Burundi modele:sélective génocide in Burundi

Report N° 20 MINORITY RIGHTS GROUP

Price 45p

___________________________________________________________________________________________________________________

The MINORITY RIGHTS GROUP is an international research and information unit registered in Britain as an educational trust under the Charities Act of 1960. Its principal aims are -

• To secure justice for minority or majority groups suffering discrimination, by investigating their situation and publicising the facts as widely as possible, to educate and alert public opinion throughout the world.

• To help prevent, through publicity about violations of human rights, such problems from developing into dangerous and destructive conflicts which, when polarised, are very difficult to resolve; and

• To foster, by its research findings, international understanding of the factors which create prejudiced treatment and group tensions, thus helping to promote the growth of a world conscience regarding human rights.

The Minority Rights Group urgently needs further funds for its work. Please contribute what you can. MRG is eligible to receive a covenant if you prefer.

The report that follows has been commissioned, and is published, by the Minority Rights Group as a contribution to public understanding of the problem which forms its subject. It does not necessarilyrepresent, in every detail and in all its aspects, the collective view of the Group.

For details of the other reports published by the Minority Rights Group, please see the back cover.

____________________________________________________________________________

___________________________________________________________________________________________________________________

- 2 -

of Human Rights, All human beings are born free and equal in dignity and rights.

adopted by the General Assembly They are endowed with reason and conscience and should act

_________________________________________________________________________________________________________________________________

- 3 -

_________________________________________________________________________________________________________
HISTORICAL NOTE
_________________________________________________________________________________________________________

Rwanda Urundi (1899 1962)______________________________

Burundi and Rwanda became part of German East Africa in 1899, at the zenith of colonial expansion in Africa. In 1916, during World War I, Belgian forces from the former Belgian Congo defeated the Germans and occupied Burundi and Rwanda. In 1923, Burundi and Rwanda became the Belgian mandated territories known as Rwanda-Urundi and were administered as a single unit. In 1946, the territories came under the United Nations Trusteeship Council, with Belgian administration. Limited self government was initiated, which culminated in the attainment of full independence for Urundi in 1962 as the Kingdom of Burundi, under King Mwambutsa IV. Between 1963 and 1964, during a succession of short lived governments, the monetary customs union with 'twin sister' Rwanda was dissolved, and Rwanda gained her independence from Belgium.

Rwanda and Burundi had been closely bound economically and otherwise to the Belgian Congo (now Zaïre), and each was managed and administered during the colonial period by Belgian officials. Events in one country never failed to have serious repercussions in the other. In each country, there has been a long standing history of violent rivalry between the Hutus and the Tutsis.

In Rwanda between 1955 and 1958, Tutsi extremists, viewing Belgian political reforms as a threat, repressed the Hutu movement; in fact, they murdered several Hutu leaders. In 1959, however, the Hutus struck back and in a bloody revolt overthrew the Tutsi minority. Indeed, Tutsis suffered heavy casualties; it is reported that approximately 120,000 fled to Burundi and other neighbouring countries.

In 1960, leaders of the Hutu Emancipation Movement (PARMEHUTU) established a provisional government. In 1961, Belgium recognized the PARMEHUTU regime, but the United Nations, hoping to preserve the ethnic economic union of Burundi and Rwanda after independence, ruled it unlawful and ordered free elections. These elections resulted in an overwhelming PARMEHUTU victory, and in 1962 a United Nations resolution ended the Belgian trusteeship and granted Rwanda full independence.

In Rwanda in 1962, the Hutus expelled the Tutsi minority in a successful coup. In Burundi, however, the dominant Tutsi minority has been able to stay in power in spite of attempted coups by the Hutus by controlling the police, the military, and other vital organizations of the government.

In 1963, there was an abortive Tutsi invasion in Rwanda, which originated from Burundi with the collaboration of some Rwanda Tutsis. The result was disastrous for the Tutsis. In the massacre that followed, as many as 12,000 Tutsis in Rwanda were killed. A renewed and intensified Tutsi exodus from Rwanda began, and the relationship between Burundi and Rwanda deteriorated accordingly.

It should be repeated that one can understand the factors contributing to hostilities within each country and also between the countries when one bears in mind that each country is controlled by the rival tribal ethnic group Rwanda by the Hutus, and Burundi, despite the fact that the population is about 85 per cent Hutu, by the Tutsis.

[ __William J. Butler and George Obiozor, 'The Burundi Affair 1972', IDOC N. America, 1973 ]

___________________________________________________________________________________________

- 4 -

______________________________________________________________________________________________________________________________________

- 5 -

______________________________________________________________________________________________

There are few parallels to the human holocaust that took place in Burundi in 1972 in the wake of a tortuous competitive struggle between the country's two major ethnic groups, the Hutu and the Tutsi . Scarcely noticed (let alone understood) by public opinion anywhere, the killings are conservatively estimated to have caused between 80,000 and 100,000 deaths. Approximately 3.5 per cent of the country's total population (3.5 million) were physically wiped out in a period of a few weeks. In comparative terms this is as if England had suffered a loss of 2 million or the United States about 8 million people. To speak of "selective genocide"*' to describe the outcome of such large‑scale political violence seems scarcely an exaggeration.

What the long‑term consequences will be for Burundi society as a whole is impossible to determine. That the country has undergone something of a metamorphosis as a result of these events is nonetheless undeniable. It has become the only state in independent black Africa to claim the appurtenances of a genuine caste society; a country in which power is the monopoly of a dominant ethnic minority (Tutsi) representing less than 15 per cent of the total population. On the basis of cultural and regional criteria alone, this percentage might drop to less than 4 per cent. Racial differences aside, the nearest parallel to this situation is provided by South Africa, Rhodesia and the Portuguese territories of Angola and Mozambique. The pattern of dominance extends to virtually all sectors of life, restricting access to material wealth, education , status and power to representatives of the dominant minority. For anyone even remotely familiar with the relatively open and flexible system of stratification that once characterized Burundi society the transformation is little short of astonishing.

The full story of what is now piously referred to in Bujumbura as " les évènements" will probably never be known. The chain of events leading to the crisis is as complex as the motives which prompted each community to decimate the other. Sorting out truth from rumour is made more difficult still by the intensity of feelings displayed by participants and observers alike over the atrocities committed by each side, the mixture of fact and fiction conveyed through official statements, and the reluctance of eyewitnesses to report what they saw. Nonetheless, there is enough evidence available to produce a reasonably accurate account of the circumstances that led to the massacre, and in so doing to dispel some of the more prevalent misconceptions about Burundi society and the roots of its recent agony.

______________________________________________________________________________________________

Situated in the Central African rift valley, in the very heart of the continent, Burundi is roughly the size of Belgium (11,000 sq. miles). Along with Rwanda, its neighbour to the north, it has one of Africa's highest population densities (185 per square mile in 1955). The growing pressure of over‑population on the land, together with the general scarcity of natural resources, lie at the root of the country's economic and social problems. What mineral resources exist, aside from small deposits of cassiterite, have yet to be exploited, and much of the economy consists of subsistence agriculture. With the recent discovery of substantial nickel deposits in the southeast the economic picture may change drastically in years ahead; so far, however, no concrete steps have been taken to tap this otherwise promising industrial potential. Coffee is the main cash crop, generating approximately 80 per cent of the country's foreign exchange (the equivalent of about $14 million annually), to which must be added such marginal crops as tea, cotton and rice. Agricultural output is as yet incapable of meeting the demands of Burundi's fast‑growing population, let alone of yielding the surplus production required for rapid economic growth.

Economic scarcity is of course as much of a reality to‑day as it was in precolonial and colonial times, when Burundi was just one of several traditional kingdoms spread throughout the interlacustrine zone.

_____________________________________________________________________________________________

To day, however, perceptions of economic scarcity are increasingly filtered through the prism of regionalism and ethnicity, thus adding a radically different dimension to the political environment. To appreciate the significance of this transformation, at least passing reference must be made to Burundi's traditional system of social stratification, one of the most complex and least understood in the whole of Africa.

The standard image of Burundi society conveyed by much of the colonial literature is that of an ethnic pyramid in which the cattle herding Tutsi, representing 14 per cent of the population, held the commanding heights of power and influence; next in rank came the Hutu agriculturalists, forming the bulk of the population (85 per cent); at the bottom of the heap stood the pygmoid Twa, a group of relatively little significance numerically (1 per cent) and otherwise. Presumably reinforcing this hierarchy of rank and privilege were the physical characteristics commonly attributed to each group: Proverbially tall and wiry, the Tutsi have been said to "possess the same graceful indolence in gait which is peculiar to Oriental people"(2); the Hutu, on the other hand, were seen as "a medium sized type of people, whose ungainly figures betoken hard toil, and who patiently bow themselves in abject bondage to the later arrived yet ruling race. the Tutsi".(3)

However satisfying to most European observers, such simplicities can only convey a highly distorted view of Burundi's traditional social system. Not only do they conceal the existence of major differences within each group, but they also tend to exaggerate the depth of cultural discontinuities among them. These distortions are closely connected. Neglect of intra ethnic cleavages is liable to obscure the basis for cross ethnic links among each group at the same time that it reduces their respective physical and cultural characteristics to a parody of reality.

Attention must be drawn, first, to the existence of two separate categories of Tutsi the 'lower caste' Tutsi Hima group, and the 'higher caste' Tutsi Banyaruguru, literally, "those who came from the north". Note, however that the term ruguru has other connotations, meaning "from above", and hence from regions of high altitude or, figuratively, from high ranking status, i.e. "close to the Court". Outside observers have unduly emphasized the geographical derivation of the term, to the point: *of equating all Banyaruguru with northern Tutsi, which is far from being the case; the Banyaruguru are found in both northern and southern provinces, and this is also true of the Hima. At the time of writing (1974) the present Governor of the Ruyigi province is a de frocked Anglican deacon named John Wilson Makokwe, a Hima from Buhiga, a northern locality. To assume that the Hima are inevitably from the south and the Banyaruguru from the north, as many observers have been prone to do, would be a gross exaggeration. The former are said to have migrated into the country from the eastern borderlands in the 17th or 18th century, about two or three centuries later than the Banyaruguru, who generally hold them in deep contempt, supposedly because of their 'upstart' attitudes and innate resourcefulness. Nevertheless it is the 'lower caste' Tutsi Hima from the south who are politically dominant, "The Himas" writes Father Rodegem, "seem gifted for leadership and direct action",(4) a statement wholly consonant with the emergent pattern of leadership in contemporary Burundi: a substantial number of civilian and military elites are recruited from the Hima stratum, and the President of the Republic (Michel Micombero) is himself of Hima origins. The Banyaruguru, by contrast, though represented in the government are virtually powerless.

Cutting across this and other cleavages are different social rankings attached to the various patrilineages (imiryango) within each group, Tutsi, Hutu and Twa. The usual distinctions are between the 'very good' families, the 'good' families, those that are neither good nor bad and bad. No less than forty three different patrilineages thus enter into the Tutsi Banyaruguru segment, each in turn falling into a specific ranking of social prestige. In this fashion lineage affiliations could substantially rectify the formal rank ordering established through ethnic divisions. The degrees of social distance within the Tutsi stratum, for example, were at times far more perceptible and socially significant than ethnic differences between Hutu and Tutsi. This multiplicity of reference group identifications within the same broad ethnic stratum has created the basis for potential conflicts among clans, families and lineages; yet the sheer fluidity of such identifications is also the source of considerable ambiguity as to how one ought to be defined in terms of clan or family affiliations. This very ambiguity in turn may help to mitigate intra group conflict. A case in point is the so called Basapfu 'clan'. This is how Father Rodegem explains the origins of the Basapfu:

__________________________________________________________________________________________

Tutsi of high ranking status. They initially came from Hima Clan. But for some reason tradition has failed to ascertain, the King one day decided that they should all be exterminated. He entrusted this task to the Abongera clan, who made a clean sweep of the Abasapfu cattle, plundered their crops, set fire to their kraals and killed whoever stood in their way. One of the survivors was a small boy, who had found refuge behind a reed screen (sapfu) After the raiders had left, he was discovered by some passers by who decided to take him to King Ntare. The latter kept him at his court under his protection and called him Musapfu to commemorate his adventure.(5)

Whether the Basapfu are actually of Hima origin is open to doubt. The historical evidence suggests that they may have been of Banyaruguru origin. The significant point is that today the Basapfu identify themselves, and are often identified by others, as being neither Hima nor Banyaruguru. They are just referred to as Basapfu, as if they formed yet another reference group within the Tutsi stratum. This, and the fact that they are more or less evenly spread throughout the country, is what later enabled some of their representatives to act as the arbiters of regional conflict, and indeed of Hima Banyaruguru conflict. For if the incumbent elites are largely drawn from the Bururi based Hima led faction, within this faction some Basapfu hold key positions within the government and the army.(6)

A final point to note is that neither Hutu nor Tutsi hold traditional claims to authority. The real holders of power in the traditional society were the princes of the blood, or ganwa. Because of the special eminence conferred upon them by the accidents of history, they became identified as a separate ethnic group, whose power and prestige ranked far above that of ordinary Tutsi. They formed the core of the political elites and as such held most of the chiefly positions available under the monarchy. Despite or because of this, they never stood as a very cohesive group. Intra ganwa rivalries are indeed a recurrent theme of Burundi's pre colonial history. Out of the competing claims of rival dynasties bitter feuds periodically broke out among the representatives of different 'Houses', culminating in the middle of the 19 th century in a major struggle between the sons of Mwami (King) Mwezi Kisabo (1852 1908) and the descendants of the previous incumbent, Mwami Ntare Rugaamba (1795 1852). Temporarily held in check but by no means dissipated by the spread of the colonial pax, the late fifties saw a sudden resurgence of these antagonisms. Even at this late date political conflict did not express itself in ethnic terms, but in the form of factionalism between representatives of opposed unilineal descent groups.

What gave a measure of unity and cohesiveness to this otherwise highly fragmented social order is that below the ganwa stratum no single line of cleavage could be said to govern the allocation of social status, wealth or power. Ethnic divisions were largely irrelevant to the distribution of social prestige, and of only marginal significance with regard to wealth. And although power was in theory the monopoly of the princes, the record shows that subchiefs and palace officials were sometimes recruited from among Hutu and Tutsi. What is more, the competitive relationships which developed among the princes made it imperative for them to seek the support of both Hutu and Tutsi hence substantiating Simmel's observation that "conflict may also bring persons and groups together which otherwise have nothing to do with each other". In this case, however, Hutu and Tutsi were not nearly as compartimentalized as the foregoing might suggest. Through the institution of clientship (bugabire) Hutu and Tutsi were caught in a web of interlocking relationships extending from the very top of the social pyramid to its lowest echelons, with the Mwami acting as the supreme Patron which in turn underscores the unifying role of the monarchy, both as a symbol and an institution. Through the use of specific symbols, ceremonies and rituals the monarchy imposed itself as a major focus for popular loyalties. No other source of legitimacy was as compelling as the Royal Drum (Karyenda) in holding society together.

The point of this discussion is that although the traditional society contained a great many potential sources of conflict, in practice conflict was seldom if ever activated along ethnic lines. To view the recent holocaust as "an extreme case of the old African problem of tribalism”8 is indeed difficult to square with the realities of traditional Burundi society. If the term "tribalism" has any meaning in this context it is a very recent phenomenon, traceable to the social transformations introduced under the aegis of the colonial state and the consequent disintegration of those very structures and mechanisms that once gave cohesiveness to society as a whole.

______________________________________________________________________________________

Dimensions of Conflict
_______________________________________________________________________________________________________

In its most acute and devastating manifestations the Hutu Tutsi conflict was the last in a series that spread over a period of at least twelve years, covering almost as wide a range of potential oppositions as the traditional society had to offer. Grafted onto this were the tensions arising from the introduction of new forms of political self expression, i.e., parties, trade unions and parliamentary institutions. Out of this combi nation of traditional and modern types of opposition developed an extraordinary hybrid and complex polity.

The introduction of the vote in 1956, six years before independence, initiated a process of political mobilization which gradually reached every sector of society, activating one group after another, pitting princes against princes, monarchists against republicans, army men against civilians, north against south,Hutu against Tutsi . At first, traditional cleavages tended to act as so many breakwaters, allowing the political mobilization of one group at a time. In contrast to what happened in Rwanda, where the mobilization of the Hutu masses was greatly facilitated and accelerated by the existence of a sharp, vertical split between the Tutsi aristocracy and the Hutu masses, in Burundi the mobilization of the population along ethnic lines was significantly delayed by the complexity of the traditional social system, and by the fact that the monarchy was relatively free from ethnic bias. Even when ethnic loyalties were stirred into action, this did not eliminate the play of narrower loyalties. One of the most
striking aspects of the country's recent political evolution is the extent to which ethnic self perceptions have tended to coexist with, and at times to become subordinate to, residual attachments to the region or to the clan. As environmental threats shifted from the ethnic to the regional or clanic level, corres¬ponding shifts of identification occurred among political actors.

This said, it is only fair to recognize that the seeds of ethnic conflict were planted long before the occurrence of violence. Tempting though it may be to emphasize the traditional dimensions of the recent slaughter, the evidence on this score is very scanty. Meyer's statement that "as long as the Batussi [sic] are masters in the country, spiritual and cultural progress is impossible for the Barundi people, for it is only the present low position of the Bahutu, kept in seclusion for centuries, that ensures the Batutsi their dominance"(9) does not seem too convincing as an argument, confusing as it does political and social (or economic) dominance while failing to distinguish between a potential basis for conflict and conflict itself. As we already stressed, although the traditional society offered a potential basis for ethnic conflict, it never experienced such conflict on a scale even remotely approaching what happened after independence.

Of far greater relevance is the process of social transformation set in motion during and after the colonial interlude. The external dimensions of this phenomenon are especially important to bear in mind, in at least two senses. The Rwanda revolution, for one thing, had a decisive psychological impact on ethnic self perceptions in Burundi. The coming to power of Hutu politicians in Rwanda led many of their kinsmen in Burundi to share their political objectives, in turn intensifying fears of ethnic domination among the Tutsi of Burundi. Thus by giving the Burundi situation a false definition to begin with, a definition patterned on the Rwanda situation, Hutu politicians evoked a new behaviour both among themselves and the Tutsi which made their originally false imputations true. Ethnic conflict thus took on the quality of a "self fulfilling prophecy".(10)

The next point is in the nature of a qualification to the foregoing: in some respects the Burundi situation had already been defined by the Belgian colonizer as one approximating to Rwanda, with the result that something of a caste structure had already begun to emerge during the colonial period. Long before aspiring Hutu politicians sought to emulate the goals and strategies of their ethnic brothers across the border, Belgian policies in Rwanda served as a model for colonial administrators in Burundi. It was both simpler and more efficient to view Burundi as consisting of a Tutsi aristocracy and a Hutu peasantry and pursue a policy of indirect rule that would maintain the dominance of one over the other. Few efforts were made during the colonial period to extend educational facilities to the Hutu masses, or for that matter to provide them with what few opportunities were available for a political apprenticeship. Student enrolment at the Ecole des Frères de la Charité (better known as the "Groupé Scolaire " of Astrida ) between 1946 and 1954 shows a clear predominance of Tutsi over Hutu a disproportion which becomes even more striking of course in the case of Rwanda (see table over).

__________________________________________________________________________________________

                                                                                    Table 1

                      Ethnic Distribution of Student Enrolment at the Groupe Scolaire (1946-54)
-------------------------------------------------------------------------------------------------------------------------------------------------------
Year*                            Tutsi                                     Hutu
                                                                       Rwanda                         Burundi                       Congolese
-------------------------------------------------------------------------------------------------------------------------------------------------------
1946                             44                            1                                          8
1947                             42                            2                                        10
1948                             85                            2                                        11                                 2
1949                             85                            5                                          9
1953                             68                            3                                        16
1954                             63                            3                                        16                                  3
----------------------------------------------------------------------------------------------------------------------------------------------------------
*Enrolment figures for 1950 52 unavailable; the data above is drawn from the enrolment records of the Groupe Scolaire at Astrida (now Butare).
----------------------------------------------------------------------------------------------------------------------------------------------------------
Likewise, the conciliar organs set up in 1962 at the countrywide and district (or territoire) levels, known respectively as the Conseil Supérieur du Pays (CSP) and the Conseils de Territoire (CT), were largely dominated by Tutsi or ganwa elements. A study published in 1959 gives the following ethnic breakdown for each set of institutions:

                                                                                        Table 2

                            Ethnic Distribution of Seats in CSP and CT: Burundi and Rwanda (1959)
---------------------------------------------------------------------------------------------------------------------------------------------------------
Representative Institution*            Ethnic Distribution
------------------------------------------------------------------------------------------------------------------------------------------------------
                                                              Tutsi (or ganwa)         Hutu                         Total
CSP
          Rwanda                                        31                                    2                               32
          Burundi                                        30                                    3                               33

CT
         Rwanda                                        125                                 30                            155
         Burundi                                        112                                 26                            138
----------------------------------------------------------------------------------------------------------------------------------------------------
*CSP: Conseil Supérieur du Pays; CT: Conseils de Territoire. Source: Aloys Munyangaju, L'Actualité Politique du Ruanda-Urundi (Bruxelles: 1959), p.20.
---------------------------------------------------------------------------------------------------------------------------------------------------------

The result is that on the eve of independence relatively few Hutu could claim the status of a modern élite, and those who did were all the more anxious to translate their egalitarian commitments into reality. Yet precisely because of the nature of their commitments, their access to positions of authority could only be viewed with the greatest suspicion by the Tutsi minority. Extension of the vote, on a per capita basis, evoked similar apprehensions. Just as social equality spelled the end of Tutsi supremacy, majority rule for many Tutsi was seen as synonymous with Hutu rule.

Even in its most restrictive sense (implying equal representation of ethnic interests in key governmental and bureaucratic posts) equality never became a reality of post independence politics. A mere glance at the ethnic distribution of top civil service positions in 1965 shows the extent of Tutsi predominance in the political system (see Table 3 following).

_________________________________________________________________________________________

- 10 -

                                                                                       Table 3

                             Ethnic Distribution of Top Ranking Civil Service Positions, July 1965

Ministries and Directions*        Ethnic Distribution          Ministries and Directions     Ethnic Distribution

                                              Ganwa   Tutsi   Hutu     Other                                                  Ganwa Tutsi Hutu Other

Prime Ministership                                                                   Justice (Secretariat)
Director General (DG)                                   1                         DG                                                                  1
Directors (D)                                        2                                     D                                                            3
Deputy Directors (DD)                      1          1                         DD                                                         4       1
Total                                                      3          2                         Total                                                      7       2

Finance                                                                                        Information
DG                                                          2                                     DG                                                                    1
D                                                             1          2                         DD                                                                    1
DD                                                                      3
Total                                                      3          5                          Total                                                                2

Economic Affairs                                                                       Social Affairs
DG                                                          1                                      DG                                                                    2
D                                                             2          1                          D                                                               2      2
DD                                                          1                                      DD                                                             2     1
Total                                                      4           1                         Total                                                          4     5

Agriculture                                                                                   Foreign Affairs
DG                                                          3                                      DG                                                              1
D                                                             4           1                         D                                                                 3     1      1
DD                                                          2           1                         DD                                                              2     1
Total                                                       9           2         &n