manhunt conducted by the Rwandan army assisted by AFDL units.55 There is no need to speak of a “humanitarian hysteria”56 to describe the concern of the international community in order to recognize the self-destructive consequences of the disastrous policies pursued by Hutu extremists and their Zairian allies. There can be no denying that Hutu leaders did organize armed raids into Rwanda. Nor is there any doubt about the diverting of humanitarian aid, presumably to use it as a bargaining chip with civilian refugees, or about Mobutu's role in arming refugee factions. Just how serious were the risks thus posed to the host country became clear after the Rwandan military began a series of attacks against the camps on October 18, 1996.
Arming refugees is one thing; disarming them is a far more difficult undertaking. Thus, if Laurent Kabila found it expedient in 1998 to arm Hutu refugees in his campaign against the Rwanda-backed RCD, disarming them after they have outlived their strategic usefulness was only one of the many headaches facing his son on the eve of multiparty elections.
As the foregoing suggests, strategies based on a calculus of short-term advantages may entail heavy costs a few years down the road. Similarly, alliances that seem perfectly logical one day may turn out to be utterly counterproductive the next. Kagame's experience with his less than obedient AFDL ally is just one example of the inherent fickleness of political clients.
A Fluid Landscape
The 1997 anti-Mobutist crusade did little more than replace one dictatorship with another. The 1998 war, on the other hand, marks a sharp break in the region's history. It ushered in one of the bloodiest wars recorded in recent times, the effects of which are still tragically visible in many parts of the country. Much of this violence, however, has gone un-reported. It unfolds not along a well delineated battle line but in oil-slick patterns in separate provincial and subprovincial arenas. Each episode has its own logic, each its own set of actors. And each is in some measure traceable to the reversal of alliances that followed in the wake of Kabila's decision to challenge the overrule of his Tutsi “protectors.”
SWITCHING SIDES
How the Kigali-sponsored attack against the refugee camps morphed into a full-scale, externally supported invasion aimed at the overthrow of Mobutu is beyond the scope of this discussion.57 Suffice it to note that the AFDL could not have reached Kinshasa with such speed and relative ease without the critical support it received from Rwanda, Uganda, and Angola. Their shared dislike of Mobutu, based on a realistic assessment of their respective interests, proved a fragile glue in the face of the new challenges raised by Kabila's revolt. Thus, if Angola remained Kinshasa's most trusted ally and would soon be joined by Zimbabwe—whose mercenary motives are well established—Rwanda and Uganda needed little prodding to turn against their renegade client. And whereas Burundi displayed, in Lanotte's felicitous phrasing, a “tolerant complicity” in the fight against Kabila,58 eventually Namibia, the Sudan, and Chad all joined Luanda in giving their half-hearted support to Kinshasa.
Basically, the old axiom according to which “the enemy of my enemy is my friend, and the friend of my enemy is my enemy” provides the essential logic behind the making and unmaking of alliances during the 1997 and 1998 wars.59 This applies to interstate as well as domestic alliances. In each case, the pattern is one in which friends and enemies reverse roles in response to their changing perceptions of the other's motives. What Rwanda and Uganda saw as a betrayal, Angola perceived as a legitimate move. Uganda's support of the Sudan People's Liberation Army (SPLA) made it a matter of realpolitik for Khartoum to join hands with Kinshasa. In sending 2,000 troops to battle against Jean-Pierre Bemba's Mouvement de Libération du Congo (MLC), Chad felt obligated to heed Khartoum's prodding (until the cost in human losses proved too onerous) in recognition of Sudan's past support to Idriss Déby.
Of the many alliances of convenience formed during the first and second Congo wars, none seemed more durable than the one between Rwanda and Uganda. Both viewed Mobutu with equal distaste, and both saw treasonable behavior in Kabila's volte-face. Yet by late 1999, the alliance had all but disintegrated. The bitter infighting that erupted in Kisangani in 1999, 2000, and 2001 over access to the rich mineral deposits of Bafwasende and other localities did more than spell the end of a friendly relationship. It brought the former allies to the brink of a full-scale war.
Much the same switching of partnerships can be seen at the domestic level. As long as he owed fealty to his Rwandan patrons, Kabila thought nothing of covering up the murder of Hutu refugees; nor did he shrink from blocking the work of the UN forensic investigation team in 1997, even demanding the sacking of its president, Roberto Garreton, “after he produced a preliminary sixteen-page report that identified forty sites where Kabila's AFDL was suspected of having committed atrocities.”60 A year later, however, those refugee leaders who managed to survive the carnage had become Kabila's best friends in his fight against Rwandan “rebels.”
The mixed fortunes of the RCD are another case in point. Despite or because of its murky origins—having been conceived, created, nurtured, and supported by Kigali to defend its interests in eastern Congo—the RCD today is a weak version of its former self. In addition to its core constituency, made up of Banyamulenge,61 it was able at first to recruit a number of influential politicians of different ethnic and provincial origins. Many have since left the movement. The old-guard Mobutists, Lunda Bululu, Lambert Mende, and Alexis Thambwe have joined other parties; Arthur Zaidi Ngoma has founded his own political formation (Camp de la Patrie), and so has Mbusa Nyamwezi (Forces du Renouveau); Wamba dia Wamba—whose early defection led to the first of many RCD clones, the so-called RCD-Kisangani, as distinct from the RCD-Goma—has resumed his academic career after a calamitous series of setbacks. What is left of the party is something of an empty shell, with its formal head, Azarias Ruberwa, in Kinshasa, desperately trying to stem the tide of dissidence.
Of the many defections suffered by the RDC, perhaps the least expected was initiated by a leading Banyamulenge personality, Manasse Ruhimbika. His short-lived Forces Républicaines Fédéralistes (FRF) made plain the lack of internal cohesion among the Banyamulenge and the growing resentment harbored by many, including Ruhimbika, over Rwanda's dominance of the RCD. The FRF break-away was only the harbinger of a far more serious split, which by 2002 had turned into a full-scale rebellion against the RCD. Led by a former RCD commander, Patrick Masunzu, the insurrection quickly spread through the Itombwe plateau, the traditional homeland of the Banyamulenge, and for a while reportedly coalesced with the FDLR and Mai-Mai elements. Masunzu's forces fought pitched battles against RCD troops, resulting in heavy losses on both sides, and some 40,000 displaced.62 Reflecting on the lessons of the insurrection, one observer commented, “Banyamulege opinion is now profoundly divided. Some still back the RCD; many feel it has abandoned their interests.”63 This is as true today as it was in 2003.
Thus, if fragmentation is indeed the most salient characteristic of Congolese politics, this is in part due to the persistence of highly divisive issues, having to do with disagreements over the extent and legitimacy of the Rwandan connection, the sharing of resources among allies, the choice of tactical alliances, and so forth. But this is only one aspect of a more complicated reality.
PATTERNS OF FRAGMENTATION
The political vacuum created by the sudden collapse of the Mobutist state must be seen as a key factor in the rapid fragmentation of the political arena. On the debris of the state, overnight a host of civil society organizations and militias mushroomed, of which the Mai-Mai militias are the most notorious for their propensity to fragment and proliferate. Pinning them down long enough to analyze their contours is not easy. Nonetheless, the political dynamic behind the surge of armed factions seems reasonably clear.64
Unlike what can observed in the case of the RCD, where fragmentation starts at the top, the efflorescence of Mai-Mai factions is a locally rooted phenomenon. It stems in part from the Mobutist legacy of playing one ethnic community off against another—sometimes referred to as the “géopolitique”
