salience of the Hutu-Tutsi dichotomy to be a permanent feature of the political landscape.
The danger of Hutu-Tutsi reductionism is all the more evident when one considers the ethnographic composition of the region. Rwanda and Burundi, after all, are not the only countries whose ethnic maps reveal the existence of Hutu and Tutsi. Kinyarwanda or Kirundi-speaking people also are found in eastern Congo, southern Uganda, and western Tanzania. According to the best estimates, the region claims roughly twelve million people speaking Kinyarwanda, and nearly twenty million if Kirundi, a language closely related to Kinyarwanda, is included. Although many migrated from Rwanda and Burundi to neighboring states during and after the colonial period, the presence of Hutu and Tutsi in eastern Congo reaches back to precolonial times. What needs to be underscored is that migrations from Rwanda or Burundi did not occur only one time but were staggered through the centuries. Length of residence, ecology, and history have shaped identities in ways that defy simple categorizations such as Hutu and Tutsi. It is with reason, therefore, that some analysts, in coming to grips with the politics of North Kivu, insist on drawing distinctions between the Hutu from Bwisha, those from Rutshuru, and those from Masisi and Kalehe.17
Map 2. Areas of confrontation in North Kivu. Reproduced from F. Reyntjens and S. Marysse, eds., L' Afriquw des Grands Lacs: Annuaire 2005–2006 (Paris: L'Harmattan, 2006), p. 37, by permission of the editors.
Such nuances went largely unnoticed by Belgian civil servants. In its effort to make more “legible,” the complex ethnic configurations of the region, the colonial state contributed significantly to formalizing and legitimizing the Hutu-Tutsi polarity. Thus the 1959 census figures for what was then the Kivu region (now divided into North and South Kivu and Maniema) designate, oddly enough, as “Bantous Hamites” and “Hamites,” respectively, the 184,089 Hutu and 53,233 Tutsi registered, thus making the “Banyarwanda” the third-largest group after the Banande (390,704) and the Bashi (382, 572).18 Time and again historians have drawn attention to the perverse effects of the colonizer's recourse to Hamitic and Bantu labels, as if to impose its own normative construction on Hutu and Tutsi.19 Equally striking is the phenomenon described by James Scott in Seeing Like a State, namely, “the state's attempt to make a society legible, to arrange the population in ways that simplified the classic state functions.” These state simplifications, he adds, “did not successfully represent the actual activity of the society they depicted, nor were they intended to…they were maps that, when allied with state power, would enable much of the reality they depicted to be remade.”20
Nonetheless, such attempts at remaking social realities did not obliterate all distinctions within Hutu and Tutsi. Nor did they erase the persistent tension between them, on the one hand, and the so-called “native” Congolese on the other. Many Kinyarwanda speakers, Hutu and Tutsi, trace their families' origins to precolonial times and have every right to claim the status of Congolese citizens. This is certainly true of the Banyamulenge (“the people from Mulenge”) established in the high plateau area of South Kivu since the nineteenth century if not earlier,21 and of the Hutu of Bwisha, many of whom lived in this area long before the onset of colonial rule, whereas others came as agricultural laborers from Rwanda in the 1930s to supply the European plantocracy of North Kivu with cheap labor. In short, as social categories, the terms “Hutu” and “Tutsi” cannot do justice to the registers of historical experience that lie at the heart of these many subloyalties.
THE CASE OF THE BANYAMULENGE
A rather unique case of ethnogenesis, the Banyamulenge are a perfect example of how geography, history, and politics combine to create a new set of identities within the larger Banyarwanda cultural frame.22 Heavily concentrated on the high-lying plateau of the Itombwe region of South Kivu, estimates of their numerical importance vary wildly. The figure of 400,000 cited by the late Joseph Muembo is grossly inflated.23 A more reliable figure would be between 50,000 and 70,000. Their name derives from the locality (Mulenge) whence they are said to originate. The term, however, has been the source of much controversy because it became increasingly used in the late 1990s as an omnibus label to designate all Tutsi living in North and South Kivu. It came into usage in 1976 as a result of the efforts of the late Gisaro Muhoza, a member of parliament from South Kivu, to regroup the Banyamulenge populations of Mwenga, Fizi, and Uvira territories into a single administrative entity. Although his initiative failed, the name stuck, and by 1996 it was often used by ethnic Tutsi and Congolese to designate all Tutsi residents of North and South Kivu.
Though much of their history is shrouded in mystery, most historians would agree that the Banyamulenge are descendants of Tutsi pastoralists who migrated from Rwanda some time in the nineteenth century, long before the advent of colonial rule (a fact vehemently contested, however, by many Congolese intellectuals). They are culturally and socially distinct from the long-established ethnic Tutsi of North Kivu and the Tutsi refugees of the 1959–62 Rwanda revolution. Many do not speak Kinyarwanda, and those who do, speak it differently. Their political awakening can be traced back to the eastern Congo rebellion of 1964–65. Many initially joined the insurgency, only to switch sides when they saw their cattle being slaughtered to feed the insurgents. Their contribution to the counterinsurgency did not go unnoticed in Kinshasa. Many were rewarded with lucrative positions in the provincial capital, and more and more of their children flocked to missionary schools. From a primarily rural, isolated, backward community, the Banyamulenge would soon become increasingly aware of themselves as a political force.
It is impossible to tell how many joined the Rwanda Patriotic Front (RPF) in the 1990s. What is beyond doubt is that they formed the bulk of Kabila's Alliance des Forces Démocratiques pour la Libération du Congo (AFDL) in 1996 and after the fall of Bukavu, they filled most of the administrative positions vacated by the Congolese, some of whom were ardent Mobutists. Equally clear is that they suffered very heavy losses during the anti-Mobutist crusade, as well as during the 1998 crisis when many were massacred by Kabila's supporters in Kinshasa and Lubumbashi. It was said that in the late 1990s, Bukavu claimed a larger number of Banyamulenge widows than any other town in the region.
The creation in mid-1998 of the Rassemblement Congolais pour la Démocratie (RCD), under Rwandan sponsorship, was meant to provide the Banyamulenge with a vehicle for the defense of their interests—and of anyone else's who cared to join—but as it evolved into an instrument for the defense of Rwandan interests in eastern Congo, many felt aggrieved and alienated. The feeling that they have been “instrumentalized” by Kagame is shared widely among them. As we shall see, this sense of grievance against Rwanda is in large part responsible for the internal strains and divisions suffered by the community. All this, however, does not detract from the fact that as a group, the Banyamulenge are profoundly aware of their cultural distinctiveness. Few people have been dealt a harsher blow by history: many argue, with justice, that they have been twice victimized; first by Kagame, who used them as cannon fodder for the defense of Rwanda's strategic interests; and then by Congolese extremists, as happened in Bukavu in June 2004 when in the wake of an abortive Banyamulenge-led coup, hundreds perished. Today Bukavu is virtually “free” of Banyamulenge.
OTHERNESS
What defines the “other” as an ally or an enemy? Several objective criteria come to mind: language (e.g., “Rwandophones” or Kinyawanda speakers), country of origin (“Banyarwanda”), place of settlement (“Banyamulenge,” the people of Mulenge), ethnicity (Hutu and Tutsi), to which must be added morphologie, or body maps, a reference that increasingly crops up in newspaper articles in Goma and Bukavu. None of the above, however, tells us why one set of criteria should be more relevant than another at any given time: why, for example, Hutu and Tutsi generally were lumped together as the embodiment of a “Banyarwanda” menace from 1963 until 1994, when the label quickly dissolved into Hutu-Tutsi enmities; or why the term Banyamulenge, as distinct from Tutsi, carried such threatening overtones among other communities from 1996 onward.
Otherness has less to do with objective identity markers than with the perceived threats posed by one community to another. Whether real or imaginary, such threats do not materialize out of thin air. They are intimately related to changes in the national and regional