DECOLONIZATION in Africa unfolded along two different trajectories, setting apart the process of decolonization in settler colonies from that in colonies without settler minorities. Where settler minorities vied for political power against both the native majority and the imperial power—as in South Africa during the Boer War, Kenya at the time of Mau Mau, and Zimbabwe following the Unilateral Declaration of Independence (UDI)—it took an internal struggle, always extralegal and usually armed, for natives to win state independence. Where settlers did not exist as a group, or none claimed political power, the colonial power had a larger margin of maneuver and was able to differentiate between moderate and militant nationalism, to play one against the other, and to usher in a mode of independence in line with imperial notions of harmony of interests in postcolonial Africa. The two trajectories of decolonization—one armed, the other nonviolent—were at the same time testimony to different modes of colonization in these places.
I have argued that Rwanda was anything but a standard colony, that it was more of a halfway house between a direct and indirect-rule colony. The mode of decolonization in Rwanda, too, did not clearly follow one of the two patterns I have outlined above. On the one hand, the movement toward decolonization was more like the kind of process in settler colonial contexts, unfolding more through a set of internal convulsions than through a direct confrontation with the colonial power. Precisely because Hutu and Tutsi had, under colonialism, become synonymous with an indigenous majority and an alien minority, decolonization was a direct outgrowth of an internal social movement that empowered the majority constructed as indigenous against the minority constructed as alien. Recall that the majority declaration was called “the Bahutu Manifesto,” not “the Rwandan Manifesto.” It claimed that “the conflict between Hutu and Hamites—i.e., foreign-Tutsi” was the heart of the Rwandan problem and called for a double liberation of the Hutu: “from both the ‘Hamites’ and ‘Bazungu’ (white) colonization.” As in settler colonies, political independence with majority rule would require a “revolution” at the local level.
On the other hand, this revolution did not have to be violent. Three developments testified to this possibility. The first was that state independence was not just a Hutu demand; it was also made by Tutsi elites. The distinction between the two was not that one called for state independence and the other opposed it, but the perspective each had on the society that would follow colonialism. The second was that, in spite of the divide between Hutu/Tutsi political elites, one could identify political tendencies cutting across the same elites. The third was that most of the violence in 1959–63 occurred not at the time of the revolution of 1959 but in response to subsequent attempts at restoration.
The year 1959 saw the first major political change in colonial Rwanda. Ushered in by political violence, it led to the routing and dismantling of Tutsi power at the local administration level. It also triggered broader constitutional and political developments that led to a transfer of governmental power from a Tutsi to a Hutu elite. In a study that aimed to reflect on “long-term transformations as they related to Rwandan Revolution,” Catharine Newbury provided an influential defense of “the Hutu revolution,” arguing that “ultimately, an appeal to Hutu solidarity became, for Hutu leaders, the most effective rallying point for revolutionary activity.”1 In writings of this period, 1959 was celebrated as a “revolution.” In a major political analysis of the revolution, René Lemarchand compared 1959 Rwanda to 1789 France, and the Hutu revolutionaries to French Jacobins.2Lemarchand saw the problems of the revolution not in terms of what it had accomplished, but in terms of what remained to be done: “Now that the initial phase has been completed, there remains the more fundamental task of social and economic reform…. Until this is done, the Rwandese revolution must be regarded, in Marx’s terms, as ‘a partial, merely political revolution, which leaves the pillars of the building standing.’ ”3 I shall argue the reverse: the revolution’s achievements were more in the economic and the social realm, its problems more political. The Revolution not only left standing, but reinforced, the political identities created by colonialism. In the history of African decolonization, however, 1959 Rwanda most closely resembles 1961 Zanzibar.4 Both ushered in a transfer of power through political violence. Neither can be dismissed as simply a change of elites. Because 1959 changed the nature of power and had a significant consequence for the mass of the Rwandan people, it needs to be seen as a “revolution.” Unlike those who tended to celebrate it, however, I argue that it needs to be problematized.
But before I do so, I find it worthwhile to point out the way in which the political violence that ushered in 1959 marks a significant departure from political violence in the preceding period. The key difference lies in the direction of political violence, in how it demarcated its agents from its target. For the first time in the history of the Rwandan state, the violence demarcated Hutu from Tutsi. The polarization of Hutu and Tutsi in 1959 contrasted dramatically with the presence of Hutu and Tutsi on both sides of the firing line during the Nyabingi revolt only a half century earlier. I argue that this single fact is proof enough that the real turning point in the history of political conflict and political violence was not colonization at the turn of the century, or even the replacement of German by Belgian rule at the beginning of this century, but the reorganization of the colonial state from 1926 to 1936.
To highlight the significance of this shift, we need to recall that the genesis of the Rwandan state can be traced to sometime in the middle of the fifteenth century. In spite of at least four dynastic changes over the next three and a half centuries and no less than ten successions that flouted ritual norms, the fact is that the political elite of Rwanda had been remarkably successful in achieving the raison d’être of any state, that is, to ensure order.5 One cannot but contrast this outstanding record with the dismal record of the colonial state: in but a handful of decades following the colonial reorganization of 1926–36, Rwanda imploded in a revolution that pitted one section of the population (Hutu) against another (Tutsi). How is one to explain this dramatic shift, from long-term political stability to short-term political breakdown?
The root causes of the 1959 Revolution need to be explored in the changes wrought by colonialism, and not in the precolonial legacy. We have seen that when Mwami Rwabugiri centralized the state toward the close of the nineteenth century, he also made it the custodian of Tutsi privilege. Belgian rule had contradictory consequence for the Tutsi: on the one hand, it branded the Tutsi as not indigenous; on the other hand, it consolidated Tutsi privilege by a double move that affected all strata among the Tutsi. Up above, it made chiefship a Tutsi prerogative with the fused authority of the chief accountable to none but the colonial power; down below, it exempted the petits Tutsi from coerced labor. It is precisely because colonialism underwrote Tutsi privilege in law that the Tutsi, beginning with the elite, embraced the racialization of their own identity as nonindigenous. The claim that the Tutsi were nonindigenous Hamites was considered necessary for their privileged treatment in law. Not surprisingly, mainstream Tutsi nationalism presented the colonial construction of custom and customary power—specifically, Tutsi privilege—as authentic “tradition” and demanded that independence be a return to tradition. This was the standard independence rhetoric of nationalism in colonial Rwanda. Unlike in other African colonies, however, standard independence rhetoric, directed only or even mainly at the colonial power, was a marginal phenomenon in colonial Rwanda.
The colonial impetus on Rwanda was contradictory: it tended to stiffen the state while dynamizing society. As an energized society tended to generate new forces, a hardened state structure proved unresponsive to them. These contradictory tendencies led to an escalating and dramatic confrontation between state and society. On the one hand, the state was organized and nurtured as so many localized despotisms. Each of these saw itself as a Tutsi power, lording it over subservient Hutu subjects who were in turn sealed from the world of Tutsi privilege by the requirement to carry an identity card and by the legal impossibility—no matter what their life circumstances—of a ritual rise to Tutsi status. On the other hand, the same colonial power introduced a money economy and school-based education, processes that generated new influences and new opportunities, and in time gave rise to a Hutu elite. Locked into a subordinate status by a legally enforced identity, this socially frustrated group developed—for the first time in the history of Rwanda—into a political counterelite. In the changed context of a post-Second World War Rwanda, the Hutu counterelite was poised to tap the grievances of the Hutu peasantry against local despots who claimed their power was not a colonial imposition but a right by custom.
THE DEVELOPMENT OF A HUTU COUNTERELITE
The Hutu counterelite developed from three social locations. The first location was the precolonial elite in the independent non-Rwandan principalities: those elites in the north who were forcibly incorporated into the Rwandan state and subjugated to a Hutu status by the newly forged alliance between German colonialism and the mwami (king). The second major source was the market economy, in particular the labor market in nearby Congo and Uganda, which made it possible for Hutu peasants to escape the demands of servitude inside Rwanda. The third source was school-based education pioneered by the alliance of missionaries and the colonial state.
René Lemarchand has argued that the set of events known as the 1959 Revolution was in reality a confluence of two distinct social processes, one in the north, the other in the center of the country.6 While “the revolution in central Rwanda was a social revolution in the sense that it developed its dialectic from the social inequities of the caste system,” he argues that a more “retrogressive” attitude shaped the revolutionary outlook in the north. The difference was this: “In seeking to evict the Tutsi oligarchy from its position of power the northern Hutu did not aim so much at the creation of a new social order as to revert to the social existence prior to the intrusion of Tutsi conquerors.”7 Lemarchand thus distinguished the key impulse behind the revolution in the north as ethnic, from that in the center as democratic.
“Northern Hutu” refers to the Bakiga, who lived in the former territories of Ruhengeri, Gisenyi, and Byumba. In the already-quoted words of P.T.W. Baxter, their “proud boast,” was “that they were never, as a people, subjugated by either Tutsi or Hima.”8 And yet, we need to keep in mind that the fiercely independent spirit of the Bakiga did not always automatically translate into an anti-Tutsi orientation. This orientation was the result of a historical development under specific circumstances.
The context that shaped the “northern Hutu” perspective was marked by at least three features. First, there was the historical difference between the incorporation of the north and that of the south and the center in the Rwandan state. Unlike the Hutu of central Rwanda who had been a part of the central court for centuries, and the southern Hutu who were subordinated to central rule before colonialism, even if only in the second half of the nineteenth century, the Bakiga of the north were subjugated to Rwandan state authority only with the onset of Western colonization. Second, while it is true that the Bakiga experience of colonial domination was coterminous with their experience of Tutsi domination, it is not true that their opposition to colonial and Tutsi domination automatically translated into an anti-Tutsi hostility. To confirm this, one needs to look at the actual historical revolt of the Bakiga against colonial and Tutsi domination at the onset of colonial rule, one that goes by the name Nyabingi. We have seen that this revolt was in reality a coalition of two forces: the section of the Tutsi aristocracy excluded from power at the death of Rwabugiri, and the Bakiga newly subjugated to this hardening Tutsi power. The revolt of the Bakiga was led by members of the Tutsi aristocracy who were bitterly opposed to the usurpation of power by the Abeega clan at the death of Rwabugiri, the most prominent of these being Muhumusa and Ndungutse. Key to its agenda was opposition to forced labor tribute (ubareetwa) freshly imposed on the newly colonized Bakiga. Third, this protracted rebellion against colonial authority and its Tutsi quislings made for a more arbitrary chiefly authority in the north than was imposed anywhere else in colonial Rwanda.
The outright defeat of the Nyabingi cult created conditions for the emergence of a different type of anticolonial revolt in the north. It was led by an indigenous Bakiga elite that could identify both colonialism and the Tutsi with the arbitrary and oppressive rule of Tutsi chiefs. This elite had roots in those who had controlled access to land in the precolonial period, whether as outright owners of the land (bakonde) or as clients (bagererwa) who controlled access to it. With colonial repression, both were replaced, the owners by incoming Tutsi and the clients by Bakiga quislings. The language of the day made a distinction between the two groups, referring to the former clients as “traditional bagererwa” and the latter as “political bagererwa.” Colonial repression was never effective enough to erase the claims of the precolonial hierarchy. As a result, the two hierarchies coexisted in a growing tension. According to Lemarchand, the older generation of bakonde tended to resort to forms of protest associated with “social banditry.” In contrast, the new generation, those “substantially more westernised and better-educated than their predecessors,” were able to integrate the traditional claims of the bakonde into a modern revolutionary movement, the one that developed in tandem with the revolution of 1959.9
It is understandable that the offspring of the precolonial elite would seek to restore a freedom very much real within living memory. For them, the revolution was more of a “national” than a democratic affair. Yet, it seems to me that Lemarchand is so preoccupied with the interests and motivations of this precolonial elite that he tends to ignore those of the ordinary peasants on the ground. Had they not been able to tap the widespread antagonism of Hutu peasants toward Tutsi chiefs, both the traditional bakondeand bagererwa would have remained isolated and weak forces. For that same reason—that he tends to downplay the tension between peasants and chiefly power—Lemarchand goes too far when he contrasts the perspective of the northern and central Hutu elites in terms that oppose a northern preoccupation with “restoration” with a central commitment to “revolution.”10 If the notion of a north preoccupied with “restoration” sidesteps the perspective of the northern peasants, then that of a south committed to “revolution” errs in the opposite direction: it underplays the point of view and interests of southern elites. To do so is to shortchange the northern initiative and to romanticize the revolutionary thrust in the center. It is generally accepted that the revolution exploded as mainly a hill-level confrontation between Hutu peasants and Tutsi chiefs. At the same time, this tension-ridden relationship cut across the distinction between the north and the south. If anything, chiefly rule in the north was even more arbitrary than in the south—a fact no one has done more to underline than Lemarchand. If we accept that it was the demand for dismantling chiefly despotism that provided the democratic kernel of the revolution, it is difficult to see how anyone can then argue that democracy was a southern demand.
Hutu labor migrants in the colonial period found two major destinations. The first was Congo, the second Uganda. Both opportunities opened up in the interwar period. A thriving labor market developed in Congo around the mines in Katanga and the plantations in Kivu. Similar opportunities opened up in Uganda with the growth of a prosperous rich peasant coffee economy in Buganda and sugar plantations in Busoga, followed by the copper mines that were opened in Kilembe after the Second World War. As one would expect, migrants originated mainly from areas adjacent to Congo and Uganda: the west and southwest for those going to Congo and the north and northeast for those heading to Uganda. We have two studies from which to draw general conclusions about the migrant experience. The first is a study of immigrant labor in Buganda, a team effort led by anthropologist Audrey Richards.11 The second is a study by Mararo Bucyalimwe of land conflicts in Masisi, eastern Congo, following the “transplanting” of peasants from Rwanda.12Since these focus mainly on the migrant experience in Uganda and Congo, we shall return to them in the chapters on Uganda and Congo. In this chapter, I shall turn to Catharine Newbury’s excellent study of state-society relations in Cyangugu in southwestern Rwanda, the source of most migrants to Congo.13 Since Newbury’s focus was on the home territory from where the migrants originated, and to which they returned, it is a source of fruitful insights into how the migrant experience shaped the anticolonial struggle.
For many of the labor migrants from Cyangugu to Congo, contract work for Europeans was a way of escaping forced labor imposed by Tutsi chiefs and the local authorities at home. Corvée included both ubureetwa and akazi, the former performed for individual chiefs and the latter for the public authority, which required it mainly for public works projects. Belgian authorities were quite aware that forced labor for local authorities and low-paid work for European employees were alternatives for the border population. And they worked hard to ensure that Tutsi chiefs not disturb this arrangement by imposing tribute on families of those who had opted to work for Europeans. In 1938, the territorial administrator in Cyangugu wrote three subchiefs under him warning them to keep the whip away from those who worked for the Nyungwe mines: “The chiefs and sub-chiefs have no right to require the workers at Nyungwe to perform forced labour associated with ubuletwa.” And then: “The chiefs and sub-chiefs do not have the right to require that the wives and children of the workers carry out [obligatory] cultivation as provided for by Regulation 89.”14 Almost a decade later, in 1946, the territorial administrator of Cyangugu made sure to include the same cautionary note in what was otherwise an exhortation to Tutsi subchiefs to ensure that the Hutu peasantry labored according to the following instructions:
We inform the subchiefs who are present of the results of yesterday’s meeting with the settlers and the chiefs. We expect more firm collaboration from the subchiefs so that the indiscipline so evident among the Hutu will cease. It is necessary that the native authorities become aware of the fact that they represent the State, and they must rule those they administer with justice and firmness. And in this regard, they must require from all Hutu [who are] not working for Europeans the completion of all duties with regard to [obligatory] crops, the struggle against erosion, and the maintenance of the roads.15
For those who migrated to Uganda from north and the northeast, the turning point was the year 1924, “when Belgian authorities empowered residents to compel natives to carry out the cultivation of food-stuffs and economic crops.” The first ten months of 1925 recorded the first large-scale Rwandan emigration into Uganda, being the great majority of the 11,771 laborers recruited by the Labour Department from the southwest of the country. Three years later, Rwanda experienced a prolonged famine; immigration into Uganda reached “formidable proportions” as an estimated 35,000 crossed the border.16 When asked for reasons they had decided to leave home, the answers highlighted the regime of forced labor and compulsions: “I left because in Rwanda a man and his wife have to work from early morning to late at night for his chief,” or “Ordinary men work for their chiefs and when they find they have nothing to wear, they leave their country to look for money.” Some complained of beatings: “If I was not beaten I would never have come to Uganda” or “I left home because I wanted a job without beatings.” One man had returned to Rwanda on three successive occasions—1938, 1941, and 1944—to see “if there were still beatings” and had felt it necessary to return to Uganda each time. Others complained that “women as well as men are liable to communal labour.” And yet others explained that they had decided to go home and bring their wives to Uganda because, when they were away, their wives were asked to carry out labor obligations of their own, on top of having to make up for the obligations of the absent husband.17
The migrant experience was the source of fresh, new insight, often subversive of hierarchy in the existing order. Newbury recorded testimony from numerous respondents, noting “arbitrary action of the powerful as a principal reason they and others went to work for Europeans.” She also pointed out that many of the returned migrants provided leadership in the protest that mushroomed over the first postwar decade: “Many of the early leaders of Hutu protest activity in Kinyaga were former wage earners who had taken up trading enterprises of some kind—gaining both economic security and a network of contacts that later proved useful for political party organization.”18
Not everything under the colonial political system was hard. Two broad processes were under way: the expansion of a money economy, and school-based Western education. Together, they would erode the social supremacy of the Tutsi while, for a time, leaving intact their political supremacy. Although the cattle-based wealth of the Tutsi aristocracy remained largely uncommercialized, Belgian officialdom made every effort to get the Hutu peasantry to grow cash crops for export. In opening up opportunities for enrichment through other than the ownership of cattle, the money economy weakened the bonds of pastoral servitude that had been the colonial ubuhake contract between patron and client. It is in this context that the expanding school system of the 1940s and 1950s provided the structural basis for the emergence of a Hutu counterelite.
The impact of the school system on the few Hutu who managed to enter its corridors was contradictory and explosive. On the one hand, it reproduced the political and social distinction between Tutsi and Hutu at an intellectual level by operating a two-tier system: the Tutsi were introduced into a “civilized” French-medium education, but the Hutu were confined to a “nativized” second-rate Kiswahili-medium education. On the other hand, the same school system was the source of merit-based impulses that could not but generate egalitarian ideas, even if the curriculum included a heavy dose of the Hamitic hypothesis.
We can get an idea of how small were the numbers of Hutu who managed to gain access to secondary education in colonial Rwanda by a second look at the figures for those enrolled at the Groupe Scolaire d’Astrida, the leading secondary school in the country. Students came from the three Belgian colonies of Rwanda, Burundi, and Congo.19 Until 1945, students from Rwanda and Burundi were registered as a single group, but were identified as Tutsi or Hutu. After 1945, the Hutu were further classified into those from Rwanda and those from Burundi, whereas the Tutsi were still registered as a single group. The figures show that Rwandan Hutu were virtually excluded from the school before 1954: between 1946 and 1954, sixteen Hutu were admitted from Rwanda, as opposed to seventy-one from Burundi. In contrast, 389 Tutsi were admitted from both Rwanda and Burundi during that same period. Only in 1956 did the proportion of Hutu students begin to increase substantially.
Ironically, the first Rwandan student to graduate with a university education was a Hutu, and he graduated from the Centre Universitaire de Kisantu (Congo-Kinshasa) in 1955. Anastase Mukuza was to become a leading figure in the postrevolutionary government of Kayibanda.20 His example is illustrative of the kind of social frustration that pushed the first generation Hutu elite into the front ranks of 1959 revolutionaries. Mukuza attended the Grand Séminaire de Nyakibanda in Rwanda, and then joined the Centre Universitaire de Kisantu in Congo, where he completed a degree in administrative and political sciences. On return to Rwanda in 1955, he paid a visit to Mwami Mutara to explore the possibility of government employment. His request was turned down. Next, he went to the Institut pour la Recherche en Afrique Centrale (IRSAC) at Astrida, looking to be a research assistant. Here again he was rebuffed. He then went to see the directeur de l’enseignement in Bujumbura in Burundi, only to be told that the administration would not recognize his diploma. He ended up as a typist (candidat commis) in Kibuye, promoted to administrative assistant in 1957, first in Cyangugu, and then in Kigali. By then, he was a potential revolutionary. Lemarchand’s comment is apropos of the significance of this case of the first Rwandan university graduate: “Like other educated Hutu, he derived a burning sense of grievance from the monopoly exercised by the Tutsi caste over all sectors of the administration and the economy; to break the hold of this monopoly became a central objective of the Hutu intellectuals on the eve of the revolution.”21 This point of view was expressed in a popular postwar play. After commiserating aloud on the injustices of buhake, one of the Hutu characters in Naigiziki’s play,L’optimiste, asks his companion, “How long shall we have to wait until our injustices are redressed?” The interlocutor replies, “Until the Hutu no longer has the soul of a serf. For that he must be reborn.” The midwife of that rebirth was a political movement of the Hutu counterelite.
Unsurprisingly, most of the leading personalities of the Hutu movement were former seminarians. They had studied for the priesthood, either at Kabgaye or at Nyakibanda. For the Hutu who managed to ascend the Church hierarchy, every climb up the ladder put them in a context dominated by Tutsi priests. The influence of the Western Church—much like that of the Western school system—was contradictory. As an institution, the Church had been the primary force advocating the “civilizing” role of the Tutsi as Hamites. Accordingly, there was preferential entry for Tutsi into the priesthood, at least until after the Second World War. But as an ideology, Christianity was a source of an egalitarian impulse for the Hutu, not just for the masses who entered the Church, but particularly for the few who did manage to enter the priesthood.
The contradiction between Christianity as an ideology and the Church as an institution came to a head in the postwar period as the attitude of the European clergy went through a major shift. With the defeat of Nazism, its collaborators were discredited everywhere in Europe. Most institutions, including the Church, experienced a democratic resurgence. The clergy coming to Rwanda after the war were a changed lot, strongly influenced by antiracist ideological currents. Unlike Monseigneurs Classe or Hirth, early leaders of the Church who were upper-class Flemish men with conservative views, the newcomers were likely to come from le petit clergé. Of “relatively humble origins,” and with a “previous experience of social and political conditions in the French-speaking provinces of Wallonia,” they were “more generally disposed to identify with the plight of the Hutu masses.”22
The Church was also the location from which the Tutsi intelligentsia defended “racial” privilege. Though its depth went no further than the colonial period, they defended it as a “tradition.” Members of the predominantly Tutsi Rwandan clergy were among the first to express public anxiety over the spread of egalitarian ideas among the new Hutu elite. The loudest warning came from Abbé Alexis Kagame, then Rwanda’s foremost historian. Without mincing words, Kagame wrote as early as 1945: “Certain egalitarian tendencies are advocated in front of those elements who are sometimes referred to as ‘child-like grown-ups’, without proper intellectual formation, which are bound to run counter to the common sense principles of most if not all of them.” Warning that “the path of progress cannot stray away from our traditional heritage,” he went on to champion a type of progress that would not question traditional authority. “Regardless of the type of socio-political system adhered to, one must avoid humiliating traditional authorities, either by disregarding their claims to leadership or casting discredit upon them in front of their subject under the pretext that everybody is equal. The conclusion the masses are likely to draw from all this is that progress, freedom, in short everything, implies contempt for traditional authorities.”23
When the Hutu graduates of seminaries and of the Groupe Scolaire d’Astrida (now Butare) entered the job market in the mid-’50s and found there were few places open for an educated Hutu, they turned to the Church for opportunities. Literally shut out of jobs in the civil service and the private sector, they looked at their new positions not just as ways of making a living but also as opportunities to articulate their major social grievance: the institutionalized exclusion of Hutu from a Belgian-supported Tutsi monopoly over all avenues of social advancement. With the support of a sympathetic clergy, they took over Church publications—the most important being the Kinyarwanda-language magazine Kinyamateka—and began to address whoever would listen sympathetically, mainly Hutu masses below and visiting United Nations Commissions above.
REFORMS AND THEIR LIMITS
Though administered by Belgium after the German defeat in World War I, Rwanda was a UN trust territory. Under UN tutelage, the process of decolonization unfolded as a series of electoral reforms, beginning in 1952. The backdrop to the reform process was a series of UN decolonization missions that were regularly dispatched to its trust territory, at least once every three years, sometimes more often, from 1949.
The first ever visit of a UN decolonization mission coincided with a dramatic reform that promised to abolish the hated ubuleetwa and replace it by a mandatory money payment. That was in 1949, yet respondents in Kinyaga told Catharine Newbury that they continued to perform ubuleetwaservices until the revolution. There was clearly a big difference between the promise of a reform and the fact of its implementation, between the wider propaganda effect of the announcement of a reform and the social impact of its implementation on the ground. The Hutu learned the same lesson when the mwami and the Conseil Supérieur (High Council) decided to issue another reform decree to coincide with the visit of the third UN decolonization mission in 1954.24 The decree provided for the progressive dissolution of ubuhake ties and the distribution of cows held under it to former clients. Once again, the impact of the decree fell short of its promise. In the absence of a corresponding reform redistributing grazing land monopolized by Tutsi patrons, it left Hutu owners of cattle dependent on former patrons for access to pasturage. Just as with the previous abolition of ubuleetwa, the reform of ubuhake did not undo the ties that bound former Hutu clients to Tutsi patrons in a relationship both unequal and coercive. Not surprisingly, peasant protest against the arbitrary use of power by chiefs became the stuff of popular press reports in the postreform period.25
It was the taste of reform, and not the absence of reform, that convinced the Hutu intelligentsia that nothing less than radical change was likely to bring an end to the social plight of the Hutu. That taste was developed through an overall encounter with social and political reform. Political reform began with local elections in 1953 and a general election in 1956. The 1953 elections were the result of the decree of 14 July 1952. The elections were wholly indirect: not only was the role of elected councils “advisory,” but the electoral choice was limited to “suitable candidates” nominated by chiefs and subchiefs. In a context where the administrative power of Tutsi chiefs was still intact, the result was not an election but an opportunity for subchiefs and chiefs to register their power. Two tendencies testified to this outcome. On the one hand, Tutsi tended to predominate in the councils, more so the higher one went up the administrative ladder. So that whereas 52 percent of council seats at the lowest administrative level were filled by Tutsi, the proportion reached a whopping 90.6 percent when it came to the Conseil Supérieur du Pays, the highest council of the land. On the other hand, when Hutu were nominated to councils, they were inevitably Hutu abagaragwa (clients) of Tutsi shebuja (patrons).26
The final opportunity for reform from above was squandered in 1956 when Mwami Rudahigwa joined the conservative Tutsi tendency to defeat a proposal to provide separate representation for Hutu on the Conseil Supérieur. To appreciate the significance of this proposal, one needs to recall two facts. One, the Conseil Supérieur was the highest advisory body of the state and was expected to become the legislature of an independent Rwanda. Two, in that crucial period from 1956 to 1959, this body included only three Hutu, comprising less than 6 percent of its membership.27
The 1956 elections introduced a two-tier system: an all-male universal suffrage at the lowest administrative level, the subchiefdom, while all higher councils continued to be voted indirectly through electoral colleges whose members were nominated by corresponding chiefs. The outcome highlighted the difference in the method employed in each case. There was a clear victory for Hutu candidates at the subchiefdom level, where the vote was direct, but not at higher levels, where the vote was indirect. The contradictory and limited nature of the reform was clear for all to see: it combined participation for Hutu at lower levels with guaranteed power for Tutsi at higher levels. From the abolition of ubuleetwa in 1949 to the general election of 1956, nearly a decade of experience with reform convinced the Hutu political elite that nothing short of political power would crack the Tutsi hold on social, economic, and cultural resources.
THE 1959 REVOLUTION
Two rival documents greeted the visiting UN decolonization mission in 1957. The documents dramatized the growing ideological polarization between Hutu and Tutsi. Anticipating the Mission’s visit, the mwami’s High Council proclaimed an all-Rwandan emancipation program. Called Mise au Point, the program called for a rapid transfer of power to the king and his council. This, it argued, was crucial to end racial tensions between blacks and whites.28 A month later came the Hutu response, in the form of the Bahutu Manifesto. Signed by Kayibanda and eight other Hutu, and originally titled Notes on the Social Aspect of the Racial Native Problem in Rwanda, the Bahutu Manifesto maintained that the heart of the problem in Rwanda was “the conflict between Hutu and Hamitic—i.e., foreign—Tutsi.”29 The authors called for a double liberation of the “Hutu from both the ‘Hamites’ and ‘Bazungu’ (whites) colonization.” It identified the “indigenous racial problem” as the “monopoly which is held by one race, the Tutsi”:
The problem is above all a problem of political monopoly which is held by one race, the Tutsi; political monopoly which, given the totality of current structures becomes an economic and social monopoly; political, economic and social monopoly which, given the de facto discrimination in education, ends up being a cultural monopoly, to the great despair of the Hutu who see themselves condemned to remain forever subaltern manual labourers and still worse, in the context of an independence which they will have helped to win without knowing what they are doing.30
The difference between the two documents could not have been sharper. And yet, though written from different standpoints, each was a claim for power. Independence first, the view of the Tutsi elite, was the claim that their prerogatives were actually “traditional” (precolonial) and should be restored. Democracy before independence, the view of the Hutu counterelite, spelled out their demand for power based on the claim that they represented the indigenous majority. One put forth a “nationalist” claim, the other a “subaltern” demand. Out of this subaltern agitation, the Hutu counterelite created a popular nationalism—a nationalism from below, to rival the Tutsi nationalism from above. Both traced economic and social problems among the poor to a “racial” tension. The difference was that while one highlighted the racial contradiction as only between foreign black and white, the other underlined it as a contradiction mainly between Hamites (Tutsi) and indigenous Bantu (Hutu).
Elections set the context in which the Hutu counterelite forged their consciousness against the Tutsi elite. Such a consciousness emerged from the throes of a political contest. Forged with the creation of the Rwandan state and sharpened with Rwabugiri’s centralizing reforms in the late nineteenth century, Tutsi identity had long preceded Hutu identity. In that context, Tutsi consciousness was a consciousness of power, while Hutu consciousness would come to be one of lack of power and of a struggle for power. Like almost everything else about colonialism, colonial power did not erase precolonial realities but added to them: on the one hand, it so sharpened the late nineteenth century contrast between Tutsi power and the Hutu absence of power as to accentuate it as a one-dimensional reality; on the other, it stigmatized Tutsi power as alien rule.
The development of a Hutu consciousness was a protracted affair, stretching from the time of Rwabugiri through the entire span of the colonial period. As late as independence in 1962, the “Hutu” of the northwestern region insisted on being considered Bakiga—like their neighbors in southwestern Uganda—not Hutu. Hutu consciousness developed in phases: before the Second World War, it was a consciousness of subjecthood that transcended all locally anchored identities; in the 1950s it became the consciousness of a people reaching for power. This development required a confluence of two movements: a genuinely popular movement of (Hutu) peasants against the local despotism of (Tutsi) chiefs; and, for the first time in the history of Rwanda, the emergence of a Hutu counterelite. Propelled center stage by a series of electoral contests, this counterelite put forth a program for the Hutu to seize power to overcome their identity as a subject people. Branded with a subject identity—“Hutu”—the counterelite emerging from the ranks of the socially oppressed held it up as a badge of pride: Hutu Power! In turning a chain into a weapon, Spartacus-style, it was neither the first nor would it be the last. One only needs to think of a related example: Black Power.
That consciousness, and the organization it came to wield, was forged in the institutional context of the Church.31 Both Grégoire Kayibanda, who later became the president of PARMEHUTU, the party of the revolution, and Aloys Munyangaju, initially its vice president and then president of APROSOMA, the party of the alternative Hutu political tendency, achieved prominence as journalists/editors for Catholic periodicals. Besides serving as personal secretary to Monsignor Perraudin, the apostolical vicar of Rwanda, Kayibanda became lay editor of Kinyamateka, the Church-owned Kinyarwanda-language paper, in 1955, and then its editor-in-chief in 1956. In December of that year, Church authorities founded a cooperative: Travail, Fidélité, Progrès (TRAFIPRO). Kayibanda became the president of its board of directors. The expanding ground-level organization of TRAFIPRO came to serve as cells for the development of the Hutu movement. It is from this organizational base—the editorship of Kinyamateka and the presidency of the board of TRAFIPRO—that Kayibanda launched a cultural association called the Mouvement Social Muhutu (MSM) in June 1957.
The more assertive the Hutu counterelite grew, the more it provoked a shrill reaction from those Tutsi who had swallowed wholesale the venom that was the Hamitic hypothesis and who were bent on defending colonial privilege as a time-tested tradition. The response came in two public letters in May 1958 from fourteen senior Tutsi notables at the mwami’s court—called the mwami’s clients, the abagaragwa bakuru b’I bwami. They claimed there was no question of any fraternity between Hutu and Tutsi since Kigwa, the ancestor of the Abanyiginya dynasty, had reduced the Hutu by force. In rejecting the demand for Hutu participation in public affairs, they evoked the tradition of conquest: equal rights were out of question “because our kings conquered the land of the Hutu, killed their ‘little’ kings and thus subjugated the Hutu: how then can they now pretend to be our brothers?”32 The second letter rejected the demand that the ibikingi, the landed property held by Tutsi lords, be abolished. The letter defended its continuation as “the custom of the country.” When Mwami Mutara III Rudahigwa summarized the deliberations of the Conseil Supérieur on 12 June, his response to Hutu agitation was recorded in the minutes as follows: “It is a damaging increasing noisy propaganda spread by a small group acting under foreign influence with communist ideas. Their intention is to divide the country. They would not succeed to divide a country whose national unity and secular political force organisation has annihilated the most powerful attackers. The country is reunited to identify, cut down, eradicate, and burn that ill tree which is infecting its life. Then a motion was voted ‘to ask that the colonial government remove from the official documents the terms Hutu, Tutsi and Twa.’ ”33
The claim that Tutsi power be restored as tradition tended to boomerang in the context of postwar Rwanda: it only succeeded in further mobilizing the Hutu and in discrediting the Tutsi cause. The Hutu response focused on the prime symbol of Tutsi power, the Kalinga drum. Rather than its symbolic association with the crown, it was attacked for signifying a permanent vision of Hutu inferiority: was not the Kalinga drum, after all, decorated with the sexual organs of defeated Hutu kings? The frontline Hutu press called for an end to “the idolatry surrounding the Karinga.”34
The more the Rwandan polity—and society—began to unravel, the more room was created for different internal tendencies. We must not be misled into thinking that the backdrop to the revolution was no more than a gelling of two polarized tendencies: one Hutu, the other Tutsi. That it was, but there was also a growing differentiation inside each. While the revolution was an outcome of the growing polarization between Hutu and Tutsi in Rwandan society, the outcome of the revolution was shaped very much by the contest between different points of views within each camp. To understand the postrevolutionary outcome, we need to go beyond the Hutu/Tutsi polarization to the ideological contest between different tendencies on each side of the Hutu/Tutsi divide.
Historically, the Tutsi political elite was united around the court and constituted a conservative tendency that equated Tutsi power with tradition. The development of a Hutu counterelite and its growing self-assertiveness brought Tutsi unity under pressure. Those who questioned the basis of this unity took initiatives to go beyond its narrow and short-term orientation. In doing so, they both crystallized the plurality of views within the Tutsi elite and gave it organizational expression. The expanded arena of Tutsi politics came to be defined by two rival political parties, Union Nationale Rwandaise (UNAR) and Rassemblement Démocratique Rwandais (RADER), one conservative, the other reformist. Similarly, it was the tension between the militantly Hutu PARMEHUTU and APROSOMA’s search for a broader constituency that came to define the arena of Hutu politics. To explore each of these tendencies is to understand both the limits and the choices—the limited menu—that the revolution of 1959 placed before the people of Rwanda.
Created in August of 1959, the UNAR was a party of monarchists most identified with the “traditionalist” point of view in Rwandan society. Under the nominal presidency of an outspoken Hutu, François Rukeba,35 it was mainly—but not wholly—a Tutsi party. UNAR’s leadership read like a Who’s Who of Tutsi high chiefs. The “progressive” tendency in UNAR best expressed its Janus-faced opposition to the Hutu internally and the Belgians externally. Wedded to Tutsi supremacy at home no less than were the conservatives, UNAR progressives championed a nationalism in external policy. In pursuit of this nationalism, they forged several alliances: with militant nationalists like MNC-Lumumba regionally, with the Communist countries in the UN Trusteeship Council, and with the People’s Republic of China outside it.
The UNAR’s external alliances were not simply another case of the Cold War making for strange bedfellows. UNAR nationalism gave genuine expression to the national grievance that the Tutsi came to feel the most, since their advance was directly blocked by the “racial” privileges of Europeans in the colony. Anticipating the founding of a Hutu party (which indeed came to be three days later), UNAR had distributed a circular on September 16: “Rwandese! Children of Rwanda! Subjects of Kegeri, rise up! Let us unite our strengths! Do not let the blood of Rwanda be spilled in vain. There are no Tutsi, Hutu, Twa. We are all brothers! We are all descendents of Kinyarwanda!”36
Because UNAR began to receive money and diplomatic backing from Communist countries in the UN Trusteeship Council, the antagonism between the Tutsi and the Belgian authorities deepened further. During 1959–60, the UNAR leadership-in-exile courted the support of MNC-Lumumba. Rumors spread that local MNC branches in Congo, especially in Goma and Bukavu, were giving financial and military assistance to the Tutsi leadership to fight their way back into the country. The more these suspicions hardened into certainty, the more Belgium became convinced that in keeping UNAR from the reigns of power, it was fighting both feudalism and communism.37
The day after UNAR had held a mass rally in Kigali vowing to “restore customs” and “shake off the yoke of Belgian colonialism,” Chief Bwanakweri created another Tutsi party with an opposite message: to shake off custom but not the ties with Belgium. RADER was born on 14 September 1959.38 Its leader, Chief Bwanakweri, came from a small number of Groupe Scolaire graduates who were determined to “avoid the worst by easing the burden of the peasantry through social and constitutional reforms.” Chief Bwanakweri not only stood out among his cohorts as a man of his word, but he also stood rather alone. In 1956 he had already gone beyond declarations and dared to translate these aspirations into radical social reforms in his chiefdoms, thereby “filling the Hutu with hope.” Though Chief Bwanakweri had significant support among university students, his combined call for internal reform of Tutsi power and a soft line against Belgian power was enough to brand RADER as a pro-Belgium party among the Tutsi.39 Then came Mwami Mutara’s transfer—some would say internal exile—of Bwanakweri to Kibuye, a remote locality in western Rwanda. From then on, RADER, the only organized democratic tendency among Tutsi, was found on the margin of Rwandan politics.
For a while, RADER participated in several preelection meetings jointly with the two major Hutu parties, APROSOMA and PARMEHUTU. In the end, however, RADER had little influence on the outcome of the revolution. Faced with UNAR’s dogged determination to uphold Tutsi power, it came to be identified with a moderate tendency—not any moderate tendency, but a moderate Tutsi tendency. Its fate testified to the narrow social base of Tutsi reformism in 1959 Rwanda.40
The Mouvement Démocratique Rwandais/Parti du Mouvement et de l’Emancipation Hutu was created in October 1959 when Kayibanda transformed the old cultural movement, Mouvement Sociale Muhutu, established in June 1957, into a political party. It is not an accident that while the main Tutsi party claimed to be both “Rwandese” and “nationalist” in name, the main Hutu party claimed to be “Hutu” and “democratic” in the same name. It is also worth noting that PARMEHUTU did not start out as an antimonarchist party. It envisaged the possibility of a constitutional monarchy, but “insisted on a genuine democratisation of all existing institutions before the granting of independence.”41
L’Association pour la Promotion Sociale de la Masse distinguished itself from PARMEHUTU, both in name and in program: it claimed to be a party of the “masses” and not just a “Hutu” party. Created in November 1957, APROSOMA was a genuinely populist party whose appeal was aimed at the poor, at Hutu as well as the petits Tutsi.42 In the rapidly polarizing context of Rwanda in 1959, where power was Tutsi and the insurrection Hutu, it could not retain its original identity: “Since most Hutu were poor and the vast majority of poor were Hutu, the party ended up as a primarily Hutu party.”43
What were the differences between the two Hutu parties? We can identify three, based on the regional basis of their core support, the ideological content of the program each advanced, and the social character of the leadership.
The two Hutu parties had an overlapping yet distinct regional basis. PARMEHUTU derived its strength mainly from two parts of the country: the northern prefectures of Ruhengeri and Gisenyi, and Kayibanda’s home region of Gitarama in central Rwanda. APROSOMA’s activities were mainly focused in Butera in south-central Rwanda and Kinyaga in southwestern Rwanda. The difference partly reflected the contrasting historical trajectories leading to the incorporation of different regions into the Rwandan state. Whereas APROSOMA focused more on regions like Kinyaga, where Tutsi presence had preceded Tutsi power and thus had a history of Hutu/Tutsi relations preceding the polarization of the colonial period,44 PARMEHUTU’s stronghold was on those parts incorporated into the Rwandan state only on the eve of colonization. But this beginning did not remain a defining feature of PARMEHUTU. The more it became the party of the revolution, the more PARMEHUTU outgrew its regional beginnings and developed into a loose countrywide coalition of different locally based groups.45
The two parties differed in the program each advanced, particularly in their definition of who was the enemy. “The only point of divergence among the Hutu,” said a petitioner in 1959, “is whether the campaign should be directed against all Tutsi without distinction, against the high aristocracy, or against the specific abuses committed by certain representatives of the Hamitic race. Hence, it is mainly a question of tactics rather than of doctrine.”46 Not surprisingly, this difference in tactics proved key to deciding the future of Rwandan society. The more events highlighted the weakness of a reform tendency among powerful Tutsi, as they did with the internal exile of Chief Bwanakweri to Kibuye in late 1959, the more credible seemed the argument that the question of Tutsi supremacy was the core political problem facing Rwanda on the eve of independence. Beyond the immediate yet stubborn fact of Tutsi supremacy lay a second question of larger and deeper significance: What would be the place of the Tutsi in postrevolutionary Rwanda?
René Lemarchand has argued that behind the difference between the leadership of PARMEHUTU and APROSOMA lay a difference between two categories of intellectuals: the ex-seminarians and the Astridiens.47 Whereas the Astridiens possessed state skills and could expect to benefit from a constitutional transfer of power, the ex-seminarians did not and could only stake a future in the state through a revolutionary upheaval. The difference between PARMEHUTU and APROSOMA signified more than just a preference in style, the former standing for revolutionary methods and the latter for a constitutional process. In reaching out to unite Hutu and Tutsi poor against Tutsi privilege, the Aristidiens of APROSOMA stood for a popular Rwandan nationalism and held out the possibility of transcending Hutu and Tutsi as colonially constructed political identities. In contrast, PARMEHUTU sought to build on the colonial heritage by organizing the Hutu—all Hutu—against the Tutsi. APROSOMA failed, at least in part, because Tutsi privilege and Tutsi wealth were not the same thing. Although the wealthy were a minority among the Tutsi, Tutsi privilege was a legal/political arrangement that affected all Tutsi.
It is a historical tragedy that the conservative Tutsi continued to have the upper hand in court: it triumphed on 28 July 1959, in what has come to be known as the Mwima coup.48 When the heirless mwami died on 25 July, suddenly and mysteriously, the abiirupronounced a 24-year-old half-brother of the deceased the new mwami. As a man of poor political judgment, he was an ideal choice for a political figurehead. The appointment was made without any prior arrangement with Belgian authorities in an atmosphere filled with great suspense. This coup set the stage for the chain of events that contemporaries called the social revolution of 1959.
It is in the aftermath of the Mwima coup—and in anticipation of the next round of elections—that PARMEHUTU was created as a Hutu political party out of the old cultural association on 19 October 1959. Almost immediately there followed confrontations between PARMEHUTU militants on one side and militants of the promonarchy Tutsi party UNAR, and Tutsi chiefs in charge of the local state apparatus, on the other. These came to a head in and around Gitarama the next month: when news spread that a group of UNAR militants had attacked the PARMEHUTU leader Dominique Mbonyumutwa, violence spread over the country. The visiting UN Mission of 1960 estimated the killings at two hundred but added that “the number may be even higher since the people preferred to bury their dead silently.”49 The focus of the revolt were the Tutsi chiefs in the local authorities. Some of the chiefs were killed; others were forced to resign. Faced with clear indications that the Tutsi in power were about to unleash repression, Belgium declared a state of emergency and put the country under the command of Colonel B.E.M. Guy Logiest.
Before the first round of violence could swing from a Hutu revolt to a full-scale Tutsi repression, it was checked by emergency military action by the colonial power. But the emergency action did not stop at restoring order. It continued and took on the dimensions of a coup d’état in the local state. Arguing that the presence of Tutsi as subchiefs and chiefs “disturbed the public order,” Guy Logiest began to replace Tutsi with Hutu chiefs, thus shepherding a “revolution” against what had hitherto been the colonial power’s own local authorities. More than three hundred Hutu chiefs and subchiefs replaced Tutsi incumbents who had been deposed, killed, or had fled what was fast developing into a peasant revolt. The Belgian military decided to go a step further and augment a Hutu administration with an embryonic Hutu-dominated armed force: an indigenous military guard of 650 men was formed in May 1960.50 It was composed of 85 percent Hutu and 15 percent Tutsi. Without this reconstitution of the local state hierarchy as a Hutu hierarchy, it is difficult to explain the dramatically different outcome of subsequent elections—for it is the local administration that controlled the ballot boxes. Two tendencies gelled, one around PARMEHUTU, the other around UNAR. UNAR boycotted the 1960 communal elections, and was routed: PARMEHUTU, the party of Hutu power, won 70.4 percent of the votes as against 1.7 percent for UNAR, the party of the Tutsi monarchy.
With Hutu chiefs in charge of most local authorities, the newly elected advocates of Hutu power at the center were finally in a position to reorganize the central state. Known in Rwandan history as the coup of Gitarama, this reorganization was carried out on 28 January 1961. On that day, dozens of trucks from all around the country converged on the town of Gitarama, bringing to that destination precisely 3,126 communal councillors and burgomasters. According to Gatwa, another 25,000 people “spontaneously assembled to hear about the unusual event.” They were addressed by members of Rwanda’s provisional government: the minister of interior, the president of the Provisional Council, and the provisional prime minister. The gathering abolished the monarchy and proclaimed a republic. Then, sitting as a constituent assembly, the councillors and burgomasters proceeded to elect a president of the Republic. The coup was complete when the president called upon Grégoire Kayibanda, as prime minister, to form the future government.51 Most observers of Rwandan politics have presumed—rightly so, in my view—that such an open reorganization of the central state could not have happened without Belgian support.
How are we to understand the role of the “external” forces that intervened on the side of the revolution?52 Few would deny the internal significance of three “external” agents: the UN missions, the European clergy, and the colonial government. The triennial UN missions acted as catalysts, each time inviting a regular outpouring of grievances from different quarters. The European clergy came to function more or less as a backup force for the Hutu counterelite, providing it with everything from ghostwriters for manifestos and UN petitions to external contacts. On its part, the colonial government literally surrendered control over local government to the insurgents. Both the opportunity provided by the triennial UN missions and the support rendered by the European clergy and the colonial government were real and, at times, even critical. Without that support, there may have been no revolution, only a peasant revolt, joined to middle-class discontent. Yet, none of this made the 1959 Revolution any less real.53 We shall later see the same explanation surface in hostile accounts of the Rwanda Patriotic Front (RPF) capturing power in 1994: as with Belgium and Guy Logiest in 1959, the role of Uganda and Museveni would be highlighted to explain away both the RPF invasion of October 1990 and its capture of power in 1994 as the outcome of an external conspiracy. While the role of external forces was real in both cases, in neither case can it substitute for an understanding of the internal dynamics of social processes.
The year 1962 saw a change in government in Belgium. The new government agreed to cooperate with the UN demand for a fresh general election and a referendum on the monarchy in Rwanda—given UNAR’s rejection of the Gitarama coup. When the UN-initiated general election followed, UNAR decided to participate. By now, conditions had changed dramatically: the machinery that organized and oversaw the election was no longer Tutsi. As one would expect, Tutsi power was routed once again: PARMEHUTU won 77.7 percent of the votes against 16.8 percent for UNAR.54 A referendum held simultaneously led to an equally massive rejection of the monarchy in favor of a republican system of government.
THE BIRTH OF HUTU POWER
Their political fortunes dramatically reversed, the Tutsi political elite splintered between those who went into political exile and those who remained at home. Those who stayed tried to work out a rapprochement with the new power. In contrast, exiles were a mishmash of those determined to return to the old order and those who left because they feared there would be no room for Tutsi in the new political order. At the same time, this division corresponded to an ongoing discussion within the new Hutu political elite: between those determined that the new political community exclude Tutsi, and those exploring ways of going beyond the “racial” divisions inherited from the colonial order. To understand reasons for the shift of opinion on either side, we need to be aware of shifts on both sides.
Even though Hutu Power (in capital letters) emerged as a formal tendency only during the civil war of the early 1990s, I have used the term Hutu power (with “p” in lower case) also to characterize the tendency that stood for the making of a “Hutu Revolution” in the late 1950s—just as I have used the term Tutsi power to speak of those who argued that Tutsi had the traditional right to exercise power, even if they did not use the term themselves. The thread that unites both expressions of Hutu power, the informal and the formal, is an overriding conviction that the Rwandan nation is Hutu and, therefore, power in an independent Rwanda must also be Hutu. Tutsi may live in Rwanda, but only as a resident alien minority, at sufferance of the Hutu nation. For the Hutu who disagreed, Hutu and Tutsi, majority and minority, belonged to a single nation—Rwanda.
Two political tendencies—one accommodationist, the other exclusionist—vied for supremacy between 1959 and 1964. These tendencies did not correspond to the political divide between the Tutsi and the Hutu political elite, or between the revolution and the counterrevolution. Rather, both tendencies could be found on either side. The exclusionists included the adherents of Hutu power as well as the champions of Tutsi power, just as the accommodationists included all those who believed it was possible for Hutu and Tutsi to be part of a common political community. Seen from this point of view, the period from the beginning of the revolution in 1959 to the last exile armed attack in 1964 was marked by several turning points. As the balance of forces shifted, so did the center of gravity of state politics. Like a pendulum, it moved from exclusion in 1959 to accommodation in 1962 and back to exclusion in 1964.
1959 began with a sharp split between Hutu and Tutsi political leaders. Kayibanda, by then the leader of the revolution, called for Hutu power and for the exclusion of the Tutsi from political life. Faced with loss of political power, the Tutsi political elite moved into exile and began preparation for an armed return to power. The ground shifted in 1962 with UN intervention leading to the New York Accord and the formal announcement of state independence that same year. The New York Accord split UNAR into two rival factions: accommodationists and restorationists. As accommodationists returned to Kigali to participate in a coalition government, restorationists persisted with preparations for an armed invasion. The coalition government of the First Republic enjoyed eighteen months of relative peace before the restorationists mounted major armed raids in 1963. Countrywide repression followed in 1964. From 1959 to 1964, the center of gravity of Tutsi politics shifted from home to exile, as did the mode of its opposition from political to armed struggle. At the same time, there was a shift in the Hutu point of view from accommodation to exclusion. The outcome of this double shift was to consolidate Hutu power as the dominant tendency in the state apparatus and to restore Tutsi power as the dominant tendency in exile. To understand the dynamics behind this outcome, we need to look more fully at the developments between 1959 and 1964.
The call for Hutu power came from Grégoire Kayibanda, the leader of the revolution and the future president of the First Republic. Immediately after the November events that came to be known as the 1959 Revolution, he proposed that Hutu and Tutsi be “segregated” into two separate zones as a first step toward a “confederal organisation.” Citing Disraeli, he compared Hutu and Tutsi to “two nations in a single state”: “Two nations between whom there is no intercourse and no sympathy, who are as ignorant of each other’s habits, thoughts and feelings as if they were dwellers of different zones, or inhabitants of different planets.”55 Between the revolution of 1959 and independence in 1962, Kayibanda’s exclusionist point of view was moderated in the face of a growing coalition in favor of accommodation.
The outcome of the communal elections of June/July 1960—which UNAR boycotted—seemed to favor the exclusionist point of view. After the elections—in which PARMEHUTU won 70.4 percent of the votes, APROSOMA 7.4 percent, RADER 6.6 percent, and UNAR a paltry 1.79 percent—a provisional assembly of forty-eight members replaced the High Council (Conseil Supérieur) headed by the king. Though Kayibanda formed a “multiracial” provisional cabinet—with nine Europeans, seven Hutu, and three Tutsi56—PARMEHUTU was determined to go it alone beyond the formalities of the transition phase. Toward this end, it tried to erode the opposition through the lure of the carrot and the pressure of the stick. The result was a radical shift in the balance of forces within the country over the next three months. Alarmed that PARMEHUTU was willing to use any and all methods to consolidate its hold on power, the opposition joined hands: by November, the leaders of APROSOMA, RADER, and UNAR had come together to form a “Front Commun” in opposition to PARMEHUTU. A spokesperson of the Front Commun denounced the “dictatorial regime” of PARMEHUTU as “racist, racial and anti-democratic.” Accusing the regime of “deliberately attempting to crush all other parties through corruption and intimidation,” the statement concluded: “This kind of feudalism is worse than the old one.”57
Cutting across the Hutu/Tutsi divide, opposition unity not only brought pressure on proponents of Hutu power, it also brought important sections of UNAR into the fold of constitutional politics. The result was that when the UN-initiated parliamentary elections were held in September 1961, UNAR agreed to participate. The outcome—with PARMEHUTU gaining 78 percent of the votes and UNAR 17 percent—was polarized, the final result giving more or less a count of Hutu and Tutsi bodies in the country. When the new parliament elected Grégoire Kayibanda as president, UNAR abstained from the vote. Kayibanda, however, appointed a government with members coming from PARMEHUTU, UNAR and APROSOMA.58 Behind the scenes, discussions continued between the constitutionalist faction in UNAR, and PARMEHUTU. Following the New York Accord of February 1962, the leader of the “progressive” faction in UNAR, Michel Rwagasana, agreed to UNAR participation in a formal coalition government. The coalition took office in May—and remained intact for another eighteen months, until November 1963. UNAR had two ministries: public health and cattle. Rwagasana, its secretary-general, spoke publicly of peace on the hills of Rwanda. He pledged: “Our party can assure you that it will spare no effort in working for the achievement of a genuine understanding between the majority and the opposition, which, by virtue of its entry into the government, can no longer be considered an opposition, but rather a partner.”59 It was the clearest indication that the tide of postrevolutionary politics was turning from exclusion to accommodation.
The countertide also unfolded with the events of November 1959, which, as mentioned above, scattered most of the Tutsi notables into exile. The most complete account available of the period60 categorizes the exiles into three factions: the monarchists, the progressives, and the activists. Themonarchists comprised those who had been close to the court in Rwanda. The progressives came from the younger, Western-educated Tutsi chiefs, with some attracted to socialist ideas. The activists provided the hard-core guerrilla fighters for the armed incursions that began in 1960 and lasted until 1964. Until the February 1962 New York Accord that paved the way for the coalition government of May, “progressives” were in formal control of party affairs. The New York Accord broke the cohesion of the “progressive” faction; some, like the party secretary-general, joined the government coalition, but others remained in exile.
I have characterized the Tutsi political elite as either accommodationist or restorationist. The accommodationists both came to accept the 1959 Revolution as a fait accompli and banked on a peaceful internal political and constitutional process through which to work out their political future in the country. In contrast, the restorationists wished to undo the 1959 Revolution, holding out the prospect of a return to power through armed exile action. As the internal political process gathered momentum—as it did with the formation of the opposition front (the Front Commun) in November 1960, the elections of September 1961, and the coalition government that followed the accord of February 1962—anti-accommodation factions infiltrated small bands of armed guerrillas into border localities. Known as the inyenzi, or cockroaches, the armed guerrillas undertook as many as ten known raids into Rwanda.61 From the outset, the raids targeted the officials of the new power in the local authorities. However low their rank, Hutu officials were considered a legitimate target. But border raids invited cruel repression. And repression, too, began to assume a standard form: it targeted the local Tutsi population as active or potential support for the inyenzi. The cumulative outcome was to set in motion a dynamic counter to that of the internal political process.
The worst case of repression took place after the New York Accord of February 1962 but before the coalition government took office in May. Its location was the prefecture of Byumba. Two successive raids in 1962 had led to the death of two policemen in February and one policeman, two civil servants, and an ordinary Hutu in March. The reprisal came the day after the March raid: “Between 1000 and 2000 Tutsi men, women and children were massacred and buried on the spot, their huts burned and pillaged and their property divided among the Hutu population.”62 As the coalition government took office, the Byumba repression came to be an anomaly. But when the life of the coalition ended amid the expanded raids of 1963, the repression of Byumba provided a norm: a raid turned into a signal for the massacre of the local Tutsi population, and for the distribution of their property among those organized as the local self-defense group. Worse than anything that had happened during the revolution, repression joined political violence to redistribution of property, rewarding perpetrators with benefits.
During the first eighteen months of the First Republic, from May 1962 to November 1963, raids were a localized affair, as were the reprisals that followed. Local reprisals went alongside the politics of accommodation at the national level. The balance shifted radically with the inyenzi invasions of November and December 1963. Known as the Bugesera invasion, this particular raid reached nearly twenty miles outside of Kigali. The repression was swift, and it was concentrated in the prefecture of Gikongoro, which not only had a very high density of Tutsi, but also contained the former royal residence of Nyanza and was the core area of Tutsi opposition. Available reports indicate both that the killings began at the instigation of the local prefect, backed by higher state authorities, and that they involved enthusiastic popular participation. The prefect, André Nkeramugabe, is reported to have told an improvised meeting of burgomasters and PARMEHUTU propagandists: “We are expected to defend ourselves. The only way to go about [it] is to paralyze the Tutsi. How? They must be killed.”63
How many were killed? The figures vary widely, depending not only on the source but also on the time of writing. Before the 1990 civil war, estimates of the number killed had ranged from 750 to 5,000. The government officially estimated killings at around 750, but no other source believed it. Africa Contemporary Record reported unofficial estimates “nearer 5,000.”64 After the 1994 genocide, estimates tended to be much higher, ranging between 10,000 and 20,000. An international team65 estimated that between 5,000 and 8,000 were killed in Gikongoro alone, that is, 10–20 percent of the total Tutsi population of the préfecture. Catharine and David Newbury estimated that between 10,000 and 14,000 people were killed in the first few years after decolonization.66 Human Rights Watch put the figure at as many as 20,000.67
Many have claimed that the seeds of the genocidal violence that enveloped Rwanda in 1994 lie in the revolution of 1959.68 But the revolution was not a bloodbath. The highest contemporary estimate from a credible source of Tutsi deaths during the revolution is around two hundred. The fact is that it was not the revolution, but attempted restoration and the repression that followed, that opened the gateway to a blood-soaked political future for Rwanda.
Politically, the invasion gave the upper hand to the Hutu power tendency. And its proponents acted swiftly. Once the invasion was checked militarily, they arrested some twenty leading Tutsi personalities in the country and executed them a week later in the town of Ruhengeri in the northern part of the country. The victims included one of the two UNAR members of government (Etienne Africa), and its president (Rudisitwarane) and secretary-general (Rwagasana) inside the country. It also included the president (Bwanakweri) and vice president (Ndazaro) of RADER. As opponents of Tutsi power who had chosen to return home to work in postrevolutionary Rwanda, they were killed because they were Tutsi determined to participate in post-1959 politics as Rwandans. The repression meant an end to organized Tutsi politics in Rwanda until the political reforms under the Second Republic. But the real impact of the repression touched both Hutu and Tutsi. By killing leading Tutsi champions of the cause of accommodation and reform, those who had fought restorationists among the Tutsi in a tooth-and-nail struggle, the repression strengthened, at a stroke, the proponents of Hutu power within the country and those of Tutsi power in exile. The former heralded a native postrevolutionary republic in which the Tutsi would be tolerated only so long as they remained outside of the political sphere, whereas the latter held on to the notion that the Tutsi were a civilizing influence with a right to rule precisely because they were different. In reality, these postcolonial twins, Bantu and Hamite, were ideological offspring of Rwanda’s poisoned colonial past.