Pan-ethnicity and memories of violence in pre-1994 Burundi and Rwanda

On 6 April 1994 the plane carrying President Juvénal Habyarimana of Rwanda and his Burundian counterpart Cyprien Ntaryamira was shot down while approaching Kigali airport. Returning from a regional summit in Tanzania, the two statesmen died when two surface-to-air missiles downed the aircraft. The incident remains unprecedented in African history and is memorable in at least two ways. It not only unleashed Rwanda’s machinery of genocidal violence that massacred hundreds of thousands of Batutsi and moderate Bahutu, but also symbolises the close interlocking of the political histories of Rwanda and Burundi. Since achieving independence, the two societies seem to have been bound together by cycles of political violence and enduring ‘ethnic conflict’ that transcends national borders.

This article tries to disentangle these pre-1994 histories and analyse the narratives and processes that have shaped the political paths the two countries have taken since. Instead of providing a mere comparison, this paper wants to explore the ‘points of interaction’ before, during and after conflict. To this end, it sheds light on two particular aspects; First, the formation of seemingly irreconcilable ethnic imagined communities (Anderson 2006) that span both societies, linking Rwandan Batutsi and Bahutu with their Burundian counterparts and vice versa. These pan-ethnicities laid the foundations for ‘ethnic’ mobilisation that pitted two perceived kin-groups against each other. Second, this paper attempts to analyse memories of violence and persecution that became engrained in the psyches of each of these communities, furthering pan-ethnic solidarity on the inside, and antagonisms towards others. Memories of past injustice nurtured long-lasting grievances which consequently translated into the renewed flare-up of conflict. Political assassinations are an example of such events that became the tipping point of pent-up animosities to open tit-for-tat violence. Bloodshed in one of the two countries often served as a catalyst for escalating political strife, group suspicion, and retributive killings in the other.

Socially constructed ethnic binaries that were conceived during colonialism identified Batutsi as alien ‘settlers’ and Bahutu as their ‘native’ victims. These narratives contributed to a social climate in which violence and counter-violence could lead to dangerous vicious circles and cycles of violence.

COLONIALISM AND THE INVENTION OF ‘KIN-COUNTRIES’

Rwanda and Burundi are often studied presuming they share an almost identical culture and history. In fact, the similarities are striking. Prior to colonialism, both had monarchical systems in which a king (Mwami) and a royal court ruled over the peasantry, belonging to the social categories of Bahutu (85%), Batutsi (14%) or Batwa (1%) (Newbury 1995, 12). When the first colonisers arrived, they elevated this social hierarchy to quasi-universal validity, declaring the racial superiority of the Batutsi. The threefold classification was particularly common in the Rwandan case. In neighbouring Burundi, these social relations were more complex, accommodating a princely elite (Baganwa) and a further division between lower-class (Bahima) and upper-class Batutsi (Banyaruguru) (Uvin 1999, 255). Within each society, Bahutu and Batutsi were also fragmented along regional or class lines, precluding any notion of pan-identity. German and Belgian authorities treated the two kingdoms as a single polity by the colonial name of ‘Ruanda-Urundi’. Rwanda’s strict hierarchical order, allowing for less mobility between Bahutu and Batutsi, was also introduced in Burundi and served as a template for making sense of its purported ethnic puzzle (Uvin 1999, 255). Differences were dismissed as anomalous; a disorder to be dealt with through social engineering. Colonial policies sought to harmonise social relations by emphasising ‘racial’ difference and an administrative centralisation of power. This also entailed the annexation of previously autonomous micro-kingdoms in Burundi (Newbury 2001, 262-265). Between 1927 and 1936, Belgian authorities institutionalised racialism by granting different legal status to Bahutu, Batutsi and Batwa in both trust territories [1]. This rigidification of social identities stood in sharp contrast to more flexible customary identity markers, such as status, occupation, or economic dependency which had been prevalent before (Mamdani 2001, 87f.; Van de Meeren 1996, 253-6). Further, the colonial enterprise produced knowledge that reflected the popularity of scientific racism in 19th century Europe (Pels 1997, 174). Batutsi minorities in Rwanda and Burundi were conceived as ‘Hamitic’ invaders that had presumably subdued the culturally ‘inferior’ Bantu majority. Accordingly, ordinary Batutsi received more political and social rights than their underprivileged Bahutu counterparts. Since, colonial narratives have served as justifications for structural discrimination against either group, making these myths the “handmaidens of war propaganda” (Vansina 1998, 38).

Relations between the pre-colonial kingdoms were ambiguous. Rwanda felt encircled by its close-by rivals of which Burundi was one of the strongest. Mutual raiding expeditions and short-term territorial conquests characterised their interactions, making long-term alliances rare (Vansina 2004, 156). Socio-political differences outbalanced relative cultural proximity. In Rwanda, clientship for land (ubukingi) and cattle (ubuhake) was more exploitative in character than similar institutions in Burundi (ubugabire). The latter’s more sophisticated social compartmentalisation found no equivalent in Rwanda. By attempting to standardise these diverging social orders, the colonial powers virtually invented the very ‘kin-countries’ they believed to have discovered. Producing such knowledge was the sole prerogative of the colonial machinery that was in the position to (re-)define and interpret what they perceived as ‘ethnic’ hierarchies (see Pels 1997). By introducing legal documents to classify their subjects, the colonisers practically mummified previous social identities and imposed their version of history. Although colonial policy played a role in exacerbating social inequalities and legalising inter-group relations, it did not create them. The tempting conclusion that inter-communal conflict in Burundi and Rwanda is entirely the product of external meddling is therefore a fallacy.

In the late 1950s, the Belgian administrators and clergies in Rwanda shifted their support from the Batutsi minority (and the monarchy) to an emerging Bahutu intelligentsia (Longman 2001, 168). The latter had experienced discrimination under colonial rule and now rallied against the Batutsi monarchy as an instrument of oppression while leaving the Belgian colonial administration intact. In 1957, they had formulated a so-called ‘Bahutu Manifesto’ that blamed the Batutsi for a system of indirect rule that the Europeans had set up. Two years later, a ‘Hutu rebellion’ (1959-1962) deposed the monarchy and turned Rwanda into a republic. For the first time, Batutsi were targeted as a group and labelled ‘collaborators’ and proponents of minority rule, making an estimated 120,000 flee the country and settle across the border in Burundi and Uganda (Van der Meeren 1996, 252). Rwanda’s political upheaval and anti-Tutsi pogroms set a dangerous example for its closest neighbour Burundi where the monarchy had endured. Polarisation and violence in Rwanda made the Burundian Batutsi anxious about suffering a similar fate (Lemarchand 1995, 60). Despite allusions to ethnicity, the exclusivist political systems and the synergy between the two countries were the causes of recurrent conflict (Lemarchand 1997, 7). Burundi’s multi-ethnic (but Batutsi-dominated) ruling party Union pour le Progrès National (UPRONA), under the charismatic leadership of Prime Minister Louis Rwagasore, was no longer able to thwart political extremism (Lemarchand 1995, 59). Rwagasore was murdered in October 1961, being the first victim in a long row of political assassinations.

PAN-ETHNICITY, THE SECURITY DILEMMA AND ‘1972’

Both Rwanda and Burundi achieved independence from Belgium in 1962. At the time, Burundi was already home to over 50,000 Batutsi fleeing violence and ‘ethnic’ pogroms in Rwanda. Political instability in Burundi was further fuelled by an attempted military coup staged by Bahutu officers in 1965 (Lemarchand 1995, 60). Taking advantage of the volatile situation, the Defense Minister Michel Micombero seized power and established a military junta dominated by his own Batutsi-Hima. A year after the putsch, his regime abolished the Baganwa monarchy, consecutively excluding the Bahutu majority and the privileged Batutsi-Nyaruguru from political power. Batutsi refugees from Rwanda hoped to benefit from this regime change that had produced a government under Batutsi leadership. Regardless of the rivalry between Bahima and Banyaruguru factions, Burundi’s military was perceived as a potential patron for all Batutsi in the region, particularly in the midst of Rwanda’s lingering confrontations (Lemarchand 1995, 61). Bahutu extremists took the fact that Batutsi refugees found a safe haven in Burundi as proof of an existing ‘Batutsi plot’. In turn, Micombero’s regime depicted the Bahutu majority as a sleeping giant that might, akin to Rwanda’s ‘rebellion’, overpower and victimise the Batutsi. Hence, colonial narratives of ‘settlers’ and ‘natives’ (see Mamdani 2001) resurfaced, pitting one group against the other, reminiscent of a ‘Darwinistic struggle’ for supremacy and survival.

Grégoire Kayibanda’s regime in Kigali became aware of Bujumbura’s obvious toleration - if not support - for Batutsi “refugee warriors” from Rwanda (Adelman 1998). The refugee camps along the Burundian border became hotbeds of Batutsi resistance against Kigali. Military cross-border incursions (1961-66) by Batutsi guerrillas prompted immediate reprisals from Rwanda’s army (Long 2012, 220f.; Reed 1996, 481). The Rwandan government portrayed the insurgents as the long arm of Bujumbura’s regime and blamed all Batutsi collectively for the attacks. The incrimination of the entire group was symptomatic for the emergence of corporate identities. Pan-ethnic sentiments, which had been absent in the pre-colonial era, developed rapidly in the wake of inter-communal violence. Bahutu and Batutsi in both Rwanda and Burundi feared each other siding with their respective ‘kin’ across the border (Bhavnani & Lavery 2011, 241). Although unfounded, this theory gained currency among ordinary citizens and the elites. Colonial fascination with the ‘Hamitic myth’, coupled with a social revolutionary re-interpretation of Batutsi-Bahutu relations in pre-colonial times, was the basis for Bahutu suspicion towards the Batutsi, and vice versa the fear of extermination. Conspiracy theories reinforced the claim that both groups collaborate with their ‘ethnic kin’, creating what Appadurai calls “the fear of small numbers” in Rwanda and an anxiety towards the majority in Burundi (Appadurai 2006; Byford 2011, 121). Pan-ethnic “communities of fate” (Okamoto 2003, 814) evolved, based not on actual, but imaginary loyalties.

Kaufman considers this situation as a nascent ‘security dilemma’. He regards group suspicion and a logic of inevitability as key obstacles to the peaceful co-existence of two identity groups (1999, 139). Despite the initial non-existence of transnational Bahutu and Batutsi identities, the fact that the two groups intermittently perceived each other as such generated a notion of imagined communities (Anderson 2006). Further, these bonds of imagined kinship materialised through the conflict-driven interaction between Rwanda and Burundi, making them appear irreconcilable and hostile. During the post-colonial era, this dilemma played out as political violence motivated by events in either of the two countries. Lemarchand accordingly describes this process as a “perverse dialectic” (1995, 30). Kigali and Bujumbura resorted to invocations of an ‘ethnic other’ or a threatening ‘fifth column’ (ibyitso) that could legitimise their repressive policies. Indeed, mobilisation along (pan-) ethnic lines was used to feign political cohesion domestically.

In the early 1970s, Burundi’s president Micombero grew wary of his own elites and began arresting senior officers and civil servants affiliated with the Batutsi-Nyaruguru rival faction (Lemarchand 1995, 88). As a response to Kigali’s increasingly discriminatory anti-Tutsi policies, he cracked down on the country’s Bahutu majority, spreading rumours of a ‘Hutu conspiracy’. In turn, Rwanda’s government took the Burundi-based inyenzi raids as a pretext to further marginalise their Batutsi population, forcing many into exile (Reed 1996, 481). In April, small bands of Bahutu insurgents seized the opportunity of intra-Tutsi power struggles to attack government posts and military facilities in Burundi’s Nyanza-Lac and Rumonge. They made quick territorial gains while massacring thousands of Batutsi. Soon, they were overpowered by the Batutsi-led army that in revenge began to indiscriminately kill Bahutu civilians and ravage the surrounding villages (Lemarchand 1995, 91f.). What began as a counter-insurgency operation rapidly turned into a “selective genocide” against the educated Burundian Bahutu population, prompting the disappearance of students, teachers, officials and public figures (Lemarchand 2008, 4; 1995; Abrams 1995, 148). Aside from the military, the youth league of UPRONA - Jeunesse Revolutionnaire National/ Rwagasore (JRN/JRR) – took part in the anti-Hutu campaign and thereby foreshadowed the type of popular participation that became emblematic of Rwanda’s 1994 genocide. A massacre against the Bahutu, which had hitherto mainly existed as anti-Tutsi propaganda aggravating the security dilemma, promptly became reality. The number of people who perished during the genocidal campaigns is usually estimated at 100,000-200,000, whereas some 150,000 escaped to Tanzania and Zaire (Loft & Loft 1988, 91; Lemarchand 2002, 551). Lemarchand claims that the 1972 killings constitute the root cause of the regions cycles of violence (1998, 3). This interpretation, however, understates the significance of the ‘Hutu rebellion’ thirteen years earlier. In fact, both events were uniquely remembered and thus fuelled cycles of violence. These dynamics will be discussed in-depth below.

PRELUDE TO THE NEXT GENOCIDE: MEMORIES AND CYCLES OF VIOLENCE

Collective Memories of Violence

Memories define human action. They retain knowledge, and are a constant reminder of past success and failure, however, they can also change over time. Durkheim’s student Morice Halbwachs explores the way that memories can extend beyond the individual (Shaw 2010, 251). For him, the only possible memories are in fact ‘collective memories’, as “the mind reconstructs its memories under the pressure of society” (Halbwachs 1992, 51). Argenti and Schramm add that ‘collective memories’ are often transmitted from one generation to the next and thus transcend a single temporal sphere (2010, 26f.).

Especially during conflict, memories are retained collectively and articulate with social reproduction (Shaw 2010, 252). In Rwanda, the eruption of violence occurred in cycles that correlated with parallel events in neighbouring Burundi and vice versa. Hence, atrocities were stored in social memories that widened the gap between Bahutu and Batutsi. Many experienced violence, exile and oppression first hand. Others, especially those brought up or born in exile, were merely socialised with the traumas of their (grand)parents. Their political engagement was largely shaped by these secondary ‘memories of violence’ and limited by the boundaries of a common consciousness (Argenti & Schramm 2010, 5). In the diaspora, this collectivisation of memory was reinforced by the geographical concentration of refugees.

Batutsi expulsions during the Rwandan Bahutu ‘rebellion’ (1959-1962) aroused fears of group extermination (Long 2012, 218). They marked a watershed in the development of genocidal tendencies that began to crystallise throughout the region. For Batutsi, especially refugees, the ‘Hutu rebellion’ became a historical point of reference, a demarcating line between two eras. The period was remembered as a “dramatic rupture in the continuity of historical space and time” (Masalha 2009, 37), echoing similar experiences of Palestinians in 1948. In a way, the events were reproduced and remembered as a Batutsi nakba [2], the catastrophe that ushered in spirals of violence and ultimately genocide. Similarly, the atrocities of ‘1972’ became the symbol of Bahutu remembrance. Overshadowed by ‘1959’, Burundi’s ethnic relations were strained by the fear of Batutsi persecution. The organised killing campaigns of 1972 created equally strong memories and traumas for the Bahutu (Weinstein 1972, 28; Schramm 2011).

The past was seen as a warning for the future. Huyssen (1995) states that remembrance is therefore not the unmediated reconstruction of the past but a reflection of the context and aspirations of those that remember. Ongoing crises and political oppression endorsed remembrance as an act of resilience or even resistance (Olzak et al 1996). Metanarratives, stressing a purportedly ‘historical struggle’ between Batutsi and Bahutu, reverberated in both communities and marginalised alternative versions of history, until the present day (see Jefremovas 1997). This constant re-negotiation and revision of the past gave rise to new disputes over truth and meaning that began to translate into new violent conflicts shortly after (Wilce 2002, 159).

Political Turmoil and the 1988 Massacres

In 1973, barely half a year after Burundi’s massacres, Rwandan President Kayibanda was ousted by Juvénal Habyarimana as a result of intra-Hutu power struggles (Mamdani 1996, 17). Although discrimination against the Batutsi lessened under the new regime, political instability extended to neighbouring Burundi. In 1976, Micombero was overthrown in a coup by his rival Jean-Baptiste Bagaza whose anti-Western and decisively anti-clerical policies stemmed from Belgium’s support for the Rwandan ‘Hutu rebellion’ that was still vividly remembered. While Habyarimana tried to ease ethnic tensions, Bagaza effectively laid the foundations for a “system of ethnic apartheid” (Abrams 1995, 150). This, and the memories from 1972, augmented feelings of martyrdom among the exiled Bahutu (Turner 2013, 18), resulting in the creation of the pan-Hutuist Parti pour la libération du peuple Hutu (PALIPEHUTU) in a refugee camp in 1980 (Scherrer 2002, 47; Lemarchand 1995, 104). Refugees not only preserved their grievances, but also converted them into concrete political action. Hence, remembering became a source of conflict in itself.

Due to Burundi’s isolationist policies and its ailing economy, Bagaza was eventually deposed by Pierre Buyoya in 1987. Buyoya immediately faced re-emerging Hutu activism. In Kirundo and Ngozi provinces, close to the Rwandan border, the Bahutu opposition was strongest. As in Rwanda, the global coffee price inflation also contributed to an economic downturn in Burundi which predominantly affected Bahutu producers (Lemarchand 1995, 122). Radical politicians on all sides exploited the socio-economic insecurity by spreading rumours of imminent inter-ethnic violence and secret plots. Mobilisation and paranoia reached a peak in summer 1988. Social unrest in the communes Marangara and Ntega evolved into the killing of Batutsi and the burning of their homes, provoking a devastating response by the Batutsi army. The retaliatory attacks were disastrous, claiming the lives of tens of thousands and displaced more than 50,000 Bahutu (Scherrer 2002, 41-43; Lemarchand 1989, 22; Loft & Loft 1988, 92; Bhavnani & Backer 2000, 285). Unlike 1972, the incidents remained geographically limited and the number of victims (an estimated 20,000) was comparably ‘low’. Nevertheless, the events evoked unease among Bahutu, who feared a repetition of ‘1972’.

Rwanda’s elite profited from the continuously violent conduct of Burundi’s army. Since independence, the two countries had mirrored each other’s behaviour in terms of ethnic policies, outbreaks of violence, sponsoring of pogroms and expulsion. Often, action and re-action unfolded in cycles of violence over longer periods of time. After 1988, while Buyoya re-gained international respectability with a “new approach” to ethnicity, Habyarimana’s Rwandan government radicalised noticeably under the pressure of extremist “Hutu Power” forces (Daley 2006, 670; Prunier 2010). Through the interplay of long memories and transnational solidarities, the resurgence of violence in one country cast its shadow onto the other. The beginning of the 1990s confirmed this tendency.

Until The Next Genocide

In October 1990, the Rwandan Batutsi exiles of the Rwandan Patriotic Front (RPF) began their invasion of Rwanda (Lemarchand 1995, Prunier 2010). Rwanda’s crippling economy, domestic pressure within Uganda, as well as rising ethnic polarisation created a window of opportunity. For Habyarimana and the akazu [3], the RPF was the embodiment of modern ‘Hamitic’ invaders, actualising the fears that ‘Hutu Power’ propaganda had previously spread. Kagame’s rebels were further perceived as the reincarnation of Burundi’s Batutsi military regime that supposedly also tried to capture the Rwandan state. Both pan-ethnic fears and memories of past violence played out in the dialectic fashion that Lemarchand (1995) describes. Habyarimana’s government radicalised its language on ethnicity (ubwoko) to send a strong signal against the ‘Batutsi threat’ that it saw in the RPF (Andersen 2000, 448-50; also Langford 2005, 13). Kuperman (2004) captures this dynamics when stating that the RPF in fact “provoked” genocide, knowing that Rwandan Batutsi were collectively seen as their natural accomplices (ibyitso). In Burundi, a similar paranoia was at work, only with reversed roles. Bahutu exiles in Tanzania and Zaire continued infiltrating Burundian territory, sparking unrelenting government retaliation, as in November 1991 (Daley 2006, 670). With an armed opposition against both regimes, the ‘security dilemma’ unfolded inexorably, characterising the transformation from “refugee-generating conflicts into conflict-generating refugees” (Lemarchand 1997, 9). The 1993 Arusha Accords signed between the Habyarimana government and the RPF only temporarily improved regional security.

During the implementation phase of the accords, neighbouring Burundi was plunged into further political distress over the murder of democratically elected Muhutu president Melchior Ndadaye on 21 October 1993. Ndadaye’s death also had serious repercussion for Rwanda, as it sent a message to Bahutu extremists to never trust Batutsi again (Hintjens 1999, 277; Lemarchand 1995b, 10). However, political murder was not an entirely new phenomenon. In Burundi, assassinations had become the weapon of choice throughout the post-colonial era, making the deaths of Louis Rwagasore [1961">, Pierre Ngendandumwe [1965"> and lastly Ndadaye permanent reminders of an ‘existential threat’ (Ndikumana 1998). This method was less common in Rwanda, but also there targeted killings were used to instil fear in the Batutsi community, extending the list of remembered atrocities (Vansina 1998, 38). In 1993, regional dynamics were sent into a tailspin, catapulting Burundi into civil war and perfecting Rwanda’s descent into genocide which consumed innumerable lives and inflicted unprecedented suffering such as rape, mutilation and displacement on individuals (De Forges 1999, 13). The crisis was not simply internal, it was the interlude of a wider conflict characterised by regional spill-overs from Burundi to Rwanda and vice versa (Bhavnani & Lavery 2011, 234). By mid-July 1994 Rwanda’s genocide had ended. Meanwhile, a Burundian “genocide by attrition” (OHCHR 1996) was taking place which effectively lasted until 2005. Just as before, genocide and civil war produced new memories and narratives that continue to shape the lives of the affected communities, their present and their future.

CONCLUSION

The bond of political violence between Burundi and Rwanda symbolically culminated in the infamous plane crash on 6 April 1994. This article has discussed the way spiralling levels of violence and spill-overs have created seemingly self-sustaining conflict dynamics. For this, two driving forces are regarded as paramount; the development of pan-ethnic imagined communities of Bahutu and Batutsi in both Burundi and Rwanda, and the memories of past violence that have sustained existing grievances and have been transmitted from one generation to the next. Minority rule by Batutsi officers in Burundi, and a Rwandan Bahutu regime posed as each other’s mirror images and illustrated the dangers of ethnocratic rule. Rigidified social identities served as the basis for a social discourse in which inequalities were exacerbated, ethnicised and feelings of injustice remained unaddressed. This lack of reappraisal for the past led to the construction of corporate identities around the ‘collective memories’ of cataclysmic disasters in 1959-62, 1972, 1988 and 1993. While commemoration is a key component of transition from war to peace, politicised memories rather perpetuate conflict than raise the chances of reconciliation. Political agitation repeatedly fostered the resurge of conflict in one country and cast its shadow onto the other. Mobilisation along ‘ethnic’ lines within Burundi and Rwanda played an important role in forging pan-ethnic communities, or at least in facilitating their social imagination. The kin-country narrative and the ‘harmonisation’ through colonial policies has contributed to the interlocking of the two countries’ social and political structures. Thus, the false assumption that Burundi and Rwanda are a ‘natural pair’ has favoured comparative approaches over a necessary analysis of interaction.

After the 1994 genocide, Rwanda chose a strategy of ‘ethnic denial’ that had been previously professed by pre-1993 regimes in Burundi. Reyntjens has gone so far as to suggest that there was a post-1994 “Burundisation of Rwanda” (Reyntjens 2009, 29). Burundi’s civil war ended with the 2000 peace agreement and the establishment of an equitable quota system which recently reached a deadlock. Either way, cycles of violence have not been curbed. On the contrary, with the exodus of Bahutu civilians after the victory of the RPF in 1994, pan-ethnic imagined communities were projected into the eastern provinces of the DRC. Rwanda’s cross-border military operations, largely motivated by resource interests, also disclose pan-ethnic tendencies that make Congolese Kinyarwanda-speakers a constituency or proxy in Kigali’s wars. Most recently, this became salient during the ‘M23 rebellion’ in 2012. Thus, narratives that had previously fuelled conflict in Burundi and Rwanda, such as the ‘pan-Hamitic’ plot, were reactivated in the Congo. Memories of the Rwandan genocide and the unaddressed atrocities committed by the RPF are passed on to coming generations without proper closure. With the cessation of armed hostilities, and the ebbing of conflict, violence is translated into ‘social memories’. Thereby, grievances are not resolved, but remain merely suspended.

NOTES

[1] After the First World War Germany was deprived of its colonies, including Deutsch-Ostafrika (German East Africa). Ruanda-Urundi was subsequently administered by Belgium.
[2] The Nakba (Arabic: catastrophe) is widely remembered by Palestinians as the beginning of exile from their homeland in 1948 and as the underlying cause of the current conflicts. For more insights see Sa'di, Ahmad H. and Lila Abu-Lughod (2007): Nakba, Palestine, 1948, and the claims of memory, (New York: Columbia University Press).
[3] The so-called akazu (small house) was the inner circle of political power during Habyarimana’s rule. Key members of this clique were Habyarimana himself, his wife Agathe and her relatives.

REFERENCES

Abrams, Jason S. (1995): ‘Burundi: Anatomy of an ethnic conflict’, In: Survival, 37
(1), pp.144-164.

Adelman, Howard (1998): ‘Why Refugee Warriors are Threats‘, In: The Journal of
Conflict Studies, 18 (1), available: http://journals.hil.unb.ca/index.php/jcs/article/ view/11672/12401 (accessed: 06.03.2014).

Andersen, Regine (2000): ‘How Multilateral Development Assistance Triggered the
Conflict in Rwanda’, In: Third World Quarterly, 21 (3), pp. 441-456.

Anderson, Benedict (2006): Imagined Communities: Reflections on the Origin and
Spread of Nationalism, (New York/London: Verso).

Appadurai, Arjun (2006): The Fear of Small Numbers: An Essay on the Geography of
Anger, (Durham/London: Duke University Press).

Argenti, Nicolas, and Katharina Schramm (eds) (2010): Remembering Violence:
Anthropological Perspectives on Intergenerational Transmission, (New York/Oxford: Berghahn Books).

Bhavnani, Ravi, and David Backer (2000): ‘Localized Ethnic Conflict and Genocide:
Accounting for Differences in Rwanda and Burundi’, In: The Journal of Conflict Resolution, 44 (3), pp. 283-306.

Bhavnani, Ravi, and Jerry Lavery (2011): ‘Transnational Ethnic Ties and the
Incidence of Minority Rule in Rwanda and Burundi (1959–2003)’, In: Nationalism and Ethnic Politics, 17 (3), pp.231-256.

Daley, Patricia (2006): ‘Ethnicity and political violence in Africa: The challenge to the
Burundi state’, In: Political Geography, 25, pp.657-679.

Des Forges (1999): ‘Leave None to Tell the Story: Genocide in Rwanda’, Human
Rights Watch Report, http://www.hrw.org/reports/pdfs/r/rwanda/rwanda993.pdf (accessed: 05.04.2014).

Halbwachs, Morice (1992): On Collective Memory, (Chicago: University of Chicago
Press).

Hintjens, Helen M. (1999): ‘Explaining the 1994 Genocide in Rwanda’, In: Journal of
Modern African Studies, 37 (2), pp.241-286.

Huyssen, Andreas (1995): Twilight Memories: Marking Time in a Culture of Amnesia,
(New York/London: Routledge).

Jefremovas, Villia (1997): ‘Contested Identities: Power and the Fictions of Ethnicity,
Ethnography and History in Rwanda’, In: Anthropologica, 39 (1/2), pp. 91-104.
Kaufman, Chaim (1999): ‘Possible and Impossible Solutions to Ethnic Civil Wars’, In:
International Security, 20 (4), 136-175.
Kuperman, Alan J. (2004): ‘Provoking genocide: a revised history of the Rwandan
Patriotic Front’, In: Journal of Genocide Research, 6 (1), pp.61-84.

Langford, Peter (2005): ‘The Rwandan Path to Genocide: The Genesis of the Capacity
of the Rwandan Postcolonial State to Organise and Unleash a project of Extermination’, In: Civil Wars, 7 (1), pp.1-27

Lemarchand, René (1989): ‘Burundi: The Killing fields revisted’, In: A Journal of
Opinion, 18 (1), pp.22-28.

Lemarchand, René (1995): Burundi: Ethnic conflict and Genocide, (Cambridge/New
York: Woodrow Wilson Center Press/Cambridge University Press).
Lemarchand, René (1995b): ‘Rwanda: The Rationality of Genocide’, In: Issue: A
Journal of Opinion, 23 (2), pp. 8-11.
Lemarchand, René (1997): ‘Patterns of State Collapse and Reconstruction in Central
Africa: Reflections on the Crisis in the Great Lakes’, In: African Studies Quarterly, 1 (3), pp.5-22.

Lemarchand, René (1998): ‘Genocide in the Great Lakes: Which Genocide? Whose
Genocide?’, Yale Center for International and Area Studies Working Paper Series, GS03, pp.1-13.
Lemarchand, René (2002): ‘Le génocide de 1972 au Burundi: Les silences de
l'Histoire’ (The 1972 Genocide in Burundi: Historical Silences), In: Cahiers d'Études Africaines, 42 (167), pp.551-567.
Lemarchand, René (2008): ‘The Burundi Killings of 1972’, Online Encyclopedia of
Mass Violence, http://www.massviolence.org/IMG/article_PDF/The-Burundi-Killings-of-1972.pdf (accessed:03.04.2014).

Loft, Francis, and Frances Loft (1988): ‘Background to the Massacres in Burundi’, In:
Review of African Political Economy, 43, Feeding Africa: What Now?, pp.88-93.

Long, Katy (2012): ‘Rwanda's first refugees: Tutsi exile and international response
1959–64’, In: Journal of Eastern African Studies, 6 (2), pp.211-229.

Longman, Timothy (2001): ‘Church Politics and the Genocide in Rwanda’, In:
Journal of Religion in Africa, 31 (2), Religion and War in the 1990s, pp. 163-186.

Mamdani, Mahmood (1996): ‘From Conquest to Consent as the Basis of State Formation: Reflections on Rwanda’, In: New Left Review, 216, pp.3-36.
Mamdani, Mahmood (2001): When Victims become Killers: Colonialism, Nativism,
and the Genocide in Rwanda, (Princeton: Princeton University Press).
Masalha, Nur (2009): ‘60 Years after the Nakba: Historical Truth, Collective Memory
and Ethical Obligations’, In: Kyoto Bulletin of Islamic Area Studies, 3 (1), pp.37–88.
Ndikumana, Léonce (1998): ‘Institutional Failure and Ethnic Conflicts in Burundi’, In:
African Studies Review, 41 (1), pp. 29-47.

Newbury, Catherine (1995): ‘Background to Genocide: Rwanda’, In: Issue: A Journal
of Opinion, 23 (2), Rwanda, pp. 12-17.

Newbury, David (2001): ‘Precolonial Burundi and Rwanda: Local Loyalties, Regional
Royalties’, In: The lnternational Journal of African Historical Studies, 34 (2), pp.255-314.
OHCHR (1996): ‘Situation of Human Rights in Burundi’, Sub-Commission
Resolution 1996/4, Office of the United Nations High Commissioner for Human Rights (OHCHR), http://www.unhchr.ch/Huridocda/Huridoca.nsf/ffe4b27219c6 151f80256639004b1970/b190b72b8b40087080256656004f1bb4?OpenDocument (accessed: 05.04.2014)
Okamoto, Dina G. (2003): ‘Toward a Theory of Panethnicity: Explaining Asian
American Collective Action’, In: American Sociological Review, 68 (6), pp. 811-842.

Olzak, Susan and Suzanne Shanahan (1996): ‘Deprivation and Race Riots: An
Extension of Spilerman's Analysis’, In: Social Forces, 74, pp.931-961.

Pels, Peter (1997): ‚The Anthropology of Colonialism: Culture, History, and the
Emergence of Western Governmentality‘, In: Annual Review of Anthropology, 26, pp. 163-183.

Prunier, Gérard (2010): The Rwanda Crisis: History of a Genocide, (London: Hurst &
Company).

Reed, William C. (1996): ‘Exile, Reform, and the Rise of the Rwandan Patriotic
Front‘, In: The Journal of Modern African Studies, 34 (3), pp 479 -501.

Reyntjens, Filip (2009): The Great African War: Congo and Regional Geopolitics,
1996-2006, (Cambridge: Cambridge University Press).

Scherrer, Christian P. (2002): Genocide and Crisis in Central Africa: Conflict Roots,
Mass Violence, and Regional War, (London: Praeger).

Schramm, Katharina (2011): ‘Introduction: Landscapes of Violence: Memory and
Sacred Space’, In: History and Memory, 23 (1), pp.5-22.
Shaw, Rosalind (2010): ‘Afterword: Violence and the Generation of Memory’, In:
Argenti, Nicolas, and Katharina Schramm (eds): Remembering Violence: Anthropological Perspectives on Intergenerational Transmission, (New York/Oxford: Berghahn Books), pp.251-260.
Turner, Simon (2013): ‘Mirror Images: Different Paths to Building Peace and Building
States in Rwanda and Burundi’, Danish Institute for International Studies (DIIS) Report, pp.1-44.

Uvin, Peter (1999): ‘Ethnicity and Power in Burundi and Rwanda: Different Paths to
Mass Violence’, In: Comparative Politics, 31 (3), pp. 253-271.

Van der Meeren, Rachel (1996): ‘Three Decades in Exile: Rwandan Refugees 1960
1990’, In: Journal of Refugee Studies, 9 (3), pp.252-267.

Vansina, Jan (1998): ‘The Politics of History and the Crisis in the Great Lakes’, In:
Africa Today, 45 (1), pp. 37-44.

Vansina, Jan (2004): Antecedents to Modern Rwanda – The Nyiginya Kingdom,
(Oxford: James Curry).

Weinstein, Warren (1972): ‘Tensions in Burundi’, In: Issue: A Journal of Opinion, 2
(4), pp. 27-29.

Wilce, James (2002): ‘Genres of Memory and the Memory of Genres: “Forgetting”
Lament in Bangladesh’, In: Comparative Studies in Society and History, 44 (1), pp.159-185.

* Hanno Brankamp is currently completing his postgraduate studies in International Security at the University of St Andrews, UK. He also holds a B.A. degree in Area Studies Asia/Africa from Humboldt University in Berlin with a specific focus on conflicts and security in Eastern Africa, including the Great Lakes and the Horn.

* THE VIEWS OF THE ABOVE ARTICLE ARE THOSE OF THE AUTHOR/S AND DO NOT NECESSARILY REFLECT THE VIEWS OF THE PAMBAZUKA NEWS EDITORIAL TEAM
* BROUGHT TO YOU BY PAMBAZUKA NEWS
* Please do not take Pambazuka for granted! Become a Friend of Pambazuka and make a donation NOW to help keep Pambazuka FREE and INDEPENDENT!
* Please send comments to editor[at]pambazuka[dot]org or comment online at Pambazuka News.