LENIN once chided Rosa Luxemburg about becoming so preoccupied with combating Polish nationalism that she could not see beyond it: he said she risked getting trapped in the world of the cat and the rat. This is a world in which no other animals matter. For the rat, there is no animal bigger than a cat: not lion, nor tiger, nor elephant. For the cat, there is none more delicious than the rat. In the presence of the other, neither has eyes nor ears for any other animal. It is in this sense that the world of the Hutu and the Tutsi was like the world of the rat and the cat. For the subaltern Hutu, as for the nationalist Tutsi, no other political reality was more definitive than that of the other.1
The 1959 Revolution was an outcome in the world of the rat and the cat. If you are not a rat, then you understand that there are many animals bigger and more dangerous than the cat: the lion is neither as distant nor as benign as it may seem to a rat. But, if youare a rat, then the cat is truly menacing and dangerous. When rats do triumph against cats, it is at once like the world turning upside down. The danger is that this upside-down world may not change in its relationships. It may still continue to be the same world: the world of the rat and the cat. But at second glance, one is not so certain: can the world where rats have belled cats remain quite the same world? How does the nationalism of the rat differ from the nationalism of the cat? How does the nationalism of the majority differ from that of the minority? Even if the majority should in time give rise to a self-interested elite, does not the very process of doing so evoke a difference between nationalism from below as opposed to that from above, between a popular as opposed to an elite nationalism, with different consequences for the society at large?
These questions help illuminate different points of view one finds on the 1959 Revolution in contemporary literature. For all the difference in emphasis, they reflect two standpoints. First, both major texts written in the shadow of the revolution take seriously the claims of the revolutionaries. Catharine Newbury highlighted the majoritarian credentials and egalitarian aspirations of those who took power after 1959, as did Lemarchand herald the majoritarian character of the revolution. Second, there are those who see 1959 as nothing but the displacement of the Tutsi elite by the Hutu counterelite, with the same world continuing, only upside down, and the revolutionary talk of the insurgent elite as so much opportunist demagoguery. Thus Gérard Prunier concluded his analysis of the 1959 Revolution in a book on the 1994 genocide: “What would later be touted as a ‘social revolution’ resembled more an ethnic transfer of power.”2 In contrast to Prunier’s dismissal of 1959 is the standpoint of Tharcisse Gatwa, a critic of the revolution’s practicewho nonetheless sympathizes with its perspective: the revolution, he says, was inspired by “a noble goal.”3 If Catharine Newbury and René Lemarchand fail to problematize the revolution, Prunier seems to shake it off a little too easily with a cynical shrug. If the innocent optimism of the former reflected the hopes of a revolutionary dawn, then the pessimism of the latter is cast in the shadow of the genocide. As I stated earlier, I find it important to take the revolution seriously, not to embrace it nor to shake it off, but to understand its limitations in spite of its real gains.
To understand the saving grace of the revolution, one would need to take the majoritarian aspirations of the revolutionaries seriously, and recognize that revolutionaries could not have seized and held power without bringing in at least some changes in response to popular aspirations. Without reform, revolutionaries could not have moved the majority into action. Such an analysis would need to go beyond the cynical observation that all the revolution did was to bring a counterelite to power; it would need to define the consequences of their coming to power to those not in power.4The point of view that focuses exclusively on the new elite risks turning into a tautology: since every social order is divided between elite holders of power and the mass subject to that power, this fact is taken as sufficient evidence that no society is different from another. To break out of that cynicism, but without innocently capitulating to revolutionary mythology or demagoguery, one needs both to accent the change in power and to highlight the consequences of the new power for those subject to it.
The revolutionaries made a threefold claim to underline the legitimacy of 1959. They heralded the revolution as national, as democratic, and as the harbinger of social justice. First, the mere shift in the identity of power from Tutsi to Hutu, they claimed, was a shift in representation. It was, obviously, a shift from a minority to a majority. More significantly, though, the claim was that the shift identified power with the indigenous Hutu rather than the alien Tutsi. Second, 1959 introduced elections at both the local and the national level. After the first direct election of burgomasters and their councils through a secret ballot in 1960, direct popular elections were held in each commune at regular three-year intervals.5 Finally, 1959 carried out a number of social reforms. The revolution abolished ubureetwa (forced labor) in 1959, as it did the right of chiefs to have any other kind of forced labor. It also appropriated igikingi, land assigned to Tutsi notables as pasturage by the king, and redistributed it to the landless. Unlike colonially appointed Tutsi chiefs who had the power to exact forced labor and personal services from Hutu—in addition meting out severe punishment to recalcitrant subjects—burgomasters and prefects under the First Republic lacked all these powers.
None of these claims were bogus. Yet, each was subject to a critique, more so as time passed. Each set in motion a countertendency that grew in time and led to a growing disenchantment with the new order.
REPRESENTATION AND JUSTICE IN THE HUTU REPUBLIC
Known after the name of his home village as the “Hermit of Gitarama,” Grégoire Kayibanda, the president of the First Republic, rarely made public appearances and hardly ever traveled abroad.6 He was said to be a model of frugality, one who wore shabby clothes and patched shoes. When he was driven, it was in a Volkswagen. The republic he created did not just claim to represent the majority; it claimed to be the republic of the entire nation, that is, the Hutu nation. After 1964, Tutsi presence was forcibly removed from the political arena; the Tutsi were found in education, in business, in the church, even in government employment, but not in the political arena. The political sphere was confined to the Hutu, members of the Hutu nation.
For the postrevolutionary power that was the First Republic, Rwanda was exclusively a Hutu state. The rationale for this was disarmingly simple, disarmingly so because it simply turned upside down the logic of the colonial state: the Hutu were indigenous, the Tutsi were alien. Whereas the Tutsi had been treated preferentially by the colonial state as a nonindigenous civilizing influence, the First Republic considered this claim reason enough to treat them as politically illegitimate. The Tutsi thus continued to be officially defined as a “race,” never as an “ethnic group.” The implication was crucial. The language of race turned around the distinction “indigenous/alien”: a racial difference could only be with foreigners, whereas an ethnic difference was with locals. As “race” was said to distinguish the indigenous from the nonindigenous, “ethnicity” was said to separate different groups among the indigenous. As such, the political distinction between a minority and a majority could only be relevant within the ethnic domain. As a race, the Tutsi were not a political minority; they were politically foreign, as it were, resident aliens. Only the indigenous—ethnic groups—could rightly belong, fully, to both civil and political society. Those classified as nonindigenous would be seen and treated as civil beings, with limitations (discriminations) depending on context, but never as political beings. From this point of view, the view of the First Republic, the Tutsi could expect to participate in civil society, but must not transgress into political society.
The Hutu Republic clearly distinguished between internal and external Tutsi, as one would between resident and nonresident aliens. This was so from the time of the inyenzi attacks of 1963–64. Those outside were considered more of a political than a cultural diaspora. Seen as organizing to overthrow the Hutu republic and to establish Tutsi power, they were defined as a permanent threat to the state and were treated as permanent outsiders. In contrast, the internal Tutsi were tolerated as civic, but not political, beings who could aspire to rights within civil society, but not in political society. Once the internal Tutsi parties had been liquidated and their leaders killed, Kayibanda’s political ammunition was directed at APROSOMA, the one Hutu-led revolutionary party that had tried to recruit and organize the petits Tutsi into its ranks. Between 1964 and 1967, APROSOMA stalwarts were “slowly but surely eased out of any political or administrative responsibility.”7 Thus Kayibanda reorganized the Rwandan state as exclusively a state of the Rwandan nation.
Ironically, the Kayibanda regime faced growing criticism in the 1960s that it had not done enough to advance Hutu representation in civil society. The critique first came from unemployed Hutu school leavers.8 Many had left primary school for lack of resources and were ploughing urban streets looking for employment. Others, in spite of being degree holders, lacked employment, to which they no doubt felt entitled. The combination created a pool of agitators, ready to be tapped by an ambitious politician with a keen sense of fresh grievances. This pool of educated discontent had grown sufficiently by the mid-1960s to surface publicly. In 1966, the party newspaper Urmuli rwa Demokrasi (Light of Democracy) dismissed them summarily as comprising one of four categories: “PARMEHUTU zealots, imbued with the ideas of 1959”; those “conceited and selfish (and) who only seek to accumulate honours, privileges and wealth”; “wind-cocks, who have one foot in PARMEHUTU and the other in the air”; and “those whose only objective is to destroy PARMEHUTU.”9
The critique focused on government policy in education and employment. It was said that the government was not doing enough to advance Hutu representation in education: while Hutu were the majority in schools, university enrollment in the middle and late 1960s was nearly 90 percent Tutsi.10 This was so in spite of government policy restricting Tutsi enrollment in postsecondary institutions to 10 percent of the overall figure. The critique no doubt put great pressure on government to take control of the educational system that was mainly under Church control. It is ironic that the political leadership nurtured by the Church in the 1950s established state control over church-directed education in the 1960s and took the initiative to secularize the educational system. To make sense of this, one needs to keep in mind that whereas the political leadership nurtured by the Church in the 1950s was Hutu, the Church leadership in postrevolutionary Rwanda continued to be predominantly Tutsi. The Church was one of the two institutions—the other being business—with a Tutsi preponderance in the postrevolutionary era.
The consequence of the school-leavers’ agitation was the law of August 1966, which established state control over education through four key provisions.11 First, it declared as state property all school buildings ever constructed with state subsidies. Second, it placed the hiring and firing of all personnel, lay and religious, in all state-subsidized private schools under the supervision and control of the state. Third, it removed the admission, promotion, and expulsion of students in these schools from exclusive control of school authorities. Finally, it also removed the choice of textbooks and curriculum content from the sole jurisdiction of school authorities. The aim was to bring the entire private—i.e., predominantly Catholic—educational system under state control. It was also the effect. For the Church-educated Hutu leadership of the state, the 1966 law provided an instrument for Hutu-izing control over a Tutsi-dominated educational system.
When it came to employment, critics held that the government’s record was even worse. Since there was no official policy requiring adequate Hutu representation in employment, even the modest increase in Hutu school enrollment registered in a few years as a dramatic expansion in the number of unemployed Hutu school leavers. The trend gathered momentum over the years. By early 1970, for example, there were roughly three hundred Hutu students among the five hundred at the National University. The more Hutu enrollment increased in postsecondary institutions, the more dramatic was the rise in the numbers of educated Hutu who were unemployed. It is this volatile group that triggered the dynamic that led to the coup of July 1973. The context for the crisis was created by the massacre of hundreds of thousands of Hutu by the mostly Tutsi army in neighboring Burundi. Those butchered were mostly school-going youth and were estimated at around 200,000.
The effect was to reignite the “racial” tension in Rwanda’s middle class.12 Hutu students at the National University in Butare began to agitate against their Tutsi colleagues. Lists of “blacklisted” Tutsi students, signed by anonymous and self-appointed Committees of Public Safety, appeared on notice boards at the university, at the National Teacher Training College, and at a secondary school. Among those targeted were children of mixed marriages (ibyimanyi) and “cheaters” (abaguze ubwoko) who were accused of having changed their “racial” affiliation. The threat was enough to get the bulk of the two hundred Tutsi students at the university to leave. Soon, lists began to proliferate at places of employment—banks, parastatal companies, private businesses—and even at an embassy. Once again, many of the named Tutsi employees left. Radio Rwanda was used to broadcast appeals inciting the Hutu to rise up and avenge themselves. Some extreme proponents of Hutu Power began publicly to call for a “final solution” to the Tutsi question. In the rapidly inflamed situation, a number of Tutsi were murdered. The estimates of the dead ranged anywhere from six to five hundred. The government responded sluggishly. Only after disturbances had gone on for weeks did it move into action. It first warned students to keep protest within “reasonable limits,” then threatened to expel unruly students, then warned of “a subversive group which is trying to cause anarchy,” and only then did it declare that it would not tolerate “discrimination against individuals because of their race.”13
The agitation expanded in concentric circles. Just as it had begun in educational institutions and moved into ministries and enterprises, it grew by ripple effect, with each ripple bringing into the fold yet another tension in Rwandan society. The paralysis of power brought to surface the tension both within power and within society: the former between Hutu of the north and those of the south, and the latter between the poor and the rich. The general population began to expand the attack against the Tutsi into an attack on the rich. In the dual context of a popular agitation against deprivation and a student critique of Hutu unemployment, the split in power between the north and the south became the source of imposing a new order from above. Major General Juvénal Habyarimana led the army to carry out a bloodless coup on 5 July 1973. By all accounts, the intervention was received with great relief by all concerned, whether the population at large or the students, whether the business community or officialdom; whether Hutu or Tutsi. Thus was born the Second Republic, which immediately declared itself the custodian of the revolution and the protector of all its children, Hutu as well as Tutsi.
The Second Republic claimed to complete the “national” revolution of 1959 through a “moral” revolution.14 The key change from the First to the Second Republic—a change that seems to have gone unnoticed by many an observer of Rwandan politics—was a shift in the political identity of the Tutsi from a race to an ethnic group. While the First Republic considered the Tutsi a “race,” the Second Republic reconstructed the Tutsi as an “ethnicity” and, therefore, as a group indigenous to Rwanda. We have seen that the language of race turned around the distinction indigenous/nonindigenous. The political distinction between a majority and a minority had little relevance within the domain of “race.” For, as a race, Tutsi were simply foreign. Their numbers were of little significance. Once reconstructed as an ethnicity, however, the Tutsi became Rwandan and their numbers became significant, just as the minority/majority distinction also became of great relevance. As a “race” under the First Republic, the Tutsi had been confined to the civic sphere and barred from the political sphere; as an “ethnicity” under the Second Republic, however, they were allowed participation in the political sphere, but limited to a scope said to befit their minority status.
The Tutsi faced discrimination, not just in the political sphere, but in the civic sphere, too. From the point of view of the Second Republic, the Tutsi were not just any minority, but a historically privileged one. As a minority defined statutorily and identified legally, its participation in civil and political life was regulated by state policy. The regulation had two purposes: (1) to redistribute through affirmative state action, and (2) to limit political participation. The Second Republic followed a “national” goal and sought to arrive at a balance between two tension-ridden objectives: justice and reconciliation. Reconciliation with the Tutsi was to be in a context of justice for the Hutu. The mode of justice would be through a system of redress within hitherto Tutsi-dominated institutions, particularly the Church, education, and employment. In some instances, as with the Church, this included direct state pressure. Such, for example, was said to be the reason for the “resignation” of Father Muvala a few days before he was to be ordained as bishop of the Catholic diocese of Butare.15 In other instances, such as education and employment, it was mainly through state-enforced quotas. The rationale was to redress historical wrongs. As such, this mode of justice was extended not only to the Hutu, but in particular to Hutu from the northern region, who were considered historically the most underprivileged.
The affirmative action program was summed up as “équitable, ethnique et régional.” The two moments of the affirmative action program, the “ethnic” and the regional, reflected a tension between the two moments of the program for justice: justice as appropriation, and justice as redistribution. The moment of appropriation targeted the Tutsi, and tended to unite Hutu against Tutsi. The moment of redistribution, however, distinguished between Hutu on the basis of regional affiliation: it thus tended to drive a wedge between Hutu from the north and Hutu from the center and the south. The 1985 law on education captured the unifying character of the politics of appropriation:16 it stipulated that selection into schools will take into account the “ethnic” affiliation of the child; the Hutu will receive over 85 percent of the places, the Tutsi between 10 and 15 percent, and the Twa 1 percent. The quota system that defined the parameters for individual appointments to civil service posts—whether by law or by decree—summed up the divisive character of the politics of redistribution.17 Allocation of posts was to take place, first on a regional basis and then on an “ethnic” basis. In practice, then, 60 percent of posts would be allocated to northerners and 40 percent to southerners. Within each region, allocation would be divided between Hutu and Tutsi/Twa, the former receiving 90 percent and the latter 10 percent.18
In practice, the quota system was subject to a number of limitations. Among the small coterie of Rwanda specialists, there is disagreement as to how strictly the quota system was enforced in practice. Citing statistical evidence from L. Uwizeyimana, Fillip Reyntjens maintains that the quota system was not rigorously enforced, as reflected by the fact that the proportion of Tutsi in public, parastatal, and private sector employment greatly exceeded the quotas, which were informal, anyway.19 Yet, some members of the predominantly Tutsi business community did complain in 1977 that government promotion and hiring practices tended to give disproportionate weight to regional and ethnic qualifications rather than to merit. In response, Habyarimana made it clear that he would not relax his quota policies, arguing that it was important to overcome historical socioeconomic disequilibria in Rwandan society.20 Finally, the quota system did not apply to the foreign diplomatic and business community in Rwanda, an important source of lucrative employment in the nongovernment sector. On the basis of her observations in Rwanda in the early 1980s, Claudine Vidal reported that a disproportionate number of Tutsi were employed by foreigners in Rwanda.21
Even as he pursued an agenda of redress for historical wrongs, Habyarimana was publicly committed to a policy of “reconciliation” between Hutu and Tutsi.22 The Second Republic marked the day of the coup, 5 July, as “a day of peace and reconciliation.”23This was not an empty boast or a rhetorical gesture. Although the state had yet to work out a policy of reconciliation toward the Tutsi in exile, it took several concrete steps toward reconciliation between Hutu and Tutsi within Rwanda. Official vocabulary began to speak of Hutu and Tutsi as “ethnicities,” no longer as “races.” The meaning of the shift was clear: the Tutsi within were there to stay. As opposed to the last year of the First Republic, when many Tutsi were being forced out of their jobs, Tutsi fears were being allayed in the first year of the Second Republic. The new regime rejected the First Republic’s “national Hutuism.” From being banished from the political sphere under the First Republic, the Tutsi were brought back within the political fold. When Habyarimana announced the formation of his cabinet on June 1, 1974, it included, for the first time since 1964, a Tutsi: André Katabarwa.24As astute an observer of Rwandan politics as René Lemarchand ended an analysis of the Second Republic in 1975 with the following conclusion: “If power in Rwanda is still the monopoly of a specific ethnic segment, identified with the Hutu sub-culture, the prospects of a Hutu-Tutsirapprochement, both within and outside Rwanda, have never been brighter since independence.”25
The distinction was no longer between civil and political rights but between participation and power. As a statutorily defined minority, the Tutsi could have rights, but must give up all thought of any meaningful participation in power. For power must remain Hutu, since Hutu was the identity of the statutorily defined majority. Tutsi participation in politics was regulated, subject to an informal quota, as was Tutsi participation in the formal economy. When speaking to Philippe Decraene of Le Monde about relations between Hutu and Tutsi, Habyarimana explained that “hatred cannot dissolve overnight” and clarified his notion of reconciliation: “It is not a question of bringing the Tutsi back to power, which would be equivalent to re-establishing the pre-1959 situation; but each ethnic group has its place in the national fold. There is a Tutsi minister in my government; there are Tutsi senior civil servants in the administration; and Tutsi officers in the army.”26
Katabarwa remained a member of the cabinet until the reshuffle of January 1979, when he was dropped.27 The first major internal struggle within the regime took place in 1979–80. Major Théonaste Lizinde, the security chief, was replaced in November 1979 and was accused in April of the following year of plotting a coup d’état. Significantly, Lizinde was also the author of a book entitled The Discovery of the Kalinga, an anti-Tutsi tract describing the search for the Karinga drum. The Lizinde faction was “considered to be violently anti-Tutsi and unhappy with Habyarimana’s efforts to reconcile the Hutu and the Tutsi.”28 While Habyarimana turned the coup attempt into an opportunity to liquidate physically the southern opposition—literally “the entire generation of revolutionaries from the south”—this did not halt his search for reconciliation between Hutu and Tutsi.29 In the January 1982 elections for the National Development Council—in effect the national legislature in a single party arrangement—the ruling party, Mouvement Révolutionnaire National pour le Développement (MRND) offered the electorate twice as many candidates as there were seats. It was meant to be an exercise in inner-party democracy. Among the 128 candidates put forth, there were two Tutsi, two Twa, and nine female militantes. Concluding his annual survey of Rwanda, René Lemarchand observed: “Rwanda remains one of the very few states in Africa where democracy retains a measure of reality.”30
Even if limited and qualified, Tutsi participation in the political sphere continued. In October 1990, when the Rwanda Patriotic Front (RPF) invaded Rwanda, there was one Tutsi minister in a nineteen-member cabinet, one Tutsi ambassador, two Tutsi deputies in a seventy-seat national assembly, and two Tutsi in the sixteen-person central committee of the country’s only party, MRND.31 The flip side of the Tutsi presence in the central state was that the Tutsi were carefully kept away from the organs of power: the army and the local state. While there was one Tutsi officer in the army, members of the army were prohibited by regulation from marrying Tutsi women. Similarly, there was an almost total absence of Tutsi from the organs of the local state: there was only one Tutsi prefect, the prefect of Butare who was killed in the genocide, and not a single Tutsi burgomaster.32 Though the Tutsi were defined as an ethnic minority and integrated as such within the central state, they were denied group recognition and were almost totally excluded from power in the local authority. Here, the gain of the 1959 Revolution—which had replaced Tutsi chiefs by Hutu—held. Tutsi gains under the Second Republic were more in the civic sphere than in the local authority. Its significance can be grasped from the fact that no major anti-Tutsi political violence was reported from Rwanda between the time Habyarimana came to power in 1973 and the onset of the war with the RPF in 1990. Many thought Habyarimana popular among the Tutsi of the interior throughout that period. Popularly, he was considered “the protector of the Tutsi.”33
Only a Hutu, someone not a Tutsi, would be a “protector of the Tutsi.” The flip side of “protecting” the Tutsi was to keep the Tutsi legally identified as the Tutsi. Thus, Habyarimana defined the prospects and limits of the Second Republic. The prospect was to rehabilitate the Tutsi to being Rwandans, alongside the Hutu. The shift was signified both by the change in the official designation of Tutsi from a “race” to an “ethnicity” and by lifting the ban on Tutsi participation in the political sphere. The limit was that Tutsi and Hutu would remain alive as political identities. If Habyarimana had the political courage to come to grips with the colonial racial legacy, he lacked the political foresight to transcend fully the combined legacy of Rwandan state formation—colonial and pre-colonial—which had crystallized Hutu and Tutsi into binary political identities.
REVOLUTION AND THE COLONIAL STATE
The processes that gave birth to the First Republic included revolutionary pressures from below. At the same time, the state was reorganized to make revolution from above. Whereas the revolutionary legacy was best captured in the ideology of the state, the colonial legacy was best reproduced in its institutions. The distinction is analytical. In practice, of course, there was no Chinese Wall dividing ideology from institution and language from practice. Just as there were institutions which claimed a revolutionary legitimacy, there were others that were considered part of an ongoing tradition and continued to function in a more or less humdrum manner. In achieving a synthesis of colonial tradition and revolutionary initiatives, the First Republic had more of a transitional significance. It is the Second Republic that represented the crowning achievement of the contradictory construct that was the Hutu Republic.
Rwanda under the First Republic was an ideological state, somewhat like Cuba, North Korea, and Israel, as Prunier says.34 But Prunier is wrong to think that this was equally true of the Second Republic. True, when the Second Republic set up the single party, the party did organize community-level displays of loyalty and support, called “animation,” with symbols borrowed from North Korea.35 But the analogy with North Korea misses a key shift in the political agenda of the Second Republic, a shift that made the agenda far less ideological. Unlike the First Republic, the Second Republic accepted the Tutsi within the political sphere, including them in the single party once it was organized.
In the ideological construction of the Hutu Republic, no institution had a greater significance than did the army. Colonial Rwanda—like colonial Somalia—had no army; in times of emergency, order was supposed to be ensured by the Force Publique of colonial Congo. The Rwanda army was built overnight, on the eve of independence, and from scratch, in a context when Belgian policy had shed its pro-Tutsi stance and neighboring Congo, itself on the verge of independence, could not be expected to ensure order in times of crisis. The 650 who formed the core of the Garde Territoriale in May 1960 comprised 85 percent Hutu and 15 percent Tutsi. As the one institution of the state that claimed to be authentically Hutu from birth, the army was both a child and a defender of the revolution. From the outset, the army claimed the mantle of the “nation-in-arms.”36
While the ideological construction of the revolutionary state was most visible in the language of the central state—particularly in the army and the single party—the institutional reproduction of the colonial state was clearest in the administrative reorganization of the local state. The politico-administrative apparatus of both the central and the local state was reorganized following a protracted constitutional discussion after the 1973 coup. The discussion led both to the founding of the single party and to the adoption of a new constitution that put a seal of approval on the reorganization of local state administration. The MRND was established as the single party in 1975.37 Legislative, executive, and judicial powers were exercised through the MRND and were centralized in the hands of the president. At the summit, the president appointed the Central Committee of the party. At the base, the party embraced every Rwandan, since every Rwandan was by law a member of the party, as indeed was the case in neighboring Tanzania and Zaire.
A draft constitution was promulgated in 1977 and was adopted on 17 December 1978. As a complete text, the draft constitution38 drew on three different experiences: the French constitution for the structure of government, the Belgian model for human rights provisions, and the Tanzanian model for the organization of a single-party state. But the interim discussion focused mainly on which “model” to follow in constructing the party-state: the Zairean or the Tanzanian. While both models involved the building of a party-state, the difference was this: whereas the party controlled the state in the Zairean model, the accent in the Tanzanian model was on the state rather than the party. Habyarimana favored the Tanzanian model and managed to convince his comrades of its virtues.
The Tanzanian influence proved key to defining the character of the local state. It provided the model for the communal reform that was initiated in 1960 and completed under the Second Republic. Inspired by the ten-cell system in Tanzania, the reform divided the country into ten préfectures, and these into 143 communes of about 30,000 persons. Each commune comprised eight collines (hills), with each hill divided into secteurs, and each sector into two cellules. Each cell was to comprise fifty familles. As Reyntjens observed in 1985, few African countries could claim to be so well organized administratively as was Rwanda under the Second Republic.39
The reorganization both reproduced the administrative structures of the colonial state and built on them. In doing so, it did away with the electoral reforms of the First Republic. Neither the prefect nor the burgomaster, in charge of the prefecture and the commune, respectively, was elected; both were administratively appointed from above. The prefect was like the colonial chief: he decided how many acres of coffee should be cultivated in each commune, and how many people should be put to work on the roads and for how long. He alone was responsible for maintaining public order and tranquility and he had the authority to suspend the execution of ministerial orders when circumstances so required. Granted more or less the powers of the colonial subchief, the burgomaster all too easily tended to reproduce practices familiar in the colonial era: he demanded gifts in return for administrative services, from settling a case to penning a signature.
We shall later see that the administrative machinery of the local state was key to organizing the series of massacres that constituted the genocide of 1994. Between 1973 and 1994, however, this machinery grew on a steady diet of coercive practices, whose means were justified as “customary” and ends as “developmental.”
CUSTOM AND THE DEVELOPMENTAL STATE
Postrevolutionary Rwanda could point to important economic achievements. It had come a long way from 1976, when it had a per capita income lower than that of any of its neighbors. By 1990, however, the World Bank estimated that the per capita income of Rwanda was higher than that of any of its neighbors. By 1987, Rwanda had the lowest debt, the lowest inflation rate, and the highest rate of growth of the Gross National Product (GNP) of any country in the region. The share of primary activities—mainly subsistence agriculture—in the GNP had declined from 80 percent in 1962 to 48 percent in 1986. At the same time, secondary activities had risen from 8 percent to 21 percent, and services from 12 percent to 31 percent. The rate of mortality was down. Hygiene and medical care indicators were improving. The proportion of children in school had gone up from 49.5 percent in 1978 to 61.8 percent in 1986. There had been no political executions since 1982, and there were fewer political prisoners than in most African countries. The record was impressive.40
The record was particularly impressive in three sectors: agriculture, reforestation, and infrastructure. Each of these developments was in some way linked to a set of relationships set in motion by the revolution. The most benign of these was the development in infrastructure, the result of a government tapping possibilities opened up by changes in Rwanda’s external relations, mainly the shift from a bilateral colonial relationship with Belgium to diversified foreign relations. The main roads branching from Kigali to the country’s frontiers were asphalted with the assistance of foreign aid. This included the road to the Burundi border (European Development Fund aid), to the Uganda border (World Bank aid), to the Tanzania border (Chinese aid), and to the Zaire border (West German aid).41 Less benign were developments that were triggered mainly by internal changes ushered in by the revolution. New relations, or newly recast relations, set new energies in motion, underpinning dramatic developments in different sectors. But then each entered a period of crisis either as the relationship in question turned sour, or as the energies unleashed earlier got exhausted. The clearest example of this kind of postrevolutionary dynamism is that of agricultural production, which increased through individual effort, and reforestation, which was the result of communal effort.
Agricultural production increased dramatically in the first two decades and suffered sharply in the following decade. Between the early 1960s and the early 1980s, Rwanda was one of only three sub-Saharan countries that succeeded in increasing total food production per capita.42 This is extraordinary, especially if one keeps in mind that Rwanda had experienced seventeen years of famine between 1900 and 1950. The Rwandan population by the middle of the 1980s was roughly five times the average during the colonial period, over eight million as opposed to between one and two million; yet, it was better nourished and had been free of famine for thirty years.43 How did this happen? It happened mainly through the expansion of crop area. The 1959 revolution abolished the “customary” control of Tutsi chiefs over extensive land areas, particularly the igikingi areas that had been reserved as pasture for cattle. This made possible two developments.44 First, large numbers of landless and land-poor peasants were brought in from the hills to settle in the drier, low-lying savanna areas in the south-center and southeast of the country. Second, fertile but swampy valley bottomland, previously reserved as dry-season pasture, was drained and brought under cultivation. L. Cambrezy estimates that, in 1978 alone, roughly 800,000 of the population of 4.8 million had been so resettled.45
By the close of the 1980s, however, the World Bank was citing Rwanda as one of the three worst performing sub-Saharan countries when it came to food production.46 By then, there was hardly any land left for crop expansion. In the absence of any technological breakthrough, and in the presence of an increase in sheer numbers, soil fertility was decreasing. In response, there was a shift away from cereals and beans toward root crops: the food basket was becoming protein-poor and starch-laden. Yet, there still remained a potentially large area for crop expansion, one that would not be tapped until after 1994. This was the land taken up by the national parks system. Estimated at more than 20 percent of the country’s total land mass, it was said to constitute the highest proportion of land mass reserved as park land in any single country on the entire continent of Africa.47
If the revolution did away with one kind of custom—a monopoly over land for Tutsi chiefs—it built on another kind of custom: the right of the state to call on corvée from peasants. “Inactivity” was one of the three social evils the Second Republic pledged to eradicate at the very outset.48Asked by the Paris daily Le Monde to elaborate on his plans for democracy, Habyarimana said: “First the population must get down to work—the Government and myself want to emphasize the value of work on the land. Thus we shall devote each Saturday to tilling the soil with hoes in our hands.”49 Forced labor for the state, usually once a week and usually on Saturday, became an institution, a part of the “custom” initiated by the revolution. Called umuganda, it was formally launched by the president in February 1974,50 and was often explained away in the literature as “cooperative communal labour.”51 This bit of compulsion was used to marshal unpaid labor for public projects, such as planting forests, constructing terraces to fight erosion, and building bridges.
Umuganda went alongside paysannat, another standard feature in the agenda of developmental dictatorships that constituted the “radical” path in Africa’s postcolonial trajectory. The point about developmental dictatorships was that they seldom claimed to be democracies. Their historical justification for soft-peddling democratic rights was always that they were the best instrument to bring about development. Thus, even when a parliament was created in Rwanda in 1978—not until five years after the closure of the National Assembly—it was called Conseil National de Développement, nothing less and nothing more than a National Council for Development. Both umuganda and paysannat were justified as developmental initiatives. The difference was this: while umugandamarshalled unpaid labor toward public works projects, the paysannat was akin to the compulsory villagization that had become standard practice in “radical” states in the region such as Tanzania and Ethiopia. In Rwanda, where the population was relatively concentrated—unlike in Tanzania and parts of Ethiopia where it was not—the paysannat was less a way of concentrating population, more a way of bringing wasteland under cultivation.52
The radical trajectory in postcolonial Africa tended to reproduce the coercive and enabling aspect of the colonial tradition as “revolution.” From this point of view, umuganda was really, as Scott Grosse observed in his report to USAID, “a program of forced investment by farmers along the lines of coercive colonial-era conservation policies.” Each public works project was a compulsion, and peasants were often doubtful of its social utility, more so as time went by. This much became clear with the onset of “democracy” in 1991 and 1992: farmers revolted against forced communal work, and the government abandoned the umuganda system.53 In some cases, as in the south of the country, peasants even destroyed the communal wood lots they had been coerced into planting and maintaining.54 The peasants’ revolt unmasked official claims that umuganda was cooperative communal labor. Ironically, the Rwanda section of the 1993 Human Rights Report of the U.S. Department of State, issued in February 1994, had this to say on the subject of “forced and compulsory labor”: “Forced labor is prohibited by law and is not known to occur in practice.”55
The Second Republic began to unravel from about the end of the 1980s. The context of that development was both internal and external. The external dimension feeding the post-1985 resource crunch in Rwanda accelerated with the multiplication of forces that fed it: coffee prices plummeted from 1989, a Structural Adjustment Programme was imposed from outside in 1990, and military spending rose dramatically following the RPF invasion, also in 1990. Coffee prices dropped by about 50 percent in the summer of 1989. At the same time, a disease affecting coffee trees had begun to spread in some parts of the country. In spite of government regulations banning the cutting down of coffee trees, farmers uprooted an estimated 300,000 such trees in the 1990s.56 Income from coffee exports fell from $144 million in 1985 to a meager $30 million in 1993.57 For the coffee farmer, however, the decline in coffee receipts was the outcome of two factors: not only a decline in the volume of coffee exported and its international price, but also a sharp devaluation of Rwandan currency following the imposition of an International Monetary Fund (IMF)-designed Structural Adjustment Programme (SAP) in 1990. When the currency was devalued by 67 percent in 1990, the government was simultaneously able to reduce drastically the real price of coffee to the farmer and to disguise this reduction by limiting the decrease in the nominal price of coffee. In this context, even a modest shielding of the coffee growers put the government in conflict with the IMF. The government thus reduced the nominal price from 125 to 100 Rwandan Franc (RWF) per kilo in 1990 and then unilaterally raised it to RWF 115 per kilo in 1991, even though the impact on the budget was adverse. The deficit increased from 12 percent of the Gross Domestic Product (GDP) in 1990 to 18 percent in 1992 to 19 percent in 1993, though, the Stabilization Programme had set 5 percent as the target for a reduced budget deficit in 1993.
Even this modest effort to cushion the impact of sharply declining prices on the coffee producer exposed the government to the ire of the IMF, and to the charge that it was in breach of the stabilization agreement. With two key measures in the Structural Adjustment Programme—eliminating coffee subsidies and reducing the budget deficit to target—not fully implemented, the Bretton Woods organizations refused to follow the stick of adjustment with the carrot of financial resources.58 The World Bank refused to provide the second tranche of structural adjustment credit. Rwanda’s resource crunch worsened. According to estimates provided by the Economist Intelligence Unit, Rwanda’s real GDP fell by 5.7 percent in 1989, bringing it below the 1983 level. The slide in GDP continued, by a further 2 percent in 1990, and another 8 percent in 1993 alone. By then, Rwanda’s debt had rocketed to nearly a billion U.S. dollars—precisely $941 million—up from $189 million in 1980.59 The GDP per capita, estimated at $330 in 1985, plummeted to $200 in 1989.60 It was a 40 percent fall in only four years. These are, however, averages. For working people, the consequence was much worse, not only because averages hide internal disparities between the rich and the poor. To make matters worse, an increasing share of declining resources was diverted to the war after 1990. Military spending quadrupled from 1989 to 1992, from 1.9 percent to 7.8 percent of GDP. Even before the sharp decline in coffee prices registered in 1989, the south and southwest of Rwanda had already suffered a serious famine, the first since 1943. As adverse internal and external trends came together from 1989 onwards, it was as if Rwanda was plunging free fall into a nightmare. And yet, we need to keep in mind that there was nothing exceptional about the economic crisis that beset Rwanda. It was shared by many an African economy undergoing IMF-supervised structural adjustment. It is not the economic, but the political, crisis that would set Rwanda apart from its neighbors in the decade to come.
FROM REVOLUTIONARY UNITY TO DEMOCRATIC OPPOSITION
To understand the critique of the revolution after 1959, one needs to begin with the revolutionary critique before 1959. Representation and justice were not only the foundation stones of the Hutu revolution and the Hutu republic, they were also the basis of the critique leveled at the postrevolutionary republics. As in other instances, postrevolutionary Rwandan history can be read as one of an unraveling tension between ideals and practice. After all, the claim of the two republics, their raison d’être, was that their core energies had been devoted to translating the principles of 1959 into policies. But implementation inevitably bred its own consequences, not only those intended but also those not. In time, these provided the basis of a critique of the postrevolutionary republic. I shall make a distinction between the internal and the external critique. The internal critique began from within the ranks of the regime; as reform set in, it found other locations within society. In contrast, the external critique came from Tutsi exiles; it gathered momentum with the RPF invasion of October 1990. We shall focus on the internal critique in this chapter, leaving the external critique for the next chapter.
Organizationally, the internal critique was made from a variety of locations, at first from within the ruling party, then from a mushrooming of civil society organizations, and finally, from an ensemble of political parties. The critique came in two different phases. The first phase began with student protests under the First Republic. Critics charged that the government had confined the revolution to the political sphere, failing to end Tutsi privilege and redress Hutu grievances in civil society. After the coup of 1973, this kind of critique led to coup attempts by disaffected members of the Habyarimana government who opposed the redefinition of the Tutsi from a race to an ethnicity as a prelude to their rehabilitation in Rwandan society, claiming that this was evidence that the regime was “pro-Tutsi.” The best-known example of this is the Lizinde coup attempt in 1979.
While the first phase of criticism focused on the theme of representation, the second phased focused on the question of justice. The more the beneficiaries of the revolution narrowed into a small and identifiable elite, the more they appeared as its usurpers. The basis of the internal critique was not that 1959 was historically unjustified, but that it had been betrayed and usurped by a narrow elite.
Revolutionary justice was social justice. Key to it was the righting of historical wrongs. As I have argued, the politics of redress involved two contradictory moments. The moment of appropriation targeted the beneficiaries of the old order. In doing so, it tended to unify its victims. In contrast, the moment of redistribution disaggregated this latter group—because it created out of this group the beneficiaries of the new order. Whereas the target of appropriation were the Tutsi in general and its intended beneficiaries all the Hutu, the actual beneficiaries were not all the Hutu. Since the process of redistribution made a distinction between different Hutu, it dissolved the revolutionary unity of the Hutu into postrevolutionary distinctions: such as, for example, a regional distinction between the Hutu of the north and those of the south, and a class distinction between propertied and propertyless Hutu.
The two moments produced two different sets of effects. In distinguishing the new from the old order, the moment of appropriation united the nation against its colonial usurpers. Indigenization was in this case the battle cry of the colonized; against a backdrop of colonial conquest, the demand for justice turned into a demand for indigenization. In this, the politics of the Hutu Republic was no different from that of the regime of Idi Amin in Uganda, or from the politics of indigenization in Nigeria, or from that of “authenticity” in neighboring Congo. In all instances, the demand for indigenization was closely tied to the politics of race as defined through the colonial period: race distinguished those indigenous from those not. The politics of indigenization was the politics of deracialization. From the point of view of an Idi Amin or a Grégoire Kayibanda or a Joseph Mobutu, indigenization was not racialism. Rather, it was deracialization.
The moment of redistribution highlighted a different kind of distinction. In distinguishing the beneficiaries of the new order from those who simply identified with it and formed its supporters, it defined the custodians of the new order. Thus, republican austerity gave way to an ethic of consumption, and the myth of the “egalitarian republic” got corroded. The Presidential Instruction of 1975 (no. 556101) gave civil servants permission to do private business without any restrictions, including owning rented houses, purchasing rented vehicles, and having interests in mixed economy or commercial enterprises.61 The instruction made official that which had been practice for some time. Corruption in and through state office became the subject of more and more writing in the 1980s. By the end of Kayibanda’s rule, the locus of corruption had been the cooperative Travail, Fidelité, Progrès (TRAFIPRO). As the national producer/consumer cooperative that had provided the most valued organizing site for revolutionaries of 1959, TRAFIPRO turned into a favored site of accumulation for businessmen and politicians in the First Republic. Its countrywide monopoly was said to strangle other forms of economic enterprise everywhere. Yet, its control was in select southern hands.62 In the Second Republic, massive corruption came to public light with the dismissal of the governor of the National Bank in April 1985. The new literature referred to the “Zairization of Rwanda,” with an isolated urban elite said to be out of touch with the mass of rural population.63
The two moments gave rise to two different kinds of literature. The literature that focused on the moment of appropriation tended to be more celebratory: it highlighted the moment of revolutionary transition, of virtuous sacrifice and of intense battle. It put the indigenous identity center stage and celebrated it as the new revolutionary identity. In contrast, the literature that focused on the moment of redistribution tended to highlight the moment of corruption, of déjà vu, of regression—even capitulation—to the ways of old. While the literature of the revolution had a touch of innocence, postrevolutionary literature tended to shade into caricature.
The revolution had defined its beneficiaries as Hutu, but critics implicitly distinguished between beneficiaries and usurpers, whom they defined in regional terms. By identifying the regional identity of key appointments in each period, observers tried to pinpoint both key regional rivalries and the shifting regional focus of the regime. It is in this sense that the center of the First Republic was said to gravitate around a factional competition between two sets of regional elites, from Gitarama prefecture in the center and from Ruhengeri in the northwest. With the coup of 1973, the center of elite competition and control was said to have gravitated to the northwest. By the late 1980s, the political elite of the Second Republic came mainly from two northern perfectures, those of Gisenyi and Ruhengeri64—even though they accounted for hardly a fifth of the country’s population and exactly a fifth of the ten prefectures in the administrative division of the country.65 A survey taken in the early 1990s showed that thirty-three of sixty-eight public institutions were under the directorship of individuals coming from either Gisenyi (nineteen posts) or Ruhengeri (fourteen posts). Befitting the home préfecture of the president, Gisenyi clearly played the lead role.66 Rwanda watchers noted that already, by the mid-1980s, nearly a third of the eighty-five most important posts in the republic—as well as almost all the leading positions in the army and the security services—were occupied by individuals from the prefecture of Gisenyi. In Gisenyi itself, Habyarimana’s home area, Bushiru, was said to compete favorably with Bugoyi.67
The critique that state offices tended to be monopolized by elite groups from a particular region resonated most with elite factions left out in the cold in the competition for office and position. That kind of critique needs to be distinguished from the kind of critique whose target was the single party order identified with the Second Republic, and which for that very reason had a wider audience. The source of this more general and political critique was the small—but influential and growing—number of literate urban people, ranging from professional and salaried middle-class persons to the more articulate workers and artisans in the capital city of Kigali. To be sure, Kigali was hardly more than a medium-sized town, whose population at the time of the 1978 census was around 120,000 in a country of 4.8 million. Yet, as soon as these voices began to express organized discontent with the kind of political order created by the single party, their significance far outstripped their numbers. This was for two reasons, one systemic, the other conjunctural.
The systemic reason had to do with the nature of a political system that held the vast majority of its population—its rural population—in the grip of myriad local authorities whose powers were literally unlimited and unaccountable to any but their superiors. We have seen that this tightfisted dictatorship combined administrative, executive, legislative, and judicial powers. It had been sanctioned as “customary” in the colonial period and then was reformed through elections after 1959. Anointed as “revolutionary” under the Second Republic, this local authority was without any electoral check-and-balance. Instead, it was reorganized and rationalized after the model of the ten-cell system in Tanzania. So long as this dictatorship held—which it did until the pressures of the civil war cracked it—it could only be critiqued from an external vantage point, that of the city. The conjunctural reason was more global. The late 1980s was both a time when Rwanda was becoming more vulnerable to the external environment and a time when winds of change were beginning to blow from without. Just when the collapse of the international coffee price, and even more so of its foreign exchange revenue, was making the Kigali regime more vulnerable to external pressures, these pressures began to take on a more political form. While Western creditors—self-described as the donor community—had been content to push the IMF-designed Structural Adjustment Programme as a standard reform package around the African continent through the decade of the 1980s, the collapse of the Soviet bloc saw demands for political reform take the front seat. It was a context in which any articulate internal critique of single party rule was bound to have a privileged hearing internationally.
This is precisely what happened in the second half of 1989, when critical periodicals and associated human rights NGOs began to surface in Kigali.68 Peter Uvin, a development administrator in Rwanda at that time, estimates that “by the beginning of the 1990s, Rwanda had one of Africa’s highest density of NGOs.” He calculates that “there was approximately one farmers’ organisation per 35 households, one cooperative per 350 households, and one development NGO per 3,500 households.”69 The internal voices of protest were reinforced by pressures for reform from the outside. These pressures came from two major sources: the assembly of Francophone states and the Vatican. As late as January 1989, Habyarimana considered any political change feasible only within the one-party system; but only a year and a half later, on 5 July 1990, he agreed to the necessity of a separation between the party and the state, possibly within the context of multipartyism. In September 1990, the Catholic archbishop of Rwanda resigned from the Central Committee of the ruling party, for the first time formally dissociating the Catholic Church from the MRND. The occasion was the pope’s visit to Rwanda; according to most observers, the resignation came as a result of direct pressure from the Vatican.
In September of 1990, Habyarimana established the Commission Nationale de Synthèse. It was given two years in which to make recommendations for a new democratic national charter. That, however, was before the Rwanda Patriotic Front (RPF) invasion of 1 October 1990, and the civil war that followed. We shall discuss the invasion and the civil war in the next two chapters, but let me make two observations here: (1) the onset of internal political reform in Rwanda preceded the RPF invasion; and (2) the immediate impact of the invasion was to accelerate the reform process. A month after the RPF invasion, Habyarimana spoke to the legislature, declared his support for the establishment of a multiparty system and instructed the National Commission to complete a draft national political charter by the end of the year. In a sharp departure from previous policy, he also acknowledged the right of refugees to return to Rwanda—without conditions, at least in theory. As directed, the commission published its draft in December. Also, as expected, the draft endorsed a multiparty political arrangement. This period also saw a fuller realization of press freedom: by March 1991, over a dozen newspapers and magazines, mostly critical of the official line, were in circulation. There followed public discussions of the charter’s proposals. In another six months, the draft constitution was approved and entered into force. That was in June 1991.
Opposition parties were permitted to register in March 1991. By July, four opposition parties had formed a coalition whose common purpose was to bring down the one-party government. The four-party coalition comprised the MDR, the PSD, the PL, and the PDC.70 Of these, the Mouvement Démocratique Républicain (MDR), was the first to organize as a party. It was a reformed version of the old MDR-PARMEHUTU. In dropping the appellation PARMEHUTU from its name, the MDR moved away from its old anti-Tutsi affiliation; at the same time, in reestablishing the MDR, the party stressed its affiliation to the central region and its opposition to the northwestern allegiance of the Habyarimana regime. The MDR was rivaled by the Parti Social-Démocrate (PSD). Ideologically positioned on the center-left, the PSD selfconsciously tried to attract an educated stratum that was both Hutu and Tutsi. From this standpoint, the Parti Libéral (PL) was even more interesting: it had no specific geographical base, but it attracted both Tutsi and those of “mixed” parentage who openly scoffed at “ethnic” politics.71 The last and possibly the least significant of the opposition parties was the Parti Démocrate-Chrétien (PDC), a Christian-Democratic group that had great difficulty establishing its oppositional credentials given the history of close relations between the Church and the single party.
The MRND, too, went through a metamorphosis in this year of change. To begin with, the party took advantage of its extraordinary congress on 28 April 1991 to adapt its statutes to the emerging multiparty context. And then on 5 July—exactly a year after the police had suppressed the strike that had signaled public opposition to single-party rule, and the very day Habyarimana agreed to the separation of the party and the state—the party modified its name. The new name signaled an expansion of its core objectives: it added “democracy” to “development.” The MRND was now the MRNDD: the Mouvement Révolutionnaire National pour le Développement et la Démocratie.
To look back at the year of the reform—from 5 July 1990 to 5 July 1991—and to identify the trends it brought to the surface—is to understand the kind of social forces that were being held in check by the single-party closure and were seeking room for expression. The opposition organized along two different lines: regional and ideological. The regional dimension highlighted the demand that power be anchored in a base broader than simply the northwest of the country. Ideologically, the opposition seemed to call for a more liberal arrangement, in both politics and economics. Neither the power nor the opposition was organized along “ethnic” lines. While one of the opposition parties, the Parti Libéral, had a mainly Tutsi leadership, it was not a Tutsi party. By the time of the 1990–91 reform, even in the first year that followed the RPF invasion, the regional question was far more politically volatile than was the “ethnic” question.
Unlike the First Republic—which was more inward-looking and saw itself as a regional exception, a product of Rwanda’s distinctive history of “race” relations—the Second Republic tended to situate itself in a regional context defined by two reference models, Tanzania to the left and Zaire to the right. It had borrowed from both and even invited comparisons with one or the other. In its potential capacity to engineer political reform and to charter a peaceful development from single to multiparty rule—while retaining the political initiative—the Second Republic seemed closer to Tanzania than to Zaire. On the eve of the RPF invasion of October 1990, the Rwandan polity was healthier than many others in the region. It had a better record of dealing with the political opposition than did most countries in the region of the Great Lakes, Tanzania being the notable exception. At the same time, there was a demonstrated willingness and capacity to undertake internal political reform. But the fate of the Second Republic turned out to be radically different from that of Tanzania. This was not because the internal critique turned out to be more radical than expected. The unexpected factor, rather, was the critique from without—the critique which stemmed from the RPF and which articulated the aspirations of the mainly Tutsi political diaspora. This, indeed, is where the difference with Tanzania was telling. While Tanzania was the one state in the region that did not drive entire groups into political exile, independent Rwanda was the one state whose very birth was linked to the phenomenon of group exile leading to a mushrooming political diaspora.
The Second Republic’s greatest single failure was that it was unable to even pose the question of how to integrate the Tutsi diaspora within the postcolonial polity. To integrate it would require less coming to terms with a political opposition than with the claim that the Tutsi were as much a part of the Rwandan political community as were the Hutu. Its failure was testimony to a past it could not come to terms with, because to do so required nothing less than to shed the presumption of its being a state of the Hutu nation. This single fact distinguished Rwanda from all other states in the region, even Burundi. All these states were organized more or less along the model of cultural pluralism, whose units were called ethnicities in Africa but nations elsewhere. This model presumed that, no matter how dispersed, every ethnicity would have an ethnic home somewhere in the region. That home was not only to be seen as the source of dispersal in the past but also a point of return any time in the future. At the minimum, the existence of the home was said to provide a fallback in times of crisis. Even if this return never happened, its very possibility—the very assurance that everyone had a home, no matter how remote—ensured that no one was orphaned in times of crisis.
It is in this sense that 1959 orphaned the Tutsi who left Rwanda. Even when Habyarimana rehabilitated the internal Tutsi from a nonindigenous “race” to an indigenous “ethnicity,” he had no intention of extending this “reconciliation” to the Tutsi diaspora. This created, for the first time, a group that was from the region but not of the region, which was part of the region but without belonging to any particular part of it. The distinctive feature of the Rwandan Tutsi diaspora was that its members were ethnic strangers everywhere; they had no ethnic home. As a group, they had nothing to lose, not even a home. The 1959 Revolution had made of the Rwandan Tutsi diaspora a group akin to the Jews of prewar Europe.
The Rwandan Tutsi diaspora of the sixties may have pined to go home, but their children who were born in exile and grew up in the seventies, away from the ancestral home, were determined to make a home where they were. As events would show, this was the truest of those in Uganda. Faced with a state that was hostile to “strangers,” at first racial, and then ethnic, it is in Uganda that the Banyarwanda—for that is how the second-generation exiles and refugees thought of themselves—made the most concerted effort to make a home for themselves by staking nothing less than their lives in the process. That effort was the guerrilla war in the Luwero Triangle of Uganda from 1981 to 1986. The irony was that it was the guerrilla struggle that showed convincingly the limits to which ethnic strangers could make themselves at home in a state that defined “home” as an ancestral—indigenous—abode for “natives,” keeping at bay all those considered nonindigenous, no matter what their commitment or predicament, as “settlers.” Forced to confront the political fact that in postcolonial Africa one’s political home was equated with an ancestral home, and an ancestral home with the precolonial home, the Tutsi diaspora turned to their ancestral home, Rwanda.
Their attempted return was the source of the external critique. This is why the external critique was not really a response to the policies of the Habyarimana regime; its roots were much deeper. It expressed the agony and the predicament, the impossible situation, of those without a political home in postcolonial Africa. This is why we need to look at developments beyond Rwanda, particularly those in Uganda, if we are to understand the nature, the timing, and the strength of that critique. For the RPF invasion was less a response to post-1959 developments in Rwanda than it was an outcome of developments within the region, one that the growing post-1959 Banyarwanda diaspora had crisscrossed and knit together. Only that explained the tenacity with which the RPF invasion persisted, more or less without regard to the tenor or pace of developments within Rwanda, or even whether these developments were moving in the direction of reform or of degeneration.