The Burundi Killings of 1972

27 Juin, 2008
Lemarchand René

In the spring of 1972 the small (10,747 sq miles), overpopulated (7 million), poverty-stricken State of Burundi experienced massive bloodletting. Burundi’s agonies did not begin nor end with what is sometimes referred to in Burundi as ikiza, the “scourge”. Nonetheless, there is nothing in the country’s turbulent history comparable to the scale of the 1972 killings. Although the number of victims will never be known, estimates range between 150,000 to 300,000 (Kiraranganya, 1985: 76) To reduce a complicated drama to its simplest common denominator, the vast majority of those killed were of Hutu origins, representing approximately 80 per cent of a total population then numbering approximately four million; the perpetrators were drawn overwhelmingly from the Tutsi minority, accounting for some 15 per cent of the population, its representatives holding full control over the armed forces and the government.

Not all Tutsi were perpetrators, however, nor were all of the victims Hutu. Hutu and Tutsi were both victims and perpetrators -- but each at separate time intervals and with very different scales of involvement. The triggering factor behind the bloodbath was a Hutu-led rural insurrection aimed at seizing power from the ruling Tutsi minority. The fulcrum of the rebellion was in the southern province of Bururi, its leadership consisting of a small group of radicalized Hutu intellectuals, most of them operating from neighboring Tanzania. To the extent that it claimed an ideology, its overtones were militantly anti-Tutsi. In a matter of days, hundreds (possibly thousands) of Tutsi lives were lost. The ensuing repression, however, went far beyond the province most directly affected by violence; its avenging furor swept across the entire country and lasted for months after it had been brought under control.

Its extensiveness and extreme brutality against all Hutu elites have prompted some commentators to refer to a “selective genocide” (Lemarchand, 1974). Yet, to this day, scholars disagree as to whether the 1972 killings should be described as a double genocide, a selective genocide, a genocide or a case of ethnic cleansing run amok. Despite the wealth of data made available by recent research (Chrétien and Dupaquier, 2007) many of the questions raised by these tragic events defy a simple answer.


A - The Context

Like Rwanda, its neighboring “false twin” to the north, pre-colonial Burundi was an archaic kingdom, hierarchically organized, where kingship served as the main focus of popular loyalties. Unlike Rwanda, however, where a Hutu uprising destroyed the monarchy shortly before independence, its formal political organization underwent relatively few changes under German (1896-1916) and Belgian colonial rule (1916-1962). And while both have experienced frightening levels of ethnic violence, the 1972 drama never received anything approaching the extensive media exposure generated by the far more devastating Rwanda genocide. Furthermore, while there is little doubt about the genocidal quality of the Rwanda bloodbath, the case of Burundi has been the source of intense controversy both within and without the country.

To properly grasp the complexity of the events leading to the carnage, something must be said of the regional context. Burundi has many things in common with Rwanda besides a one hundred kilometer boundary: both States share strikingly similar ethnic configurations, social customs and languages. Therefore, it is not surprising if their histories, though propelling their societies in radically different directions, should have influenced each other to a degree unparalleled in any other two States in the continent. The critical difference between their post-independence trajectories lies in the rise to power of distinctive ethnic hegemonies. While Rwanda crossed the threshold of independence as a Hutu-controlled republic, until 1966 Burundi stood as a constitutional monarchy ruled by a government of mixed origins. These contrasting itineraries have had a profoundly divisive impact on the society of Burundi. For if the Rwanda model served as a powerful pole of attraction for many politically conscious Hutu, for the Tutsi elites the same model was a source of intense fear. For a growing number of Hutu republicans, Rwanda was the ideal political formula for Burundi; for the Tutsi it was the embodiment of a nightmare scenario, to be avoided at all cost.

The nightmare almost became reality when, on October 19, 1965, a group of Hutu gendarmerie and army officers unsuccessfully tried to overthrow the regime by attacking the royal palace in the heart of the capital city (Lemarchand, 1994). Although the mutiny quickly collapsed, this did not prevent the outbreak of inter-ethnic violence in the countryside. As armed bands of Hutu vented their anger against innocent Tutsi civilians, hundreds were reported killed in and around Muramvya, north of the capital. In a series of repressive measures, 38 Hutu officers were arrested and shot two days later; again, on October 28, ten leading Hutu politicians were summarily tried and executed; in the following weeks, 86 death sentences were handed down by improvised military tribunals. The next major purge occurred in September 1969, when rumors of another Hutu-engineered coup led to the arrest and execution of scores of influential Hutu personalities. By then the Hutu leadership had been virtually decapitated.

While the attack on the royal palace made dramatically clear the danger of a Hutu take-over, the inherent weakness of the monarchy was cruelly demonstrated when, in the wake of the coup, King Mwambutsa quickly fled to Switzerland, leaving the throne vacant. The accession of Mwambutsa’s heir to the throne on July 8, 1966, under the name of Ntare, did little to restore the prestige of the Crown, or assuage Tutsi fears. After a brief trial of strength between the government, headed by Captain Michel Micombero, and the King, the monarchy was formally abolished on November 28, 1966, making Ntare’s reign the shortest in the history of the country. With the advent of the First Republic and the appointment of a predominantly Tutsi government, Tutsi elements became dominant in the army and the government, while Hutu elites were virtually denied participation in the institutions of the State. By 1967, out of nine provincial governors, only one was a Hutu; the mayors (administrateurs communaux), likewise, were overwhelmingly Tutsi. The army high command would soon follow suit.

The steady rise of Hutu-Tutsi enmities, accompanied by the more or less systematic exclusion of Hutu from the institutions of government, must be seen as a central element in the background of the 1972 killings. Their timing, however, draws attention to another underlying fault-line. Cutting across Hutu-Tutsi polarities was another major social cleavage, involving Tutsi vs. Tutsi. At the root of this intra-ethnic struggle was the long-standing, traditionally based opposition behind the Tutsi-Banyaruguru and the Tutsi-Hima: the former, considered to be closer to the court, were generally seen as ranking considerably higher in the traditional pecking order. With the rise to power of the army, however, something of a reversal of status occurred in this hierarchy of rank and privilege. The real holders of power within the army and the government, including President Micombero, were Tutsi-Hima, many from the Bururi province in the south, which served as a major recruiting ground for the armed forces. Ethno-regional tensions rose rapidly in 1971, as a number of leading personalities of Banyaruguru origins were accused of fomenting a plot to bring back Ntare to the throne.

It is not by chance that the outbreak of a Hutu-led insurrection, in late April 1972, happened at the height of the Hima-Banyaruguru crisis. In the months preceding the Hutu uprising, the government seemed to be tottering on the brink of anarchy. Already in July 1971, charges of conspiracy were brought against a group of leading Banyaruguru personalities, and on January 14, 1972, a military tribunal issued nine death sentences and seven life sentences. With rumors of plots and counterplots spreading across the land, the ruling Hima clique, headed by Micombero, saw its legitimacy plummet. It was precisely when the government’s popularity was at its lowest ebb that the insurgents chose to strike. Ironically, nothing could have done more to solidify its ethnic underpinnings than the looming threat of a violent Hutu uprising.


B - Decision-Makers, Organizers and Actors

On April 29, like a bolt out of the blue, anti-Tutsi violence swept across the lakeside town of Rumonge. In a matter of hours, the rebellion spread to other localities along the shore of Lake Tanganyika, including the southernmost town of Nyanza-Lac, where roving bands of Hutu attacked Tutsi civilians. Countless atrocities were reported by eyewitnesses. In the provincial capital of Bururi, all military and civilian authorities were killed. After taking control of the armories in Rumonge and Nyanza-Lac, the insurgents proceeded to kill every Tutsi in sight, along with a number of Hutu who refused to join the rebellion. A short-lived “Martyazo Republic” was proclaimed in Vyanda, in early May, an experiment quickly brought to an end by government troops sent out to crush the rebellion. Although the uprising lasted only a few days, its cost in human lives, according to missionary sources, is believed to have ranged between 800 and 1200. (Chrétien and Dupaquier, 2007: 106)

Considerable uncertainty surrounds the identity of the instigators. The names that are most frequently cited as constituting the brain trust of the insurrection are those of three former students at the University of Bujumbura, known for their militant radicalism – Celius Mpasha, Albert Butoyi and Daniel Ndaburiye – and the former parliamentarian, Ezechias Biyorero (aka Yussuf Ibrahim), all of them living in Tanzania since 1969. (Kiraranganiya, 1985). Of these, Biyorero is said to have played the key role in planning and organizing the insurrection. The overall impression one gets, from what little reliable data is available, is that of a loosely knit group of young radicals trying to mobilize the more underprivileged segments of Hutu society – workers and peasants -- through a mix of racist and revolutionary slogans. While some of the perpetrators came from Tanzania, others were locally recruited among the unemployed youth. Both operated in small bands of insurgents, varying in strength from 50 to 100 and more, armed mostly with spears and machetes. According to some eyewitnesses, they were joined in some places, as in Nyanza-Lac, by groups of “Mulelistes”, a label evocative of the late Pierre Mulele, the key organizer of the Kwilu rebellion in Zaire (now the Democratic Republic of Congo [DRC]) in 1964-65; in this context, however, the so-called Mulelistes were probably remnants of Gaston Soumialot’s rebel army from eastern Congo. Just as in the mid-1960s, many of Soumialot’s warriors had recourse to drugs and magic to claim invulnerability to bullets, similar rituals and whistle-blowing commands were observed in Nyanza-Lac and other localities. Just how many were involved remains unclear.

A French pilot who flew helicopter missions on behalf of the Burundi army put the total number of insurgents at one thousand, “including the majority of committed or conscripted Hutu, Zairian Mulelistes in the middle and the Hutu organizers at the top” (Hoyt, 1972, cited in Lemarchand: 204, 158).

The government response, at any rate, was immediate, and its wide-ranging follow-through devastating. Micombero’s decision to dissolve his government, on April 29, hours before the eruption of Hutu violence, while still unclear, may have been motivated by his desire to have a free hand in dealing with a foretold insurrection. On April 30, while counterattacks were launched against insurgents, elements of the armed forces began to coordinate their efforts to exterminate all Hutu suspected to have taken part in the rebellion. The repression would soon reach a nation-wide scale, with purges affecting every sector of the civil society.

The men in charge of coordinating military operations were high-ranking officers: Chief of Staff Thomas Ndabememye, most prominently, along with Albert Shibura, André Yanda, and Joseph Rwuri, all with close personal and regional ties to Micombero; while some held ministerial portfolios -- Shibura served both as Minister of Interior and Justice, while Yanda combined the functions of Minister of Information and Executive Secretary of the ruling party -- others, like Rwuri, served as military advisers to the President. The principal architect of mass murder, however, was the Minister of Foreign Affairs, Artémon Simbabaniye, who bears much of the responsibility for the government’s strategic shift from repression to countrywide purges. After his promotion to the rank of itinerant ambassador with plenipotentiary powers, on May 12, Simbananiye had a virtually free hand to organize the killings in the provinces.

More than their official positions, their regional ties served as a powerful source of solidarity. Frequent references by analysts to the “Bururi lobby” underscore the significance of the phenomenon. They all had roots in Bururi province, and many came from the same locality (Matana). More often than not, however, these ties tended to cut across formal institutions, penetrating deeply into the ruling party, the Parti de l’Union et du Progrès National (Uprona) and ancillary organizations. Of these the notoriously militant youth wing, the Jeunesses Révolutionnaires Rwagasore (JRR), was by far the most violent.

Along with the army, the JRR played a central role in perpetrating mass murder -- but not until its Hutu membership, numbering in the hundreds, had been eliminated. From an auxiliary force in support of the army during the clean-up operations in the south, as the repression expanded northwards into the capital city and the provinces, the JRR asserted itself as a largely independent force, rounding up and arresting hundreds and thousands of Hutu. Many of the summary executions conducted in the Bujumbura prison were the work of the JRR units, while dumping of corpses in mass graves was a chore shared by the army and the JRR. According to some sources, additional support in the rounding up and execution of Hutu in the north apparently came from the Tutsi refugee camps, hosts to thousands of Tutsi families expelled from Rwanda during the Hutu revolution (1959-1962). Precisely to what extent is impossible to tell.


C - Victims

If the exact number of deaths is anyone’s guess, there can be no doubt about the murder of the single most prestigious victim: late on the evening of April 29, shortly after his return from six years of exile, ex-King Ntare was assassinated in his royal residence in Gitega. This was not an act of random vengeance. Whether the order to kill Ntare came from Shibura, Simbabaniye or Micombero], all three shared, with other members of the Bururi lobby, a pathological fear of a monarchical plot. And if such indeed was the sub-text of the Hutu insurrection, regicide was the only option.

Even more haunting than the monarchical menace was the nightmare of an impending genocide of Tutsi. Hence the appropriateness of mass murder as a preventive strategy. Extreme threats required extreme solutions. After the “clean-up” of rebel-held positions, Hutu troops, numbering approximately 750, were systematically murdered by their Tutsi commanders. Those few holding officer’s rank, like the highly popular Cabinet Minister Martin Ndayahoze, were the first to be killed. The same fate befell an estimated 300 Hutu gendarmes. Next in line were Hutu students and teachers in secondary and technical schools. By June, approximately 45 per cent of primary school teachers and the near totality of secondary school teachers were reported missing. In a scenario that recur time and again, groups of soldiers and JRR would suddenly show up in classrooms, call Hutu students by name and take them away. Few ever returned. At the Université Officielle in Bujumbura, about one third (120) of the students disappeared in such circumstances. The école normale of Ngagara, a suburb of Bujumbura, lost more than 100 students out of a total of 315. Of the 415 students enrolled at the école technique of Kamenge, 60 are believed to have been killed, while another 100 fled. Out of 700 students enrolled at the athénée (secondary school) of Bujumbura, at least 300 disappeared, some killed and others fleeing to avoid being killed. At the athénée of Gitega, some 40 students were massacred, raising the total of missing students to 148; at the institut technique agricole, also in Gitega, 40 students out of a total of 79 were listed as missing, of whom 26 were said to have been executed. (Lemarchand, 1994, 2002) The list extends to confessional schools, both Catholic and Protestants. In short, not only Hutu elites but potential elites were physically liquidated. Nor was the Church spared. Reporting from Bujumbura in early June, New York Times journalist Marvine Howe noted that “twelve Hutu priests are said to have been killed, and thousands of Protestant pastors, school directors and teachers”. What we are dealing with was not so much a repression as a ghastly slaughter of Hutu populations. The carnage went on for months. By the end of August as many as 200,000 people were reported killed.

This is not meant to obscure the atrocities committed by Hutu insurgents, but to underscore the contrast with the rather more improvised nature of the massacre of Tutsi civilians during the rebellion. Aside from the fact that the number of victims was only a fraction of the Hutu killed at the hands of the army and the JRR, the killings of Tutsi were noticeably more indiscriminate and uncoordinated. In contrast with the carefully targeted nature of the repression and its follow-up, neither gender, age, religion or ethnicity for that matter made a difference in the choice of victims. Admittedly, civil servants and military establishments were a key target, as shown by the systematic massacre of provincial authorities in Bururi and Rumonge. But the overall impression is one of random assassinations of Tutsi civilians, along with those Hutu who refused to heed the call of rebels. While some Tutsi were killed, others were spared (Chrétien et Dupaquier, 90). In some localities victims were picked on the basis of pre-established listings, but not in others. Sometimes victims who could bribe their prospective executioners were spared, while others were killed. In some places, Tutsi houses were torched, and shops looted, but not elsewhere. The “Muleliste” dimension – in the form of slogans such as “Mai Mulele”, “Mai Muhutu” and appeals to magic – was much in evidence in some localities (Kigwena, Mugara, Vyanda), but not in others. For all their brutality and anti-Tutsi thrust, the evidence strongly suggests that the killings of Tutsi civilians were not nearly as well organized, systematically implemented and carefully planned as the annihilation of the Hutu elites.


D - Witnesses

Considerable caution is in order when it comes to evaluating the accuracy of eyewitnesses’ accounts. Ethnic biases are to be expected, as are the failings of memory decades after the act. The richest source of such materials is found in Chrétien and Dupaquier’s extensive interviews with Hutu and Tutsi respondents, but these were conducted some thirty years after the “events”: with the passage of time, questions are bound to arise as to the reliability of individual memories.

Not all Hutu testimonies, however, are necessarily biased or deliberately selective, and the same is true of Tutsi accounts. On a number of points there is indeed a remarkable degree of congruence. This is particularly true of the circumstances surrounding the repression and its aftermath, including the magnitude of the killings, the role of the JRR and the army, the frantic efforts made by some to save their neighbors’ lives irrespective of their ethnic identity. Consider, for example, the following account by a 40 year old Hutu from Nyanza, which brings out the auxiliary role of the JRR along with attempts – mostly in vain -- made by some Tutsi to protect their Hutu neighbors: “[After the repression] interrogations were conducted and every Hutu supposed to have assisted the Mulele was taken away, they were mainly educated Hutu. In these investigations there were Tutsi notables who protected Hutu. At the time, it was the JRR and policemen who rounded up the Hutu and took them to the commune, sometimes fifty at a time.” (Chrétien et Dupaquier, 2007: 140-141) Another Hutu gave this description of the relentlessness and free-for-all character of the repression in the Bukurira region, in the south: “The repression was terrible. The army moved in. Then two camps emerged: Hutu and Tutsi… My father was killed. Anyone could cause someone to be arrested. Even a Tutsi alone, even a little Tutsi kid with a small stick could issue a death sentence against anyone, and the victim had to dig his own grave… To kill and to loot went hand in hand. Corpses were all over… People would say ‘I did nothing wrong’ and hoped to be set free. Since they were innocent, they did not resist. They could have fled, or go into hiding… The repression did not respect the previous co-existence (of Hutu and Tutsi)” (Ibid 145).

No less horrendously revealing of the scale of the killings – but note here the “patriotic” motivations of the killers -- is this narrative by a Tutsi, JRR militant, recorded in 1972: “Every morning we got into a truck, with other JRR. We did it in a really patriotic and nationalist spirit, I can assure you. It was terrible, an incredible carnage. I remember especially in Minago, we wore gloves and breathing pads, but the odor, you know the odor, the road was covered with corpses, which had been decomposing under the sun for a week… In Kabezi and Nyanza-Lac, the only living things we saw were dogs and goats and abandoned cows, but one day we saw a small child, 9 or 10 years old, in a tree. He had been up there for ten days, climbing down from time to time to catch a piece of manioc. When he saw us he said he didn’t care what happened to him, but he wanted his father, his mother, and his three brothers, killed by the rebels, to be buried…” (Ibid. 142-3).

Hutu accounts of the dynamics of the insurrection are relatively scarce, if only because so few survived to tell their story. Nonetheless, what few are available confirm many of the traits observed by other witnesses. The Muleliste component, for example, is clearly brought out in this commentary by a Hutu from Bururi: “[In Rumonge] people wearing palm leaves appeared with machetes and spears. They shouted ‘Mai Mulele!’ and started cutting up people. They just hit their victims with their machetes once and never looked back. They were drugged. They showed up around 6:00 pm, after the reunion. We went into hiding. They attacked everybody, but mainly Tutsi” (Ibid. 99). Although the elimination of the Tutsi emerges as a central theme in the testimonies of Hutu and Tutsi, there is also widespread agreement that a number of Hutu fell under the blows of the rebels. One Hutu trader from Bururi put it this way: “They killed everybody, Hutu and Tutsi, and even women and children” (Ibid. 101). That many Hutu took great risks in protecting Tutsi lives is also confirmed by another Hutu source: “In general, Hutu notables refused to participate in the massacres. If anything they facilitated the flight of Tutsi towards the hills of Mumirwa, where there are many more Tutsi than in the (Imbo) plain”. (Ibid. 101-102)

Some of the more chilling stories are from European witnesses, i.e. aid workers and missionaries. A French aid worker who happened to be in Nyanza-Lac when the rebels took over the town, reports what happened on April 30: “At 6:00 pm the Tutsi, along with the civil servants and the army men were massacred by machete-wielding insurgents. The communal building and the military camp were sacked… Greek and Pakistani traders were witness to the atrocities going on in their shops. Some victims agonized all night long. On May 1, the rebels started looking for those who went in hiding. Twenty six were found and massacred in the afternoon… The ‘rebels’, as they were called, were Hutu, but they were joined by Muleliste mercenaries who probably came from Tanzania, but many were already settled along the banks of Lake Tanganyika… They had their own rules of behavior: they were forbidden to drink alcohol (therefore no beer), to steal (except for the communal treasury and ammunitions from the military camp), and to have contacts with women. These rules were always observed; the traders went on with their business. Before they left (the Mulelistes) attended a display of sorcery, designed to make the combatant invincible” (Ibid. 87-88).

Missionaries were among the very few Europeans who were witness to the subsequent massacre of Hutu populations. Their reports find a recurrent echo in the cables sent out at the time by the Deputy Chief of Mission of the US Embassy, Michael Hoyt, to the State Department. Here are some examples: “No respite, no letup. What apparently is genocide continues. Arrests going on around the clock (May 26)… The liquidation of Hutu goes on apace. Catholic missionaries are increasingly disgusted. Stories which can only be called sickening reach us every day; many Hutu are being buried while still alive. One informant calculates that between 1,400 and 1,500 Hutu males were killed in reprisals in Rutovu alone. The most normal means of execution have been sledgehammers. In the region between Mwaro and Bukirasaz, the army was called in and killed all Hutu males it could find; Tutsi civilians killed the women and children. The death toll is in the thousands” (June 10, 1972)… “In two days following July 14, three new ditches filled with Hutu bodies were spotted near Bujumbura airport” (July 21)… “Repression against Hutu is not simply one of killing. It is also an attempt to remove them from access to employment, property, education, and the general chance to improve themselves” (August 11) (Hoyt, 1972).

Boniface Kiraranganya, a distinguished civil servant of princely origins (ganwa) who witnessed day after day how scores of school children, students, civil servants and professional people were rounded up and sent to their graves, described the carnage as “the paroxysm of madness, the most perfect example of what human beings are capable of doing when freed of all obligations to control their destructive instincts” (Kiraranganya, 1985: 76).


E - Memories

Filtered through the prism of ethnicity, collective memories only bear a distant relationship to the facts. The experience of traumatic events creates its own universe of discourse, where the “other” is seen as belonging to a sub-human category. This is also true of individual memories, and it applies to both Tutsi and Hutu. But given that the loss of Hutu lives was vastly superior to the Tutsi’s, it is easy to see why Hutu memories have been particularly prone to Manichean representations.

The phenomenon is nowhere more vividly captured than in Liisa Malkki’s discussion of “the mythico-history of atrocity”, based on open-ended interviews with Hutu refugees in Tanzania (Malkki, 1995: 91 ff). She shows how memories of atrocity mutate into a sickening, ghoulish depiction of the tortures inflicted by Tutsi: “There was a manner of cutting the stomach (of a pregnant woman). Everything that was found in the interior was lifted out… The cadaver of the mama, the cadaver of the baby of the future, they rotted on the road. Not even a burial. The mother was obliged to eat the finger of her baby… The Tutsi girls were given bamboos. They were made to kill by pushing the bamboo from below [from the vagina] to the mouth” (Ibid. 91). The aim here is the demonization of all Tutsi; in the minds of her interviewees the thread of evil goes way back, from the history of Tutsi penetration into the country to the present time: “The dehumanization of the Tutsi at this level acted as a culmination of earlier assertions in the mythico-history that the Tutsi did not belong to the ‘nation’ in its pure, ‘natural’ state” (Ibid, 93). Stripped of their hype, Malkki’s interviews capture a recurrent theme in the discourse of Hutu extremists: the innate perversity of the Tutsi “race”. While the past is instrumentalized to demonstrate the evil dispositions of the Tutsi “race”, the present is re-interpreted in a way that confirms the fundamental goodness of the Hutu: there is no reference to the challenge posed by the Hutu insurrection, let alone to the acts of courage displayed by Tutsi to protect their Hutu neighbors. Much of this, along with a lengthy reconstruction of Burundi’s pre-colonial past as a story of good guys vs. bad guys, is articulated in what later became the manifesto of Hutu extremists, an undated document by the late Rémi Gahutu titled “Persecution of the Hutu of Burundi” (Gahutu, n.d.). Insofar as they grossly amplify an all too tragic reality, Hutu memories bring into sharp focus the traumatic psychological impact of the experience of mass violence.

For some Hutu analysts the so-called rebellion is yet another irrefutable proof of Tutsi perfidy. Not only did the government fail to take any action to nip the insurrection in the bud – despite its awareness that it was being planned – but in fact did everything it could to facilitate the explosion, the better to justify the ensuing genocidal bloodbath. What happened in last days of April was the tragic outcome of a government-inspired manipulation, aimed at justifying the next step – the mass eradication of Hutu populations. Such, in brief, is the argument set forth by the historian Augustin Nsanze (Nsanze, 2003, 214-218), which to this day forms a key leitmotiv in the imaginings of Hutu extremists.

Where Hutu memories frequently elude references to the Hutu insurrection, Tutsi memories, on the other hand, give pride of place to the mortal threat posed to their community by Hutu rebels. Genocidal intent is the key presumption. As if to strengthen the force of the argument, the number of Tutsi casualties is deliberately inflated, and so, also, the number of rebels, with the figure of 25,000 sometimes cited as fact (Shibura, 1973: 96). To mobilize so many people requires “a lot of time and colossal resources”, writes Albert Shibura, a key participant in the repression (Ibid.) The unstated assumption is that a large number of people killed is irrefutable proof of genocide. Another is that it has happened before. Thus the massacre of hundreds of Tutsi (here again there are no reliable estimates) by the authors of the abortive 1965 coup are described by Shibura as “the genocide of October 19, 1965”, abetted by outside actors, including the CIA (Ibid. 58). Allegations of a Hutu-inspired plot in 1969 are again portrayed as involving “an attempted genocide” of Tutsi (Ibid. 76). Every instance of Hutu violence is thus cited as proof of a genocidal thread running from 1965 to 1972. Furthermore, despite “the barbaric conspiracy of some 25,000 nationals and foreigners”, to quote from the Burundi ambassador to Brussels, Laurent Nzeyimana, “there is no ethnic or tribal problem in Burundi”. His argument is straightforward: if the externally abetted plot has in no way diminished the cohesion of Hutu and Tutsi, this is because inter-ethnic harmony is inherent in Burundi society, and because the majority of the people know that the prophylactic measures taken by the government were only directed at those elements involved in the conspiracy.

With the passage of time, many of these outrageous claims are receding from public and private memories. Furthermore, beginning with the Arusha conference and the opening of the political arena to opposition groups, a genuine transformation has taken place in the country’s political climate. Most importantly, ethnicity has ceased to operate as the central axis of Burundi politics. All of which, along with the promise of a Truth and Reconciliation Commission, raises hopes that accusations based on invective and twisted evidence may in time yield to a more dispassionate exchange.


F - General and Legal Interpretations of the Facts


Divergences in scholarly interpretations of the 1972 tragedy revolve around the concept of genocide: Should the killings of Hutu by the Tutsi army and the JRR be described as genocide? If so, on what grounds can one rule out the use of the term to describe the wanton killings of Tutsi civilians by Hutu insurgents?

The debate about genocide has tended to generate more heat than light, in part because of the absence of a consensus of opinion among scholars as to what constitutes genocide, in part also because the lack of solid empirical evidence about specific aspects of the tragedy. To this must be added the astonishing indifference of the international community about the events of 1972, which is all the more striking when compared to the sustained attention given to the Rwanda genocide. With few exceptions, the tendency among observers to hold up the case of Rwanda as a major reference point has all but excluded the case of Burundi as a legitimate candidate for the analysis of genocide.

Nonetheless, few knowledgeable observers would deny that the genocidal dimension is deeply inscribed in the atrocities committed in 1972. In his monumental history of the Great Lakes region, Jean-Pierre Chrétien refers to the repression as “a veritable genocide of Hutu elites” (Chrétien, 2000: 277), and in a 1974 report to the London-based Minority Rights Group, this writer used the term “selective genocide” (Lemarchand, 1974). This is in contrast with the tendency of Rwanda scholars to consistently refer to the Burundi as an example of “ethnic massacres” -- and to the Hutu rebellion as a “purported [sic] rebellion” (Leonard and Straus, 2003: 73) -- as if to speak of a Burundi genocide might somehow diminish the horrors of its vastly more destructive Rwanda counterpart.

When describing Burundi as a case of genocide, or “partial” or “selective” genocide, scholars generally have in mind the mass killings of Hutu by Tutsi; the massacre of Tutsi by Hutu, on the other hand, is seldom, if ever characterized as genocide. A major exception is the work of Jean-Pierre Chrétien and Jean-Francois Dupaquier, titled Burundi 1972: Au bord des génocides (2007). The substance of their argument is inscribed in the book’s intriguing sub-title: the implication is that we are here dealing not with one but two “near genocides”. The crimes committed by Hutu rebels are presented as a projet génocidaire, aimed at the physical liquidation of the Tutsi community; the main difference with the killings of Hutu, presumably, is that the projet never had a chance to fully materialize. But the element of “brinkmanship” in the sub-title suggests that neither in fact qualifies as a full-fledged genocide.

The Chrétien-Dupaquier argument raises some obvious questions: while uncritically endorsing the official version of the Burundi authorities at the time – the notion that the crimes committed by Hutu rebels were the harbinger of a genocide – the authors are at a loss to offer solid empirical data in support of their thesis and at the same time all too prone to turn a blind eye to counterfactual evidence. As shown above, the targeting included many Hutu as well as Tutsi; the participation of former Congolese ‘rebels’ (simba), or described as such by many observers, suggests the improvised character of the rebellion; so does the widely scattered, piecemeal aspect of the killings, and the absence of a coherent genocidal ideology other than what might be construed from anti-Tutsi slogans. To see in the killings of approximately one thousand Tutsi by scores of rural insurgents manipulated by a handful of radicalized intellectuals a parallel with the planned extermination of anywhere from 200,000 to 300,000 Hutu carries little conviction, no matter how much genocidal intent might be read in the anti-Tutsi sloganeering of the Hutu rebels.

As noted earlier, there is no commonly accepted definition of genocide. Even among scholars who subscribe to the 1948 UN Convention, many would agree that its principal criteria – intent, targeting of racial, ethnic or religious groups, and the destruction “in whole or in part” of such groups – defy precise measurement. Nonetheless, they provide important analytic touchstones for differentiating the character of the Hutu insurgency from the ensuing carnage. The element of intent emerges with tragic clarity from the deliberate, systematic elimination of all Hutu elites and potential elites; ethnic targeting (except for the killing of Ntare) was far more consistent at every stage of the repression than during the rebellion; and while in both cases the targeted group was only “partially” eliminated, the virtual annihilation of Hutu elites is a commentary on the vastly more destructive consequences of the repression.

If only because of its “selective” character – the elimination of an ethnically defined elite group – the case of Burundi does not fit into the Holocaust (or the Rwanda) paradigm. It cannot be described as a total genocide, and for that reason some may quibble about the appropriateness of the genocide label. Jacques Sémelin’s definition – “that particular process of civilian destruction that is directed at the total eradication of a group, the criteria by which it is defined being determined by the perpetrator” (Sémelin 2007, 340) – might conceivably offer conceptual ammunition to those who would challenge the view that anything like a genocide has been committed against Tutsi or Hutu. By the same token, as defined by the perpetrator as the group to be eradicated, there can be little doubt that the extermination of the Hutu elites stands as a tragic illustration of the genocidal urge to “purify and destroy”(Ibid.) Once all is said and done, no amount of retrospective ratiocination about the applicability of the genocide label can ever erase from their collective memories the agonies suffered by Hutu and Tutsi in the time of ikiza.


G - Bibliography

CHRETIEN, Jean-Pierre and DUPAQUIER, Jean-Francois, 2007, Burundi 1972 : Au bord des génocides, Paris: L’Harmattan.

CHRETIEN, Jean-Pierre, 2000, L’Afrique des Grands Lacs : Deux Mille ans d’Histoire Paris: Flammarion.

GAHUTU, Rémi, N.d., Persecution of the Hutu of Burundi, Mimeo: N.p.

HOYT, Michael, 1972, US Embassy Cables from Bujumbura to the State Department, Unpublished materials available from the author’s collection at the University of Florida, Gainesville, Website

KIRARANGANIYA, Boniface, 1985, La Verité sur le Burundi. Sherbrooke (Canada) : Naaman.

LEMARCHAND, René, 2004, “The Burundi Genocide”, Century of Genocide: Critical Essays and Eyewitness Accounts, Samuel Totten, William Parsons and Israel Charny (eds.), New York and London: Routledge, pp. 321-337.

LEMARCHAND, René, 2002, “Le génocide de 1972 au Burundi: Les silences de l’histoire”, Cahiers d’Etudes Africaines, 167 XLII-3, pp. 551-567.

LEMARCHAND, René, 1994, Burundi: Ethnic Conflict and Genocide. Cambridge and New York: Cambridge University Press and Woodrow Wilson Center Press.

LEMARCHAND, René, 1974. Selective Genocide in Burundi, Minority Rights Group, London.

LEONARD, David and STRAUS, Scott, 2003, Africa’s Stalled Development: International Causes and Cures. Boulder and London: Lynne Rienner.

MALKKI, Liisa, 1995, Purity and Exile: Transformations in Historical-National Consciousness Among Hutu Refugees in Tanzania. Chicago: University of Chicago Press.

NSANZE, Augustin, 2003, Le Burundi Contemporain, Paris: L’Harmattan.

SÉMELIN, Jacques, 2007, Purify and Destroy : The Political Uses of Massacre and Genocide, London: Hurst and Company.

SHIBURA, Albert, 1993, Témoignages, Mimeo : Bujumbura.

Citer cet article

Lemarchand René, The Burundi Killings of 1972, Mass Violence & Résistance, [en ligne], publié le : 27 Juin, 2008, accéder le 17/05/2021,, ISSN 1961-9898
Retour en haut de page