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

by Gerard Prunier

  The Forces Armées Congolaises were supposed to have 120,000 men at the time of the peace agreement, but a large number of those were “ghost” soldiers who existed on paper only to enable the officers to collect their salaries from the Ministry of Defense. The RCD-G had perhaps 40,000 men, the MLC around 20,000, and the RCD-ML about 8,000. In addition, the eastern militia groups collectively known as Mayi Mayi, who ranged from the Ituri all the way down to northern Katanga, were an unknown quantity which regrouped anywhere between 50,000 and 300,000 combatants, depending on the criteria used to define that word.117

  The FARDC were set up to fuse the various armed groups into one national army. But the various warlords tended to keep some troops outside FARDC as a kind of military insurance policy, or else to “incorporate” their men in name only, putting them into homogeneous “former rebel” units where the ex-belligerents retained their separate command structures, even within the supposedly “unified” forces. Hence the all-important role of mixage and brassage, the two steps of the integration of the former troops into supposedly homogeneous units. To this day, mixage (when troops of various ethnic origins serve in the same units) and brassage (when troops are stationed in parts of the country well away from their ethnic homeland) are still far from completed, in the east particularly, and many units of FARDC are “national” in name only.

  The goal was for FARDC to eventually number about 120,000, even though military experts (particularly the South Africans) believed that a tighter, more professional, and better paid force of 60,000 to 70,000 would probably be better adapted to the country’s needs. But the main problem remained that of the ghost soldiers. Payments were allocated for 240,000 men, a figure that was obviously cooked.118 But the foreign military experts had problems deflating these numbers because their Congolese counterparts systematically manipulated the figures, cooked the raw data, and tampered with the computer programs. The government was spending $8 million monthly on soldiers’ salaries and at least half of that might have been embezzled. On the official scale junior officers were paid $50 per month and privates $10, a powerful incentive for officers to embezzle and for soldiers to loot. The officer corps was still very confused. At the top levels former FAZ generals who had taken twenty years to reach that rank and had studied in foreign military academies had to cohabit with Mayi Mayi “generals” who were much less educated (if at all) and who had been promoted from bush fighter to brigadier in one year. Various foreign military missions and training teams were trying to put some kind of order and professionalism into all this.119

  But a measure of the incoherence of the foreign efforts at military restructuring is evident when one considers that the 10,600 MONUC soldiers, who had only limited rules of engagement, cost roughly $1 billion per year, while the 120,000 to 150,000 Congolese soldiers of FARDC had a budget of less than $150 million, even when taking into account all the cheating around ghost soldiers. Army living conditions120 and salaries were so poor that discipline was and will remain a problem for the foreseeable future.

  With about nine thousand men the national police was supposed to be the elite of the police force. In spite of having been trained by the Angolans, the EU, and the French it had remained poorly disciplined and without direction. On November 21, 2006, when a gaggle of fewer than two hundred pro-Bemba demonstrators entered by force the Supreme Court to burn it down,121 the Police d’Intervention Rapide122 ran away because some of the demonstrators had guns and were firing in the air. Not only did they abandon their equipment (batons, shields, helmets), they even stripped down to their underwear for fear of being recognized by the population and lynched.123 As for the 29,000 members of the territorial forces, they were supposed to have been trained by MONUC and by South Africa, but in fact fewer than 5,000 had been. Their salaries were often stolen by their superiors, they are underprofessionalized, and they could not be relied upon to keep law and order. The result was that civil disturbances were usually handled by the FARDC or by special units like the Groupe Spécial de la Sécurité Présidentielle (GSSP), which tended to use disproportionate force and cause massive casualties.

  To understand the security problems of the transition, one has to play them on that background. Then they suddenly appear much more understandable and their confused handling much more excusable.

  If we forget for a moment the Kigali-induced late 2004 crisis examined earlier, what were the recurrent security problems during the transition period?

  • In terms of sheer violence, the Ituri situation. Following the efficient but short-lived French intervention in 2003,124 MONUC had moved in with about four thousand men and aerial support. It took its Chapter VII seriously and adopted an offensive stance, slowly wearing down the militias. By early 2005 they were sufficiently weakened to allow the arrest of several of the leaders to take place.125 From then on the situation started to wind down, even if some holdouts remained, such as “Peter Kerim” in the Mahagi area. Over ten thousand militiamen had been disarmed in the Ituri over the past twelve months. But the problem of DDRRR was that the last two “Rs”—reinstallation and reinsertion—were usually missing. At about $50 the demobilization package offered to individual soldiers was insufficient to start any kind of business or even to go back to farming. As a demobilized and not reinserted militiaman told the UN, “It is very difficult for us to survive without the guns. The movement was our income. Now they have to give us jobs.”126 But there were no jobs. So what often happened is that militiamen would get demobilized to get the $50 and a bit of food, and then buy or steal another gun, or dig up one they had buried, and go back to the bush.

  • Another area of insecurity was eastern Katanga, where the mystical Mayi Mayi commander known as Gédéon terrorized the Dubie-Kato-Kilwa area.127 MONUC being overstretched, he was hunted down by one of the first “integrated” FARDC brigades, which behaved with ruthless disdain for human rights and caused almost as much havoc as the man they were supposed to capture.

  • A ghost from the past resurfaced in the east: the old Ugandan ADF guerrilla unit. MONUC hunted it down and, in cooperation with FARDC and the UPDF lying in wait on the order side of the border, was able to kill eighty-six ADF combatants in late December 2005.128

  • A bit in the same way as Rwanda before, but somewhat less dangerously, Uganda tried one last time to sponsor a “rebel” movement in the east. In June 2005, duly steered by operatives of ESO, a number of Congolese guerrilla remnants129 met at the Crested Crane Hotel in Jinja and given birth to the Mouvement Révolutionnaire Congolais (MRC). In a rare display of boldness, President Museveni declared on March 19,2006, “If the LRA attacks any part of Uganda we shall follow them into the Congo, with or without approval.”130 On May 25 MONUC, FARDC, RPA, and UPDF officers met in Kinshasa to try to defuse the multiple tensions centered around the LRA presence in Garamba, the ADF skirmishes around the foothills of the Ruwenzori, and the fairly large-scale operations of the Ugandan-sponsored MRC all the way from the Ituri down to the Semliki. The atmosphere was tense, with strong Congolese hostility toward the Ugandans, who were put in the position of having to justify themselves and did not manage to do it very convincingly.131 A few days later FARDC traded shots with infiltrated Ugandan elements in Garamba;132 this was the pathetic ending of a miserable attempt, as Bosco Taganda himself soon started to negotiate with the Congolese authorities.

  • Last but not least, there was the problem of the FDLR. This last incarnation of the old Rwandese génocidaire forces was an altogether different element. The FDLR was a well-organized, nearly conventional army with ranks, paperwork, leaves of absence, and pay for its soldiers.133 Kigali did not know how to deal with it. The government had failed to uproot it, did not really want to use it as an excuse for offensive operations in the eastern Congo anymore, but could not tolerate it on its borders for fear that one day it might develop into a real threat if something went wrong inside Rwanda itself.134 So both MONUC and the Congolese regime were asked to deal with it, something they
were neither capable of nor really willing to do since it would have meant a large-scale military operation in difficult circumstances and with a doubtful outcome. Non-Banyarwanda FARDC officers had lingering sympathies for a group that had been their ally against Rwanda during the war, and MONUC did not really have the stomach for a large military offensive that seemed a bit much even for their Chapter VII mandate.135

  So the transition lived on, with its troubled former battlefield areas that it was never able to fully pacify. There were two Congos: the former government territory, which grumbled and complained but lived roughly in peace, and the former war zone, which wondered at times if the war had really ended.

  The elections

  By early 2006 election fever had started to grip the Congo. The fighting in the east, the security problems, the last-ditch attempts at destabilization were all—rightly—perceived as remnants of the past. But the looming future was both full of hope (perhaps too much; the elections had turned into a Holy Grail) and full of threats. All efforts converged at successfully organizing an enormously difficult exercise. MONUC was both carefully deploying its troops to minimize security risks and preparing to organize a massive logistical operatiori.136 In April the European Union contributed a $21 million auxiliary military force of two thousand men under a Franco-German coordinated command. Its main point of deployment was Kinshasa itself (MONUC was mostly busy in the east), where the Bemba-Kabila rivalry had created a tense climate. In a much more discreet way, Angola was also preparing for problems at election time, and it had carefully deployed two brigades along its border with the western DRC.137 Dozens of new “political parties” were springing up (parliamentary elections were due to be held at the same time as the presidential); these were parties in name only since they were mostly tribal or regional gatherings around the name of one or two well-known local politicians. The president had his own party, of course, the Parti Pour la Reconstruction et le Développement (PPRD); but it was not built on the model of “presidential parties” elsewhere in Africa. It was a fairly modest thing, and its secretary-general, Vital Kamerhe, was only one of the president’s men.138 There were thirty-three presidential candidates, seven of them considered to be “serious”: Joseph Kabila, Jean-Pierre Bemba, Oscar Kashala, Pierre Pay Pay, Antoine Gizenga, Azarias Ruberwa, and Diomi Ndongala.139 But everybody knew that, barring a last-minute surprise, the basic contest was between Kabila and Bemba. So both of them tried to gather momentum and attend to the small but vital elements that would make a decisive difference. This in itself was a momentous shift in the political landscape; albeit in a crude way, the politics of democratic seduction had entered the picture. There was still gunplay in the east, there were all kinds of rumors of plots and destabilization, and the former actors of the war all kept fingering their weapons. But Africa’s core quasi-continental “country” had entered a new world, that of rough but democratic politics.

  This was immediately evident in the choice of the candidates’ public relations maneuvers. On June 17 transitional president Joseph Kabila married his long-time girlfriend, Marie-Olive Lembe da Sita. This was a masterful tactical move since Kabila’s support came from the east, and the west resented a man seen as representing “the ignorant Swahilophones.”140 “Madame Olive,” as she quickly became known, was a Mukongo. She was also goodlooking, politically savvy, and a good public speaker who quickly became a tremendous asset for her husband during the campaign and helped him minimize his losses in the western DRC. Socially, though Kabila was not even a Roman Catholic and though he and Lembe da Sita had been “living in sin” for the past six years (they had a five-year-old son), he nevertheless got married in a lavish Roman Catholic ceremony blessed by Kinshasa’s Cardinal Frédéric Etsou, a key Catholic power broker. To make it more ecumenical, President Kabila, a free-thinker but nominal Anglican, had his marriage blessed by a Protestant bishop as well.

  Both candidates pulled out all the stops; every argument, including unfair ones, was used. Bemba’s slogan was “Vote Mwana Mboka,” meaning “Vote for a native son,” this to play on the rumors of Kabila’s supposedly “foreign” (i.e., Rwandese) ancestry. The cannibalism practiced by Bemba’s troops in Ituri during the war resurfaced as an electoral issue. Each candidate flexed his muscles, just this side of organized violence. Kabila let the men around him (not himself, he never did it himself) discreetly “remind” the electorate that the Maison Militaire Présidentielle (Presidential Military Establishment), not in theory but in practice, gave orders to all the civilian and military secret services and to the fifteen-thousand-strong GSSP, the only semi-serious military outfit in the country. As for Bemba, he used his popularity in the capital to the hilt and even a bit further. On July 27, three days before the polls, he organized a huge meeting of 100,000 supporters at the Tata Rafael Stadium in Kinshasa. The crowds were ecstatic and they trashed the Haute Autorité des Médias building to punish the commission for what they considered the media’s unfair coverage of their candidate’s campaign. When the police tried to interfere, Bemba’s guards opened fire, killing six. Demonstrators burned down a bar, raped a journalist, and ransacked a church.141

  On July 30 eighteen million voters (out of the twenty-five million registered) went to the polls. Then the suspense started; in that huge country, it would take weeks to know the results. Meanwhile, the pressure mounted, so when the results were finally known on August 20, things came to a head. First the PPRD stalwarts, frustrated at the fact that their candidate had not won an absolute majority in the first round (Kabila had 44.81 percent of the vote to Bemba’s 20.03 percent), tried to push the Electoral Commission to “rectify” the results so that the president would be immediately declared a winner.142 They also sent the GSSP to try to prevent Malu Malu from announcing the results. Soon armed clashes broke out in front of Bemba’s CCTV station, causing six fatalities; the next day, August 21, tanks were out in the streets, with the PPRD diehards hoping to provoke a level of violence sufficient to justify a cancellation of the election in order to proclaim Kabila the winner. The CIAT ambassadors rushed to Bemba’s residence to try to get him to publicly accept the score and quiet things down, but as they were inside the house, GSSP forces arrived on the scene and opened fire. The whole thing nearly ended up in a diplomats’ holocaust as the house was riddled with bullets and Bemba’s helicopter set on fire.143 Kabila managed to rein in his overzealous followers and save the diplomats. There was still some light fighting on the 22nd, but Bemba and Kabila met on the 23rd and the tension was defused. All in all, twenty-three people had been killed and forty-three wounded.

  The main surprise in the election results was how well veteran Lumumbist politician Antoine Gizenga had done. With 13 percent of the vote he came third behind Kabila and Bemba and thus assumed a kingmaker position. His bastion was Bandundu, where he got more votes than all the other candidates put together, and both frontrunners’ camps courted him. Nzanga Mobutu, one of the late dictator’s sons, had managed to get 5 percent of the vote by coming in second in Equateur behind Bemba. Another result of note (which was not a real surprise) was the collapse of the RCD-G and its candidate, Azarias Ruberwa. The man who was identified with the hated Rwandese invaders got barely 2 percent of the vote. But the really worrying thing was the deep division in the country: a Swahili-speaking east and south solidly behind Kabila and a Lingala-speaking west and north (almost) solidly behind Bemba. If one calculated only demographics, Kabila’s victory was unavoidable, as “his” regions simply had more people than Bemba’s, But the problem came from Bemba’s popularity in the capital and the neighboring areas, which gave him a massive nuisance capacity.

  As for the legislative elections, the five-hundred-member Parliament inaugurated on September 22, 2006, had two positive characteristics: it was reasonably representative of the state of public opinion in the Congo and it was freely elected. It also has two negative characteristics: it did not have a UDPS representation and it was incredibly fragmented.

  Fi
rst there were two (relatively) big chunks of MPs directly linked to the two presidential candidates: the PPRD had won 111 seats (about 21 percent) and Jean-Pierre Bemba’s MLC had obtained 64 seats (around 13 percent). But it was significant that the two main presidential contenders had, between them, managed to attract only 34 percent of the electorate, while a majority of the MPs either belonged to small or even tiny parties or to no party at all.144

  Then there were eleven medium to small parties who had MPs. By order of diminishing importance, there was the Parti d’Action Lumumbiste Unifié (PALU), Gizenga’s party, which had 34 seats (7 percent), mostly won in Bandundu, where PALU got 80 percent of the vote. PALU was part of the new Alliance pour la Majorité Présidentielle (AMP) coalition, Kabila’s new anti-Bemba loose coalition. The Mouvement Social pour le Renouveau, led by Pierre Lumbi, who was not a presidential candidate, had 27 seats and represented about 5 percent of Parliament. It was also a member of Kabila’s AMP. The Forces du Renouveau, Mbusa Nyamwisi’s party, had 26 seats, and the Rassemblement Congolais pour la Démocratie (RCD), led by former vice president and rebel leader Azarias Ruberwa, had gained only 15 seats (3 percent). The various Christian Democratic parties together got 37 seats, but being divided among four different organizations they failed to exercise the influence they could have had if united. The Union des Démocrates Mobutistes (UDEMO), led by Nzanga Mobutu, won 9 seats in Equateur Province, where it represented a minority vote among the Bangala, competing with J. P. Bemba’s MLC for the same tribal vote. The last of the “real” parties, the Union Nationale des Fédéralistes du Congo (UNAFEC), led by rabble-rousing Katangese politician Kyungu wa Kumwanza, was the direct heir to the old Mobutu-era UNAFER anti-Balubakat party. It won only 7 seats but nevertheless represented a real danger in Katangese politics because of the demagogic and violent populist tactics of its leader. UNAFEC was a member of the presidential AMP, but a fairly unpredictable one.