poverty, overpopulation, the collapse of the Zairian state, the ambitions of politicians who kept manipulating local feelings and contradictions for their own benefit—everything converged into making the Kivus a dangerous and volatile environment, with many links to the radical evil of the genocide that had just happened in Rwanda. It was obvious that the irrupt
ion of masses of desperate and belligerent refugees in that environment would create a major problem, fraught with potentially devastating consequences for the future. Beyond simplified ideological explanations, we can now look at the situation created by this refugee exodus.

  The first point to remember is that the numbers involved were huge: during July and August 1994 a total of 850,000 refugees arrived in the Goma area of North Kivu, another 332,000 moved to Bukavu in South Kivu, and another 62,000 came to the Uvira zone, where 255,000 Burundese had also been settled since the 1993 explosion in Burundi. The second point to keep in mind is the peculiar nature of their “refugee” character. Although the great mass of the people were indeed genuine refugees, they were manipulated by a highly ideological political and military leadership that had no intention of sitting idle in Zaire while waiting for the “humanitarian emergency” to resolve itself. The refugees were seen by their leaders as a political trump card that could be used to manipulate the international community, seduce President Mobutu, and threaten the new government in Kigali. The third point is that the whole thing was extremely expensive. Figures are hard to compute given the intricacy of bilateral, multilateral, and NGO financing, but the figures in the table give an idea of the order of magnitude.