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

by Gerard Prunier

  Actually, the whole process was much more typical of the workings of modern transnational capitalism than of the old style imperialism proposed by conspiracy theorists. Yes, the Americans were involved in the fall of Mobutu; no, it was not to control mineral riches, which they could have acquired perfectly well in a simpler way—unless one wants to believe that President Clinton started a war in order to enable former President Bush to steal something he already owned.157

  This does not mean that there were no backdoor dealings with the AFDL during the war. But those started only after the rebels had taken the mining capital Lubumbashi and when it looked increasingly likely that the Mobutu regime would not last much longer.158 The first to come was Nicholas Davenport, head of De Beers in Kinshasa, who flew to Lubumbashi together with some Union Miniére officials and a World Bank representative on April 18, 1997. But Boulle, true to his freelance approach, had been in contact with the rebels for the previous three months. He even facilitated similar contacts between Tenke Mining and the AFDL when the rebel forces started marching south.159 On April 16 Boulle signed the mythical contract, and, according to experts, the up-front cash he put up at the time was less than $40 million. But he also loaned his private Learjet to both Kabila and his “minister of mines,” Mawampanga Mwana Nanga,160 something that was of precious help during the war. The unconfirmed rumor is that he made $400 million two years later when he resold his rights to Anglo-American.

  By early May a real charter flight of twenty-eight financiers (representatives of Morgan Grenfell, First Boston Bank, Goldman Sachs, and others) flew into Lubumbashi when it became obvious that Kabila was going to be the next president.161 But it looked more like a case of bandwagon jumping than like a long-standing conspiracy.

  By then the AFDL had acquired new confidence. No more ultimatums and empty declarations, but actions: in Mbuji-Mayi the Lebanese diamond traders were asked for $960,000 in back taxes,162 and De Beers had a nasty shock when it tried to take “its” latest diamond consignment, worth about $3 million, out of town: Mawampanga told the South African team that they must tender for the lot. When De Beers protested that it would be a breach of their purchase agreement the “minister” replied that they themselves were in breach of contract since they had been buying diamonds from black market dealers who had mined them illegally on MIBA concessions. The De Beers men asked for orders from headquarters and were told to pay $5 million. They also did not contradict Mawampanga’s subsequent public statement when he said that the South African company would break with the Mobutu regime and deal with the AFDL in the future.163 A few weeks later Mobutu had become a memory. And, according to the mining expert Jean-Claude Sémama, who as a Frenchman should not be too soft on the “Anglo-Saxons,” “The political change did not seem to have benefited U.S. mining companies. There are two possibilities: either they tried to appropriate resources but did not manage to do it; or else the media image is simply false.”164 But once the mining myths were disposed of there remained another major problem concerning the war: What had happened to those refugees who did not return to Rwanda in mid-November 1996?

  The fate of the refugees.

  As we have seen, refugees fleeing the RPA-AFDL attacks in the Kivus mainly took two directions: a very large group of perhaps 350,000 started to walk toward Lobutu, while a smaller group of around 150,000 went down toward Shabunda. A very small group of mainly former ex-FAR and Interahamwe walked up to the Sudan.165 And, perhaps most neglected of all, many (perhaps nearly 100,000) remained in the Kivus, hiding in the forests on the foothills of the Virunga or trying to take refuge among native Congolese Hutu populations around Masisi and Walikale. In the words of Jean-Hervé Bradol, then a member of the Médecins Sans Frontieres mission in Kigali, “They had entered the Bermuda Triangle of politics.”166 Those who suffered the most in the short run were the ones who remained in the east without being able to get to the refuge of the forests. The RPA-AFDL troops were hunting them down like rabbits and were helped in their task by Mayi Mayi guerrillas, especially in North Kivu, where they were simply carrying on with the pre–Rwandese genocide Masisi war that had restarted with the arrival of the Rwandese refugees in the summer of 1994.167 This resulted in, among other acts of violence, the Nyamitaba massacre of November 21, 1996, when several thousand refugees were killed by the Mayi Mayi168

  The main group of refugees started to walk toward Walikale after the Interahamwe had killed several of their numbers to deter them from turning east.169 They managed to reach the Lobutu area, where they settled in two huge camps, at Amisi and Tingi-Tingi, with a combined population of at least 170,000 people. Another 100,000 refugees were rolled in a southwesterly direction toward Shabunda. Both groups eventually settled, in very difficult circumstances. The frenzied refugees number game slowly died down after the MNF idea collapsed. The Americans had continued to underestimate the refugees left in Zaire,170 while the French kept clamoring that “an international force . . . is more than ever necessary.”171 But everybody knew that the MNF issue was dead. The Rwandese government even gloated about it, Patrick Mazimpaka declaring, “Those refugees about whom the international community is so worried, we found them, we repatriated them. And the international community should reward us for all the savings we allowed it to make.”172 In typical U.S. newspaper clipped style, an American journalist gave a simplified summary of the situation: “In the last two months a rebel force few had ever heard of before solved a problem in Eastern Zaire that the United Nations and western democracies could not solve in two years. Using quick and decisive military strikes (and, some unconfirmed reports claim, leaving piles of massacred bodies) the rebels have separated hundreds of thousands of Rwandese refugees from the Hutu militants who have virtually held them hostage and have sent the refugees packing back to Rwanda.”173

  End of story. At least, that was the story sold in the media. But at about the same time things started to move for the refugees in Tanzania. After the mid-November exodus Tanzanian President Ben Mkapa had declared to the press, “Repatriation of refugees now seems more feasible,”174 and the Tanzanian Permanent Secretary to the Ministry of Home Affairs Colonel Magere had told UNHCR, “Following the mass return of Rwandese refugees from Eastern Zaire and the developments which have taken place, the Rwandese refugees in Tanzania have no longer any legitimate reason to continue to refuse to return to Rwanda.”175 The refugees knew it and they immediately started to flee the camps, walking to Rusumo, hoping to get anywhere—Kenya, Malawi, Zambia—except to Rwanda.176 But the pressure for repatriation was tremendous. UNHCR issued a joint communiqué with the Tanzanian authorities telling the refugees to return. The Tanzanian army was brought in; two Italian priests from Rulenge Parish who were telling the refugees that they had a right to decide whether to go or stay were arrested and deported to Italy, and UNHCR vehicles were used to truck the refugees back over the border. By the end of December practically all the Tanzanian refugees were home; there had been 3,200 arrests at the border and only two deaths.177 The Tanzanian repatriation was considered again as “voluntary,” which reinforced the notion that whoever was running away from the Rwandese army in Zaire must be a génocidaire.

  The refugees who settled around Lobutu and Shabunda enjoyed a precarious respite of a few months while the Alliance was sorting itself out militarily and politically and consolidating its eastern base. But by early 1997 the AFDL and the Rwandese forces started to move west. The first attack occurred at Katshunga, forty-five kilometers northeast of Shabunda, on February 4 and sent 30,000 to 40,000 refugees fleeing toward Pangi.178 Most of them were overtaken by the AFDL-RPA forces, which were moving faster than the refugees because they had vehicles and the road was passable. The attackers immediately separated out the young males and “shots were heard.” The survivors were then herded back toward Bukavu.179 Those who managed to escape started on a long southwest trek that eventually took them all the way to the Angolan border by way of Kindu, Lusambo, Mbuji-Mayi, and Luiza.180 But they w
ere harassed all the way, at times by innovative tactics. Because the refugees were hungry and exhausted, the Alliance forces would let the humanitarians take the lead, and when the refugees heard that some help had arrived, they would come out of the forest. Then the military, who were not far behind, would pounce and kill them. An RPA commander told a worker with Médecins Sans Frontiéres, “All those who are in the forest have to be considered enemies.” Some of the humanitarian NGOs chose to withdraw, fearing that their help was in fact lethal for the refugees.

  Meanwhile the Alliance was also attacking around Lobutu; the sharp fighting on the Osso River had mostly been carried out by Rwandese on both sides. In Kigali U.S. Ambassador Gribben, as usual strongly in support of the attackers, made short shrift of the humanitarian quandary, writing in a cable to Washington, “We should pull out of Tingi-Tingi and stop feeding the killers who will run away to look for other sustenance, leaving their hostages behind… . If we do not we will be trading the children in Tingi-Tingi for the children who will be killed and orphaned in Rwanda.”181 Such a message implied that helping the refugee camps would somehow sustain some kind of war effort by the ex-FAR against Rwanda itself. In fact, the truth was almost completely the opposite. The fleeing ex-FAR had stuck with the refugees, partly out of simple solidarity, partly to use them as human shields and reservoirs of humanitarian aid, partly because they wanted to keep their constituency. But when it became obvious that the situation of the camps might end in absolute collapse they started to think about alternative strategies. Gen. Augustin Bizimungu then flew back to Kinshasa,182 while Col. Gratien Kabirigi reorganized his best fighting troops (about five thousand men) and turned them around, going back eastward through the forest to avoid both the fleeing refugees and their pursuers.183

  The humanitarian situation in Tingi-Tingi and Amisi was terrible, with thirty to thirty-five people dying every day, mostly small children. By early February a cholera epidemic had started and the humanitarian workers were short of just about everything.184 On February 14 Sadako Ogata, the UNHCR high commissioner, begged the Alliance forces not to attack Tingi-Tingi; when they finally did anyway most of the refugees fled toward Kisangani and others ran into the jungle. From that moment on what had been a very difficult humanitarian situation became a particularly atrocious one.185 To make everything more confusing, this was the moment when Father Balas’s testimony about the killings in Kivu was released. This testimony, which was vital, was ill timed in terms of helping the public understand the situation because it did not refer to what was then happening near Kisangani but to events that had happened four months earlier in Kivu. Public opinion was in upheaval, with pro-Hutu and pro-Tutsi hurtling accusations at each other in perfectly bad faith.186

  By late February the former Tingi-Tingi refugees had resettled somehow in a string of camps stretching over a hundred kilometers south of Kisangani, from Ubundu to Lula by way of Kasese and Biaro. There were between 120,000 and 140,000 people, in a state of total dereliction. Some Médecins Sans Frontières doctors said of their charges that they were “not so much patients but rather pre-cadavers.” In March a representative of the World Food Program who flew over the Kisangani-Tingi-Tingi road said it was “strewn with hundreds of bodies lying along the roadside, apparently dead or dying.”187 At the same time, at the other end of the trail, bedraggled survivors of the November camp fighting who had been hiding in Kivu since then finally emerged starving from the bush; on April 1 20,000 to 30,000 refugees appeared out of nowhere at a village called Karuba at the northern end of Lake Kivu. They were former residents of Katale and Kahindo who had fled to the bush when the artillery fire started coming down and who were finally surrendering.188 They (and others later) were careful to wait until some UNHCR personnel were around so as to avoid being killed upon emerging.

  Around Kisangani the main humanitarian problem was the same that had already turned working in Tingi-Tingi into a nightmare, that is, constant and capricious restriction of access to the refugees. Expatriate staff were not permitted to stay overnight in the camps and were forced to commute every day over very bad roads, reducing their working time to absurdly short hours. Mortality kept rising; the health status of the refugees was described as “catastrophic” by Médecins Sans Frontiéres. Permission to airlift the worst cases was denied. Even airlifting unaccompanied sick children was refused by the Alliance189 under the pretext that there were still ex-FAR in the area.190 On April 17 about two hundred Rwandese soldiers were flown into Kisangani and immediately replaced the Katangese Tigers who had so far escorted the humanitarian workers. They were part of a killer team dispatched directly from Kigali. They were reputed to carry in their knapsacks small cobbler’s hammers which they used to silently and efficiently smash skulls.191 First they did a bit of psychological preparation, telling the local population that the six villagers who had been killed near the Kasese camp on April 20 had been killed by the refugees. This enabled them to recruit some of the villagers, whom they needed for the large burying job they were about to carry out. Then, on April 21, after the humanitarian workers were forbidden to come, they hit the camps. Humanitarian workers were not allowed back in until the 25th, and by that time the camps were totally empty. There had been about 85,000 people between Kasese 1, Kasese 2, and Biaro camps.192

  Reports of these abominable events were filtering out, but only weakly.193 Belgian Secretary of State for Cooperation Reginald Moreels gave an interview in which he talked about the massacres, but without giving sufficient detail.194 There seemed to be a great reluctance (Bradol’s “Bermuda Triangle of politics”) at officially condemning what the RPA-Alliance forces were doing for a variety of reasons: first, as Patrick Mazimpaka had pointed out, the West was relieved that the blocked refugee situation was finally “taken care of,” albeit in atrocious fashion; second, given the fact that the refugee problem was in itself a product of the 1994 horror, there seems to have been an unspoken compact among the various Western actors195 not to prevent the Rwandese from carrying out their revenge since it was the West’s lack of reaction during the genocide that had made it possible in the first place. But this compact was not universal, and some elements of the international community were trying to react, albeit in a state of complete and contradictory disarray.196 The first so-called Garreton Report on the human rights situation in Zaire was a step in the right direction,197 but it caused violent protests from the Alliance-Rwandese forces, which did not appreciate its severe judgment. The situation became even more conflictual when Garreton was sent back to Zaire in April as the situation was worsening. He flew to Kigali only to be refused entry by the AFDL for being “partial.” He was not able to carry out his mission and had to fly back to New York after hearing what he could from those who dared to approach him. Still, some of his conclusions can stand for the whole horrible episode:

  One cannot of course ignore the presence of persons guilty of genocide, soldiers and militia members, among the refugees… . It is nevertheless unacceptable to claim that more than one million people, including large numbers of children, should be collectively designated as persons guilty of genocide and liable to execution without trial… . The accounts heard or read by the mission show that most of the acts of violence attributed to AFDL were carried out against refugees inside the camps… . Very often the targets were neither Interahamwe combatants nor soldiers of the former FAR. They were women, children, the wounded, the sick, the dying and the elderly, and the attacks seem to have had no precise military objective. Often the massacres were carried out after the militia members and former FAR soldiers had begun to retreat.198

  This Rwandese tragedy played out in the Congo was going to have grave diplomatic consequences for Laurent-Désiré Kabila after he finally assumed power. But before closing off the topic, I have one last remark about the casualty figures, which have been such a bone of contention. In its final tally of the situation the UNHCR decided on the following figures:199

  Number of Rwandese refugees
in eastern Zaire (Sept. 1996):

  1,100,000

  Repatriations from eastern Zaire:

  Spontaneous:

  600,000

  By land:

  180,000

  By air:

  54,000

  Total:

  834,000

  Refugees located in Angola:

  2,500

  In Congo-Brazzaville:

  20,000

  In the Central African Republic:

  3,800

  Total:

  26,300

  Refugees remaining in Zaire:

  26,300

  Total refugees location unknown:

  213,400

  Over 200,000 refugees had thus officially disappeared. But even this already large figure rests on the assumption that there were really 600,000 returnees in November 1996. And as we have seen, that is likely to have been an optimistic evaluation. So if we consider the fact that UNHCR admits to another 35,500 Burundian refugees with a “location unknown” status, which is a polite way of saying they are dead, and if we take into account the very probable exaggeration of returnee numbers, the total refugee death toll should be considered to be around 300,000, an estimate that UNHCR High Commissioner Ogata considered possible.200

  5

  LOSING THE REAL PEACE (MAY 1997–AUGUST 1998)1