In 1994 an incredibly fast genocide took place in the small, impoverished, and hitherto internationally obscure African country of Rwanda. Since many people around the world first heard of that country during the shocking events of that year or through subsequent books and movies about them, the name “Rwanda” has almost become synonymous with the horror of total genocide. Within 100 days, roughly 800,000 people were slaughtered, mostly by state-sponsored mobs wielding simple weapons like machetes and clubs. Although there were no mechanized means of mass murder such as gas chambers or strategic bombers, modern communication methods—particularly the radio—played an important role in orchestrating the deadly campaign. That episode was one of the most appalling and infamous incidents of mass violence in the late 20th century. It was also strongly consistent with the 1948 international legal definition of “genocide,” as there was an intentional and planned program of murder meant to exterminate a distinct section of the Rwandan population completely, based solely on their Tutsi identity. Of course, many other people were killed. Why did that terrible event happen? In answering that question, the international media often fell back on racist stereotypes of Africa and offered easy but inadequate explanations of innate ethnic hatred and “tribal warfare.” While some academics have suggested that Rwandans had been psychologically conditioned to become particularly obedient people who responded to terrible orders, others have pointed to intense overpopulation and competition over land as a contributing factor.1 An important point to remember is that the 1994 event was not the first genocide to have taken place in the history of Rwanda. There was an earlier and much less well-known massacre of Tutsi in 1963–1964, seen as a genocide by international observers at the time, which strongly foreshadowed the events of 30 years later. This chapter will discuss the historical context of the two genocides against the Tutsi that took place in postcolonial Rwanda. It will look at a combination of elements that made Rwanda’s history distinct from other African countries and perhaps contributed to those terrible events. Those factors include a long history of violence, a culture of fear and state control that dates from the precolonial era, and the rise of a colonial racial hierarchy that combined with such other issues as the country’s small geographic size and population density.
By the time of colonial conquest in the late 19th century, the Great Lakes region of east-central Africa was dominated by a series of centralized states, including Buganda, Bunyoro, Nkore, Burundi, and Rwanda. Given the limited nature of African oral tradition and the lack of documentary sources, little is known of the detailed history of those hinterland states before roughly 1800. Archaeological evidence broadly illustrates that people in the region had been practicing settled agriculture and metallurgy for several millennia. Furthermore, archaeological evidence also suggests that roughly a thousand years ago cattle keeping became increasingly popular and that a relatively sudden change in pottery style across the region seems to indicate the arrival of immigrants from another area. However, the lack of impact on local Bantu languages suggests that this group of important newcomers may have been somewhat small and eventually absorbed into local communities. During the 1400s and 1500s CE, within the context of Later Iron Age developments associated with state formation, the people of the Great Lakes formed the area’s first-known organized kingdoms, with some focusing on cultivation and others on pastoralism, depending on local environmental conditions. Much earlier, between around 200 BCE and 1000 CE, Nilo-Saharan-speaking (commonly called Nilotic) pastoral people from northwest of Lake Turkana, in what is now South Sudan and southern Ethiopia, had gradually moved southward into the Great Lakes region. Their arrival may have been related to the changes discussed above. That process accelerated during the 1400s and 1500s CE, as other groups of Nilo-Saharan-speaking pastoralists of similar origin, probably reacting to drought, also moved south. Some of those pastoralists moved east near Lake Victoria to form the basis of well-known East African pastoralist communities such as the Karamojong of Uganda and the Maasai of Kenya and Tanzania, who pushed Bantu-speaking cultivators off the grasslands and into the highlands of the Great Rift Valley. Other Nilo-Saharan pastoralists, Luo-speakers from what is now South Sudan’s Sudd floodplains, also moved south around the same time and broke into small groups, with some settling northeast to establish Luo communities in what is now western Kenya and others moving west around Lake Victoria. It is well known that the last element of Luo migrants was instrumental in the creation of such centralized states as Bunyoro and Buganda on the northwest side of Lake Victoria in the 1500s and 1600s. In this area, the Luo merged with the preexisting population of cultivators. Historians have long recognized that similar Luo pastoralists may have also formed the nucleus of the distinct pastoralist minorities who came to dominate cultivating communities in the highlands between Lake Victoria and Lakes Edward and Kivu and eventually formed the states of Nkore, Burundi, and Rwanda. Since the historic origins of the Tutsi have become a controversial topic (particularly within Rwanda), it is important to remember that the southward movement of Nilo-Saharan pastoralists is a well-established feature of East African history and certainly affected the Great Lakes region.
The precolonial Kingdom of Rwanda (often called the Nyiginya Kingdom after the name of the royal clan) represented a compact but militarily powerful state located in the southern portion of the modern country of Rwanda. By 1700, the kingdom had developed an effective, centralized royal court with control over land and labor and a permanent army, which was a rare and potent institution in precolonial East and Central Africa. Although there is no doubt that Rwanda’s Tutsi, Hutu, and Twa identities existed when the first German colonists arrived in the mid-1890s, the origins and nature of those categories has been contested. The lack of certainty mostly relates to the nature of historical evidence, as there was no literature concerning Rwanda before the advent of German colonial rule. As in most of the rest of precolonial Sub-Saharan Africa, Rwanda was a non-literature society, in which information was passed orally from person to person and generation to generation. Rwandan oral traditions recorded in the 1950s, more than half a century after colonization, relate strongly to the Tutsi-dominated traditional kingdom that still existed at that point and, as all oral sources, are malleable and symbolic. As such, there is almost no direct evidence regarding the history of the Tutsi, Hutu, and Twa identities before 1895. In discussing postcolonial genocide, one may ask why the precolonial origin of those identities matter as they clearly existed in people’s minds and were the subject of Rwandan state policy in the early 1990s. As will be discussed below, however, the Tutsi, Hutu, and Twa identities have a long and dynamic history in Rwanda (and Burundi) and are thus deeply rooted in that society. Understanding that history, as Mahmood Mamdani explains, serves to contextualize popular participation in the genocides against the Tutsi that took place in 1963–1964 and 1994 and helps clarify “how the unthinkable becomes thinkable.”2
A basic narrative of Rwanda’s early precolonial history was initially put forth in literature written during the colonial era. It explains that Twa hunter-gatherers were the original inhabitants of the area, which was subsequently settled by Bantu-speaking and iron-using Hutu farmers at some point in the distant past. Then, perhaps no more than 500 years ago, Tutsi pastoralists arrived and used their military prowess and systems of patronage created by loaning out cattle to the existing inhabitants to gradually establish the centralized Nyiginya state. In the colonial view, different regional origins meant that members of each group were believed to possess stereotypical physical characteristics: the Twa who were short, the Hutu who were of medium height and stocky build with round facial features, and the Tutsi who were tall and thin with sharp facial features. Although colonial scholars believed that those identities and their associated traits had always existed, more recent historians point out that the names Tutsi, Hutu, and Twa probably crystallized over time, especially as new groups moved into the area and began to make distinctions between themselves and others. For example, the ancestors of late-19th-century Hutu probably did not always call themselves by that name and only came to do so as they fell under the rule of the Rwandan kingdom. Jan Vansina, an eminent historian of Central Africa who conducted extensive original research on precolonial Rwandan history in the 1950s and 1960s, believes that this process happened over a long time. Choosing his words carefully, Vansina explains that “there is no doubt that present-day Rwandans really encompass three different biological ‘populations’ and that whichever scenario is adopted to account for this fact, the differences among the groups run so deep that they must extend back millennia rather than centuries.”3 Vansina proposes that the polarization of Tutsi and Hutu identities may have begun within the 18th-century Rwandan military as warriors were recruited from elite cattle-herding families called Tutsi and noncombatant support personnel originated from less-prestigious agricultural families and were referred to as Hutu, which meant that they were servants. Eventually, those terms and their hierarchical pastoral and agricultural connotations spread within the broader society. Precolonial marriage practices and artificial selection may also have contributed to the rise of those identities and their concomitant physical traits. Twa hunter-gatherers were long shunned by other groups, married and reproduced among themselves, and took to making pottery as a niche economic activity. As the Nyiginya state expanded during the 1700s, the pastoralist Tutsi imitated their royal family and avoided marrying into less-respected agricultural lineages.4 In the 19th century, that practice became much stricter, resulting in the emergence of a recognizable hierarchy of three social identities—Tutsi, Hutu, and Twa—which was clearly observed and recorded by the first Europeans who visited Rwanda at the turn of the 19th and 20th centuries. Similar processes happened in neighboring kingdoms such as Burundi to the south with the Tutsi/Hutu/Twa division and Nkore in the northeast with Hima pastoral warriors who dominated Iru cultivators.
Since the terrible events of 1994, it has become popular to reject the idea that Rwanda’s historic Tutsi, Hutu, and Twa identities originated with the movement of different groups into the area at different times, which is criticized as a divisive colonial theory. As such, Rwanda is often presented as the ancestral home of all its people. It follows then, that the three identities must have emerged from a single community because of local factors, or perhaps later colonialists either invented them entirely or changed them in some way to facilitate divide-and-rule. In the early 1970s, Walter Rodney, a polemical historian of the transatlantic slave trade, speculated that different diets may have caused the development of dissimilar physical traits among the Tutsi, Hutu, and Twa. In other words, the pastoral Tutsi grew taller because of their protein-rich diet of milk and meat, whereas the Hutu and Twa did not enjoy such luxuries. Rodney presented no evidence to support this view.5 In the 2004 Hollywood film Hotel Rwanda, which is about the 1994 genocide, several Rwandan characters tell visiting foreign journalists that the Tutsi and Hutu identities were simply made up by the Belgian colonial rulers. In a book on the military in Rwanda, former Rwandan Patriotic Front (RPF) officer Frank Rusagara presents the Tutsi, Hutu, and Twa identities as irrelevant in the country’s precolonial history.6 In contemporary postgenocide Rwanda, the common-origin theory has become the official history of the country and is enforced by antigenocide ideology laws that limit freedom of expression, especially around issues of identity and history. The Tutsi, Hutu, and Twa identities do not officially exist, yet they are forever on people’s minds. The common-origin version of history is mobilized to promote national unity and reconciliation in a postconflict society. It is also a useful mechanism of control, given that the ban on discussing identities serves to obscure the fact that many top government officials are from the Tutsi minority and thus might not be seen as representing the interests of the majority.
Several realities of contemporary Rwandan culture seem to support the common-origin theory. There does not appear to be obvious linguistic evidence that different groups moved into the country from other areas, because today every Rwandan speaks the same language, Kinyarwanda. In a lengthy theatrical performance entitled Rwanda 1994, an authoritative Belgian commentator presents a long lecture on the history of Rwanda, in which he states categorically that the ancestors of the Tutsi could not have come from outside the country, as no foreign conqueror, in all human history, has ever adopted the language of the conquered. Of course, there are many examples from world history (particularly China) that contradict that sweeping statement. One only has to look as far as Malawi to see that the Ngoni people, who conquered part of that area in the middle 1800s, have now almost lost their original Nguni language in favor of the Chinyanja language of their former subjects. If the movement of new groups into what is now Rwanda took place millennia ago, then other languages that were present there may well have been replaced by a broad common language. Today’s Kinyarwanda does have such variations as the slight difference between the language spoken in the north and south of Rwanda, and the distinctive accent of the Twa. It is also important to note that Kinyarwanda, historically, is not a language that is only associated with the precolonial Kingdom of Rwanda as mutually intelligible versions such as Kirundi (the language of Burundi) are spoken by people in neighboring areas. Another fact presented in support of the common-origin theory is that every Rwandan is a hereditary member of one of 18 different clans, each of which includes people of Tutsi, Hutu, and Twa identities. However, the 18 clans may have originated as Tutsi institutions into which their subjects and servants were incorporated.7
In contextualizing Rwanda’s two postcolonial genocides, it is often forgotten that precolonial Rwanda had a particularly violent history in comparison to other African states and societies. The legendary founder of Rwanda, Ruganzu Ndori, is said to have spent most his life fighting in different military campaigns, and, after he died from a battle wound to his eye, his spirit was believed to inhabit the royal Kalinga drum that became the state’s symbol. Ndori created Rwanda’s military system, which allowed for something like a standing army and represented a major advantage over the kingdom’s neighbors. In another distinct military innovation, many armies had matching and named “cattle armies,” which functioned as a commissariat and consisted of livestock from warriors’ families and war booty. According to Vansina, “The deepest effect of this new military organization was the institutionalization of militarism and martial violence that finally permeated the whole of Nyiginya culture as the armies became the foundation of the administrative structure of the realm.”8 Successful warriors, who cut off the testicles of their victims to show that they had killed men, were revered and displayed symbols of their violent accomplishments. Killing 7 men in combat earned a warrior an iron necklace with bells; after 14 total kills, he was given an iron and brass bracelet; and 21 resulted in a mountaintop ceremony during which he became a national hero.9 During the 1700s and 1800s, Rwanda was embroiled in almost-constant warfare, including attacks on neighbors meant to reduce them to tributary status and expansionist campaigns west toward Lake Kivu and north toward the volcanoes. There were also terrible internal conflicts between rival claimants to kingship, as there was never a peaceful transition of power. Relations with Burundi to the south were intensely violent over a long period. By the 1790s, the Nyiginya Kingdom had organized 30 armies and could field about 12,000 combatants, which made it a principal regional military power. Between 1800 and 1870, Rwanda doubled its size. This history of warfare inspired local sayings such as “Rwanda attacks, it is not attacked,” and “Tutsi share blood, not milk.” Although Rwanda began to engage in increasing regional trade in the 1850s, the state remained somewhat isolationist, as Swahili/Arab caravans protected by gunmen and carrying goods from the East African coast were banned from its borders. As such, the kingdom’s arrogant pastoralist warriors did not show much interest in acquiring newly arriving firearms.
Perhaps the most aggressive precolonial ruler in the history of Rwanda and potentially the entire Great Lakes region was Kigeri IV Rwabugiri, who led the kingdom from his coming of age in 1875 to his death in 1895. He launched an average of two major military campaigns per year, which, for a precolonial African state, was extremely ambitious. Given the pressure that this put on the kingdom’s primarily Hutu farmers to feed the ever-expanding and active armies, Rwabugiri had to crush three major provincial rebellions and many smaller insurrections. Furthermore, he formed eight new armies and altered the kingdom’s military system so that he gained more authority and a greater share of loot taken during those wars. It is likely that Rwabugiri’s increased centralization of the Rwandan state and the concentration of cattle into the hands of a small number of wealthy pastoralist families furthered the division between Tutsi pastoralists and Hutu farmers.10 Recognizing this history should not be seen as a vilification of Rwanda or Rwandans (many countries have a violent history) but as a reminder that extreme violence was not new and did not end with the beginning of colonial rule. It is often said that Rwanda had a peaceful history up until 1994, but that is simply not true.
During the early 1890s, as the European Scramble for Africa was in full swing, the Kingdom of Rwanda was weakened by drought and locusts that ruined crops, rinderpest and foot-and-mouth disease that killed cattle, and small pox and jiggers that affected people. At that time, Rwanda was positioned between three expanding colonial powers. To the east and southeast, the Germans had taken control of a portion of the East African coast and were moving inland through what is now Tanzania; to the west, the Belgians of Leopold II’s Congo Free State were establishing their authority over the vast Congo River Basin; and to the north the British had intervened in the Buganda civil war and were creating the colony of Uganda. In May and June 1894, a German colonial expedition under Gustav Adolf von Gotzen entered Rwanda and visited Rwabugiri, who was embroiled in a war with neighboring Nkore in what is now Uganda and attempting to quell resistance in the subordinate area of Bushi in what is now the Democratic Republic of Congo (DRC). On leaving Rwanda, the German expedition used its firepower to repel a surprise attack by some of the kingdom’s warriors, though it is unclear if Rwabugiri had ordered the action. With the sudden death of Rwabugiri in September 1895, the territories he had conquered, including Ijwi Island in Lake Kivu and Bushi, rebelled and threw off Rwandan dominance. The next year, after Rwabugiri’s son, co-ruler, and heir, Mibambwe IV Rutarindwa, ordered what turned into a disastrous attack on a Belgian colonial post at Shangi on the south end of Lake Kivu, violence broke out at the royal residence of Rucunshu, in which hundreds were killed and the ruler committed suicide.
In February 1897, Kanjogera, a widow of the late Rwabugiri and a rival of Rutarindwa, emerged victorious from the Rucunshu massacre and arranged for her teenage son Yuhi V Musinga to become king, followed by a purge of potentially disloyal military commanders. Since this coup represented the rise of Kanjogera’s Bega clan over that of the royal Nyiginya clan, the new regime would struggle with legitimacy. The next month, a German expedition with 300 soldiers under Captain Hans von Ramsay arrived from Lake Tanganyika, and Kanjogera and Musinga accepted a German protectorate over Rwanda in exchange for military support against internal rivals and other external forces like the Belgians. Almost immediately after the Germans left, Kanjogera and her supporters had to contend with a series of rebellions led by rival claimants to the kingship and by Hutu in the north who had been recently conquered by the late Rwabugiri and wanted to reclaim their independence. In suppressing those uprisings, Kanjogera’s warriors devastated many communities, including some that had not rebelled, and many people fled into the hills. In 1900 the mostly French-speaking Catholic missionaries of the White Fathers arrived in Rwanda. The royal court secured the loyalty of the Tutsi by initially forbidding them from attending missionary instruction, but many Hutu flocked to the newcomers, whom they saw as offering protection against the exploitive kingdom. Some began to refer to missionaries as the “kings of the Hutu.” At that point, the Germans exerted little authority in Rwanda, as their only military post was in the far south of the country, and their regional headquarters was in Bujumbura in Burundi. In 1901, at the request of Musinga’s court, a German military detachment moved up from Bujumbura to suppress a rebellion in the eastern region of Gisaka. Many dissident leaders fled to exile in Burundi. In April 1904, the Bega faction that controlled the court consolidated its power by slaughtering a large number of rivals from the Nyiginya clan. Around the same time, the White Fathers, leading armed groups, pushed into the mountainous northern frontier, where they raided local Kiga communities to compel them into providing workers for building missions. At that point, the term “Kiga” (plural Bakiga) referred to the people of the mountains who were not Tutsi or Hutu, as they were not yet under the control of the Rwandan state. In 1904 a German military expedition led by Werner von Gravert toured Rwanda to impress the royal court and ventured north to inflict terrible vengeance on the Kiga for attacking a mission.
After Musinga came of age in 1905, he relied increasingly on German colonial troops to suppress rebellions and to extend his rule northward into agriculturally rich and previously autonomous areas in which Kiga farmers were subjugated by the Rwandan kingdom and therefore became known as Hutu. The Germans were eager to cooperate with Musinga because they were being challenged elsewhere in German East Africa by the Maji Maji Rebellion, and they had few personnel on the ground in Rwanda. The Germans also began to force Rwandans to cut and transport timber for the construction of a new colonial capital called Kigali in the center of the territory. Furthermore, Musinga gradually emerged from the dominance of his mother’s Bega clan and began to reestablish the primacy of his paternal Nyiginya clan.
Violence plagued Rwanda’s northern frontier as several interrelated rebellions challenged the advancing dominance of Musinga and the Germans. In the early 1900s, the prophetic Nyabingi movement led by Muhumuza, a woman who claimed to be a widow of the late Rwabugiri, fought a guerrilla war against Musinga’s rule in Ndorwa, as well as against German, Belgian, and British colonial rule in what became a colonial tri-border zone. In 1908 the Germans imprisoned Muhumuza for two years, and in 1911 a British expedition captured her and exiled her to Uganda. Between 1905 and 1912, a force of Twa led by Basebya used the forest and swamps of northern Rwanda as a staging area for guerrilla war against the Rwandan and German forces sent into the region. In 1910 Rukara, a Hutu and leader of the Barashi, the strongest Kiga group in the north, killed a European missionary and rebelled against Musinga. At the start of 1912, Ndungutse, who had previously been involved with Muhumuza’s activities and similarly claimed royal links to the late Rwabugiri, led a rebellion in north-central Rwanda that encouraged Kiga to refuse demands for labor and tax from the Rwandan state and to expel its agents. Ndungutse became popular with the newly defined Hutu of northern Rwanda and posed a significant threat to Musinga’s rule. In April 1912, a detachment of German soldiers and 3,000 of Musinga’s warriors moved north, where they stormed the rebel leader’s stronghold and killed him. The soldiers then conducted a reign of terror in the north by destroying homes and crops, seizing livestock, taking hostages, and conducting massacres. A missionary wrote, “The war continues; the batutsi massacre, are without mercy, half of the population of Bumongo [a neighboring region] will be destroyed. Groups of women are led away and will become the booty of the Great Chiefs.” It was at that time that the Twa leader Basebya was captured and executed by a German patrol. Rukara was also captured by the Germans, and, although he was shot to death while trying to escape on his way to the gallows, his corpse was hanged to dissuade further Kiga resistance. In the aftermath of this reign of terror, Musinga sent more officials, mostly Tutsi, to administer the north, including some areas never before under his kingdom’s authority. A victory celebration was held at Musinga’s capital of Nyanza, where the king, given his dependence on colonial support, was compelled to parade his royal cattle before a German official as a sign of his submission.11
During the First World War, Musinga cooperated with the Germans as Kiga resistance in the newly conquered north resurfaced. In 1916, as part of a general invasion of German East Africa, Belgian colonial troops from neighboring Congo occupied Rwanda and Burundi. Initially, the Belgian occupation forces seemed to undermine the central court by interacting directly with subordinate Tutsi officials and Hutu commoners to acquire resources for the war effort. However, within a year, the danger of anarchy prompted the Belgians to adopt the German system of governing indirectly through Musinga’s court. In the north, Belgians and Tutsi court officials gradually suppressed the decentralized resistance of the Kiga, which furthered their transformation into Hutu subjects of the Rwandan state.
With the end of the global conflict in 1918, the League of Nations divvied up defeated Germany’s former colonial territories to the victorious powers as mandates that were theoretically meant to one day become independent countries; in practice, those mandates were treated like other African colonies. Within this context, Belgium formally took over the administration of Rwanda and Burundi in the early 1920s. Immediately after the war, the Belgians adopted the contradictory policies of ruling Rwanda indirectly through the traditional kingdom—as the British did in parts of neighboring Uganda—and simultaneously “civilizing” its society through the promotion of Christianity and individual rights. Since the royal court was hostile to Christianity, the Belgian administration introduced secular education for Tutsi young men at the king’s capital of Nyanza in the south. Hutu who sought education attended mission schools. In the mid-1920s, the Belgian administration began to strip the royal court, which was divided between traditionalist and pro-European factions, of its power to appoint officials, collect tax, and settle judicial cases. As the Belgians became more influential, members of the Tutsi court became more interested in Western education, and many converted to Christianity, which, given the existing network of patron-client relations, led to the popularization of schools and churches throughout the country. At the same time, the colonial capitalist economy developed through the mining of tin and growing of coffee for export.
To govern and exploit Rwanda more efficiently, the Belgians simplified the administrative system, creating a hierarchy of Tutsi chiefs and subchiefs who reported to them. In 1926 the Belgians eliminated Rwanda’s existing administrative structure, in which each hill community had three chiefs—one for each the land, people, and cattle—and in which Hutu and Tutsi could occupy those positions. A new and much more authoritarian system was created, whereby each hill was under one chief, who was always Tutsi. The Hutu were completely deprived of avenues for advancement and redress of grievance. Without the legitimacy of the historic monarchy, the Belgians emphasized racial myths, whereby the supposedly intelligent Tutsi were seen as natural rulers and the allegedly less-intelligent and submissive Hutu were seen as natural workers. The new system provoked resistance. Anxiety was also worsened by a famine in the late 1920s that was caused by drought and colonial labor demands that hindered agriculture and led to the deaths of at least 40,000 people. In 1928 the Kiga in northern Rwanda rallied around a leader called Semaraso and attacked Tutsi officials, killing about a dozen and seizing cattle. After colonial troops and Tutsi warriors easily suppressed the rebellion by killing and jailing several dozen people and destroying the shelters and crops of around 1,000 more, Semaraso fled into Uganda. In other parts of the country, anti-European millenarian movements such as Nyabingi gained popularity among the Hutu and were repressed by Belgian authorities.
In November 1931, Belgian officials, with approval from their superiors, deposed the traditionalist Musinga, who had subtly resisted Westernization, Christianization, and growing colonial authority.12 The Catholic Church, led in Rwanda by Bishop Leon-Paul Classe, a longtime advocate of Tutsi racial supremacy, supported the removal of Musinga, because he had been making contacts with rival Protestants and Seventh-day Adventists. Musinga was immediately replaced by his son, Mutara III Rudahigwa, who had always been interested in European ways and who, upon his baptism in front of his subordinate chiefs, became the first Christian king of Rwanda. This transition prompted a final wave of religious conversion throughout the country, which further enhanced the power of the Catholic Church. Seen by many as the “King of the Whites,” Rudahigwa became a mostly symbolic ruler and a model of Christian values. Tutsi officials and overseers enforced intensified and individualized taxation, obligatory labor on the building of infrastructure such as roads and terracing, and the compulsory growing of certain crops under antifamine and development policies through corporal punishment against Hutu peasants. As such, colonial Rwanda was turned into “a vast forced labour camp” geared toward the production of agricultural exports that profited Belgian businesses.13
A colonial racial hierarchy was imposed. In 1933 and 1934, the Belgians conducted an official census in Rwanda that categorized all inhabitants as Tutsi, Hutu, or Twa and required people to carry identity cards on which their status was inscribed. Any flexibility in the application of those identities was lost. With privileged access to Western education and exemption from forced labor, the Tutsi minority were transformed from the historic pastoralist-warrior elite into the new colonial intellectual-administrative elite tied to the Catholic Church. The Hutu majority, subject to coercion and taxation, became the peasant producers of cash crops. As in precolonial times, the Twa minority continued to be reviled, although now this low status was reinforced by European racial ideology. The categorization of Rwandans in the census was based on personal information provided by the Catholic Church—physical appearance and ownership of cattle. Contrary to popular belief, the census did not classify any Rwandan with ten cows or more as a Tutsi.
Colonial racial theories such as Social Darwinism—the idea that human societies had experienced a process of natural selection and evolution—were applied to the history of Rwanda to explain the development of this hierarchy of identities. Although the European application of those ideas to Rwanda dated back to the late 19th century, they now became incredibly influential within Rwandan colonial society. The Twa were seen as the almost subhuman original inhabitants of the country, who had been conquered by the agricultural Hutu, who were understood as biologically and technologically similar to other Bantu-speaking Africans of sub-equatorial Africa. In turn, the Twa and Hutu were subjugated by the supposedly racially superior Tutsi pastoralists, who were believed to have originated from somewhere outside of Africa. In the colonial mind, the Tutsi were seen as evidence of the Hamitic theory, which postulated that indigenous Africans were incapable of organizing complex states and thus the existence of precolonial kingdoms could only be explained by the arrival of mysterious non-African outsiders termed “Hamites” (named after the son of Noah from the Bible) at some point in the remote past. The defining feature of colonial rule in Rwanda became race. The Hutu were imagined as the inferior, indigenous Bantu race, and the Tutsi as the alien and superior Hamitic race. Unlike many other colonies in Africa in which indirect rule was imposed, the inhabitants of Rwanda were not broken down into different ethnic groups with separate traditional political structures and customary laws. The Europeans taught colonial ideas about race and history in Rwanda, which were accepted by the emerging Tutsi intelligentsia, who attended a special Tutsi school in Astrida (Butare), as proof of their superiority. With separate educational streams, Tutsi students were taught in French and prepared for administrative careers, whereas the few Hutu students were given an inferior education in the broad East African language of Kiswahili.14
In the 1950s, the ideology of “Hutu Power” emerged to challenge the Tutsi-minority domination of Rwanda. The context was provided by the Second World War, which saw the global circulation of powerful ideas about national self-determination and democracy. The authors of Hutu Power were from Rwanda’s small but nascent Hutu counter-elite, who had managed to gain Western education and access to publication venues through the Catholic Church, in which there was new interest in social justice issues. Some of the first Rwandans to express interest in Christianity were Hutu, and in 1919 the first group of Hutu priests were ordained. The new Hutu Power movement of the 1950s reinterpreted colonial ideas about race and history. From their perspective, the Tutsi were presented as foreign usurpers and decadent exploiters of the hardworking and pious Hutu majority. A leading member of the new Hutu elite was teacher and former Catholic seminarian Gregoire Kayibanda, who attended a Catholic youth conference in Belgium in 1950. In 1953 Kayibanda became the first lay editor of a Catholic monthly magazine called L’Ami (the friend), and in 1955 he was promoted to the editorship of the Catholic Kinyarwanda weekly newspaper Kinyamateka (about history). In March 1957, Kayibanda and Swiss-born Catholic Archbishop Andre Perraudin published the ten-page “Bahutu Manifesto” that explained the problems of Rwanda in terms of the racial division between Hutu and Tutsi and called for the emancipation of the Hutu and the implementation of racial quotas in education and employment that would clearly favor the majority. Several months later, in June, Kayibanda was instrumental in forming the Muhutu Social Movement (MSM), which advocated for the general improvement of conditions for the Hutu majority. In November 1957, while Kayibanda was again visiting Belgium to build contacts with Catholic politicians and labor leaders, the extremist anti-Tutsi activist Joseph Gitera formed the Association for the Social Promotion of the Hutu Masses (APROSOMA) in the south of Rwanda. Around the same time, conservative Tutsi launched their own political grouping that, in 1959, transformed into the Rwandan National Union (UNAR). Within the broad context of decolonization discussed below, the emerging African political parties of Rwanda were clearly divided along Hutu and Tutsi lines.15
Across Africa, during the late 1950s, the moderate Westernized associations of the interwar years morphed into modern mass nationalist movements that demanded immediate independence from the colonial masters. The new and growing cities of Africa, which had swelled during the Second World War, provided a consolidated venue for nationalist leaders to spread their message and rally support, often among the unemployed and disappointed. The subsequent and rapid decolonization of Africa in the late 1950s and 1960s happened for many reasons. Those reasons included the spread of international anticolonial sentiment, the failure of British and French military intervention in Egypt in 1956, anticolonial insurgencies in Kenya and Algeria, the domination of global politics by two officially anticolonial superpowers, and the decolonization of Asia in the late 1940s, which created a growing anticolonial block within the new United Nations. As such, Britain granted independence to Ghana in 1957, and at the same time, negotiations were underway for the independence of Nigeria, which took place in 1960. Belgium came under increasing pressure, both internationally and within the Congo, to move toward similar decolonization. Reluctant to abandon its interests in Africa, the Belgian government attempted to grant independence quickly to unprepared and friendly African regimes that Brussels could control. As such, the vast Belgian Congo, with just six months of preparation, became independent in 1960. The country rapidly broke down through a mutiny by African soldiers against Belgian officers who were still in charge and through declarations of independence by separatists in the regions of Kasai and Katanga, which resulted in civil war.
In late-1950s Rwanda, Belgian officials began to see that the emerging Hutu political leadership were potentially more cooperative postcolonial partners than the Tutsi intellectual and administrative elites. As in other parts of Africa, many members of the established, Westernized, educated elite of Rwanda—who were mostly Tutsi—had embraced anticolonialism, African nationalism, and aspects of socialism. The predominantly Tutsi UNAR seemed to represent those ideals that were seen as threats to continuing Belgian interests. On the other hand, the small Hutu leadership was deeply Catholic, strongly anticommunist, and could convincingly claim to represent the majority in a future democratic system. This fear of socialism and communism must be understood within the international Cold War context in which Belgian officials, businesspersons, and church leaders worried about losing influence to future postcolonial African regimes friendly to the Soviet Union or Communist China. In early 1959, the Belgian government, copying what it was trying to do in Congo, announced that elections for a territorial government would take place in Rwanda toward the end of the year. Such embryonic Hutu and Tutsi political associations as APROSOMA began to declare themselves as formal political parties that would run for electoral office. When King Mutara Rudahigwa was suddenly admitted to hospital in Burundi and died of an apparent cerebral haemorrhage, many Tutsi believed that he had been murdered by the Belgians. Subsequently, they launched UNAR and demanded immediate independence under the existing Tutsi-dominated monarchical system. Around the same time, some moderate Tutsi formed the Rwandan Democratic Rally (RADER), which advocated cooperation with the Belgians and the Hutu, but it quickly became marginalized in the extremist environment. In October 1959, Kayibanda transformed MSM into the Party for the Emancipation of the Hutu (PARMEHUTU). Its manifesto called for the end of Tutsi colonialism and feudalism, rejected hostility toward Europeans, thanked the missionaries for their educational endeavors, advocated gradual democratization rather than immediate independence, and supported private property rights and a constitutional monarchy.16
Violence was triggered when Belgian authorities attempted to depose three Tutsi chiefs who were also UNAR supporters. At the beginning of November 1959, a Hutu subchief who was active within PARMEHUTU was attacked by UNAR enthusiasts because he refused to condemn the dismissal of the three Tutsi chiefs. Within a week, violence spread across most of the country, with Hutu mobs burning Tutsi homes and the newly installed King Kigeri V Ndahindurwa forming a monarchist/UNAR militia that counterattacked and killed Hutu leaders. In mid-November, Belgian Governor Jean-Paul Harroy declared a state of emergency and placed Colonel Guy Logiest—commander of a Force Publique colonial military detachment in neighboring Congo—in charge of the territory. Troops from eastern Congo had already begun to occupy Rwanda and used their firepower to restore order in a way that favored the Hutu gangs. Preferring the idea of a future independent Hutu republic to a Tutsi monarchy, Logiest replaced more than half of the Tutsi chiefs with Hutu ones, many of whom were PARMEHUTU members. Although UNAR activists were prosecuted for crimes committed during the violence, supporters of APROSOMA and PARMEHUTU were not. Given the violence, Belgian officials rescheduled local government elections for June 1960, despite objections from a UN delegation, which witnessed Tutsi homes being burned and which called for negotiation between all major parties before a vote could proceed. PARMEHUTU won a stunning electoral victory, taking 160 out of 229 legislative seats, with the Tutsi-oriented parties, including UNAR, winning just 19. In turn, local officials of the new Hutu government took power from the chiefs, and periodic violence against Tutsi continued into the next year. The deposed King Kigeri fled the country in July. Since Congo became independent in June 1960, the Force Publique withdrew from Rwanda and Burundi, and the Belgians replaced it with the new, locally recruited Territorial Guard of Ruanda-Urundi (GTRU). Although this new military in Rwanda was supposed to reflect the population by consisting of 14 percent Tutsi and 86 percent Hutu, in practice it was almost entirely Hutu and led by Belgian officers. Ignoring further UN protests, Belgium granted independence to the Republic of Rwanda in July 1962, with Kayibanda as the country’s first president and PARMEHUTU in control of the legislature. Characterized by a dramatic reversal of power relations, the decolonization process in Rwanda between 1959 and 1962 would be celebrated by the new Hutu regime as the “social revolution.”
From 1959 to 1964, some 336,000 people, mostly Tutsi, fled Rwanda to escape violence and settled as refugees in neighboring Burundi, Tanganyika (Tanzania from 1964), Uganda, and Congo, all of which were in the middle of a decolonization process. Since it had a similar Tutsi monarchy that was eventually replaced by a Tutsi-dominated military regime, Burundi was the most welcoming to the Rwandan Tutsi refugees. In 1961 some of those refugees began to form armed groups, bent on retaking power at home. The exiled movement’s leadership was plagued by internal quarrels between monarchists such as Francois Rukeba, socialists like Gabriel Sebyeza, and apolitical military activists. Some Rwandan exiles went to Communist China for guerrilla warfare training, and the Chinese provided the movement with money to buy weapons. More arms were stolen from a police station in Tanganyika, and a training camp was established at Kigamba in Burundi that became independent in 1962. Given that the guerrillas operated at night, the Rwandan government called the Tutsi rebels inyenzi, or cockroaches, which eventually became a dehumanizing term for all Tutsi. In March 1962, as revenge for a series of cross-border raids by exiles in which some police and civil servants were killed, Rwandan state forces slaughtered 1,000 to 2,000 Tutsi civilians in the northern area of Byumba. The victims’ homes were burned, and their land and possessions were given to local Hutu residents. In late November 1963, some 1,500 Rwandan exiles in Burundi, armed mostly with spears and bows and arrows, marched toward the Rwanda border but were intercepted and disarmed by Burundian authorities.
In late December 1963, the disorganized exiled leadership launched a poorly planned offensive, partly meant to take advantage of problems within Rwanda’s Hutu government. The Rwandan National Guard (GNR), Rwanda’s Belgian-led and almost exclusively Hutu military, was waiting for them. Some 600 insurgents crossed the border from recently independent Uganda and were stopped by the GNR, which killed 300 and handed the rest over to the Ugandan army. Another force from eastern Congo, where the exiles were allied with local Simba rebels, crossed the southern border near Cyangugu and was defeated by the GNR, which executed 90 prisoners. The anticipated contingent from Tanganyika did not show up. A unit of 200 to 300 insurgents crossed the Burundi border at Nemba, captured a police station where they seized weapons and ammunition, and advanced to Nyamata, where they were welcomed by Tutsi inhabitants of a refugee camp. By the time the force arrived at Kanzenze Bridge, just 15 kilometers south of Kigali, it had grown to 1,000 poorly armed people, who the GNR easily gunned down and dispersed. The attempted invasion had failed.
Rwanda’s Kayibanda administration responded to the incursions by organizing Hutu civilian militias that carried out a wave of reprisals against the Tutsi population. The regime encouraged violence by promoting an atmosphere of panic about further Tutsi incursions. Although the Rwandan government claimed that 750 people were killed in this period, other sources put the figure at around 10,000 to 14,000 victims, and some claim it was as high as 20,000. Among the dead were all the moderate Tutsi political leaders still inside the country, who were arrested and taken to Ruhengeri in the north, where they were executed. The most intense violence took place in the Gikongoro district in the south, which included the old royal capital of Nyanza, where between 5,000 and 8,000 people were killed, who represented between 10 and 20 percent of the district’s Tutsi population. The prefect of Gikongoro reputedly told a meeting of mayors and PARMEHUTU supporters that “we are expected to defend ourselves. The only way to go about it is to paralyse the Tutsi. How? They must be killed.”17 The UN remained mute; Burundi, given its sympathy for the exiles, was the only state to protest. The violence prompted more Tutsi to leave Rwanda, and, by 1990, around 1 million Rwandans lived in exile. In 1966 the Tutsi exile movement launched a few more raids from Burundi but collapsed shortly thereafter amid leadership quarrels. Many of its members became caught up in the insurgency in eastern Congo that was eventually suppressed by American-sponsored mercenaries and Belgian paratroopers. Within Rwanda, the exile raids allowed the Kayibanda government to overcome internal divisions and bolstered the reputation of the GNR as the heroic defender of the nation. As Mamdani has noted, the elimination of the Tutsi leaders within the new regime meant that the PARMEHUTU government could portray itself as “a native post-revolutionary republic in which the Tutsi would be tolerated only so long as they remained outside of the political sphere.”18
Although there was no official international protest, the massacres of Tutsi in Rwanda during late 1963 and early 1964 inspired comparisons with the Holocaust of the Second World War and were repeatedly described as genocide. In January 1964, two UN officials working in Rwanda resigned because they could no longer remain in a country “which is practising genocide.”19 In early February, a British newspaper reported that British subjects who had visited Rwanda and were staying in Nairobi, Kenya, had made “allegations that the Rwanda Government is engaged on a deliberate policy of genocide against the country’s former rulers.”20 A few days later, prominent British philosopher Bertrand Russell, writing for a French newspaper, described the killing as “the most horrible and systematic massacre since the extermination of the Jews by the Nazis.”21 Around the same time, Vatican Radio broadcast that “since the Genocide of the Jews by Hitler, the most terrible systematic genocide is taking place in the heart of Africa.”22 A British newspaper reported that “local observers, who can move around fairly freely, consider that genocide is not too strong a word to describe what is going on.”23 Margery Perham, a British pioneer of African history, urged that “at least Rwanda should be expelled from the United Nations for such an appalling breach of the convention on human rights and genocide.”24 Of course, that did not happen. There were also those who believed that it was inappropriate to categorize the massacres as genocide. The Kayibanda government accused its critics of slander and neocolonial agendas.25 Thomas Jamieson, the director of field operations for the UN High Commission for Refugees (UNHCR), thought that the reports of killings in Rwanda were exaggerated.26 Burundian prime minister Pierre Ngendandumwe, who was trying to calm the situation and convince the new Organization of African Unity (OAU) to convene a conference on the matter, called Russell’s statement an “inaccurate generalization” and claimed “that while the majority of those being killed were Watutsi, many Wahutu had also been killed because the massacres were directed primarily at the opposition UNAR party [National Rwandese Union].”27 The next year, Ngendandumwe, the first Hutu prime minister of Burundi, was assassinated by a Tutsi refugee from Rwanda. In retrospect, the massacres of Tutsi in 1963–1964 would seem to correspond to the international legal definition of genocide; they were intentional and aimed at the extermination of at least part of a group defined along racial lines.
In newly independent Rwanda, President Kayibanda became a secretive and authoritarian ruler who demanded the same unquestioning obedience as the old monarchy and ruled by manipulating divisions within the dominant elite. Kayibanda generally favored Hutu elites from the southern part of the country and played them against those from the north. During the middle-to-late 1960s, a potential rival political faction was eliminated when the southern Hutu leaders of APROSOMA were gradually pushed out of government. Ultimately, Kayibanda depended increasingly on a small circle of supporters from his hometown of Gitarama in central Rwanda. The PARMEHUTU regime promoted an official ethos of strong Catholic faith and pride in the hardworking but penurious life of the Hutu peasantry. The president led by example through his austere personal lifestyle—unusual among African heads of state. Popular support among the Hutu was garnered through the abolition of forced labor and the redistribution of pasture land formerly controlled by the old royal court and now devoted to cultivation, which increased food production and ended famine. The colonial system of local administration through chiefs was replaced by a hierarchy of elected officials. Some academics celebrated the majoritarian gains of the revolution, which had overturned Tutsi colonialism. At the same time, Kayibanda’s government continued the racial system of the Belgian colonial rulers: Rwandan citizens were required to carry identity cards that indicated a racial status of Tutsi, Hutu, or Twa, and the regime adopted a quota system whereby a racial group’s percentage of the total population determined their access to civil service jobs and public education. Allocated 9 percent of the quota, Tutsi opportunities were limited. Tutsi could be active in church, business, and education sectors and in the civil service, but they were excluded from politics and portrayed as essentially foreign. Kayibanda’s regime saw Rwanda as an indigenous Hutu nation with a Hutu republic. Within the region, Rwanda became isolated during the late 1960s and early 1970s; it stayed out of the civil war in eastern Congo, feared the Tutsi military government in Burundi, and could not credibly engage in the anticolonial rhetoric of Julius Nyerere’s Tanzania. Rwanda’s main external sponsor remained Belgium, the former colonial power, and it developed ties with Christian nationalist politicians in West Germany. Known as the “Hermit of Gitarama,” Kayibanda rarely left Rwanda. At the time, Rwanda was one of the poorest countries in the world, and its economy was extremely vulnerable to fluctuations in the world price of its coffee and tea exports.
Events in the adjacent countries of Burundi and Rwanda have often influenced each another with tragic results. In April 1972, an incursion by Tanzanian-based Hutu insurgents into Burundi was defeated, which led to the Burundian Tutsi military regime and its militias killing between 100,000 and 200,000 Hutu citizens. As a result, several hundred thousand Burundian Hutu refugees fled to neighboring countries, including Rwanda, which, given its Hutu government, was obviously seen as sympathetic. Kayibanda attempted to use the situation to address several internal problems faced by his regime. For a few years, there had been discontent among Hutu students who saw that, despite government regulations, Tutsi still dominated the educational sphere, amid the increasing number of Hutu school graduates who could not get jobs because the government racial quotas did not apply to the private sector. In late 1972 and early 1973, the increasingly out-of-touch Kayibanda tried to exploit the Burundian refugee situation to recreate the anti-inyenzi fervor of 1963: he launched a government campaign to enforce the 9 percent quota on the Tutsi for government jobs and school placements. Many people were expelled from the civil service, and, at the National University of Rwanda in Butare, students’ family histories were scrutinized to determine if they were truly Hutu. Radio broadcasts incited the Hutu to defend themselves, and extremists called for a final solution to the Tutsi problem. The oppressive campaign, which resulted in around 750 deaths, led to another major exodus of fearful Tutsis from the country. Kayibanda’s plan backfired, however, as violence between rival Hutu factions broke out in rural areas. The regime seemed to be losing control.28
In early July 1973, Major General Juvenal Habyarimana, commander of the Rwandan military, staged a coup that seized political power and restored order. Kayibanda was imprisoned and probably starved to death. Hutu from the southern and central parts of the country were removed from government and replaced by Hutu from the north, Habyarimana’s home region. The new military regime abandoned the old rhetoric of racial ideology, which was anathema to most independent African countries and portrayed the Tutsi, Hutu, and Twa identities in terms of the familiar concept of ethnicity. The Tutsi were no longer seen as a foreign race but as a previously advantaged minority ethnic group that had to be restricted for the benefit of others, while at the same time protected. In the same vein, northern Hutu were seen as a previously disadvantaged majority who became the beneficiaries of affirmative action. As such, the identity cards and quota system were retained, though not as rigidly supervised as before, and a very few Tutsi were given posts within the central government and diplomatic corps. However, Tutsi were almost completely excluded from local government and the military, where Hutu officers were forbidden to marry Tutsi women. An unofficial understanding developed whereby the Tutsi stayed out of politics and were allowed to prosper in private business and rise to prominence in the Kigali social scene. Although it seems ironic now, some referred to Habyarimana as the “protector of the Tutsi.”
In 1978 Habyarimana transformed his military regime into a civilian one-party state under the National Revolutionary Movement for Development (MRND), which all citizens, regardless of ethnicity, were required to join. The new constitution was modeled on that of Tanzania, which, like many other African countries at the time, employed a one-party system. In the Rwandan elections of 1983 and 1988, Habyarimana was the only presidential candidate and was reelected with over 99 percent support. Some critics of the regime quietly disappeared, but there were no massacres. Under the guise of a traditional practice called Umuganda, the state reintroduced a version of forced labor that obliged citizens to engage in several days of work per month on public projects, which sometimes included estates owned by government elites. In terms of local administration, the MRND returned to the practice of the colonial state, with the central government appointing local officials who held dictatorial power over their communities, including the right to call for Umuganda. Internationally, the hitherto isolationist Rwanda joined the French-sponsored Economic Community of the Great Lakes Countries (CEPGL) in 1976 and the World Bank–backed Kagera River Basin Organization in 1977, and the amount of foreign aid money and expatriate aid workers increased substantially. The country’s economic, health, and educational systems all improved. In 1978 Habyarimana signed a defense agreement with France, which was seeking to bolster its alliance of Francophone African countries, mostly former French colonies, against imagined encroachment by their Anglophone neighbors. Habyarimana became a close ally of Mobutu Sese Seko, the similarly anticommunist and notoriously corrupt dictator of Zaire (formerly the Belgian Congo), who had also shifted from Belgian to French sponsorship. Although elements of Habyarimana’s regime such as the one-party state, the rhetoric of ethnic cooperation, compulsory communal labor, and select political murder were common in postcolonial Africa of the 1970s and 1980s, the degree of social control in Rwanda was unusual for a noncommunist country.29
During the Kayibanda and Habyarimana eras, a large and growing refugee population of mostly Tutsis developed outside the country. By the end of 1964, they numbered around 360,000, with 200,000 in Burundi, 78,000 in Uganda, 36,000 in Tanzania, and 22,000 in Congo. By the end of the 1980s, the total figure had increased to around 500,000 and may have been as high as 700,000, with diasporic communities in Europe and North America. There are disagreements over the total number of Rwandan refugees at different times, because they were sometimes confused with other people who had left Rwanda as labor migrants during the colonial era and who had assimilated into such countries as Zaire and Uganda. In addition, not all Rwandan exiles were officially registered as refugees in host countries, which made counting them difficult. The number of exiles was also exaggerated by refugee activists who wanted to vilify the Kigali regime for excluding so many of its citizens and by the successive Rwandan governments who wanted to make it seem impossibly difficult to accommodate so many returnees. Both the Kayibanda and Habyarimana regimes refused to address the exile issue. Throughout the 1970s and 1980s, Rwandan exiles, many of whom had been spirited out of Rwanda as small children or were born in other countries, developed their own identity through a network of associations and periodic publications and promoted a fantasy of Rwanda as a paradise from which they had been expelled.30
During the 1960s, the postcolonial government of Milton Obote and his Uganda People’s Congress (UPC) discriminated against Rwandan refugees in Uganda. Many Rwandan refugees lived in Uganda’s Nkore (Ankole) area, which shared a border with Rwanda, where they became involved in familiar political disputes between local Hima pastoralists, who had been favored by the outgoing British colonial regime, and Iru agriculturalists, who were coming to dominate local government. Whereas the Iru mostly supported the UPC, the Rwandan exiles, who were mostly Tutsi, sided with their Hima counterparts and backed the Uganda’s Democratic Party (DP), which became the opposition after Uganda’s 1962 independence. Hostility from both the new Obote central government and the Iru majority of the border region meant that the armed movement among Rwandan exiles in Uganda was greatly restricted. By 1969, Prime Minister Obote was planning to conduct a survey that would be used to exclude Rwandan exiles from Ugandan politics and eventually to eject them from the country. Therefore, Rwandan exiles welcomed the military coup that brought Idi Amin to power in 1971 and Amin appeared to favor them as an anti-Obote group. However, while a few Rwandan exiles served in Amin’s notorious secret police, they still faced discrimination and obstacles to citizenship.31
Among the many people who fled Amin’s brutal dictatorship in Uganda during the 1970s were some of Rwandan origins, who joined the Tanzania-based Front for National Salvation (FRONASA), led by Yoweri Museveni, who was from a Nkore Hima background. As such, Rwandan exiles were among the Ugandan exiles who participated in the 1979 Tanzanian invasion of Uganda that vanquished Amin. Nevertheless, after Obote’s return to power in 1980, his new regime engaged in active repression against the Rwandan community. In 1982 some 40,000 people in southern Uganda were driven toward the Rwandan border, with thousands caught in no-man’s-land for many months. A few thousand more Rwandans were driven into Tanzania. Those evictions overlapped with continuing tensions within the Ugandan Nkore area, in which Iru, who were supporters of Obote, sympathized with Rwandan Hutu aspirations, and Hima, who were associated with Rwandan Tutsi exiles, were similarly victimized. In 1979 Rwandan exiles in Uganda formed the Rwandese Refugee Welfare Association (RRWA) to assist those who were victimized after the fall of Amin. The next year, the group changed its name to the Rwandese Alliance for National Unity (RANU), which reflected a more militant political stance, including the ambition of an eventual return to Rwanda. However, the restoration of Obote’s administration in Uganda meant that in 1981 RANU had to relocate to Nairobi, Kenya, where it remained until 1986. The Kenya-based RANU facilitated the recruitment of Rwandan exiles into Museveni’s new National Resistance Movement/National Resistance Army (NRM/NRA), which waged a guerrilla struggle against the Obote regime in southern Uganda during the first half of the 1980s. Famously, Museveni personally led the small NRA unit that launched the campaign in 1981, and the unit included Rwandan exiles Fred Rwigyema and Paul Kagame. Obote’s denunciation of the NRA as a foreign Rwandan group—which included the portrayal of Museveni as a Rwandan invader because of his pastoral Nkore Hima heritage and the victimization of Uganda’s Rwandan refugees, many of whom languished at border camps because Rwanda would not admit them—encouraged more exiles to join Museveni. Although the Habyarimana government grudgingly agreed to accept some Rwandan exiles to relieve pressure on the Uganda border camps, it is unclear how many were actually resettled, and Obote’s security forces continued to harass the refugees. When the victorious NRA seized Kampala in January 1986, its 14,000 personnel included some 4,000 Rwandans, many of whom were now experienced military leaders and soldiers. In 1987 RANU headquarters returned to Kampala with the popular Rwigyema as the group’s president. It recruited more Rwandan exiles from across the region and changed its name to the Rwandan Patriotic Front (RPF), which illustrated an even greater militancy that sought to overthrow the Habyarimana regime and to repatriate Rwandan refugees. The next year, a world congress of Rwandan exiles in Washington, D.C. passed resolutions about a “right of return,” a term borrowed from Israel, which were communicated to Rwanda’s Habyarimana government.
Within Museveni’s new NRM administration in Uganda, the upper strata of the military contained a disproportionally high number of officers of Rwandan origin, as they had risen to prominence and gained combat experience during the insurgency. These included Major-General Rwigyema, who was both deputy defense minister and deputy army commander-in-chief; Major Kagame, who was head of military intelligence; Peter Bayingana, who was head of medical services; and Major Chris Bunyenyezi, who was a brigade commander. While it initially appeared that Museveni would grant citizenship to Rwandan exiles as a reward for their contribution in the liberation of Uganda from a series of bloody regimes, rising xenophobia among Ugandans and the NRM quest for political legitimacy made this difficult. Furthering the problem was the accusation that two Rwandan NRA officers, Bunyenyezi and Stephen Nduguta, had committed atrocities during counterinsurgency operations in marginalized northern Uganda in the late 1980s. In the south, economic competition between incoming Rwandan exiles and local people also became a factor. As a result, Museveni began to remove Rwandans such as Rwigyema and Kagame from important positions. Although many long-term Rwandan refugees in Uganda had resolved to stay in that country, the reemergence of local hostility toward them stimulated a dramatic move to reclaim a lost homeland. It was an opportune time to strike, as a recent decrease in world coffee prices had hurt the Rwandan economy, rendering the country vulnerable to attack. Furthermore, Habyarimana had just announced plans to allow more exiles to return, which meant that the RPF had to act quickly before their grievances lost legitimacy.32
During September 1990, Rwigyema, using the cover that he was organizing a Ugandan Independence Day parade, gathered Rwandan members of the Ugandan army near the Rwanda border. On October 1, 1990, some 2,500 out of 4,000 Rwandan NRA soldiers, led by Rwigyema and other senior officers, seized the border post at Kagitumba and invaded Rwanda. Calling themselves “Inkotanyi,” or “those who struggle together,” the newly constituted RPF army was equipped with small arms, heavy weapons, and communications vehicles taken from the Ugandan army. They had no armored vehicles or heavy artillery, which would have been difficult to acquire surreptitiously, and carried limited fuel and ammunition because they expected to seize Kigali quickly. Although Habyarimana’s 5,200-strong Rwanda Armed Forces (FAR) possessed armored cars, artillery, and helicopters, they had no recent combat experience and were caught off guard by the incursion. As such, the RPF quickly advanced 60 kilometers south of the border to capture the town of Gabiro. However, the RPF’s problems mounted: supply lines from the Ugandan border became overstretched; the Ugandan government erected roadblocks to block other NRA Rwandan soldiers from joining the expedition; and civilian Rwandan refugees, trying to return home, flooded the roads near the border. On the second day of the operation, given a dispute over tactics, Majors Bunyenyezi and Bayingana murdered Rwigyema. They were later arrested and returned to Uganda, where they were executed. During October, the FAR staged a phoney RPF attack on Kigali as pretext to arrest thousands of suspected rebel sympathizers, most of whom were Tutsi, and FAR soldiers massacred Tutsi civilians.
France quickly intervened, given its defense agreement with the Habyarimana regime. French authorities attempted to discredit and vilify the RPF by calling them “Khmer Noir” (Black Khmers), a reference to the genocidal Khmer Rouge in 1970s Cambodia, and portrayed them as part of a broad plan to expand Anglophone, particularly American, influence in the region. Paris sent heavy weapons to the FAR and launched Operation Noroit (North Wind), in which over 500 French paratroopers were dispatched to Rwanda. Habyarimana put the head of the French military mission, Lieutenant-Colonel Gilles Chollet, in charge of operations, which effectively made him the unofficial FAR commander. Some 400 Belgian troops were sent to Rwanda to protect Belgian citizens, but they were soon withdrawn. Mobutu directed 1,000 elite Zairean soldiers to help his ally in Rwanda, but their looting, raping, and brutalizing of civilians turned them into a liability, and they were sent home. Supported by French-piloted attack helicopters, an FAR counteroffensive began on October 7, and, by the end of the month, the leaderless RPF, enduring heavy casualties and desertions, had been pushed back to Akagera National Park in the northeast.
The repulsed RPF consolidated and continued its struggle. In late October, Kagame returned from a military course in the United States, took command of the beleaguered RPF forces, and gained permission from Museveni to lead his men through Ugandan territory to the Virunga Mountains in northern Rwanda, where they regrouped. Colonel Alexis Kanyarengwe, a Hutu former FAR officer exiled after a failed coup attempt in 1980, was appointed RPF chairman to reduce the appearance that the movement was predominantly Tutsi. Recruits from the Rwandan diaspora flocked to the RPF—at first mostly from Burundi, but eventually from Zaire, Tanzania, Europe, and North America. The RPF grew from 5,000 in early 1991, to 12,000 at the end of 1992, to 25,000 in April 1994. Museveni’s administration looked the other way as Ugandan army officers supplied their former colleagues with ammunition, weapons, and equipment. Funded by the Rwandan diaspora and facilitated by Uganda, the RPF also purchased bargain-price ammunition and weapons from recently independent and penniless former Soviet republics.
As the war dragged on, the FAR was dramatically expanded from 5,200 in October 1990, to 30,000 by the end of 1991, to 50,000 by the middle of 1992. The recruits were mostly landless Hutu peasants or unemployed youth, who were given just 15 days training before deployment. The lack of training meant that French military advisors had to adopt an active role in combat. Additionally, the war drained the resources of the Rwandan state. Although Rwanda received a credit of $41 million (U.S.) from the International Monetary Fund in 1991 and imposed a special tax to pay for the war, by the end of the 1991 financial year, there was a budget deficit of $188 million (U.S.).33
In January 1991, the RPF launched a guerrilla campaign with a raid on a prison in the northern town of Ruhengeri; the raid was embarrassing to Habyarimana and his inner circle as that was their home area. The Rwandan state replied by massacring between 300 and 1,000 Tutsi civilians in the northwest. Based in the north, the RPF closed the road to Uganda, which meant land-locked Rwanda’s imports and exports had to be transported via the longer and more costly route through Tanzania. Both sides lacked commitment to negotiations and cease-fires; the FAR was confident in French military assistance, and the RPF believed it could not be dislodged from its Virunga mountain sanctuary. After the failure of an FAR offensive at the end of 1991, the RPF launched its own offensive in early 1992, which took control of the agriculturally rich Byumba area. Mutinous and ill-disciplined FAR soldiers undertook massacres of Tutsi, separate from state-ordered actions. At the same time, Hutu peasants feared revenge by the Tutsi and fled RPF-occupied territory, which revealed the movement’s lack of popular support.
In April 1992, Habyarimana introduced multiparty democracy, which was influenced by both the immediate Rwandan conflict and political reforms sweeping across post–Cold War Africa. In Rwanda, that democracy resulted in the emergence of anti-Tutsi extremist parties such as the Coalition for the Defense of the Republic (CDR) and pushed the RPF into serious negotiations because it could no longer claim to be fighting a one-party dictatorship. During 1992, Hutu Power groups formed “civil defense” militias such as the ruling MRND’s Interahamwe (those who work together) and the CDR’s Impuzamugambi (those with a common purpose) trained by French military advisors. Negotiations in Arusha, Tanzania, produced an unstable cease-fire that lasted from July 1992 to February 1993 and was monitored by a Neutral Military Observer Group (NMOG) with personnel from Francophone and Anglophone African countries outside the region (Senegal, Nigeria, Zimbabwe, and Mali) and that reported to an Organization of African States (OAU) joint military-political commission in Ethiopia. The negotiations enraged Hutu extremists, including many within the government and military, who disapproved of their president’s engagement with the Tutsi enemy, with whom they might have to share power. It was likely at that point, perhaps sometime in early 1992, that Hutu extremists began discussing the possibility of a mass slaughter of the Tutsi to avoid that eventuality. Within the Rwandan state, shadowy Hutu extremist organizations were formed such as the Zero Network, which focused on assassination and those known as “bullets” within the army who believed the war was not being waged with enough intensity. In November 1992, Leon Mugesera, an MRND official in Gisenyi, told party supporters that “the fatal mistake we made in 1959 was to let them get out . . . They belong in Ethiopia and we are going to find them a shortcut to get there by throwing them into the Nyabarongo river. I must insist on this point. We have to act. Wipe them all out!”34
A breakdown in the Arusha negotiations led to the January 1993 killing of 300 Tutsi civilians in northern Rwanda. In early February, as a result, the RPF mounted a successful offensive in which it captured Ruhengeri, doubled the territory under its control, and advanced to within 23 kilometers of Kigali. On February 20, the RPF unilaterally declared a cease-fire, as it did not want a direct confrontation with French forces defending the capital, and a major battle over Kigali would unite the city’s inhabitants behind Habyarimana and scuttle the chance of a negotiated settlement. Almost the entire population of the areas newly occupied by the RPF—some 800,000 people—fled south to avoid what the state claimed was a return of Tutsi feudal oppression. In retaliation for previous massacres, the RPF killed local Hutu officials and their families. In April, both sides returned to the Arusha negotiations; Habyarimana’s forces had been largely defeated, and the RPF was under pressure from the UN, which intervened in the crisis for the first time. The RPF agreed to withdraw from the territory it had conquered in February, which was designated a demilitarized zone (DMZ) and initially patrolled by NMOG, and the small UN Observer Mission to Uganda-Rwanda (UNOMUR) tried to ensure that the RPF was not being supplied via the Uganda border.35
Both sides in the war employed print and radio propaganda, but Hutu Power media had a much greater impact. In 1987 a newspaper called Kanguka (wake up), launched by a Tutsi businessman in Rwanda, began to criticize the Habyarimana government and openly sympathized with the RPF, even after it had invaded the country. In 1990, shortly after the initial RPF invasion, Habyarimana’s MRND funded a French and Kinyarwanda newspaper called Kangura (wake others up) as a foil to Kanguka. In December of that year, Kangura published the “Hutu Ten Commandments.” The commandments discouraged Hutu men from associating with Tutsi women; prohibited Hutu from doing business with Tutsi; claimed all important government and administrative positions for Hutu; demanded the majority of educational jobs be occupied by Hutu; emphasized that all military personal must be Hutu and that soldiers must not marry Tutsi; warned Hutu not to have mercy on Tutsi; urged Hutu to unite; and insisted on the proselytizing of the Hutu ideology of the 1959–1962 social revolution. Once the CDR was formed, Kangura and its funders shifted toward that more-extreme movement and criticized Habyarimana for betraying the Hutu cause in negotiations. From its inception, Kangura published articles hinting at or directly calling for the extermination of RPF sympathizers, who were generally assumed to belong to the Tutsi identity, and, in January 1993, it predicted that Habyarimana would be assassinated in March. In mid-1991, the RPF established Radio Muhabura (leading the way) in Uganda, which encouraged armed resistance to the Rwandan state, read official statements by RPF leaders, encouraged FAR soldiers to desert, accused Habyarimana’s military and militia of committing massacres, and denied RPF involvement in atrocities. However, the station’s impact was limited, given its vague reporting, its inability to broadcast as far as southern Rwanda, and its use of English language. Radio Muhabura generally avoided the terms “Hutu” and “Tutsi,” and, when it responded to the January 1993 massacre of Tutsi by stating that “the Kigali regime has now embarked on genocide,” it did not specify whom the genocide was targeting. In July 1993, a radio station called Radio-Television Libre des Mille Collines (RTLM, or Thousand Hills Radio-Television) was established in Rwanda to broadcast anti-RPF and anti-Tutsi propaganda. With some of the same Hutu extremist funders as Kangura, it condemned the Habyarimana government for participating in the Arusha process and the UN for supposedly siding with the Tutsi.36
In early August 1993, Habyarimana’s regime, moderate-opposition groups, and the RPF signed the Arusha Accords, in which they agreed to form a transitional government of all political parties, allow the return of hundreds of thousands of refugees, and create integrated security forces. Since the RPF-occupied area was in the north and thus removed from the capital, its representatives in Kigali’s transitional parliament were protected by a 600-strong RPF battalion. Some 2,500 peacekeepers of the UN Assistance Mission for Rwanda (UNAMIR), commanded by Canadian Lieutenant-General Romeo Dallaire, would supervise implementation of the agreement and monitor the DMZ. Woefully unprepared, UNAMIR lacked an intelligence unit, had no ability to investigate violence, and maintained only two combat-capable infantry battalions—one Belgian and another Ghanaian. Dallaire had not heard of Rwanda before he was sent there, and he arrived in the country with only a tourist map and no staff. In October 1993, as UN peacekeepers began to arrive in Rwanda, the Tutsi-dominated army in Burundi murdered the democratically elected Hutu president and massacred tens of thousands of Hutu, which forced approximately 200,000 Hutu refugees into Rwanda. For Rwanda’s Hutu radicals, that act demonstrated the dangers of a possible Tutsi-RPF takeover. As stipulated in the Arusha agreement, France’s Operation Noroit concluded in December with the withdrawal of the main French intervention force, but French military advisors and mercenaries remained with the FAR. In January 1994, Dallaire reported to UN headquarters in New York that an informant had told him of a Hutu militant plan to assassinate members of the transitional government and eliminate the Tutsi population, and he outlined a plan to seize weapons caches and requested reinforcements and additional equipment for UNAMIR. However, given the lack of strategic interest in Rwanda by Western powers and the recent disastrous American intervention in Somalia, the UN head of peacekeeping and later secretary-general Kofi Annan ignored the warning and forbade UNAMIR from taking action. That would turn out to be a catastrophic mistake.37
During the early evening of April 6, 1994, Rwanda’s presidential jet—carrying Habyarimana and Burundian president Cyprien Ntaryamira—was approaching Kigali airport when it was destroyed by a surface-to-air missile. The identity of the assassins has been contested hotly. From the time the incident took place, Hutu Power advocates accused the RPF of shooting down the plane in a bid to restart the war and take over the country. A decade later, in 2004, French judge Jean-Louis Bruguiere, who investigated the incident as a result of the deaths of the aircraft’s French crew, accused nine RPF leaders, including Kagame—by that time president of RPF-ruled Rwanda—of complicity in the destruction of the aircraft. In 2006 the same judge issued international arrest warrants. The RPF strongly denied those claims and attributed the accusation to an attempt by the French state to conceal its complicity in the 1994 genocide, which is discussed below. Indeed, that dispute prompted Kagame’s government to break diplomatic ties with Paris in 2006. However, another French judge later criticized Bruguiere’s report for relying too heavily on the testimony of a small number of former RPF soldiers, who subsequently changed their stories. In 2010 Rwanda’s RPF government issued the findings of its own investigation, which concluded that Hutu Power supporters within the FAR shot down Habyarimana’s aircraft. France’s reestablishment of diplomatic links with Rwanda that year seemed to indicate its acceptance of the report. Another French investigation, the results of which were released at the start of 2012, stated that the missile could not have been launched from the RPF base in Kigali but that it had come from the FAR’s Kanombe barracks, which was the home of Habyarimana’s presidential guard. In response, former high-ranking members of the RPF, now exiled from Rwanda and involved in opposition politics, stated that they knew back in 1994 that Kagame had indeed ordered the downing of the plane.38
It seems most likely that the assassination of Habyarimana represented the start of a Hutu extremist coup within the Rwandan state and military led by ministry of defense official Colonel Theoneste Bagosora. After the aircraft was destroyed and the killing of Tutsi began, some fighting broke out between FAR units, as not all officers were party to the conspiracy that seemed to mostly involve those from the north—the home area of Habyarimana’s inner circle. Within a week of the assassination, the conspirators formed a transitional government that implemented a policy of eliminating the Tutsi population to avoid power sharing through the Arusha Accord. Almost immediately after Habyarimana’s death, Hutu militias and FAR units erected roadblocks in Kigali and began the systematic slaughter of Tutsi civilians. The process was facilitated by the identity card system, which indicated a person’s status as Tutsi, Hutu, or Twa. Radio broadcasts by RTML encouraged the killings and announced the names and addresses of potential victims as well as suspected hiding places. The massacres quickly spread to other parts of the country, and, over the next three months, almost one million people, mostly Tutsi, were murdered. Neighbors killed neighbors, and killing was often conducted in public. While some have claimed that a high percentage of the adult male Hutu population took part in the murders, a specialized study by Scott Straus maintains that only around 200,000 people were involved, and, among those, only a small minority conducted most of the killings. In addition to the established militias, killers were organized through the Umuganda communal labor system, which had been implemented in the 1970s, and the murderers often referred to killing as “work.” Among the murderous mobs, there was a variety of motivations: a few were hard-line Hutu radicals, eager to kill; others feared retribution if they did not participate; some saw opportunities for looting property or grabbing land, and freely flowing beer removed inhibitions. The weapons were generally cheap, everyday objects used in an agricultural society—the most infamous of which was the machete. Rape and other kinds of sexual assault became common and were influenced by the particular hatred that the Hutu Power groups reserved for Tutsi women, who had been seen in the colonial mind as possessing superior beauty. There were not many places to hide or escape in small, densely populated, and highly controlled Rwanda. Some people survived by concealing themselves among corpses, taking shelter with a sympathetic neighbor or relative willing to take a massive risk, or fleeing into wilderness areas, where some were pursued by mobs with hunting dogs. There were ambiguous situations in which people who were engaged in killing also acted to save a relative, friend, or even a stranger for whom they felt momentary pity.39
During the unfolding genocide, many Tutsi sought protection at local churches, which facilitated massacres, as the victims were concentrated and trapped. Since the Catholic Church had a long association with state power in Rwanda, many priests and nuns participated in the killings. For example, in April 1994 Father Athanase Seromba, parish priest at Nyange in western Rwanda, persuaded 2,000 Tutsi civilians to hide in his church. He then urged local militia members to burn the building, and he personally shot people who tried to escape and ordered the church’s remains crushed with a bulldozer. Leaders of Protestant denominations were also involved in massacres. However, a few clergy and nuns worked against the genocide, including Hutu nun Felicitas Niyitegeka, who concealed Tutsi in a church facility in Gisenyi, smuggled some into nearby Zaire, and insisted on dying with them when they were discovered. Since Sister Felicitas’s brother was a senior army officer, she could have avoided retribution.40
On April 7, the day after Habyarimana’s death, the FAR presidential guard disarmed, tortured, and killed 10 Belgian UNAMIR soldiers assigned to protect interim Prime Minister Agathe Uwilingiyimana, a moderate Hutu who had been prevented from going to the radio station to address the country and who was also murdered. The killing of the Belgian soldiers precipitated the sudden withdrawal of the Belgian contingent from UNAMIR, and, despite Dallaire’s request for a 5,000-strong force to stop the massacres, the UN ordered the mission reduced to just 270 personnel. Nevertheless, Ghanaian Brigadier Henry Kwami Anyidoho, UNAMIR deputy commander, and the Ghanaian battalion of 456 troops remained in Kigali, where they protected UN headquarters and several thousand people who sought safety at Amahoro Stadium and escorted humanitarian relief convoys. Another factor that prevented more robust action by UN peacekeepers was that some of their unarmed colleagues, who were spread out in small detachments across the country, became virtual hostages.41
On April 8, French advisors with the FAR cleared vehicles from the Kigali airport’s runway to allow the nighttime landing of a French transport aircraft carrying a 544-strong intervention force. Over the next two days, about 500 Belgian paracommandos and a company of Italian paratroopers landed at the Kigali airport and were joined by the Belgian battalion that had left UNAMIR. Called Operation Amaryllis by the French and Operation Silverback by the Belgians and Italians, the action was meant to evacuate European civilians quickly. Despite French influence among Rwanda’s Hutu politicians and military officers and the popularity of French soldiers among the Hutu militia, those French forces did little to save Tutsi people from violence. By April 14, when RPF forces began mortaring the airport, French, Belgian, and Italian forces had flown out with 2,000 civilians, including Agathe Habyarimana, the late president’s widow and a central figure in the coup, and her entourage.42
The United States was one of a few countries in the world that possessed the military capability to intervene decisively to stop the unfolding genocide in Rwanda. However, from early April, the Clinton administration decided against intervention, given Rwanda’s lack of strategic importance and fear that American troops might get involved in fighting as they had in Somalia the previous year. Despite later claims that they were not immediately aware of the scale of the killing, classified documents reveal that senior officials of the Clinton administration had accurate and timely information on what was happening. Almost from the moment it began, the administration privately discussed the “genocide” and the “final solution” in Rwanda, and the president and vice president were informed. However, in late May, Washington officials played word games by publically describing events in Rwanda as “acts of genocide” to prevent the American public from demanding action. At a press briefing, journalist Alan Elsner asked State Department spokesperson Christine Shelley, “How many acts of genocide does it take to make genocide?” Shelley could not respond. Furthermore, the Clinton administration used its position within the UN Security Council to block the reinforcement of UNAMIR and refused to use American technological capability to disrupt the RTML radio broadcasts that were encouraging and directing the death squads.43
The war between the Rwandan state and the RPF resumed immediately after the downing of Habyarimana’s aircraft. On April 7, the RPF detachment in Kigali left parliament and established a defensive position along the northern route into the city, from where they repelled an FAR attack. The next day, the main RPF force in northern Rwanda launched a three-pronged offensive. One element moved west to Ruhengeri, where it engaged the FAR and protected the right flank of the main southward advance. A second element moved down the eastern frontier, encountered little resistance, and reached the Tanzanian border on April 22, which secured the main force’s left flank. The principal RPF force moved south, and, by April 11, its vanguard had begun to encircle Kigali, while the RPF battalion already there threatened the airport. Consequently, Rwanda’s interim government fled to Gitarama. The RPF forced the closure of the Kigali airport on May 6, cut the Kigali-Gitarama road on May 16, and captured the Kigali airport and military camp on May 22. Attempting to relieve pressure on its new headquarters at Gitarama, the FAR launched a counteroffensive south of Kigali on June 6 that the RPF easily defeated. With the subsequent capture of Gitarama on June 13, the RPF mounted an offensive against Kigali from the northern, eastern, and southern suburbs, with heavy fighting continuing until July 3, when the FAR, its ammunition spent, withdrew from the city, along with most of its civilian inhabitants. During June, the RPF’s eastern prong reached the southeastern corner of Rwanda, turned west to push FAR units along the Burundi border, and, at the start of July, captured Butare. With Kigali secured and the FAR routed, Kagame turned his main force north to seizure Ruhengeri and Gisenyi in mid-July. In late May, with 15,000 of its 20,000 to 25,000 troops engaged in the battle for Kigali, the RPF had begun to recruit young Tutsi genocide survivors and exiles from Burundi, who were eventually blamed for revenge killings of Hutu.44
In mid-May, UN authorities reversed their decision to downsize UNAMIR and ordered its expansion to 5,000 personnel. However, it would take time to assemble international troops and transport them to Rwanda, where the genocide continued. As such, the UN approved a French interim measure called Operation Turquoise, which was launched on June 23 and aimed, at least officially, to create a refugee sanctuary in southwestern Rwanda. The Turquoise zone was occupied by a force of 2,550 French soldiers and 500 others from Francophone African countries, who were flown into a staging area at Goma in neighboring Zaire by the French Air Force, who rented transport aircraft. The United States refused to provide airlifts. The establishment of the Turquoise zone blocked the RPF’s westward advance along the Burundian border. Among the hundreds of thousands of refugees who passed through the French-occupied area en route to eastern Zaire were retreating FAR elements, complete with weapons and equipment, Hutu Power militia, and senior planners of the genocide. Indeed, the genocide continued within the Turquoise zone, with some French troops assisting Hutu militia to locate hiding Tutsi. In late June, a French reconnaissance detachment arrived at Bisesero near the western town of Kibuye, where they convinced Tutsi, who had used bows and arrows, spears, and clubs to fight off Hutu militias, to come out of their forest and hillside hiding places. The French troops then abruptly left the area and returned three days later to find between 1,000 and 2,000 Tutsi murdered. In late August, the French and their allies withdrew and were replaced by Ghanaian UNAMIR peacekeepers, who watched the RPF complete their occupation of the country. Although Operation Turquoise saved between 10,000 and 17,000 people, the small and poorly armed UNAMIR rescued twice as many.45
In mid-July, with the genocide effectively suppressed, the RPF and the Republican Democratic Movement (MDR), the main opposition party under the old regime, and a number of smaller moderate parties formed a government of national unity in the spirit of the Arusha Accords. Pasteur Bizimungu, a Hutu RPF member, became president; Kagame became vice president/minister of defense; and Faustin Twagiramungu, a Hutu MDR member, became prime minister. In the initial cabinet, the RPF held 8 positions, which represented the largest of any party, and there were 15 Hutu and 6 Tutsi ministers. The RPF morphed into the Rwandan Patriotic Army (RPA), the new Rwandan military, which meant that Kagame directly controlled the coercive means of the state. Tensions quickly developed within the government when ministers such as Seth Sendashonga, a Hutu RPF member, began to ask questions about allegations that the RPF was terrorizing the population through massacres and select assassinations.
In April 1995, the Kibeho camp, located in the south of Rwanda and home to 150,000 internally displaced people who were afraid to return home, was forcefully dismantled by two RPA battalions that fired into the crowd. While the new government stated that 338 people had been killed in the massacre, Australian UN peacekeepers present at the camp put the figure at around 5,000. That attack would serve as a model for a subsequent and much deadlier RPA campaign against refugee camps inhabited by Rwandan Hutu in eastern Congo, which will be discussed in another chapter. The Kibeho massacre enflamed tensions within the unity government, which effectively broke down in August with the dismissal of Twagiramungu, Sendashonga, and three other cabinet ministers. In 2000 the politically impotent Bizimungu resigned the presidency, which was assumed by Kagame, who had always been the real power behind the throne. Bizimungu, who formed an opposition party that was hastily banned, was imprisoned for inciting violence and corruption. During the 2003 election, the first under a new constitution, Kagame represented the RPF and gained a remarkable 95 percent of the vote, while his main competitor, Twabiramungu, whose MDR party had been outlawed, gained just under 4 percent. The authoritarian RPF state tolerated the existence of token opposition groups and orchestrated the assassination of exiled political leaders such as Sendashonga, who had moved to Kenya.46
Shortly after coming to power, Rwanda’s RPF-led government formulated new laws for the prosecution of genocide-related offenses. The initial approach emphasized retribution. In 1998 the Rwandan state publically executed 22 people for genocide crimes, with some having been condemned by summary trials without legal assistance. By 2000, there were around 130,000 suspected genocide perpetrators in overcrowded Rwanda prisons awaiting trails that could take over a century to complete within the established Western-style legal system inherited from the colonial era. Compounding the problem was the fact that Rwanda’s postgenocide legal system was in tatters, as many judges and lawyers had been killed or fled the country.
In 2001 the RPF regime instituted the Gacaca system, which was based on Rwandan traditional law and represented a new approach to the issue of genocide-related justice. The principle of revenge was replaced with reconciliation. Gacaca involved just over 12,000 community courts throughout the country that would hear 1.2 million cases. In addition to accelerating the justice system, the Gacaca courts were meant to discover the truth of what had happened in 1994, including finding the location of victims’ remains, eliminating impunity, and encouraging reconciliation and national unity. While the alleged leaders of the genocide and those accused of rape remained in prison, awaiting trial by ordinary courts, the Gacaca system dealt with people accused of murder, attempted murder, torture, assault, and property offenses. Councils consisting of nine elected members had the authority to impose prison sentences ranging from 30 years’ imprisonment for murder to community service for property crimes. The many criticisms of the Gacaca process included complaints of the defendants’ lack of right to legal representation, whereas the councils were advised by professional state prosecutors; allegations that some of the untrained and unpaid councillors engaged in corruption; suspicions that some of the councils included suspected genocide perpetrators; and the prohibition on discussing crimes allegedly committed by the RPF during 1994. Some genocide survivors disliked Gacaca, as it seemed like a way for perpetrators to avoid punishment by seeking forgiveness, and they believed the RPF government was sacrificing justice for political expediency to cultivate legitimacy among the Hutu majority.47
In November 1994, the United Nations established the International Criminal Tribunal for Rwanda (ICTR), which began work in Arusha, Tanzania, the following year. It was mandated to judge people accused of genocide, crimes against humanity, and war crimes in Rwanda during 1994 and developed as a trial venue for those considered to have been leaders of the genocide. In 1998 the ICTR found Jean-Paul Akayesu—mayor of Rwanda’s Taba commune during 1993 and 1994 who had been extradited from Zambia in 1996—guilty of nine counts of genocide and crimes against humanity and sentenced him to life imprisonment. The trial was significant in a number of ways: it was the first instance in which a court enforced the 1948 UN Convention on the Prevention and Punishment of the Crime of Genocide; genocide was clearly distinguished from violations of the Geneva Convention, which regulates warfare; and sexual assault within this context was included among acts of genocide. In 2003 three men involved with RTML broadcasts were convicted of genocide and incitement to genocide and crimes against humanity; they were sentenced to some 30 years’ imprisonment. By May 2015, the ICTR, much criticized for its slowness, had indicted 93 people and completed 55 trials involving 75 accused.48
Since many genocide perpetrators had fled Rwanda in 1994, the justice systems of other countries became involved in the issue. Under the recently adopted principle of “universal jurisdiction,” which extends the ability to prosecute beyond national borders, trials of suspected genocide criminals from Rwanda were held in Belgium, Switzerland, Germany, Canada, Finland, Norway, Sweden, the Netherlands, and France. In the United States, alleged Rwandan genocide perpetrators were tried for immigration offenses, which usually involved gaining access to the country based on false statements. In 2001 a Belgian court convicted two Catholic Rwandan nuns of murder and war crimes for their participation in burning to death some 5,000 people who had taken refuge at their convent in southern Rwanda in 1994. They were sentenced to 15 and 12 years imprisonment, respectively. Both were released after serving half their sentences, which is not unusual in the Belgian justice system, and returned to their Belgian monastery. In 2009 a Canadian court convicted Desire Munyaneza—a shopkeeper in Butare in 1994, who had later fled to Toronto—of genocide, war crimes, and crimes against humanity and sentenced him to life imprisonment with no possibility of parole for 25 years. In Germany in February 2014, Onesphore Rwabukombe, a former mayor in Rwanda, was found guilty of aiding and abetting genocide and sentenced to 14 years in prison. Since France harbored many genocide planners and perpetrators, including Agathe Habyarimana, it was notable when, in March 2014, Pascal Simbikangwa, former intelligence head in the Habyarimana regime, was prosecuted by a new war crimes department, found guilty of genocide and complicity in crimes against humanity, and imprisoned for 25 years. In 2007 Rwanda abolished the death penalty to facilitate the extradition of suspected genocide perpetrators from other countries in which capital punishment did not exist. In 2011 the ICTR transferred its first case to Rwanda, and the same year a decision by the European Court of Human Rights allowed genocide suspect Sylvere Ahorugeze to be extradited from Sweden to Rwanda. In January 2012, after a long legal struggle that went to the Supreme Court of Canada, academic Leon Mugesera was deported to Rwanda, where he was to stand trial for a 1992 speech, partly quoted above, which encouraged the extermination of the Tutsi. The Mugesera case was significant because he had left Rwanda before the 1994 genocide.49
In 2008 the RPF government of Rwanda created a National Commission for the Fight against Genocide to coordinate commemoration activities, develop strategies for fighting genocide and genocide ideology, counter genocide denial and trivialization, and assist genocide survivors. Beginning in 1996, the RPF regime developed a network of officially recognized genocide memorial sites across the country. In 2004 a national genocide memorial and museum was opened at Gisozi in Kigali and was funded and designed by a British-based organization called Aegis Trust, which was involved in Holocaust education. Throughout Rwanda, genocide memorials are often located at former churches or schools in which victims had gathered to seek safety only to be trapped and killed. They often feature various displays of human remains, ranging from neatly piled bones in an ossuary basement at a church in Nyamata to rows of lime-covered desiccated corpses laid out on tables at a former school in Murambi. Those sites became venues for annual memorial ceremonies held every April and have become “dark tourism” attractions for visitors to the country. Advocates for the exhibition of human remains argue that it provides irrefutable proof that the genocide happened—which is important, given genocide denial—and represents an essential element of ensuring that it will never happen again. Critics of the practice point out that it violates the dignity of the dead, has no precedent in Rwandan culture, and presents an obstacle to reconciliation, as the Hutu majority is constantly reminded of its collective guilt. While the Tutsi-led RPF government banned the Tutsi, Hutu, and Twa identities in favor of a broad Rwandan national identity, the memorials and events associated with them serve to remind people of those divisions. Furthermore, the RPF regime stands accused of using the memorials and related events to create an historical narrative that legitimizes and entrenches its power. Western countries are reminded that, given their failure to intervene to stop the 1994 genocide, they do not have the moral authority to criticize the RPF. Similarly, the stark nature of the memorials enables the RPF to portray its critics on such issues as human rights and freedom of speech as dangerous genocide deniers seeking to return Rwanda to the horrific days of 1994. Rwanda has been compared to Israel, which is seen as using the memory of the Holocaust to justify repression of the Palestinians and to silence critics. Indeed, President Kagame has used annual genocide commemoration ceremonies to disparage political opponents and international critics of the RPF, and even to boast about the assassination of a former colleague in South Africa.50
Survivors of Rwanda’s 1994 genocide became active in forming organizations to pursue their interests. Rwandan communities in different parts of the world were central to the effort. In August 1994, survivors and relatives of victims living in Belgium formed an organization called Ibuka (remember) to commemorate the victims, seek prosecution of perpetrators, and assist survivors of the genocide. In May 1995, a similar group formed in Switzerland. The Rwandan-based version of Ibuka launched in November of 1995 and eventually developed into an umbrella organization for different branches and similar groups in Rwanda and others in countries such as France and Canada. Although Ibuka initially cooperated with the ICTR in Tanzania, the relationship deteriorated in the late 2000s over concerns for the safety of witnesses, the acquittal of some suspects, and the reduction of some sentences on appeal. Within Rwanda, Ibuka champions the interests of the roughly 300,000 genocide survivors, who, as a minority within a population of around 12 million, are seen as vulnerable and sometimes feel ignored by the government in the pursuit of national unity. A number of more-specialized survivor groups were also formed. The Association of Widows of the Genocide: Consolation (AVEGA Agahozo) was established in 1995 and by 2002 had 25,000 members and 49 employees. It facilitates medical and psychological treatment and legal assistance and raises money for housing and small-business projects. Other survivor groups include the Association of Orphan Household Heads (AOCM), which was founded in 2000 and had 6,000 members by 2006, and the Association of Student Survivors of Genocide (AERG) formed at the National University of Rwanda in 1998, with graduates then establishing the Group of Former Student Survivors of Genocide (GAERG) in 2002.51
The 1994 genocide in Rwanda has become one of the most written about and filmed events related to modern African history. It represents the subject for an ever-growing literature that includes specialized academic studies, accounts by eyewitnesses and survivors, and even some fictional novels. Furthermore, there are many films relating to the genocide in Rwanda, including several dozen documentaries (some based on books) and more than half a dozen feature films, some of which include internationally acclaimed actors. A number of those films appeared in the mid-2000s, near the tenth anniversary of the genocide. Some of those films have been criticized for inaccurate portrayals: in particular, Hollywood’s 2004 Hotel Rwanda, which tells the supposedly true story of a Hutu hotel manager in Kigali, Paul Rusesabagina, who heroically offered sanctuary to potential victims. However, it has now become known that Rusesabagina—who later denied that there had been a genocide against the Tutsi—allegedly extorted money from those he was protecting, and that the hotel residents were spared because of media attention and the nearby UN presence. Other films such as Shooting Dogs and Sometimes in April, both released in 2005, have been received more positively.52
Despite overwhelming evidence from eyewitnesses that the 1994 genocide in Rwanda targeted people of Tutsi identity, several alternate and often far-fetched theories have been put forth. Those claims have been called “genocide denial.” A very few people, including former UN representative to Rwanda Jacques-Roger Booh-Booh, have claimed that no genocide took place and that the massacres were greatly exaggerated. A few others maintain that the predominantly Tutsi RPF shot down President Habyarimana’s aircraft and then embarked on a genocide against the Hutu, which was subsequently covered up by the U.S. government, which sought to install a friendly regime in Rwanda. As such, it is claimed that most of the hundreds of thousands of corpses that littered the streets and countryside of Rwanda were not really Tutsi but Hutu. Of course, that represents a ridiculous conspiracy theory for which there is no proof. Another view accepts that a genocide took place against the Tutsi, but provocatively blames the mainly Tutsi RPF for purposefully provoking it as part of its strategy to gain power.
There is also a “double genocide” theory that claims that, although Hutu Power extremists did attempt to exterminate the Tutsi population in 1994, the almost-simultaneous RPF counteroffensive represented another genocide against the Hutu. Statements to that effect were made in the press in 1994 by French foreign minister Alain Juppe and French president Francois Mitterrand, key allies of the Habyarimana regime, and subsequently became a common theme in the legal defense of alleged genocide perpetrators before various courts, including the ICTR. The total number of deaths during the genocide has also raised controversial questions about the nature of the genocide. According to a 1991 Rwandan government census, 600,000 Tutsi lived in Rwanda at that time. To use rough figures, if around 350,000 Tutsi survived the genocide, that means somewhere around 250,000 must have been killed. It followed then, that if around 800,000 or more people in total were killed in the genocide, as many sources indicate, the remaining 550,000 victims were not Tutsi. Is it possible that they were Hutu massacred by the RPF? That line of thinking is based on the assumption that the 1991 census was accurate, and it seems entirely possible that the Habyarimana government underestimated the size of the internal Tutsi population to minimize their claims to political representation. It is also likely that the total number of genocide victims has been overstated, with the figure of about 500,000 advanced by some. While it seems certain that RPF elements conducted revenge massacres of Hutu during the fighting in Rwanda in 1994, that cannot be labelled genocide under the international legal definition, as there is no evidence of an RPF plan to exterminate part or all of the Hutu population.53 However, as discussed in the next chapter, there is a possibility that RPF killings of Rwandan Hutu refugees in eastern Congo in 1996 constituted genocide.
The genocides that occurred in Rwanda in 1963–1964 and 1994—similar events though different in the scale of killing—were the result of a combination of factors, some of which date back to the precolonial era. The country has a long history of violence and warfare that intensified during the rule of late-19th-century leader Rwabugiri and continued through the periods of German and Belgian colonial rule. The history of violence was important, because it created an historical memory of terror that was exploited by postcolonial regimes. A series of paranoid and manipulative regimes imposed tight controls: the Nyiginya monarchy, the Belgian colonial rulers, and the two postcolonial Hutu republics. Arguably, the post-1994 RPF regime has continued the tradition. In a way that was distinct from most of the rest of colonial Africa, in which ethnic and regional divide-and-rule became the norm, Belgian officials and missionaries imposed the concepts of race and racial hierarchy on the long-existing Rwandan identities of Tutsi, Hutu, and Twa. The binary of an alien Tutsi minority and an indigenous Hutu majority—the former favored in colonial times and then oppressed by postcolonial states—came to dominate Rwandan society and politics through most of the 20th century and beyond. The genocides of the late 20th century were certainly avoidable and were the result of decisions made by specific leaders. However, those choices were heavily influenced by a dangerous combination of historical factors, including violence, control, fear, and racial ideology within the context of a small and densely populated country. Such a mixture of elements was rare in colonial and postcolonial Africa, which is likely why there are few comparable cases of such clearly defined and total genocide in the continent’s history.