CONVENTIONAL wisdom in Goma and Bukavu has it that Kivu Province in eastern Congo is where losers in Rwanda traditionally end up, and it is from Kivu that they prepare to return to power in Rwanda.1 A civil society activist in Bukavu explained to me the long-term effect of Rwandan conflicts that tend to “spill over into our country”: “These ethnic conflicts are cyclic with each ethnic group taking turn in power and misfortune. The fate of one today is the fate of the other tomorrow. The consequence for us are the refugees of the conflict. Another consequence of cyclical fortunes is that when they return, not everyone returns, some remain. Those who remain become Congolese.”2
The RPF victory set off a massive exodus of Hutu from Rwanda. Like everything else about Rwanda those days, the rate and extent of the exodus was without precedent: over two million crossed Rwanda’s borders in a week, dividing roughly between Congo and Tanzania. Over a million spilled over into Congo, mainly into North and South Kivu, a region that hosted most of the Kinyarwanda-speaking population in Congo. As they crossed the Congo-Rwanda border in mid-1994, the million-plus refugees literally brought the trauma of postgenocide Rwanda to the region of Kivu. The impact was volcanic, and its effects have yet to ebb. The escalating crisis in Rwanda introduced a double tension in Kivu, both external and internal, both a tension between Kivu and the power in Rwanda and a tension within Kivu society. This tension grew in intensity as the Kinyarwanda-speaking refugee and exile population in Kivu grew in size, increasing the weight of refugees and exiles while blurring the distinction between them and earlier immigrants. In turn, this fed the tendency on the part of many “indigenous” Congolese to refuse to distinguish between Kinyarwanda-speaking Congolese and the mix of refugees and exiles from Rwanda. If the 1990 RPF invasion of Rwanda from Uganda was born of the confluence of citizenship crisis on both sides of the Uganda-Rwanda border, the RPF invasion of Congo in 1996 needs to be understood as an outcome of a similar confluence, except that this time it joined the citizenship crisis on the Rwanda and the Congo sides of the border. To explain this is the first purpose of the chapter.
The Banyarwanda of Congo, who then numbered fewer than a million, were already in the political limelight by the early 1990s. Historically, the Congolese Banyarwanda were divided between those in North Kivu and the rest in South Kivu, the former mainly Hutu and the latter mainly Tutsi. There was also a political difference between the two parts of Kivu. North Kivu had been home to a long-simmering citizenship crisis, stemming from the fact that the Banyarwanda of Masisi, previously recognized as indigenous, had been systematically disenfranchised over the three decades beginning on the eve of independence. In South Kivu, though, the crisis of citizenship had been late in coming, and did not really surface until the expectation of elections in the early 1990s. When they organized to defend their citizenship rights, the Banyarwanda of Congo tended to come together on a linguistic and regional basis, seldom as Hutu or Tutsi. The shift in political identity from Kinyarwanda speakers of a particular locale to Hutu and Tutsi across different locales is a distinct development of the 1990s. To understand that development is a second purpose of this chapter.
Not surprisingly, when the RPF connected with Banyarwanda across the border, it was with those in South Kivu, and not with Kinyarwanda speakers in North Kivu, in spite of the fact that the citizenship crisis in the north had been brewing for much longer than in the south. This was proof enough that the RPF had already begun to think of itself as Tutsi and the power in Rwanda as Hutu, for the simple fact about Kivu was that the Banyarwanda of the south were predominantly Tutsi, whereas those of the north were predominantly Hutu. But for anyone taking a longer view of things, there was a great historical irony in this. We shall later see that the Tutsi influx into South Kivu had been prompted by two late-nineteenth-century developments: the centralization of kinship in Rwanda at the expense of independent Tutsi aristocratic families, and the bitter factional struggle in the Tutsi elite on the death of Rwabugiri. The Tutsi who left did so voluntarily. For them, Rwanda was no subject for a romantic construction of home. Rather, it signified the suffocating tentacles of a centralizing power, something from which to stay away. This negative historical memory was reinforced by later developments. Since the 1959 Revolution, and especially since the 1972 massacres in Burundi, the Tutsi of South Kivu had made great attempts to distance themselves from the explosive world of Hutu and Tutsi in Rwanda and Burundi, instead seeking to define their place in the ethnic kaleidoscope called Congo. Like their counterparts in Uganda, the Tutsi in Congo also attempted to change from a descent-based to a territorially based political identity: from being Banyarwanda, those ancestrally belonging to Rwanda, they had tried to become Banyamulenge, those living in the hills of Mulenge in South Kivu. But this attempt boomeranged, as had the Ugandan reform, testifying to the regional character of the crisis of citizenship.
Finally, this chapter should allow the reader to compare the nature of political identity in Congo with that in Uganda, and contrast both with the nature of political identity in Rwanda. The difference arises from the history of state formation in the colonial period. The Congolese state—like the Ugandan state—distinguished between those Banyarwanda who had been on Congolese territory when it was first colonized and those who came later. The former were considered nationals, but the latter were divided into colonial-eramigrants and postcolonial refugees. The nationals were presumed to have a right to their own Native Authority, whereas migrants and refugees were regarded as ethnic strangers and were denied the group right to a “customary” home and a Native Authority. The divide between ethnic and civic dimensions of citizenship obtained in Congo and Uganda, but it did not in Rwanda. Whereas the Native Authority in Uganda and Congo was ethnically defined, the Native Authority in Rwanda was simply demarcated territorially, as a local authority, but without an ethnic identity. This difference, of course, stemmed from the defining feature of the Rwandan state, that it recognized only races, and not ethnic groups, as political identities.
To understand the impact of the million-plus refugees who streamed into North and South Kivu from postgenocide Rwanda, we need to sketch the contours of ethnic relations in North and South Kivu on the eve of the genocide. As in Uganda, the Banyarwanda of Congo comprised three distinct groups: nationals, migrants, and refugees. Nationals could claim the greatest historical depth: they were already resident in the territory that Belgian colonialism demarcated as Congo in the late nineteenth century. Migrants crossed the border at different times during the colonial era, either voluntarily in search of a livelihood or under compulsion. Refugees, in contrast, were wholly a postindependence phenomenon. They testified to the mercurial instability of postcolonial politics in the region.Whereas refugees were part of a volatile political diaspora, nationals and migrants were part of a more stable cultural diaspora. Before the great overflow of Hutu refugees in 1994, nationals and migrants far outnumbered refugees. But after 1994 the relationship was reversed. As the numbers of refugees began to exceed those of nationals and migrants, the political diaspora came to dominate and define the life circumstances of the cultural diaspora.
The predicament of the Banyarwanda in Congo flowed directly from the political arrangement put in place by Belgian colonialism in the colonial period. As elsewhere in colonial Africa, the law distinguished between the indigenous (natives) and the nonindigenous (nonnatives). But this is where the similarity between Rwanda and Congo ended. Whereas political identities in Rwanda tended to correspond to those under direct-rule colonialism (such as in preapartheid South Africa), those in Congo were more characteristic of indirect-rule colonialism. We have seen that group identity in Rwanda was racialized for all: the Tutsi as Hamites and the Hutu as Bantu. In contrast, group identity in Congo was both racialized and ethnicized: unlike in Rwanda where the majority defined as indigenous was pressed into a racialized identity called Bantu, single and homogenized, this majority in Congo was further divided into multiple ethnicities, each with its own “customary” home, “customary” law, and a “customary” authority to enforce it.
The Banyarwanda of Congo thus fell between the stools of the bifurcated world created by indirect-rule colonialism. Racially considered the same as native Congolese—“Africans”—those who had come to Congo after its conquest by Belgians were set apart ethnically as being nonindigenous to Congo. This historical circumstance left them without a claim to a Native Authority in Congo. Without an ethnic patch of their own on Congolese soil, they were treated as ethnic strangers in every Native Authority. This dilemma became even more acute after independence.
The world of the racialized citizen and the ethnicized native changed after independence. All postindependence regimes were determined, to one degree or another, to do away with the stigma of race that they associated with colonial rule. The tendency of the postcolonial state was to deracialize civic identity. Civic citizenship ceased to recognize any difference based on race or place of origin. That is where similarities ended and differences sprouted among different kinds of post-colonial reform agendas. The conservative variant of the postcolonial state—to which belong the experiences of Congo and Uganda—continued to reproduce the native identity as ethnic. The irony was that deracialization without deethnicization continued to reproduce a bifurcated citizenship since not every civic citizen, a citizen of the state of Congo, could claim ethnic membership of a Native Authority. Even if the civic sphere ceased to make a distinction between citizens who were indigenous and those who were not, the ethnic sphere continued to reproduce this distinction. To be recognized as ethnically indigenous meant to have an ethnic home (a “Native Area”) governed by an ethnic administration (a “Native Administration”). To understand the practical significance of being a civic but not an ethnic citizen, it is worth exploring further the distinction between civic and ethnic dimensions of citizenship.
Civic citizenship is a consequence of membership of the central state. Both the qualifications for citizenship and the rights that are its entitlement are specified in the constitution. Under deracialized civic law, these rights are mainly individual and are located in the political and civic domain. In contrast, ethnic citizenship is a result of membership in the Native Authority. It is the source of a different category of rights, mainly social and economic. Further, these rights are not accessed individually but by virtue of group membership, the group being the ethnic community. The key socioeconomic right is the right to use land as a source of livelihood. Herein lies the material basis of ethnic belonging, particularly for the ethnic poor. The immediate practical consequence of being defined a citizen of nonindigenous origin is this: nonindigenous citizens are denied “customary” access to land since they do not have their own Native Authority. To access land in “customary” areas, they are compelled to pay tribute to “customary” authorities in these areas. To understand why “nonindigenous” citizens in rural areas should persistently call for a Native Authority of their own, we need to begin with a fuller understanding of the Native Authority in Kivu.
The Native Authority in Kivu is three tiered. At the lowest level is the chief of the locality. Then comes the second-level chief, the chef de groupement, and then finally the mwami of the collectivité. Those considered nonindigenous and living in rural areas may, and usually do, have a chief of the lowest order from among their own ranks, one who is answerable to the higher authority for their immediate governance. Only those considered indigenous, however, have the right to a chief of the second and third tier from one of their own. The distinction is crucial for customary power really rests at the level of the chef de groupement and the mwami. They have the power to confirm ethnic belonging and to issue identity cards, oversee administration, allocate customary land for livelihood, hold tribunals through which customary justice is meted out, run local markets, and so on.3
To understand the growing dilemma of the Kinyarwanda-speaking minority in Kivu, we need to understand their changing relationship to the Native Authority system. That nationals could claim a precolonial tie with the land, migrants a connection that went to the colonial period, and refugeesonly a postcolonial link turned out to be a fact of not just historical but also great political significance. The key term differentiating civic from ethnic citizens, indigeneity, was defined wholly in reference to the colonial experience: the indigenous were those who could demonstrate a tie to the land predating colonial occupation. This distinction also divided the Banyarwanda residents of Congo into two: those considered indigenous to Congo and those not. For reasons of history, the balance between the two groups was different in North from South Kivu.
The Kinyarwanda-speaking minority in Kivu consists of three territorially distinct groups, the Banyarutshuru and the Banyamasisi in the north, and the Banyamulenge in the south. As we shall see, the Banyarutshuru and the Banyamasisi are predominantly Hutu, whereas the Banyamulenge are Tutsi. Furthermore, because their presence on Congolese soil preceded its demarcation as a Belgian colony in the 1880s, the Hutu of Rutshuru are considered “indigenous” and so are entitled to a Native Authority of their own.4Matters, however, are not as straightforward with the Banyamasisi of North Kivu or the Banyamulenge of South Kivu, the two groups on whom we shall focus.
THE BANYAMASISI OF NORTH KIVU
The neat division between nationals of precolonial vintage and colonial migrants, the former indigenous and the latter not, tended to break down in Congo because two entire groups seemed to fall between the cracks. The first of these were the Banyamasisi of North Kivu, and the second were the Banyamulenge of South Kivu. The claim of the Banyamasisi to Congolese citizenship became a bone of contention in the decades that followed independence. The contention stemmed from the difference between two types of colonial migrations, separating “labor migrants” of an earlier period (1926 to 1937) from those “transplanted” after 1937.
Labor migrants were a response to the arrival of Belgian settlers in Kivu. The Kivu highlands had fertile land, a mild climate, and no tse-tse flies, an ideal location for white settlement. To prepare the highlands for just that fate, the fundamental law of 1908 made a distinction between “indigenous” and “state” land: all land actually cultivated by “natives” was declared “indigenous” and the rest was declared “vacant” and seized by the state. It did not matter that this “vacant” land was actually used for grazing, or hunting or foraging, as was the forest around Masisi by the local Bahunde people. Then, companies—three of them—were granted land in millions of hectares. These generous grants were periodically reviewed and reduced if the company concerned did not show sufficient evidence of putting the land to use. After two such reviews, a single company, CNKI, was granted the overall authority to parcel out land to individual white settlers as and when they arrived and asked for land. From 58 in 1928, the number of settler families increased to 330 in 1949.5
The corollary of white settlement was a labor problem. The local Bahunde were given to a hunting and foraging life. As the land was divided and the forest cleared, hunting and foraging entered a period of crisis. To force the Bahunde into an alternative, colonial authorities passed a 1917 ordinance that required them to grow food and crops. Following that, forced labor practices were “generalized in Masisi,” specifically between 1914 and 1928. The response of the Bahunde was to run from encroaching administrative authorities.6 In this context of incoming white settlers and fleeing locals, Belgium turned to the more thickly populated colony of Rwanda for labor. A decree of 19 July 1926 authorized Rwandans to seek employment freely outside their country and legally opened the country to labor recruiters from the outside. Three types of recruiters came into Rwanda: Union Minière, the mining conglomerate (1925–31); CNKI, the plantation oligopoly (1928–32); and individual settlers.
But this solution did not fully work. The hitch was that labor migrants had an option between Belgian-dominated Congo and the British colonies in East Africa. Figures told that the preference was for the latter: in 1931, for example, 6,869 men worked in Uganda and Tanganyika, but only 4,170 in Kivu.7 Thus began the era of “transplantation.” The idea was to go beyond encouraging the migration of individual laborers whose options were limited by the thought of returning one distant day. The new approach was to get entire families to move, with no thought of return. To achieve this goal, Belgian administration decided to manipulate the “push factors,” in particular a combination of overpopulation and famine: the more famine spread, the more transplantation could be put forth as a humanitarian response to a growing tragedy. Though the immigration was supposedly voluntary, Catharine Newbury’s research in Kinyaga suggests that a measure of force was involved: Gishari in Kivu, one respondent told her, is “the place where chiefs sent the people they didn’t like.”8 The immigrants were a trickle in 1937—only 691. The big increase was the result of the big Rwanda-wide famine (nyirahuku), which stimulated an increase from 8,492 in 1942 to 24,448 in 1945.9 Even though Gishari was declared saturated in 1945, the Mission d’Immigration Banyarwanda (MIB) was created in 1948 to transplant immigrants to other parts of Masisi. The population of Masisi territory increased threefold between 1936 and 1969, and rose further from 273,920 in 1970 to 482,007 in 1983.10
Whereas labor migrants were presumed to have a home away from where they were resident, those “transplanted” were not. The whole idea was that they were uprooted from home, and had to make another. In line with this thinking, the transplanted Banyarwanda were granted their own Native Authority in Gishari in Masisi. This made for a tension with the local population. To begin with, the Bahunde had been opposed to the influx because they saw it as increasing competition for land. Local chiefs, on the other hand, were happy to see more coming in: given that ethnic strangers would have to give the chief an extra payment in return for the temporary right to use land, every new immigrant meant an additional source of tribute. To escape that very tribute, post-1936 immigrants insisted on having their own Native Authority, which they got.11 This single development brought Bahunde peasants and Bahunde chiefs together on an “indigenous” basis. The more the numbers of the Bahunde shrunk in proportion to those of the transplanted—by 1990, the Bahunde were but 15 percent of the population of Masisi12—the more they asserted the one political right that the colonial state recognized as a native prerogative: ethnic belonging as custom. The Collectivité Gishari was established in 1938 and disestablished in 1957, just as the colonial power prepared to turn the corner to independence. This dissolution set the stage for the postindependence crisis of citizenship.
THE DIALECTIC BETWEEN ETHNIC AND CIVIC CONFLICT
When Collectivité Gishari ceased to be a Kinyarwanda-speaking Native Authority in 1957, the Banyarwanda in Masisi lost any ethnic space to express their political preference. With the coming independence of Congo, however, an alternate political space began to open up for natives. The opening of civic space was marked by oncoming provincial and municipal elections. Being the majority, the Banyarwanda in Masisi won the collegial elections of 1958. The response of the Bahunde elite was to use their ethnic prerogative as the population “indigenous” to the collectivité to hound most Banyarwanda from positions of influence in the local state: “They systematically removed Rwandan immigrants from important positions they held in local administration and maintained only a few loyalists in minor positions.” This practice became “common elsewhere in the former MIB zones by 1960,” so much so that “it became common to see, in each locality, only one Hunde family, that which had been brought in to rule over the Rwandan immigrants.”13 Outside the political sphere, the indigenous prerogative translated into an assertion of “customary” control over all land designated as “indigenous” so that only those immigrants paying tribute to “traditional” authorities were allowed to continue to till customary land. The widespread slogan udongo ya baba (literally, earth of the father, or fatherland) summed up the point of view that the Hunde were landowners and the immigrant Banyarwanda tenants. It is the acceleration of these trends that led to the eruption of armed conflict in Masisi in 1963–64. Called “La Guerre du Banyarwanda,” this conflict was in reality a popular Banyarwanda uprising against abuse by Bahunde chiefs.
So sensitive was the nationality status of the Kinyarwanda-speaking minority as independence approached that even the Roundtable Independence Conference in Brussels was unable to fix the juridical status of this minority. This was in spite of the fact that conference participants included representatives of the Banyarwanda from Masisi. The Fundamental Law left the citizenship status of the minority unresolved, stating that the Congolese people will themselves decide this issue. Even though Mobutu Sese Seko abrogated the Fundamental Law when he usurped power in 1965, the Brussels outcome came to introduce an element of insecurity in the juridical status of the Kinyarwanda-speaking minority.
In the three decades that stretched from the end of La Guerre du Bayarwanda in 1963–64 to the beginning of the Rwandan genocide in 1994, a complex of processes unfolded in Kivu. In retrospect, one can see how this dynamic produced the environment that incubated the post-1994 crisis of Kivu. The more they felt blocked at the local level, the more the Kinyarwanda-speaking minority looked to the civic sphere—both the market and the central state—for alternate strategies, economic as well as political. Unable to access land as did the “indigenous” Congolese, as a “customary” right, those with resources devoted them to purchasing as much land as possible through the market. Frustrated from exercising power locally by the ethnic character of the local authority, they made every effort to access positions at higher provincial and national levels, whether through elections or through connections. This, in turn, provoked a response from among the “indigenous” majority. Afraid that the Banyarwanda would use national representation to acquire power locally, “indigenous” Congolese came to oppose citizenship rights for them. When their citizenship was questioned and their right to run for office denied, the Kinyarwanda-speaking minority—particularly the Tutsi, a minority within the minority, without a “home” anywhere in the region, one they could count on as a fallback in times of crisis—developed a strategy of entry into organs of the state, particularly the security apparatus.
Three key decisions marked the course of the spiraling crisis of citizenship that fed the insecurity of the Kinyarwanda-speaking minority. Each had a vital impact on the future of this minority. The first was Mobutu’s 1972 Citizenship Decree. The second was the 1981 Citizenship Law passed by an elected Parliament, and the third was a resolution by the 1991 Sovereign National Conference upholding the provisions of the 1981 law. To understand the nature of the movement from 1972 to 1981 to 1991, we need to grasp the changing political context over these decades.
The 1972 Citizenship Decree
The context of the 1972 Citizenship Decree was the first major postindependence crisis of regional proportions that sent thousands of refugees streaming into Kivu. This development took place as the aftermath of the massacre of about 200,000 Hutu in Burundi in 1972. Faced with a growing refugee influx, the local population began to see themselves as an imperiled “indigenous” majority. This, in turn, made the position of the Kinyarwanda-speaking minority even more insecure. As the “indigenous” majority responded with pressure on the Kinyarwanda speakers, the minority looked to the central state for adequate protection. This was the context in which the Mobutu regime extended citizenship to those who had come as refugees from Rwanda in 1959–63. This measure was introduced as the 1972 Citizenship decree. Its effect, though, was to alarm the local majority who saw the decree as a direct outcome of growing Tutsi influence within the state apparatus. Many believed that Mobutu had signed the decree under the influence of Bisengimana, his chef de cabinet who was himself said to be a 1959 Tutsi refugee.
The following year, 1973, the Mobutu government passed the General Property Law. A measure similar to that passed by the Amin regime in Uganda, it nationalized all land, including both the land under the control of “traditional” authorities in the rural areas and land controlled by white settlers. While the state was unable to implement the provision with regard to rural land under “customary” control, it was able to transfer settler-controlled land to Zairean citizens. The result was to usher in “a newly formed class of rural Congolese capitalists.” This is the context in which the more prosperous among the Kinyarwanda-speaking population cashed in on their newly acquired civic citizenship to gain property rights. Soon, some of the biggest plantations in North Kivu passed into the hands of Banyarwanda and were turned into ranches. A study conducted in 1991 showed the long-term result in Masisi: 512 families—of which 502 were Banyarwanda—controlled about 58 percent of the available land. Of these 502 Banyarwanda families, most were Tutsi-Banyarwanda.14 To many in Kivu Province, the 1972 Citizenship Decree came to symbolize not simply an inclusive citizenship policy but one so undiscriminating that, if followed in practice, it would surely turn Kivu into an open sanctuary for the surplus population from Rwanda and Burundi. “What can’t be accepted,” a prominent civil-society leader concluded in a conversation on citizenship in 1997, “is an order whereby every immigrant who comes in is granted citizenship automatically—a practice that came in with Bisengimana becoming Chief of Staff to Mobutu.”
The 1981 Citizenship Law
It is not until the legislative elections of 1977 that the “indigenous” majority developed a strategy equal to countering the minority strategy of penetrating the security and party apparatus of the Mobutist party-state. The prospect of election brought home the realization that sheer numbers could be translated into political power, so that the majority could get access to power even if it was shut out of appointments in the state party, the Mouvement Populaire Révolutionnaire (MPR). The “indigenous” majority followed a single guideline: better not elect another Tutsi if you want to balance out against them. When one was elected—as was Gisaru, a Munyamulenge, as deputy of Uvira in South Kivu—the response of the local majority was to accuse him of having manipulated the election. Not surprisingly, the parliament that came out of the 1977 elections passed a new citizenship law hostile to the Kinyarwanda-speaking minority. The 1981 law was said to have been passed under strong pressure from Nande and Hunde politicians from North Kivu:15 it stipulated that only those persons who could demonstrate an ancestral connection to the population residing in 1885 in the territory then demarcated as Congo would qualify to be citizens of Congo.
It was one thing to pass the law, quite another to implement it. By the time of the 1985 provincial assembly elections, the question of citizenship was still unsettled, though the 1981 law remained on the books. In this context, the “indigenous” majority improvised a solution: the Kinyarwanda-speaking population may vote in the elections, but none of its members may run for office. The solution seemed to compound the problem; for the first time, all Kinyarwanda speakers were lumped together into a single group, regardless of how long different sections had been on Congolese soil. The response of the Kinyarwanda-speaking minority, particularly the Tutsi, was to smash ballot boxes. As a result, no provincial assemblies were elected in North or South Kivu.
The Sovereign National Conference (Conférence Nationale Souveraine, CNS) and the Resolution on Citizenship, 1991
The Sovereign National Conference marked a turning point in the political history of postcolonial Congo. Countrywide, the CNS heralded the coming together of an internal opposition to the Mobutist party-state. In Kivu, however, it had a double significance: it also marked an important step in the constitution of a self-conscious political majority. Though this majority transcended ethnic lines, it constituted more of an interethnic rather than a nonethnic majority, seeing itself as an “indigenous” majority threatened by a “nonindigenous” minority growing through the periodic influx of refugees.
The CNS took place at a time when the Banyarwanda minority was once again gripped by anxiety about their citizenship status. Following the RPF attack on Rwanda in October 1990, many young Tutsi in Kivu decided to cross the border into Uganda and join the RPF. The Mobutu regime responded with Mission d’Identification de Zaı¨rois au Kivu, authorized to carry out an on-the-ground verification of who among the Kinyarwanda speakers was Zairean and who was not—because their families had come after the Berlin Conference. As a result, many Hutu and Tutsi from 1936 were not verified as Zaireans. This in turn increased the flow of Tutsi youth crossing into Uganda to join the RPF. By the time the CNS met in 1991, citizenship had become a hot issue, particularly in the region of Kivu. Not surprisingly, the delegations from North and South Kivu urged the CNS to give priority to the citizenship issue. In response, the Haut Conseil de la République, an organ of the CNS, adopted the 1981 Citizenship Law.
To understand why a majority constituted through the democratic process would appear as a threat to the minority, we need to take a brief look at the history of the internal opposition. The history of organized peaceful opposition goes back to the formation of the Union pour la Démocratie et le Progrès Social (UDPS) in 1982. This formation was the result of several impulses, including calls for reform from thirteen parliamentarians in 1980. Whereas the tendency before 1980 was for opponents of the regime to flee into exile, the tendency after 1980 favored the growth of a peaceful internal opposition. Combined with the reform wave that followed the end of the Cold War, the gradual development of this internal opposition led to Mobutu’s “opening up” speech of 24 April 1990. In that speech, he promised political reform. As a first step, Mobutu promised to relinquish the presidency of the MPR and thereby ensure the separation of the party and the state. In the two weeks before he took that promise back, the idea of holding the Sovereign National Conference had caught the imagination of the political opposition. The CNS opened officially on 7 August 1991.
The proceedings of the CNS were televised throughout urban Congo. It was enough to inspire further initiatives. There was a mushrooming of civil society organizations, thickening the texture of the internal political opposition.16 The overall thrust of the CNS was to deepen and to coordinate the internal opposition to the Mobutist state. At the same time, the CNS impacted on the provinces in different ways. In the region of Kivu, it tended to crystallize two related trends, one in the “indigenous” majority, the other in the Kinyarwanda-speaking minority. It accelerated the majority tendency to differentiate Tutsi from Hutu and to lump together all Tutsi, regardless of the depth of their presence on Congolese soil, into a single group. The tendency to use the term “Banyamulenge” as a generic term for all Congolese Tutsi really gathered momentum with the CNS. Correspondingly, it was during the CNS that the Banyamulenge found out that, even though they had moved to Congo in the nineteenth century—much earlier than the post-1959 Tutsi immigrants to North Kivu—their situation was not very different from that of the Tutsi of Masisi and Goma in North Kivu.
The Sovereign National Conference brought several contradictory political tendencies to a head in Kivu. While the Kinyarwanda-speaking minority—particularly the Tutsi—continued to look to organs of the state party, including its security organs, for protection against the “indigenous” majority, the majority continued to invest in representative processes both as protection from the arbitrary rule of the party-state, and as guarantee that they would prevail against the minority. The very democracy that tended to create a majority across ethnic lines tended to pit a self-consciously “indigenous” majority against what many increasingly came to think of as a “nonindigenous” minority, one they saw as not only Kinyarwanda-speaking but also owing political allegiance to Rwanda. Within the Kinyarwanda-speaking minority, the Hutu began to differentiate themselves as “indigenous” from Tutsi as “nonindigenous.” It testified to a double development: the growth of politics of indigeneity in Congo, and growing Hutu/Tutsi tensions in the region. This is why it should not be surprising that the very fact that the CNS began to discuss the question of citizenship raised a fear in the minority—particularly the Tutsi—that it was about to lose its citizenship status. The CNS also marks the point after which the course of ethnic developments in South Kivu began to converge with those in North Kivu, particularly Masisi.
THE BANYAMULENGE OF SOUTH KIVU
The minority question in South Kivu is less complex than that in the north. Whereas the situation in North Kivu has a longer history and is intimately affected by what happens in Rwanda, the situation in South Kivu is of more recent origin but is influenced by developments in both Rwanda and Burundi. The long-standing immigrant minority in South Kivu originates more from Burundi than from Rwanda. North Kivu has always had a Banyarwanda Native Authority in Bwisha, and one in Gishari from 1938 to 1957, but South Kivu has never had a Banyarwanda Native Authority. Those living in South Kivu and Burundi have tended to shift back and forth between two adjacent valleys: the Imbo Valley in Burundi and the Ruzizi Valley in South Kivu. Today, the population of both valleys is Kirundi speaking. Thus, the Kirundi-speaking population in South Kivu is considered “indigenous” to the region. Like the Banyarwanda in Rutshuru in North Kivu, the Barundi in Ruzizi Valley have also had their own customary chief, in a collectiviténamed Barundi.17
The Banyamulenge are mainly Tutsi. It is said that their arrival in South Kivu dates to the 1880s, when Rwabugiri ruled the central kingdom of Rwanda. Two explanations are advanced for the movement of Tutsi away from the kingdom. The first relates to Rwabugiri’s determination to gather more tribute from the rich, the second to the bitter conflict of succession that took place at his death, an event named after the place where he was buried, called Rucunshu. In Rwanda and Burundi, an aging king does not publicly proclaim his successor. The result is a struggle for succession at the king’s death. When the conflict is particularly bloody, those who lose are compelled to move away. The two explanations do not rule out the possibility that both may be true.18
The claims about when and why the Banyamulenge moved are many and have multiplied as the political crisis has intensified. After the original migration of the Banyamulenge—whether during the reign of Rwabugiri or as an aftermath of the succession conflict at his death—there were successive migrations. Labor migrants followed in the colonial period. The impetus began with labor recruitment in Ruanda-Urundi by the Union Minière du Haut Katanga, which began in 1925 and continued to 1929. More than seven thousand workers are said to have been recruited in that five-year period. Government reports note the steady trickle of labor migrants from Rwanda to South Kivu, particularly Bukavu, from the 1930s.19 The next big influx was that of Tutsi refugees in 1959–60. Unlike the bulk of the Banyamulenge who lived on the high plains and were pastoralists, both the labor migrants of the colonial era and the 1959–60 refugees tended to live in urban areas or in refugee camps, such as Kalonge near the airport.
Unlike the Barundi, however, the Banyamulenge have never had their own Native Authority. Banyamulenge chiefs were confined to the first level, the chief of the locality. For access to land, they paid homage to existing chiefs where they settled. The area in which the Banyamulenge resided covered three territorial administrations: (a) the territoire of Mwenga inhabited by the Balega, (b) the territoire of Fizi inhabited by the Babemba, and (c) the territoire of Uvira inhabited by the Bavira and the Bafuliro. The territoireis a fourth level of administration, after the localité, thegroupement, and the collectivité; it comprises several collectivités. The territoire of Uvira thus comprises three collectivités, called the Bavira, the Barundi, and the Bafuliro. The name of each collectivité is taken from the name of the ethnicity considered “indigenous” to it.
Unlike in North Kivu, the Banyamulenge in South Kivu were seen as one among many ethnicities, and not as a nonindigenous minority set apart from the indigenous majority—at least until the holding of CNS in 1991 raised expectations about forthcoming elections. We need to recall that the politics of indigeneity in Masisi developed locally, in a context shaped by two factors: the macro factor of colonial legislation which made a clear distinction between those indigenous and those not, and the appropriation of land from the Bahunde whose very mode of livelihood was destroyed at the local level. In South Kivu, the Banyarwanda migration preceded the establishment of colonial authority and proceeded on the basis of mutuality between different groups. As the colonial period came to a close and the politics of Rwanda took on explosive dimensions, the Banyarwanda immigrants in the hills of Mulenge began to distance themselves from their ancestral world and define their identity and thus their future more in line with their new home. Thus was born the identity “Banyamulenge.”
There is no agreement on when the term “Banyamulenge” came into general usage and why. Some historians I spoke with in South Kivu said the triggering event was really the genocidal killing of Hutu in Burundi in 1972; after it, the Tutsi became very unpopular in the entire area, that is why the Congolese Tutsi began to distance themselves from Rwanda.20 To the colonial construction of political identity tied to ethnic origin, was counterposed a radically different notion of political identity, this time tied to territorial residence. To understand the nature of the shift, we need to begin with the understanding that the point of political identity is to claim (or to deny) political rights. From the point of view of an immigrant population stigmatized as “nonindigenous,” the fact that power is identified in ethnic terms—say, as Bafuliro—means that rights are also restricted to those who belong ethnically, in this instance to the Bafuliro, thereby disenfranchising all others considered immigrants. The claim to shift identity from the ethnic (the Banyarwanda) to the territorial (the Banyamulenge) must, in this context, be seen as an attempt to define a more inclusive basis of rights, based on residence rather than ethnicity.
This shift did not become contentious until after 1991, when the CNS passed the 1981 Citizenship Bill. Before that, group relations in South Kivu were defined more along ethnic lines, less along a divide defined by the notion of indigeneity. Unlike in Masisi, the politics of indigeneity in South Kivu was more of an import—at first from the central government and then from the democratic movement—than a local construction. The central government’s contribution was summed up as the principle of géopolitique, whereby the Mobutu government argued that “all positions of authority could only be awarded to those indigenous to the region concerned.”21 Faced with the prospect of democratic elections in 1991, local authorities launched a campaign to identify Zairean nationals. Local politicians, too, got converted in the face of elections as they also began to discover the political uses of indigeneity.
More than any other, one single shift marked this change in context. When Congolese Tutsi tried to distance themselves from the socially explosive world of Tutsi and Hutu in Rwanda and Burundi, this suggested a deeper and more sinister agenda to “indigenous” ears. Why otherwise, many asked, would immigrants seek to hide their “real” identity? It is precisely the creative side of the Tutsi initiative—that a change from a descent-based to a residence-based political identity would mean a radical shift in the definition of the subject of rights—which sounded sinister to “indigenous” ears. The consequence, an indigenous politician pointed out to his growing audience, would allow all local residents without exception, even those ethnically not indigenous, to claim rights. Later, the Bafuliro would point out that Mulenge is the name of the place, the groupement, where the Tutsi were first allowed to settle by the Bafuliro. In 1924 they had asked for permission from the colonial power to occupy the high plateau farther south. When permission was granted, they moved south, which is why some claim the Banyamulenge really arrived in 1924. The term may seem innocent on the surface, but the Bafuliro claim it is not, for it really sums up the Tutsi claim to “own” Mulenge, which actually was “owned” by Bafuliro. From this point of view, the Tutsi have developed a disturbing tendency to call themselves by the name of the place where they have settled. The more the Tutsi move, the more they seem to sprout place-based identities such as the Banya-tulambo, and Banya-minembwe, and so on. To “indigenous” ears, then, immigrant claim to a place-based identity really masks an immigrant strategy to lay claim to local land. Why else, many ask, would the Banyamulenge seek to distinguish themselves from the Banyarwanda, except to mask their history, the fact that they came from Rwanda? To tolerate this, they point out, is to encourage any Kinyarwanda-speaking person to follow the same strategy and claim to be a Munyamulenge, and thus a Congolese, no matter how shallow their presence on Congolese soil.
Indeed, the citizenship status of the Kinyarwanda-speaking minority was a matter of lively debate among civil society leaders in Kivu when I was there in 1997. Who should be a citizen and who not? Should the Mobutu/Bisengimana Decree of 1972, allowing all refugees of 1959 to become Congolese citizens, prevail? Or should there be an affirmation of the 1981 law, passed by Parliament and affirmed a decade later by the Sovereign National Conference, that only those with a proven connection to an ancestor resident in the territory demarcated as Congo in 1885 be verified as Congolese? Or should all those currently resident in Congo who pledge political allegiance exclusively to the Congolese state be considered Congolese—regardless of their parentage, place of birth, or duration of stay in Congo?
Interviews with civil society leaders in 1997 brought two strong convictions to light. The first related to civic citizenship, the second to ethnic citizenship. The more one pressed home the link between the mounting political crisis in Kivu and the citizenship question, the more civic leaders tended to agree that the more inclusive option may also be the more prudent. The other side to this growing consensus—that all those resident in Congo before the Rwanda genocide of 1994 be recognized as its citizens, meaning civiccitizens—was an equally firm consensus that ethnic citizenship must be restricted only to “indigenous” Congolese. While the first tendency was a source of hope, it is the “indigenous” consensus reflected in the second that gives real insight into the crisis of citizenship in contemporary Kivu.
TWO CROSS-CUTTING TENSIONS
For the decade and a half that stretched from the end of La Guerre du Banyarwanda to the Citizenship Law of 1981, the nationality conflict in North Kivu revolved around two pivots. The first pitted the “indigenous” majority against the Kinyarwanda-speaking minority, whether immigrant or not. We have seen that the more this tension grew, the more it tended to blur all historical distinctions among different groups of Banyarwanda: between immigrants and nonimmigrants, and between different groups that had come at different times. As a consequence, all Kinyarwanda speakers came to be considered nonindigenous. The second pivot of conflict was internal to the Banyarwanda; it pit Tutsi against Hutu. As the tension between Hutu and Tutsi increased in Rwanda, it also did in North Kivu. This is clear from the impact of the 1959 “social” revolution in Rwanda. As the group that came in 1959–63 began to organize to return to power in Rwanda, relations began to sour, both with the “indigenous” majority, and between Hutu and Tutsi in the Kinyarwanda-speaking minority. When articulated with the first, there was a tendency for the Hutu, who had either been there before the colonial period or came during its heyday, to claim an indigenous status against the Tutsi, most of whom arrived in North Kivu after 1959.
The shift from a mainly Banyarwanda immigrant identity to an identity highlighting the difference between Hutu and Tutsi is reflected in the breakup of Umoja, a common Banyarwanda organization, into two separate bodies, one Hutu, the other Tutsi, in the 1980s. Umoja was formed as a Banyarwanda organization in the aftermath of the 1981 Citizenship Law, which classified as noncitizen all Banyarwanda who came to Congo after its colonial boundaries were drawn up in 1885. Its second president, Senzeyi Ryamukuru, claimed that Umoja was formed at the behest of Bisukero, the first president of the Provincial Assembly of Kivu.22 After a consultation between Bisukero and Ryamukuru, two young men, one a Hutu (Sekimonyo Cosmos) and the other a Tutsi (Munyamakuo David), were given the task of bringing local Hutu and Tutsi together in a single organization. Umoja was born as an organization of all Congolese Banyarwanda from Goma, Rutshuru, and Masisi. Sekimonyo became its first president in 1983. In 1985, Sekimonyo (a Hutu) became the president of the Regional Assembly, and Ryamukuru (a Tutsi) became president of Umoja. In another few years, however, Umoja was no more.
Umoja disintegrated in 1988 and was replaced by separate Hutu and Tutsi organizations. With the direct financial support of President Habyarimana of Rwanda and the political support of President Mobutu of Congo, the Hutu in Rutshuru built links with the Hutu in Masisi and formed a common organization of Hutu in North Kivu called Maghrivi (Mutualité des Agriculteurs du Vironga). It was said that part of Mobutu’s electoral strategy was to identify “indigenous” Hutu through Maghrivi so as to grant them citizenship. The main message of Maghrivi was that there are no “indigenous” Tutsi in Congo. The proof, it was said, was the Native Authority, which was Hunde in Masisi and Hutu in Rutshuru. Maghrivi called for elections of all chiefs. It figured that an electoral strategy would both neutralize the Bahunde claim to be “indigenous” and translate the numerical majority of the Hutu into local political supremacy over both the Bahunde and the Tutsi. In response, the Tutsi leaders of Umoja founded SIDER (Syndicat d’Initiative pour le Développement de la Zone de Rutshuru) as an exclusively Tutsi organization. SIDER was later absorbed into the ADP, the Alliance Démocratique des Peuples; the difference, according to Sekimonyo, was that ADP was an organization of all Congolese Tutsi. By the middle of the 1990s, not only were Hutu and Tutsi organized across localities in Congo, but Hutu and Tutsi associations crossed state boundaries and began to function as regional networks.
Class and Ethnic Conflict in Masisi
At the heart of the conflict was the question of land. Pitting poor against rich Hutu from the outset, the land conflict soon turned into an ethnic confrontation between Hutu and Bahunde in Masisi. To understand the shift from one to the other, we need to focus yet again on the two ways of acquiring land under the system inherited from colonialism. One is through a market transaction, a way that by its very nature is open only to the well-off, those with means to register a preference on the market. The other is by asserting one’s “customary” right as a member of a Native Authority. This is the more political way, and it is the only one open to the poor. The land conflict in North Kivu began in Masisi in 1993 as a class conflict among the Hutu, and then turned into an ethnic conflictbetween the Hutu and the Bahunde over whether the former should have the right to their own Native Authority. At the outset, the Tutsi joined the “indigenous” Bahunde (and the Banyanga) against the Hutu. By the end of the year, however, as the conflict came to focus on the question of who was entitled to a customary right to land through a customary authority, it pitted the “indigenous” (the Bahunde and the Banyanga) against the “nonindigenous” (the Hutu and the Tutsi).
Masisi is an area with a Hutu immigrant majority, said to be around 75 percent of the population by the early 1990s. The conflict began when rich absentee Hutu (and Tutsi) landlords began taking over the lands of mostly poor Hutu (and some Bahunde) in Masisi. The displaced poor, said to be around one thousand, fled to Walikali, where they demanded the right to elect their own ethnic leaders. Since the Wanyanga held that this “customary” right could be exercised only by those indigenous to the soil, the claim led to a clash between the one thousand Hutu and the Wanyanga in Walikali. The poor one thousand then returned to Masisi, where they made the same “democratic” claim, except that this time they also had the backing of their richer kin, the rich Hutu, and the general Kinyarwanda-speaking population. The claim led to a conflict with the Bahunde in Masisi. According to leaders of the civil-society-based Peace Campaign in Goma, the emerging Hutu point of view was strongly shaped by the Hutu organization Maghrivi.23
The response of the Mobutist state to growing conflict in Masisi was to send in units of the DSP and the Garde Civile. Neither, however, was provided with means of sustenance. All were forced to live off the local population, which they did. The difference was that while the DSP lived off the more prosperous Hutu, the Guard Civile lived off both the Bahunde and the ordinary Hutu. The army ended up protecting the land claims of the “nonindigenous” (mainly the Hutu) against the “indigenous” (mainly the Bahunde) population, while the conflict grew into a bloody affair. When asked to give an idea of the intensity of the conflict, a Xavérien Father in Bukavu estimated that between 10,000 and 20,000 were killed, while some 200,000 Bahunde, Hutu, and Tutsi must have run away in the process.24 This was the context in which a million-plus refugees streamed from Rwanda into North and South Kivu.
POSTGENOCIDE RWANDA COMES TO KIVU
The numbers of Kinyarwanda-speaking people in Kivu Province exploded with the genocide of 1994 in Rwanda. In North Kivu, in particular, their arrival coincided with a rapidly escalating ethnic conflict. To understand fully their impact on the local situation, we need to bear in mind that the refugee question actually mediated a relationship between three different factors. These were: the local conflict in Kivu; the explosive dynamic of the Hutu/Tutsi conflict in Rwanda; and the pernicious role played by major foreign powers, particularly France and the UN.
The first set of refugees actually came before the Rwandan genocide began: from Burundi into South Kivu in late 1993 and early 1994. These were mainly Hutu fleeing the terror of the army after the assassination of Ndadaye in October 1993. They numbered roughly 50,000 and were unarmed. Not so, however, the million or so Hutu refugees who poured in from Rwanda in mid-1994. They lived in armed camps, controlled by the ex-FAR (the Rwandese National Army, or Forces Armées Rwandaises) and the Interahamwe, who continued to be supplied militarily by the French. The armed soldiers and militia were said to number some 20,000 in Bukavu and 30,000 to 40,000 in Goma. According to a local priest working with the Catholic Relief Services (Caritas) in Bukavu, there was agreement between the French and Mobutu that the Congolese forces will not disarm soldiers of the defeated Rwandan army. The refugee question allowed a crisis-plagued Mobutu to resurface politically by posing as the protector of refugees in central Africa.
Both the UN system and U.S.-based NGOs, however, continued to treat these armed camps as exclusively refugee settlements. Along with mainly Northern-funded international NGOs, the UNHCR continued to provide daily provisions for the inhabitants of these camps, advertising this as a humane and charitable act. Asked who he thought bore moral responsibility for that situation, the Congolese priest who worked with Caritas, also the local partner of UNHCR in providing assistance to refugees, answered categorically. The responsibility, he said, lay with UNHCR since it had a real choice in late 1994. That choice was to ask member states to disarm the camps so as not to have to feed what was fast turning into an army. The contrast with Tanzania, which also had to shoulder the burden of a million-plus refugees, makes this clear. Unlike Tanzania, which had a functioning central state and army, Zaire did not. In the absence of a functioning central state in Mobutu’s Zaire, it was clear that only the international community was in a position to impose a solution on Mobutu.
From this point of view, the larger responsibility lay with France and with the UN. The French had deliberately and effectively used humanitarianism as a cloak for the defense of narrow state interests. Through Opération Turquoise, France had gone out of its way to create a protective corridor to save those politically responsible for the genocide in Rwanda. The UN had watched the unfolding of the genocide in Rwanda without so much as lifting a finger. In similar fashion, they watched with complacency as refugee camps were established in the vicinity of international borders, and then as they were turned into camps to arm and train refugees.
The setting up of armed camps of Hutu refugees made life hell for the Tutsi in North and South Kivu. Already, the threat of being declared non-citizens by the 1991 Mission d’Identification de Zaı¨rois au Kivu had increased the cross-border movement of young Congolese Tutsi going to join the RPF for military training. This movement lent credibility to the notion spread by some “indigenous” organizations, including Maghrivi, that the Congolese Tutsi were really Rwandese, not just culturally but also in their political allegiance. Yet, the fact was that the vast majority of Congolese Tutsi had stayed behind in spite of the bloody fighting in Masisi in North Kivu for the simple reason that they had everything to lose and little to gain by moving. That period came to an end in 1994. On the one hand, the Tutsi of Kivu felt physically endangered by the influx of over a million Hutu in armed camps; on the other, they felt a vacuum in Rwanda, to which they could retreat in safety. Even then, not all left willingly. While the 1959 refugees hoped to reacquire their properties upon return, Rwanda held little promise for earlier immigrants who showed little desire to return. When they left, they did so because they felt they had to, because everyone seemed to want them to leave. As if to underline this development, the High Command of the Republic (HCR)—the Parliament of transition—sent a member of Parliament, Mambweni Vangu, to review the situation in Kivu following the genocide of 1994. The Vangu Commission was stacked with anti-Banyarwanda extremists. All Kinyarwanda-speaking people, Hutu or Tutsi, are refugees and must return home—such was the verdict of the commission. Anzuluni Mbembe, the co-speaker of Parliament, joined the chorus when in April 1995 he signed an HCR resolution branding the Banyamulenge as recent refugees and including a list of Banyamulenge to be expelled from Congo.25 The situation in North Kivu reached a climax between March and May 1996, when the remaining Tutsi from Masisi and Rutshuru were identified and taken to the border. They were chased out, not killed. They moved into refugee camps in the Rwandan border town of Gisenyi. In these same camps, one also found the Bahunde, because the Hutu had decided to go after all their enemies: the Banyanga, the Tutsi, the Bahunde. This was the peak of the crisis. It is also when the First Rebellion broke out, leading to the end of the Mobutu regime.
In contrast to North Kivu, the citizenship problem in South Kivu seemed forced until the 1994 refugees came in. Only then did the local administration begin to appropriate Tutsi property in the valley, openly supported by Anzuluni Mbembe. Under pressure from armed Hutu in the camps and from soldiers of the Congolese army, the Banyamulenge began to forge links with the RPF to acquire arms. Many in the valley population blamed Gapangwa, the bishop of Uvira, for colluding in the arming of the Banyamulenge population. A similar split occurred in the Protestant Church, and even in the NGO population. An academic sympathetic to the plight of the Banyamulenge recalled that period: “For anybody in the NGO world, to be publicly sympathetic to the Banyamulenge was to court death. The rationale was: how can you sympathise with those arming when the opposition is unarmed?”26 The reference to being unarmed was to the Congolese opposition, not to the Interahamwe in the armed refugee camps, nor to the Congolese army.
The insertion of a million-plus refugees in camps that were armed and resourced from the outside had a devastating effect on civilian life in Kivu. First, it led to the dollarization of the economy. This bitter truth is best conveyed in the words of the Bukavu-based priest who participated in this humanitarian effort. “One talks of all the humanitarian organisations that came here but one doesn’t talk of how they ruined our economy through its dollarization, its rents going up, local Zairois finding life increasingly beyond their reach. In short, amazing resources were deployed in an unreachable endeavour, one which did not correspond to our vision.”27 To talk to civil society leaders in Kivu about the experience of hosting a million-plus refugees resourced through international NGOs was to listen to a litany of troubles—criminality, ill health, increased prices, lowered production, mounting insecurity—all traced to that single experience.
The second effect of armed refugee camps was to accelerate the tendency to militarize ordinary life. From Kivu, the genocide in Rwanda loomed like a volcanic eruption. As the spillover into Kivu translated into armed refugee camps, the people of Kivu began to experience the violence of Hutu/Tutsi antagonism directly. Subjected to a regime of terror by armed Interahamwe based in refugee camps, more and more Congolese Tutsi crossed the border into Rwanda. In response, the RPF trained and armed Congolese Tutsi. As the Interahamwe roamed the countryside, they began collaborating with the Congolese army. In response, more and more Native Authorities created their own militia. The anatomy of political life in Kivu began to resemble that in Rwanda. As in Rwanda, where every political party had come to have its own militia by the genocide of 1994, so in Kivu every Native Authority began to acquire its own militia in the postgenocide period.
The origin of the militia phenomenon lies in the Mulelist rebellion against the Mobutist coup of the early 1960s, when the militia from around Fizi and Uvira joined the Mulelists. The commander of the Rwanda force in Congo, James Kabarebe, traced the militia operating in Kivu to the remnants of the “Mulele wars”:
When Mulelists were defeated, the leadership fled to Europe, and peasants retreated into the countryside to defend themselves. They added more weapons to theirs, but failed to forge a political organisation. When they are asked to sing, they sing about Lumumba. When you tell them about Kabila, they ask, “‘Nani Kabila? Huyu ni mtoto wa Lumumba?’” (“‘Who is Kabila? Is he the son of Lumumba?’”) When you say yes, they dance with joy.28
The original militia tended to have a number of factors in common. First, faced with superior military technology, they tended to rely on supernatural resources. How else could men and women with no more than bows and arrows overcome guns and bullets, except with the help of a pantheon of spirits that would transform lethal bullets into harmless raindrops? Just as in the Mayi Mayi resistance in pre–World War I Tanganyika and many a rebellion thereafter, so in the Mulelist uprising in Congo of the 1960s magic turned out to be a key component of rebellion. Rebel forces marching into battle chanted “Mulele Mai! Mulele Mai!” evoking the power of Mulele the leader to turn bullets into Mai, literally, water.29 Second, the militia tended to combine “a very strict military code” with frequent “experiments with more egalitarian forms of social organisation for self-help and protection.”30 Though every militia developed in tandem with ethnically defined Native Authorities, it also functioned at an arm’s length from the Native Authority. Stemming from the old order, they seemed to reach out to define possibilities of a new one.
The land conflict in Masisi was another important turning point in the development of local militias. The Congolese head of the First Rebellion, Laurent Kabila, traced the development of an “ethnic militia” in North Kivu—unlike in South Kivu—to the formation of Maghrivi by Hutu partisans.31 Gradually, the nature of the militias changed to favor youthful members. The more that fighting turned into a mode of earning livelihood, the more militia membership became a refuge for many a marginalized youngster and school dropout. It is in this period that, faced with an “indigenous” militia, the Hutu developed their own countermilitia, called Les Combattants. After 1994, it collaborated freely with the Interahamwe. In a parallel movement, the Tutsi, concentrated in South Kivu, consolidated their organizations under a single umbrella, the ADP, in November 1996. The ADP, as already pointed out, was an organization of the Congolese Tutsi.
While the biggest of the militias in Kivu were those of the Congolese Hutu and Tutsi, at least four militias operated on an “indigenous” basis in North Kivu by 1997. The first of these, the Mayi Mayi,32 was said to be based in the central area of Masisi and Walikali. Its recruits came mainly from two ethnicities, the Bahunde and the Batembo. The second was the Ngilima. Based in the northern areas of Lubero and Beni, it drew members mainly from the Banande. The Banande were also the main force in the third militia, the Kasingien. The difference was that the Kasingien was a cross-border militia, its members coming from Congolese living on both sides of the Uganda-Congo border. With its headquarters at the foot of Mount Ruwenzori, the Kasingien freely cooperated with the Ngilima. Like the Ngilima, the Kasingien also claimed to have found a mystical antidote that would render humans safe against bullets. The last militia to organize by 1997 was Katuko. With mainly young Banyanga recruits, it operated in the area stretching from Kale in the south to Walikali in the north.
Just as the term “Banyamulenge” has become a generic term for all Congolese Tutsi, so the term “Mayi Mayi” became a generic term for all militias in Kivu Province linked to “indigenous” Native Authorities. The forces that mounted the first rebellion against Mobutu were a coalition of recruits from various ethnic militias, both “indigenous” and “nonindigenous”—on the one hand the Mayi Mayi, on the other the Banyamulenge. In an interview at the height of the First Rebellion, when he was trying to downplay the numbers of the Banyamulenge among his forces, Laurent Kabila said, “The Banyamulenge are no more than 4,000 in our forces, which are more than 15,000.”33 One was struck by the almost exact parallel with the numbers of the Banyarwanda among the National Resistance Army (NRA) as it took Kampala almost a decade ago. There was also a second parallel. If it was the disenfranchisement of the Banyarwanda in the NRA that created the immediate context of the RPF invasion of Rwanda in 1990, it was more or less a similar threat to the Banyamulenge in the new Congo army that would bring Rwanda to back a Second Rebellion, this time against the government of Laurent Kabila. The Second Rebellion found the various “indigenous” Mayi Mayi and the “nonindigenous” Banyamulenge on opposite sides.
The Mayi Mayi joined the First Rebellion in Congo, the rebellion against Mobutu, but opposed the rebellion when it came to power. The reason is simple. They joined it when the rebellion targeted the Interahamwe and the allied Congolese army. And they opposed the rebellion when they saw it turn into the spearhead of a likely Rwandese occupation. Most civil-society leaders in Kivu shared this fear. To explain it, they cited two developments. The first was that the commander of the Rwandese army was formally appointed the commander of the Congolese National Army. Second, this army had begun to intervene directly in Congolese affairs, actively supporting demands by the Congolese Tutsi: that the Banyamulenge be given a separate Native Authority in South Kivu, and that the Hutu head of the Native Authority in Rutshuru (North Kivu) be replaced by a Tutsi, so as to return the situation to what it had been before 1918. From the point of view of “indigenous” ethnicities in Kivu, the postgenocide Rwandese army was an armed expression of Tutsi power that would be used to give teeth to Tutsi claims for an “indigenous” status in Congo or, worse still, to annex Kivu to Rwanda.
The ethnic situation in Kivu went from bad to worse with the success of the First Rebellion against Mobutu. The opportunity for removing a long-standing dictatorship in Kinshasa was turned into revenge-seeking in Kivu. No sooner had the war begun than revenge killings started to happen in Goma (North Kivu). The first half of 1997 was marked by lots of killings—with even more people displaced—particularly in Masisi as the Tutsi of North Kivu settled accounts with the Hutu in Maghrivi. A prominent civil-society leader claimed that approximately six thousand Hutu must have been killed in Goma alone in the short space of a week.
The situation in South Kivu was worse. The “Banyamulenge”—I put the term in quotes since we don’t really know who these were—entered the Ruzizi Valley in September 1996. A prominent Bukavu-based intellectual, otherwise sympathetic to the citizenship claims of the Banyamulenge, described the situation in words that one would have dismissed as an exaggeration had they come from a stranger.
The Banyamulenge conquered their rights by arms but the rift between them and the local population has grown. The attitude of the Tutsi soldiers—the Rwandese and the Banyamulenge—during and after the war has made them more detested by the population due to killings, torture. For example, they will go into the village, raid all the cattle, tell the population—since when have you learned to keep cattle; we are cattle; we know cattle. In Bukavu, they went into and stole from houses. Not so much in Goma. The result is the population is increasingly getting concerned over the question of the Tutsi presence.
Two tendencies seemed to be coming together in this assault on the “indigenous” population. For the Congolese Tutsi, it seemed an opportunity to settle scores with local opponents. The Rwandese Tutsi, however, seemed to have generalized their hatred of thegénocidaires, first to all Hutu and then to the “indigenous” population in Kivu, seeing it as a willing host to armed camps of the génocidaires. But their actions fed wild fears in the local population, creating an incredibly tense situation. Some thought that Tutsi power in Rwanda was trying to annex Kivu and turn it into a homeland for Hutu. Others were convinced that a plan was afoot to kill the “indigenous” elite, such as intellectuals and business people, and that lists had already been compiled for that purpose. Several dates were in circulation. Bukavu seemed in a state of grand panic in September 1997. “Today,” a highly respected academic claimed, “it is being said that Ugandan and Rwandan soldiers are digging trenches all around the city, with guns aimed at the city. Everybody is preoccupied with security, not with how to improve relations with one another.”
The more the crisis grew, the more it gave rise to stereotypes which, in turn, nourished the crisis. When asked to reflect on possible solutions to the conflict, a peace activist in Goma mused: “One needs to ask the indigenous whether they can chase away all the Rwandese, and ask the Rwandese whether they can kill all the autochtones.”34 Unwittingly, he had thrown light on the kinds of fears that fed popular stereotypes: the Rwandese fear that they may be chased away by the “indigenous,” and the “indigenous” fear that they may be killed by the Rwandese.
Militarization spread two tendencies in Kivu, indeed as it had in Rwanda. First, the link forged between militarization and genocidal tendencies inside Rwanda spread across its borders. The First Rebellion led to an indiscriminate slaughter of Interahamwe, of unarmed Hutu refugees, of the Hutu in Maghrivi, and even of those Hutu not connected to Maghrivi. Those responsible for that slaughter became part of the post-Mobutu government and were part of the forces that opposed a UN inquiry into the matter. Those who carried out indiscriminate massacres of Hutu in Kivu are today a part of the military forces of the Second Rebellion. The Second Rebellion, in turn, evoked from the Kabila government an exhortation to the “indigenous” population in Kivu to slaughterindiscriminately not only invading forces from Rwanda, but also the Congolese Banyamulenge in the rebellion, and even all Congolese Tutsi civilians. While Rwanda armed the Congolese Tutsi to beef up the Second Rebellion, the Kabila government armed Congolese Hutu as a countermeasure. Each seemed determined to liquidate the other—physically.
The second effect of militarization was to reduce all credible politics to armed politics. The result was to undermine politics as a civil activity. Once again, this tendency developed in a consolidated form in pregenocide Rwanda, where each political party felt compelled to organize its own militia as a matter of self-defense. As in postgenocide Rwanda, those in power tend to demonize all oppositional politics—regardless of its political character—as génocidaire, and, as if on cue, opposition takes on an armed character.
CUSTOMARY SPHERE AND ETHNIC CONFLICT
The depth of the crisis in eastern Congo cannot be understood unless we see it as the result of a confluence of two distinct processes: the social crisis of postgenocide Rwanda and the citizenship crisis in the entire region. The genocide has given rise to a diasporic state, Rwanda. Two convictions underline the diasporic character of postgenocide power in Rwanda. The first is an overwhelming sense of moral responsibility for the very survival of all remaining Tutsi, globally. The result is that postgenocide power is defined by a diasporic, rather than a territorial, notion of political obligation and political community. The second—also a direct outcome of the experience of genocide—is the conviction that power is the condition of Tutsi survival. As the Congolese Tutsi legal adviser to the secretary-general of the Alliance put it, “In Rwanda, the Tutsi have reached a conclusion that power is the only guarantee for their right to life, otherwise they will be killed by Hutu.”35 The newly appointed Rwandese commander of the Congolese army echoed that same thought: “The Tutsi are just a scared group, from 1959, 1973, 1994. They will feel no assurance until they are protected by Tutsi themselves. That is natural.”36
The crisis in Kivu was immediately triggered by the spillover of the Rwandese genocide across the border into eastern Congo. For that reason, its external aspect has been more dramatic. Yet, the internal aspect of the crisis, generated by the partial and incomplete reform of the colonial state, is the more salient. While the reform deracialized the civic sphere, it left the ethnic character of the customary sphere intact. The rationale was that this would preserve the authenticity of Congolese custom. The consequence, though, was to contaminate the deracialized civic sphere with conflict generated in the ethnic sphere. So long as the customary sphere distinguished between citizens on the basis of whether they were ethnically indigenous or ethnic strangers, it continued to generate conflict in ethnic terms. Those denied rights in the customary sphere on ethnic grounds turned to the civic sphere to marshal resources to establish customary claims—just as those considered ethnically indigenous sought to turn customary prerogatives into an advantage in the civic sphere. So long as the customary sphere is not deethnicized as part of a broader reform, deracialization of the civic sphere will only lead to a spillover—even to an explosion—of ethnic conflict in the civic realm.
We can now see the location of the political problem called ethnic. Whereas “customary” power—the power at the level of the collectivité—is defined monoethnically, the population resident on the ground is multiethnic. Thus, for example, while the power and the locality were defined as Bafuliro, the resident population included both “indigenous” Bafuliro and Banyarwanda immigrants. I have suggested that it is the struggle of the Banyarwanda immigrants in this area to change from a descent-based (Banya-rwanda: those from Rwanda) to a residence-based (Banya-mulenge: those from Mulenge) political identity that illuminates the nature of this dilemma. And further, that it is the reasons for the failure of this initiative in the context of a rapidly regionalizing crisis that may suggest ways out of this conflict.
In attempting to change from indigeneity to residence as the basis of rights, the Banyamulenge tried to undo an important legacy of colonialism. Their initiative paralleled that of the NRA in the Luwero Triangle in Uganda. Though there is no proof of any contact between the two, both responses were born of similar postcolonial state histories. But there was also a difference. In the Ugandan context, the initiative came from a leadership of both migrant and “indigenous” origins. Locked in a civil war against a despotic state, this majority tried to construct a political umbrella under which they could mobilize all residents, indigenous and nonindigenous. In the Congolese case, however, the initiative came exclusively from those considered not indigenous as part of their effort to find accommodation as residents. Without a further initiative underlining a broader commonality of interests among residents, however, this initiative appeared as a “nonindigenous” attempt to usurp “indigenous” resources. The civil war in Congo thus took a turn very different from the earlier civil war in Uganda: instead of a coalition of all residents against a despotic state, as was the earlier tendency in the Luwero Triangle in Uganda, the war in Congo pit the “indigenous” majority against a divided nonindigenous minority, with the Congolese Tutsi allied to the postgenocide state in Rwanda and the Congolese Hutu opposed to it just as firmly. The war in Congo thus crystallized two volatile regional diasporas—one Hutu, the other Tutsi—each determined to set the region on fire if the demands it considered legitimate were not met.