Racketeering Militaries and the Promise of Stability:
Justifying Dictatorship in Rwanda and Burundi
Timothy Longman
Vassar College
WORKING DRAFT NOT FOR CITATION
The threat of ethnic conflict and other forms of public disorder remains among the
most persistently evoked arguments against multiparty democratic government in Africa.
As many scholars have noted, the arbitrary drawing of colonial borders left nearly
all African states with a multiplicity of linguistic, religious, and ethnic groups,
and political mobilization has frequently occurred along lines of identity, often
aggravating tensions between identity groups. In the 1960s and 1970s, many African
heads of state justified moves to centralize their power and limit civil liberties as necessary
to maintain national unity and avoid divisive and bloody conflicts like the Congo
Crisis and the Biafra War.(1) Military officers, regardless of their actual reasons for staging coups, commonly
justified their seizures of power as necessary to unify their countries, and in many
cases populations welcomed military rulers because they believed in the ability of
the military to restore order.(2)
In recent years, dictatorship and single-party government have fallen into disfavor
in Africa. Three decades of authoritarian rule failed to bring either economic growth
or stability to Africa, so in the 1990s popular movements have emerged throughout
most of the continent to demand political liberalization. Nearly every African state
has instituted institutional and political reforms since 1990. At the same time,
many authoritarian leaders have continued to attack democracy as dangerous and divisive
and, even after acceding to demands for democratic reform, they have attempted to maintain
their own domestic and international support by characterizing their opponents as
champions of parochial interests and instigators of disorder.(3) Many scholars and international policy-makers have been reluctant to endorse multiparty
democracy in Africa largely because they fear that democracy can potentially ignite
ethnic conflict and other social disturbances. The emphasis on "governance," for
example, is explicitly offered as an alternative to discussing "democracy," which,
it is implied, may not be appropriate for the African context.(4) The fear of descent into chaos haunts much of the current academic discussion of
political reform in Africa.
It is my contention in this paper that opponents of democratic government have actively
exploited the fear that democracy encourages division and disorder in order to justify
a retention of or return to authoritarian rule. Charles Tilly has argued persuasively that states have historically used the threat of war to justify their centralization
of power, even though it is states themselves that create the threat of war. According
to Tilly, states have built their power through a system of racketeering, creating the very threat from which they promise to protect their populations.(5) Military officers and entrenched authoritarian rulers have used a similar racketeering
tactic within their territories to undermine democratic reform and justify authoritarian
rule. In Rwanda, Burundi, Kenya, South Africa, Ivory Coast, and other African states, opponents of democracy have (generally covertly) incited ethnic violence and
encouraged an expansion of criminal activity, then blamed the growing public disorder
on democracy and offered themselves as the only forces capable of restoring order.
Since those promising to bring order were themselves the primary instigators of disorder,
they did in fact possess the ability to follow through on their promises and bring
the disorder to an end. In these cases, the primary threat to public order was not
democracy itself, but rather the exploitation of the belief in democracy's potential
for disorder. By buying into the myth that democracy creates division and authoritarian
rule brings order, scholars, international diplomats, and local politicians alike
are enabling the enemies of democratic government to undermine democracy and serve
their own political interests by encouraging disorder.
In this paper, I analyze the cases of Rwanda and Burundi to demonstrate the use of
racketeering by supporters of authoritarian rule in order to undermine democratic
reforms. In both cases, opponents of democratic government secretly supported the
formation of civilian ethnic militia and employed the militia to heighten ethnic conflict.
In both cases as well, officials used the police selectively, encouraging a growth
in robbery and other crimes to contribute to a popular sense of declining order.
In Rwanda, a core group of northern Hutu were able to refortify their hold on power by discrediting
multiparty reforms and dividing and disempowering opposition parties by blaming them
for the growing violence in society. In Burundi, by encouraging inter-ethnic gang violence in Bujumbura and elsewhere, military officers who opposed rule by the
Hutu majority justified the return to power of former President Pierre Buyoya as
the only means of restoring order. Sadly, many people, both within the countries
and in the international community, accepted the proposed interpretation of events, and in
both cases the result has been the loss of possibilities for democracy and an unfortunate
loss of lives under the repressive power of authoritarian states.
Rwanda: Genocide in the Name of Order
In the first decade of Rwanda's independence, ethnic violence between the majority
Hutu population and the formerly dominant minority Tutsi group regularly disturbed
public order. In 1973, after a wave of attacks on Tutsi throughout the country,
the leader of the armed forces, Major-General Juvénal Habyarimana, took power in a coup d'état,
pledging, among other things, to bring an end to ethnic strife. For more than a
decade, Habyarimana effectively maintained order in Rwanda. He appeased Hutu ethno-nationalists by maintaining quotas on Tutsi in education and employment and expanded
the economy at a respectable rate by attracting substantial international development
assistance.
By the late 1980s, however, public discontent with the regime had grown. Power had
become increasingly concentrated in the hands of a small group of trusted associates
from Habyarimana's home region in the northwest, many of them relatives of Habyarimana's wife, a descendent of an historically important Hutu family. This group, known
as the presidential akazu
, or "little house," dominated positions in the military and government, as well as
important positions in business and religious institutions, and concentrated an increasing
portion of the nation's wealth in its hands. Newspapers began to carry stories exposing official corruption, and in 1990 those groups excluded from power --primarily
Hutu from the south and center and Tutsi from throughout Rwanda --began to organize
to demand liberalization of the political system. In mid-1990, responding to growing
public protest, Habyarimana announced his intention to implement reforms, and a year
later he accepted a new multi-party constitution. Over the next few months, opponents
of the president organized several new parties, which formed a coalition and pressured Habyarimana into naming a multi-party government of transition under an opposition
prime minister.(6)
When the opposition political parties entered into the government in 1992, the akazu
and its supporters initiated a campaign to discredit both them and the entire process
of multipartism
by encouraging an expansion in disorder. The Interahamwe, the youth wing of the
president's party, became increasingly radicalized and violent. They staged protests
that periodically shut down the capital, violently broke up opposition party rallies,
and harassed members of opposition parties and others they perceived as enemies of the
regime. Both the Interahamwe and military personnel became increasingly involved
in violent crime, empowered by the failure of the police to respond to commit increasingly bold acts of banditry such as armed robbery and carjacking. Habyarimana and other
leaders of his party, though they actually maintained most political power themselves,
including control over the police and armed forces, blamed the opposition for the
decline in public order.(7)
While the general public did not necessarily believe that the opposition was in fact
responsible for the growing insecurity, interviews I conducted in Rwanda during this
period revealed that the conflicts between the political parties and the expansion
of violent crime did make people disenchanted with multiparty democracy. What one unemployed
carpenter living near the southern town of Butare said in response to a question
about political reforms was typical:
The people involved in politics have profited from these changes, but the peasants
suffer a great deal from these changes. Brothers no longer get along, because they
are not in the same party. When there are victims from the demonstrations, they
are always peasants.(8)
An elderly woman in a rural area in the center of the country, where Habyarimana's
party was believed to have little support, reported,
I want Habyarimana to continue to govern, even though many people don't share his
political views. Multiparty government should never have been implemented, because
people want to kill their neighbors, they want to burn the homes of their companions.(9)
In addition to the expansion in crime and the increasingly violent clashes between
parties, Habyarimana used heightened ethnic tensions very effectively to increase
public insecurity and discredit the opposition. In October 1990, just as pressures
for democratic reform were reaching a peak, a group composed primarily of Tutsi refugees
based in Uganda attacked northern Rwanda. The Rwandan Patriotic Front (RPF) claimed
to want a democratic transition and demanded the right to return to Rwanda for Tutsi
refugees, most of whom had fled the country during the violence of the 1960s. Although
the Rwandan army easily repulsed the initial RPF invasion, Habyarimana's supporters
effectively used the RPF threat for propaganda purposes, warning the population that
the Tutsi wanted to reestablish the dominant and exploitative position they had enjoyed
under Belgian colonial rule and presenting themselves as the defenders of Hutu interests.
As the RPF made increasingly successful incursions into the country, displacing tens of thousands of people from areas along the Uganda border, Habyarimana's supporters
sought to portray their opponents as Tutsi sympathizers and traitors. Beginning
in early 1991, Habyarimana's inner circle secretly gave approval for local government officials to organize ethnic massacres in several areas. These massacres, which
they portrayed as spontaneous public uprisings, served to heighten ethnic tensions
in the country.(10)
The war with the RPF allowed Habyarimana to appeal not only for domestic support but
also for assistance from the international community. France in particular provided
the government with extensive military aid, making possible not only a substantial
arms build up but an expansion of the armed forces from fewer than 6,000 to more than
35,000 within a period of only three years. Soldiers were stationed not only in the
northern prefectures, location of the RPF advance, but throughout the country, where
they harassed and intimidated the population. To justify the nation-wide presence of
the armed forces, supporters of the president staged guerilla attacks within the
country, which they attributed to the RPF. On the night of October 4, 1990, just
days after the initial RPF assault, extensive shooting was heard in Kigali which the government
blamed on the RPF; evidence has since demonstrated fairly conclusively that government
troops themselves staged the "attack" to convince Belgium, France, and Zaire of the urgency of sending troops to defend the capital. Around the time the multiparty
government took power in 1992, a series of bomb attacks began that over the next
year targeted buses and public places, such as markets and dance halls. The military
consistently blamed these bombs on the RPF, but opposition politicians claimed that they
were in fact planted by Habyarimana supporters seeking to destabilize the multi-party
government.(11)
The various divide-and-rule tactics employed by the Habyarimana regime eventually
succeeded in dividing and coopting portions of the opposition parties, intimidating
most critics of the regime into silence, and quelling public support for democracy.
Members of the akazu
organized a secret military force, dubbed the Zero Network, that carried out assassinations
of several opposition politicians. A new extreme Hutu nationalist party, the Coalition
for the Defense of the Republic (CDR), intensified the level of anti-Tutsi and anti-opposition rhetoric. By late 1992 members of the CDR and the Interahamwe
began to organize as militia, receiving military training from the armed forces.
With the war escalating and both ethnic tensions and general insecurity increasing,
Habyarimana's supporters presented themselves as the only force that could bring peace to
the country while defending the interests of the majority. This logic was carried
to its extreme in the 1994 genocide.
Following a successful offensive, in which the RPF occupied a large section of northern
Rwanda, and under intense international pressure, Habyarimana signed a peace accord
with the RPF in August 1993, that provided for the integration of the RPF into the
army and a new government of transition that would include the RPF. Allies of the
president were scandalized by the accord. Members of the akazu
established a new radio station that stirred up public fears of Tutsi domination
and broadcast regular denunciations of "enemies of the state," including both Tutsi
and moderate Hutu. The military stepped up its training of civilian militia and
began to provide them with arms. When President Habyarimana was killed in an April 1994 plane
crash, a plan was already in place to eliminate opposition to the regime through
a massive campaign of violence. Within hours after the crash, the presidential guard
and other elite troops spread out into the capital with lists of prominent Tutsi and
moderate Hutu to kill, and in subsequent weeks, the new Hutu-nationalist government
carried the violence into the rest of the country. Between April and July, when
the RPF seized control of the country, soldiers and civilian militia killed an estimated 800,000
people, the vast majority of them Tutsi. To this day, many of those who organized
and carried out the genocide continue to deny that it was a genocide, claiming that
it was a defensive action carried out in the name of public order.
Burundi: From Dictatorship to Democracy and Back
The case of Burundi shows even more clearly the ability of the military leaders to
exploit the perception that the authoritarian rule brings order. Burundi, like Rwanda,
has a majority Hutu population and a Tutsi minority that monopolized power in the
colonial period, but in contrast to Rwanda, the Tutsi retained power after independence.
In 1966, after the monarchy made overtures to the Hutu majority, the largely Tutsi
military took power to guarantee continued Tutsi hegemony. For two decades the military used extensive violence and repression to quell Hutu political aspirations, including
the systematic slaughter of some 200,000 Hutu in 1972, mostly intellectuals and community
leaders. In the early 1990s, however, President Pierre Buyoya changed strategies, seeking to appease the Hutu population by diminishing official repression and
bringing Hutu into his government. Seeing the pressures of democratization sweeping
across Africa and apparently believing that he could win an election, Buyoya organized
multi-party elections in June 1993. When he lost to a Hutu candidate, Melchior Ndadaye,
Buyoya accepted the results and turned over power.(12)
Although President Ndadaye established a multi-party, multi-ethnic government and
named a Tutsi prime minister, many Tutsi opposed the transfer of power, fearing that
it was a first step toward their ultimate annihilation. The armed forces remained
almost entirely Tutsi and became a haven for Tutsi supremacists. In October 1993, less
than four months after Ndadaye took power, a group of military officers attempted
to stage a coup. Although they failed to seize power, they assassinated President
Ndadaye and several other government leaders. Hutu in various parts of the country, in many
cases led by local government officials, took revenge against Tutsi civilians, and
the army responded by slaughtering thousands of Hutu civilians, without regard for
their involvement in the anti-Tutsi violence. An estimated 30,000-100,000 people died
in the violence, roughly equal numbers of Hutu and Tutsi.(13)
After several weeks of political uncertainty in the capital, during which various
groups contended for supremacy, the surviving officials of the Ndadaye regime were
able to reconstitute a new government. They named another moderate Hutu president,
but only a few months later, he too was killed, in the same Kigali plane crash that killed
President Habyarimana. Once again Burundi's political system was thrown into turmoil.
Given the terrible bloodletting taking place in Rwanda, this time Tutsi supremacist factions in the parliament and military refused to accept another Hutu president
without substantial concessions to protect their power, including naming a prime
minister from their own ranks and conferring expanded powers on the military. After
several months of negotiation, a Convention of Government was reached with a Hutu president,
Sylvestre Ntibantunganya, but a powerful new National Security Council severely limited
his authority, while Tutsi supremacist parties gained a de facto veto over government decisions.(14)
Over the next two years, security conditions throughout Burundi gradually deteriorated.
A Hutu rebel army, the Forces for the Defense of Democracy (FDD), formed by leaders
of Ndadaye's party who fled Burundi after his assassination, refused to accept the new government, claiming that it lacked real power and represented a capitulation
to Tutsi extremists. They expanded guerilla activities into many parts of the country,
attacking primarily military targets. In a what became a generalized pattern, the
armed forces would respond to the FDD raids, in which FDD forces would typically attack
in guerilla style, looting a military outpost then fleeing into the hills, by taking
revenge on Hutu civilians. In nearly all cases, the armed forces would then report
that the FDD itself had killed the civilians, but in research I conducted in Burundi
in 1997, I found that in nearly all the cases I investigated, civilians were killed
by the military after the FDD had fled the scene. The military also began to carry
out summary executions and other targeted attacks on Hutu they viewed as threats to security,
businesspeople, the educated, party activists. Typically this violence, too, was
attributed to the FDD and its supporters, but again my research found these claims
to be ungrounded.(15)
The violence that proved of primary concern for the international community, which
was concentrated overwhelmingly in the capital, involved attacks between youth gangs
in Bujumbura and other cities. While historically Burundi's two main ethnic groups
had lived together in multi-ethnic communities, the violence following Ndadaye's assassination
started a process of ethnic segregation. Tutsi fled from the countryside into refugee
camps, where they could be protected by the armed forces, and into cities. Beginning in 1994, Tutsi youth began to form gangs, generally known as the Sans Echecs
or the Sans Défaite
, that attacked Hutu civilians, seeking to drive them out of multi-ethnic neighborhoods.
Hutu youth formed gangs as well, and for a number of months in 1995, gang warfare
shook the capital, but the Tutsi gangs, since they received militia training and
arms from the armed forces, were easily able to dominate. Furthermore, the military sought
very selectively to enforce order, pursuing Hutu gang members while ignoring offenses
by the Tutsi gangs. By the end of 1995, nearly all Hutu residents had been driven out of Bujumbura, either into the countryside to live with relatives or into refugee
camps around the edges of the city. The gangs, however, continued to disturb public
order by engaging in armed robbery and other violent crimes.
Turmoil prevailed in the political realm as well, as Ntibantunganya's limited power
prevented him from effectively responding to the growing crisis in the country.
The Tutsi supremacist factions placed the president in a Catch-22 situation, allowing
him little power, inciting public disorder, then condemning him for failing to control
it. The hardline Tutsi parties organized public protests and forced the resignation
of both the speaker of the parliament and the prime minister, whom insisted by replaced
by one of their own. A number of Hutu and moderate Tutsi political officials were
assassinated, probably by members of the armed forces, including five governors,
each of whom was replaced by a military officer. The killings contributed to a sense
of growing chaos, which Tutsi supremacist politicians used as grounds to call for military
intervention. Eventually in July 1996, fearing a coup, Ntibantunganya and a number
of his ministers sought refuge in international embassies, and a few days later Pierre
Buyoya official returned to power, declaring himself president and promising to restore
order.
Buyoya's return to power is a clear case of the principle of racketeering at work.
By supporting gang and militia activity and by engaging in assassinations, the military
contributed to a sense of declining order which they claimed only they had the power to stop. Many people in the international community, while publicly condemning
the coup, privately expressed relief. Several expatriates working in diplomatic missions
and the United Nations system privately expressed to me strong support for Buyoya,
claiming that he had restored order.
While it may be true that Buyoya has brought order, particularly to the capital, it
is not difficult to understand how he has done so. Now occupying political offices
themselves, the armed forces no longer had an interest in carrying out assassinations.
After taking power, the armed forces drafted members of the youth gangs into the
army, effectively removing them as a threat to public order, and they enforced the
law much more rigorously. As the ex-foreign minister under Ntibantunganya told me,
"The army is charged with protecting the institutions and the population. How is it that
they were incapable of doing so under Ntibantunganya but are capable under Buyoya?"(16) As the foreign minister implies, the armed forces were not incapable of enforcing
order under a civilian government, but they chose not to do so, because the disorder
served their political purposes.
Democracy and Public Order
Anti-democratic forces in Rwanda and Burundi are not unique in creating or encouraging
disorder in an effort to undermine popular and international support for democracy.
Many observers of Kenyan politics have accused President Daniel arap Moi of involvement in violence in the Rift Valley between ethnic groups that have supported his
party, such as the Masai, and those that have opposed him, such as the Kikuyu. The
role of the white South African police and defense forces in fostering black-on-black
violence to attempt to discredit moves toward majority democratic government has now been
well documented. Examples of such calculated exploitation of chaos can be found
outside of Africa as well, as in Chile, where the armed forces and other conservative
elements played a major role in creating instability during the regime of Salvador Allende,
then claimed instability warranted the 1973 coup.
That the "forces of order" should be involved in creating disorder is not, of course,
an earthshattering observation. The policy of divide-and-rule is one of the oldest
political strategies on the books. Nevertheless, dictators continue to benefit greatly from the myth that they bring order while democratic governments breed conflict.
While I would challenge the very nature of the "order" brought by military rulers
-- for example, though the Burundi military's killing thousands of unarmed Hutu civilians and rounding up more than 300,000 Hutu into regroupment camps may have limited activity
by rebel forces, it hardly represents "order" for those who are living as virtual
prisoners under constant fear of arbitrary arrest, torture, or execution -- my point
here is that conservative militaries, embattled dictators, and other authoritarian
forces effectively use the belief in their ability to bring order to undermine democracy.
They encourage disorder such as crime and ethnic violence so that they can discredit democratic reforms and regimes, then they step in and gain popular support by creating
order, which they are able to do by simply putting an end to their own support for
disorder.
I am not so foolish as to propose that democracy is necessarily more orderly than
authoritarianism -- there are too many cases of democratic competition degenerating
into conflict for that to be true. Nevertheless, I do strenuously object to the
necessary association of dictatorship with order. To assume that countries must choose between
the disorder of democracy on the one hand and the order, but repression, of dictatorship
on the other is to play into the hands of those who wish to restrict civil liberties and promote their personal political power. For those concerned about order,
the choice is not between democracy and dictatorship but between responsible leaders
who seek to serve the public will and irresponsible ones who seek to exploit social
divisions for their own political gain, and there are examples of both democrats and
dictators in each category.
Endnotes
(1)David E. Apter and Carl G. Rosberg, "Changing African Perspectives," in Apter and
Rosberg, eds., Political Development and the New Realism in Sub-Saharan Africa
(Charlottesville: University of Virginia Press, 1994), pp. 24-28; Donald Rothchild
and Michael W. Foley, "African States and the Politics of Inclusive Coalitions,"
in Donald Rothchild and Naomi Chazan, eds., The Precarious Balance: State and Society in Africa
(Boulder: Westview, 1985), pp. 233-264.
(2)J. Gus Liebenow, "The Military Factor in African Politics: A Twenty-Five-Year Perspective,"
in Gwendolen M. Carter and Patrick O'Meara, eds., African Independence: The First Twenty-Five Years
(Bloomington: Indiana University Press, 1985, pp. 126-159.
(3) Yoweri Museveni of Uganda has been Africa's most outspoken recent opponent of multi-party
democracy, invoking Uganda's history of divisive and violent politics to justify
a "no-party" state, though he has in fact instituted competitive elections (c.f.,
Nelson Kasfir, "'No-Party Democracy' in Uganda," Journal of Democracy
vol. 9, no. 2 (1998), pp. 49-63). Presidents Bongo of Gabon, Bédié of Côte d'Ivoire,
and Biya of Cameroon all cited growing unrest in the society as reasons to justify
their hold on power, even while allowing elections to take place. (See articles
by David E. Gardinier, Robert J. Mundt, and Joseph Takougang in John Clark and David E
Gardinier, eds., Political Reform in Francophone Africa
(Boulder: Westview, 1997).
(4) C.f., Goran Hyden, "Governance and the Study of Politics," in Goran Hyden and Michael
Bratton, eds., Governance and Politics in Africa
(Boulder: Lynne Rienner, 1992), pp. 1-26.
(5) Charles Tilly, "War Making and State Making as Organized Crime," in Peter B. Evans,
Dietrich Rueschemeyer, and Theda Skocpol, Bringing the State Back In
(Cambridge: Cambridge University Press, 1985), pp. 169-191.
"Since governments themselves commonly simulate, stimulate, or even fabricate threats
of external war and since the repressive and extractive activities of governments
often constitute the largest current threat to the livelihoods of their own citizens,
many governments operate in essentially the same way as racketeers" (p. 171).
(6) Catharine Newbury, "Rwanda: Recent Debates over Governance and Rural Development,"
in Goran Hyden and Michael Bratton, eds., Governance and Politics in Africa
(Boulder and London: Lynne Rienner, 1992), pp. 193-219; Filip Reyntjens, L'Afrique des grand lacs en crise: Rwanda, Burundi: 1988-1994
(Paris: Karthala, 1994), pp. 89-130. On economic development and disparity, see
Ferdinand Bézy, Rwanda: Bilan socio-économique d'un régime, 1962-1989
(Louvaine-la-Neuve, Institut d'étude des pays en développement, 1990).
(7)Africa Watch, "Rwanda, Talking Peace and Waging War: Human Rights since the October
1990 Invasion," Short Reports Vol IV, no. 3 (New York: Africa Watch, February 27,
1992); Association rwandaise pour la défense des droits de la personne et des libertés
publiques (ADL), Rapport sur les droits de l'homme au Rwanda, septembre 1991-septembre 1992
(Kigali: ADL, December 1992).
(8)Interviewed at Mbazi, Butare, January 25, 1993, by C.K., in Kinyarwanda, translated
into English by author.
(9)Interviewed in Biguhu, Kibuye, December 26, 1992, by O.N., in Kinyarwanda, translated
into French by author.
(10)Fédération Internationale des Droits de l'Homme, Africa Watch, et al, Rapport de la commission internationale d'enquête sur les violations des droits de
l'homme au Rwanda dèpuis le 1er octobre 1990 (7-21 Janvier 1993)
(Paris: FIDH, March 1993); Africa Watch, "Beyond the Rhetoric: Continuing Human
Rights Abuses in Rwanda," (New York: Africa Watch, June 1993); ADL, Rapport sur les droits de l'homme
.
(11) Human Rights Watch Arms Project, "Arming Rwanda: The Arms Trade and Human Rights
Abuses in the Rwandan War," (New York: Human Rights Watch, January 1994).
(12) René Lemarchand, Burundi: Ethnocide Discourse and Practice
(Cambridge: Cambridge University Press, 1994); Reyntjens, L'Afrique des grand lacs en crise
, pp. 17-87, 227-248.
(13) Fédération Internationale des Ligues des Droits de l'Homme, Commission internationale d'enquête sur les violations des droits de l'homme depuis
le 21 octobre 1993
(Paris: FIDH, July 1994); Gaëtan Sebudandi et Pierre-Olivier Richard, Le drame burundais: Hantise du pouvoir ou tentation suicidaire
(Paris: Karthala, 1996); André Guichaoua, Les crises politiques au Burundi et au Rwanda (1993-1994)
(Lille: Université des Sciences et Technologies de Lille, 1995).
(14) I have discussed this period in greater detail in Timothy Longman, Proxy Targets: Civilians in the War in Burundi
(New York: Human Rights Watch, March 1998), pp. 10-24.
(15) Ibid, pp. 15-80.
(16) Interview June 25, 1997, by the author in French.